Journal of the Department of Wildlife Conservation 2006-1: 113-118 SOCIAL PROBLEMS AND ECONOMIC POTENTIALS OF DOMESTICATION OF ELEPHANT AS A MEAN OF ELEPHANT CONSERVATION IN SRI LANKA A.M.S.T. ATHAUDA* Department of Agricultural Economics and Business Management Faculty of Agriculture,University of Peradeniya, Sri Lanka ABSTRACT Loss of elephant habitats appears to be the main reason that leads to human elephant conflicts in most parts of the Dry Zone in Sri Lanka. Capturing and domestication of wild elephants is identified as one of the means of elephant conservation which promote the coexistence between humans and elephants that existed for generations. However, with the recent structural changes in the society and the economy such as mechanization and increased wage rates, captive elephant conservation may not be attractive as it was before. With this background, this study aimed at identifying the social as well as economic problems, and suggesting possible remedies that help to promote domestication of elephants. The study was based on collecting the data from a survey and discussions with key informants. The survey was comprised of a mail survey based on structured questionnaire and face-to-face interviews. The study found that since elephants are reared in Wet Zone, feed availability is not a major problem. However, most of the elephant owners have to make payments for the feeds. Indigenous, western or both veterinary practices are used for captive elephants. Generally, a single mahout is used to control an elephant and most of the mahouts live below the poverty line. Most of the elephants in homegarden are reared for prestige, while temples and devalas keep elephants for ceremonies. Hotels and orphanages rear elephants with the intention of attracting tourists. When compared costs and revenues, orphanages and hotels enjoy with marvelous profits. Thus hotels that receive profit can rear captive elephants, where there is a demand for elephant safaris, rides, etc. In addition to that, tamed captive elephants can be used to establish an elephant park, which is a win-win solution for both owners and people who take pleasure from elephant related activities. INTRODUCTION The Asian elephant (Elephas maximus) represents one of the most seriously endangered species of large mam in the world. It is one of the few terrestrial megaherbivores extant in the world. In Sri Lanka elephants are declared as a protected species in 1937, under the enactment of the Fauna and Flora Protection Ordinance. During past five decades, the Sri Lankan population declined drastically mainly due to loss of habitats caused by deforestation and forest fragmentation. Domestication of elephants for various purposes including logging and cultural purposes has bear a practice in Sri Lanka and other African countries. Capturing and domestication of those wild elephants can be considered as a one of the measure of elephant conservation. The number of purposes with regard to rearing of captive elephants are gain income, social prestige, use in ceremonies for promote tourism activities. In practice, mostly captive elephants are kept for one more of alternative purposes. Presently, there are about 15,000 elephants in captivity in the world (Sukumar, 1986) while 214 captive elephants in Sri Lanka (Santiapillai, 199 114 Mahouts are the people who handle the elephants. In early times mahouts are available from generation to generation and they are well-trained people. However, now adays people reluctant to work as mahouts. Present status of captive elephant conservation in Sri Lanka At present, four types of captive elephant conservation methods are formed in Sri Lanka. They are individual people keep in their home gardens, temples and devalas keep in their garden, hotels keep captive elephants for tourism and keep captive elephants in the orphanages. Jainudeen and Jayasinghe (1970) stated that the more wealthy and aristocratic families were living in those areas and they keep elephants for prestige. The studies on captive elephants clearly indicate that captive elephant population is declining rapidly. There were 736 captive elephants in 1946(Santiapillai & De Silva), while in year 1955 there were 670 (Dereniyagala, 1955), According to the DWLC reports there were 344 domesticated elephants and this was declined to 214 by 1997(Jayewardene, 1997). The elephants in captivity decline over the time due to a number of reasons. Most of the captive elephants are too old thus their lifetime declines rapidly, absence of practicing breeding programmes among captive elephants, government banned to capture elephants from the wild and lack of veterinary practices and lack of availability of well-trained mahouts are rare of there. Most of the captive elephants feed kitul, jack and coconut. Elephants need ample amount of feed per day. Dry matter need of the elephant is 0.5% of the body weight and daily ration should be according to the age groups (Krishnamurthy, 1992). Therefore, finding of this amount of feeds is much more difficult. Because growth rate of those trees are lower than the feed requirement rate of the captive elephants. However compared to the world captive elephant population, Sri Lanka’s captive elephant population is fairly low. Large numbers of captive elephant population are found in Myanmar, Thailand, India and Lao. Fifty percent of world captive elephant population lives in those countries. In Myanmar, captive elephants are used for many works such as, timber extraction (logging), transportation (as baggage elephants in hilly forests), religious ceremonies and processions, capturing wild elephants, state functions and ceremonies, tourism and agricultural activities in difficult areas. India possesses 20 percent of captive elephant population of Asia and they are used of the cultural activities in India and provide lot of care for them. Problem justification As discussed earlier, deprivation of habitats for elephants appear to be the main problem of elephants. Given the limited forest coverage of 22% of total land area in Sri Lanka, it has been identified domestication of wild elephants as one of the solution of mitigating human elephant conflict in Sri Lanka by a number of experts. 115 Though the practice of capturing and rearing has been done in Sri Lanka, for generations, given the limited land reserves and substitution by mechanical equipments, this option has to be justifiable economically. Of new areas of captive elephants, though tourism appears to be somewhat attractive, its economic potential has not been investigated. Therefore, in order to recommend elephant domestication on measures of conservation, it is of paramount importance to investigate the economics and the social aspects of the various means of captive elephant conservation Against this background the objective of this study are to get an insight on present socio economic condition of captive elephant conservation so as to identify the problems of the elephant owners and then information would be helpful to develop meaningful recommendation to promote domestication as a mean of elephant conservation in Sri Lanka. METHODOLOGY The study is manly based on data collected through primary survey. The data were collected using a structured questionnaire which includes general information, information on feeding, veterinary practices, bathing facilities, mahouts, income and specific information based on different type of captive elephant conservation methods. Face-to-face interviews and mail surveys were used to collect the data based on snowball sampling. The target population of the study was composed of people who own elephants. Basnayake Nilames or chief incumbent of temples, chief of devalas, hotel owners who rear elephants and the officer-in-charge of the Pinnawala Elephant Orphanage. The sample consisted, nineteen elephant owners, eight Buddhist priest or Basnayake Nilames, two hotel owners and the officer- incharge of the Orphanage at Pinnawala. The data analysis includes computation of the descriptive and cross tabulation. Statistical software Minitab was used for the analysis. RESULTS AND DISCUSSION The situation presents the current status of rearing of elephants such as sources of feed, feed type, feed availability, feed cost, veterinary practices, and mahouts, bathing facilities, cost and revenue. Source of receiving elephants DWC does not allow capturing elephants from the wild. As a result, remaining captive elephant number increase or renew only through breeding programmes. However, breeding programmes are not generally practice in domestic elephants. The survey finds that the most of the elephants (54.44%) reared in homegarden are reared by generations (Figure 1). In temples and devalas, most of the elephants were purchased from outside. This is in additions to the donation by Presidents and Priministers that is account for 25% of the 116 domestic elephants. In temple of tooth relic, this was about 100%. The elephants in Pinnawala are directly captured by the DWC. However this was done by early stages and most of that elephants were captured due to bad health conditions and some were translocated from war areas. According to the different studies there are more number of male elephants in the wild and capturing of males will not be a big problem. In addition to that orphanage increase captive elephant number through breeding programmes since they have huge potential of spending ability and veterinary practices. As a result captive elephant number in orphanage increases with time. Though private owners show a high demand for tamed elephants and prefer adults, as those are directly profitable it is very difficult to do it more pragmatically. Capturing of elephants from the wild is a very huge task and need much cost and man power to do it. Well trained people are needed to that activity. After capturing translocation have to practice and it is very difficult task. The most common method for catching elephants is tranquillizing and translocation them by truck. As a result direct capturing cannot be practiced by individual people and temple and devala, who need elephants. Percentage based on type of conservation 60 50 Reared by generation Percentage 40 Out side purchase Direcly captured 30 Puchased from an auction Other 20 10 0 Home garden Temple & devala Hotel Orphanage Type of conservation Figure 1. Source of receiving elephant Types and sources of feed Feed type Dung studies indicate that elephants eat around 140 different species, approximately 120kg per day. They prefer grasses, thorny and seasonal plants which have less chemical defenses. Among those feeds most common types of feeds are Kitul, coconut and jack. Captive elephants in homegarden are mainly fed on Kitul (32.73%), coconut (27.27%) and jack (25.45%) (Figure 2) Due to the weakness of some captive elephant, owners give sustenance feed for them. 117 Eventhough feed cost is high, owners provide those feeds, because of keeping elephants for social prestige. This special feed ration include mixture of feeds, eight breads, federal, cytoxine, protinex, soya flour(1kg), glucose, mung flour(500g), rice flour(1kg), four samaposa packets, sesame, rulan and Nestomalt. This feed ration give daily with the purpose of improving health conditions of the weak elephants. Consequently, elephants in hotels require much more feed compared with others. Orphanage also had a special feed ration, which include kitul leaves, kitul trunk, jack leaves, coconut leaves and mixture of other leaves which increase taste. These are bought from outside contractors. There are five baby elephants also in the orphanage. These baby elephants were fed in milk, in addition to leaves. Each baby elephant receive a packet of Lactogen per milk time and there were five milk times per day. Percentage based on type of conservation 45 40 Percentage 35 Kitul 30 25 Coconut 20 Jack 15 other 10 5 0 Home garden Temple & devala Hotel Orphanage Type of conservation Figure 2. Feed types Feed availability According to the results of the survey, feed availability is high among areas where captive elephants are reared. Most of these areas belong to wet zone, thus vegetation is available throughout the year. However, some of people who rear captive elephants in homegarden indicated difficulty in finding feeds while others experienced that it is extremely difficult to find the feeds (Figure 3). Though temples and devalas possess high amount of land area, they do not keep elephants in those lands, because of lack of water availability and feeds. Orphanage normally has enough amounts of feeds, as they purchase them suppliers from outside. Since orphanage is a government institution, their budget is covered through government fund allocation and in addition to that they earn huge income which could be able to spend buy feeds from out side suppliers. 118 Percentage based on type of conservation 120 Very high Percentage 100 High 80 Difficult 60 Extremly difficult 40 20 0 Home garden Temple & devala Hotel Orphanage Type of conservation Figure 3. Feed availability Source of finding of feeds Elephant owners find the feeds for elephants either from freely or with some payment. All types of elephant owners buy feeds for their elephants (Figure 4). Since orphanage possesses seventy captive elephants, they cannot find such amount of feeds with freely, also they have enough amounts of funds to allocate feeds and as a result they totally depend on outside feed suppliers. Most of the people rear elephants in homegarden and temple and devala, owned large land areas. Thus they can find some amount of feeds with freely. Some elephants that are in hotels are reared by generation, hence they possess large land areas, and thereby they obtain part of elephant feeds with freely. However all types of elephant owners buy large proportion of feeds from outside suppliers and spend huge cost on feed. Pecentage based on type of conservation 120 Percentage 100 80 Free 60 Paym ent 40 20 0 Hom e garden Tem ple Hotel Type of conservation Figure 4. Source of finding of feeds 119 Orphanage Feed cost Pinnawala orphanage incurs a very high feed cost per elephant relative to other conservation strategies, which is around Rs.3,64,285 (Figure 5). Temples and devalas also spend high cost for feeds of elephants, because they possess paddy fields and other income sources. As a result, they spend much on feeds. Since hotels concern about profit, they spend less amount of money on feeds. However, the elephant feed costs of hotel is Rs.300-500 /elephant/day and around 4-5 people take care for the food supply. Most food comes from the coconut plantations nearby. Feed cost 142642 364285 214735 Home garden Temple & devala Hotel Orphanage 179666.5 Figure 5. Feed cost Veterinary practices Type of veterinary practices Both western as well as indigenous veterinary practices are used. However, use of indigenous veterinary practices diminishing due to number of reasons such as, lack of Finding of indigenous veterinary doctors are difficulties of finding of these medicines, Western veterinary practices are popular among elephant owners. Recovery rate is high with veterinary practices and western veterinary practices are highly available throughout captive elephant conservation areas. People, who rear elephants in home gardens and temples and devalas, mainly use both veterinary practices due to low cost while orphanage totally depends on western veterinary practices (Figure 6). There are two veterinary doctors attached to Pinnawala elephant orphanage and they give special vaccines to the elephants. In addition to that keep litters in very clean. Regular check ups are practiced by the veterinary doctors in orphanage. 120 P ercen tag e b ased o n typ e o f co n servatio n 120 Percentage 100 80 60 40 W e stern 20 Ind e ge ne o us 0 H om e garden T em ple & devala H otel O rphanage B oth T yp e o f con servatio n Figure 6. Type of veterinary practices People, who rear elephants in home gardens and temples and devalas, mainly use both veterinary practices due to low cost while orphanage totally depends on western veterinary practices (Figure 6). There are two veterinary doctors attached to Pinnawala elephant orphanage and they give special vaccines to the elephants. In addition to that keep litters in very clean. Regular check ups are practiced by the veterinary doctors in orphanage. Veterinary cost Highest veterinary cost is spent in homegardens, because there are two elephants that are weak and old thus owners spend high cost to improve their health status (Figure 7). Orphanage also spends much cost on veterinary practices. Elephants in orphanage receive vermicides in every four-month and when they sick, doctors give penadols and one dosage include twelve penadols. People rear elephants in home garden for social prestige are wealthier people. Therefore, they spend much cost on veterinary practices. Hotels do not much concern on veterinary practices, because of their main purpose is to obtain income. Since elephants are large animals, providing of veterinary practices are cost effective, because owners have to give large dosage to recover elephants. Elephants are aggressive therefore providing veterinary practices incorporate with danger; as a result, owners have to spend much on the doctor’s fee. Veterinary cost 20000 28052 Hom e garden Tem ple & devala Hotel Orphanage 13333.5 17562.5 Figure 7. Veterinary cost Bathing facilities 121 Elephants mainly bathe in rivers. Since elephants are large animals, they need large water bodies for bath (Figure 8). Most of the captive elephant rearing areas belong to wet zone in Sri Lanka. As a result, water is available throughout year in those water sources. Difficulties in finding water sources are not a main problem among captive elephant owners. Rivers, lakes streams are found in those areas, such as Mahaweli River, Maha oya, etc. However finding of good quality water is difficult during rainy season is a problem among owners because of mixing of mud with water. Percentage based on type of conservation 120 Percentage 100 River 80 Stream 60 Lake 40 Other 20 0 Home garden Temple & devala Hotel Orphanage Type of conservation Figure 8. Bathing facilities Mahouts The role of mahouts is crucial in captive elephant conservation. In most cases, owners hire one or two mahouts to handle the elephants in captivity. Since the elephants are so aggressive during musth period two mahouts are used to handle them. Captivity elephants need to be chained; however this will make them aggressive towards mahouts. Home garden and temple and devala use two mahouts to control an elephant, while hotels and orphanage use only one mahout. Hotels and orphanage mainly concern about their profit. For that reason keeping two mahouts per elephant is an additional cost for them. As a result, they keep only one mahout to handle their elephants. However, some mahouts are coming from generation; therefore they possess enough skills to handle more than one elephants. In the orphanage, a mahout controls several captive elephants. Number of mahout per elephant, is an important issue to provide better care and attention for elephants. Wages of mahouts are important issue due to a number of reasons Low wages leads to low level of living of mahout population and it leads to less care to elephants by mahouts. On the other hand, high mahout wage rate will alter the profits of keeping elephants. The study revealed that mahouts serve in hotels and the orphanage, receive Rs.3500- 6000 wage per month (Figure 9). When elephants participate to processions, owners receive income and most of the time this income goes to the mahouts. Most of the mahouts in hotels receive around Rs.6000.00 per month. However, it was found that some temples and devalas do 122 not provide desired wage for mahouts, because these mahouts are coming from generations and temples and devalas give lands for those mahouts on behalf of wage. Instead they are asked to find work and earn money. Elephant owners give high wage rate to mahouts otherwise opportunity cost of being mahout is high. Wage rate is a sensitive factor, because most of the mahouts receive low income, which is not enough to cover cost of living. Consequently, they live below the poverty line. In addition to that social level of being mahout is low, so the perception management is needed to improve the social status of mahouts. This can be done by giving them a uniform and a higher salary or simply by changing the name of the job. Percentage based on type of conservation 120 Percentage 100 <3000 3000-3500 3500-6000 >6000 80 60 40 20 0 Home garden Temple & Hotel Orphanage devala Type of conservation Figure 9. Mahout wages Purpose of rearing captive elephants Mixed use strategies, allowing for one or more purposes will yield positive impact of domestication of elephants. So that most of the elephants rear in home garden (43%) for social prestige and they keep elephants as symbol of their wealth (Figure 10). In addition to that, some elephant owners (21.62%) keep elephants for logging and tourism activities. Most of devalas and temples rear elephants to use in pageants. Number of elephants in procession is declining with time. As a result attractiveness of that procession’s get reduces and most of the temples and devalas want to rear more number of elephants. The new trend of captive elephant rearing is, use these elephants for tourism activities. Setting up eco-tourism projects could be a source of income for farmers during the dry season. Hotels and orphanage earn more revenue directly and indirectly due to rear elephants. The elephants in orphanage are going to train for procession like cultural activities and they are very concerned about the well being of our animals and that is the reason why they will not sell elephants to the private owners. Tourism profits were not the main goal to start the safaris. The general idea was to safe the domestic elephants that had been used in the logging industry. In fact most elephant owners did not have enough money to take good care for their elephants. 123 Percentage based on type of conservation 80 70 Percentage 60 Tourism 50 Prestige 40 Ceremonies 30 Income 20 Other 10 0 Home garden Temple & devala Hotel Orphanage Type of conservation Figure 10. Purpose of rearing captive elephants Revenue and costs The economic justification for rearing captive elephants can be found by considering revenue and costs. In Orphanage and hotels revenue exceeds the costs thus hotels and Pinnawala elephant orphanage enjoy profits by rearing captive elephants (Figure 11). Orphanage possesses seventy captive elephants and it is the only place in the world where people can see such a large herd of captive elephants. As a result most of foreigners as well as local people visit the Pinnawala elephant orphanage to get an experience from this rare occasion and most of them prefer to see feeding on baby elephants and bathing of elephants. In orphanage huge amount of income comes from the entrance fee. The price of ticket of adult foreigner is Rs.500, child is Rs.250, local adult is Rs.50 and from local child is Rs.25. In order to bring video cameras they pay Rs.500-1500. During season, about 500 foreigners and about 500 local people visit this place daily. In addition to that earnings10% of the total annual income of the Maximus shop goes to Pinnawala. Medical costs, feeding and other management, without the salaries, is between Rs.50 and 75 thousand a day. Hotels organize some elephant related activities such as elephant rides, safaris, processions, mock war, etc and they pay Rs.400 from foreign and Rs.50 from local person. In addition to elephant related activities they practice some other income related activities which gain profit for the hotel. Temples and devalas received least income due to keep elephants. Most of the temples and devalas and some of people who rear elephants did not use elephants to earn profit. Because those people keep elephants for prestige and temples and devalas keep elephants to participate ceremonies. In addition to that, people participate their elephants to participate ceremonies, however do not obtain income from those activities. Also some people possess vehicles to transport their elephants. As a result, maintaining cost of elephants is high. 124 The temple of the tooth relic used to have 100 elephants. However they are becoming old (60% is older than 50). As a result temples need more elephants. Specially, during July to September most of temples in Sri Lanka held processions and elephants are used to decorate them. As a result during that period temples suffer lack of elephants. In Sri Lanka, company started called ‘Maximus’ produce recyclable paper using elephant dung. One elephant produce around 80kg dung a day and with 10kg dung can produce 120 sheets of rough paper equal to 720 A4 size sheets. Thereby it gives economic values to elephants dung and creates more awareness for elephant protection. In future they expect to collect elephant dung form areas where people suffer from human elephant conflict. This will create positive intention on elephant conservation among the people suffered human elephant conflict. The annual turnover of the Maximus is around US$ 250,000 and it will contribute to national income of the Sri Lanka. A captured elephants costs now around 2 million rupees and a tusker between 4 and 5 million. There is a huge demand for captured elephants and are very difficult to get, because the owners do not want to sell their elephants and it is not allowed by DWLC to capture more elephants from the wild. In 1985 the Captive Elephants Owners Association (CEOA) was founded and there has been a demonstration against the governmental policies to get more elephants to the private owners. The CEOA wanted the possibility for the owners to buy elephants from the Pinnawela elephant orphanage, captured from the wild, or otherwise imported elephants. Revenue and Cost 600000 500000 Rupees 400000 Average cost 300000 Average revenue 200000 100000 0 Home garden Temple & devala Hotel Orphanage Type of captive elephant conservation Figure 11. Revenue and costs 125 Importance in captive elephant conservation in Sri Lanka Table 1. Significance of captive elephants in Sri Lanka Keeping ƒ Economics Religion Urban Education Ecology Total system Owners Mahouts Social tourism forestry Temple 2 1 3 0 1 0 7 Homegarden 3 1 3 2 1 2 12 Hotel 3 1 1 3 1 1 10 Orphanage 3 2 0 3 3 0 11 0=not significant 1=low 2= moderate 3= relatively high A highest total mark is given for elephants keep in homegarden and it is followed by the elephant’s keep in Pinnawala elephant orphanage (Table 1). Elephants in home garden are mostly used for religious activities and also for logging, elephant safaris and rides like income generating activities. As a result, owner receives somewhat high income from the elephants. Most of the elephants in Pinnawala elephant orphanage frequently use for educational and research activities, income generating activities like tourism. In addition to that the elephant dung in orphanage is used as fertilizer for the coconut palms, banana and rubber trees. The orphanage possible attractions could be elephant watching, and specially on feeding of baby elephants and everyone wants to feed them. For entertainment, elephant watching works already very well. Elephants in hotels create lot of money for the owners; however they pay less for mahouts. Mainly these elephants are used for tourism activities and rarely use for educational and research activities. Temples and devalas use these elephants mainly for religious activities and rarely use for educational, tourism like activities. Temples and devalas do not use these elephants to generate income. Problems associated with captive elephant rearing Both elephant owners and mahouts face a number of problems, when keeping and handling elephants. When handle some elephants, there are some incidences that some the mahouts and indigenous veterinary doctors were killed when elephant becomes musth. During this time elephants damage to the properties of other people and owners have to bear financial compensation for damages. As a result of reduction of forest coverage, finding of feeds for elephants were much more difficult. Since elephant is a large mammal, should give large dosage of medicines compared with other animals which is much expensive for elephant owners they need large water bodies to bath. Therefore, finding of bathing place during drought period and rainy period is a much more difficult. Indigenous veterinary practices are shared by all among captive elephants, nevertheless finding of veterinary practices become difficult among elephant owners. Some mahouts who handle elephants in temple and devala do not receive wage from the owners consequently mahouts attempt to get more 126 works from the elephants. Some people give liqueur to mahouts and endeavor to get more work from elephants, which exceed their carrying capacity. This will result early death among captive elephants. Some mahouts put less care on elephants and mistreat the elephants that lead to early deaths among captive elephants. With the popularization machinery, the number of work available per elephant declined with time and owners receive lower income and keeping an elephant become much more difficult. Although Pinnawala elephant orphanage is a better place for elephants, number of mahouts in this place is very low and elephants do not receive required degree of care and attention from the mahouts. DWC wants to register every elephant owner for Rs.500; however owners will get nothing in return. For killing an elephants have to pay Rs.5000 and most of the time murders are set free without charge. As a result, elephant owners do not register in DWC. CONCLUSION This study aimed at investigating the economic and social aspects of captive elephant rearing as measures of conservations of elephants in Sri Lanka. The following conclusions can be generated according to the results of the study. Rearing of elephants in homegarden and temple and devala, face a number of problems such as finding of feeds, finding of mahouts, high maintenance cost, difficulties in finding of water during drought and difficulties in finding good quality water during rain and mahouts are less care and mistreat the elephants, etc. Nevertheless, they prefer to rear one more elephants with the intention of social prestige and to use for ceremonies. As stated in results and discussion, still there are ample amounts of feeds available throughout the captive elephant conservation areas. The average income of mahout range between Rs. 3000.00 – 7000.00 which is not adequate. Hotels and orphanage enjoying profit. The annual profit of the orphanage is about 5 million, while hotels gain Rs. 80,000.00 annually from an elephant. Orphanage spends Rs.3, 64,285.00 annually for feeds of an elephant and spends highest feed cost per elephant. People who rear elephants in homegarden spend remarkably highest cost on veterinary practices of an elephant (Rs.28, 052) As indicated in problem justification, captive elephant conservation methods should be economically sound However with the mechanization, the elephant gets less draft work thus captive elephants become wild elephants. Consequently, since hotels enjoying with profit, can recommend rearing elephants in hotels, which has high demand for elephant safaris, rides, etc. The captive elephants in orphanage are not tamed. As a result, people unable to move closer to elephants. Therefore, tamed captive elephants can be used establish an elephant park, which is economically sound. A park could create awareness of both tourists and local community, creating income possibilies for community, mitigating human elephant conflicts in the short term and contribute to the conservation. Establishing an elephant park is win win situation for both elephant owners and people who take pleasure from elephant related activities. 127 ACKNOWLEDGEMENT I express my sincere gratitude and appreciation to my supervisor Dr. L.H.P.Gunaratne, Senior Lecturer, Department of Agricultural Economics and Business Management, Faculty of Agriculture, University of Peradeniya, for his valuable advises, encouraging guidance and persistent throughout this study. REFERENCES Chadwick, D.H. (1991). Elephants –Out of time and space: National Geographic. Quoted in T.N.Samarasinghe. An economic assessment of damage caused by the wild elephants in villages around Ritigala strict nature reserves. B.Sc.thesis, University of Peradeniya, 2000. Eltringham, S.K.(1992). The Return of the illustrated Encyclopedia of elephants. Tiger paper. Quoted in T.N.Samarasinghe. An economic assessment of damage caused by the wild elephants in villages around Ritigala strict nature reserves. B.Sc.thesis, University of Peradeniya, 2000. Gokula, V. and Vardharajan, M. (1996). Status of temple elephant management in Tamil Nadu, Southern India.Gajah15:37-40. Jayewardene, J. (1998). Captive elephants in Sri Lanka: status, distribution and numbers. Nature 4:13-18. Kurt, F. and Maze, K.U. 2003.Guidlines for the management of captive Asian elephants and the possible role of the IUCN/SSC Asian Elephant Specialist Group. Gajah22:30-41. Lair,R.C. (1997). The care and management of the Asian elephant in domesticity. Rome: FAO, 1997. Thailand. Williams, A.C. and Jahnsingh, A.J.T. 1996. Elephant capture in Meghalaya, Northeast India- the past and the future. Gajah17:1-5. The care and management of domesticated Asian elephants in Sri Lanka[on line]. Thailand: Jayantha Jayawrdena, 2001[cited 16 December 2004]. Available from internet http://www.fao.org/DOCREP/005/AD031E/ado31eov.htm 128 ECONOMIC POTENTIAL OF THE TOURISM INDUSTRY TO CONSERVE THE ENDANGERED ELEPHANTS IN SRI LANKA K.A.D.K.S.D. BANDARA* and L.H.P. GUNARATNE Faculty of Agriculture, University of Peradeniya ABSTRACT The Sri Lankan elephant population shows a marked reduction during last five decades, mainly due to destruction of elephant habitats by development projects and illegal encroachments. With the limitations of the lands available for elephants and escalating human-elephant conflicts, domestication of the elephants appear to one of the viable way of conserving elephants. However, with the low demand for draft, domestication of elephants should be targeted on more economically attractive options such as tourism. With this background, the economic potential of utilization of elephants in tourism industry is investigated in this study. The degree of interest on elephant related activities by the tourists, and the status of involvement of tourist hotels and guest houses in those activities are also evaluated. The potential of generating revenue via the integration of the elephants to the sector was also considered. The necessary information for the study was collected through two surveys carried out simultaneously during the period of January to February 2005. One survey was carried out targeting the tourists that come to see the elephants, based on a structured questionnaire prepared in English, French and Japanese. The random intercept sampling method was used. Meantime, a mailed survey was executed focusing the tourist hotels and guest houses. Tourist hotels, guest houses within 40-km boundary of parks, protected areas and orphanage were selected to the sample. The descriptive statistical analysis revealed that, about 8% of the sampled tourists have selected the Sri Lanka as their tours destination mainly to see the elephants while around 20% tourists arrive to Sri Lanka with an intention of watching elephants. The study revealed that more than 80% tourists like to experience the elephant rides whereas more than 60% tourists prefer experience on the elephant safaris. The average willingness to pay for these activities are 20 US$ and 28 US$ respectively. The survey carried out with the managers shows that more than 70 % of hotels and accommodations that are located vicinity of the parks have involved in any kind of elephant related activities as keeping elephants for tourism, promotion of elephant related activities and facilitating elephant watching. Study found that there is a substantial potential in exploit elephants in tourism industry, which will generate more income to the elephant owners thus domestication can be used as one of the means of conservation of endangered elephants in Sri Lanka. 129 INTRODUCTION The Elephas maximus maximus in Sri Lanka is the type specimen of Linnaeus (Crusz, 1986). Shotake et al. (1986) showed that it is genetically quite distinct from the Indian sub species Elephas maximus ibengalensis. The elephants have been influencing Sri Lankan socio-cultural environment since the ancient time. There is an enormous cultural and religious significance of elephant to the both Buddhists and Hindus in Sri Lanka. Status of elephants in Sri Lanka The Sri Lankan elephant population has undergone a marked reduction from the mid nineteen century (Santiapillai and Jackson, 1990; De Silva, 1998). Fragmentation and loss of habitats are the factors contributing to this decline and this lead to human elephant conflict (Desai, 1998). Also this situation largely results from the ad-hoc development projects carried out during the last five decades (Weerakoon, 1999) and is exacerbated by the lack of coordination between different government departments and wildlife authorities. Moreover poor integration of economic aspects and lack of attention to public preferences for elephant conservation compound the problem. The elephant has been protected in Sri Lanka since 12th century A.D. (Wicramasinghe, 1928). Nevertheless, large numbers were captured to be used as war elephants or for export to other countries. The systematic slaughter of elephants in general and of tuskers in particular began with the arrival of the colonial powers and the introduction of firearms. The situation was exploited to such an extent that a government ordinance in 1891 banned the “wanton destruction” of elephants (Olivier, 1978). The animal was given full legal protection in 1937. But today only about six percent of the males have tusks in Sri Lanka, in contrast to southern India, where 90% of the bulls are tuskers (Sukumar, 1986). If this trend exists the Sri Lankan elephant population will extinct from earth for ever within one or two decades. Conservation efforts At present, there is a growing concern over conservation of Sri Lankan elephant by various government organizations and non-government organizations. Some of these measures adopted are: establishment of new national parks and protected areas, establishment of elephant corridors, habitat enrichment, translocation of elephants and elephant drives, electric fencing and promoting exsitu conservation. The decline of the Sri Lankan elephant population has been largely results of lacking socio-economic and political considerations. Therefore it is important that its management and conservation should take in to account, the social and economic value of its existence. According to Santiapillai. (1997) and Dublin,(1997) the key to finding out a long term solution to the human elephant conflict in Sri Lanka is two fold. To 130 encourage the adoption land-use strategies (such as selective timber extraction, controlled livestock grazing) that are compatible with elephant conservation, and where elephants and humans do overlap, that people derive tangible benefits from their presence (such as eco-tourism). This study has focused towards the assessing of potential economic benefits that can be taken via tourism industry to conserve the endangered elephant in Sri Lanka. Moreover, whether conservation is capable with the incorporation of elephants to the tourism industry. Tourism Tourism is one of the world largest and fast growing industries. According to the recent statistics tourism provide 10% of the world’s income and employs one tenth of the world work force. It contributes 6% of the global GNP and 13% of consumptive expenditure. In Sri Lanka, tourism has become a very important factor in foreign exchange earning. Gross earnings from the tourist industry increased by 39 percent in 2003, from US dollars 248 million in 2002 to US dollars 324 million in 2003. (Central Bank of Sri Lanka, 2003).Tourism has been significant in the case of many aspects, a very tangible element which can use for conservational activities. Tourism play a key role in economic and social benefits like attracting foreign currency, more funds for the conservation, improve world understanding .etc. Huge number of tourists arrives per year to Sri Lanka. Table 1 show the number of tourists arrived in last two years. Table 1. Tourists arrival by region Tourism industry involves various stakeholders in addition to the tourists such as accommodation establishments and restaurants owners, travel agents, tour operators. Table 2 shows the income distribution among various stakeholders. Table 2. Revenue from the tourism- 1993 to 2002 (In Rs. Million) Source: CTB publication, 2003 Source of Revenue Hotels & Restaurants Travel Agencies Shops 1995 79.5 28.0 10.8 1996 59.6 26.4 9.3 1997 116.0 39.1 12.1 131 1998 143.6 58.8 14.4 1999 164.5 67.3 16.5 2000 162.7 61.7 15.1 2001 128.4 52.6 13.9 2002 149.7 64.5 14.8 Embarkation Tax Cultural Triangle Botanical Gardens Zoological Gardens National Parks Museums TOTAL 201.6 168.7 25.2 11.4 38.5 1.2 564.9 151.1 121.2 16.8 14.4 27.8 0.6 427.2 183.1 186.0 21.8 15.1 34.5 0.7 608.4 190.5 225.0 26.7 18.9 41.2 0.1 719.8 218.2 300.5 32.6 23.1 56.0 0.9 879.6 200.2 276.0 29.4 29.2 60.2 0.5 835.0 269.3 222.0 21.2 22.3 54.4 0.6 784.7 393.2 242.8 26.3 30.2 58.0 0.8 977.8 This study is mainly focused towards assessing the potential contribution of accommodation establishments (tourist hotels and guest houses) and restaurants in addition to the tourists. The hotels and travel agencies earn large portion of revenue as shown in table 3. Table 3. Foreign exchange earnings form tourism -2001 & 2002 (In Rs mill) Source 2001 * 2002 ** Banks Travel Agencies Shops Hotels Others TOTAL 6,718.1 4,026.0 4,379.4 2,614.4 1,125.4 18,863.3 8,523.3 5,146.5 5,615.2 3,386.4 1,530.6 24,202.0 * Revised ** Provisional Tourism as a mean of elephant conservation The Sri Lankan elephant population is being declined in a crucial rate. Therefore, conserving of this unique sub species of Asian elephant is of paramount importance in substantiation of economical, social and cultural benefits due to its existence in the long term. Domestication of elephants especially to the tourism industry is considered as a one of the solution to conserve elephants, which again promote the long term coexistence of human and elephants. Per year about 500,000 of tourists arrive to Sri Lanka and many of tourists visit wildlife parks to observe the elephants in the wild. Also the captive elephants in tourist hotels used for tourists attractions as elephant rides thus generating huge amount of money. The accommodation establishments situated near the park generate revenue via supplying foods, lodging and other facilities to the tourists who come visit the parks. In this sense, tourism can perform a major role in conserving elephants. Though tourism has that much significance for elephant’s conservation, its potential should be examined economically. Therefore, investigation of economic potential of tourism industry to conserve the endangered elephants is very important as a prior task. The main objective of this study is to evaluate the economic potential of tourism industry for using in elephant conservation in Sri Lanka. The willing to pay for various elephant related experiences by tourists and those factors influencing their willingness to pay is also found. In addition, the ways and means by which tourist hotel owners contribute to conserve the elephants is also investigated. 132 METHODOLOGY Both primary data and secondary data were used in this study. Primary data was collected during the period of January 2005 to March 2005. Adopted methodologies for the process of data collection were mainly questionnaire surveys, literature review and informal discussions with key informants. Relevant articles in journals and magazines as well as the research publications were reviewed to obtain additional information. Data collection Two parallel surveys were conducted incorporating the tourists and managers/owners of tourist hotels, guest houses and accommodations. The field survey of tourists The relevant data to the study were collected by interviewing the tourists using a structured questionnaire. The interviewing of tourists was done so as to represent a group of tourists by several individuals of that group. The questionnaire was comprised of three sections. They were arranged to get the personal information, general information of visit to Sri Lanka and their responses on experienced elephant related activities with preferences. The questionnaire was prepared in English and translated into French and Japanese in order to collect data from the non-English speaking tourists. Total of 120 questionnaires were used for the survey and 70 of them were in English. The number of French and Japanese were 30 and 20 respectively. Pre testing was carried out prior to commence of the field survey. The target population was the tourists who visit the nature sites. Therefore a sample of tourists was interviewed using the questionnaires at Uda Walawa, Wasgamuwa, Minneriya, Bundala, Yala and Pinnawala where major portion of tourists come to see the elephants. Mail survey of managers of the hotels, guest houses and accommodations In this part of the study, it is assumed that the tourists who are interested in elephants arrive the areas closer to the wildlife parks and the orphanage. Therefore, initially the parks and protected areas where the elephants are present were found. Then the area within a boundary of 40 km around the parks and protected areas were chosen was for the study. The hotels, guest houses and accommodations were selected using the list of “Accommodation Guide” published by the Ceylon Tourist Board which was updated in November-December, 2004. The relevant data to the study from managers of the relevant establishments was collected using a structured questionnaire. The questionnaires were mailed with the cover letters to the managers of the hotels, guest houses and accommodations and follow up telephone calls were given, increase the response rate. Meantime, some of the establishment were visited and directly interviewed the managers. The questionnaire was prepared in English which was comprised of three sections, section 1 was prepared to get the general information of the 133 establishments. The section 2 was included to get the elephant related activities, which they have involved. The section 3 was used to the information about their involvement to conserve endangered elephants. The necessary secondary information was collected from the publications of several institutions and their web sites such as Ceylon Tourist Board, Sri Lanka Hotel Corporation, Institute of Policy Studies and Central Bank. Some information as list of tourist hotels, guest houses and accommodations, number of tourists arrived in past few years, revenue earn by the tourism industry in relation to each year national income, amount of revenue earn by each stakeholder in the tourism sector were collected from these sources. Analytical framework Most of the data were collected through the survey of tourists and managers of accommodation establishments, which can use to assess the interest on elephants as well as elephant related activities. Therefore both descriptive data analysis and regression analysis were employed. Assuming a linear relationship between willingness to pay for the elephant related activities and other independent variables, the regression model was fitted to estimate the relationship between selected attributes and the willingness to pay amount. Conceptual model WTP= ƒ (NAT, AGE, DAYS, INC, EDU, SEX, EXP, ACT, INV) WTP=Willingness to pay for the elephant related activities (Rs) Variable DAYS=Days stay in Sri Lanka AGE=Age (Years) EDU=Educational level INV=Whether experienced such type of activities NAT=Nationality SEX=Sex EXP=Expected daily Expenditure in Sri Lanka INC=Annual income ACT=Number of elephant related activities like to experience Hypothesis Expected relationship is the tourists who stay longer period experienced such activities and willing to pay more. Young people are more interested on elephants and elephants related activities. Therefore they like to pay more. Educated people are expected to be more towards enjoying the elephants related activities which are nature based activities. Experience tourists expected to be spend more on that activities. Developed nations have already deteriorated their nature and biodiversity when achieve the development. Therefore they like to pay more for these activities. Per capita income of each country used as an indicator of level of development Males are more prefer that activities. High expending people expected to be having high willingness to pay. Higher income visitor are expected to be spend more money. More interested tourists on like to experience more number of activities. 134 RESULTS AND DISCUSSION Visitor characteristics Based on the sample survey, it was found that the average age of the tourists who came to see the elephants is 56 years with the range of 21 to 77 years. The average number of days stayed in Sri Lanka is 15 days, with a maximum of 30 days and a minimum of 6 days. The respondents’ average annual income of the sampled tourists is 36,954 US $. The random intercept survey is comprised of 16 different nations. Their distribution is given in the table 4. Table 4. Nationality distribution of the sample Number of Tourists Nationality % British 36 36 % Irish 4 4% Australian 2 2% Finland 1 1% Danish 1 1% Italian 1 1% Canadian 2 2% Dutch 4 4% Swiss 3 3% American 7 7% German 8 8% New Zealand 4 4% Poland 4 4% Franchise 9 9% Swedish 5 5% Japanese 7 7% Chinese 2 2% n = 100 135 Status of education Most of the tourists come to see the elephants are mid-career professionals who hold a degree and more than 75 % belong to those categories. More than 18% of visitors have completed secondary education. Almost all the respondent has at least primary education. It was noted that the tourists who have some kind of understanding and interest about environment visit these areas to see elephants. Education Level Distribution : Tourists Sample No formal education No primary edu. 0% 0% No formal edu. 1% 6% 18% 48% 27% Followed primary education Completed primary education Followed secondary education Completed secondary education Followed graduate studies A graduate degree Figure 1. Status of education in tourist’s sample Reason for selecting Sri Lanka as a travel destination The question 1 of section B in the questionnaire of tourists was directed to get the most important reason when selecting Sri Lanka as a travel destination. Table 5 shows the details of the tourist’s responses. Table 5. Most important reason for selecting Sri Lanka as a travel destination Visiting friends and/or relatives See the Sri Lankan elephants Business/Convention Sun and beaches Archaeological sites Cultural/Native history Nature history Sight seeing 10% 19% 10% 16% 3% 6% 7% 28% 100% 136 n = 87 According to the responses, a major portion of tourists (28 %) indicated sight seeing as the most important reason when selecting Sri Lanka as a travel destination. In addition to that seeing elephants in Sri Lanka has a quite significant. Because 19 % of tourists have chosen Sri Lanka as a travel destination due to presence of elephants. Degree of Influence, presence of elephants in Sri Lanka to choose as a travel destination Figure 2 shows the degree of importance of Sri Lankan elephants to the tourists when selecting Sri Lanka as their travel destination. Influence of Sri Lankan Elephants to Choose Sri Lanka as a Traval Destination Not important 21% Main reason 8% Main reason Important Important 46% Somewhat important 25% Somew hat important Not important Figure 2. Degree of influence elephants to choose Sri Lanka as travel This shows that main reason of 8 % of sample tourists is seeing the Sri Lankan elephants in journey to Sri Lanka, according to their responses. They might be the researchers or eco-tourists. Tourist’s involvement of nature related activities Table 6. Involvement of nature related activities by the tourists Activity Number % Jungle excursion Visiting cultural sites Bird watching Wildlife Botany Fishing Visiting indigenous communities Others 44 64 29 45 31 4 7 6 48.4 70.3 31.9 49.5 34.1 4.4 7.7 6.6 n=91 As shown in the table 6 most of the tourists arrive to Sri Lanka had opportunity to visit the cultural sites (70 %) also around 50 % of tourists involve in wildlife related activities as elephant safaris, deer safari etc. 137 Experiences of elephant related activities: tourists According to the responses of tourists, more than 80 % have an interest on experiencing the elephant rides. Also more than 60 % tourists like to experience the elephant safaris. The portion of tourists who have interest on elephant circuses, races etc are very low. In addition to this some of the tourists like to participate activities such as visit orphanage, looking at bathing of elephants, visit the temples which keeping elephants, hiking in the natural habitat of elephants. This portion of tourist is 11 %. The categorized responses are shown in table 7. Table 7. Experiences of elephant related activities by tourists & their willingness to pay for them Percentage of tourists like to experiences Average willingness to pay in Rs Elephant ride Elephant safaris Elephant circuses Cultural events which use elephants 82% 66% 4% 7% 5% 11 % 1942 2831 1341 1241 1921 1297 (20 US $) (28 US $) (13 US $) (12 US $) (19 US $) Elephant races Others Willingness to pay for the elephant related activities: tourists Survey respondents were asked to mention their willingness to pay (WTP) for the each elephants related activities they would like to experience. According to the responses of tourists, indicate in the table 7, average willingness to pay for the elephant ride is 20 US $ with minimum of 8 US $ and maximum of 51 US $. Also the amount willingness to pay for the elephants safaris is 28 US $ with minimum of 6 US $ and maximum of 50 US $. Willingness to pay amount reflects the tourist’s valuation of the each elephant related activity. The amount of willingness to pay can be utilized to get an idea about the degree of significance each activity and potential of getting benefits economically through the tourism industry via the incorporation of elephant related activities to the schedules of tourist packages. Involvement of elephants related activities by tourist’s hotel, guest houses and accommodations. Nature related activities: tourist hotels, guest houses and accommodations According to the responses of managers of the tourist hotels, guest houses and accommodations the percentage of establishments that has involved in each activity is indicated in the table 8. Table 8. Nature related activities Practice by the tourist hotels, Guest houses & accommodations 138 Nature related activity % of hotels & guest houses Jungle excursions 40 % Visiting cultural sites 65 % Bird watching 60 % Wildlife 70 % Visiting botanical gardens 15 % Fishing 15 % Visiting indigenous communities 5% Others 10 % n=32 As shown in that table 70 % of the ventures in the sample have involved in wildlife related activities. Among the establishments that has engaged in wild life related activities for earning revenue, a high number of the tourists hotel, guest houses and accommodations have involve in elephant related activities. (Figure 3) Wildlife Related Activities: Tourist hotels & Guesthouses % of Hotels & Guesthouses Involved 80 71 % 65 % 70 60 47 % 50 40 30 20 29 % 29 % 12 % 10 0 Others Deer safarisBird watching Touractivities to game sanctuaries Elephant related Tours to national zoological garden Figure 3. Wildlife Related Activities Involve by the Tourist Hotels & Guest houses for the Tourism Elephant related activities of the tourist hotels, guest houses and accommodations The involvement of these establishments on elephant’s related activities can be categorized in to 3 types as shown in the table 9. Table 9. Elephant related activities involved by hotels & guest houses 139 a) Keeping Elephants for Tourism Activities (Exhibitions, circuses, riding, carry tourists …etc.) b) Promote Elephant Related Activities for Tourism (Safaris, tours, seeing perahara… etc) c) Facilitate Elephant Watching (Situated near parks, orphanages… etc) 21 % 79 % 36 % According to that more than 75 % of enterprises have involve in promoting elephant related activities as safaris, arrange tours to orphanage and protected areas…etc. Only 21 % hotels and guest houses keep elephants for tourism activities. Also more than 35 % of accommodation establishments situated near the wildlife parks and orphanage, which facilitate watching the elephants. That kind of enterprises earns revenue by supplying accommodations and foods for the tourists. The hotels and accommodations keeping elephants have involved in various activities. Those activities can be categorized as in the table 10. Table 10. Various uses of elephant that are taken by the hotels & guesthouses, which are keeping the captive elephants % of Hotels & Guest houses 40% 60% 0% 40% Activity of Elephants use As an exhibit for tourists Use for elephant rides Use for elephant circuses Use for cultural activities Give for cultural activities in other 40% places 20% Use to get some works done Conservation activities: tourist hotels and guesthouses According to the responses of the managers, most of the tourist hotels and accommodations have involved in elephant conservation activities. That activities and the level of involvement are summarized in the table 11. This indicates that there is a potential to get participated the organizations that has not engaged in that kind of activities so far. Table 11. Participation in conservation activities of elephants by Hotels & guest houses Conservation activity Contribute to trust fund support for conservation Aid for the orphanage Help for conservation campaigns Others % of Hotels & Guest houses 67% 13% 23% 22% 140 Estimates of the regression model The factors influencing the expenditure by tourists for various elephant related activities were identified. The estimates of the multiple regression models are given in Table12. WTP= NAT +AGE + DAYS + INC + EDU + SEX + EXP + INV Table 12. Estimates of the OLS regression model Variable Constant Nationality Age No. of days stay Annual income Education Sex Expected expenditure Number of activities like to experience Experienced with similar activities Coefficient Std. Error 13981 - 0.16 - 85.2 0.003 - 0.005 0.61 0.06 0.25 17833 39.45 0.224 0.141 1356 3883 3309 - 0.008 9164 - 0.12 5317 2 Adjusted R = 0.52 The model was free from perfect mulicollinearity with reasonably satisfactory fit (adjusted R2). Among the considered variables, nationality, days spend in the Sri Lanka, whether the tourists have experienced such activities and level of education were significant at 5 % level. National income was used as a proxy for the variable, nationality so that contrary to the expectation, tourist with high income countries are less interested in spending money on elephant reacted activities. However, tourists who had plans to stay more were found to be more enthusiastic in spending on similar activities. Also it was found that young are more willing to pay for these activities. 141 CONCLUSION The study aimed at getting an insight into the make use of elephants for tourism and identifying the future potential of this so as to promote tourism as an alternative mean of elephant conservation in Sri Lanka. The study was carried out with tourists who visit the national parks and the owners/managers of guest houses/hotels located closer to these sites. This study found that there is a substantial potential of incorporating elephants into the tourism industry, which will foster the coexistence of human and elephants as in the past. This study revealed that there is a huge demand for the elephant related activities more than identified so far, such as elephant rides and safaris. The present rates charged for elephant related activities are much less than their willingness to pay for those activities. According to the findings, an average willingness to pay for the safaris and rides are approximately 20 and 28 US $, respectively. This shows the potential revenue that can earn through these activities. It may cost 20,000-25,000 rupees per month to maintain a captive elephant. Given the substantial cost of maintaining a captive elephant and less demand for draft work, there is a need to find better alternatives; the study indicates that much more amount of revenue can be generated per month when the elephants are used for the tourism activities under well managed situation. The recent studies carried out by the Institute of Policy Studies has identified that the revenue from the protected areas could be increased significantly if the Ceylon Tourist Board (CTB), Department of Wildlife Conservation and the Forest Department assist the private sector to promote “value added” activities around protected areas. This would include hotels and tourist attractions such as elephant safaris. The findings of this study corroborate this fact by identifying the potentials in a scientific manner. The nature tourist market is a particularly attractive niche market as there is some evidence that nature tourist have higher incomes than the average tourist, stay longer and spend more on locally produced products and services (Vidanage, 1995). With the problems of the tourist industry due to the war, nature tourism represents a niche marketing, which can command a premium. Many private tour operators feel that nature tourism has considerable growth potential. Promotion of nature tourism would also accord with the three main objectives of the tourism Master Plan prepared by the Ceylon Tourist Board (CTB) of moving away from low budget mass beach tourism by selecting activities that upgrade existing attractions and product, diversify the product mix and its capacity and develop new circuits and product packages inland, including new tourist areas. The elephant related activities are in the category of existing tourists attractions. Therefore, the value of them in the sense nature related activities can be identified from this study and it is important to take an active role to promote elephant related activities with a good product mix to attract among the tourist who are the potential customers of them. This requires the Ceylon Tourist Board taking an active role in promoting elephant related activities. On the other hand, with the limited carrying capacity of the protected areas and growth of the rural populations, the human elephant conflict will escalate in the future, so that there is a growing need to find feasible solutions. Therefore, use of elephants in the tourism industry by capturing and taming of young elephants can be considered as one of the counteractive measure. Then a mechanism could 142 bedeveloped to transfer a part of the revenue generated from such activities to compensate the farmers to mitigate human elephant conflict, as a mean of conservation. REFERENCES Central bank of Sri Lanka. 2003. Annual report. Colombo. Central Bank. Daniel, J.C., 1993. The Asian elephant: future prospects. Gajah11:02-15. De Silva, M., 1998.Status and conservation of the elephant and the alleviation of man-elephant conflict in Sri Lanka.Gajah 19:01-25. Gujarati, D.N. 1995. Basic Econometrics, 3rd ed. McGrow Hill Inc, New York. Gunatilaka, H.M. 2003. Environmental valuation; Theory and Application, PGIA. University of Peradeniya. Handavitharana, W., Dissanayaka, S., & Santiapillai, C. 1994. The survey of elephants in Sri Lanka. Gujah. 12:1-13. Jayewardene, J. 2001. Biodiversity & elephant conservation trust Gujah 20:73-77. Krishantha, W.P. 2001.Comparative analysis of nature tourism and conventional tourism in Sri Lanka. Project report. (B.Sc.). University of Peradeniya. Lair,R.C. 1997. The Care and management of the Asian elephant in domesticity. Rome: FAO, 1997. Thailand. Samarasinghe, T.N. 2000. An economic assessment of damage caused by the wild elephants in villages around Ritigala strict nature reserves. Project report. (B.Sc.)., University of Peradeniya. Steel, P. and Siva Kumar, M.. 1998. A strategy for nature tourism management in Sri Lanka. Colombo. Institute of Policy Studies. Tourism and the Economy of Sri Lanka. Economic Review (people’s Bank), 1994. 20(5):2-13. Vidanage, S.P.1995. Potentials and factors affecting ecotourism in Sri Lanka., Practicum report. (M.Sc.).PGIA. University of Peradeniya. Wicramasinghe, W.M.D.T.2002. Promotion of ecotourism in Upper Hantana mountain range., Kandy., Practicum report.(M.Sc.).PGIA. U 143 STRUCTURE AND COMPOSITION OF VEGETATION IN THE IFS -POPHAM ARBORETUM, DAMBULLA M.A.A.B. DILHAN*, T.D. WEERASINGHE and J. AMARASINGHE IFS - Popham Arboretum, Dambulla ABSTRACT The Dambulla Arboretum is a unique place, where a wasteland has been converted into a sanctuary of tropical trees by using a simple silvicultural method. The Arboretum consists of 3.6 ha arboretum along with a 10.8 ha woodland. The goal of the present survey was to document the abundance of dry zone key plant species and to update the checklist of plants in terms of seasons. The abundance of plant species, which were naturally regenerated, was enumerated in six 20 m × 20 m experimental plots. The girth at breast height (gbh) of individuals over 10 cm was measured within main plots, while individuals below 10 cm gbh were counted in 20 m ×10 m sub plots within the main plots. Five random plots 2 m × 2 m quadrates were used to measure the cover value of ground vegetation. In addition, annuals and biannual during the wet and dry season and undocumented perennials were encountered in an ad hoc manner addition to the experimental plots. A total of 101 plant species belonging to 91 genera and 42 families were identified. Among the taxa identified 45 had medicinal value, 14 timber value and 7 both timber and medicinal value. There were 45 tree spp., 27 shrub spp., 18 herbaceous spp. and 11 climber spp. Of them five were endemic viz., Diospyros oppositifolia, Diplodiscus verrucosus, Canarium zeylanicum, Cassine balae and Micromelum minutum. The leading dominant species and families based on important value index (IVI) were Syzygium cumini and Myrtaceae (IVI = 29 & 32), Grewia damine and Tiliaceae (IVI = 23 & 28) and Memecylon umbellatum and Rubiaceae (IVI = 16 & 25) respectively. In the ordination diagram, the plots in the arboretum and woodland are clearly separated by four distinguishable clusters and were assigned into plant communities. The species associated with cluster A were mainly trees viz., Diplodiscus verrucosus, Diospyros ferrea, Lepisanthes tetraphylla and Sapindus emarginata along with a shrub. While Cluster C resulted in five tree species. The survey has shown that the arboretum provides refugee to many valuable timber trees, including endemic and endangered plant species. INTRODUCTION The dry zone forests of Sri Lanka have experienced a boost in large-scale depletion of forest land by slash and burn agriculture, illegal felling, fire, irrigation schemes and agricultural expansion resulting from a growing population (Samarasinghe, 1995). As a remedy reforestation and conservation have been carried out by the Forest Department on degraded lands of the dry zone. Mr. F. H. Popham introduced a sustainable silvicultural method namely the Popham method (Popham, 1993), which facilitates the natural regeneration of indigenous species. This leads to succession towards woodlands dominated by small trees and bushes managing the growth of creepers and thorny shrubs. The IFS - Popham Arboretum in Dambulla is a typical example of natural regenerated forest. It has also contributed to rich biodiversity comprising 192 plant species (Cramer, 1993), 72 birds and 35 butterflies (Arboretum Newsletter, 1996). Research into the regeneration dynamics of silvicultural assisted scrub vegetation (Samarasinghe, 1995) in relation to different burning histories and regeneration strategies of selected forest species (Weerawardane, 1999) has been carried out during the past decade. The goals of this study are to document plant 144 diversity of silviculturally treated vegetation in the IFS-Popham Arboretum and to update the checklist of plant species documented by Cramer (1993). MATERIALS AND METHODS The IFS - Popham Arboretum is located close to Dambulla, 2.9 km away from it on the Kandalama Dambulla road (Figure 1). The total extent of the IFS Popham arboretum, is 14.4 ha comprising 3.6 ha in the arboretum and 10.8 ha of woodland (70 51’ 34" N and 800 40’ 28" E). AR1 AR3 AR2 WL1 WL2 WL3 Figure 1. Location of the IFS - Popham Arboretum. Colored circles indicate the sampling Plots. Plots in the arboretum = AR; and woodland = WL. The permanent plot method was used for sampling in this study. Six plots, each 20 m × 20 m were demarcated within the woodland and arboretum. Three plots each from the woodland and arboretum were selected and each plot was divided into 20 m × 10 m sub plots to facilitate sampling. Individuals over 10 cm girth at breast height (1.3 m above the ground) in the main plots (20 m × 20 m) were enumerated and were marked with numbered aluminum tags. Sub plots of size 20 m × 10 m were demarcated within the main plot to enumerate individuals 145 below 10 cm gbh and above 1 m in height. Individuals less than 1 m in height were enumerated by 2 m × 2 m quadrate sampling. A number of characteristics of the vegetation, viz., density and gbh of all woody species were recorded. The Importance Value Indices (IVI) of all the species were calculated using relative basal area and relative density for individuals over 1 m in height. Relative cover value was used for individuals less than 1 m in height. The flora of the arboretum documented by Cramer (1993) was revised by collecting plants in an ad hoc manner representing all vegetation types and microhabitats during the dry and wet season as shown in appendix 1. Herbarium specimens were prepared from the plant material collected in the field. Each individual sampled was identified as to its species as far as possible with help from the National Herbarium, Peradeniya, use of the Hand Book to the Flora of Ceylon by Dassanayake and Fosberg (1990-1991), Dassanayake et al. (1994-1995) and Dassanayake and Clayton (1996-2000). An ordination of the vegetation gradients in the dataset that could be compared against their spatial distribution on the ground was carried out using multivariate analysis (Detrended Correspondence Analysis) with the help of PCORD4 software. RESULTS STRUCTURE OF VEGETATION The density, over 1 m height and basal area greater than 10 cm gbh, of trees and shrubs in the arboretum are shown in Table 1. The total density of individuals was 2208/ha and the basal area was 93.2 m2/ha. However, the density and basal area of twenty leading tree and shrub species were 1795/ha and 74.8 m2/ha respectively. The highest density of trees was recorded in Grewia damine and followed by Diospyros ferrea, Pterospermum suberifolium and Chloroxylon swietenia etc. The highest contribution of basal area was recorded of Syzygium cumini followed by G. Damine and Mitragyna parvifolia etc. Memecylon umbellatum showed the highest density and basal area and Phyllanthus polyphyllus and Allophylus serratus followed respectively. 146 Table 1. List of first ten leading tree and shrub species represented in the vegetation size class above 10 cm gbh with their basal areas, density and IVI values. Species IVI Value BA m2/ha Tree species Syzygium cumini Grewia damine Chloroxylon swietenia Mitragyna parvifolia Pterospermum suberifolium Diospyros ferrea Bauhinia racemosa Diplodiscus verrucosus Lepisanthes tetraphylla Pleurostylia opposita Total 29 23 14 11 9 8 7 6 5 5 24.90 ± 4.36 13.11 ± 0.23 8.30 ± 0.34 9.14 ± 3.31 3.44 ± 0.17 2.20 ± 0.09 4.05 ± 0.31 2.20 ± 0.13 1.41 ± 0.05 0.70 ± 0.05 69.45 ± 2.65 46 ± 0.51 192 ± 0.20 113 ± 0.81 25 ± 0.03 117 ± 1.28 129 ± 0.42 50 ± 0.23 88 ± 0.71 83 ± 0.62 92 ± 1.85 935 ± 41.35 Shrub species Memecylon umbellatum Phyllanthus polyphyllus Allophylus serratus Flueggea leucopyrus Eugenia bracteata Ochna obtusata Dichrostachys cinerea Catunaregam spinosa Flacourtia indica Capparis brevispina Total 16 8 6 5 3 2 1 1 1 1 1.51 ± 0.01 0.89 ± 0.03 0.88 ± 0.03 0.76 ± 0.01 0.34 ± 0.02 0.03 ± 0.00 0.45 ± 0.05 0.25 ± 0.05 0.18 ± 0.02 0.09 ± 0.03 5.38 ± 1.62 321 ± 3.0 163 ± 2.62 108 ± 1.22 96 ± 0.73 54 ± 0.60 54 ± 1.82 17 ± 0.06 17 ± 0.12 13 ± 0.07 17 ± 0.58 860 ± 26.32 Total individuals above 1m height Total twenty dominant species Den/ha 93.2 74.8 2208 1795 Girth class distribution in relation to percentage individuals are given in Figure 2. The girth size class 10- 20 showed the highest number of species along with the highest species richness. Species richness of first and third gbh size class is more or less similar, while the proportion of individuals is quite high in the gbh size class 20-30. (34) 50 40 30 (8) (17) (9) (11) > 50 10 40 - 50 (18) 20 30 - 40 20 - 30 10 - 20 0 0 - 10 % individuals in different girth classes 60 147 Girth size class distribution (cm) Figure 2. Girth class distribution of individuals greater than 10 cm gbh (trees and shrubs) in the Popham Arboretum in Dambulla. The number of species that were present in each size class is given in parentheses. FLORISTIC COMPOSITION OF VEGETATION A total of 101 species belonging to 91 genera and 42 families were identified. Among the taxa identified 45 had medicinal value, 14 timber value and 7 both timber and medicinal value. There were 45 tree spp., 27 shrub spp., 18 herbaceous spp. and 11 climber spp. comprising 5 endemics (Diospyros oppositifolia, Diplodiscus verrucosus, Canarium zeylanicum, Cassine balae and Micromelum minutum) were recorded. Species and family dominance On the basis of IVI values, the leading dominant in the vegetation size class >10 cm gbh were S. cumini, G. damine and M. umbellatum etc (Figure 3). The contribution of basal area and density of those species showed alternative variation. Compared to basal area density of constitute species showed more or less similar among the leading species. 35 % Relative Basal Percent IVI Value 30 % Relative Density 25 20 15 10 5 V. altissima I. pavetta P. opposita F. leucopyrus L. tetraphylla A. serratus D. verrucosus B. racemosa D. ferrea P. polyphyllus P. suberifolium M. parvifolia C. swietenia M. umbellatum G. damine S. cumini 0 Plant species Figure 3. Variation in IVI values of the first ten leading species in Popham Arboretum in Dambulla. The most dominant family based on IVI value was Myrtaceae followed by Tiliaceae and Rubiaceae. However the two leading families constitute only two species, while Rutaceae, Fabaceae, Euphorbiaceae and Rubiaceae contribute higher species richness (Figure 4). % Relative Basal area 30 (2) % Relative Density (2) (5) 20 (7) (6) (1) (6) (3) 10 (4) (1) Sterculiaceae Ebenaceae Sapindaceae Fabaceae 148 Melastomataceae Euphorbiaceae Rutaceae Rubiaceae Tiliaceae 0 Myrtaceae Percent IVI Value 40 Family Figure 4. Variation in IVI values of the first ten leading families in Popham Arboretum in Dambulla. Number of species in a given family is given in parenthesis. POPULATION SIZE Considering the population size classes, species richness was high in the vegetation size class over 10 cm gbh (Figure 5). The population between 2 -10 recorded the highest number of species for both vegetation size classes. Vegetation > 10 cm gbh 20 15 10 5 Vegetation < 10 cm gbh No. of species in each size classes 25 41 - 50 31 - 40 21 - 30 11 - 20 2 - 10 1 only 0 Figure 5. Number of species in each of the population size classes in the vegetation of the Population size classes Popham Arboretum in Dambulla. TRENDS AND RELATIONSHIPS Figure 6 gives the species association in terms of their occurrence in the arboretum and woodland. Cluster A and B associated with nine species including one species of borderline negative (Meum), while cluster C and D associated with eleven species including one species of borderline positive (Phpo). In the TWINSPAN diagram WL3 and WL1 formed a single distinguishable cluster, while AR1, AR2 and AR3 formed a single distinguishable cluster along with WL1 cluster. Therefore, species very common to both arboretum and woodland occur at the WL1 plot. 149 A R2 A R1 Chsw E = 0.20 Cluster C Cluster D Bema Stnu Ixpa Axis 2 E = 0.18 Plop WL2 Cadi Sycu Alse Dive Cluster A Bara Dife A R3 Eubr Ocob Ptsu Meum Lete Borderline negative Saem Phpo WD3 E = 0.19 Grda Borderline positive Cluster B Flle WL1 Axis 1 Figure 6. The ordination diagram based on species abundance in terms of plots. Species are denoted by four letters (two letters from generic name of particular species and two letters from species name) and plots are denoted by two letters and numerical number referring to arboretum (AR) and woodland (WL). Four clusters identified by TWINSPAN were subjectively marked in the ordination diagram, where cluster A and B denoted the positive branch of the cluster, while cluster C and D denoted the negative branch of the cluster. Letter E indicates the Eigen value of each cluster. Cluster A was composed of four typical dry zone tree species such as D. verrucosus, D. Ferrea, L. tetraphylla and S. emarginata with typical shrub viz., A. serratus. All species associated in cluster C (Plots of AR1 and AR2) were trees such as C. swietenia, P. opposita, S. nux-vomica, Bauhinia racemosa and Ixora pavetta. Cluster D were associated with mostly shrub species such as P. Polyphyllus, Ochna obtusata, Benkara malabarica and Eugenia bracteata and trees such as P. suberifolium and Canthium dicoccum var. umbellatum. 150 DISCUSSION Perera (1998) carried out a vegetation survey at Sigiriya sanctuary and recorded over 5,000 individuals of trees and 5,000 individuals of shrubs per hectare. This was quite a high density compared to density of trees and shrubs at the arboretum. This is because of the contribution of immature seedlings and sapling in the Sigiriya sanctuary. Furthermore, size class distribution of individuals at Sigiriya sanctuary showed very few large trees and the majority has smaller diameters. But in the Dambulla arboretum, the predominance of mature trees indicates that the forest in Dambulla arboretum is higher in its regeneration state. Samarasinghe 1995 showed that Chloroxylon swietenia was the most dominant tree based on its relative density (RD =3.4 in year1993) followed by Grewia damine (RD = 0.1), Diospyros ferrea (RD = 0.1) and Pterospermum suberifolium (RD = 0.5). The present survey found that these are the leading tree species however; their dominance within the survey was altered. On the other hand, Samarasinghe recorded Phyllanthus polyphyllus was the leading dominant, whereas it was the second dominant species in the present survey. Relatively high stem densities, particularly stems in the smaller girth size classes is due to a dominance of species of typically small size i.e. Diospyros ferrea, Memecylon umbellatum and Ixora pavetta and coppicing of shrubs. Cramer in 1993 documented 192 plant species of which 88 species were trees composed of rare, endemic and endangered species. The vegetation in the arboretum, when compared with primary forests in the dry zone has climax species in the canopy layer. Reverse J shaped curve observed in the vegetation of the arboretum indicates that these are in the late regenerating phase of secondary vegetation. The vegetation of the arboretum was regenerated via silvicultural management since 1963 and therefore it harbors mature trees, while the vegetation of the woodlands declared for conservation since 1989 by the Institute of Fundamental Studies has both mature and immature trees. Bauhinia racemosa, G. damine and P. suberifolium that play an important role in the damana forest, adapted to withstand fire can be seen here. On the other hand, M. hexandra, C. swietenia, Drypetes sepiaria and Diospyros ebenum are climax forest species in the dry zone. The occurrence of fire tolerant species as well as climax forest species indicates that the forest is in a late succession stage of regeneration. The IVI values of C. swietenia are fairly good. This tree species was considered as the key species in monsoon forests, which in the past, was classified as M. hexandara - D. sepiaria - C. swietenia association based on their abundance in the forest. However, the present survey recorded that these species in the arboretum still provide refuge for remarkably rare species such as Diospyros ebenum, Diospyros malabarica and Diospyros oppositifolia. Of the shrub species M. umbellatum recorded the highest importance value. Fernando (1996) has suggested that the very high relative importance value of these species in forests of Maduru Oya national park indicate that there has been minimal disturbance. This is further confirmed because the arboretum was initiated in the year 1963 and disturbance has not taken place since and therefore an abundance of this species is obvious and to be expected. 151 The three most dominant families by number of species in the arboretum and woodland were Rutaceae, Fabaceae and Euphorbiaceae. Plants in these families are adapted to grow under dry and stressful environmental conditions. Some members of the Euphorbiaceae and Rutaceae have thorns, latex, cladodes and/or relatively small leaves to conserve water and escape herbivory. Some members of Fabaceae can fix Nitrogen and therefore, they can perform well in the dry zone forests. The present results further confirm Samarasinghe’s findings of diverse families viz., Euphorbiaceae, Rubiaceae and Rutaceae. The ordination diagram clearly shows that the plots of the arboretum and woodland are divided into two because of different regeneration stages. The species inhabitants in the arboretum are mainly trees and shrubs typically found in the primary dry zone forests. Cluster A resulted in basically dominant species in the woodland. On the other hand, Cluster C dominated by valuable timber such as Chloroxylon swietenia and Pleurostylia opposita and mainly inhabits the arboretum. CONCLUSIONS The study proves the importance of the arboretum with regard to its fairly rich flowering plants including endemics, valuable timber trees and plentiful medicinal plants. Furthermore, arboretum provides refuge for typical dry zone key plant species through natural regeneration from the soil seed bank. Application of Popham method is recommended for degraded land to convert the land into productive ecosystems. ACKNOWLEDGEMENTS The authors extend their thanks and appreciation to Mr. Sam Popham, Creator of this Arboretum, Mr. Nigel Billimoria - Senior Manager, Ruk Rakaganno - A National NGO and current manager of the Arboretum, and the Institute of Fundamental Studies for permission to carry out research. And last but not least we wish to thank the field assistants K.G. Sumane Banda, U.G. Rathnasiri, K.G. Palis and M.G. Jamis. 152 REFERENCES Anon. (1996). Newsletter. Dambulla arboretum and woodland. In: J. Samarasinghe, (ed.) Issues 2, IFS Popham arboretum. Cramer, L.H. (1993). A forest arboretum in the dry zone. Institute of Fundamental Studies, 241. Dassanayake, M. D. and Fosberg, F.R. (1990-1991). A Reversed Hand Book to the Flora of Ceylon. Amerind Publishing Co. Ltd., New Delhi. 1-6. Dassanayake, M. D., Fosberg, F. R. and Clayton, W.D. (1994-1995). A Reversed Hand Book to the Flora of Ceylon. Amerind Publishing Co. Ltd., New Delhi. 8-9. Dassanayake, M. D. and Clayton, W. D. (1996-2000). A Reversed Hand book to the Flora of Ceylon. Amerind Publishing Co. Ltd., New Delhi. 10-12. Fernando, H.S.K.F. (1996). A comparative study on the ecology of woody vegetation of forest types in the Maduru Oya National Park. M.Phil. thesis, University of Peradeniya, Sri Lanka. Jayaweera, D.M.A. (1982). Medicinal plants (indigenous and exotic) used in Sri Lanka (I edition). The National Science Council, Sri Lanka. 1-4. st Perera, G.A.D. (1998). Vegetation and the regeneration of moist deciduous forests at Sigiriya, Sri Lanka. Photo. 5 (1): 9-16. Popham, F.H. (1993). Dambulla. A Sanctuary of tropical trees. Sam Popham foundation, UK, 72 p. Samarasinghe, J. (1995). Regeneration dynamics of silviculturally assisted dry zone scrub vegetation at Dambulla arboretum. In: H.S. Amarasekera and S.G. Banyard (Editors), Proceedings of the Annual Forestry Symposium, Dept. of Forestry and Environment Science, University of Sri Jayewardenepura, Sri Lanka. 291-299. Senaratna, L. K. (2001). A check list of the flowering plants of Sri Lanka. National Science Foundation, Sri Lanka. 451 p. Weerawardane, N.D.R. (1999). Natural regeneration of some dry zone forest species assisted by silvicultural management in dry zone woodland at Dambulla. The Sri Lanka forester. 23 (3&4): 7-17. 153 Appendix 1 Species list of the vegetation in the IFS Popham Arboretum (T = Tree, S = Shrub, H = Herb, C= Climber). Species/Family Acanthaceae Andrographis paniculata* Justicia procumbens Amaranthaceae Aerva lanata* Amaryllidaceae Crinum sp:* Apocynaceae Aganosma cymosa* Carissa spinarum Ichnocarpus frutescens Araceae Amorphophallus paeoniifolius* Arecaceae Borassus flabellifer* Asclepiadaceae Wattakaka volubilis* Asteraceae Elephantopus scaber Eupatorium odoratum Boraginaceae Carmona retusa Cordia dichotoma Ehretia laevis Burseraceae Canarium zeylanicum* Capparaceae Capparis brevispina Celastraceae Cassine balae* Pleurostylia opposita Clusaceae Garcinia morella* Convolvulaceae Ipomoea littoralis* Dioscoreaceae Dioscorea alata* Dracaenaceae Sansevieria zelanica* Ebenaceae Diospyros ebenum Diospyros ferrea Diospyros malabarica Diospyros oppositifolia Diospyros ovalifolia* Euphorbiaceae Bridelia retusa Croton officinalis* Dimorphocalyx glabellus Code Local Name Medicine-M/ Life form Timber-T Anpa Jupr Heen-bin-kohomba M Mayani H H Aela Pol pala H M Crin H Agcy Casp Icfr Heen-Karamba Kiri-wel Ampa Kidaran H Bofl Tal T Wavo Anguna Elsc Euod Et-adi Podisinghomaran H S Care Codi Ehle Heen-tambala Lolu Walangasal S T T Caze Dik-kekuna Cabr Wellangiriya M S Caba Plop Neraloo Panakka M T T T Gamo Gokatu T Ipli Tel-kola C Dial Hingurala C Saze Niyanda H Dieb Dife Dima Diop Diov Kaluwara Kalu habarala Timbiri Kalu-mediriya Kunumella Brre Crof Digl Katakala Weli wenna 154 M M M H S H C T T M T M/T T T T T T T S S Drypetes sepiaria Euphorbia antiquorum* Euphorbia heterophylla* Flueggea leucopyrus Givotia moluccana* Margaritaria indicus Phyllanthus amarus Phyllanthus polyphyllus Fabaceae Bauhinia racemosa Cassia fistula Cassia roxburghii Derris benthamii Dichrostachys cinerea Mimosa pudica Tamarindus indica Flacourtiaceae Flacourtia indica Lamiaceae Ocimum americanum* Lauraceae Alseodaphne semecarpifolia Litsea glutinosa Linaceae Hugonia mystax* Loganaceae Strychnos potatorum* Strychnos nux-vomica Melastomataceae Memecylon umbellatum Osbeckia sp. Meliaceae Azadirachta indica Chukrasia tabularis Moraceae Streblus asper Myrtaceae Eugenia bracteata Syzygium cumini Ochnaceae Ochna obtusata Oleaceae Chionantus zeylanica* Jasminum angustifolium Periplocaceae Hemidesmus indicus Poaceae Setaria sp. Rhamnaceae Scutia myrtina* Ziziphus oenoplia Rhizophoraceae Cassipourea ceylanica* Rubiaceae Benkara malabarica Canthium coromandelicum Canthium dicoccum var. umbellatum Catunaregam spinosa Ixora pavetta Mitragyna parvifolia Drse Euan Euhe Flle Gimo Main Pham Phpo Wira Daluk M Heen-katu-pila M Karawu Pita-wakka Kuratiya M Bara Cafi Caro Disc Dici Mipu Tain Maila Ehela Ratu-wa Kala-wel Andara Nidi-kumba Siyambala Flin Uguressa S Ocam Heen-tala H Alss Ligl Wewarani Bomee T M T T Humy Bu-getiya M C Stop Stnu Ingini Goda kaduru M M T T Meum Osbe Kora-kaha Bowitiya M T S Azin Chta Kohomba Hulan-hik M/T M/T T T Stas Geta-netul Eubr Sycu Daeduwa Madan Ocob Galkera Chze Jaan Geratiya Wal-pichcha M H C Hein Heen-iramusu M H M M/T M M M/T T T H S H T H S T T T C S H T T M/T T T S Seta H Scmy Zioe Heen-eraminiya Cace Pana Bema Caco Cadi Cats Ixpa Mipa Getakula Kara Bokutu Kukurman Goda ratmal Helamba 155 M S S S M M M M M/T S S T S T T Morinda umbellata* Oldenlandia corymbosa* Tarenna asiatica Rutaceae Acronychia pedunculata Atalantia ceylanica Chloroxylon swietenia Clausena indica* Glycosmis mauritiana Limonia acidissima Micromelum minutum Pleiospermium alatum Toddalia asiatica Sapindaceae Allophylus serratus Cardiospermum halicacabum* Lepisanthes tetraphylla Sapindus emarginata Sapotaceae Manilkara hexandra Sterculiaceae Helicteres isora* Pterospermum suberifolium Tiliaceae Diplodiscus verrucosus Grewia damine Microcos paniculata* Verbenaceae Gmelina asiatica Lantana camara Vitex altissima Vitaceae Cissus quadrangularis* Cissus vitiginea Moum Maha-kiri-wel Olco Wal-patpadagam Taas Tarana M Acpe Atce Chsw Clin Glma Liac Mimi Plal Toas Ankenda Yakinaran Buruta Migon-karapincha Bol-pana Divul Wal-karapincha Tumpat-kurundu Kudu-miris Alse Caha Lete Saem Kobbe Penala-wel Dambu Penela M M S C T T Mahe Palu T T Heis Ptsu liniya Welan S T Dive Grda Mipa Dik-wenna Daminiya Keliya T T S Gmas Laca Vial Demata Ciqu Civi Hiressa Wal-niviti Milla M M T C H T M M M M T S T S S T S T C M M T S S T M C C *Plants, which are not documented in Cramer’s checklist in the arboretum. Endemic species are in bold. Source: Jayaweera (1982), Cramer (1963) and Senaratna (2001). 156 HABITAT QUALITY AND AVAILABILITY OF THE WESTERN CEYLON SLENDER LORIS, Loris tardigradus IN THE KOTTAWA ARBORETUM S. N. GAMAGE1*, D. WEERAKOON2 and A. GUNWARDENA1 1 Department of Animal Science, University of Ruhuna, Mapalana, Kamburupitiya. 2 Department of Zoology, University of Colombo, Colombo-03. ABSTRACT The red slender loris, Loris tardigradus is one of the three primate species endemic to Sri Lanka. Currently there are two recognized subspecies of the red slender loris, L. t. tardigradus and L. t. nycticeboides. Of these L. t. tardigradus (Western Ceylon slender loris) inhabits rainforests in the southwestern region of the island while L. t. nycticeboides (Ceylon mountain slender loris) is restricted to the montan zone. Kottawa Arboretum harbors one of the few remaining L. t. tardigradus populations in the country. This study was conducted to determine the population density, habitat selection criteria and to assess the habitat availability of L. t. tardigradus in the Kottawa Arboretum. Using the line transect method, 34 sightings were made over a period of 21 months. Based on these observations the density of Western Ceylon slender loris in the Kottawa Arboretum and habitat selection criteria were ascertained. The calculated density of L. t. tardigradus in the Kottawa Arboretum is 41animals/ km2. The average height of trees preferred by L. t. tardigradus is 13.97 m ± 6.02. Most of the time lorises were observed at a height range of 3.5-15 m above the ground level. Average height from the ground level where L. t. tardigradus were observed to occupy the tree was 8.64 m ± 5.00. Of the 50 tree species recorded in the Kottawa Arboretum, L. t. tardigradus was found to utilize only 16 species. KEY WORDS: Density, habitat availability, habitat selection, L. t. tardigradus, low land rain forest INTRODUCTION Sri Lanka, with a total land area of 65,610 km2 is a tropical island situated in the Indian Ocean. The southwestern region of Sri Lanka, encompassing approximately 20,000 km2, is the only aseasonal ever wet region in the whole of South Asia (Ashton & Gunatilleke, 1987; Gunatilleke et al., 2005). This region is referred to as the wet zone of Sri Lanka and receives up to 3000 mm of rainfall annually. Wet-zone of Sri Lanka along with the Western Ghats of India is designated as one of the world’s 11 biodiversity “hyperhot” hotspots, in demand of extensive conservation investment (Myers et al., 2000; Brookes et al., 2002). However, agro ecosystems and human settlement cover most of the land area in the wet-zone of Sri Lanka (Pemadasa, 1996; Ashton et al., 1997; Gamage, 2005). A burgeoning human population, demand for subsistence land, and a high proportion of endangered and endemic species within Sri Lanka’s wet zone have resulted in its being declared a critically endangered eco-region (Mill, 1995; Nekaris et al., 2005). The slender loris (Loris) is a small nocturnal prosimian primate endemic to Sri Lanka and South India. The forests of Sri Lanka are home to two species of slender loris (Loris tardigradus and L. lydekkerianus), with four currently recognized subspecies, L. t. tardigradus, L. t. nycticeboides, L. l. nordicus, and L. 157 l. grandis (Osman Hill, 1953; Groves, 2001; Nekaris & Jayewardene, 2003). The red slender loris, Loris tardigradus is endemic to Sri Lanka (Groves, 2001; Nekaris & Jayewardene, 2003; Nekaris et al., 2005). The conservation status of this species (L. tardigradus) has since been elevated to the endangered category (IUCN, 2004). According to Nekaris & Jayewardene, (2003) the Western Ceylon slender loris, L. t. tardigradus only inhabits rainforest in the southwestern region of the island while the other sub species L. t. nycticeboides is restricted to the montane region above 2000 m. Preliminary abundance estimates of L. t. tardigradus showed that these lorises are patchily distributed, even within a single forest reserve (Nekaris & Jayewardene, 2003; Nekaris et al., 2005). The Kottawa Arboretum is a part of Kottawa-Kombala forest reserve and is classified as a lowland rainforest (Pemadasa, 1996; Ashton et al., 1997). Kottawa Arboretum harbors one of the few remaining Western Ceylon slender loris L. t. tardigradus populations in the country. Very few studies have focused on the ecology of L. t. tardigradus except one study that has been done in Masmullah proposed forest reserve (Nekaris et al., 2005; Pers. Com. Bernede). Thus, the overall aim of this study is to determine the density, habitat selection criteria and habitat availability for L. t. tardigradus in the Kottawa Arboretum. MATERIALS AND METHODS Study site Kottawa Arboretum is a part of Kottawa-Kombala forest reserve, which is situated in the southern region of Sri Lanka and belongs to Yakkalamulla Divisional Secretariat Division of Galle District (6°05’ N and 80°18’ E). The extent of the Arboretum is approximately 20ha and it is classified as a lowland rainforest (Pemadasa, 1996). Remnants of Dipterocarpus forest occur in the Arboretum site. Secondary forest occurs where the original forest cover has been removed due to logging. Some of the logged areas are replanted with pinus trees (Pinus caribaea), by the Forest Department (Gamage, 2005). The study taxon L. t. tardigradus is the smallest of the slender lorises, weighing 85-172g (Nekaris et al., 2005). On the basis of museum specimens, Groves (2001) recently distinguished it from other slender lorises. It occurs only in the southwestern region of Sri Lanka (Nekaris, 2003; Nekaris & Jayewardene, 2003; Nekaris et al., 2005). 158 Population density This study was carried out from August 2003 to April 2005, using the fixed line transect method. Five line transects, each 200 m long separated by 50 m was marked in the study site. These transect’s were repeated 19 times during the sampling period. Along the transect the distance to the animal was visually observed and the angle between the animal and transect was measured using a compass. Density was calculated using the following equation. D = f ∑ ni / 2L (Sutherland, 1996), where D is density, L is length of the transect; ni is numbers of animals recorded in the recognized zones, and f = a1 + 1/ dt where a1 = 2 (∑ cos gi / dt ∑ ni) and dt is distance beyond which data were truncated; gi = π ni di / dt; and di = zi sin øi, where zi is distance of the ith animal when first observed, øi is angle to ith animal when first observed, di is perpendicular distance from transect line to ith animal. Observation of loris habitat use Focal animal instantaneous point sampling method was used to obtain behavior data (Charles-Dominique & Bearder, 1979; Nekaris et al., 2005). Headlamps fitted with red filters were used to minimize disturbance to the animal. Data regarding habitat use was recorded upon first spotting an animal. The type of data collected included substrate size, substrate angle, height from the ground level, tree height, and tree type (Nekaris, 2001; Nekaris et al., 2005). Vegetation sampling The plot-less sampling technique (Sutherland, 1996) was used to ascertain the density of tree species of the study sites. 30 sampling points were chosen randomly in the study site. At each point two sticks were placed perpendicular to each other to demarcate four quadrates. In each quadrate the nearest tree (girth >10cm) from the point and the nearest neighbor of the tree (girth > 10cm) in the same direction was identified. Then identity of the plant species, distance from the point to the tree, distance between plant and its nearest neighbor, circumference at breast height and the estimated tree height was recorded. Furthermore, the percentage arboreal continuity of each tree was measured using the following scale 0-1.5, 1.5-3.5, 3.5-5.0, 5.0-10.0, 10-15 & 15 < meters (Nekaris et al., 2005; Gamage, 2005). In addition microhabitat characteristics of each of the following strata 0-1.5, 1.5-3.5, 3.5-5.0, 5.0-10.0, 10-15 & 15< meters (nature, size, orientation and the presence or absence of vines and epiphytes within the strata) was recorded. In each quadrate percentage of saplings on the ground was determined using the Braun-Blanquet scale (Sutherland, 1996). Floral density was calculated using the T-square method (Sutherland, 1996). The equation used was D = m2 / (2.828 ∑ xi ∑ zi), where D is tree density (trees/ha), m is number of sampling points, xi is distance from the sampling point (m), and zi is distant to the nearest neighbor (m). A test of random distribution was determined using the equation, ‘t’ = {∑ [xi2 / (xi2 + zi2 / 2)] – m / 2} (12/m), where a value greater than 1.96 indicates a non-random distribution (Sutherland, 159 1996). Basal area was calculated using the equation, Ba = (2 x CBH/ 4) x D, where Ba is the basal area, CBH is circumference at breast height, and D is tree density (Sutherland, 1996). RESULTS Population density During the observation period a total of 36 sightings (data points) of L. t. tardigradus were recoded however two sighting (data points) were omitted for calculations due to difficult to identify the tree. The unit density of L. t. tardigradus at the Kottawa Arboretum was found to be 41-animals/ km2. Therefore the estimated population size of L. t. tardigradus in the Kottawa Arboretum (extent 20 ha) is approximately 8 animals. Floral composition of the study area The 240 trees surveyed during the vegetation study represented 50 species that belongs to 25 families. Of these 37 species (74%) are endemic to Sri Lanka while the remaining 13 (26%) can be defined as native species (Table 1). The most abundant tree species recorded was Acronychia pedunculata (n = 27; 11.3%). Other relatively common species included Lijndenia capitellata (n = 21; 8.8%), Agrostistachys coriacea (n = 11; 4.5%), and Mangifera zeylanica (n = 10; 4.2%). All other tree species were encountered less than 10 times during the survey period. The calculated tree density of all trees in the sample was 1917 trees/ha. The calculated‘t’ value for the test of random distribution was +21.48. The average height of trees in the sample was 8.19 m ± 6.55, with a minimum of 2.5m and a maximum of 40m. The average CBH was 34.38 cm ± 52.64; with a minimum of 10 cm and a maximum of 305 cm. The tallest tree (40 m) recorded from the study site was Dipterocarpus hispidus with a CBH of 305 cm. The most common tree species, Acronychia pedunculata, had an average height of 4.96 m ± 0.84, and a CBH of 17.48 cm ± 4.03. The average basal area of the trees was 94.0 ± 220.4 m2/ha. Density of ground cover between nearest neighbor trees was determined 120 times using the Braun-Blanquet scale. The average density was 2.1 ± 1.8 %. Table 1. Tree species (>10 cm CBH) recorded using the plot-less sampling technique in the Kottawa Arboretum. (Abbreviations: En = endemic, Na = native). 160 Family Anacardiaceae Anacardiaceae Anacardiaceae Anacardiaceae Anacardiaceae Annonaceae Arecaceae Burseraceae Celastraceae Celastraceae Clusiaceae Clusiaceae Clusiaceae Clusiaceae Clusiaceae Dilleniaceae Dilleniaceae Dipterocarpaceae Dipterocarpaceae Dipterocarpaceae Dipterocarpaceae Dipterocarpaceae Dipterocarpaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Flacourtiaceae Flacourtiaceae Icacinaceae Lauraceae Lauraceae Melastomataceae Melastomataceae Moraceae Myrataceae Myrataceae Myristicaceae Myristicaceae Myristicaceae Ochnaceae Oleaceae Rhizophoraceae Rhizophoraceae Rubiaceae Rutaceae Sapotaceae Sapotaceae Symplocaceae Thymelaeaceae Verbenaceae Species name Campnosperma zeylanicum Mangifera zeylanica Semecarpus nigro-viridis Semecarpus subpeitata Semecarpus walkeri Xylopia chamionii Caryota urens Canarium zeylanicum Bhesa ceylanica Kokkoona zeylanica Calophyllum bracteatum Calophyllum moonii Calophyllum thwaitesii Garcinia quaesita Mesua thwaitesii Dillenia retusa Schumacheria alnifolia Dipterocarpus glandulosus Dipterocarpus hispidus Dipterocarpus zeylanicus Shorea affinis Stemonoporus canalicuculatus Vateria copallifera Agrostistachys coriacea Bridelia moonii Chaetocarpus castanocarpus Homalium zeylanicum Hydnocarpus octandra Stemonurus apicalis Cryptocarya wightiana Lisea gardneri Lijndenia capitellata Memecylon capitellatum Artocarpus nobilis Syzygium makul Syzygium neesianum Horsfieldia irya Horsfieldia iryaghedhi Myristica dactyloides Ochna lanceolata Chionanthus zeylanica Anisophyllea cinnamomoides Carallia brachiata Timonius flavescens Acronychia pedunculata Palaquium grande Palaquium petiolare Symplocos coronata Gyrinops walla Vitex altissima 161 Common Count name Aridda 8 Atamba 10 Gatabadulla 6 Kabarabadulla 7 Badulla 6 Dathketiya 1 % Status 3.33 4.17 2.50 2.92 2.50 0.42 En En En En En En Kithul Kekuna Pelan Kokum Walukeena Dombakeena Batukeen Goraka Diyana Godapara Kakiriwara Dorana Buhora Hora Beraliya Mandora Hal Beru Pathkela Hedawaka Liyan Waldeul Uruhonda Gulmora Thalan Pinibaru Velikaha Badidel Aluboo Panukera Eriya Ruk Malaboda Bokera Gerieta Velipenna Dawata Angana Ankenda Kiripedda Kirihambiliya Uguduhal Walla Milla 0.42 1.25 1.67 0.83 1.25 0.83 3.33 2.08 2.08 2.92 3.33 0.83 1.25 2.92 1.67 1.25 0.83 4.58 2.50 2.08 0.83 0.42 1.67 0.42 0.42 8.75 1.25 0.83 1.25 0.83 3.75 2.93 1.25 0.83 0.42 0.83 1.67 0.42 11.25 0.83 0.83 0.83 2.08 1.67 Na En En En En En En En En En En En En En En En En En En Na Na En En Na En En En En En En Na Na Na Na Na En Na En Na En En En Na Na 1 3 4 2 3 2 8 5 5 7 8 2 3 7 4 3 2 11 6 5 2 1 4 1 1 21 3 2 3 2 9 7 3 2 1 2 4 1 27 2 2 2 5 4 A total of 28 climbers were identified. They are, Dalbergia lattifolia (Fabaceae) (n=6; 21.4%), Salacia reticulata (Hippocrateaceae) (n=4; 14.3%), Pandanus sp. (Pandanaceae) (n=4; 14.3%), Dalbergia pseudo-sisoo (Fabaceae) (n=3; 10.7%), Tetracera sarmentosa (Dilleniaceae) (n=3; 10.7%), Coscinium penistratum (Menispermaceae) (n=3; 10.7%), Gyrinops walla (Thymelaeaceae) (n=3; 10.7%), Smilax zelanica (Smilacaceae) (n=1; 3.6%) and Entada phaseoloides (Fabaceae) (n=1; 3.6%). Usage of trees by L. t. tardigradus Of the 50 tree species recorded in the Kottawa Arboretum, L. t. tardigradus was found to utilize only 16 species (Figure 1). Among the trees species preferred most are Dillenia retusa, Chaetocarpus castanocarpus, Horsfieldia iryaghedhi and Mesua thwaitesii. Of the four most abundant tree species, Acronychia pedunculata, Lijndenia capitellata, Agrostistachys coriacea and Mangifera zeylanica only two species (Lijndenia capitellata, and Mangifera zeylanica) were used by L. t. tardigradus and even then these trees were used sparsely. A detailed description of the trees on which L. t. tardigradus was seen is given in table 2. Characteristics of substrate used by L. t. tardigradus The average height of the trees used by L. t. tardigradus was 13.97 m ± 6.02 (range 3-28 m). Average height from the ground level where L. t. tardigradus were observed to occupy the tree was 8.64 m ± 5.00 (range 1-22 m). Most of the time (n = 27; 79%) lorises were seen to occupy a position in the range of 3.5-15 m above the ground level (Figure 2). Only on 2 occasions were loris seen at a height greater than 20 m from the ground level. 20 18 16 14 % 12 10 8 6 4 2 % Usage by lorises % Availability Figure 1. Tree availability at Kottawa Arboretum and frequency of utilization by L. t. tardigradus (n = 34). 162 H al D om bak eena Batuk eena Kok um D athk etiy a Aridda M illa Pinibaru Liy an Pelen Etam ba Bedidel D iy anaa R uc k H edaw ak e G odapara 0 Table 2. Frequency and percentage at which slender lorises were encountered on different plant species (n = 34). Species Dillenia retusa Chaetocarpus castanocarpus Mesua thwaitesii Horsfieldia iryaghedhi Artocarpus nobelis Vitex altissima Homalium zeylanicum Lijndenia capitellata Mangifera zeylanica Bhesa ceylanica Campnosperma zeylanicum Xylopia chamionii Kokkoona zeylanica Calophyllum thwaitesii Calophyllum moonii Vateria copallifera Common name Frequency of usage Godapara Hedawake Diyanaa Ruck Bedidel Milla Liyan Pinibaru Etamba Pelen Aridda Dathketiya Kokum Batukeena Dombakeena Hal 6 4 3 3 2 2 2 2 2 2 1 1 1 1 1 1 Average height Tree Occupied 13.8 ±3.5 14.0 ±4.9 14.3 ±4.0 11.3 ±3.1 20.5 ±6.4 13.5 ±2.1 12.0 ±2.8 3.0 ±0.0 17.0 ±11.3 9.0 ±1.4 15 7 26 17 28 16 9.0 ±2.8 8.2 ±3.3 8.0 ±2.0 6.0 ±1.0 14.0 ±8.5 9.5 ±2.1 6.5 ±0.7 1.5 ±0.7 9.0 ±7.1 4.5 ±0.7 15 3 18 12 22 14 The available substrate was grouped into four categories according to the diameter as twigs (≤ 1 cm), small branches (2 – 5 cm), medium sized branches (6 – 10 cm) and large branches (≥ 10 cm). L. t. tardigradus showed a higher preference for branches or twigs (n = 27; 76 %) followed by vines (n = 6; 15 %). L. t. tardigradus were seen rarely on tree trunks (Figure 3). The available substrate was grouped into three categories according to the orientation as vertical, horizontal and oblique. L. t. tardigradus showed a higher preference to oblique and horizontally oriented substrates compared to vertical oriented substrates (Figure 4). 45.0 40.0 Percentage 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 0-1.5 1.5-3.5 3.5-5 5-10 10-15 Height classes (m) Figure 2. Vertical distribution of L. t. tardigradus 163 >15 50 60 40 50 40 % 30 % 30 20 20 10 10 0 Vertical 0 < 1cm 2-5 6 - 10 Horizontal Oblique Substrate orientation > 10 cm Available at the study site Size category Availability at the s tudy s ite Loris utilisation Loris utilisation Figure 4. Orientation of the available substrate and usage by L. t. tardigradus Figure 3. Substrate size availability and usage by L. t. tardigradus Habitat structure The vertical axis of the Kottawa Arboretum was divided into six height classes and the tree availability, substrate continuity, substrate availability and the nature of the available substrate within each of these height classes was investigated to ascertain the habitat availability of L. t. tardigradus within the Kottawa Arboretum. Tree availability: Percentage of trees that reach the maximum height of each of the six height classes was determined to asses the density of the available substrate. All the trees studied were taller than 1m. After 1m the number of trees reaching the maximum height of the height class decreased gradually. A significant reduction occurs after 3.5m (figure 5). % Tree availabilty 100 80 60 40 20 0 0.15m 1m 3.5m 7m 10m 15m Heigt Class (m) Figure 5. Percent tree availability at different height classes in the Kottawa Arboretum Substrate Continuity: Substrate continuity was present in all of the height classes. The height classes 3.5-5m (23%) and 5-10m (22%) had the highest percentage of substrate continuity (Figure 6). 164 Deleted: aroboretum Substrate contiuity 25 20 15 10 5 0 0-1.5m 1.5-3.5m 3.5-5m 5-10m 10-15m 15m< Height Class Figure 6. Percent substrate continuity at different height classes in the Kottawa Arboretum Habitat quality and availability: The habitat quality of Kottawa Arboretum was analyzed based on the habitat selection criteria (type of substrate, orientation and the girth) observed in this study for L. t. tardigradus. The habitat quality for lorises was found to be highest between 3.5 m – 10 m height range (Figures 7, 8 and 9). However, the habitat quality between 10m to 15 m was also found to be suitable for lorises. % Substrate type 100 80 60 Branch Trunk Vine 40 20 0 0.15m 1.5m 3.5m 5m 10m 15m Height Class Figure 7. Percentage availability of different substrate types at the different heights of the vertical axis at Kottawa Arboretum. 165 Deleted: aroboretum 100 90 % Orientation 80 70 60 50 40 Vertical 30 Horizontal 20 Oblique 10 0 0.15m 1.5m 3.5m 5m 10m 15m Height Class Figure 8. Percentage orientation of substrate types in the different heights of the vertical axis at Kottawa Arboretum. 100 90 % Substrate size 80 70 60 < 1cm 50 < 5 cm 40 30 < 10 cm > 10 cm 20 10 0 0.15m 1.5m 3.5m 5m 10m 15m Height Class Figure 9. Available substrate size at different heights of the vertical axis at Kottawa Arboretum. 166 DISCUSSION Population density of L. t. tardigradus recorded in the Kottawa Arboretum during this study (41-animals/ km2) is three times greater than the population density recorded at Massmulla Proposed Forest Reserve (13 animals/ km2) by Nekaris & Jayewardene (2004) who notes that it is the highest population density of L. t. tardigradus recorded in Sri Lanka. This indicates that the habitat quality of Kottawa Arboretum is much better than Masmulla Proposed Forest Reserve even though it is smaller in size. Therefore, a detailed investigation of the habitat selection criteria of lorises was conducted. Based on these observations habitat quality and availability in the Kottawa Arboretum for lorises was evaluated. The floral sampling results show that the Kottawa Arboretum has high species diversity. Climbers, which provide good substrate for lorises (Nekaris et al., 2005) were associated with more than 10% of the trees sampled in the study site. However, increased number of climbers indicates that the forest had been subjected disturbance (Ashton et al., 2001). Furthermore, the basal area values recorded are lower than the values expected from a primary forest which once again indicates that the area has been subjected to selective logging (Bhuyan et al., 2003). However, compared to Massmulla Proposed Forest Reserve, Kottawa Arboretum appears to be less disturbed due to two reasons. First, the basal area value is higher than the values recorded for Massmulla Proposed Forest Reserve (Nekaris et al., 2005). Second, no introduced plant species were recorded in the tree sample (n=240) of Kottawa Arboretum. Based on the tree usage by L. t. tardigradus it can be concluded that they show a higher preference towards trees such as Chaetocarpus castanocarpus and Dillenia retusa, which generally grow in disturbed forests. Analysis of habitat preferences of L. t. tardigradus showed a higher preference for small branches and twigs that are obliquely or horizontally oriented. Nekaris (2001) and Demes et al., (1990), argues that continuity of arboreal substrate is important for slender loris locomotion. An analysis of the three dimensional structure of the Kottawa Arboretum in terms of continuity of habitat and habitat characters such as type, orientation and girth of the available substrate indicates that the highest habitat quality and availability is in the height range 3.5m to 15m. This is consistent with the field observations where the highest number of loris sightings were made at this height range with the average height from the ground level where L. t. tardigradus were observed to occupy was 8.64 m ± 5.00. A few potential predators were also observed at the study site such as golden palm cat (Paradoxurus zeylonensis), rusty spotted cat (Prionailurus rubiginosus), fishing cat (Prionailurus viverrinus), brown fish owl (Ketupa zeylonensis), forest eagle owl (Bubo nipalensis) and Indian python (Python molurus).Even though the size of the estimated population of western Ceylon slender Loris, L. t. tardigradus at Kottawa Arboretum is small, the high density observed indicates that the habitat quality is very high. Thus Kottawa Arboretum can be considered as an important site for the conservation of western Ceylon slender Loris, L. t. tardigradus. However, at present Kottawa Arboretum exists as a small isolated forest patch as the Galle-Udugama road separates it from closest large forest track, the Kombala Kottawa forest reserve. Thus long term conservation of this population may require linking of Kottawa Arboretum with the Kombala Kottawa forest reserve through a suitable land use type. 167 ACKNOWLEDGEMENTS We wish to acknowledge K.A.I. Nekaris for providing the necessary literature and equipment for the survey as well as her helpful advice and guidance throughout this study, Lilia Bernede for providing valuable literature and comments, Wasantha Liyanage for invaluable assistance provided in the field, Department of Forest Conservation for granting permission to conduct the study, National Science Foundation Grant Number NSF/RSP/UOR/A/02/2001 for providing the necessary financial support. REFERENCES Ashton, P.M.S. & Gunatilleke, C.V.S. (1987). New light on the plant geography of Ceylon I. Historical plant geography. Journal of Biogeography. 14: 249-285. Ashton, P.M.S., Gunatilleke, N., Zoysa, M.D.De, Dassanayake, N., Gunatilleke, I.A.U.N. & Wijesundera, S. (1997). A field Guide to the Common Trees and Shrubs of Sri Lanka. W.H.T. Publications Ltd, Colombo. Ashton, P.M.S., Gunatilleke, C.V.S., Singhakumara, B.M.P. & Gunatilleke, I.A.U.N. (2001). Restoration pathways for rain forest in southwest Sri Lanka: review of concepts and models. Forest Ecology and Management 154: 409-430. Bhuyan, P., Khan, M.L. & Tripathi, R.S. (2003). Tree diversity and population structure in undisturbed and human-impacted stands of tropical wet evergreen forest in Arunachal pradesh, Eastern Himalyas, India. Biodiversity and Conservation. 12: 1753-1773. Brookes, T.M., Mittermeier, R.A., Mittermeier, C.G., Fonseca, G.A.B.Da, Rylands, A.B., Konstant, W.R., Flick, P., Pilgrim, J., Oldfield, S., Magin, G. & Hilton-Taylor, C. (2002). Habitat loss and extinction in the hotspots of biodiversity. Conservation Biology. 16: 909-923. Charles-Dominique, P. & Bearder, S.K. (1979). Field studies of lorisid behavior: Methodological aspects, in Doyle et al. (eds), The Study of Prosimian Behavior. Academic Press, New York. 5629-5667. Demes, B., Jungers, W.L. & Nieschalk, U. (1990). Size and speed related aspects of quadrupedal walking in lorises: a comparison of gait characteristics and locomotor stresses in Loris tardigradus and Nycticebus coucang, in Jouffroy et al. (eds), Gravity, Posture and Locomotion in Primates II. Sedicesimo Publishers, Florence.175-197. Groves, P.C. (2001) Primate Taxonomy. Smithsonian Institution Press, Washington D.C. Gamage, S.N. (2005). A comparative study on Biodiversity of selected manmade and Natural habitats in low country wet zone of Sri Lanka. MPhil thesis. Faculty of Agriculture, University of Ruhuna. Kamburupitiya Sri Lanka. Gunatilleke, I.A.U.N., Gunatilleke, C.V.S. & Dilhan, M.A.A.B. (2005). Plant biogeography and conservation of the southwestern hill forests of Sri Lanka. The Raffles Bulletin of Zoology, Supplement No. 12: 9-22 IUCN, (2004). Redlist. Available at: < URL: http://www.redlist.org Myers, N., Mittermeier, R.A., Mittermeier, C.G., Fonseca, G.A.B.Da & Kent, J. (2000). Biodiversity hot spots for Conservation Priorities. Nature. 403: 853-858. 168 Nekaris, K.A.I. (2001). Activity budget and positional behavior of Mysore slender loris (Loris tardigradus lydekkeriaus): implications for “slow climbing” locomotion. Folia Primatologica 72: 228-241. Nekaris, K.A.I. (2003). Observations on mating, birthing and parental care in three subspecies of slender loris in the wild (Loris tardigradus and Loris lydekkerianus) Folia Primatologica. 72: 228-241. Nekaris, K.A.I. & Jayewardene, J. (2003). Pilot study and conservation status of the slender loris (Loris tardigradus and Loris lydekkerianus) in Sri Lanka. Primate Conservation. 19: 83-90. Nekaris, K.A.I. & Jayawardena, J. (2004). Survey of Slender loris (Primates, Lorisidae Grey, 1821: Loris tardigradus Linnaeus, 1758 and Loris lydekkerianus Cabrera, 1908) in Sri Lanka. Journal of Zoology. 262: 327-338. Nekaris, K.A.I., Liyanage, W.K.D.D. & Gamage, S.N. (2005). Influence of forest structure and composition on population density of the red slender loris Loris tardigradus tardigradus in Masmullah proposed forest reserve, Sri Lanka. Mammalia 69(2): 201-210. Mill, R.R., (1995). Regional overview: Indian Subcontinent, in Davis et al. (eds), Center of Plant Diversity: a Guide to Strategy for their Conservation. Volume 2: Asia, Australia and the Pacific. World Wildlife Fund for Nature and the IUCN Press, Cambridge: 62-135. Osman Hill, W.C. (1953). Primates. Comparative Anatomy and Taxonomy. I. Strepsirrhini. Edinburgh University Press, Edinburgh. Pemadasa, A. (1996). The green mantle of Sri Lanka. Tharanjee Prints, Nawinna, Maharagama, Sri Lanka. 242 p. Sutherland, J.W. (1996). Ecological census techniques. Cambridge University Press.UK. 336 p. 169 SOCIAL RELATIONSHIPS OF WILD JUVENILE ASIAN ELEPHANTS (Elephas maximus) IN THE UDAWALAWA NATIONAL PARK, SRI LANKA D. JAYANTHA1 *, P.N. DAYAWANSA1, U.K.G.K. PADMALAL2, W.D. RATNASOORIYA1 and J.A. WEERASIGHE3 1 2 Department of Zoology, University of Colombo Department of Zoology, The Open University of Sri Lanka 3 Department of Wildlife Conservation, Colombo 7 ABSTRACT Social relationships of juvenile elephants (3-6 years old) in the Udawalawa National Park were studied. Focal animal sampling was employed to quantify behaviour of 450 individuals. Nearest neighbour (NN) and nearest neighbour distance (NND) were recorded to analyse social relationships. Adult females and juveniles were the NN of the study group during 50.67% and 37.55% of the total observed time respectively. The mean NND was 1.62m (SD±2.8) while it was less than 5m in 98% of the time. 33% of the time the study group was touching (NND<1m) the NN. There was a significant difference between NND categories (p<0.05). Being the NN, 80% of the infants stayed at a touching distance and was cared or allomothered by the juveniles under discussion. Time allocated for different behaviour pattern by the study group varied with the NN. When the study animals were accompanied by age-mates, they spent 17% of time for social playing and another 3% for non-play social contacts. It was only 1% for each behaviour pattern when the adult females were in close proximity. Maximum social contacts were observed between study animals and infants. The findings suggest that juvenile elephants more frequently associate adult females and near–age mates while they show social relationships in a varying degree with different associates. High play and social contacts of juveniles provide a great opportunity to develop skills and social confidence well needed for the survival in future. KEY WORDS: Social Relationships, Juvenile Elephants, Udawalawa National Park, Sri Lanka INTRODUCTION Mammals that live long have a lengthy childhood and adolescence. It is perceptible extended childhood is necessary for learning and development of the growing animal. The fact is observed in the elephant society, which is one of the most advanced mammalian social organizations (Sukumar, 2003). Young elephants spend long years in physical and behavioral development and diverse behaviours exhibited by adult elephants reflect their long history of social interactions and learning (Sukumar, 2003). Social interactions or social relationships among elephants are maintained by communication, behaviour and proximity (Sukumar, 2003). Nearest neighbour (NN) of an individual and distance to the nearest neighbour (DNN) are some of the parameters of proximity (Garai, 1997). Elephants show particular relationships with group members; for an example, juvenile African elephants under eight years old in the Amboseli national park, Kenya were within 5m distance from their mothers 80% of the observed time (Lee, 1986). In spite of biological need of suckling in young calve, juvenile interactions with their associates provide an opportunity for learning and thereby cognitive and motor skills of the individual is improved (Sukumar, 1994). 170 Different behaviours expressed by social animals also contribute for the long run learning process. Playing is one such behaviour pattern restricted to higher form of life and frequently observed in juvenile age. It is a training programme for the strength and skills that younger will require in adult life (Morris, 1990). Sukumar (1994) states that playing helps to accelerate the development of brain and nervous system. Many of other interactions describe social relationships of elephants. Juvenile as well as adult animals communicate by entwining trunks and inserting trunk tip to each others mouths when they are under stress. A calf would show the same behaviours to learn about feeding from its mother. Continual contacts between mother and calf reassure the psychological well being of the growing animal Sukumar (2003 & 1994). Such interactions among elephants invariably increase the social bonding which will help in better survival. In this context, it is important to study the way juvenile Asian elephants establish relationships with its group members. This will help to understand the social skillfulness and confidence of a particular juvenile compared to age mates which will then explain the social health of the animal. A study was conducted in the Udawalawa national park, Sri Lanka with the objective of describing social relationships of juvenile elephants METHODOLOGY Study area The Udawalawa national park (UNP) is in the intermediate zone of the Southern Sri Lanka (N 060 24’-060 35’ E 0800 45’- 0810 00’) and currently it has 30, 821 ha of scrublands, grasslands and dry-mixed evergreen forests as dominating vegetations (Department of Wildlife Conservation, 2005). Extensive areas of Panicum maximum dominated savanna type grasslands (Jayantha et al, 2005) have been resulted from shifting cultivation practiced before declaration in 1972. Along with seasonal grasslands adjoining the Udawalawa reservoir, elephants of UNP heavily utilize savanna grasslands. The authors believe that UNP harbors a healthy breeding population of elephants exceeding 500 individuals (Jayantha & Dayawansa, 2006). Study group Juvenile animals of estimated age 3-6 years old were observed in the study. This age group is partially parallel to ‘young juveniles’ as explained by Santiapillai in 2004. Based on Sukumar, (1994) the elephant population in the UNP was categorized into eight different groups in the field level. (Figure 1) 1. INF: Infants – animals of shoulder height up to the level of ventral abdomen of an average adult female; approximately day 1-1 ½ years old. 2. JVI: Juveniles (Class I) - shoulder height varying between ventral abdomen and neck level of the adult female; approximately 1 ½ -3 years old. 3. JVII: Juveniles (Class II) - shoulder height varying between neck and eye level of the adult female; approximately 3-6 years old. 4. JVIII : Juveniles (Class III) - shoulder height varying between eye and dorsal canthus of ear opening of the adult female; approximately 6-10 years old. (Juvenile males would be slightly taller than juvenile females of the same age) 171 5. SAF : Sub adult females - shoulder height varying between dorsal canthus of ear opening and shoulder level of the adult female; approximately 10-12 years old. 6. SAM: Sub adult males - shoulder height is just below or as the same that of the adult female; approximately 10-15 years old. 7. AF: Adult females – grown females pregnant, lactating or weaned; appearance of mammae (whether suckled or not) is considered when needed in deciding this. 8. AM: Adult males – grown males of shoulder height more than that of an average adult female; approximate age more than 15-20 years. Figure 1. Relative height of young female elephants in relation to full grown female. (Sukumar, 1994) Study protocol Selected animal group was observed in UNP from April 2004 to March 2005. Focal animal sampling and continuous recording (Martin & Bateson, 1993) were conducted every other week to quantify behaviour of the juveniles encountered at 450 different occasions. Total sampling time was 4500 minutes. Hides and distant observation (using 8x40 binoculars) were employed to minimize Hawthorn effect (Jayantha & Dayawansa, 2006). Nearest neighbour (NN) of the focal animal and distance to the nearest neighbour (DNN) was recorded together with different behaviours expressed. Descriptive statistics was used to describe the findings. 172 RESULTS 1. Nearest Neighbour Frequency Juveniles of age 3-6 years old were observed almost half of the time (50.67%) with adult females, possibly their mothers. They spent 37.55% of the time collectively with juveniles showing no affiliation towards a particular age or size group. Sub adult females and sub adult males were the NNs of the focal group nearly at the same frequencies, 4.22% and 4.00% respectively. Infants were seen near to the study group 3.11% of the total occasions and it was the adult males that made least NN frequency (0.44%) (Figure 2). 0.44% 3.11% 10.44% INF JV I JV II 14.22% JV III 50.67% SAF SAM AF 12.89% AM 4.22% 4.00% Figure 2. NN frequencies of different categories as a proportion of the total encounters. 2. Nearest Neighbour Distance Mean NND was 1.62m (SD±2.8, range 0-40m). Mean NND for different NN categories varied significantly (ANOVA, one-way, p<0.05). A clear pattern of mean NND was evident when young animals (0-10 years old) considered as the NN collectively; with increasing age (and body size) NND increased (Figure 3). Out of total encounters, at 98% times the NN was within 5m away from the focal animals and during 33% of the encounters they were touching the NN (NND<1m). Out of their total associations with the study group, infants stayed in close proximity (NND<1m) 80% of the time. It is <2% of the observations that study group stayed between 5-10m away from the NN and stayed very rarely at a distance >10m (Figure 4). 3.5 Mean NND (m) 3 2.5 2 1.5 1 0.5 0 -0.5 INF JV I JV II 173 JV III SAF SAM AF NN category Figure 3. Mean NND of different NN categories AM % frequency of NN distance 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% >10m 6-10m 1-5m <1m INF JV I JV II JV III SAF SAM AF AM NN category Figure 4. Percentage frequency of NND in relation to NN categories 3. Social behaviours in relation to the Nearest Neighbour The juveniles under discussion expressed different behaviour patterns; only three patterns were directly related to current context i.e. social play, non-play social contacts and agonistic interactions. Since only four occasions were recorded (hit by adult females twice, hit a sub adult female and a Class II juvenile) for the total of 450 encounters, agonistic behaviour pattern is not considered here. Several elements were described under social play and non-play social contacts (Annexure 1).Time allocated for each pattern calculated as a proportion of the total time the focal animals spent in association of different NN categories (Figure 5). Propotion of time spent 0.18 0.16 0.14 0.12 0.1 PLAY SOCIAL 0.08 0.06 0.04 0.02 0 INF JVI JVII JVIII SAF SAM AF AM NN category Figure 5. Proportion of time spent in relation to NN category 174 Social play is generally high when the study group had juveniles as the NN; the time spent was maximum with the age mates (17%). Non-play social contacts were high between infants and the study animals. Focal animals shared the same lesser amount of time for social play and non-play social contacts when the NNs were sub adult females, sub adult males and adult females. They had no social contacts with adult males. DISCUSSION Young juveniles of elephants spend more time with the adult females, particularly with their mothers (Kurt, 2002; Lee, 1986; McKay, 1973 & Sukumar, 2003). The fact is further emphasized by the current study. The association must have more relationships in social context other than the mere biological need of suckling as only six encounters were there on suckling focal animals during the study period. Apart from that, they had substantial associations with near-age juveniles (1-10 years old) next to the adult females proportionately. This can be clarified by ‘peer socialization’, during which much of the juvenile contacts made with members of the group other than the mother (Sukumar, 2003 & Moss, 1998). It was observed during the study that juveniles of 1-10 years old forming social groups within the cow-calf units. Infants were also recorded to accompany the study group parallel to Lee’s (1983) and Moss’s (1998) observations of how juvenile females accompany younger siblings. The term allo-mothering (Santiapillai, 2004) would describe this association when the juveniles take care of its younger. Gunawardene et al reported in 2004 that nursing infants stay with their mothers 100% of the time; it is worthwhile to note the possible overlapping of age & size categories of the two studies. The focal animals stayed in close proximity with their nearest neighbours. However, the mean nearest neighbour distance for different age & size categories varied showing a particular trend among growing animals. From infant to sub adult age or size, the distance increased gradually. This observation indirectly explains how young elephants move away from their nearest neighbours with increasing age and explore its environment (Sukumar, 2003 & Gunawardene et al, 2004). According to Garai (1997), close proximity of juvenile elephants is a sign of less social confidence to spend a solitary or independent time. Out of the social relationships of a juvenile, playing is an important aspect of learning. Playing in general is acrobatic (primates), exploratory (felids) or social ((Morris, 1990). Social play in juvenile elephants allows them to recognize kin and formation of social bonds useful in future (Sukumar, 2003). Juveniles of 3-6 years old spent most of their play time with near-age animals. The same observations have been made on Amboseli elephants (Lee, 1986). With infants, they allocated more time for non-play social contacts which supports the association of allomothering. Focal animals had less social interactions with sub adults and adult females compared to young animals. 175 CONCLUSIONS Close proximity of juvenile elephants of 3-6 years old with the adult females and near- age juveniles coincides with social interactions. Majority of the time they stayed within 5m distance from their nearest neighbours showing a less social confidence to do so. Social play and non-play social contacts were the behaviour patterns of interest regarding social relationships. The first pattern was frequent among near-age juveniles and the second was predominantly between infants and the juveniles under discussion. Young animals form juvenile groups of near-age members and young juveniles sometime play allo-mothering role for infants. Play and other social interactions experience by juvenile elephants would help in acquiring social skills expressed in later life. ACKNOWLEDGEMENTS The authors extend their sincere gratitude to the Department of Wildlife Conservation, especially to the staff, Udawalawa national park. The Born Free Foundation, UK is acknowledged for the financial support. REFERENCE Department of Wildlife Conservation, Sri Lanka. (2005). Management Plan- Udawalawa National Park. Garai M.E.(1997). The development of social behaviour in translocated juvenile African elephants, Loxodonta africana (Blumenbach). Ph.D. Dissertation, University of Pretoria. Gunawardene, M.D., Jayasinghe, L.K.A., Janaka, H.K., Weerakoon, D.K., Wickramanayake, E. and Fernando, P. (2004). Social Organization of Elephants in Southern Sri Lanka. In Jayawardene, J. (Ed.). Endangered elephants; past, present & future. Biodiversity & Elephant Conservation Trust. 66 p. Jayantha, D. & Dayawansa, P.N. (Eds.) (2006). Airavana. Serial Publication of the of the project ELEMONI-SL; Juvenile elephant monitoring project, Udawalawa National Park. University of Colombo & Department of Wildlife Conservation, Sri Lanka. 1(1): 5. Jayantha, D., Dayawansa, P.N., Padmalal, U.K.G.K, Ratnasooriya, W.D. (2005). Behaviour of Juvenile Asian Elephants in Panicum maximum Dominated Grasslands in the Udawalawa National Park. Proceedings of the 10th Annual Forestry and Environment Symposium. 53 p. Kurt F. (2002). Physical and social development in captive-born and orphaned Asian elephants of the Pinnawela Elephant Orphanage (Sri Lanka). In Proceedings of Workshop on Captive Elephant Management. Trichur, India. Lee, P.C. (1986). Early social development among African elephant calves. National Geographic Research. 2: 388-41. Martin P. & Bateson P. (1993). Measuring Behaviour: An introductory guide. 2nd Ed. Cambridge University Press. McKay G.M. (1973). Behaviour and Ecology of the Asiatic Elephants in Southeastern Ceylon. Smithsonian Contribution to Zoology. 125: 69. Morris, D. (1990). Animal Watching. Arrow Books Limited. 230 p. 176 Moss, C. (1988). Elephant Memories; Thirteen years in the life of an elephant family. The University of Chicago Press. Pp 145-174. Santiapillai, C (Ed.). (2004). Gajah; Journal of the Asian Elephant Specialist Group. 23:16. Sukumar, R. (2003). The living Elephant; Evolutionary ecology, behaviour and conservation. Oxford University Press. Pp 125-190. Sukumar, R. (1994). Elephant Days and Nights; Ten years with the Indian Elephant. Oxford University Press. 86-109. Survey Department of Sri Lanka. (2002). 1:50,000 topographical map of Sri Lanka. Sheets 76, 81 and 82. . 177 ADOPTION OF A WILDLIFE CONSERVATION PLAN BY CROP AND LIVESTOCK FARMS IN CANADA: WHAT FARMER AND FARM CHARACTERISTICS MAKE A DIFFERENCE? J.M.U.K. JAYASINGHE 1* and A. WEERSINK 2 1 Department of Agribusiness Management, Faculty of Agriculture and Plantation Mgt. Wayamba University of Sri Lanka, Makandura, Gonawila (NWP), Sri Lanka. 2 Department of Food, Agricultural & Resource Economics, University of Guelph, Guelph, Ontario, Canada, N1G 2W1. ABSTRACT Management Survey (2001) conducted by Statistics Canada and Agriculture and Agri-Food Canada. The target population consists of 21,000 active farms in Canada with sales greater than $10,000. The farms responded to the survey (Nt = 16,053 with 76.4% response rate) were classified into three major categories: (1) “crop farms” (Nc = 5,425), (2) “livestock farms” (Nl = 2,250) and (3) “mixed farms” (Nm = 8,378) with both crops and livestock. The results indicate that rate of This paper examines the impact of various farmer and farm characteristics on the adoption of a Wildlife Conservation Plan (WCP) – “a formal written document prepared by an expert that describes the measures to be taken by an agricultural operation to conserve natural land and wildlife habitants adjacent to it” - by crop and livestock farms in Canada. Those characteristics considered in the analysis include: human capital (age, sex), financial (profits, non-farm income, farm assets), farm structure (size, ownership), and social (degree of urbanization, population density). It uses data collected in the Farm Environmental adoption of WCP is comparatively less (13.9%) as compared to others, including manure, fertilizer, pesticide, water, and grazing management plans. The results from a Logit Regression analyses suggest that age, profitability, farm size, and degree of urbanization affect significantly on this behaviour in all farm types, however with varied size and signs. It highlights the importance of taking into account of voluntarily private-action of the farming community to formulate publicregulation aiming an environmentally friendly and conservative agriculture farm setting. KEY WORDS: Crop and livestock farms, Environmental management systems (EMS), Wildlife conservation plan(WC) 178 INTRODUCTION Other than short and long-term financial performance of the firm, the management of which may be provoked by moral concerns for quality of the environment where the firm is located. Consequently, the management might decide to adopt various Environmental Management Systems (EMS) – an environmental-friendly production practice that documents a firm’s activities that affect environmental performance – as a guide to reducing its ecological impact. There is no exception to this with respect to the firms operate in agriculture sector in Canada. The Farm Environmental Management Survey (FEMS) conducted by the Statistics Canada in 2001 in collaboration with the Agriculture and Agri-Food Canada identified a number of such EMS adopted by farms operate in this sector, including nutrient management plans (NMP), fertilizer management plans (FMP), pesticide management plans (PMP), water management plans (WMP), wildlife conservation plans (WCP), grazing management plans (GMP), and nutrient management plans (NMP) etc. The basic activities carried out by the management of farms that implemented these EMS are summarized in Table 1. A firm’s decision to invest its scarce resources (that possess higher opportunity costs) on the adoption of a particular EMS may be a result of, from one hand, the direct and/or indirect private or social benefits that it can “obtain by adoption” and/or will be “forgone by non-adoption”. The direct and/or indirect private or social costs that are “incurred by adoption” and/or can be “skipped by not-adoption” may also have a significant impact on this behaviour, on the other. Table 1. Definitions for various environmental management systems (EMS) Type Definition MMP Explains the types of liquid, solid/semi-solid manure storage systems use (e.g. unlined lagoon, open tank, sealed, covered tank etc.), frequency of storage and use of manure; specific treatments use (e.g. aeration, additives, separation, drying etc.), and odor control systems etc. FMP Explains the measures use to apply fertilizer (e.g. broadcasting, banded, post-plant top/side dressing etc.); mix of legume and chemical fertilizer to be used in each season, and their frequencies etc. PMP Explains certain information with respect to different application strategies of herbicides, insecticides, and fungicides; sprayer calibration techniques, and alternative methods other than chemical pesticides to control weeds, insects ad diseases etc. WMP Explains the sources and total volume of water to be used on a per acre basis; methods use to irrigate the land (e.g. sprinkler, drip, surface flooding etc), and ways and means of domestic water testing etc. WCP Explains any measures taken to conserve natural land and wildlife habitants that are adjacent to the agricultural operation (e.g. livestock fencing, cultivation of perennial forage, trees, bushes etc). GMP Explains any measures taken to conserve natural wetlands including rotational grazing for livestock and practices such as “carry-over” and “re-seeding”. NMP Explains the methods of testing nutrient content of the farm’s liquid or solid/semi-solid manure before applying it to the land; consideration of nutrient carry-overs; distance to water ways, and timing of applications etc. 179 Source: Agriculture and Agri-Food Canada One of the direct and private benefits to the firm by adopting an EMS may be, for example, higher revenue earned through increased market share or price premiums. Another private benefit may be, although rather indirect, that an EMS provides a credible signal to its existing and potential customers that it is an “environmentally-friendly” firm. As reported by Henriques and Sadorsky (1996), many Canadian firms adopt environmentally friendly quality management systems to improve its public image and reputation with the community. In terms of costs, a firm may concern about conserving the factors of production used in its day-to-day operations (i.e. reduction of direct costs) and/or reducing the waste generated in such activities by adoption of a particular EMS. Costs could be further reduced as an EMS by means of lower interest rates charged by financial institutions; lower premiums charged by insurance companies, and lower liability risks (e.g. compensation, legal fees) by minimizing the risk of involvement with the judiciary to solve the cases related to the environmental quantity (Khanna and Anton, 2002). According to Wall and Weersink (2001), an EMS is used in many developed countries as evidence of due diligence which is often the only acceptable defense in a legal challenge stemming from an environmental accident. Eventually, a firm’s motivation for adopting EMSs, individually or collectively, without investing such resources on other competitive ends will be determined by the size of the margin between these benefits and costs to the adopter (JayasingheMudalige and Weersink, 2004), which may be highly subjective to the characteristics of the firm and the entrepreneur (Buchanan, 1969). There are a number of studies that examined empirically the motives for adoption of certain agri-environmental practices by various types of agricultural operations in developed countries. Many of these studies have been focused on the factors affecting the adoption of such practices that generate direct and or indirect “private” benefits to the farmer (see, for example use of specific vaccines by cattle producers by Bhattacharyya et al., 1997; various types of fertilizers and pesticides by Smith and Smithers, 1992). In consequently, there exists a gap in the environmental economics literature, with respect to the studies that examined the impact of various human and socio-economic characteristics associated with farmers and of farms on the adoption of individual Best Management Practices (BMP) and/or certain EMS that possess “public goods characteristics”, for example a well-formulated and comprehensive plan that explains the actions the farm can undertake to protect the wildlife in and around the farmland (exceptions, include adoption of modern soil conservation techniques by Rahm and Huffman, 1984, and integrated pest management techniques by D’Souza et al., 1993). To the best knowledge of authors, this phenomenon was not examined empirically to date with respect to the farms operate in the agriculture sector in Canada. The purpose of this study is to examine the impact of various farmer and farm characteristics on the adoption of a wildlife conservation plan (WCP) by crop and livestock farms in Canada. Further, it contrasts and compares whether that behaviour of farmers is associated with their desire to adopt any other plans, including those primarily generate “private benefits” (e.g. fertilizer and pesticide management plans) and “social benefits” (e.g. manure and water management plans). 180 METHODS Theoretical framework A rational farmer will invest on adoption of an EMS if it helps her, directly or indirectly, to maximize the profits of the farm. Alongside, the farmer may also decide to adopt certain other EMS (e.g. WCP) in her farm and/or area adjacent to the farmland that generates greater social benefits than the private benefits. In light of this, for the farms that involved with production of crop and livestock in Canada we can hypothesize that “the motivation for the management of the farm to invest its scare resources on a WCP that generates greater social benefit (in compared to the private benefits) will depend on the human-capital (e.g. age, sex), financial (e.g. profits, non-farm income), operational and structural (e.g. size, ownership), and social (e.g. distance, population pressure) characteristics of the farmer and the farm”. The following theoretical model was specified to test this hypothesis: WCPi = α0 + Σβij Xij + εi Where, WCPi describes whether the farm in question has adopted a WCP or not (i = 1, 2, …n). Xij is a vector of j explanatory variables included in the model. The terms βij denote regression coefficients corresponding to the explanatory variables (j = 1, 2, …l). Further, α0 and εi denote the intercept and random error terms, respectively. Data collection and analysis The secondary data included in the FEMS, which was a voluntary national survey focusing on the level of adoption of environmental management systems (EMS) and best management practices (BMP) by livestock and crop operations in all Provinces across Canada and on numerous other issues were used to estimate the coefficients of variables included in the model (Table 2). Table 2 . Variables used in the model 181 Variable Description (expected sign) Human capital characteristics AGE SEX TMA Age of the farm household head in years (+) Gender of the farm household head (Male = 1; Female = 0) Time allocation of the farm household head for farming (Full-time = 1; Part-time with off-farm work = 0) Financial characteristics PFT INC AST Overall profitability of the farm - calculated by taking the ratio of: total gross farm receipts of the operation in 2000 / total farm business operating expenses of the farm in 2000 (+) Income earned by the head of the farm household through non-farming operations, such as retail business and factory work (Non-farm income = 1; No non-farm income = 0) Total fixed assets of the farm – calculated by taking the total present market value of land, buildings, and farm machinery in thousands of Canadian dollars (+) Operational and structural characteristics Organizational arrangement # OSP Sole proprietorship = 1; Other arrangements (i.e. for partnerships / corporation) = 0 OFC Family corporation = 1; otherwise = 0 ONF Non-family cooperation = 1; otherwise = 0 Land ownership pattern OWN Ratio of own land to the total land area of the farm (+) LLG Ratio of land leased from the government to the total land area of the farm FSZ Area of the farm in hectares (+) Social and regional characteristics DIS PPD The distance in kilometers “as a crow flies” from the farm operation to the nearest Census Metropolitan Area (CMA) (+) The population density of the Census Sub-Division where the farming operation is located measured as the number of people per square km (-)## Notes: # A farm without any formal agreement with any partner was assumed to be the base scenario. ## The tendency to have a WCP decreases with increasing population density since there is low chance to have wildlife habitats in urbanized areas. All active farms with sales greater than C$10,000 as included in the “Agriculture Division’s Farm Register in Canada” (n = 21,000) were considered to be the target population for this particular survey. The response to survey was significantly high with 16,053 questionnaire were returned with 76.4 percent response rate 1. The data from the 2001 Census of Agriculture in Canada were also tied to the FEMS database to obtain more accurate, up to date and comprehensive data. For the purpose this analysis, farms that responded to the questionnaire were categorized into three major categories: (1) “crop farms” (NC=5,425), (2) “livestock farms” (NL=2,250), and (3) “mixed farms” (NM=8,378), which possesses both crop and livestock in the farm in various proportions. Considering the dichotomous nature of the dependent variable, a Logit Regression analysis was used to estimate the coefficients of variables explained in Table 2, for example: adoption of a WCP = 1; non-adoption of a WCP = 0. 1 The responses rate was high because two prominent government institutions in Canada, which have close connections with the country’s farming sector – the Statistics Canada, and the Agriculture and Agri-Food Canada, backed it. 182 RESULTS Descriptive statistics There were 8764 farms out of 16053 that responded to the survey (i.e. 55 percent) did not adopt any EMS. Another 11.7 and 8.6 percent of individual farms adopt a single and two EMSs, respectively. As shown in Figure 1, only 13.9 percent of farms possess a WCP, and which is only second to the percentage of farms that adopt a NMP. Type of Plan No Livestock Crop Mixed N = 2,250 N = 5,425 N = 8,378 % Ran No % k WCP 215 9. 4 6 65 12 3 .0 Ra N % nk o 5 13 1 65 6. Ra nk 6 3 MMP 309 13 2 .7 25 4. 5 7 6 21 2 87 6. 1 1 NMP 7 0. 7 3 80 14 4 .8 4 12 1 63 5. 7 1 FMP 66 2. 5 9 14 27 93 .5 1 21 2 63 5. 2 8 PMP 55 2. 6 4 14 27 78 .2 2 17 2 62 1. 5 0 WMP 261 11 3 .6 98 18 8 .2 3 18 2 38 1. 4 9 GMP 320 14 1 .2 15 2. 5 9 7 20 2 81 4. 8 183 3 PERCENTAGE 25 23.2 19.2 20 17.1 15.9 15 12.9 20.5 13.9 10 5 0 NMP WCP GMP MMP WMP PMP FMP Figure 1. Percentage of farms that possesses a WCP in the total sample The number and percentage of crop, livestock and mixed farms that adopt the seven types of EMSs are reported in Table 3. Table 3. Adoption of various types of EMS by farms in Canada: Source: FEMS database – Statistics Canada It shows that there were only 9.6, 12.0, and 16.3 percent of these farms respectively adopt a WCP. As a whole, mixed farms have the highest adoption rates in general across the seven EMSs considered while livestock farms have the lowest. The percentages of livestock farms with a FMP, PMP and NMP are lower than the WCP (i.e. 2.9, 2.4 and 0.3). We may suggest that the activities included in these EMS (see, Table 1) are “not important” to maximize the profit of a livestock farm, and consequently, the cost of adoption of which in these farms cannot be not justified. Although such activities included in a GMP are considered to be “very important” for livestock farms, a large difference with respect to the adoption rates of GMP and the WCP (i.e. 14.2 - 9.6) cannot be observed. In the context of crop farms, the percentages of MMP (4.7%) and GMP (2.0%) were less than the WCP (12.0%). In fact, like in the previous case, the adoption of MMP and GMP, individually or collectively, may not play a significant role in crop farms. To verify whether there is any overlap with respect to farmers’ understanding and interpretation of the specified tasks in each EMS and in turn to adopt it together with others, the correlations between each of the six different EMS considered in this analysis with the farms having a WCP across all farms responded to the survey were examined (Figure 2). The correlation coefficient exceeds 0.5 only in WMP indicating that farmers with “more or less same characteristics” are likely to adopt both wildlife conservation and water management plans. This indicates that a farmer who has desire to conserve wildlife in the area also takes action to preserve natural resources such as water. 184 Correlation Coeffcient 0.6 0.556 0.492 0.5 0.427 0.430 0.377 0.4 0.3 0.2 0.153 0.1 0 NMP MMP FMP PMP WMP GMP Figure 2. Correlation coefficient of WCP with other EMS: Estimates of coefficients The results from the Logit analysis that produce logged odds (logits) of parameters are reported in Table 4 for three models representing the farm types. All the models were significant at a level of 0.01. Further, the relatively higher Pseudo R-square values (0.7310, 0.7156 and 0.7240 for livestock, crop, and mixed farms, respectively) suggest that the models performed well. There are three variables developed to explain the effect of age of the farmer (AGE), overall profitability of the farm (PFT), and size of the farm (FSZ) were significant at the 1 percent probability level in all three models, and possess the expected sign. This indicates that as the farmer gets matured (i.e. age increases by one-year) and earns higher profits (i.e. the ratio calculated in this respect increases by one unit), and the size of the farm gets larger (i.e. increases by a onehectare) the logged odds (logits) of adoption of a WCP would be increased or decreased by the value of respective coefficients included Table 4, for example 0.892, 0.0675 and 0.0877 for crop, livestock and mixed farms, respectively for AGE (see, Borooach, 2002 and Pampel, 2000 for interpretation of results from Logit Regressions). 185 Table 4. Estimates of coefficients for livestock, crop, and mixed farms: Variables Estimates LIVESTOCK (N = 2250) CROP (N = 5425) MIXED (N= 8378) 0.0892*** (0.0083) -0.0147 (0.0292) 0.0974* (0.0422) 0.0675 *** (0.0042) -0.0396 (0.0211) 0.0635 (0.0475) 0.0877 *** (0.0072) -0.0233 (0.0192) 0.0887 (0.0496) PFT 0.1264 *** (0.0399) INC -0.2180 * (0.0732) AST 0.0497 ** (0.0201) Operational & Structural Characteristics 0.1206 *** (0.0469) -0.2221 * (0.0543) 0.0549 * (0.0204) 0.1254 *** (0.0479) -0.2197 * (0.0524) 0.0459 * (0.0209) OSP 0.0027 * (0.0013) -0.0365 (0.0182) -0.0022 (0.0741) 0.0588 ** (0.0281) -0.0745 (0.0998) 0.0344 *** (0.0089) 0.0024 ** (0.0014) -0.0378 (0.0203) -0.0031 (0.0731) 0.0479 ** (0.0171) -0.0621 (0.0741) 0.0445*** (0.0058) 0.0142 * (0.0087) -0.0069 ** (0.0032) 0.8475 *** (0.0366) 0.0139 ** (0.0078) -0.00612 ** (0.0023) 0.5673 *** (0.0374) Human Capital Characteristics AGE SEX TMA Financial Characteristics 0.0036 ** (0.0011) OFC -0.0344 (0.0191) ONF -0.0033 (0.0723) OWN 0.0489 *** (0.0155) LLG -0.1024 (0.0970) FSZ 0.0576 *** (0.0017) Social & Regional Characteristics DIS PPD Constant 0.0129 ** (0.0079) -0.0076 *** (0.0024) 0.9960 *** (0.0372) R–square 0.7310 0.7156 Note: ***, **, and * denotes the 1, 5, and 10 percent significant levels, respectively. The probability of adopting a WCP in all types of farms increases as the income of the farmer (INC) and the sustainability of the farm measured in terms of use of capital assets (AST) increases by a unit. Therefore, all the variables used to characterize the financial situation of farms characteristics, including the PFT, INC and AST had a significant impact on this behaviour. The results also suggest that farmers who “own” the most of their land were likely to adopt a WCP along with other EMS indicating their long-term commitment to their land as opposed to short-term interests. 186 0.7240 Both DIS and PPD were significant at various levels in all three samples. Majority of these also possessed the expected sign. As expected, farmers who remote to a major urban center (DIS) were likely to adopt a WCP suggesting that wildlife is a factor that affect farming in these areas. Similarly, adoption decreases with the population density of the region in which the farm is located (PPD) increases. CONCLUSIONS The outcome of the analysis suggests that a number of factors show a significant impact on farm’s decision to adopt a wildlife conservation plan to protect the flora and fauna in and around the farming environment. In abstract, the young and rich farmers with sufficiently large farms and assets tend to adopt a WCP. Although the percentage of adopting so is low in the farming population, those farms with a WCP is not significantly difference from the farms with other plans that generate higher private benefits, for example fertilizer and pesticide management plans. The results provide some useful insights into the formulation of effective public policy aiming environmental protection, in general, and conservation of wildlife, in particular. The conversion of natural wildlife habitats into profitable agribusiness ventures through the development of crop and livestock farms creates much profit to the farming community. However, such an action has a number of consequences, for example ever-increasing human-wildlife conflicts for limited resources such as water and place to live (similar to human-elephant conflicts in Sri Lanka), loss of biodiversity, and air and water pollution etc. to both farming and non-farming communities. In the wake of this understanding, a number of Municipalities in certain Provinces in Canada, for example in Alberta and Quebec, have made it mandatory to adopt some of these plans, for example NMP and MMP, in the farm. However, adoption of a WCP is not compulsory in any of the 10 Provinces in Canada. This highlights the fact that those farmers with a WCP in place possess positive incentives to behave environmentally friendly “voluntarily”. The respective governments should take into account of this factor to avoid, or at least to minimize, such bad effects arising from commercial farming systems to the environment. For example, depend on its needs, a particular Municipality can make it a policy to have a WCP in place and can promote farmers to adopt other plans such as FMP and PMP together with a WCP and/or a WMP. These way, whilst deriving “private benefits” through the implementation of former, farmers can be motivated to “pay some sort of compensation” to the society for utilization of the public goods such as wildlife habitats by implementing the later. At the same time, these institutions can have an appropriate program to capture those farmers who do not possess a any of these plans into the system in order to prevent that “voluntary” action of a farmer not become a barrier to be competitive in the marketplace. 187 ACKNOWLEDGMENTS The authors wish to express their gratitude to Mr. Mike Trant and Mr. Martin Beaulieu, Agriculture Division of the Statistics Canada for given permission to assess to the FEMS database and their continuous support during April to July 2004 to collect necessary information for this analysis. REFERENCES Bhattacharyya, A., T. R. Harris, W. G. Kvasnicka, and M. Veserat. (1997). Factors Influencing Rates of Adoption of Trichomoniasis Vaccine by Nevada Range Cattle Producers. Journal of Agricultural and Resource Economics 22(1): 174-190. Borooach, V. K. (2002). Logit and Probit: Ordered and Multinomial Models. Series: Quantitative Applications in the Social Sciences. No. 138. Thousand Oaks: Sage Publication. Buchanan, J. M. (1969). Cost and Choice: An Inquiry in Economic Theory. Markham. Chicago. D’Souza, G., D. Cyphers, and T. Phipps. (1993). Factors Affecting the Adoption of Sustainable Agricultural Practices. Agricultural and Resource Economics Review (October). 159-165. Henriques, I. and P. Sadorsky. (1996). “The Determination of an Environmentally Responsive Firm: An Empirical Approach”. J. of Environmental Economics and Management 30(3): 381-395. Jayasinghe-Mudalige, U. K., and A. Weersink (2004) “Factors Affecting the Adoption of Environmental Management Systems by Crop and Livestock Farms in Canada”, Sri Lankan Journal of Agricultural Economics, 6(1). Khanna, M. and W. R. Anton. (2002). “Corporate Environmental Management: Regulatory and Market-based Pressures”. Land Economics 78(4): 539-558. Pampel, F. C. (2000). Logistic Regression: A Primer. Series: Quantitative Applications in the Social Sciences. No. 132. Thousands Oaks: Sage Publications. Rahm, M. R. and W. E. Huffman. (1984). The Adoption of Reduced Tillage: The Role of Human Capital and Other Variables. American Journal of Agricultural Economics 66: 405-413. Smith, B. and J. Smithers. (1992). Adoption of Soil Conservation Practices: An Empirical Analysis in Ontario, Canada. Land Degradation and Rehabilitation 3: 1-14. Wall, E. and 188 A SURVEY OF AMPHIBIANS IN THE NILGALA FOREST RESERVE AND ITS VICINITY D.M.S.S. KARUNARATHNA 1, U.T.I. ABEYWARDENA 1, M.D.C. ASELA 1 and D.G.R. SIRIMANNA 1, L.D.C.B. KEKULANDALA2 1 The Young Zoologists’ Association of Sri Lanka, National Zoological Gardens, Dehiwala, Sri Lanka 2 IUCN – The World Conservation Union, No: 53, Horton Place, Colombo 07, Sri Lanka ABSTRACT This paper presents the findings of a preliminary study on Amphibian diversity in Nilgala Forest Reserve, in the Monaragala District, which is situated in the intermediate zone of Sri Lanka. Our aim was primarily to study the Amphibian species diversity and various threats they face in this remote, unstudied area. We were able to record 19 species of amphibians, which is about 18% of the total Sri Lankan Amphibians described to date. We were also able to record several species for the first time in Nilgala Forest Reserve. There were 6 (31%) endemic species and 5 (26%) nationally threatened species among them. Hence Nilgala Forest Reserve supports a high Amphibian diversity. However, this important forest is threatened by harmful human activities such as man made fire, illegal logging, chena cultivation and road kills. INTRODUCTION Sri Lanka and Western Ghats of India is a biodiversity hotspot, rich in herpetofaunal assemblages (Bossuyt et al., 2004; Meegaskumbura et al., 2002). Favorable environmental factors such as high rainfall and humidity, high density of undergrowth found in this region support a rich diversity of herpetofauna. Based on published sources, a total of 184 species of reptiles (Bahir & Maduwage, 2005; Bahir & Silva, 2005; Batuwita & Bahir, 2005; Das & De Silva, 2005; De Silva, 1996; Manamendra-Arachchi & Pethiyagoda, 1998; Manamendra-Arachchi & Pethiyagoda, 2001a; Manamendra-Arachchi & Pethiyagoda, 2001b) and 103 species of amphibians (De Silva, 1996; Dutta & Manamendra-Arachchi, 1996; Manamendra-Arachchi & Pethiyagoda, 2005; Meegaskumbura & Manamendra-Arachchi, 2005) have been recorded from the island to date. The poikilothermic nature of herpetofauna restricts their distribution to areas with high rainfall such as lowland wet zone forest areas and montane forest areas (Giri & Chaturvedi, 2001). According to published literature wet zone forests harbors more than 60% of the indigenous herpetofauna of Sri Lanka. Furthermore, high percentage of endemism can be seen in the southwest lowland forests where almost 90% of the endemic vertebrates are concentrated (Bambaradeniya et al., 2003; Senanayake et al., 1977; Wijesinghe & Dayawansa, 2002). However, herpetofaunal diversity in the dry and intermediate zone forest areas has not been studied well (De Silva et al., 2004). Reptiles and amphibians play a vital role in the functioning of an ecosystem. Their diet of insects and small animals contributes to control of pests in human habitats, including crops and homes (Knopf, 1998). Frogs are among the most misunderstood group of vertebrates. Hence many people treat them with almost the same revulsion as snake, despite their being ecologically important as pest controllers 189 (Manamendra-Arachchi, 2000). Certain herpetofauna species can be considered as indicator species of environmental change. Nilgala forest reserve (NFR) is one of the largest and important forest areas in Monaragala District. However, the amphibian fauna of NFR is poorly studied. There are several preliminary amphibian surveys carried out in NFR (De Silva et al., 2004; Hettige et al., 2000). The present survey attempted to document the amphibian fauna of NFR through extensive field surveys made in 2004 and 2005. One of the biggest drawbacks for conserving amphibian fauna of the country is the lack of knowledge of their distribution and ecology, since only a fraction of the Amphibian species present in different areas of the country is hitherto known to science. Hence, it is essential to gather information on the diversity of amphibian fauna in different areas of the country, as a first step towards conservation. This paper would contribute to enhance the current knowledge of amphibian diversity within the Nilgala Forest Reserve. MATERIALS AND METHODS Study area Nilgala forest reserve (NFR) is a unique forest ecosystem covering 12,432 hectares in the Bibile divisional secretariat division. According to the Gunatilleke & Gunatilleke (1990) the major vegetation type is lowland tropical dry mixed evergreen forest (Figure 13). Commonly found trees are Aralu (Terminalia chebula), Bulu (Terminalia bellirica), Nelli (Phyllanthus emblica). NFR is located between 7° 08` – 7° 14` Northern latitudes and 81° 16` – 81° 20` Eastern longitudes, approximately 11 km southeast from Bibile town. The altitudinal range varies from 200 m to 700 m above sea level within the boundaries of the NFR (De Silva et al., 2004). The general climatic conditions in the Nilgala area is described as moderately cool, which turns humid during the northeast monsoon season. The average annual rainfall is around 1750 mm, with most of the rainfall occurring from December to March, with occasional rains in some months. The weather gradually becomes very dry from August to December and highest temperatures are recorded in August. The mean annual temperature in the NFR is 28°C with maximum of 32°C and minimum of 24°C. NFR is also important as a major watershed for Gala Oya and Panmedilla Oya throughout the year. There are several peaks within the NFR and “Yakun Hela” the highest peak (700 m). Methodology The present study was carried out during the period of 2004 to 2005. A total of 56 days (10 hrs/ day) were spent for fieldwork during the two years covering the wet and dry seasons. General area surveys were carried out in different habitat types within the NFR. Surveys were conducted both day and night and flashlights were used at night. All habitats such as water bodies, under the rocks, logs and decaying vegetation, and arboreal amphibians in trees and bushes up to 5 m, were thoroughly searched for the presence of specimens. All collected species were examined carefully and noted down before released back to the same habitats. The diagnostic keys given by Dutta and Manamendra-Arachchi (1996) and Manamendra-Arachchi and Pethiyagoda (2005) were used for species identification. Furthermore basic environmental parameters were collected at locations, where specimens were collected. 190 Eleven habitat types were identified and sampled during the survey and a brief description of these habitats is given in (Table. 01). Table 1. Description of habitats in NFR. Habitat Type 01 Chena 02 Home Garden 03 Paddy field 04 Riverine forest 05 Road Side 06 Rock-outcrop 07 08 Savannah forest Shrub / Bush area 9 Small Pond 10 11 Stream Tank Description of Habitat Trees belong to family Rutaceae is dominant and grows up to 5 m, scattered bushes present, main cultivation is maize, banana and finger millet. Leaf litter is very low. Mixed cropping with woody plants like a Mangifera indika, Chloroxylon swietenia, Schleichera oleosa, Tamarindus indika, trees grows up to 15 m, shade is about 50%, and Leaf litter content is high and wet. Paddy fields are moderate in extent (about 1 acre), wallowing sites are frequent along the fields, field bunds are narrow. Shade 80% with large tall trees growing up to 20 m, Mangifera ceylanica, Maduca longifolia, Terminalia chebula Diospyros ebenum and Diospyros malabarica are the dominant species, thick wet leaf litter layer available, Shady Forest, decaying logs are common. Generally consist of small bushes growing up to 2 m. Species such as Maduca longifolia, Terminalia bellirica and Mangifera zeylanica can also be found in several areas. Large rock boulders and grassy areas with seasonally moist cascade habitats. Shade 20% with tall trees. Only Terminalia chebula, Terminalia bellirica and Phyllanthus emblica Forest. 1m to 2m Tall and randomly distribute with open soil. Seasonally flooded, mud pond, gem pits, Agricultural wells, drinking wells, mud pits, clay pits Perennial flowing water bodies, 1m to 10m wide, visibility high, and turbidity low. Open water bodies, covered by macrophytes (25 %) Figure 1. Paddy fields in near NFR. 191 Figure 2. Rockout crops (Yakun Hela) in NFR. Figure 3. Savannah forest in NFR. RESULTS AND DISCUSSION During the survey, 19 species of amphibians (Annex 01) belonging to four families were recorded from NFR. This includes 13 genera. There were 6 endemic and 5 nationally threatened species among them (IUCN Sri Lanka, 2000). Ranid frogs (dominated by Fejervarya limnocharis, Philautus regius, Euphlyctis cyanophlyctis, Euphlyctis hexodactyla) were the most abundant amphibian group in NFR according to the present survey (Figure 04), while Caecilians were the least abundant. When considering the primary mode of living there were nine terrestrial, four arboreal, four aquatic and 2 fussorial species. Several species were recorded for the first time in NFR and it is one of the significant findings of the present survey. They are Philautus regius and Philautus fergusonianus. Furthermore three unidentified species were also recorded during the survey, which probably include new species belonging to genus Nannophrys. 192 Ichth 5% Ranid 58% Bufon 11% Micro 26% Figure 4.Species composition of Amphibians families in NFR. When considering the proportional representation of species, the highest abundance was shown by Fejervarya limnocharis (13.44 %) and Philautus regius (10.67 %). The high abundance of Fejervarya limnocharis is possibly related to abundance of aquatic and semi aquatic habitats (Small ponds, streams and paddy fields) within the study area. Hence this species is one of the most common species (Table 02) encountered in the NFR. Philautus regius is also a common species associated with shrub vegetation. The shrub vegetation is very high in the NFR and provides the habitat for this species. It is interesting to note that this species was recorded for the first time in NFR and it is considered as rare in the dry and intermediate zones. Table 2. Species richness of amphibians in NFR. Scientific Name Bufo atukoralei Bufo melanostictus Kaloula taprobanica Microhyla ornata Microhyla rubra Ramanella variegata Uperodon systoma Hoplobatrachus crassus Fejervarya limnocharis Euphlyctis cyanophlyctis Euphlyctis hexodactyla Rana gracilis Sphaerotheca breviceps Sphaerotheca rolendae Philautus fergusonianus Philautus regius Polypedates cruciger Polypedates maculatus Ichthyophis glutinosus Total Number of Individuals 3 17 8 11 13 3 16 19 34 24 24 4 7 9 3 27 12 18 1 193 Proportional Representation 1.19 6.72 3.16 4.35 5.14 1.19 6.32 7.51 13.44 9.49 9.49 1.58 2.77 3.56 1.19 10.67 4.74 7.11 0.40 Figure 5. Rana gracilis Sri Lanka Wood frog. Figure 6. Kaloula taprobanica Common Bull Frog. Figure 7. Unidentified Nannophrys species in NFR. 194 Figure 8. Sphaerotheca rolendae Marbled Sand Frog. Figure 9. Philautus fergusonianus Ferguson’s shrub frog. When considering the proportional representation of species in each habitat type (Figure 10), the highest species richness occurred in home gardens (21.05 %) followed by Roadside habitats (19.74 %) and Riverine forests (13.16 %), while Rock outcrop habitats showed the lowest species richness (1.32 %). The high species richness in the home gardens might be due to several reasons. The high amount of leaf litter and shade and also the availability abundant of food items such as insects attracted to electric light would have been the reasons for high species richness in home gardens. Furthermore the sampling time allocated to home gardens was significantly higher than the time allocated to other habitats. This would have lead to the record of more species from home gardens than other habitat types. Nevertheless, these home gardens with large number trees, high shade and high amount of leaf litter, in the vicinity of NFR is vital refuge for amphibian fauna. It appears that a higher species richness of amphibians occur in disturbed habitats such as home gardens, chena and roadside shrub, and this may be due to the ‘edge effect’ phenomenon. The species richness is also high along the road side (On Bibila – Ampara main road). This may also due to the increased sampling effort in this particular habitat type. The lowest species richness (1.32%) was shown by Rock Outcrop habitats. These rocky outcrops is exposed to the sun hence the high temperature is not suitable for amphibians (Pough et al., 2004). 195 Proportional Representation of species 25.00 20.00 15.00 10.00 5.00 0.00 na a he e C ar ut st ko re oc fo R ah nn st va re Sa fo e in er iv R ld fi e y dd h us Pa /B b ru Sh e id S d oa R am re St G nk d Ta on lP al n de ar e om Sm H Habitat Types Figure 10. Species richness of Amphibians in various habitat types in NFR. Threats During the survey period several threats to the amphibian fauna in the NFR was observed and recorded. These include forest fires, logging, and extensive use of chemicals for agriculture, forest clearing for chena cultivation and road kills. People living around the NFR frequently make fire to clear the underbrush, prepare the ground for the next cultivation cycle and to hunt animals. These fires are very frequent in the months of August and September. This activity destroys the habitats of amphibians. Illegal logging activities take place around and within the NFR and this seriously affect the quality of the forests and in turn its inhabitants. The local communities are involved in paddy and chena cultivation in the vicinity of the NFR. These people use chemical fertilizers, pesticides and weedecides. This extensive use of chemical is a threat to the many animals including amphibians. Another significant threat to the amphibian fauna of the area is road kills. Bibila – Ampara main road is cutting across the NFR and many animals subjected to road kills. The team has observed and recorded that amphibian mortality is particularly high on this road especially after a rain. The team has carried out a separate survey to document the impacts road kills. 196 Figure 11. Forest fire inside the NFR. Figure 12. Road kill (Microhyla ornata on Bibile – Ampara Main Road within NFR) 197 Figure 13. Map of the Nilgala Forest Reserve and Roads. CONCLUSION This preliminary investigation of the amphibian fauna of the Nilgala Forest Reserve and its vicinity clearly shows that NFR is an important location in terms of amphibian diversity. It also evident that NFR act as an important refuge for threatened amphibians, in the intermediate zone belonging to the Uva province. The NFR and its surrounding habitats have not been studied well and it is evident from the fact that 6 species were new sight recordings for NFR. Therefore detailed studies have to be carried out to document the amphibians in this site. There are several human activities that have a negative impact to the amphibian fauna as well as to other faunal and floral groups such as fire, logging, gem mining and road kills. Therefore suitable mitigatory action has to be initiated to conserve Nilgala Forest Reserve and its surrounding habitats. 198 ACKNOWLEDGEMENTS The authors wish to thank Dr. Channa Bambaradeniya (IUCN – The World Conservation Union) for reviewing the manuscript. We would also like to thank Mr. Mendis Wickramasinghe (IUCN – The World Conservation Union) and Mr. Kelum Manamendra-Arachchi (WHT) for sending valuable literature. Our heartfelt Thanks goes to the following persons for their intimate support; Mr. Naalin Perera, Mr. Dilup Chandranimal, Mr. Sarath Ekanayake, Mr. Sampath Gunatilleke, Mr. Roshan Rodrigo, Mr. Prasanna Samarawickrama, Mr. Sandun Perera, Mr. Suranjan Fernando and Mr. Nalinda Peiris, Mr. Vimukthi Weerathunga, Mr. Pradeep Samarawickrama (IUCN – The World Conservation Union). Finally, we thank Mr. Chamila Soysa, Mr. Toshan Peiris, Mr. Panduka Silva, Mr. Asanka Udayakumara, Mr. Anushka Kumarasinghe, Mr. Dimuthu Wickramasinghe, Mr. Thasun Amarasinghe and Mr. Devaka Jayamanna (YZA – Young Zoologists’ Association) for his kind help during the field visit and other activities in Nilgala Forest Reserve. REFERENCE Bahir, M.M. & Maduwage, K.P. (2005). Calotes desilvai, a new species of agamid lizard from Morningside Forest, Sri Lanka. In: Yeo, D.C.J.; Ng, P.K.L. & Pethiyagoda, R (editors), Contributions to biodiversity exploration and research in Sri Lanka. The Raffles Bulletin of Zoology. 12: 381-392. Bahir, M.M. & Silva, A. (2005). Otocryptis nigristigma, a new species of agamid lizard from Sri Lanka. In: Yeo, D.C.J.; Ng, P.K.L. & Pethiyagoda, R (editors), Contributions to biodiversity exploration and research in Sri Lanka. The Raffles Bulletin of Zoology. 12: 393-406. Bambaradeniya, C.N.B.; Perera, M.S.J.; Perera, W.P.N.; Wickramasinghe, L.J.M.; Kekulandala, L.D.C.B.; Samarawickrema, V.A.P.; Fernando, R.H.S.S. & Samarawickrema, V.A.M.P.K. (2003). Composition of faunal species in the Sinharaja world heritage site in Sri Lanka. Sri Lanka Forester. 26: 21-40. Batuwita, S. & Bahir, M.M. (2005). Description of five new species of Cyrtodactylus (Reptilia: Gekkonidae) from Sri Lanka. In: Yeo, D.C.J.; Ng, P.K.L. & Pethiyagoda, R (editors), Contributions to biodiversity exploration and research in Sri Lanka. The Raffles Bulletin of Zoology. 12: 351-380. Bossuyt, F.; Meegaskumbura, M.; Beenaerts, N.; Gower, D.J.; Pethiyagoda, R.; Roelants, K.; Mannaert, A.; Wilkinson, M.; Bahir, M.M.; Manamendra-arachchi, K.; Ng, P.K.L.; Schneider, C.J.; Oommen, O.V. & Milinkovitch, M.C. (2004). Local endemism within the Western Ghats – Sri Lanka Biodiversity Hotspot. Science. 306: 479 – 481. Das, I. & De Silva, A. (2005). Photographic guide to the Snakes and other Rteptiles of Sri Lanka. New Holland Publishers. 144 p. De Silva, A. (1996). The Herpetofauna of Sri Lanka: a brief review. Graphic Land, Kandy. Published by Author, 99p. De Silva, A.; Bauer, A. M.; Austin, C.C.; Goonewardena, S.; Hawke, Z. & Vanneck, V. (2004). The diversity of Nilgala forest, Sri Lanka, with special reference to its herpetofauna. Lyriocephalus., 5(1&2): 164-182. 199 Dutta, S.K. & Manamendra-Arachchi, K.N. (1996). The Amphibian Fauna of Sri Lanka. Wildlife Heritage Trust of Sri Lanka. 230 p. Giri, V. & Chaturvedi, N. (2001). Preliminary survey of the Herpetofauna in the Western Ghats region of Maharashtra. Tigerpaper., 8(2): 1-7. Gunatilleke, I.A.U.N. & Gunatilleke, C.V.S. (1990). Dristribution of floristic richness and its conservation in Sri Lanka. Conservation Biology. 4(1): 21-31. Hettige, U.S.B.; Wickramasinghe, L.J.M.; Priyadarshana, T.G.M.; Gunawardena, K.; Perera, L. I. & Manorathna, A. (2000). Fauna of Gal Oya National Park. Sri Lanka Naturalist. 3(4): 55 – 61 pp. IUCN Sri Lanka (2000). The 1999 Red List Threatened Fauna and Flora of Sri Lanka. IUCN Sri Lanka. 144 p. Knopf, A.A. (1998). Familiar reptiles and Amphibians of North America. Chanticleer Press, New York. USA. 112 p. Manamendra-Arachchi, K. & Pethiyagoda, R. (1998). A synopsis of the Sri Lankan Bufonidae (Amphibia: Anura) with description of two new species. Journal of South Asian Natural History. 3: 213 – 248. Manamendra-Arachchi, K. (2000). Know your frogs. Sri Lanka Nature. 2(5): 4 – 16. Manamendra-Arachchi, K. & Pethiyagoda, R. (2001a). Ramanella nagaoi, a new tree-hole frog (Microhylidae) from southern Sri Lanka. Journal of South Asian Natural History. 5: 121 – 133. Manamendra-Arachchi, K. & Pethiyagoda, R. (2001b). Polypedates fastigo, a new tree frog (Ranidae: Rhacophorinae) from Sri Lanka. Journal of South Asian Natural History. 191199. Manamendra-Arachchi, K. & Pethiyagoda, R. (2005). The Sri Lankan shrub-frogs of the genus Philautus Gistel, 1848 (Ranidae: Rhacophorinae) with description of 27 new species. In: Yeo, D.C.J.; Ng, P.K.L. & Pethiyagoda, R (editors), Contributions to biodiversity exploration and research in Sri Lanka. The Raffles Bulletin of Zoology. 12: 163-303. Meegaskumbura, M.; Bossuyt, F.; Pethiyagoda, R.; Manamendra-Arachchi, K.; Bahir, M.M.; Milinkovitch, M.C, & Schneider, C.J. (2002). Sri Lanka: an amphibian hotspot. Science., pp. 298-379. Meegaskumbura, M. & Manamendra-Arachchi, K. (2005). Description of eight new species of shrub-frogs (Ranidae: Rhacophorinae: Philautus) from Sri Lanka. In: Yeo, D.C.J.; Ng, P.K.L. & Pethiyagoda, R (editors), Contributions to biodiversity exploration and research in Sri Lanka. The Raffles Bulletin of Zoology.12: 305-338. Pough, F.H.; Andrews, R.M.; Cadle, J.E.; Crump, M.L.; Savitzky, A.H. & Wells, K.D. (2004). Herpetology, Third edition. Pearson Prentice Hall, USA. 726p. Senanayake, F.R.; Soule, M. & Senner, J.W. (1977). Habitat values and endemicity in the vanishing rainforest of Sri Lanka. Nature. 265: 351 – 354 Wijesinghe, M.R. & Dayawansa, P.N. (2002). The amphibian fauna at two altitudes in the Sinharaja rainforest, Sri Lanka. Herpetological Journal. 12: 175 – 178. 200 REPORT ON THE STUDY OF THE BEHAVIOR OF JUVENILE ELEPHANTS RELEASED FROM THE ELEPHANT TRANSIT HOME TO THE UDA WALAWE NATIONAL PARK M.R. MOHAMED* Udawalawa National Park, Department of Wildlife Conservation, Sri Lanka ABSTRACT The behavior of several elephant calves released from the Elephant Transit Home was observed. Of the nine elephants that were released from the transit home, seven were observed to have joined wild herds, but two have not been found. These two, a male and female are believed to be missing or dead. Although the other seven elephants have joined wild herds, behavioral observations suggest that they have not been completely assimilated into the wild host herds. The observations also suggest that the human-elephant interactions and feeding regime while at the transit home may be influencing the behavior of the calves after they are released. INTRODUCTION It has been estimated that only about 2500-3000 wild elephants now remain in Sri Lanka, down from a population estimated at over 20,000 at the beginning of the 19th century (McKay 1973). Habitat loss to the rapidly expanding human population, large development projects, and other human activities such as the ongoing conflict have threatened the existence of this large herbivorous mammal needing about 100-150kg of food and 100-125 liters of water per day. With the shortage of food in their jungle habitats, the elephants have started to raid crops, home gardens, and even granaries and houses to fulfill their dietary needs. The result has been a severe escalation of the human-elephant conflict, with loss of elephant and human lives—although careful and thoughtful land-use planning can eliminate the bulk of these conflicts which are actually the result of ad hoc development plans. Elephant are commonly shot and killed by farmers retaliating against crop raids, and as is usual in conflicts, the juveniles suffer; elephant calves, orphaned by the death of their mothers, are being rescued by the wildlife officials with increased frequency. The Department of Wildlife Conservation therefore, established a facility called the Elephant Transit Home in 1995. Located near the Uda Walawe National Park, the Elephant Transit Home keeps the rescued orphaned baby elephants in a near wild situation until properly structured (age, sex) social groups are formed by these animals. When ready, these calves are then released into the wild, with the hope that they will join wild herds. The Elephant Transit Home is in the western part of the Uda Walawe reservoir. The facility consists of grassland, scrub/grassland mosaic and abandoned chena lands in different serial stages of succession, and the land runs down to the water. The baby elephants are kept in a roofed enclosure for the night and during the cooler times of the day are released to roam freely in the open areas of the reservoir bed. 201 When new calves are brought into the transit home, they are examined and treated (if necessary) by a veterinarian. The animals are fed regularly with a milk formula that has been developed at the Pinnawela Elephant Orphanage where over 50 orphaned baby elephants have been brought in and looked after from the 1970s. Over 12 baby elephants have been born at Pinnawela. As envisaged the young animals, brought together through adversity, bond together in small units spending most of their time in play. Play (fighting, wrestling, pushing etc.) is an essential ingredient for bonding. This sort of bonding is essential if these animals are to have a sense of security once they are released to the wild. Since its inception 83 elephant calves have been brought to the Elephant Transit Home. Thirty-nine died subsequently due to serious illnesses and disabilities. Some have been gifted to temples, others to public institutions and to the Pinnawela Elephant Orphanage managed by the Zoological Gardens. The objective of the transit home was achieved, when nine calves were released into the wild; a batch of four in March 1998, and another batch of five elephants in June 2000. Three elephants from each batch were radio collared, with the objective of monitoring their behavior and whether and how they become assimilated into wild herds. The objective of this was to study the behavior of the juvenile elephants released and the relationships that they formed with the wild herds that they joined. The study also tried to determine the social organization, habitat use and movement patterns of the juveniles released. Background Two batches of elephants that were released to the natural habitat in Uda Walawe National Park (UWNP) in 1998 and 2000 were considered in this study. The details at the time of release are given in Table 1. Figure 1 202 Table 1. Batch 1, Released on March 21, 1998 (three males and one female) Name M/F Age Height Collared Freq. Code Gamini M(tusker) 4.5 yrs 5' 1" Yes 12 Panduka M 5 yrs 4' 6"" Yes 14 Anusha F 4.5 yrs 4' 10" Yes 16 Anuradha M N/A N/A No - Batch 2, Released July 1, 2000 (one male and four females) Name M/F Age Height Collared Freq. Code Sandamali F 5 yrs 5' Yes 9 Komali F 5 yrs 4' 6" Yes 5 Isuru M 5.5 yrs 4' 11" Yes 8 Mattali F N/A N/A No - Emelyn F N/A N/A No - The tusker Gamini (FC12) from the first batch and the female Komali (FC5) from the second batch were not considered in this study due to the malfunctioning of their radio collars. The other collared elephants (FC14, FC16, FC9, and FC8) were tracked and their behavior studied in June & July 2001, as the first period of this study. The observations made during this first period are presented here. Fauna and flora of the Uda Walawe national park The Uda Walawe National Park is situated in southern Sri Lanka and falls partially within the intermediate dry zone. (Figure 1) 203 A large number of species inhabit the park; among them the large mammals take a prominent place. The Asian elephant (Elephas maximus) is the most important animal in the park, which has an estimated permanent population of between 350 to 400 elephants (Jayewardene 1994.) Hulankapolla, Goonawiddagala and Tibiriyamankada are the favorite habitat that these populations frequent during the dry season. Some of the other prominent mammals include the spotted deer (Cervus axis), sambar (Cervus unicolor), leopard (Panthera pardus), black-naped hare (Lepus nigricollis), grey langur (Presbytis senex), toque monkeys, wild pigs (Sus scrofa), and water buffalo. The park also supports a large and important bird fauna, among which are the Malabar pied hornbill, the Ceylon gray hornbill, peacocks, jungle fowl, red faced malkoha, to name but a few. The reserve has a wide range of vegetation types ranging from scrub jungle to open grassland (Guinea B grass) with some scattered and small dense pockets of jungle. The habitat is ideally suited for elephants. Figure 2. Protected areas under the Department of Wildlife Conservation METHODS 204 Radio telemetry was used to track elephant movements in order to determine the ranging patterns, habitat use, social organization and post-dispersal activities of the juvenile elephants. The study tracked the four animals, which were radio collared at the time of release. Namely, Panduka (FC14, male), Anusha (FC16, female), Sadamalee (FC09, female) and Isuru (FC08, male). All animals were located and observed for five days each week and their ranging pattern and positions including positions in relation to the nucleus herd and other animals were recorded. The positions were overlaid on a base map giving land use, topographical features, protected areas and infrastructure (roads, settlements etc). Unfortunately, the initial—and most important stage of the reintroduction—was not closely monitored. By the time this study got underway, the elephants had already joined wild herds, and the radio-collars were either becoming too tight or the batteries on the collars were close to becoming nonfunctional. Thus, the collars had to be removed soon after this monitoring began. History of the elephants released The nine juveniles released in two batches in March 1998 and June 2000 respectively as mentioned previously consisted of four males including a tusker and five females. Of these, three males (including the tusker) and three females have radio collars (Table 1). • Gamini (FC12): Gamini was a tusker, about four and half years old and 5’1” tall at the time of release. The tusks were about 10 inches long. He was solitary since the release, and according to the observations made by the DWLC field officers he had not joined a herd and roams in a relatively large area. In October 1998 Gamini was found by a fisherman, stuck in mud in the area where the Walawe River joins the reservoir. After being informed, the wild life officers rescued the tusker. There are reports that this elephant has been sighted in the Diyavinna area, which is beyond the NE boundary of the park. There have been no signs of this tusker over the last 12 months and no radio signal from the collar has been detected. It is doubtful whether the animal is now alive. • Panduka (FC14): Panduka, another male was released in March 1998 and was 5 years old and 4’6” tall. After few days of his release Panduka was found with a herd of about 27 elephants. It was observed that Panduka is being treated as a juvenile of that herd. This herd travels in thick, tall shrub areas and Guinea B grass areas. The collar was removed on the 15/07/2001 in Hulankapolla area as it was tightening round the neck of the elephant with the growth of the animal. If left longer it would have choked the animal. 205 • Anusha (FC16): Anusha, a four and half years old and 4’10” tall female, was released in March 1998. After few days of release she also joined a herd of about 7-9 elephants living in Mau Ara area. This herd has a few mature male elephants. This herd normally lives in areas with shrub and areas with tall trees. Her collar was removed on the 18/07/2001 in Mau Ara area as it was tightening round the neck with the growth of the animal. • Sandamali (FC09) and Isuru (FC08): Sandamali, a female, and Isuru, a male were released in July 2000. Sandamali was about 5 years old and 5’ tall while Isuru was about 5.5 years old and 4’11” tall. They joined to two separate herds. Sandamali joined to a herd of about 22 elephants and Isuru joined to a herd of about 26 elephants. The two herds live close to each other merge frequently to form a greater herd. • Komali (FC05): Komali, a female, was released in July 2000 and was about 5 years old and 4’6” tall. Few days after release the radio signals from the collar stopped and she could not be located. Since then there has been no sign of her in the park. She is also believed to be dead. • Anuradha, Manel, and Emalin Anuradha, a male, and the two other females were not radio-collared when released. It was observed that they have joined the two herds, which Sandamali (FC09) and Isuru (FC08) have joined. Ranging patterns Observations of the radio-collared animals are given in the following tables. Table 2 - FC08 Isuru Table 3 - FC09 Sadamali Table 4 - FC14 Panduka Table 5 - FC16 Anusha A map of the home ranges of the four elephants mentioned above elephants during the two-month period June/July 2001, is given in Figure 2. 206 Figure 3 Figure 4 Figure 5 207 208 DISCUSSION Observations during, June/July 2001 Of the nine juveniles released so far, apart from the male Gamini (tusker, radio collar code 12) and the female Komali (radio collar code 05,) the rest have been observed to have successfully assimilated into herds living in and around the Uda Walawe National Park. It is possible that Gamini (41/2 years at release in 1998,) though a juvenile, may have faced antagonism from other juvenile/sub-adult males within herds due to the fact that it was a tusker, as evidenced by the fact that it was never spotted within the vicinity of a herd. In the early period, when it was spotted, it was always solitary. This antagonism may have encouraged it to move to areas outside the park where there are fewer elephant. In fact there are unverified reports that it has been spotted in places like Iththagala, Diyawinna, Panahaduwa, and Madunagala etc., all outside the Park. These records need to be followed up. The fact that Gamini was originally from Wayamba, an area with a different ecology, and that it lived in the somewhat controlled environment of the transit home may have too been contributory factors affecting its post release behavior. With there being no rigorous monitoring plan along with the release in 1998, the information available about its early years is scant, and unfortunate. Panduka (5 years at release in 1998,) Sandamali (5 years at release in 2000,) Isuru (51/2 years at release in 2000) and Anusha (41/2 years at release in 1998) each joined separate herds within a few days of release. It is observed that these four prefer to associate with the adults in the respective herds and stay with the nucleus of the herd, rather than move with their peer-age group animals. This is possibly due to their association with man previously. It appears that they are more agitated and fearful of the presence of humans than their wild counterparts. Their food intake pattern too appears to be somewhat different to peer age juveniles within the same herd. On average they show a greater lethargy than the juveniles that were in the herd do. These preliminary observations need to be studied in detail in a subsequent part of the project, since it is likely that the feeding patterns and methods during their time in the transit home may affect the way they forage under natural conditions. Anuradha (released 1998,) Mattali (released 2000) and Emelyn (released 2000) have joined the same herds as Isuru and Sandamali and prefer to associate the animals released with them. They show the same behavior patterns as the released animals mentioned previously. While it is clear that the wild herds easily assimilate the released juveniles, it is necessary to ascertain the reasons for their somewhat abnormal behavior. If this is due to the management and care pattern within the transit home needs to be investigated. Equally important is to have in place a rigorous long-term post release observation plan to get more complete information on the progress made by the released animals. 209 Panduka, Sandamali and Isuru are within herds with home ranges falling within the boundaries of Gonaviddagala, Thibiriyamankada and Hulankapolla, where the vegetation is mostly grassland composed of guinea B grass. There appears to be a uniform locomotion pattern. Future observations will show whether seasonal variations in home ranges do take place. Anusha is within a herd living around the Mau Ara tank where there is a mix of medium forest cover and shrub forest. The home range is significantly smaller than that of the other herds. While it is very important to have a long-term comprehensive monitoring process, the preliminary data over the two months indicate that the released animals and their herds have sufficient food and water within the confines of their respective home ranges measured over the study period. Observations during august to November 2001 The relationship between the juveniles in the wild herd and the juveniles released in the park were observed in greater detail. The wild animals were seen to cohabit more closely and be active and playful in comparison to the newly introduced juveniles which in general tended to be more on their own with a noticeable lack of interaction with the former. The newly introduced juveniles were also seen hovering closer to the mature females in the herds, compared to the wild ones. The mature females always take care of the juveniles. They help the juveniles to stand up when fallen, pull them back closer to the herd when they wander away and when in danger alert them and surround them. However, it was noticed that the mature females are a lot more indifferent to the introduced juveniles. In times of danger, the introduced juveniles were seen to be attempting to get closer to the herd nucleus, on their own, rather than at the prodding of the mature ones. Sometimes on such occasions, the wild juveniles were seen to be aggressive towards the introduced ones by fighting and chasing them off. However, “Sandamali” (FC9, female 6.5years old) was seen to be striking back when attacked as such. Though the adults normally do not interfere during such fights, they will intervene only if the situation gets out of hand. During this period the drought ended and the park received the early rains from the northeast monsoon. The grasses began to grow and the water holes started filling up. The availability of food increased. As a result, the home ranges compared to the previous period of observation were shrinking. On rainy days, the herds tend to stay on higher ground, avoiding the waterlogged areas. No significant changes in feeding or behavior patterns were observed compared to the last period, except the fact that during heavy rain, the introduced juveniles sought shelter under trees and stopped feeding, unlike the wild animals which kept to their normal grazing pattern. During the previous period of observation, “Sandamali” (FC09, female), “Isuru” (FC08, male), “Manel” & “Emalin” were in two separate herds. But during the current period they were observed to be grouping together and hovering near the respective herds but not within them. They did not get to the nucleus of the respective herds. 210 On 8th and 9th October, the four juveniles were seen entirely separated from the respective herds and traveling together. They were observed continuously during the daytime (10 hours) on both these days and were never seen to get near or join the respective herds. They were lead by “Sandamali” (FC9, female 6.5years old) and were confined to an area of about 80 Ha. On 10th October the group was observed to have rejoined their respective herds. It is possible that during this time the male “Isuru” (FC08, 7 years old) may have been trying to separate from the herd. Normally only the males leave the herd when they reach maturity, whereas it appears that in this instance it seemed as if the females too were trying to follow the male. This maybe due to their bonding during their period in the transit home. This behavior must be studied further. The question also arises whether the male juvenile was mature enough to leave the herd. In conclusion, these observations show the importance of studying the pre and post dispersal behavior of both indigenous and introduced juveniles more methodically and thoroughly. Recommendations This release of juvenile elephants represents a pioneer effort to reintroduce large mammals into the wild. Unfortunately, however, the study was not monitored from its inception, and an excellent chance for the Department of Wildlife Conservation to contribute to a global knowledge on conservation and management has gone a begging. It is strongly suggested that any further activities of the Elephant Transit Home be done in collaboration with credible scientists so that the animals, DWLC, and the scientific world can benefit. In this context, the following recommendations are made. ƒ Monitor the behavior of the released elephant calves from the very beginning. Animals should be observed throughout most of the day and all behavior patterns should be recorded in detail. Fulltime researchers with DWLC counterparts should be assigned for this purpose. ƒ Several animals should be radio-collared so that they can be located and identified. Animals without collars should be outfitted with false collars (fastened with coir rope so that the collars will drop off after about a year). This will allow these animals to be identified without ambiguity. ƒ Juvenile males should be radio-collared and monitored until, and for sometime after, they leave the host herd. Virtually nothing is known about the dispersal of juvenile males, after they leave the herds — how far they go, where they go to, etc. This information is of vital importance for managing targeted elephant populations in a world where not each and every elephant and elephant population can be conserved; a situation that the DWLC and all Sri Lankans will have to accept, given the intense conflict and competition for land between people and elephants. 211 REFERENCE McKay, G.M. (1973) The ecology and behavior of the Ceylon elephant in south-eastern Ceylon, Smithsonian Contribution to Zoology No. 125 212 EXPERIENCE OF RESTRAINING ELEPHANTS IN MUSTH P. NIGAM and P.K. MALIK Wildlife Institute of India, Post Box # 18, Chandrabani, Dehra Dun – 248 001, Uttaranchal ABSTRACT Musth has implications for maintenance in captive populations as management of these animals is quite cumbersome and involves considerable risk. The present study highlights successful restraint of elephants in musth on three different occasions employing sedative/analgesics. Effective levels of standing sedation could be achieved using xylazine hydrochloride and ketamine hydrochloride at a dose rate of 0.13 mg/kg and 0.04 mg/kg body weight respectively for immediate restraint. The subsequent actions following sedation till complete recovery from musth events are documented. KEY WORDS: Elephant, Musth management, Xylazine hydrochloride, Ketamine ydrochloride, Yohimbine hydrochloride. INTRODUCTION Adult male Asian elephants (Elephas maximus) during sexually active period exhibit a phase of altered behaviour known as musth. This is discernable through two physical attributes-secretions from temporal glands and dribbling of urine. There are several other subtle changes in behavior resulting in rise of aggression (Kahl and Armstrong, 2002), restlessness and reduced feeding activities (Poole 1987). Physiologically elevated serum androgen levels have also been reported (Hall-Martin, 1987; Rasmussen and Schulte, 1998). Elephants in musth in captivity are well known for aggression and nonresponding to the commands of their keepers. Early detection of musth through physical signs/ altered behavior is essential for keeping such elephants restrained during the period. Non detection of musth symptom and keeper’s negligence often risks the public safety and leads to loss of life and property. We have come across three such cases of musth incidences wherein it was essential to restrain the animal for the public safety. We used chemical immobilization technique for restraining such animals and subsequent management of musth in post-restraint period. The present paper deals with drug choices, their efficacy and management experiences for handling such elephants in three different stages of musth at three different sites. 213 METERIALS AND METHODS Case 1: A captive Makhna (Tusk less bull elephant) 29 yrs of age at Jaldapara Wildlife sanctuary was reported to be in musth and had run into the forest. The elephant was not responsive to Mahout’s command and was aggressive. The animal had dried discharge marks around the temporal. It was decided to chemically restrain and tether the elephant. A total dose of 250 mg of Xylazine hydrochloride and 150 mg of Ketamine hydrochloride were used to sedate the elephant. The elephant achieved standing sedation and could be easily handled. The limbs were tied and the elephant was tethered on to a tree. The elephant was revived using 50 mg yohimbine hydrochloride given intravenously. The elephants were monitored till full recovery from the drug effect. Micturation and initiation of feeding were indicative of its recovery and self care. The ration provided to the elephant was restricted and the elephant was allowed abundant intake of water and was kept cool by frequent spraying of water. The elephant regained his normal temperament within a week. Case 2: A captive makhana elephant around 40 years of age was brought to Haridwar from Meerut to participate in an annual festival. The elephant was having slight temporal discharge since last two months that was gradually diminishing. The elephant had never shown any signs of musth except for temporal discharge throughout his life. It was reported that the elephant was exhibiting prolonged periods of penile erection along with stereotypic movements of head and whip like trunk movements and behavioral alteration since a week. Two days prior to veterinary intervention the elephant was aggressive and had charged Mahout. The elephant turned violent and chased and bashed mahout and also broke adjacent structures. The elephant was successfully restrained using 400 mg of Xylazine hydrochloride and 100 mg of ketamine hydrochloride. The elephant ran for about 100 meters and came to a halt. Reduced ear, tail and trunk movements along with relaxation of penis and snoring were indicative of desired level of sedation. The elephant was chained and rehabilitated similar to the previous case. The elephant regained normal temperament within 10 days. The situation could thus be mitigated. Case 3: A 32 year old captive makhana elephant was reported to be in musth and had become threat to life and property. The elephant had been temporarily kept in a compound along with a cow elephant within the heart of Saharanpur city. The animal was exhibiting slight temporal discharge since a week, frequent penile erection, dribbling of urine, stereotypic movements of head three days prior to veterinary intervention. The elephant had also tried mounting the cow elephant on three different occasions since morning however failed to establish sexual congress. The animal was also not responsive to Mahout’s commands and had charged mahout a day prior to the veterinary intervention. The elephant was chemically restrained using 600 mg of Xylazine hydrochloride and 150 mg of ketamine hydrochloride and secured to a tree. The above mentioned rehabilitation procedures were followed and the elephant regained normal temperament within 15 days. 214 In all the three cases the drugs were remotely injected using Dist-inject aluminum dart with 63 mm long collared needle and employing Dist Inject model 60 N projectors. Yohimbine hydrochloride and Atipamezole hydrochloride was used as an antidote for xylazine. This technique of chemical immobilization was not used as treatment for musth in elephant but was used as one of the methods of mitigating the problem by securing the animal so that it does not become a threat to life and property. The details of induction time, sedation time and recovery are tabulated below. 215 Journal of the Department of Wildlife Conservation 2006-1: 113-118 Table 1. Drug dosages used in managing elephants in musth on different occasions Sr No. 1 2 3 Location/ Estimated weight/ Age Jaldapara/ Makhana 3000 kgs/ 29 yrs Jwalapur/ Makhana 3000 40 yrs Month/Year State of Musth Xylazine HCl (mg) Ketamine HCl (mg) Induction time Sedation time Antidote Recovery time December 2002 Almost dried up temporal discharge, animal not responsive to commands and showing some degree of aggression (Post musth period) Diminishing temporal discharge, exhibiting prolonged periods of penile erection along with stereotypic movements of head, ear and whip like trunk movements, behavioral alteration (End of mid musth) Slight temporal discharge, frequent penile erection, dribbling of urine, stereotypic movements of head and non responsive to Mahout’s commands. (Mid musth period) 250 15 17 min 50 min Yohimbine 50 mg 3 min (0.08 mg/kg) (0.05 mg/kg) 400 100 7 min 27 min Yohimbine 40 mg 8 min (0.13 mg/kg) (0.03 mg/kg) 600 150 8 min 40 min Atipamezole 2 min 30mg (0.17 mg/kg) (0.04 mg/kg) July 2004 Saharanpur/ November Makhana 2004 3500 32 yrs Induction time: Time of darting to time animal showed first signs of sedation Sedation time: Time when animal showed first time of sedation to giving of the antidote Recovery time: Time when animal injected antidote till it regained consciousness Journal of the Department of Wildlife Conservation 2006-1: 145-160 RESULTS AND DISCUSSION The phenomenon of Musth has been recognized for centuries as a natural behavior among healthy adult male elephants between 15-60 yrs of age. The duration and severity of musth varies between individuals and may last from few days to several months. It is characterized by episodes of aggressive behavior and heightened sexual activity; however bulls can breed in and out of musth. In all the three cases presented, the elephants were makhanas between age group of 30-40 yrs and in the later stages of musth ranging from mid musth to post musth period. This period seems to be a critical period which needs to be handled with care. All the three elephants exhibited varied signs indicative of musth. (Table 1) Managing musth in elephants has been one of the most important challenges in captivity. Elephants in musth have been successfully immobilized employing narcotics like etorphine (M-99) or Immobilon (Jainudeen, 1970, Jainudeen et al, 1971, Cheeran et al. 2002, Sabapara and Raval, 1993, Sarma and Dutta, 1996). On three occasions, Thakuria and Barthakur (1994) controlled a captive African elephant coming into musth using diazepam and lorazepam. Xylazine HCl, due to its high therapeutic index, smooth induction and smooth recovery, ability to induce trunk immobilization and excellent analgesic and sedative properties has become a useful drug in elephant practice (Bongso, 1980, Schmidt 1983, Fowler, 1986, Nayar et. al. 1992, Pathak, 1991, Sarma and Pathak, 2001, Nayar et al 2002). The drug has been generally used only as a supportive drug to prolong the period of tranquilization following recovery from initial immobilization employing etorphine hydrochloride (Sabapara and Raval, (loc. cit.), Sarma and Dutta, (loc. cit.)). Xylazine has almost similar action as etorphine except for a delayed induction time which has relevance in difficult terrain, dense habitat and personal safety. The drug has been found to be suitable in immobilizing musth elephant as the elephant retain the standing posture enabling satisfactory chaining operation. Dutta and Pathak (1997) successfully used Xylazine for immobilizing elephant in musth using a total dose of 700 mg for a 3 ton elephant. Cheeran (1994) commented using Xylazine alone in controlling captive rogues employing 100120 mg/ton of the drug. Sarma and Dutta (loc. cit.) successfully used 500 mg Xylazine alone on three different occasions for immobilizing elephants in musth. Xylazine alone has been reported to have depressant effect on cardiac and respiratory function (Pathak, (loc. cit.), Sarma and Pathak (loc. cit)). Ketamine does not have any depressant effect on the cardiovascular and respiratory system but produces muscular tremor and stiffness of the skeletal muscle. Combination of Xylazine and Ketamine minimizes the undesirable effects of both the drugs and has been shown to produce a balanced anesthesia. However, Cheeran et al 2002, Cheeran et. al. 2003 reported photosensitization in Asian elephant sedated with Ketamine-Xylazine and did not observe as much synergism as has been observed in carnivores. Sarma and Pathak (loc. cit.) in their study also inferred that ketamine could mildly mitigate the hypotension brought about by Xylazine, while potentiating its sedative action, hence recommended their use as a combination in elephants. Pathak (loc. cit.) tried 100-150 Xylazine and 50-100 gm ketamine intravascularly to laterally recumbent elephant in 53 clinical cases and reported that the combination could produce quick, safe and dependable analgesia, anesthesia and muscular relaxation when given intravenously in different surgical conditions affecting elephants. They have also reported recovery without excitement and untoward effect. Similar findings were reported by Nayar et. al.1992. In all the three cases presented, combination of Xylazine hydrochloride and Ketamine hydrochloride were used at an average dose of 0.13 mg/kg and 0.04 mg/kg body weight respectively to achieve effective standing sedation. No adverse effects were noticed. The advantage of using both the drugs in the present case was that the elephants could be handled and chained in a standing posture and the overall dosage of both the drugs could be considerably reduced. Though the present approach of chemically restraining the elephant was not a remedy for musth, it provided effective means of mitigating the problem by securing the animal to avoid any untoward incident. CONCLUSION The present study highlights successful restraint of elephant in musth on three different occasions using Xylazine hydrochloride and ketamine hydrochloride at a dose rate of 0.13 mg/kg and 0.04 mg/kg body weight and its subsequent rehabilitation. 146 REFERENCES Bongso, T.A. (1980). Sedation of the Asian elephant with xylazine. Journal of the American Veterinary Medical Association. 177( 9):783. Cheeran, J.V. (1994). Note on Controlling Musth in Elephants. Zoo’s Print. 9(12): 27-28. Cheeran,J.V., Radhakrishnan,K. and Chandrasekharan,K. (2002). Musth Journal of Indian Veterinary Association Kerala 7(3):28-30. Cheeran, J.V., Panicker K.C., Kaimal, R.K. and Giridas, P.B. (2003). Tranquillization and translocation of captive bulls. In: Giants on Our Hands: Proceedings of the International Workshop on the Domesticated Asian Elephant. Food and Agriculture Organization of the United Nations Bangkok, Thailand. 5 -10 February 2001. Dutta, B. and Pathak, S.C. (1997). Capturing Musth Elephant (A Case Report) Zoo’s Print. April Vol. XII, No.4:12. Fowler, M.E. (1986). Zoo & Wild Animal Medicine, W.B. Saunders Company, USA, 2nd Ed. 892-893. Hall-Martin, A.J. (1987) Role of musth in the reproductive strategy of the African elephant (Loxodonta africana).S.Afr.J.Sci. 83:616-620. Jainudeen, M.R. (1970). The use of etorphine hydrochloride for restraint of a domesticated elephant (Elephas maximus). Journal of the American Veterinary Medical Association 157(5):624-626. Jainudeen, M.R., Bongso, T.A. and Perera, B.M.O.A. (1971). Immobilization of aggressive working elephants (Elephas maximus). Veterinary Record 89(26):686-688. Kahl, M.P. and Armstrong, B.D. (2002). Visual displays of wild African elephants during musth. Mammalia 66:159-171. Nayar, K.N.M., Radhakrishnan, K., Chandrasekharan, K., Cheeran, J.V., Ravindran, S., and George, P.O. (1992). Anaesthesia for surgical manipulations in the elephant. In: Silas, E.G., Nair, M.K., and Nirmalan, G. (Editors), The Asian Elephant: Ecology, Biology, Diseases, Conservation and Management (Proceedings of the National Symposium on the Asian Elephant held at the Kerala Agricultural University, Trichur, India, January 1989). Kerala Agricultural University, Trichur, India pp. 156-158. Nayar, K.N.M., Chandrasekharan, K. and Radhakrishnan, K. (2002). Management of surgical affections in captive elephants. Journal of Indian Veterinary Association Kerala 7(3):5559. Pathak, S.C. (1991). Xyalzine-ketamine anesthesia in Indian elephant (Elephas maximus indicus). – Trial on 53 clinical cases. International Seminar on Veterinary Medicine in Wild and Captive Animals, Nov. 8-10, Bangalore, India. pp:21. Poole, J.H. (1987). Rutting behavior in African elephants: the phenomenon of musth. Behavior. 102:283-316. Rasmussen, L.E.L. and Schulte, B.A. (1998). Chemical signals in the reproduction of Asian (Elephas maximus) and African (Loxodonta africana) elephants. Anim. Reprod. Sci. 53:19- 34. Sabapara, R.H. and Raval, P.P. (1993). Controlling musth elephants by tranquilzation: two contrasting cases. Zoo’s Print, Nov’93:26-27. 147 Sharma, K.K. and Dutta, B. (1996). Musth and its Management in Asian Elephant: A Discussion Based on Four Clinical Cases. Zoo’s Print. Apr; 11(4):21-22. Sarma, K.K. and Pathak, S.C. (2001). Cardio vascular response to xylazine and Hellabrunn mixture with Yohimbine as reversal agent in Asian elephants. Indian Veterinary Journal 78(5):400-492. Schmidt, M.J. (1983). Antagonism of xylazine sedation by yohimbine and 4-aminopyridine in an adult Asian elephant (Elephas maximus). Journal of Zoo Animal Medicine 14: 4-97. Thakuria, D.B. and Barthakur, T. (1994): Management of musth in a male African elephant by chemical sedative in the Assam State Zoo, Guwahati. Zoo’s Print, Sep, 94: 120. 148 EFFECTIVENESS OF ELECTRIC FENCING IN SRI LANKA IN MITIGATION OF HUMAN ELEPHANT CONFLICT: THE ROLE OF SOCIAL ASPECTS P.K. PREMARATHNE* and L.H.P. GUNARATNE Department of Agricultural Economics and Business Management, Faculty of Agriculture, University of Peradeniya, Sri Lanka ABSTRACT Human elephant conflict (HEC) has become a major environmental issue in Sri Lanka, which is at an escalating trend with more elephant and human deaths, property damages, and many recoverable and unrecoverable losses. Fragmentation and loss of natural habitats, disturbances of migratory routes due to increasing human settlements are the prominent reasons for the conflict. With a view of mitigating the HEC, a number traditional as well as introduced measure has been adopted by communities and Department of Wildlife Conservation. Among the various abatement strategies, electric fences appear to be the most preferred. Against this background, the study on the five electric fences namely, Kandaketiya, Kalagama, Herathgama, Mahaweli System G and Lunugamvehera was carried out with the objective of finding out the effectiveness of electric fences in Sri Lanka. The data were collected from 238 respondents and officers in DWLC regional offices. The study revealed that proper maintenance of the electric fences is the key factor for their effectiveness. The level of community participation for maintaining of the electric fences and acceptance levels of the fences is depicted through their attitudes and perceptions. Further, involvements of community based organizations (CBOs) were crucial in establishment and maintenance. Social factor appear to more important than technical factors. The study makes suggestions based on findings for improved level of community participation and acceptance for effective fencing. The study targets to evaluate the role of social factors to increase the effectiveness of electric fencing in Sri Lanka. The specific objectives are; to investigate the social acceptance and perception of mitigation measures of the conflict and to identify the needed measures to improve the social acceptance and community support on electric fencing. 149 INTRODUCTION Elephant is the key animal species with supremacy in the Sri Lankan wild from the ancient periods of time. Similar to most of the endangered species in the wild, extinction of elephants is at an ascending trend in the world. The main causes for this were the excessive level of habitat encroachment and poaching by human. In Sri Lanka, the present elephant population is approximated between 3,160 and 4,405 elephants (Kemf & Santiapillai, 2000). During last fifty decades the forest cover got reduced in more than 18 % with the increase of human population from 8 million to 19.5 million. As a result, elephants have to limit their territorial boundaries to national parks and some forest pockets in Northeastern, Eastern and Southern parts of the island. The coexistence between humans and elephants has been deteriorated during past few decades. Encroachment and replacement of the forestlands for slash and burn cultivation, expansion of human settlements towards elephant habitats were intensified during last few decades. Competition for the scarce land, food and water has created the conflict between humans and elephants over the time. Crops, houses and some instances human lives are frequently damaged by roaming elephants. The human encroachment at a higher degree left the elephants to complete dependency on cultivated crops. Since the traditional mitigation strategies are now lack in effectiveness in high HEC areas, farmers are forced to use different ways to harm elephants. Number of strategies such as promulgation of protected areas by resettlement of people, translocation of aggressive male elephants, elephant drives, electrical fences, compensation payments and conservation aimed other measures have been taken by the governmental as well as non-governmental authorities. Among currently practicing measures for mitigating human-elephant conflict, electric fences are considered as the most effective measure. If the electric fences have been strategically located, it acts as very effective elephant barriers (De Silva, 1998). The electrified fences appear to be the people's solution to the elephants’ problem by physically separating their territories. In this context, the effectiveness of the electric fences is mostly depending on the social acceptance. 150 METHODLOGY The theoretical framework of sampling procedure, data collection, data analysis, details of the selected electric fence and the sample areas has been presented in this chapter. Selection of the electric fences and areas for the study Based on the severity of the HEC, geographical location and involving organizations in operational activities of the fences, following five electric fences were considered for this study: Kandeketiya electric fence near VictoriaRandenigala-Rantambe Sanctuary, Herathgama electric fences near KahallaPallekele Sanctuary, electric fence around Mahaweli System G, Kalagama electric fence at Balaluwewa-Kalawewa sanctuary and Lunugamvehera electric fence at Lunugamvehera National Park. Description of the area and the electric fences Kandeketiya electric fence at Victoria-Randenigala-Rantambe Sanctuary Kandeketiya electric fence in Kandeketiya Divisional Secretariat Division has been established to prevent elephant damages coming from VictoriaRandenigala-Rantambe Sanctuary which located in Central region of the country and extends approximately 41,600 hectares. The fence was constructed during 1998 to1999 period from Uma oya to Pathagala rock extending approximately up to 9 km. Due to the Accelerated Mahaweli Development Programme (AMDP), the forest cover was extensively cleared loosing a substantial land extent from these catchments for the huge reservoirs. It has arisen a number of environmental issues like human elephant conflict. Approximately 75 to 100 elephants living in this sanctuary invade nearby villages in search of food. The study was conducted for totally 47 households from the villages protected with fence namely Akkiriyawatta, Maliyadda and Wewathenne and the villages that do not protected with electric fence such as Lemasooriyagama, Othalawa, Serasumthenne, Serupitiya, Meeriyabedda and Theripehe. Herathgama electric fences near Kahalla-Pallekele Sanctuary Herathgama electric fence in Northwestern region of the island protects mainly Polpithigama D.S.Division from the attacks by the elephants in KahallaPallekele Sanctuary. This sanctuary is located within three Districts extending 21,690 hectares. The sanctuary consists of 20 to 30 elephant’s in resident herd and 80 to100 in migratory herd. The forest is one of the resting-places of migrating herds between Wilpattu and eastern forests of the country. The fence was constructed in 2000-2001 period extending about 33 km from Siyabalangamuwa reservoir to Immihaminegama villages covering Herathgama, Irrudeniyaya, Thibbatuwewa, Pothana, Koonwewa, Siyabalangamuwa, Pothuwila, and Galahitiyawa. The survey was included 58 households from the villages in fenced areas namely Siyambalangmuwa, Siyambelewa, Hatangama, Thibbatuwewa, Galahiyawa, Herathgama, Nikawewa, Pothuwila, Irudeniyaya, Mahapitiya, Bambaragalayaya, Pansiyagama and the villages not covered by the electric fence such as Jayalanda, Meegalewa, Kankanigama, Kalankuttiya and Govigammanaya. 151 Electric fence around Mahaweli System G Mahaweli System G locates in between the Dambulukelle forest and the Wasgamuwa National Park. Being a comparatively narrow land strip, elephants migrate through this settlement area. The Electric fence of Mahaweli System G extends approximately 105 km. It was constructed in 2000-2001. The Mahaweli System G is totally covered by this electric fence. The study was conducted in three areas namely, Atthanakadawala-Seegala, Galmulla and Damanayaya covering 43 households considering before and after fence scenario. Kalagama electric fence at Balaluwewa-Kalawewa sanctuary The Kalagama electric fence in Anuradhapura District with approximately 10 kilometers in length was established in 2001. The fence lies from Konpolayagama to Undurawa along the boundary of Kalawewa- Balaluwewa sanctuary, the catchments of Kalawewa and Balaluwewa reservoirs. The sanctuary is a transit as well as habitat for many (100-128) elephants. Data was collected from the villages namely Undurawa North, Kalagama, New Balaluwewa and Dambewatuna. Electric fence at Lunugamvehera National Park The Lunugamvehera electric fence extends 17 km in length that considered for the study. It initiated from the Wilamba Wewa (reservoir) closer to the National Park and continues towards the direction of Yala National Park. Lunugamvehera National Park (23,499.77 hectares) is located in Hambantota district in the Southern region of the island. Human encroachment for cattle feeding is a severe issue at the National Park in which around 150 elephants are living. The villages namely Kiulara, Thanamalwila and Ranawarawa were considered the villages without having coverage from electric fence. Punchiappujandura, Lunugamvehera, Boogahawewa, Gestupana, Colony 1, Colony 2 and Padikepuhela, Karawile, Thammennawa and Hunathuwewa villages were considered the protection from electric fence including number of 60 respondents. Data collection A well-structured questionnaire was prepared for household survey in selected sample areas. The expected data to collect were general household information, land ownership, behavioral pattern of the wild elephants, number of crops and property damages, number of injuries, number of human deaths before and after the electric fence, severity of the damages, attitude of the villagers towards the electric fencing. HEC related information such as numbers of human deaths, numbers of injuries, numbers of property damages, numbers of plant destructions and numbers of crop damaging incidences were collected from key informants. Relevant secondary information includes total number of elephant related incidences during past few years, existing mitigating measures was collected from government institutions such as DWLC, Mahaweli Authority and divisional secretariat officials in relevant areas. The information on electric fences was also obtained 152 RESULTS AND DISCUSSION Considering the damage incidences both in fenced and non-fenced areas, the effectiveness of the electric fences have been evaluated on avoidance of damages. The Kandeketiya electric fence has not shown a substantial reduction in the number of incidences. Because of the inadequate length of coverage by electric fence and the elevated human activities in elephant habitats are the main reasons. Table 1. The effectiveness of electric fences in damage avoidance in 2003 Incidences VRR Damage reduction in each site KP Mahaweli S.G Kalagama LMV Human deaths 1 0 2 3 0 Human injuries 3 3 0 0 3 Property damages Plants destructionCoconut trees Plants destruction other trees Crop damagesPaddy lands Crop damages-other Damages to stored paddy 9 6 2 9 1 28 123 194 17 92 29 30 89 72 114 10 15 113 146 145 0 -9 21 96 81 3 13 4 5 0 According to the findings in table 01, the Kandeketiya electric fence has shown its effectiveness by avoiding human injuries. The electric fencing seems to be less effective in avoiding human deaths and damages to stored paddy. Slight increases of the incidences can be experienced in property and other crop damages. Related to the Herathgama electric fence, the number of incidences is fairly low and there are no records on human deaths or injuries. Plants destruction by elephants appeared to be fairly high even after establishing the fence may be due to weak points. Meanwhile the effectiveness of Kalagama electric fence at Balaluwewa-Kalawewa sanctuary is fairly high. However, plant destruction has not decreased satisfactorily. The Lunugamvehera electric fence shows high effectiveness in avoiding damages. Most of the fences are not functioning at expected levels of effectiveness. Some reasons that affect the effectiveness of the electric fencing are partially and incomplete maintenance, lack of community participation to various activities related to maintenance of the fence. The human elephant conflict in Kandeketiya was arisen with the implementation of Accelerated Mahaweli Project which damaged elephant habitat, migratory routes as well as the human -elephant co existence. 153 Table 2. Satisfaction of stakeholders on electric fencing as a HEC mitigation measure KahallaResponse Separation with a System VRR Lunugam Kalagama Pallekele G -vehera 44.20% 38.89% 81.60% 60.00% 62.43% 25% 20.37% 12.20% 13.50% 20.91% 11.50% - - - - - 12.96% - - 6.66% - 12.96% 6.10% 16.67% - - 7.40% - 9.83% 10.00% 15% - - - - fence Translocation of aggressive elephants Complete drive Establishment of a corridor Plantation of fodder trees Establishment of elephant conservation & management unit Relocate villages As shown in the table 02, in Kandeketiya, most of the respondents agreed to physically separate from elephants by a fence. Forty-four (44.20%) percent of the responses were with this solution. Peoples’ attitudes towards electric fencing in the Polpithigama Divisional Secretariat Division, shows the preference to separate an area for elephants by a fence (38.89%) Attitude of the respondents in the Mahaweli System G (Bakamoona) towards electric fencing says that separation of an area by an electric fence is the best solution according to the majority of respondents (81.60%). It depicts the positive attitudes on the electric fencing by the people, and in the same way pointed out the effectiveness of the electric fence related to the System G. According to the responses, majority has kept more trust on electric fences. Earlier, more attacks of elephants had been come from catchment areas. With the establishment of electric fence, the trend of frequent visits of elephants into the villages has been declined. Therefore, peoples' 154 attitudes on electric fence are at positive level, which is shown by the highest number of responses (60%) with the electric fencing as a solution. At present, there is a problem with the effective coverage of the electric fence due to poor maintenance. People are considerably much satisfied (62.43%) of the electric fences as a mitigation option. However, they are dissatisfied on the prevailing settings of the electric fences. Due to those weaknesses of location of the fence, people have an attitude that elephants are still entering into the villages. Some respondents pointed out that elephants might be visiting the covered area from other areas and other forests that are located in the same side of the Lunugamvehera electric fence. Therefore, the HEC is still a problem though it is not much severe as before. Though the functioning electric fences in study areas were able to reduce the exposure to elephant related incidences compared to the non fenced areas, peoples don’t have 100% favors at electric fencing. The reason is they still suffer from some incidences. Therefore, it seems that combination of the mitigation measures rather than single measures are much effective and sustainable in existence. The social aspects which are influencing on effectiveness of electric fences can be listed as follows; 1. Role of government organizations 2. Community support 3. Perception and attitudinal factors of the stakeholders 4. Forest links The role of the government organizations includes the maintenance of regulations related to the land and conservation policies. In some instances, the government authorities have failed to enact the regulatory measures such as relocation of unauthorized settlers from the protected areas and corridors. It is therefore, the legal operations needs to be strengthened. In Victoria- RandenigalaRantambe sanctuary area, such issue on relocation of settlers had been arisen. Community support in maintenance of the fences is a crucial factor in effective electric fencing. The degree of community interest on participation in fence maintenance is found usually negative sloping with time. For the success of the electric fences, capacity development of community based organizations can be practiced. People in the fenced areas are also interacting with the nearby forests for fulfilling their needs such as timber, firewood, herbs and cattle feeding etc. Though these are illegal activities, the communities don’t have other alternatives so that those issues needed to be considered in a reasonable approaches rather than imposing strict regulations. That kind of approaches brings the community much closer interacts with electric fences. In Lunugamvehera, the one of the influencing factor on fence failure is extensive cattle feeding in national park. Those can be limited using approaches such as establishing separate feeding areas, buffer zones for outside animals. 155 The electric fences act as the barriers for accessing in to the forest recourses by the concerning communities. Therefore, those who are loosing their prevailing lands due to acquisition of lands for protected areas need to be allowed to use some extent of the land with certain restrictions. Therefore, wherever those routes of electric fencing are demarcated, a comprehensive pre- study is needed. If not, the disputes between community and authorities will be resulted and it makes the failures of the fences. CONCLUSION The study found that even the short-term benefits of electric fencing are not fully achieved in most of the cases. However, the weaknesses related to social related aspects need to be addressed. In such situations, the perceptions of the people in the conflict were not considered in policy making. The most important is community support in maintaining the fence. Community organizations have to play a role to protect the fence and keep clean the route of the fence. This needs to be supported by authorities by provision of funds and material (e.g. posts for replacement). Inability to achieve the intended benefits of the fences can be attributed to this factor mostly. Electric fence related activities can be improved with close rapport and relationship between government authorities and community. Active village level organizations need to be given the responsibilities and authority with transferring benefits. Further, the community should be made to understand the importance of the fence is to reduce the conflict. The tasks assigned for the communities need to be in flexible manner. The activities can be allocated with responsibilities and continuous monitoring. Further, the maintenance and related operations should be assigned for the communities not only in closer areas of the fence but also to the communities within the effective distance. The effectiveness of the electric fencing as well as the social acceptance can be improved with the approach of high level of community participation in all the stages of fence establishment. Integrated approaches along with electric fencing will be the best way to improve the as well as increase the level of community involvement. 156 ACKNOWLEDGEMENT This research was conducted with the generous funding from Economy and Environment Programme for Southeast Asia (EEPSEA). Our special thanks extend towards Dr. David Glover and Dr. Hermi Francisco and Professor Nancy Olewier for their kind cooperation and also to Ms. Cathy Ndiaye in EEPSEA. Our heartiest gratitude extends to Mr. Edmond Wilson (Deputy Director, Elephant Conservation, Law Enforcement and operations), Mr. Hendavitharana (Social Scientist, DWLC) and the staff members in conservation, IT and GIS units. Also we appreciate the support of Mr. Manjula Amararathne (Assistant Director-DWLC-Northwestern region), Mr. Mahesh (Elephant Control Unit, Galgamuwa), Assistant Directors of DWLC, Central, Mahaweli and Southern regions and staff of DWLC offices at Herathgama, Anuradhapura, Galgamuwa, Meegalewa, Galkiriyagama, Keerthibandarapura, Minipe, Lunugamvehera and Elahera. We wish to extend our thanks to the staff of the Residential Office of Mahaweli System G for their support in data collection. REFERENCES De Silva, M. (1998) Status and Conservation of the Elephant (Elephas maximus) and the Alleviation of Man-Elephant Conflict in Sri Lanka. Gujah, 19. IUCN, Maps Electric Fencing. (2000) IUCN African Elephant Specialist Group, online reference at www.iucn.org. Kemf, E. and Santiapillai, C. (2000) Asian Elephants in the wild: 2000- WWF Species Status Report. WWF-World Wide Fund for Nature, Gland, Switzerland. Santiapillai, C. (1997) Elephant conservation in Sri Lanka. The Island. (Sri Lanka) August. 24.1997 157 STRUCTURAL CHARACTERISTICS AND FLORISTIC COMPOSITION OF THE HAKGALA STRICT NATURE RESERVE AT 1800m ELEVATION R.M.W. RATHNAYAKE* Department of Wildlife Conservation, Sri Lanka ABSTRACT The woody vegetation of Hakgala Strict Nature Reserve at an altitude of 1800 m was quantitatively studied by plot sampling. Twenty two species were found to be endemic. The exotic species, Cestrum nocturnum was the most abundant species at this altitude. Nineteen percent (11 species) of the species identified were present at less than 10 stems per ha density, and the most dominant gbh class was 10-20 cm. Only 19% of the species (11 species) was represented by 100 or more individuals per ha. Eighty eight per cent of species (50 species) were represented by less than 5% of the basal area. Based on the relative cover values of species, Neolitsea fuscata, Michelia nilagirica and Semecarpus coriacea were the dominant species found in the vegetation. KEY WORDS: Floristic composition INTRODUCTION Upper Montane Rain Forests (UMRF) cover the higher parts of Sri Lanka ranging from 1500 m to the highest peak Piduruthalagala (2524 m), and these forests are characterized by a peculiar floristic composition within the IndoMalaysian region (Werner, 1982). Generally a large proportion of the angiosperm flora of Sri Lanka is endemic and commonly found in the low country wet zone and montane zone. Hakgala Strict Nature Reserve (HSNR) is located in botanically rich UMRF ranging from 1600 m to 2178 m elevation. The environmental conditions prevailing at different elevations could affect the physiognomy and floristic composition of the forest at respective elevation (De Rosayro, 1958). According to Grubb et al. (1963), the woody species i.e. trees and shrubs of UMRF exceeding 10cm gbh (girth at breast height) could be considered as woody vegetation. Wijesundera (1991) studied the phytosociology of HSNR, and Rathnayake et al. (1996)a & b studied the woody vegetation of the same forest at 1600 m elevation. Before these studies, species present in the UMRF in Sri Lanka have been identified by numerous authors (Vincent (1883), Chapman (1947), Koelmeyer (1957), Nisbet (1961) & Greller & Balasubramanian (1980)). It could be considered that HSNR is at dynamic state and knowledge of forest structure and floristic is necessary to the study of forest dynamics, plantanimal interactions and nutrient cycling. The Objectives of the present study was to study the forest structure in terms of distribution of stem sizes and basal area and floristic. 158 MATERIALS AND METHODS Study site The study was carried out at mid elevation (1800m) of Hakgala Strict Nature Reserve (HSNR) which lies 6 km southeast of Nuwara Eliya in Central and Uva provinces (latitudes being between (6053’ – 6057’ N and longitudes being between 80046’ and 80050’E) (Green (1990)), and the well known Hakgala Botanic Garden is situated in the bottom of this reserve. HSNR covers an area of approximately 1,142 ha, and is surrounded by Madulsima range on the east, Haputale on the south, Pedro range on the northwest, Adam’s Peak and Elk Plains on the west and Udapussellewa range on the North East respectively. The HSNR consists of three peaks aligned in the east west direction, and it is ranging from 1600 m to 2173 m elevation. Numerous streams, streamlets, steep slopes, gullies and many rocky out crops are found in the reserve. Floristic composition and structural characteristics Plant specimens were collected from the woody vegetation and herbarium specimens were prepared following herbarium techniques (Mitra, 1957). They were then identified in the National Herbarium, Peradeniya and in the Hakgala Botanic Garden. The identifications were confirmed by referring to the books (Dassanayake & Fosberg (1980) & Wijesinghe (1994)) . For the quantitative survey, an undisturbed representative area of about 1 ha of the forest vegetation was selected. According to Yamada (1977) ten plots of 20 x 20 m were demarcated randomly in this area. Stem density, frequency and dominance (in terms of basal area) were estimated for each species of woody vegetation (Greig-Smith, 1957). Stem density of Species A in sample plot X = No. of stems of species A in sample plot X Area of sample plot X Dominance of Species A (relative cover value) = Total basal area of species A x 100 Sum of basal areas for all species 159 RESULTS Table1 shows the species composition of all species (gbh > 10 cm), life forms and endemicity. The distribution of species among taxa was uneven. A total of 57 species belonging to 43 genera and 27 families were recorded. Taxonomic richness (total number of taxa) was 127. About 12.3% (7 species) of the species belonged to the family Lauraceae; 10.5% (6 species) to Rubiaceae and 7 % (4 species) to each family of Myrtaceae and Symplocaceae. The vegetation also consisted of 77.2 % (44 species) trees and 22.8% (13 species) shrubs. About 38.6% (22 species) species was found to be endemic. The only exotic species found in the vegetation was Cestrum nocturnum. Table 1. Floristic composition of woody perennial vegetation in Hakgala Strict Nature Reserve at 1800 m elevation T= tree, S = Shrub, *- endemic, #- Exotics Family & Species Life Form 1, Aquifoliaceae Ilex walkeri Wight & Gardn ex Thw. T 2. Anacardiaceae Semecarpus coricea Thw. T* 3. Asteraceae Senecio corymbosus Wall. Ex DC. S 4. Buxaceae Sarcococca zeylanica Bail. S 5. Caprifoliaceae Viburnum erubescens Wall. Ex DC. Viburnum coriaceum Bl. T S 6. Celastraceae Euonymus revolutus Wight Microtropis Wallichiana Wight ex Thw. T* T* 7. Clusiaceae Calophyllum walkeri Wight. T* Life Form Family & Species 8. Daphniphyllaceae Daphniphyllum neilgherrense (Wight) Thw. T 9. Elaeocarpaceae Elaeocarpus montanus Thw. T* 10. Ericaceae Vaccinium symplocifolium (G. Don ) Alston S 160 11. Euphorbiaceae Glochidion coriaceum Thw. T 12. Flacortiaceae Caesaria thwaitessi Briq. Scolopia crassipes Clos. T* T* 11. Icacinaceae Apodytes gardneriana Miers Nothapodytes foetida (Wight) Sleumer T* T 12. Lauraceae Actinodaphne ambigua (Meisn.) Hook. f. Actinodaphne glauca Nees Actinodaphne speciosa Nees Actinodaphne sp. 1 Cinnamomum ovalifolium Wight. Neolitsea fuscata (Thw.) Alston Neolitsea sp. 1 T* T* T* T T* T* T 13. Magnoliaceae Michelia nilagirica Zenker T 14. Melastomataceae Memecylon rotundatum (Thw.) Cogn. In DC. Osbeckia sp. S* S 15. Moraceae Ficus microcarpa L.f. T 16. Myrsinaceae Ardisia gardneri C.B. Clarke Maesa perrottetiana A.DC. Rapanea robusta Mez. S* S T 17. Myrtaceae Eugenia mabaeoides Wight Syzygium assimile Thw. Syzyfgium revolutum Walp. Syzygium rotundifolium Arn. T* T T T* 18. Oleaceae Olea polygama Wight T Life Form Family & Species 19. Rubiaceae Canthium montanum Thw. Ixora calycina Thw. Lasianthus gardneri (Thw.) Hook. f. Psychotria nigra (Gaertn.) Psychotria zeylanica Sohmer Tarenna flava Alston T* T S* S S T 20. Rutaceae Acronychia pedunculata (L.) Miq. Euoida lunu-ankenda (Gaertn.) Merr. T T 21. Sabiaceae 161 Meliosma simplicifolia (Roxb.) Meliosma pinnata (Roxb.) Max. in Walp. T T 22. Sapindaceae Allophyllus varians (Thw.) Radlk. T* 23. Sapotaceae Isonandra lanceolata Wight T 24. Solanaceae Cestrum nocturnum L S# 25. Symplocaceae Symplocos bractealis Thw. Symplocos cochinchinensis (Lour.) S. Moor. Symplocos elegans Thw. Symplocos sp. T* T T* T 26.Theaceae Eurya ceylanica Wight S 27. Ulmaceae Celtis Cinnamomea Lindl. Ex Planch. T Unidentified sp.1 Unidentified sp.2 T T Structural characteristics Table 2 provides trees and shrub families (> 10cm gbh) with percentage of mean total basal area and density (individual’s ha-1). Overall density of live stems in all size classes was calculated to be 2641 individuals’ ha-1. At the family level, stem density and relative basal area differed between the stem size classes. Thirteen families of species were represented below 80cm gbh stems. Only two families i.e. Lauraceae and Rutaceae represented all the stem size classes. Above> 100cm gbh values were represented by 11 families. Families Caprifoliaceae, Elaeocarpaceae and Sapotaceae represented only 21 –20cm gbh stem size class. The highest relative basal area values were recorded for families, Lauraceae (22.1%), Magnoliaceae (13.7%) and Myrtaceae (11%). Except seven families i.e. Lauraceae, Magnoliaceae, Myrtaceae, Sabiaceae, Rubiaceae, Anacardiaceae and Rutaceae, other families showed less than 5% relative basal area. Families, Rubiaceae, Solanaceae, Lauraceae, Myrtaceae, Caprifoliaceae and Sabiaceae showed more than 100 individuals/stems ha-1. Family Solanaceae (367 stems/individuals ha-1) represented the highest stem density. Four families i.e. Buxaceae, Celastraceae, Elaeocarpaceae, Melastomataceae and Sapotaceae were rarely found representing less than 5 stems/individuals ha-1(Tables 2). The stem density showed a reversed “J” shaped curve (Figure 1). The highest density was recorded for 10 –20 cm gbh stem size class and the lowest density class was > 100 cm gbh stems. Further Figure 2 and Table 2 show that 54.1% (1429 individuals ha-1) fall in to the lowest girth class (i.e. 10-20 cm). Eleven species exceeded >100 cm girth (gbh) with a density of 43 individuals per ha. 162 Table 2. Relative basal area values and density (number of individuals per ha) Values in each stem size class %BA- relative basal area No. - Density (number of individuals per ha) Family 10-20 Anacardiaceae Aquifoliaceae Buxaceae Caprifoliaceae Celastraceae Clusiaceae Cornaceae Daphniphyllaceae Elaeocarpaceae Ericaceae Euphorbiaceae Flacourtiaceae Icacinaceae Lauraceae Magnoliaceae Melastomataceae Moraceae Myrsinaceae Myrtaceae Oleaceae Rubiaceae Rutaceae Sabiaceae Sapindaceae Sapotaceae Solanaceae Symplocaceae Theaceae Ulmaceae Total %BA 0.01 1.80 0.45 0.02 0.01 0.06 0.01 0.12 0.16 2.62 0.08 2.10 5.15 0.32 0.39 0.94 1.23 0.30 15.77 No. 2 83 4 1 4 6 2 5 24 130 10 121 687 16 48 206 51 29 1429 Stem size class (gbh) cm 41-60 61-80 21-40 %BA 0.06 2.0 0.01 1.05 0.24 0.04 0.03 0.07 0.14 0.12 0.17 5.90 0.12 0.15 4.85 2.32 1.58 0.85 0.43 0.01 0.95 0.22 0.61 0.37 22.29 No. 8 69 3 10 3 1 3 4 12 4 17 183 2 10 115 84 60 23 36 1 147 5 26 5 831 %BA 0.03 0.08 0.62 0.03 0.06 0.48 0.21 4.45 0.14 0.04 1.65 2.59 0.15 0.10 0.13 1.00 0.89 12.65 163 No. 2 9 5 1 3 3 2 57 2 3 17 33 6 4 14 12 4 177 %BA 1.20 0.02 0.05 0.37 0.10 0.17 2.14 4.70 0.16 0.48 2.45 0.20 1.21 1.20 0.08 14.53 No. 5 1 3 3 2 1 27 12 1 4 25 3 8 5 2 102 81-100 %BA 2.11 0.43 0.21 0.37 3.80 1.40 0.62 3.46 0.20 0.73 13.33 >100 No. 8 2 1 2 20 3 2 16 2 3 59 %BA 2.90 0.58 0.20 3.20 7.56 0.39 1.05 0.89 1.50 2.70 0.51 21.48 No. 7 2 1 9 10 1 2 1 4 5 1 43 stem density (no. of stems per ha) 2000 1500 1000 500 0 20-40 21-40 41-60 61-80 gbh classes (cm) 81-100 Figure 1. Stem density classes of woody perennial vegetation at study site The total basal area (sum of the stem cross sectional area) at breast height extrapolated to per hectare basis) for the study area was 39.8m2 ha-1. The highest relative basal area was recorded for 21-40cm gbh stem size class i.e., 22.29 and the lowest relative basal area was recorded for 41 – 60cm gbh stem size class i.e. 12.65. (Table 2). The largest trees (> 100cm gbh), which constituted only 16% of the tabulated stems accounted for 21.5% of total basal area. A dominance-diversity curve was computed at the species level by ranking the relative basal area of all stems > 10 cm gbh for trees and shrubs were identified (Figure 2). All but seven species i.e. Neolitsea fuscata (16.9 %), Michelia nilagirica (13.7%), Meliosma simplicifolia ( 9.4 %), Semecarpus coriacea (6.2%), Syzygium rotundifolium (5.9%) and Acronychia pedunculata (5.5%) and Eugenia mabaeoides (5%) constituted less than 5% of the percentage basal area. 16 14 12 10 Relative basal area 8 6 4 57 53 49 45 41 37 33 29 25 21 17 13 9 5 0 1 2 Species rank Figure 2. Dominance diversity curve for the woody perennial vegetation at study site According to the dominance values (in terms of relative basal area values) different dominant families, genera and species were found (Table 2). Below 60 cm gbh classes, the dominant families were Lauraceae, Solanaceae and Myrtaceae. Within these three families, the genera of Neolitsea, Cestrum and Eugenia and species of P.zeylanica, C.aurantiacum and N.fuscata were the dominants. The dominant families above 60 cm gbh were Lauraceae, 164 >100 Magnoliaceae and Anacardiaceae. Within the dominant genera of Neolitsea, Semecarpus and Michelia, dominant species of N.fuscata, S.coriacea and Michelia nilagirica were found. DISCUSSION From the results it was evident that the vegetation at 1800 m of HSNR was dominated by the families Lauraceae (12.3%, 7 species), Rubiacea (10.5%, 6 species), Myrtaceae (7%, 4 species) and Symplocaceae (7%, 4 species).Lauraceae commonly occurs elsewhere at this elevation, and has been reported as the dominant in the vegetation at this elevation. Dominance of Lauraceae at this site may reflect the abundance of several avian dispersers of this family. Taxonomic richness (total number of taxa) was quite high at the study site like other montane forests (Grubb et al. (1963), Heaney & Proctor (1990) & Tanner (1977)). The presence of 38.6% endemic species further reveals the importance the forest vegetation. Number of species found in the montane zone of Sri Lanka was 591, and 291 species were endemic (Abeywickrama, 1956). Trimen (1885) also has reported that about five sixth of the country’s endemic species are present in the hill flora. According to Willis (Procteor et al (1988) and de Rosayro (1958), endemic species occurred in the montane rain forests mainly due to isolation of mountains, and isolated species may have evolved and became endemic. The only exotic species i.e. Cestrum nocturnum was found with the highest abundance, and it has been introduced to Sri Lanka through Hakgala Botanic Garden in 1889 (Trimen, 1890). This exotic species was considered as a weed of montane zone (Bond, 1952). According to Rathnayake et al. (1996) a, C.nocturnum was an exotic species found at 1600 m too, but it was absent at 2000m elevation probably due to the facts that the seeds or fruits of these two species are normally dispersed by birds (Alston, 1931), and at the higher elevations the birds could not disperse them successfully. Further, the presence of extreme environmental conditions (i.e. high wind velocity and high light intensity) at higher elevations, coiled already affects the survival of this species. According to the highest abundance of C.nocturnum in HSNR it could be suggested that it could suppress the growth of other species present in the forest. At the moment this exotic species has spread through the lower elevation on to the higher elevations seriously threatening the endemic vegetation of the upper montane rain forests. Stem density showed a reversed “J”-shaped curve which is typical for a mature stand, with many small stems compared to few large ones. The large size classes of stems were more variable spatially than small stems, so a large sample area would be needed to characterize composition and structural attributes of larger trees than small trees. According to Figure 1, many individuals of woody perennial vegetation do not exceed 20 cm gbh, and smaller individuals of overstorey vegetation may allow saplings to share the space for exposing to the light. Generally, C.nocturnum shows low gbh classes and that could be helpful in the case of saplings of C.nocturnum. High gbh values (> 100 cm gbh) were recorded for Michelia nilagirica, Semecarpus coriacea, Ficus sp. and etc., and they may not allow saplings to share the space. Compared to other species therefore their relative densities were low. 165 Less than ten species showed more than 5% of relative basal area at each elevation, and they are the dominants. It could be suggested that these species might determine the future of whole forest. That means these species may affect the seed germination, growth of saplings and growth of other plants. The highest relative basal area values were recorded for Michelia nilagirica and Neolitsea fuscata at 1800 m as well as 1600 m studied6, and they were not found at 2000 m elevation (Rathnayake et al., 1996a) due to their huge and tall stems and presence of shallow soil depth and other unfavorable environmental conditions. It was not the highest basal area holding species at these two elevations, because C.nocturnum is shrub form with low gbh values. The cover values give an idea of timber volume of the forest, although timber of this forest is not important in furniture industry. Relative basal area is a useful indicator of particular species in a forest and it may reflect the importance of the species in the community dynamics. Based on the overall relative cover values of species Neolitsea fuscata, Michelia nilagirica and Semecarpus coriacea were the dominants, and they could be considered as dominant species involved in community dynamics. They are endemic, and community dynamics mainly depend on these species too. Sampled areas in comparable studies ranged from 400m2 to 4000m2 (present study). Taxonomic richness (total number of taxa) was quit high at HSNR like other montane rain forests (Grubb et al., (1963), Heaney & Proctor (1990), Abeywickrama (1956) & Nadkarni et al., (1995)). Some of families and genera found in Hakgala SNR are common at study sites of other montane, forests (Heaney & Proctor (1990), Tanner (1977), Proctoer et al., (1988), Nadkarni et al., (1995), Weaver & Murphy (1990), Edwards (1977) & Edwards & Grubb (1977)). In the present study, the densest species was C.nocturnum, and the most dominant species was Neolitsea fuscata. Therefore the results of the present study were different from those of Wijesundara (1991), and these changes could be due to the forest die back, natural death of some trees and the greater distribution of C.nocturnum. At Barva forest in Jamaica, the dominant family was Euphorbiaceae (14.5% of basal area) (Tanner, 1977), whereas this taxon comprised only 0.5% of the total basal area only at 1800 m elevation in Hakgala SNR. The dominant family at Monteverde forests in Jamaica was Lauraceae (31% of total basal area) (Edwards, 1977), comprised only 5.8% of the basal area at the Barva site (Tanner 1977). In the present study, the dominant family was Lauraceae (i.e 22.9% of total basal area). This difference may result from different forest disturbance regimes; and climatic conditions. The current data on tropical forests suggest the similar elevation and environmental conditions do not dictate similar structure and floristic in tropical montane vegetation. A larger body of information on environmental factors, especially those that influence patterns of disturbance and regeneration are necessary to explain the great variation exhibited in tropical montane forests. The results of the present study could be useful in measures pertaining to the conservation of naturally established flora in the upper montane rain forests. It is also recommended that regular floristic studies, both qualitative and quantitative, at least at 6 year intervals be carried out and in order to monitor the possible vegetation change that could occur overtime in this vegetation. 166 REFERENCES De Rosayro, R.A. (1958). The Climatic and Vegetation of the Knucles Region of Ceylon. Ceylon Forester 2 (3 & 4) : 201-260. Grubb, P.J., Lloyd, J.R., Pennington, T.D. & Whitemore, T.C. (1963). A comparison of montane and lowland rain forest in Ecuador. The forest structure, physiognomy and floristics. Journal of Ecology 51: 567-601. Rathnayake, R.M.W., Jayasekera, L.R. & Solangaarachchi, S.M (1996). A quantitative study of overstorey vegetation of an upper montane rain forest. The Sri Lanka Forester XXII (3&4):43-49. Rathnayake, R.M.W., Solangaarachchi, S.M. & Jayasekera, L.R. (1996). A quantitative study of pigmy forest at 2000 m in Hakgala Strict Nature Reserve. Proceedings of the Second Annual Forestry Symposium. Department of Forestry and Environment Science. University of Sri Jayawardenapura, Sri Lanka. pp. 215-222. Vincent (1883). The Forests of Ceylon. Tropical Agriculturist 2 (1-12):280-281. Werner, W.L. (1982). The Upper Montane Rain Forests of Sri Lanka. The Sri Lanka Forester 15 (3&4): 119-131. Wijesundara, D.S.A.. (1991). Phytosociology of a montane forest in Sri Lanka. M.Phil. thesis, Faculty of Science, University of Peradeniya, Sri Lanka. Chapman, V.J. (1947). The application of aerial photography to ecology as exemplied by the natural vegetation of Ceylon. Indian Forester 73 (7):287-314. Koelmeyer, K.O. (1957). The climate classification and distribution of vegetation in Ceylon. The Ceylon Forester 3 (2): 144-163. Nisbet, R.H.McD. (1961). A forest inventory of the Peak Wilderness-Agra Bopats Forest Area and part of the Kelani Valley Forest Area. Ceylon Government Press, Colombo. 58p. Greller, A.M. & Balasubramaniam, S. (1980). A prelimiary floristic classification of the forests of Sri Lanka. Sri Lanka Forester 14 (3-4):163-170. Green, M.J.B. (1990). IUCN Directory of South Asian Protected Areas. IUCN-The Conservation Union, Cambridge, U.K pp. 211-212. World Mitra, J.N. (1957). An Introduction to Systemic Botany and Ecology. The World Press Private Ltd., Calcutta, India. Dassanayake, M.D. & Fosberg, F.R. (1980- ). Revised Hand Book to the Flora of Ceylon. Parts IVII. Amerind Publishing Co. (Pvt) Ltd., New Delhi, India. Wijesinghe, Y. (1994). Checklist of Woody Perennial Plants of Sri Lanka. Forest Department, Sri Lanka. 20 p. Yamada, L. (1977). Forest ecological studies of the montane forests of Mt. Pangrango, West Java. IV. Floristic composition along altitude. South East Asian Studies. 15: 226-254. Greig-Smith, P. (1957). Quantitative Plant Ecology (3rd edition). Butterworth, London, England. 256 p. Heaney, A. & Proctor, J. (1990). Preliminary studies on forest structure and floristics on Volcan Barva, Cota Rica. Journal of Tropical Ecology 6:307-320. 167 Tanner, E.V.J. (1977). Four montane rain forests of Jamaica: a quantitative characterization of the floristics, the soil and foliar mineral levels, and a discussuion of the interelations. Journal of Ecology 65:883-918. Abeywickrama, B.A. (1956). The origin and affinities of the flora of Ceylon. Proc. Ann. Sess. Ceylon A.A.S. Part 2:99-21 Trimen, H. (1885). Remarks on the composition, geographical affinities and origin of the Ceylon flora. Journal of Royal Asiatic Society (ceylon Branch) 9. Willis, J.C. (1908). The floras of Hill tops in Ceylon. Annals of the Royal Botanic Garden Peradeniya 4 (4): 131-138. Trimen, H. (1890). Ceylon Administrative Reports. Royal Botanic Gardens, Peradeniya. Report of the Director for 1889 12 p. Bond, T.E.T. (1952). Wild Flowers of the Ceylon Hills. Oxford University Press. 240 p. Alston, A.H.G. (1931). A handbook to the flora of Ceylon. Vol. 6., Dulau, London. Proctoer, J., Lee Y.F., Langley A.M., Munro W.R.C. & Nelson, T. (1988). Ecological studies on Gunung Silam, a small ultra basic mountain in Sabah, Malaysia. I. Environment, forest structure and floristics. Journal of Ecology 76:320-340. Nadkarni N.M., Matelson T.J. & Haber W.A. (1995). Structural characteristics and floristic composition of a Neotropical cloud forest Monteverde, Costa Rica. Journal of Tropical Ecology 11:481-495. Weaver, P.L. & Murphy, P.G. (1990). Forest structure and productivity in Puerto Rico’s Luquillo Mountains. Biotropica 22:69-82. Edwards, P.J. (1977). Studies in a montane rain forest in NewGuinea. II. The production and disappearance of litter. Journal of Ecology 65:971-922. Edwards, P.J. & Grubb, P.J. (1977). Studies of mineral cycling in a montane rain forest in New Guinea. I. The distribution of organic matter in the vegetation and soil. Journal of Ecology 65:943-969. 168 SUBSTRATE PREFERENCE OF THE CERATOPHORA TENENTII: GUNTHER, 1834 (SQUAMATA: AGAMIDAE) IN THE NORTHERN FLANK OF KNUCKLES CONSERVATION FOREST IN SRI LANKA R.K. RODRIGO1*, U.K.G.K. PADMALAL1, I.U.P.A. SAMARAWEERA1 and A.U.L.D. JAYANTHA2 1 The Department of Zoology, Faculty of Natural Science, The Open University of Sri Lanka, Nugegoda, Sri Lanka. 2 The Department of Zoology, Faculty of Science, University of Colombo, Sri Lanka ABSTRACT A study was conducted in the eastern slope of the northern flank of the Knuckles forest range in Sri Lanka for a piod of 3 months from March to June 2002 with the objective of studying the substrate preference of C. tennentii. Data were collected between 0800 to 1800 hrs following focal sampling method. Juvenile (J), male (M) and female (F) lizards were observed. The most preferred substrates were tree trunks (Juveniles = 34.3%, males = 37.3%, females = 13.5%) and cardamom stems (Juveniles = 33.5%, males = 28.2%, females = 32.2%) throughout the study period. The animals spent less than 10% of the total observed time on ground litter, logs, moss covered logs, twigs and moss covered twigs. This is the first study of the kind, substrate preference of any agamid lizard, in Sri Lanka. KEY WORDS: Substrate preference, Cetatophora tenentii, Knuckles conservation forest INTRODUCTION Knuckles forest range comprises different unique eco- systems (De Rosayro, 1958), with a high species endemism and richness (Ministry of forestry and environment, 1999). Out of five species of the endemic agamid lizard genus, Ceratophora, C. tenentii is distinctive to these mountain forest habitats. Due to habitat destruction by clearing the under storey for cardamom plantations, logging and man made fires, change in macro and microenvironment of the species is invariably resulted. A drastic change in ambient temperature following montane habitat destruction can be highly influencing the thermoregulation activity pattern of ectotherms (Clarke, 1996) such as agamid. The study has designed to determine the substrates preference of the endemic and highly threatened lizard species (Bambaradeniya, C. N. B. & Samarasekara, V. N., 2001). C. tenentii belongs to the class reptilia, order Squamata, family Agamidae (Lizards) and sub family Lyriocephalinae. Cophotis, Ceratophora and Lyriocephalus are the three extant genera of the sub family and are endemic to Sri Lanka (Deraniyagala, 1953; Manamendra-Arachchi, & Liyanage, 1994). The genus Ceratophora is represented by five species C. tennentii, C. stodartii, C. aspera, C. eardlarnii and C. karu (Pethiyagoda & Manamendra-Arachchi, 1998). C. tennentii is distinguished from all other Ceratophorans by the complex, laterally compressed rostral appendage (Pethiyagoda & Manamendra-Arachchi, 1998), its similarity to a leaf has earned this lizard the common name, the LeafNosed Lizard (Senanayake, 1979). The total length of the adult male is about 185 mm and that of adult female is about 183 mm (Deraniyagala, 1953). 169 Predominant colour of the dorsum and sides of mature individuals is reddish brown to olive green; larger scales more greenish than smaller ones. Area around the eye and sides of neck are with black margins. There are about ten broad, dark brown bands on tail separated by narrow, lighter interspaces. Venter is usually whitish. Male is darker than female (Manamendra-Arachchi, 1990). Juvenile lizard is dark brown both dorsally and laterally (Pethiyagoda & Manamendra-Arachchi, 1998). Distribution of the C. tennentii is only in Knuckles mountain region (Manamendra-Arachchi, K. & Liyanage, S. 1994), which has been isolated from rest of the central hills by the Mahaweli River, at an altitude of 760-1220 m (Pethiyagoda & Manamendra-Arachchi, 1998). Most of the habitat of this species is now under cardamom (Elettaria cardamonium) plantations (ManamendraArachchi & Liyanage, 1994), with its concomitant clearing of undergrowth. This slow moving lizard is frequently seen arboreal (Pethiyagoda & ManamendraArachchi, 1998). Location The selected habitat “Rivers Turn” is located in the Matale district in Sri Lanka. The study site is located adjacent to the Matale - Laggala main road in the Eastern slope of the mountain range 07. 523050 N and 080.736150 E (Figure 1). Habitat The study area was montane tropical wet evergreen vegetation. A clear stratification of the vegetation could be seen in the area, consisted of continuous canopy, under storey and ground layer. Calophyllum spp., Syzygium spp., Neolitsea spp., and Symplocos spp. were the dominant plant species in the canopy. Most of the trees trunks were covered by mosses, lichens and orchids. In the understorey, Strobilanthes spp., Lisea spp. and Elettaria cardamonium were the common plants. The ground layer contained different herbs, grasses, orchids, ferns and fern allies (De Rosayro, 1958) (Figure 2). METERIALS AND METHODS All observations were made in the eastern slope of the northern flank of the Knuckles forest range (Reverse Tern) from March to June 2002. Data were collected between 0800 to 1800 hrs in the cardamom plantation. Relevant ambient temperature, substrate and weather condition were reported simultaneously. Substrate of each lizard was noted in every minute (time sampling) following focal sampling method (Altmann, 1975). Light intensity was taken in every 30 minutes using a standard light meter. Ambient temperature of the site, where the observations were made, was noted in every 30 minutes with an alcohol thermometer throughout the study period. The weather condition was recorded according to four categories viz sunny, windy, misty and rainy. Males, females 170 and juveniles were considered for the study and they were identified according to taxonomic characters, their color patterns, size (Deraniyagala, 1953), and experience of the author. Every lizard was observed approximately from 2m distance for immunized the effect to the lizard natural behavior and a pair of binoculars (“Pentax” 8x30) was used to observe lizards from distance. Source: IUCN Sri Lanka (2003) Figure 1. Box indicated the reverse turn. Figure 2. Habitat of the study area with Cardamom plants. 171 Percentage of substrate use 40 35 30 25 20 15 10 5 0 BT CT GL L LM T TM TT TTM Substrate Juvenile Male Female BT= Bamboo Trunk, CT= Cardamom Trunk, GL=Ground Litter, L=Leaf, LM=Log-Moss Covered, T=Twig, TM=Twig-Moss Covered, TT=Tree Trunk, TTM=Tree Trunk- Moss Covered. 30 25 20 15 10 5 0 8.0 0 8.3 0 9.0 0 9.3 0 10 .00 10 .30 11 .00 11 .30 12 .00 12 .30 13 .00 13 .30 14 .00 14 .30 15 .00 15 .30 16 .00 16 .30 17 .00 17 .30 18 .00 Average Ambient Temperature (°C) Figure.3. Substrates used by juvenile, male and female lizards. Time (hrs) Figure 4. Average ambient temperature in the study site throughout day. 172 RESULTS Substrate preference of the lizards The most preferred substrates were tree trunks (Juveniles = 34.3%, males = 37.3%, females = 13.5%) and cardamom trunks (Juveniles = 33.5%, males = 28.2%, females = 32.2%) throughout the study period. The lizards next used moss covered tree trunks (Juveniles = 10.7%, males = 14.5%, females = 30.5%) and bamboo trunks (Juvenile = 9.8%, male = 10%, female = 3.2%) out of the total observed time period of a day. The animals spent less than 10% of the total observed time on ground litter, logs, moss covered logs, twigs and moss covered twigs (Figure 3). Weather condition of the study period During the study period weather fluctuated through out the day from sunny to rainy conditions from 0800hrs to 1130hrs there was a gradual increasing in the sunny condition. When the noon, sunny condition was gradually started to turning to the windy condition till 1330hrs. In the afternoon the weather conditions of the study area was misty and late afternoon there was a rainy condition. Ambient temperature in the study site Ambient temperature variations between 0800 – 1800 are shown in Fig. 04. In the morning average ambient temperature of the study site was 16.8 °C. The temperature slowly increased with the time and it was highest (24.25 °C) between 1000 – 1100 am. After 1200 noon average ambient temperature of the site gradually decreased and it was about 16.25 °C at 1800 hrs (Figure 4). DISCUSSION Most used substrate by lizards was cardamom trunks (Figure 3). Flowers of the cardamom plants support many kinds of insects as a food source for C. tennenti. Further, foliation of cardamom bushes provides a good opportunity for basking as the trunks expose to the sunlight more frequently compared to other substrates (author’s comment). Therefore, lizards preferred to hang on the cardamom trunks and next most used substrate was tree trunks (Figure 3). Moss-covered tree trunks used as the substrate by female lizards many times (30.5%) compared to juvenile (10.7%) and male (14.5%) lizards. Bamboo trunks, ground litter, logs, moss Covered logs, twigs and moss-covered twigs were used less than 10% of the total observed time (Figure 3). Moss covered surfaces have been preferred least frequently; may be due to low temperature of the substrate brought about by trapped moisture unless it is a tree trunk that directly exposed to the sun light. The color of the body of C. tennentii, lizards was frequently changed when lizards were changed the substrate as most of the reptiles are doing, as a defensive mechanism (Pough, 2001). 173 During the study period weather were fluctuated sunny to rainy conditions. In the morning average ambient temperature was low (16°C) in the study site due to the heavy mist and drizzling. The selected habitat is situated in the east side of the Knuckles forest range. Therefore, during the morning hours the light intensity is gradually increased within the study site. Because of that, the average ambient temperature has been increased (highest temperature was recorded as 24.25°C around 10.30 hrs) during 0900 to 1100 hrs. After the noon the average ambient temperature was fluctuated between 20°C to 16°C till night was fallen. (Figure 4). At the night lowest average ambient temperature in the study site was recorded 14°C. Acknowledgements I would like to thank Mr. D.B.Ekanayake, the owner of the Cardamom plantation in Riverse-turn, where the study was conducted, for granting me permission to carry out the study. Mr. Mendis Wickramasinghe, the instructor of the reptiles group and my colleagues of the Young Zoologists’ Association of Sri Lanka (YZA), helped me to lead this project to be successful. The staff members of the Biodiversity Section of IUCN –The World Conservation Union- Sri Lanka are also acknowledged for providing me with references. Finally, I appreciate the help given by Dr. C.N.B. Bambaradeniya for his assistance in completion of this project and the report subsequently. 174 REFERENCES Altmann, J. (1975). Observational study of behavior: sampling methods. Allee laboratory of animal behavior, University of Chicago, Chicago, Illinois, U. S. A. Bambaradeniya, C.N.B. and Samarasekara, V.N. (2001). An overview of the threatened herpetofauna of South Asia. IUCN - The World Conservation Union, Sri Lanka country office, 53, Horton place, Colombo 07, Sri Lanka. Pp 98-101. Clarke, A. (1996). The influence of climate change on the distribution and evolution of organisms. In the Animals and temperature (Ed. Johnston. I.A. & Bennett. A.F.). Society for experimental biology, Seminar series 59,University press, Cambridge. Pp 377- 407. De Rosayro, R.A. (1958). The climatic and vegetation of the Knuckles region of Ceylon. Ceylon Forester. 3: 201-260. Deraniyagala, P.E.P. (1953). A colored atlas of some vertebrates from Ceylon. Vol. II. Ceylon National Museums, Colombo. Pp 59-63. Manamendra-Arachchi, K. and Liyanage, S. (1994). Conservation and distribution of the agamid lizards of Sri Lanka with illustrations of the extent species. Journal of South Asian Natural History. 1(1):77-96. Ministry of forestry and environment. (1999). Biodiversity Conservation in Sri Lanka. Ministry of forestry and environment, “Sampathpaya”, Rajamalwatte Road, Battaramulla, Sri Lanka. 18p. Pethiyagoda, R. and Manamendra-Arachchi, K.N. (1998). A revision of the endemic Sri Lankan agamid lizard genus Ceratophora Gray, 1835, with description of two new species. Journal of South Asian Natural History. 3(1):1-50. Pough, F.H. et al. (2001). Herpetology. Second edition. Prentice-Hall, Inc. New Jersey.USA. Pp 431-464. Senanayake, F.R. (1979). Notes on the lizards of the genus Ceratophora. Loris. 15(1):18-19. 175 COMPARISON OF HERPETOFAUNAL DIVERSITY IN THREE WETLAND IN HABITATS THE SOUTH-EAST COAST OF SRI LANKA R.K. RODRIGO1*, U.K.G.K. PADMALAL1, C.N.B. BAMBARADENIYA2, V.A.M.P.K SAMARAWICKRAMA2, N. PERERA2, M.S.J. PERERA2 and T.N. PERIES2 1 2 The Open University of Sri Lanka, Nawala, Nugegoda, Sri Lanka. IUCN-The World Conservation Union, 53, Horton Place, Colombo 7. Sri Lanka. Abstract This comparative study investigates the diversity of herpetofauna in Rekawa, Kalametiya-Lunama and Walawa wetland habitats in the South-East coast of Sri Lanka. Data were gathered in systematic manner at fortnight intervals during 1st October 2002 to 31st March 2003 in selected habitats in three sites. Fifteen amphibians of four families and 43 reptiles of 15 families were recorded in the study area. The endemic species noted consisted of one amphibian species (Rana gracilis) and five reptile species (Lankascincus fallax, Mabuya madaraszi, Sphenomorphus rufogulus, Lycodon osmanhilli and Xenochrophis asperrimus). A higher diversity of herpetofauna was found in Walawa (Shannon index = 0.233) compared with the other two sites (Rekawa=0.176, Lunama-Kalametiya = 0.223) possibly due to numerous habitats associated with the Walawe River. A number of threats to herpetofauna have been identified, among which clearance of habitats, collecting turtle eggs, spread of alien invasive flora and highway accidents can be highlighted. Sustainable development schemes, awareness programs conducted for the villagers and for the school children and mangrove reforestation are some of recommendation made to conserve the biodiversity of the area including the herpetofauna. KEY WORDS: HERPETOFAUNAL, DIVERSITY, REKAWA, LUNAMAKALAMETIYA, WALAWA ESTUARY, SOUTH-EAST COAST, SRI LANKA. Introduction Sri Lanka is an island located in the south-eastern tip of peninsular India, between northern latitudes 5° 55’ and 9° 51’ and eastern longitude 79° 41’ and 81° 51’. The total land area is approximately 65000km2. The land consists of three peneplains; the first (0-125m), the middle (125-750m) and the third (above750m) (Pemadasa, 1996). The island is divided in to four main climatic zones viz wet, dry, intermediate and arid zones according to the annual rainfall and temperature. In terms of “biodiversity per unit area”, Sri Lanka is ranked among the highest in Asia (Bambaradeniya, 2001a) and is considered as one of the 25 global biodiversity hot spots which reports diverse ecosystems resulted from a wide range of topographic and climatic variations (Myers et al., 2000). In fact, the island is a mega hotspot of herpetofauna harboring 103 amphibians and 181 reptiles, out of which several taxa are geographic relicts (Crusz, 1986; de Silva, 2001). Further more, when compared with other countries with high percentage of endemic reptilian taxa, Sri Lanka rank 4th the world (De Silva, 1996). 176 Sri Lanka having 103 species of amphibians, of which 87 of them are endemic to the island. Ichthyophiidae, Bufonidae, Microhylidae and Ranidae are the families recorded island wide from coastal areas up to highlands (Kirthisinghe, 1957; Dutta & Manamendra-Arachchi, 1996; Manamendra-Arachchi & Pethiyagoda, 1998; Manamendra-Arachchi & Pethiyagoda, 2001a; ManamendraArachchi & Pethiyagoda, 2001b, Gower & et.al., 2005; Manamendra-Arachchi & Pethiyagoda, 2005; Meegaskumbura & Manamendra -Arachchi, 2005). According to the published literature, Sri Lanka harbors a total of 81 inland tetrapod reptile species (48 endemic) (Deraniyagala, 1953; De Silva, 1980; Pethiyagoda & Manamendra-Arachchi, 1998; Greer, 1991; Bahir & Maduwage, 2005; Bahir & Silva, 2005; Batuwita & Bahir, 2005) and 82 inland serpentoid reptile species (44 endemic). In addition to the inland reptiles, 13 species of sea snakes and five species of marine turtles occur in the coastal waters of the island (De Silva, 1996). Among the tetrapod reptilia, three endemic genera of skinks (Lankascincus-6 species; Nessia-8 species and Chalcidoseps- 1 ) and three endemic genera of agamid lizards (Ceratophora-5 species; Lyriocephalus1species; Cophotis- 1 species) are considered as geographical relicts while one endemic Uropeltid genus (Pseudotyphlops- 1 species) and four endemic Colubrid genera (Aspidura- 5 species; Cercaspis- 1species; Haplocercus-1 species; Balanophis- 1 species) are recognized as that of serpentoid counterparts (Crusz, 1986; Greer, 1991). Of the total herpetofaunal species in Sri Lanka, 61% of amphibians (33 species) and 55.4% of reptiles (86 species) have been identified as nationally threatened (IUCN Sri Lanka, 2000). Study area Rekawa, Kalametiya-Lunama lagoon and Walawa estuary are situated in the South-East coast of Sri Lanka (Figure 1). These sites lie between Godawaya (N 06°06’26.2” E081°03’01.1”) and Tangalle (N6°02’19.7” E080°48’38.0”) N Walawa Lunama & Klametiya Figure 1. Map showing the study sites; Rekawa, Lunama & Kalametiya and Walawa. 177 Rekawa is situated on the border of the intermediate and dry climatic zones on the southern coast of Sri Lanka. The prominent feature of the area is that a large brackish water lagoon surrounded by the extensive mangrove vegetation, which is the dominant vegetation, bordering the lagoon extending over an area of about 250 hectares (Cooray, 1998). The two interconnected brackish water lagoons, namely Kalametiya (Kalametiya lagoon is mainly fresh water now) (606 ha) and Lunama (192 ha), both lagoons are parts of the Kalametiya sanctuary. Along the coast there is a low and narrow ridge of sand dunes separating the lagoons from the sea. The highest point of the area is Ussangoda hill, which is 20m above the sea level, and 1 km east of the Lunama Lagoon (CEA/Euro consult, 1995). Kalametiya lagoon can be roughly be divided into a silted and well-vegetated northern section and a southern part with open water surfaces. The northern section is only inundated during high floods with a maximum water depth of 1.25m (CEA/Euro consult, 1995). Lunama lagoon does not open into the sea. The lagoon is very shallow with a mean depth of 0.75m. It has a gently sloping bed, but in the south –west, the bank rises more abruptly. The bottom is covered by a thick package of organic matter and aquatic plants (CEA/Euro consult, 1995). Walawa estuary is a special habitat for the many of the flora and fauna found in the area. It makes different kinds of habits and vegetations types such as mangrove, marshlands, and bogs. Before reaching to the sea, Walawa River is running parallel with the sea from Godawaya to Welipatanwila. The special feature of the estuary is that its mouth is opened to sea only during the rainy seasons of the year. This is a preliminary study of the herpetofauna diversity in Rekawa, Kalametiya- Lunama and Walawa wetland areas. Being the first comparative herpetofaunal study in the selected area this research will enable to gather valuable baseline data. These eco-systems are highly fragmented and overexploited by man. It is very useful to monitor environmental variables in such a way that ecologists can detect environmental changes and relate the findings with the changes in herpetofaunal populations. For example, if the salinity and water level of a costal lagoon are recorded on a regular basis, an ecologist can extrapolate whether a particular measurement is within the typical range of fluctuations or a significant change has been occurred or being occurred based on ecological parameters (Sutherland, 1996). Amphibians and reptiles are very sensitive to the rapid environmental changes in the eco-systems (Pough et al, 2001); doing a survey on currently existing species of these taxa and obtaining data on ecological parameters from respective habitats, the information gathered would be applicable for the evaluation of different ecosystems in the area. Final out put of the study can also be used in continues monitoring of the herpetological diversity in the study area in future. 178 METETERIALS AND METHODS The data on herpetofaunal species were gathered from 1st October 2002 to 31 March 2003 systematically. Field sampling was carried out every other week; each sampling session spanned over six continuous days. Each selected site was sampled in the morning, noon, evening and night throughout the study period in order to avoid the time bias for a particular location in the case of transects. Rekawa, Lunama-Kalametiya and Walawa were the selected sites in the study area. Sampling locations were selected in each site considering the accessibility, representative habitats, special habitats and spatial distribution in the study area through an initial reconnaissance survey. st Pit fall traps (Sutherland, 2000) -5 baskets of 12 liter each were set along a belt transect at 20m x 5m. Traps were only activated for three days per a sampling session and were searched in the early morning, noon and late evening. Direct counts used for amphibians individuals at spawning sites in the study area was often used (Sutherland, 2000). Night Sampling was carried out especially for the survey of nocturnal herpetofauna in the study area. The sampling locations were randomly selected and the special habitats were searched from 1900hrs to 2400hrs using a spotlight. The time taken to study each location was dependant on the size of the area. Visual Encounter Method was used for herpetofauna in each site in the day and night (Wickramasinghe and Bambaradeniya, 2001). Belt Transect (100m x 5m transect) was laid in the selected locations in the study sites. Each transect was searched for a 2m height from the ground level. One replicate for each transect was done in an analogous habitat (location). Visual Observation for amphibians and reptiles, were observed during the sampling period. The total number of herpetofaunal individuals recorded at each location throughout the study period was taken into account for calculations. Following guides and keys were used for the species identification and nomenclature; for amphibians Dutta, Manamendra-Arachchi, (1996) and for reptiles Deraniyagala, (1953). De Silva, (1980), Wall, (1921). Site-specific data gathered during the survey was used to calculate Shannon Index. RESULTS Methods distinct habitats were identified in the study area for the herpetofaunal survey. The three study sites, Rekawa, Lunama-Kalametiya and Walawa, supported 8, 9 and 10 different habitats types respectively (Table 1). Table 1. Selected different habitat in the three study sites. Rekawa Coastal Sand Dune Coconut Plantation Grassland Home Garden Mangrove Associated Mangrove True Paddy Field Scrubland Lunama- Kalametiya Chena Coastal Sand Dune Grassland Home Garden Mangrove Associated Mangrove True Paddy Field Reed Bed Scrubland 179 Walawa Casuarina Coastal Sand Dune Coconut Plantation Grassland Home garden Mangrove Associated Marshland Paddy Field Salt Marsh Scrubland The twenty six herpetofaunal families recorded in Sri Lanka, the study area supported nineteen families. In Rekawa eight families of reptiles and three families of amphibians were recorded. There were nine families of reptiles and two families of amphibians recorded in Lunama-Kalametiya. The highest richness of herpetofaunal families was recorded from the Walawa site, which include families of reptile eleven and three families of amphibians (Table 2.). Table 2. Herpetofaunal families recorded from the study sites. * Present - Absent Family (recorded in Sri Lanka) Reptiles Crocodylidae Dermochelyidae Cheloniidae Bataguridae Testudinidae Trionychidae Agamidae Chamaeleonidae Gekkonidae Lacertidae Scincidae Varanidae Acrochordidae Typhlopidae Uropeltidae Boidae Colubridae Elapidae Hydrophiidae Viperidae Amphibians Ichthyophiidae Bufonidae Microhylidae Ranidae Rekawa Kalametiya-Lunama Walawa * * * * * * * * - * * * * * * * * * * * * * * * * * * * * - * * * * * * * * The study area represented 27% (15) amphibian and 25% (43) reptile species of Sri Lanka (Figure 2). The most common amphibian species recorded was Common Paddy field Frog (Limnonectes limnocharis) (242 individuals) while Fan Throat Lizard (Sitana ponticeriana) was the most common reptile (62 individuals) recorded in the study area. A rare species of turtle namely Leatherback Turtle (Dermochylus coreaceae) was observed in a sand dune in Godawaya while Scaly-fingered Gecko (Lepidodactylus lugubris) was also recorded in Rekawa. of fifteen amphibians and fourty three reptiles recorded in the study area, one species of amphibians and eight species of reptile are considered as threatened according to the IUCN national red list (IUCN Sri Lanka, 2000). 180 181 No.of Species 200 150 103 100 43 15 50 0 Amphibians Reptiles Sri Lanka Study Area Figure 2. Herpetofauna richness in the study area compared to Sri Lankan herpetofauna richness. No.of Species The number of amphibians recorded in the Rekawa, Lunama-Kalametiya and Walawa were ten, six and fourteen species respectively (Figure 3). The reptiles recorded in Rekawa, Lunama-Kalametiya and Walawa were 12, 21and 29 species respectively (Fig. 3). The Shannon diversity index for Rekawa =0.176, Lunama-Kalametiya =0.223 and Walawa =0.233. According to the Shannon diversity index highest species diversity was recorded in the Walawa and the lowest was in Rekawa. 35 30 25 20 15 10 5 0 Rekawa KalametiyaLunama Walawa Site Amphibians Reptiles Figure 3. Number of amphibians and reptiles species recorded from the different study sites. 181 Out of 38 endemic amphibian species recorded Sri Lankan Wood Frog (Rana gracilis) was the only endemic amphibian found in the study area whereas of 81 endamic reptile species recorded Common Lanka Skink (Lankascincus fallax), Red Throat Little Skink (Sphenomorphus rufogulus), Lycodon osmanhilli, Mabuya madaraszi and Common Pond Snake (Xenochorophis asperrimus) were the endemic. In all eleven different human impacts have been identified as possible threats to herpetofauna in the study area (Table 3.). Clearance of habitats for various needs including settlements, cultivations and industrial developments has affected the vegetation types of the area resulting in habitat fragmentation. Specific threats such as collecting turtle eggs, intentional killing of snakes and discriminate use of fishing nets contribute for direct loss of herpetofauna. A remarkable number of road kills occurred during late hours especially in the rainy season. However, Lunama-Kalametiya site has severely affected by human interferences compared to other study sites. Following table summarizes the threats to herpetofauna in different sites. Table 3. Comparative severity of identified threats to herpetofauna in different sites. Degree of human interference; - Nil + low ++ moderate Threat Rekawa +++ high ++++ very high LunamaKalametiya Walawa Clearance of habitats + +++ Shell mining. - ++++ - Coral Mining ++++ - - Poorly planned irrigation systems ++ +++ - ++ Discriminate use of agro-chemicals + ++ ++ Feral livestock + +++ + Collection of turtle eggs + ++++ ++++ Intentional Killing + ++ + Fishing activities + +++ ++ Highway accidents + +++ + Spread of alien species + ++++ +++ DISCUSSION The reptile fauna of the area occupies a wide range of ecological niches, such as sea (eg: turtles and sea snakes), freshwater (eg: terrapins and mugger crocodile), ground surface (eg: star tortoise and cobra), trees (eg: geckoes and green vine snake and green garden lizard) and soil (eg: Skins). Amphibians are also distributed in every niche except the sea & lagoon (Limnonectes limnocharis being aquatic, Polypedates maculatus being arboreal, Bufo fergusonii being terrestrial and Sphaerotheca rolandea being fossorial.) Some amphibian species such as Bufo fergusonii, Euphlyctis cyanophlyctis and Limnonectes limnocharis were very common in the study area while some were comparatively rare species such as (Uperodon systoma, Kaloula taprobanica, Philautus leucorhinus and Sphaerotheca rolandea.) When taking tetrapod reptiles in to the account, Varanus bengalensis, V. salvator and Calotes 182 versicolor were the most common species. Sitana ponticeriana was only recorded in Walawa study site restricted to a few habitats viz coconut and casuarina plantations and sand dunes. Dermochylus coreaceae, Crocodylus palustris, Lepidodactylus lugubris and Mabuya madaraszi were comparatively rare in the study sites. Discussions with the villagers led to conclude that D. coreaceae prefers sand dunes of Walawa rather than other sites as the area is relatively less affected by human disturbances. Python molurus, Xenochrophis asperrimus and Lycodon osmanhilli were also rare in the area and X. Piscator was the most abundant snake in the study sites. A survey on biodiversity status profile of Bundala national park conducted by Bambaradeniya et al (2002) revealed fifteen amphibians (1 sp. endemic) and forty eight reptiles (6 spp. endemic) of which one amphibian and thirteen species of and reptiles were listed as nationally threatened. It is interesting to note that Bundala being the closest national park and the wetland to the study area, to have an almost similar herpetofaunal richness. The diverse habitat types (n=10) in the Walawa river might be the possible reason for a higher diversity in the site. A new sight record of an endemic amphibian, Sri Lanka Wood Frog (Rana gracilis), was recorded from a home garden and from a casuarina plantation in Walawa. Another rare observation was made in Godawaya (Walawa) beach; a female leatherback turtle (Dermochelys coreaceae) in December 2002. Sixteen sea snakes are recorded from the coastal waters around Sri Lanka. Hydrophis cynocinctus, H. gracilis, H. Spiralis were common in Walawa. According to fishermen’s experience, abundance of these sea snakes can vary with the season. It is also worth to note that all the individuals found in the study were by catches from fishing nets. A previous study conducted in Lunama and Kalametiya reported thirty eight species of reptiles and there were no information on amphibians. These include fifteen serpentoids and twenty three tetrapod; ten species were endemic while twenty three species were nationally threatened (CEA/Euro consult, 1995). The same has recorded Bungarus ceylonicus, Crocodylus porosus, Calodactylodes illingworthorum, Cyrtodactylus sp., Hemidactylus depressus, Calotes, Dasia helianus, Mabuya beddomei, M. bibronii, which the present study could not record. The lowest herpetofaunal diversity was recorded from Rekawa (Fig.3). Five species of marine turtles (Dermochelys coriacea, Caretta, Lepidochelys olivacea, Eretmochelys imbricata and Chelonia mydas) are recorded in Rekawa (De Silva, 1997) but during the study period C. mydas was the only turtle observed. There are few studies have been conducted for sea turtles in the area. Out of the total individuals of turtles nest in Rekawa beach, 93.4% was C. mydas, 5% was D. Coriacea and 1.6% was C. Caretta during a three months period May-July 1994 (Cooray, 1998). 183 In Sri Lanka, mangroves are found scattered mainly along the north-western, north-eastern and eastern coasts bordering lagoons and river estuaries (De Silva and De Silva, 1998). The mangrove habitats support to accommodate vast number of fauna and flora and it gives many valuable products and uses such as tannin, lime, animal feed, medicine, food and beverages, firewood and timber and brush pile for lagoon fishery industry (Pinto, 1986). Most of mangrove forest areas have been subjected to human interference for a long time. As a result of human interference, Walawa, Lunama-Kalametiya sanctuary and Rekawa have been profoundly affected by the clearance of these mangrove habitats and associated vegetations for human settlements, chena cultivations and shell mining. Vast areas of undisturbed natural scrublands have been cleared for the establishment of an industrial zone in the Bata-Atha area, which is an important habitat for many rare herpetofaunal species such as Calliophis melanurus, Geochelone elegans etc. (pers.obs). In Walawa, coastal sand dunes and scrublands have been cleared for a coconut plantation and this can adversely affect the marine turtles which arrive at the beach in search for nesting habitats. A mangrove patch in Lunama area has been cleared for illegal shell mining. As a result of shell mining, large pits are created in the habitat and this has become a major problem for certain ground dwelling herpetofaunal species especially for terrapins and tortoises. Besides the direct habitat loss, species get trapped in these pits and eventually die. Fragmentation of habitats results in edges (Hunter, 1990) and this could be a major problem for the survival of herpetofauna species in the area. According to Hunter,(1990) certain species need large tracts of interior forests to outlive and edges might be detrimental for them in which competition, parasitism and predation could be higher and thus problematic. Due to human activities on these habitats the quality of composition and structure of the vegetation is lost and it has a direct impact on the abundance of herpetofauna. From a research on the effect of forest structure on amphibian abundance and diversity in the Chicago region, scientists have concluded that a high quality forest supports higher species richness and diversity than a low quality forest (Nuzzo and Mierzwa, 2000). Presence of appreciable sea grass beds, coral reefs, lagoons etc along the coastal waters around the island provide ample feeding localities for turtles (De Silva, 1997) and sea snakes (Allen and Steen, 1994). Illegal coral mining in the Rekawa coast line results in lost of feeding, breeding and nesting habitats for many sea snakes and marine turtles. Establishment of poorly planned irrigation systems in Lunama-Kalametiya area brings silt, agro-chemicals and garbage in to the lagoons and ultimately the drainage is directed to the sea and as a result coral habitats are polluted. Fishing activities, specially netting, cause direct injuries to the turtles and sea snakes following being trapped. A study on the effect of fishing on turtle populations (Jinadasa,1984) has showed that the sea turtles in the whole stretch of shore from Ambalangoda to Kalpitiya are caught unintentionally about 4-5 individuals per week during the heavy fishing seasons (non monsoon) and about 1 individual during the poor fishing season (monsoon). He further concludes the number caught for the entire country could be about 12 to 15 individuals per week during heavy fishing season and 3 individuals per week during poor fishing 184 season. Not only such fishing activities, but also collecting eggs and slaughtering for human consumption are the other threats pertaining to marine turtle populations (Santhiapillai, 2000, Wickremasinghe, 1981). This problem can be minimized by educating the villagers, fisherman and school children of the respective areas. Spread of alien invasive species is among the various threats to the biodiversity of Sri Lanka (Bambaradeniya, 2001b). Lunama-Kalametiya and Rekawa lagoons have been threatened by the distribution of Prosopis juliflora primarily and also by Opuntia dillenii and Eichhornia crassipes.The siltation decreases lagoon capacity and finally silted area is invaded by these monocultures. Due to large herds of cattle and feral buffalo in Ussangoda grassland and scrublands in Ussangoda and Kalametiya, habitat destruction is maximal. Soil erosion and disturbance to the vegetation are the direct results of unlimited grazing and physical damage caused by the hooves. Major and minor roads running through wetlands cause numerous road kills of diverse herpetofauna species. X. piscator, V. bengalensis and L. limnocharis were the most common species subjected to road kills. Such incidents were recorded more during the rainy season and often wetlands and paddy fields were the associate habitats found besides the roads (Rodrigo et. al., 2003; Bambaradeniya et. al., 2001). Some of the recommendations made for conservation of the herpetofauna in the area was; 1. Development schemes proposed in the area should be preplanned in order to minimize habitat destruction and to maintain habitats in a sustainable manner. 2. Mangrove reforestation can be implemented in the lagoons and estuaries of the area. 3. Awareness programs should be conducted for villagers and school children regarding every aspect of conservation (including herpetofauna and pertaining legislations). 4. Sign boards regarding animal crossing areas and underground tunnels connecting habitats on either side of the roads should be set to minimize the road kills. Acknowledgments Dr. Nayana Kariyawasam, the coordinator research project, the open university of Sri Lanka is acknowledged for support and encouragement. Our colleagues, Mr. Rohan Wasantha, Mr. Chamil Perera, Mr. Dayantha Withanage, Miss Rumesha Perera and Mr. Hasantha Sanjeewa are gratefully acknowledged for their generous support given during fieldwork. Mr. L.J. Mendis Wickramasinghe helped me in numerous ways during the research period thus; a special gratitude goes to him. This study was carried out as a part of a systematic biodiversity assessment project conducted in Rekawa/Ussangoda/Kalametiya coastal ecosystems by the IUCN Sri Lanka, funded by the Global Environment Facility program of the UNDP. 185 REFERENCES Allen, G. R. & Steen, R. (1994). Indo-Pacific coral reef field guide. Tropical Reef Research. Singapore. pp.351-353. Ashton, M. S., Gunathilake, S., De zoysa, N., Dassanayake, M. D., Gunathilake, N. and Wejesundera, S. (1997). A field guide to the common trees and shrubs of Sri Lanka WHT publications, Colombo, Sri Lanka. pp.9-11. Bahir, M.M. and Maduwage, K.P. (2005). Calotes desilvai, a new agamid lizard from morningside Forest, Sri Lanka. In: Yeo, D.C.J.; Ng, P.K.L. & Pethiyagoda, R (editors), Contributions to biodiversity exploration and research in Sri Lanka. The Raffles Bulletin of Zoology, Supplement No. 12: 381-392. Bahir, M.M. and Silva, A. (2005). Otocryptis nigristigma, a new species of agamid lizard from Sri Lanka. In: Yeo, D.C.J.; Ng, P.K.L. & Pethiyagoda, R (editors), Contributions to biodiversity exploration and research in Sri Lanka. The Raffles Bulletin of Zoology, Supplement No. 12: 393-406. Baldwin, M. F. (1991). Natural resources of Sri Lanka; conditions and trends. Natural Resources, Energy and Science Authority of Sri Lanka. pp.237-254. Bambaradeniya, C. N. B. (2001a). Threatened herpetofauna of Sri Lanka. in An overview of the threatened herpetofauna of South Asia .Eds. Bambaradeniya, C.N.B & Samarasekara, V.N. IUCN-The World Conservation Union, Sri Lanka. pp.91-101. Bambaradeniya, C. N. B. (2001b). The status and implications of alien invasive species in Sri Lanka. Zoos’ Print. 17: 930-935. Bambaradeniya, C.N.B., Wickramasinghe, L.J.M., Samarawickrama, V.A.P. and Kekulandala, L.D.C.B. (2001). Herpetofauna mortality in highways: A case study from Sri Lanka. Abstracts. Fourth World Congress of Herpetology. 10-11. Bambaradeniya, C. N. B., Ekanayake, S. P., Fernando. R. H. S. S., Perera, W. P. N. and Somaweera, R. (2002). A biodiversity status profile of Bundala national park- a Ramsar wetland in Sri Lanka. IUCN Sri Lanka. pp 28-29. Batuwita, S. and Bahir, M.M. (2005). Description of five new species of Cyrtodactylus (Reptilia: Gekkonidae) from Sri Lanka. In: Yeo, D.C.J.; Ng, P.K.L. & Pethiyagoda, R (editors), Contributions to biodiversity exploration and research in Sri Lanka. The Raffles Bulletin of Zoology, Supplement No. 12: 351-380. CEA/EUROCONSULT, (1995). Wetland site report and conservation management plan. Kalametiya and Lunama Kalapuwa. Central Environmental Authority, Sri Lanka. 17 p. CEA/EUROCONSULT, (1993). Wetland site report and conservation management plan. Bundala national park. Central Environmental Authority, Sri Lanka. pp.3-14. Cooray, R. (1998). The marine turtles in Rekawa; a survey on nesting activity and exploitation. Sri Lanka Naturalist. Young Zoologists’ Association of Sri Lanka. 02(1, 2): 1-8. Crusz, H. (1986). The vertebrates of Sri Lanka. Endamism and other aspects. Rep.soc. Res. Native Livestock. 11: 65-80. Deraniyagala, P. E. P (1953). A colored atlas of some vertebrates from Ceylon. Tetrapod Reptilia of Ceylon. Vol. 2. The Ceylon Government Press. 186 De silva, A. (2001). The herpetofauna of Sri Lanka: An historical overview, current status with checklists. Amphibia and Reptile Research Organization of Sri Lanka. De silva, A. (1997). Marine turtles of Sri Lanka: An overview. Lyriocephalus. 3: 16-25. De silva, A. (1996). The herpetofauna of Sri Lanka. A brief review. Published by the author. 10100. De silva, P. H. D. H. (1980). Snake fauna of Sri Lanka: with special reference to skull, dentition and venom in snakes. The National Museums Publications, Colombo. De silva, M and de silva, P. K. (1998). Status, diversity and conservation of the mangrove forests of Sri Lanka. J.South Asian Nat.Hist. Vol.3. No.01. WHT Sri Lanka, Colombo, Sri Lanka. 79-102. Dutta, S.K. and Manamendra-arachchi , K. (1996). The amphibian fauna of Sri Lanka. WHT Sri Lanka, Colombo, Sri Lanka. Gower, D.J., Bahir, M. M., Mapatuna, Y., Pethiyagoda, R., Raheem, D and wilkinson, M. (2005). Molecular phylogenetics of Sri Lankan Ichthyophis (Amphibia: Gymnophiona: Ichthyophiidae), with discovery of a cryptic species. The Raffles Bulletin of Zoology, Supplement. Supplement No. 12: 153–161. Greer, A. E. (1991). Lankascincus, a new genus of scincid lizards from Sri Lanka, with description of three new species. Journal of Herpetology. 25: 59-64. IUCN SRI LANKA, (2000). The 1999 list of threatened fauna and flora of Sri Lanka. IUCN Sri Lanka. 114p. Jinadasa, J. (1984). The effect of fishing on turtle populations. Loris. 16(6): 311-314. Kirthisinghe, P. (1957). The amphibians of Ceylon. Published by the author. Printed by the William Clowes and Sons, London. Hunter, M. L. (1990). Wildlife, forest and forestry. Regents/Prentice hall. New Jersey. 132p. Manamendra-arachchi, K. and Pethiyagoda, R. (1998). A synopsis of the Sri Lankan Bufonidae (Amphibia: Anura) with description of two new species. Journal of South Asian Natural History. 3: 213 – 248. Manamendra -arachchi, K. and Pethiyagoda, R. (2001a). Ramanella nagaoi, a new tree hole frog (Microhylidae) from Southern Sri Lanka. J. South Asian Nat. Hist. Vol.5. No. 2. WHT Sri Lanka, Colombo, Sri Lanka.121-133. Manamendra -arachchi, K. & Pethiyagoda, R. (2001b). Polypedates fastigo, a new tree frog ( Ranidae: Rhacophorinae) from Sri Lanka. J. South Asian Nat. Hist. WHT Sri Lanka, Colombo, Sri Lanka. 5(2): 191-199. Manamendra -arachchi, K. & Pethiyagoda, R. (2005). The Sri Lankan shrub-frogs of the genus Philautus Gistel, 1848 (Ranidae: Rhacophorinae), with description of 27 new species. The Raffles Bulletin of Zoology, Supplement. Supplement No. 12: 163–303. Meegaskumbura, M and Manamendra -arachchi, K (2005). Description of eight new species of shrub frogs (Ranidae: Racophorinae: Philautus) from Sri Lanka. The Raffles Bulletin of Zoology, Supplement. Supplement No. 12: 305–338. Myers, N., Mittermeier, R. A., Mittermeier, C. G., Fonseca, G. A. B., and Kents, J. (2000). Biodiversity hotspots for conservation priorities. Nature. 403: 853-858. 187 Nuzzo, V. A. and mierzwa, K. S. (2000). The effect of forest structure on amphibian abundance and diversity in the Chicago region. United States Environmental Protection Agency Great Lakes National Program Office (GLNPO). 77 W. Jackson Boulevard. Chicago. 121. Pemadasa, A. (1996). The Green Mantle of Sri Lanka. Department of Botany, University of Ruhuna, Matara, Sri Lanka. Pethiyagoda, R and Manamendra arachchi, K. (1998). A revision of the endemic Sri Lankan agamid lizard genus Ceratophora Gray, 1835, with description of two new species. J. South Asian Nat. Hist. Vol.3, No.1. WHT Sri Lanka, Colombo, Sri Lanka. 1-50. Pinto, L. (1986). Mangroves of Sri Lanka. Natural Resource, Energy and Science Authority of Sri Lanka. Colombo. Pough, F. H., Andrews, R. M, Cadle, J. E., Crump, M. L., Savitzky, A. H. and wells, K. D. (2001). Herpetology. Prentice Hall, Upper Saddle River, New Jersey.U.S.A. Rodrigo, R. K., Samarawickrama, V. A. M. P. K., Perera, N., Perera, M. S. J., Peries, T. N., Bambaradeniya, C. N. B. (2003). Herpetofauna mortality along Tangalle-Ambalanthota main road and minor roads in between in the South-East coast of Sri Lanka. Proceedings of the 23rd annual sessions of the Institute of Biology Sri Lanka. Colombo 07. 13p. Santhiapillai, C. (2000). Marine turtles of Sri Lanka; Their biology, diversity and conservation. Sri Lanka Nature. WHT Sri Lanka, Colombo, Sri Lanka. 2: 42-49. Sutherland, W. J (1996). Ecological census techniques. University press, Cambridge. Sutherland, W. J (2000). The conservation hand book: Research, management ans policy. University press, Cambridge. Wall, F, (1921). Snakes of Ceylon. Navrang Book Sellers and Publishers. New Delhi. India. Wickramasinghe, L. J. M. and Bambaradeniya, C. N. B. (2001). Composition and distribution of reptiles in the Walawa River basin of Sri Lanka. Abstracts. Fourth World Congress of Herpetology. 134-135. Wickramasinghe, S. (1981). Turtles and their conservation. Loris. 15: 313-315. 188 APPENDIX 1. Habitat, site, relative abundance and status of amphibians in the study are. Habitat: Ca= Casuarina Plantation; Ch= Chena; Co= Coconut Plantation; Cs= Coastal Sand Dune; Ho= Home Garden; Ma= Marshland; Mn= Mangrove; Pa= Paddy Field; Sc= Scrubland; Sm= Salt Marsh. Area: R=Rekawa; LK=Lunama-Kalametiya; W=Walawa Estuary. Relative Abundance: VC=Very Common; C=Common; UC=Uncommon; R=Rare; VR=Very Rare. Status in National Red List 1999: TR=Threatened * Endemic species Family Bufonidae Microhylidae Ranidae Scientific Name Bufo fergusonii Bufo melanostictus Uperodon systoma Microhyla ornata Microhyla rubra Kaloula taprobanica Sphaerotheca breviceps Sphaerotheca rolandae Limnonectes limnocharis Hoplobatrachus crassus Rana gracilis* Euphlyctis cyanophlyctic Euphlyctis hexadactylus Polypedates maculatus Philautus leucorhinus Commom Name Ferguson's Dwarf Toad Common Toad Baloon Frog Ornate Narrow- mouthed Frog Red Narrow-mouthed Frog Common Bull Frog Banded Sand Frog Marbled Sand Frog Common Paddy Field Frog Indian Bull Frog Sri Lankan Wood Frog Skipper Frog Six-toed Green Frog Chunam Tree-Frog - Habitat Sm, Ho, Ma, Co, Ch & Mn Ca & Ho Sm Ma, Co & Pa Co Ho & Co Co, Ho & Sc Gs Ma, Ho, Sc, Pa & Mn Sc, Ho, Gs & Pa Ho & Ca Ho, Pa, Ma, Sc, Mn, Gs & Ch P, Sc & Ho Gs Sm 189 Area W, LK & R W W&R W&R R W, R & K W & Ka W&R W, LK & R W, K & R W W, LK & R W, LK & R K W Relative Abundance VC C UC C UC UC C UC VC C VR VC C C VR Status in National Red List TR - APPENDIX 2. Habitat, site, relative abundance and status of reptiles in the study are. Habitat: Ca=Casuarina Plantation; Ch= Chena; Co= Coconut Plantation; Cs= Coastal Sand Dune; Ho= Home Garden; Ma= Marshland; Mn= Mangrove; Pa= Paddy Field; Ri= Riverine Vegetation; S=Sea; Sc= Scrubland; Sm= Salt Marsh Area: R=Rekawa; LK=Lunama-Kalametiya; W=Walawa Estuary. Relative Abundance: VC=Very Common; C=Common; UC=Uncommon; R=Rare; VR=Very Rare. Status in National Red List 1999: TR=Threatened * Endemic species Family Crocodylidae Trionychidae Bataguridae Testudinidae Cheloniidae Scientific Name Crocodylus palustris Lissemys punctata Melanochelys trijuga Geochelone elegans Chelonia mydas Lepidochelys olivacea Dermochelys coriacea Commom Name Mugger Flap Shell Turtle Parker's Black Turtle Star Tortoise Green Turtle Olive Ridley Sea Turtle Leatherback Turtle Varanidae Varanus salvator Varanus bengalensis Water Monitor Land Monitor Agamidae Calotes versicolor Sitana ponticeriana Heidactylus frenatus Hemidactylus brookii Hemidactylus triedrus Hemidactylus leshenaulti Gehyra mutilata Lepidodactylus lugubris Mabuya carinata Mabuya macularia Mabuya madaraszi* Lankascincus fallax* Riopa punctata Sphenomorphus rufogulus* Python molurus Ahaetulla nasutus Amphiasma stolata Atretium schistosum Boiga trigonata Cerberus rhynchops Dendrelaphis tristis Lycodan aulicus Lycodon osmanhilli* Lycodon striata Oligodon arnensis Ptyas mucosa Xenochorophis asperrimus* Xenochorophis piscator Bungarus caeruleus Hydrophis cynocinctus Hydrophis gracilis Hydrophis spiralis Naja Naja Daboia russelii Common Garden Lizard Fan Throated Lizard Common House Gecko Spotted House Gecko Termite Hill Gecko Bark Gecko Four Claw Gecko Scaly Finger Gecko Common Skink Bronze Green Little Skink Spotted Skink Common Lanka Skink Dotted Garden Skink Red Throat Little Skink Indian Python Green Vine Snake Buff-Striped Keelback The Olive Keelback Gamma Cat Snake Dog faced Water Snake Common Bronzeback The Common Wolf Snake Osmanhill’s Wolf Snake Shaw's Wolf Snake Kukri Snake The Common Rat Snake Commom Pond Snake Checkered Keelback The Common Indian Krait The Chittul John's Sea Snake The Narrow Banded Sea Snake Cobra Russell's Viper Dermochelyidae Gekkonidae Scincidae Boidae Colubridae Elapidae Viperidae 190 Habitat Ri Mn, Ma, Gs & Pa Pa & Mn Sc & Ma Cs S Cs Ca, Cs, Ma, Mn, Sc, Ho, P & Pa Sc, Gs, Mn, Ho & Ch Ho, Co, Ca, Cs, Mn, Sc & Ch Co, Ca & Cs Ho, Co, Sc & Ch Ho Gs & Sc Mn Ho Ho Ca, Co & Ho Sc Ho Ho Ch Ho Ho & Sc Ho Pa & Ho Pa Ho Mn Mn & Sc Sc Gs Ho & Sc Sc Pa & Sc Mn & Ma Pa & Ma Pa S S S Ho & Pa Pa & Ho Area W W, LK & R LK W R W W Relative Abundance R C UC R C C VR Status in TR TR TR TR - W, LK,& R W, LK,& R VC VC - W, LK,& R W W, LK,& R W LK & R LK & R K R W, LK & R W LK W LK W R W, LK & R W LK W W LK LK W W & LK W & LK W & LK W, LK & R W & LK LK W W W W & LK LK VC C VC VC C UC UC VR C UC VR UC UC VR VR UC UC R C UC UC R VR R UC UC VR C UC C VR C UC UC TR TR TR TR - EFFECT OF CLIMATE CHANGE ON FLORA AND FAUNA OF YALA AND BUNDALA NATIONAL PARKS IN SOUTHERN SRI LANKA S.P. SAMARAKOON* Department of Botany, University of Ruhuna, Matara, Sri Lanka ABSTRACT An attempt was made in 2003 to identify the climate variability and its effects on flora and fauna in Hambantota. The objectives of the study were (1) to extract the patterns of climate variability, (2) to construct a “Climate Trend Diagram (CTD)” and (3) to study the effect of climate change on flora and fauna. Rainfall and temperature data (1869 - 2002) were collected from Colombo Meteorological Department and analyzed using Statistical Packages to find the cycles of relatively similar patterns and trend lines for each cycle. Information on flora and fauna were collected from desk-top studies, and field visits. The mean annual rainfall and temperature at Hambantota were 1022.4 mm, and 27.10C respectively. Generally, the rainfall was declining and temperature was increasing since 1970s. November showed the highest rainfall of 183.1 mm and the driest months were February, July, and August. The actual rainfall and the mean temperature showed patterns of climatic cycles of 16years and 6-years respectively. Since 1970s, the CTD showed that the mean annual temperature has been increasing and the rainfall decreasing. Presumably, the prolonged drought has decreased ground water and increased scrub species. Plants and animals, which have relatively higher demand for water, were the most threatened species. The long-term effects of climate change were reflected by forest die-back, and spread of woody weeds in Bundala and Yala National Parks. INTRODUCTION According to the information available throughout the history it is apparent that the climate of different parts of the earth has been changing and such changes are generally referred to as natural climate changes or more precisely climate variability (IPCC, 2001; IPCC, 2004). Usually these changes occur in the atmosphere, but the effects of such changes affect on other environmental components including both biotic and a-biotic parts. United Nations Framework Convention on Climate Change (United Nations, 1992; Orlando, and Smeardon, 1999) defines climate change as "a change of climate which is attributed directly or indirectly to human activity that alters the composition of the global atmosphere and which is in addition to natural climate variability observed over comparable time periods”. Atmosphere is an essential part of the physical and chemical environment for life. Changes, anthropogenic or otherwise, to the physical and chemical properties of the atmosphere have the potential of affecting directly the quality of life and even the very existence of life. Climate generally defined as the pattern or cycle of weather conditions such as wind, rain, snowfall, humidity, and clouds, including extreme or occasional ones, over a large area, averaged over many years. During the past century or so with the increased use of mineral oils and emission of green house gases and the destruction of natural environment the man has been identified as the major factor behind the change in natural climatic changes. Sri Lanka as an island country in the tropics, its vulnerability to the effect of global climatic changes 191 such as increased atmospheric temperature becomes serious. This is especially true for Hambantota area (South Eastern corner of the island) compared to other areas of the country because of its inherent dry, high temperature and low rainfall situation despite its ancient historical settlements (Perera, 1965). The present study is aimed to identify the climate variability patterns and the advent of the serious anthropogenic modern climate change and the vulnerability of flora, and fauna of Yala and Bundala National Parks in Hambantota District to the drastic effects of climate changes. OBJECTIVES (1) To study the patterns of climate variability since 1869-2002 in Hambantota Area. Hambantota has weather recordings of rainfall and temperature since the year 1869. Although daily maximum and minimum temperatures appear more serious effects on the vegetation and fauna for most of the period only mean temperature values are available; mean monthly temperature gives only marginal understanding of the effects while annual mean gives still less representation of the real temperature variation. (2) To identify and extract rainfall and temperature cycles for Hambantota District. (3) To study the variation of rainfall and temperature trend lines for each of the climate cycle. (4) To construct a pictorial representation of both monthly rainfall and temperature continuous variability for 134 years from January 1869 to December 2002. (5) To identify the advent of major climate change in Hambantota. (6) To study the effect of climate change on flora and fauna of National Parks of Hambantota Area. 192 MATERIALS AND METHODS Extraction of climate cycles using mean monthly rainfall and temperature (continuous) in Hambantota Monthly rainfall and temperature data of Hambantota were collected from Colombo Meteorological Department for the period from 1869 to 2002. These data were analyzed using SPSS Statistical Package and Microsoft Excel to find the patterns of monthly changes, mean individual monthly changes, continuous monthly changes and patterns of annual changes. Periods (cycles) of relatively similar variability patterns for rainfall and temperature were extracted. Climate cycle trend lines For each climate cycle the trend lines were drawn using the Excel Programme to show the climate trends. Construction of climate trend diagram Using the extracted rainfall and temperature trend lines a “Climate Trend Diagram” was constructed for Hambantota. The effect of climate change on flora and fauna of the national parks Information on flora and fauna of the Yala and Bundala National Parks was collected as follows: i. Desk-top studies of the previous work: For the desk-top study information were collected from (a) the Department of Wildlife, and (b) my previous studies of the area and (c) Interviewing the relevant personnel of the park. ii. Field studies of the different areas of the national Parks The changes in the botanical and faunal composition during the recent past (! 990 - 2002) were noted. Special attention was paid to study the spread of woody weeds, forest die-back and the presence of large herds of domesticated and feral cattle in the park area. Information regarding the vulnerability of flora and fauna to changing climate and potential climate related problems were collected through the historical and field studies of the occurrence of wild fire, spread of woody weeds (Lantana camara, Prosopis juliflora), possibility of the spread of giant panic grass (Panicum maximum), forest dieback (Manilkara hexandra, Drypetes sepiaria, Salvadora persica, and availability of water, changes in the botanical composition, and encroachment of the parks by the feral cattle and buffaloes. The Climate Trend Diagram was used to identify the human influenced climate change in Hambantota. The points of deviation from the normal patterns of variation in the rainfall and temperature cycles were considered as the points of the commencement of severe human influence on the climate variability. 193 RESULTS AND DISCUSSION Analysis of climate data Descriptive statistics The central tendencies in terms of the mean and the dispersion in terms of range, minimum and maximum values, standard deviation and variance are shown in the Table 1. With a mean of 183 mm the month of November showed the highest rainfall at Hambantota, followed by December (128 mm) and October (127 mm). Since 1869 the minimum, maximum and mean values of monthly rainfall at Hambantota in October, November and December were 1, 10, 2 (mm); 564, 472, 486; and 127, 183, and 128 (mm) respectively (Table 1). In all the other months the mean rainfall were less than 100 mm and the lowest rainfall were experienced in the months of February, July and August with less thank 50 mm. Table 1. Descriptive Statistics for rainfall (mm) data (1869 - 2002) Hambantota Month JANUARY FEBRUARY MARCH APRIL MAY JUNE JULY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER TOTAL Minimum mm 0 0 0 1 0 1 0 0 0 1 10 2 489 Maximum mm 376 273 258 333 512 235 224 214 452 564 472 486 2063 Mean mm 83 44 63 92 93 59 44 46 64 127 183 128 1013 Standard Error 6 4 5 5 7 4 4 4 6 8 8 7 21 Standard Deviation 74 50 55 62 85 49 44 42 65 92 95 85 247 Table 2. Descriptive Statistics for temperature (0C) data (1869 - 2002) Hambantota Month JANUARY FEBRUARY MARCH APRIL MAY JUNE JULY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER Minimum (0C) 24.7 24.0 25.8 25.7 25.8 25.2 25.6 25.9 26.2 25.6 25.4 24.6 Maximum (0C) 32.4 33.2 33.7 34.1 34.5 34.4 34.1 33.8 33.2 33.1 32.5 31.9 Mean (0C) 26.6 27.0 27.8 28.4 28.5 28.2 28.3 28.1 27.9 27.7 27.2 26.6 Classification of months and years based on rainfall data 194 Standard Deviation 1.8 1.8 1.8 1.7 1.7 1.8 2.0 1.8 1.7 1.7 1.6 1.7 (a) Classification of cases (years) The Dendrogram produced by the hierarchical analysis of the mean annual rainfall and temperature data of Hambantota (1869-1980) did not show meaningful cluster separation. (b) Classification of variables (months) The Dendrogram produced by the hierarchical cluster analysis to classify the variables (months) using the mean monthly rainfall data of Hambantota (1869-1980) is shown in the Figure 1. Figure 1. Classification of months based on the rainfall data Based on mean monthly rainfall the variables (months) can be classified into 4 groups and 7 cycles (spells) as follows: Group 1: February, March. June, July, August, and September (44-64 mm of rainfall): Two cycles (spells) of drought, (a) February and March, and (b) June, July, August, and September Group 2: January, April, and May (83-93 mm of rainfall): Two cycles (spells) with Moderate rainfall (a) January, and (b) April and May 195 Group 3: October and December (127-128 mm of rainfall): Two cycles (spells) with High rainfall (a) October, and (b) December Group 4: November (183 mm of rainfall): One cycle with very high rainfall Classification of months and years based on temperature data The Dendrogram produced by the hierarchical cluster analysis to classify the variables (months) using the mean monthly temperature data of Hambantota (1869-1980) is shown in the Figure 2. ******HIERARCHICALCLUSTER ANALYSIS****** Dendrogram using Average Linkage (Between Groups) Rescaled Distance Cluster Combine Figure 2. Classification of months based on the mean monthly temperature data Based on mean monthly temperature the variables (months) can be classified into 3 groups and 7 cycles (spells) as follows: Group 1: November, December, January, February (26.6-27.2 0C): One cycle (spell) of mild temperature Group 2: March, August, September, October (27.7 – 28.1 0C): Two cycles (spells) with Moderately high temperature (a) March, and (b) August, September, and October Group 3: April May, June, and July (28.1– 28.5 0C): One cycle (spell) of high Temperature On the other hand the minimum mean temperatures were recorded for the months of November, December and January and maximum means for May, April, July, June and 196 August (Table 2). As such under the influence of low rainfall and the high temperatures the driest months in Hambantota are June and July followed by August and September. Rainfall patterns (a) Mean Annual Rainfall Except around the 9th Year (1877), generally the graph of mean annual rainfall looked somewhat arc-like (Figure 3). About 85% of the peaks appeared above the 1000 mm rainfall line and one peak has risen above the 2000 mm level. About 5 years (approximately 11% of the years) the rainfall had been very low (the inverted peaks). Generally, the shape of the curve showed that the rainfall had been declining during the last 30 years (from about the year number 100) or so indicating a severe long drought in Hambantota. Figure 3. Mean annual rainfall of Hambantota plotted against year. (b) Mean Continuous Monthly Rainfall (January 1869 to December 2002) Figure 4. Monthly rainfall of Hambantota (1869-2002) plotted against number of the month When monthly rainfall was plotted against the number of the month starting from January 1869 and ending at December 2002 some interesting patterns were resulted (Figure 4-11). The month number 69 (1874 November) showed a high peak of more than 300 mm of rainfall. Similarly, the graph showed a general pattern of recurrence of high rainfall over 200 mm, which had a cycle of approximately 25months. The frequency of the rainfalls over 300 mm is approximately 30 years. An interesting observation of Figures 4-11 is that the overall monthly variation of rainfall has distinct units of periods with consistent duration of 192 months (16 years approximately). The overall trend line of rainfall during the period (monthly values from the year 1869 to 2002) was neither positive nor negative (Figure 4). A long band of rainfall spectrum was used to extract rainfall cycles ( ) which are shown in the concise diagram. 197 Figure 5. Monthly rainfall of Hambantota (1869-1893) plotted against number of the month Figure 6. Monthly rainfall of Hambantota (1894-1919) plotted against number of the month Figure 7. Monthly rainfall of Hambantota (1920-1944) plotted against number of the month 198 Figure 8. Monthly rainfall of Hambantota (1945-1969) plotted against number of the month Figure 9. Monthly rainfall of Hambantota (1970-1994) Figure 10. Monthly rainfall of Hambantota (1995-2002) 199 Temperature patterns Monthly temperature values were plotted against the number of the month starting from January 1869 and ending at December 2002 and the results are presented in the Figure 3.8. An interesting observation of it is that the overall monthly variation of temperature has distinct units of periods with consistent duration of 75 months (6 years approximately). The overall temperature trend line during the period (monthly values from the year 1869 to 2002) was positive. As for rainfall data a long band of temperature spectrum was used to extract temperature cycles (Figure 3.9). Temperature variation patterns for 25 year periods are shown in figures from 3.9 - 3.14 sequentially. Each temperature cycle has its own inherent features however apparently the duration is the same (6-years). Figure 11. Continuous monthly mean temperature of Hambantota (1969-2002) Figure 12. Monthly temperature of Hambantota (1869-1894) 200 Figure 13. Monthly temperature of Hambantota (1895-1919) Figure 14. Monthly temperature of Hambantota (1921-1944) 201 Figure 15. Monthly temperature of Hambantota (1945-1969) Figure 16. Monthly temperature of Hambantota (1970-1994) Figure 17. Monthly temperature of Hambantota (1995-2002) Construction of climate trend diagram (Figure 4) As noted in the previous sections, after careful observations of the rainfall and temperature graphs it was possible to extract patterns of climatic cycles. It was found that there was a conspicuous rainfall cycle which occurred in 16-year (192 months) periods. Similarly, the continuous monthly temperature curve suggests temperature cycles of approximately 6-year (75 months) duration. Within a cycle, either rainfall or temperature, there was more or less a similar pattern of fluctuations. For each cycle a graph (rainfall in mm verses month, and temperature 0C verses month) was plotted and the trend of each graph (positive, negative, or neutral) was extracted and such trend lines were used to construct the Climate Trend Diagram (CTD). Climate data of all main climate stations were analyzed and a climate diagram for each was constructed in the previous studies of the variability of climate of Sri Lanka (Mueller-Dombois, 1968). However, the use of climate trend lines in the present study appears to be a new approach to obtain more information using monthly data of 133 years. 202 The main features of the CTD (Figure 4) were as follows: (1) Left vertical line (Y axis) represents monthly rainfall (2 cm = 100 mm), (2) Right vertical line (Y axis) represents mean monthly temperature (1 cm = 1 0C), (3) Lower horizontal line (X-axis) represents time in months since the year 1869. It is graduated in temperature cycles of approximately 6 years (75 months), and (4) Upper horizontal line (X-axis) represents time in months since the year 1869. It is graduated in rainfall cycles of approximately 16 years (192 months). The diagram clearly shows that during the last 30 years the rainfall has been decreasing (the thick trend lines) and the temperature increasing (the thin trend lines) in Hambantota, an indication of severe drought in the area; further the temperature trend line has surpassed the rainfall trend line during the last 10 years. Such a situation may be directly attributable to humanly influenced climate change during that period, rather than natural climatic variations. Although rainfall trend line was horizontal during the last cycle the good news is that the raw data suggests a slightly positive trend since 1997 up to the present days (not shown in the diagrams). Figure 4 Climate (rainfall and temperature) Trend Diagram (CTD) for Hambantota showing the variation of climate through the last 133 years from 1869 to 2002. Figure 18. Climate (rainfall and temperature) Trend Diagram (CTD) for Hambantota 203 Effect of climate change on flora and fauna of Yala and Bundala National Parks Principal habitats and vegetation in Yala and Bundala of and Hambantota In Yala and Bundala National Parks the principal habitat types present are (1) dense forests, (2) open forests, (3) scrub forests, (4) grasslands, (5) fresh water tanks, (6) lagoons, and (7) sand dunes (Survey Department, 1988). Grasslands in the area are of two types, viz., Pitiyas, which occur usually around the Wewa (artificial tanks) and Pelessas associated with open forests and mainly located towards the coastal belt in areas such as the Block I and Yala east (Samarakoon, 1997, 1998). Closed canopy dense forests in Yala National Park are generally located in the northern areas (Block IV and V) and along the banks of River Menik, and Kumbukkan Oya (Riverine forests) (Samarakoon, 1998). In Bundala closed canopy dense forests are not found and its forests are either the open canopy type or the scrub type. It was found that certain forests and grasslands habitats in Bundala National Park are changing at a faster rate due to introduction of serious alian weedy species (Mesquite and Lantana) coupled with sea level rise as a result of global climate change. Mesquite has already conquered many low-lying areas replacing the fringe mangroves in lagoons and the natural vegetation in the grassland park country around the lagoons. As a result a large number of previously grassland type and deciduous type scrub forest habitats with a grassland component (Pelessa) have been changed into a more or less closed canopy single species thorny forest, which is usually devoid of a grass cover on the ground. Despite the fact that mesquite is a palatable species browsed by the elephants and deer the presence of large thorns gives it a negative value; on the other hand the changed habitat does not produce sufficient feed material for the domesticated buffaloes and neat cattle. It appears the end results are: 1) migration of the domesticated grazing animals into the grasslands in the Yala National Park causing severe problems to wild animals, and 2) human elephant conflict. Flora of Yala and Bundala National Parks Within the park a total of about 280 species of flowering plants have been recorded representing large trees, shrubs, and herbs. A list of common trees and shrubs found in the Yala and Bundala National Parks is shown in the Table 3. In the present study a total of 9 tree species and 8 shrubs species were noted as commonly found in the two parks. It was noted that Lantana camara has penetrated into the deep forests, almost every part of Yala and Bundala, and has become the dominant thorny shrub particularly in the scrub forest habitats. Its presence and dominance especially in the open forest areas has displaced the natural shrub species which are a major component of the herbage eaten by the large mammals including the elephant. It has, during the last century, changed the botanical composition of the flora and the habitats of the park country. 204 Table 3. List of common trees and shrubs in Yala and Bundala National Parks * Not in Bundala and coastal areas of Yala Species Vernacular name Trees Chloroxylon swietenia* Vitex pinnata Manilkara hexandra Drypetes sepiaria Feronia limoponica Azardracta indica Tamarindus indica Berrya cordifolia* Terminalia arjuna Burutha Milla Palu Weera Divul Kohomba Siyambala Halmilla Kumbuk Species Vernacular name Shrubs Cassia auriculata Salvadora persica Carissa spinarum Randia dumetorum Gymnema sylvestre Securiniga leucopyrus Lantana camara Prosopis juliflora Ranawara Maliththan Karamba Kukuruman Mas-bedda Katupila Gandapana (Weed) Mesquite (Weed) Fauna of Yala and Bundala National Parks Animal numbers were obtained from the Yala and Bundala National Parks and the department of Wildlife Conservation in Colombo. Of the large mammals highest number estimated is for the spotted deer (5,205) and the lowest for leopard and sloth bear (Table 4). In the whole of Yala Protected Areas Complex (YPC) the estimated number of elephants is a meager 319. In addition to the large animals the author has observed many cattle (neat type), mostly domesticated, are grazing within the YPC. Lists of other mammals, birds and other animals are given in the Table 5. Table 4. Large mammals and their numbers in Yala Protected Area Complex Source: Department of Wildlife Conservation Scientific name Vernacular Name Elephas maximus Bubalus bubalis Cervus unicolor Axis axis Sus scropfa Panthera pardus kotiya Melurus ursinus Elephant Buffalo Sambur Spotted deer Pig Leopard Sloth Bear 205 Estimated number in the YPC 319 1,749 898 5,205 637 124 125 Table 5. List of the other mammals, birds (including the migratory ones), and other animals commonly found in Yala and Bundala National Parks. Scientific name Mammals: Loris tardigradus Canis aurius Tragulis meminna Vivericula indica Canis aurius Prebites entellus Paradoxurus zeylonensis Manis crassicaudata Histrix indica Lepus negricollis singhala Other Animals: Testudo elegance Calotes versicolor Calotes calotes Calotes ceylonensis Varanus bengalensis Varanus salvator Lankascinus fallax Python molurus Crocodilus palustris Crocodilus porosus Bufo athukoralei Limnonectes greenii Vernacular Name Una-hapuluwa Nariya Miminna Urulawa Hiwala Wandura Kalawedda Kaballawa Ittawa Hawa Taraka-ibba Gara-katussa Pala-katussa Thola-visithurukatussa Thala-goya Kabara-goya Dumburu-hiraluwa Pimbura Kimbula Geta-kimbula Athukoralagekurugemba Lanka-welmediya Scientific name Birds: Gallus lafayetti Galloperdix bicalcarata Ephippiorhynchus asiaticus Letoptilosjavanicus Pelecanus phillipensis Phalacrocorax niger Mycteria leucocephala Bubulcus ibis Egretta garzetta Mesophoyx intermedia Casmerodius albus Ardeola grayii Migratory Birds: Tringa hypoleucos Dendronanthus indicus Limosa limosa Pitta brachyura Pluvialis fulva Phoenicopterus ruber 206 Vernacular Name Wali-kukula Haban-kukula Ali-manawa Bahurumanawa Alu-pastuduwa Diya-kawa Latu-wekiya Gawa-koka Kuda-ali-koka Sudu-medikoka Maha-sudukoka Kana-koka Silibilla Helapenda Penda-kalugohonduwithth a Awichchiya Ranmahaolevia Siyakkaraya Vulnerability of flora and fauna Demand for water A list of major plant and animal species found in the area which demand relatively more water for their sustenance is shown in the Table 6. Table 6. List of major plant and animal species considered having high demand for water. Scientific name Plants: Vitex pinnata Manilkara hexandra Drypetes sepiaria Azardracta indica Tamarindus indica Terminalia arjuna Vernacular name Milla Palu Scientific name Animals: Elephas maximus Bubalus bubalis Cervus unicolor Vernacular name Elephant Buffalo Sambhur Weera Kohomba Siyambala Kumbuk The trees listed are evergreen and relatively use more water for transpiration. Terminalia arjuna, usually grows on the riverbanks and benefited by the availability of soil water even during the long droughts; the other species of trees are adapted to grow mainly on hilly mounds where there is more silt in with higher water holding capacity compared to sandy soils in the valleys amongst the hills, where only grass and strongly zerophytic plants (e.g., Acacia eburnean) survive. Generally, in some Blocks of Yala and almost all the areas of Bundala the vegetation is a mosaic of forested (or scrub forest) hills and valleys of grassland with scattered thorny trees. Such a system had been surviving in the area for hundreds of years in equilibrium with the climate. However, with the onset of serious climate change in 70s annual drought became more and more severe and the new climate affected the natural vegetation by way of replacing some tree species (e.g., Manilkara hexandra, and Drypetes sepiaria) which cannot survive during the severe droughts due to lack of sufficient water. Similarly, the animal species listed in the table are large and need more water for the maintenance of transpiration and normal bodily needs. Further such animals prefer to stay in water and muddy habitats at least few hours a day to avoid high temperatures during the day time. Plant and animal species, which have higher demand for water, appear more threatened by the effect of climate change. 207 Woody weeds Another consequence of habitat destruction (death of tree species) is the introduction of woody weeds such as Lantana camara, mesquit and other thorny shrubs, which demand less water and spread very fast. Further, generally they are not good feed material for the animals such as the elephants; however, I have observed that both species provide food for birds. The effect of habitat change due to climate on the large animals is immediately visible in many areas of both parks. However, although not easily observable, the effect of climate change on smaller mammals and other vertebrates for example the amphibians and the like is more drastic. Some species require moist and watery places for the completion of their life cycles. With the destruction of their habitats (depletion of water in the streams and ponds for example) such animals become extinct in the area or become imprisoned, isolated and restricted to places where the environmental condition is still remain suitable. Forest dieback Forest dieback is a phenomenon occur in certain areas, for example the Horton Plains Montane Forest) of the country, reported since 1960s, 70s, 80s and 90s (Samarakoon, 1994). I have observed the same phenomenon occurring in Hambantota and its surroundings since early 1990s. The death of patches of Maliththan (Salvadora persica) and Weera Trees (Drypetes sepiaria) was seen in Bundala National Park (Figure 5). The dead or dying patches were noted closer to the water front of the lagoon area. Since 1996 the death of Palu trees (Manilkara hexandra) was observed in that area and the death of the mangrove species Kirala (Sonneratia casiolaris) was observed in the Menik Ganga estuary at Yala (Figure 6). As at present the Weera trees are very rare in the affected areas and a large number of Palu trees are dead and similar numbers are dying. The apparent reasons behind the death of these species are the increase in the mean sea level with the increase in earth atmospheric temperature and unavailability of sufficient fresh water during the long severe droughts. 208 Figure 5. Map showing the die-back areas of Palu (Manilkara hexandra) trees in Bundala National Park Figure 6. Map showing the die-back areas of Kirala (Sonneratia caseolaris) trees in Menik River estuary at Yala 209 Crop failure In parallel with the death of forest population water availability for paddy and other cultivations in Hambantota became very low especially during the last 20 years or so. Keep in mind that a sufficient water supply could be provided for the crops government constructed vast irrigation schemes such as the Lunugamvehera and other reservoirs in late 1980s early 1990s. With the declining rainfall and increasing temperatures such reservoirs were never filled even during the rainy season. Thousands of crop fields were devastated. The farmers became poorer and poorer; some changed their way of earnings; to earn their living some secretly cultivate Cannabis sativa in the jungles; some others encroached into the national parks and hunted animals and yet others involve in some other practices. Most of these practices created social problems. CONCLUSIONS (1) Climate change associated with global temperature increase and rainfall fluctuation causes drastic changes to the habitats of Hambantota area (2) Throughout the climatic history of 134 years in Hambantota it is possible to identify rainfall cycles of approximately 16 years and temperature cycles of approximately 6 years. (3) Since 1869 to about the year 1965 the rainfall in Hambantota had been increasing and showed marked positive trends in consecutive cycles. (4) Since around 1970 the mean rainfall declined and the rainfall trends of the cycles were negative or more or less neutral. (5) The mean monthly temperatures around the year 1900 were very high and above the level of 28 0C. The temperature effected drought prevailed for a brief period of about one year. (6) In general the average temperature trend during the last 20 years or so had been on the increase. (7) Increase in sea water levels causes sea water intrusion into the lagoons and estuary systems which lead to the destruction and slow change of the existing habitats. (8) Increased temperature and relatively low rainfall for a long period of about 20 years or so in Hambantota means decrease in ground water levels, decrease in water availability for plants and animals. (9) As mitigate actions measures such as the following are suggested: (i) Rehabilitation of the ancient tanks in the area (ii) Instead of cultivating high water demanding crops to cultivate less water demanding crops (iii) Instead of practicing animal husbandry using high water demanding animals use less water demanding crops (iv) Try to make use of the dry situation as a resource of high value (E.g. to produce more salt, dry fish) 210 ACKNOWLEDGEMENTS I thank Dr. B.M.S. Batagoda, Project Director and Mr. T.K. Fernando, Consultant of the Climate Change Enabling Activity (Phase II) Project of the Ministry of Environment and Natural Resources, “Sampathpaya”, No. 82, Rajamalwatta Road, Battaramulla for providing the necessary funds for this study. REFERENCES IPCC (2001): Climate Change (2001): Synthesis Report. A Contribution of Working Groups I, II, and III to the Third Assessment Report of the Intergovernmental Panel on Climate Change [Watson, R.T. and the Core Writing Team (eds.)]. Cambridge University Press, Cambridge,United Kingdom, and New York, NY, USA, 398 p. IPCC (2004) IPCC Workshop on Describing Scientific Uncertainties in Climate Change to Support Analysis of Risk and of Options, SWITZERLAND INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE IPCC Secretariat, c/o WMO, 7bis, Avenue de la Paix, C.P. N° 2300, 1211 Geneva 2 Ireland, May 11 – 13, 2004 Mueller-Dombois, D. (1968). Ecogeographic analysis of a climate map of Ceylon with particular reference to vegetation The Ceylon Forester 8 (2-4): 39-58. Orlando, B.M. and L. Smeardon (eds.) (1999) Report of the Eleventh Global Biodiversity Forum: Exploring Synergy Between the UN Framework Convention on Climate Change and the Convention on Biological Diversity. IUCN—the World Conservation Union, Gland, Switzerland and Cambridge, United Kingdom, 46 p. Perera, L.S. (1965) Rohana Rajyaya (Kingdom of Rohana), Anuradhapura Period (Vidyankankaaa University Press. Samaranayake, R. (1983). Hambantota A Profile of a District in Rural Sri Lanka Ministry of Plan Implementation Sri Lanka Samarakoon, S. P. (1994). Dying of trees in Horton Plains Bio News 8:106-109. Samarakoon, S. P. (1997). A preliminary investigation of the plant communities of Nimalawa sanctuary in southern Sri Lanka Proceedings of the Sri Lanka Association for the Advancement of Science. 249p. Samarakoon, S. P. (1997). Vegetation types of Lunugamvehera National Park. Proceedings of the Sri Lanka Association for the Advancement of Science. 250p. Samarakoon, S. P. (1998). Fodder plants of Yala National Park Complex in southern Sri Lanka. Proceedings of the Sri Lanka Association for the Advancement of Science. 128-129. Samarakoon, S. P. (1998). Herbage production from native grasses and planted pastures under coconut in southern Sri Lanka. Proceedings of the Sri Lanka Association for the Advancement of Science. 130131. Samarakoon, S. P. (1998). Some preliminary observations on the ecology of Kirinda sand dune in southern Sri Lanka. Proceedings of the Sri Lanka Association for the Advancement of Science. 198p. Survey Department, (1988). 1:50,000 Maps compiled and published by the Surveyor General of Sri Lanka. 211 MARINE TURTLE CONCERVATION PROJECT AT BUNDALA NATIONAL PARK I.H.S.K. DE SILVA* Bundala National Park,Department of Wildlife Conservation, Sri Lanka ABSTRACT Five of the world's seven species of marine turtle come ashore to nest in the beaches of Sri Lanka. All five species are listed as endangered by the world conservation union (IUCN), and protected under national law. All five species of marine turtles visit Sri Lanka nest at Bundala beach, which is about 16 Km in length every year during the turtle nesing period. Bundala cost line host nesting endangered Olive Ridley (Lepidochelys olivaceae), Green turtle (Chelonia mydas), Loggerhead turtle (Caretta caretta), Hawksbill turtle (Dermochelys imbricata) and Leatherback turtle (Dermochelys Coriaceae) Leatherback turtle is critically endangered species in the world. Green turtles and Olive Readly turtles both visited Bundala beach for nesting during last four years Olive Readly nested three hundred and nine times (309) for the last four years. Year 2004 and year 2005 was not recorded any nesting sites of Logger head and Leatherback turtle on the Bundala beach, While Hawks bill turtle did not come last three years to Bundala beach except only single nesting site was recorded year 2004. There are one hundred nineteen (119) uncertain arrivals of individuals were recorded. Uncertain means no direct evidence for which species made the nests, but nesting sites could be found on the beach. However compare to year 2002 and 2003 number of uncertain records comparatively reduce year 2004 and 2005. Two conservation methods i:e in-situ conservation and ex-situ conservation method were applied for hatch of turtle eggs. Irrespective of the method used the success rate of hatchling of Green turtle and Olive readly were more than 85%. Lowest rate was recorded for Olive readly as 15.2%. Highest success rate were recorded in 2005 (83.4%) (Table 2) Wild boar and feral dogs are the main predators of turtle eggs on Bundala beach. To protect from predators, used concrete cylinders (dia - 30", height - 1 1/2', thickness 2") and labeled it to easy to identification, and determine the date of hatchlings come out. INTRODUCTION Bundala National Park is located in the Hambantota District (6008' - 6014N, 81 08' - 81018'E) covering an area 3698 ha. The park falls within the Southeastern Arid Zone of Sri Lanka, with a general climate that can be classified as hot and dry. The average annual rain fall for the area is about 1,074 mm, with the highest monthly rainfall occurring in November. The mean annual temperature is about 27.1C0. Topographically, the park is generally flat with sand dunes bordering the coastline. Three shallow brackish water lagoons located within the park namely Malala (650 ha), Embilikala (430 ha) and Bundala (520 ha), from a complex Wetland system that harbors a rich bird life, including several species of migratory waterfowl. Recognizing the importance of Bundala as an important habitat for wildlife, it was declared as sanctuary under the fauna and flora protection ordinance in 1969, and later upgraded to a National Park in 1992 (Wet land site report and conservation management plan, Bundala National Park-1993). Beside the bird life, the lagoons support a large number of fish and Shrimps. The reptile fauna of Bundala National Park includes many IUCN Red list species such as 0 212 estuarine crocodile; endemic frog (Bufo atukoralei) also has been recorded from the Park (Bambaradenya et al., 2001). Every year during the turtles breeding period Bundala cost line host nesting endangered Olive Ridley (Lepidochelys olivaceae), Green turtle (Chelonia mydas), Loggerhead turtle (Caretta), Hawksbill turtle (Eretmochelys imbricata) and Leatherback turtle (Dermochelys coriaceae) All five marine turtle species that nests in Sri Lanka are listed on the convention on International Trade in Endangered species (CITES) Appendix 1, of which Sri Lanka is a signatory. Sri Lanka is also a signatory to a Mou (Memorandum of Understanding) on the conservation and Management of Marine Turtles and their habitats of the Indian Ocean and South - East Asia (IOSEA) The Bundala National Park (BNP) initiated a monitoring program of nesting turtles at Bundala beach in year 2002. At present BNP get the service of volunteers from local villages to patrol the beach, to collect information about nesting and protect turtle nests. Concrete cylinders are used to protect nest by wild boars and feral dogs. The project has two main objectives. The main objective is to record the number of each turtle species nesting on the Bundala beach. The second objective is to establish suitable methods of insitu and exsitu conservation methods for turtle nests at Bundala beach. MATERIALS AND METHODS Study Area Sri Lanka has four climatic zones which include the wet zone, dry zone, intermediate zone and Arid zone. The study Area is located in the arid zone in Hambantota district in Southern Province. The study Area is located at 6008' - 6014'N, 81008' - 81018'E. The length of the Bundala beach is approximately 16 Km length. The beach contains number of sand dunes. Field survey Although the nesting beach at Bundala is 19 Km in length currently only 4 Km of this is included in the conservation area. The major threat to nesting females, the eggs in the nest and the hatchlings in this area includes natural predators such as wild boar, and feral dogs, which damage nests and prey on hatchlings as they emerge. This is combated by a team of 15 volunteers who come from the vicinity of the park to beach patrol 24 hours per day. The 4 Km stretch of beach at Bundala is monitored 24 hours a day, seven days a week so all members of staff work on and shift rotation system. There are different sets of overlapping shifts from 2.00 a.m. - 6.00 a.m., (5 staff), from 10 p.m. - 2 a.m. (5 staff), from 2 p.m. - 10 p.m. (3 staff) and 6 a.m. - 2 p.m. (2 staff) Nesting females that are observed at night are allowed to dig their nest and then recorded the number of eggs in the nest, their location and their biometric data (length of carapace, 213 width, number of scales etc.). Once the nesting process is complete, the nests are covered by large concrete cylinders (height - 1 1/2 feet, diameter 30 inch, thickness 2 inch) specifically to protect the nest from the wild boar and feral dogs that are so plentiful in the park, throughout the incubation period (Figure 1). In addition hatchling number of emerging from the nest of each species is recorded. Identification was done using a colored atlas of same Figure 1. Turtle nest protecting cylinder RESULTS Five species of turtle which are logger head, Hawk bill, Leatherback, Green turtle and Oilve redly nests Bundala beach. But Loggerhead and Leatherback turtle did not nest during last two years. In addition Hawks bill did not nest in year 2003 and year 2005. Olive redly and Green turtles nest at Bundala beach regularly during last four years (Table 1). 214 Table 1. Nesting records of five species of turtles at Bundala beach from 2002 – 2005. Year Name of species Green Turtle Olive Redly Loggerhead Hawkbill Leatherback Uncertain 2002 2003 2004 2005 Grand Total 26 92 35 18 10 44 06 152 09 05 33 06 10 01 18 03 55 24 41 309 44 19 15 119 160 No of Sp 140 120 Green Turtle 100 Olive Redly Loggerhead 80 Haw kbill 60 Leatherback 40 Uncertain 20 0 2002 2003 2004 2005 Year Figure 2. Nesting records of five species of turtles at Bundala beach from 2002 – 2005. Figure 2. Green Turtle Figure 3. Hawkbill Figure 5. Loggerhead Figure 4. Olive Redly Figure 6. Leatherback 215 Olive readly arrived three hundred and nine times (309) for last four years to Bundala beach which was recorded highest nesting. Lowest attendance was recorded for Leatherback for last four years (15 times), while Hawks bill, Green turtle and logger head were recorded 19 times, 41 times and 44 time attendance respectively (Table 1). Second highest attendance (119 times) was recorded uncertain species which were not identified particular species but could be fond nests (Table 1). Success rate of hatchling is very high irrespective of two conservation methods. Highest success rate of hatchlings of Green turtle could be fond year 2002 (96.01%) while year 2005 it was 88.31%. Year 2003 and 2004 the rate was 85.7% and 85% respectively. Highest success rate of hatchlings of Olive Readly was recorded during the year 2005 (85.09%) (Table 2). Success rate of hatchling of uncertain species, year 2005 was recorded (83.4%). Lowest rate was recorded year 2003 (67.42%) (Table 2). Table 2. Success rate of hatchlings of in-situ and ex-situ conservation method for five species of turles at Bundala beach. Year Name of species Green Turtle Olive Redly Loggerhead Hawkbill Leatherback Uncertain 2002 2003 2004 2005 96.01% 15.2% 92.07% 92.17% 83.33% 81.06% 85.71% 80.59% 71.27% 72.34% 67.92% 85% 80.07% 88.31% 85.09% 83.41% 120.00% Sp percentage 100.00% Green Turtle 80.00% Olive Redly Loggerhead 60.00% Haw kbill Leatherback 40.00% Uncertain 20.00% 0.00% 2002 2003 2004 2005 Year Figure2. Success rate of hatchlings of in-situ & ex-situ conservation method for five species of turtles at Bundala beach 216 DISCUSSION Green turtle and Olive Redly turtle came to Bundala beach for nesting throughout last four years (2002-2005). Therefore nesting of two species can be seen throughout the year in Bundala beach. Leatherback and Loggerhead did not come during last two years. Hawks bill turtle did not come during last three years except one attendance of year 2004. According to IUCN Leatherback, Logger head and Hawk bill species are critically endangered species in the world. In addition it may be the reason, which these species may nest beyond the conservation area. Turtle conservation project confined about 4 Km length of the beach. Success rate of hatchling of all five species are very high irrespective of in-situ and ex-situ conservation method. Species which are not seen at the time of nesting were recorded as uncertain species. Success rate of uncertain species is high in last four years (Table 2). Number of volunteers need to minimize the record of uncertain species. Furthermore volunteers should be given appropriate technology and knowledge to identify species using their tracks. CONCLUSION Beach of Bundala National Park provides good nesting habitats for five species of turtles in the world. Green turtle and Olive readly came to nest throughout the last four years to Bundala beach. Therefore Bundala National Park harbors rich reptile diversity. No special reason could be found for the absence of critically endangered three species of turtles Leatherback, Hawk bill and Logger head turtle, for the last three years. In situ and ex-situ conservation methods are very successful in the Bundala beach. The area of the project must be expanded for 16 Km and it will help to get more number of attendance and nesting sites. In addition the number of volunteers should be increased, so that more data could be gathered during peak nesting season. REFERENCES Archie, C. (1980). Life Nature Library. The Reptiles. Pp 9-185. Bambaradeniya C.N.B. (2001). Guide to Bundala, A Guide to the Biodiversity of Bundala of Bundala National Park - a Ramsar wetland in Sri Lanka. 54p. Caughley, G, Sinclair, A.R.E. (1994). Wildlife ecology and Management. Pp 1-300. Central Environmental Authority (1993). Wetland site Report and conservation management plan, Bundala National Park.103p. Deraniyagala, P.E.P. (1955). A colored Atlas of some vertebrates from Ceylon. Serpentoid Reptilia. National Museum of Sri Lanka. 3:14-119. Deraniyagala, P.E.P. (1993). A colored Atlas of some vertebrates from Ceylon. Tetrapod Reptilic National Museum of Sri Lanka vol 03. 217 AN ASSESSMENT OF WATER RESOURCES DEPENDENCY OF SURROUNDING COMMUNITY, IMPACTS OF AQUATIC WEEDS AND EFFECTIVE WEED CONTROL METHODS OF THE WATER BODIES IN ANAWILUNDAWA WILDLIFE SANCTUARY AND RAMSAR WETLAND IN SRI LANKA S. C. WILSON* Department of Wildlife Conservation, Sri Lanka ABSTRACT The Anawilundawa Wildlife Sanctuary (AWS) was established in June 1997 by the Department of Wildlife Conservation (DWC), Sri Lanka under the provisions of the FFPO. (Government Gazette number 97915 of 11 June 1997) Considering its biodiversity and Wetland values, the sanctuary was recognized as a wetland of International Importance by the Ramsar convention in 2002. The AWS consists of a group of seven shallow cascading reservoirs including the around terrestrial landmass. The reservoirs are the heart of the sanctuary which depends on the balance of the wetland ecosystem. The people in and around, depend on the water resource of the sanctuary for their livelihoods. Therefore the reservoirs play a significant role for the balance of the wetland ecosystem as well as the dependencies of the surrounding community. But the reservoirs are covered with aquatic weeds. The objective of the present study is to access the impacts of the weeds and to propose management practices to control the aquatic weeds. The direct observations, interviewing the community, and several pilot practices had been conducted during the study. The results reviles that, there is a seasonal pattern of water resource dependency in the study area. The extraction of non timber forest products takes place throughout the year. The water level of the sanctuary indicates a positive relationship with the water resources dependencies. There are about ten aquatic weeds in the AWS. The most impacts cause from Water hyacinth (Eichhornia crassipes), Salvinia (Salvinia molesta), and Water lettuce (Pistia stratootes). The large leaf area of weeds leads to increased water loss through transpiration. It interfere with fish life by reducing oxygen level in water, producing toxic substances such as Hydrogen Sulphide and reducing phytoplankton population, prime food sources for fish. Weeds reduce the water flow and increase the siltation. When compared with the previous records, in the full capacity condition, the water capacity of the reservoirs are eight inches lower than that of 1995. Water hyacinth also downgrades recreational water facilities by reducing the water surface, creating offensive odors and tainting of water. Mechanical control of weeds is one method, and removal of Salvinia by hand or machine is a practical control method. Biological control of Salvinia and Water hyacinth is a better method in this sanctuary as this is a protected area. In 1990, Agriculture Department of Sri Lanka has introduced Salvinia weevil to these water bodies. The DWC, the management authority of the sanctuary, organized a campaign for the removal of aquatic weeds May, 2005 during the peak drought. Surrounding community involved with this activity through their CBO. Prevention is the best form of weed control. The community can also help to prevent the spread of aquatic weeds by reporting the Department of Wildlife Conservation. Rehabilitation of water outlets of the reservoirs should be done immediately. Desilting of the water bodies is another important practice to increase the water capacity. Community participation for weed control, desiltation of the reservoirs, to control the illegal activities affecting the balance of the ecosystem of this study area could be suggested as some immediate site specific actions with appropriate strategies. 218 INTRODUCTION Sri Lanka has a golden history about Protected Area (PA) Management. The first PA, Mihintale Sanctuary was declared by king Devanampiyatissa the Great in about 200 B.C. After the Mihintale Sanctuary, Sri Lanka has 82 PAs including 3 Strict Natural Reserves, 17 National Parks, 4 Nature Reserves and 58 Sanctuaries. All these PAs are under the jurisdiction of the Department of Wildlife Conservation (DWC) which has the authority of Fauna & Flora Protection Ordinance (FFPO). The Anawilundawa Wildlife Sanctuary (AWS) was established in June 1997 by the DWC under the provisions of the FFPO. (Government Gazette number 97915 of 11 June 1997) Considering its biodiversity & Wetland values, the sanctuary was recognized as a wetland of International Importance by the Ramsar convention in 2002. The AWS consists of a group of seven shallow cascading reservoirs including the surrounding terrestrial landmass. The reservoirs are the heart of the sanctuary which depends on the balance of the wetland ecosystem. The sanctuary includes the state lands as well as private lands including home gardens and paddy fields. The people in and around, depend on the water resource of the sanctuary for their livelihoods. Therefore the reservoirs play a significant role for the balance of the wetland ecosystem as well as the dependencies of the surrounding community. All seven reservoirs of the AWS are shallow with maximum depth not exceeding 4.0 meters during full storage. (Wetland site report & conservation management plan, Anaiwilundawa tanks, 1994) But the water is covered with varied aquatic weeds. Several researches were carried out by various researchers. There are records on previous applications of biological control of Water hyacinth and Salvinia by the Department of Agriculture. But there are no previous studies on aquatic weeds and water dependencies in AWS. Significance of the study The AWS harbours a rich native terrestrial and aquatic faunal and floral biodiversity that is further enriched by the seasonal migration of waterfowls. These factors led to the declaration of Anaiwilundawa as a birds sanctuary and recognition of it as a Wetland of International Importance by the Ramsar Convention. The group of seven reservoirs, which cover the area of about 14ha, is the heart of the wetland ecosystem. 219 Research problem The water capacity of reservoirs is the most important factor for the sustainability of the sanctuary. The people living around the sanctuary use the water for different purposes. But the reservoirs are covered with aquatic weeds. It is a significant threat to the natural balance of the wetland ecosystem as well as the community which depends on the water of reservoirs. The management practices need to be applied to control the weeds. Purpose of the study AWS, the wetland ecosystem contains seven reservoirs. The water is the main factor for the balance of the ecosystem. The people on diverse needs, depends on water of these reservoirs. The existing quality of water is deteriorating. In this study the researcher is going to find the impacts of aquatic weeds of the reservoirs in AWS. The findings are important to the managers of the sanctuary who are responsible for the conservation and sustainable utilization of natural resources of this wetland ecosystem. They can be incorporated into the management plan. The proposed management practices can be applied in other wetlands as well under similar conditions. The weed control methods would be important to the DWC who are the management authority of the sanctuary. Scope of the study In this study the writer wishes to focus the attention on the aquatic weeds but not on terrestrial weeds. The list of the aquatic weeds in tank system is being prepared. The direct and indirect dependencies on tanks are being studied. The writer wishes to discuss the weed management methods. These methods can be applied by the DWC who are the management authority of the sanctuary. The writer does not plan to calculate the total area density of each aquatic weed in tanks. Anawilundawa Wildlife Sanctuary, the wetland ecosystem contains seven reservoirs. The water is the main factor for the balance of the ecosystem. The people have various dependencies on the water of these tanks. The existing condition of the water is becoming unsuitable for fauna and flora existence. The main objective of the study is to propose management practices to control aquatic weeds. In this study the researcher hopes, to discuss the impacts of aquatic weeds on wetland ecosystem of AWS, and to find out the dependencies of surrounding community on water of the reservoirs. 220 METHODOLOGY Study Area Location The Anawilundawa Wildlife Sanctuary, covering an area of 1400ha, is situated between the coast and the Negombo – Puttalam railway/road (70 42’ N, 790 49’ E) in Puttalam District of North-Western Province in Sri Lanka (CEA, 1994). The closest towns are Chilaw (10km) to the South and Puttalam (35km) to the North (Figure 1) 221area Figure 1. The study Declaration The AWS was established in June 1997 by the Department of Wildlife conservation under the provisions of Fauna and Flora Protection Ordinance. In August 2001, the sanctuary was recognized as a Wetland of Internationally Important by the Convention on Wetlands of International Importance especially as Waterfowl Habitat (Ramsar Convention), following a request made by the DWC. Figure 1 shows the study area Methodology The survey had been carried out in August and October 2005. In this survey direct observations of dependencies on reservoirs had been recorded. During the day time from 6.00am to 7.30 pm the activities of the people in the sanctuary were observed. A selected set of people were interviewed to find out the dependencies on water, to find their means of income generation activities, names of weeds in water, to identify medicinal herbs. The weeds, some shrubs, trees were identified with the guidance of herbarium and books on taxonomy. The faunal species were identified with the guidance of taxonomy books. The traditional weed control methods were identified by interviewing the village community. RESULTS Aquatic plants in AWS The list of aquatic plants found in AWS is shown in table 3. Table 1. Aquatic plants in AWS Family Acanthaceae Alismataceae Amaranthaceae Aponogetonaceae Araceae Araceae Asteraceae Ceratophyllaceae Convolvulaceae Cyperaceae Fabaceae Hydrocharitaceae Lemnaceae Lentibulariaceae Limnocharitaceae Species Hygrophila schulli Limnophton obtusifolium Alternanthera sessilis Aponogeton natans Lasia spinosa Pistia stratiotes Eclipta proatrata Ceratophyllum demersum Ipomoea aquatica Schoenoplectus Nepyunia oleracea Ottelia alismoides Lemna sp Utricularia aurea Limnocharis flava 222 Local Name Niramulliya Mukunu wenna Kekatiya Kohila Diya paradel Kikirindi Kankung Marsiliaceae Menyanthaceae Nelumbonaceae Nymphaeaceae Onagraceae Polygonaceae Pntederiaceae Pteridaceae Salviniaceae Scrophulariaceae Typhaceae Marsilia quadrifolia Nymphoides hydrophylla Nelumbo nucifera Nymphaea pubescens Ludwigia adscendens Polygonumtomentosum Pllygonom glabrum Eichhornia crassipes Monochoria vaginalis Acrostichum aureum Salvinia molesta Limnophila aquatica Typha angustifolia Kumudu Nelum Manel Diyanilla Sudu kimbul wenna Japan jabara Karam koku Salvinia Hambu pan The various dependencies of water resources in the sanctuary People surrounding the sanctuary have various types of dependencies on the sanctuary especially. Most of them are on reservoirs, or associated with the reservoirs. There are two types of dependencies on water resources i.e. direct dependencies and indirect dependencies. Direct dependencies 1. water for bathing and washing 2. drinking water for cattle and other domestic animals 3. wallowing grounds for buffaloes in tank water 4. rowing boats and swimming– entertainment 5. water for paddy cultivation 6. water for vegetable cultivation 7. fishing Indirect dependencies 1. fish and terrapins 2. edible green leaves - Kan kun, Gotukola, Mukunuwenna, Thampala, Neeramulliya, Anguna, Thora, Girapala, Pitasudu, 3. fruits – Tal, Mango, Wood apple, Tamarind, Cashew, Kon, 4. vegetables- Lotus stems, water lily seeds, Drum sticks, 5. reeds for handicrafts 6. fire wood 7. grassing lands 8. edible mushrooms 9. Erriconuts 223 The seasonal pattern of major dependencies of AWS Figure 2 shows various types of major dependencies that vary with the seasons and availability in the sanctuary. Fire wood extraction and extraction of fruits and vegetables are found throughout the year. Water for Paddy cultivation recorded in two periods April to July in Yala season and November to March in Maha season. During the dry season from March to October and September to November while water level becomes low villagers release buffaloes for grazing and wallowing. According to these facts the highest water resource dependency is recorded from February to May. DEPENDENCY Grass land Fire wood Fruits & vegetables Wallowing ground Entertainments Fishing Paddy cultivation Water for bathing Jan Feb Mar Apr May Jun Jul Aug MONTH Figure 2. Various types of major dependencies 224 Sep Oct Nov Dec Over exploitation of resources Poaching During the weekends and holidays large numbers of bird hunters and teams of poachers visit the sanctuary. The poaching of soft-shell and hard-shell terrapins supplies the demand created by hotels in the area. Extraction of clay from reservoir beds in dry period During the dry season clay mining takes place on the reservoir beds for brick making industry. Pollution Biocides are not reported to be used in huge quantities within the sanctuary, but there can be the risk of pollution through the runoff from long stretches of paddy fields and coconut plantations within the catchments area. Small quantities of plastic and polythene wastes can be seen on the road sides leading to the sanctuary. These indigestible materials accumulate in the reservoir and water canals. At present this is not a severe problem if the dumping of polyethylene is not controlled it is going to be a severe environmental problem in future. Natural threats The rainfall records in Puttalam show a paternal reduction during the recent past. The tank system remained without water for long periocls during the last few years. DISCUSSION The present study shows that about 10 weed species have extensively spread in the reservoir system of the AWS. Invasive alien weed species have caused comparatively higher impact on water than the native weed species. Almost all the villagers in and around the sanctuary depend on water resources of the sanctuary; furthermore the water is the main factor that contributes to the balance of the natural ecosystem. But large scale commercial activities making use of water recourses do not take place in the study area. Most dependencies are for private consumptions such as income generation activities of the local community such as paddy cultivation, fishing, fire wood collection, vegetable cultivation etc. Irrigation Department and the Department of Wildlife Conservation are the responsible Government Institutions for maintaining the water capacity of the sanctuary. DWC display some sign boards and information boards in the Sanctuary, but without clearly demarcated boundaries. As far back as 2004 DWC recruited and stationed number of employees for protection of the Sanctuary. The Department of Wildlife Conservation initiated a weed control programme in 2005 and removed aquatic weeds manually in Anawilundawa and Suruwila tanks. However chemical weed control methods are not recommended to be carried out in the Sanctuary. Community Based Organizations have been formed with the guidance of 225 DWC and Divisional Secretary of the area. The DWC which is the management authority of the sanctuary organized a campaign for removal of aquatic weeds in May 2005, during the peak dry period. Adjacent communities were involved in this activity through their CBO. CONCLUSION AND RECOMMENDATION Conclusion The natural balances of the ecosystem, the various dependencies of the community and wetland values are related to the water availability of this sanctuary. Issue of insufficient water supply During the draught season the water level of the reservoirs decreases dramatically. It is the core problem related to the water level. Besides water flows out to the reservoirs since the sluice gates are not maintained properly and some of them do not operate at all. Aquatic weeds The water capacity of the reservoirs is very low because of the above situation. The nutrition level and the eutrophication are high as a result the aquatic weeds grow rapidly. There are about ten aquatic weeds in the study area. The most of the impacts are caused by Water hyacinth (Eichhornia crassipes), Salvinia (Salvinia molesta), and Water lettuce (Pistia stratootes). Water hyacinth also downgrades recreational facility by reducing the water surface, creating offensive odors and tainting water. Large leaf surface of weeds increases water loss through transpiration. It also interferes with fish life by reducing water oxygen levels, producing toxic substances such as hydrogen sulfide and reducing phytoplankton population, which is prime food source for fish. They change the ecological balance and habitat of the area. Besides they reduce the water flow and thereby increase siltation. When compared with the previous records, the water capacity of the reservoirs appear to be decreasing and it was eight inches lower than that of 1995 in the full capacity condition. The major dependency levels vary with the season, due to availability of water. The highest dependency is recorded from March to July. 226 Recommendation Mechanical control of weeds is one method, and removal of Salvinia manually or by machine is a practical control method. Biological control of Salvinia and Water hyacinth is the better method in this sanctuary as this is a protected area. In 1990, Department of Agriculture, Sri Lanka introduced Salvinia Weevil to these water bodies. Community participation for weed control of this study area has proved to be effective. CBO’s have formed in each village. Through these CBO’s the community can be involved to removing the weeds, as well as to control illegal activities of the people especially from the outsiders. Repairing of sluice gates and desiltation the reservoirs should be implemented immediately by the responsible agencies to maintain the water level of the sanctuary. Sand mining, felling of timber and poaching should be controlled through Law Enforcement and by regular patrolling by the Department of Wildlife Conservation. REFERENCES Anon, Weed note No1/99, Agriculture western Australia. Anon,(2004) NRM facts, Salvinia and Water lettuce, Department of Natural Resources and Mines, The state of Queen land. Anon, (2003) Weed management guide, Natural Heritage Trust, Australi Anon,(2003),Wetland conservation in Sri Lanka,IUCN. Anon,(1994)Wetland site report and conservation management plan Annaiwilundawa tanks,Wetland conservation project, Sri Lanka,CEA. Griffiths, M.W.,Julian, M.H.,Stanley,J.N.(2001)Biologiacal control of water hyacinth Purcell, M.F., Wright,A.D.,(1981)The biology of Australian weeds,226-240. Website, www.agric.wa.gov.au 227 228 View publication stats