Congratulations on the completion of your project that was supported by The Rufford Small Grants Foundation.

Transcription

1 The Rufford Small Grants Foundation Final Report Congratulations on the completion of your project that was supported by The Rufford Small Grants Foundation. We ask all grant recipients to complete a Final Report Form that helps us to gauge the success of our grant giving. We understand that projects often do not follow the predicted course but knowledge of your experiences is valuable to us and others who may be undertaking similar work. Please be as honest as you can in answering the questions remember that negative experiences are just as valuable as positive ones if they help others to learn from them. Please complete the form in English and be as clear and concise as you can. We will ask for further information if required. If you have any other materials produced by the project, particularly a few relevant photographs, please send these to us separately. Please submit your final report to jane@rufford.org. Thank you for your help. Josh Cole, Grants Director Grant Recipient Details Your name VY TRAN NGUYEN Project title RSG reference Reporting period The ecology and distribution of two globally range-restricted Galliform species in South Viet Nam 12 months Amount of grant 5956 Your address vychim@gmail.com Date of this report 30 August 2015

2 Fully achieved Partially achieved Not achieved 1. Please indicate the level of achievement of the project s original objectives and include any relevant comments on factors affecting this. Objective Comments 1. Density and abundance of the species in Cat Tien National Park 2. Define ecological factors affecting habitat use of the species 3. Potential distribution of the Germain s peacock pheasant in cat Tien NP 4. Potential distribution map of the Germain s peacock pheasant in its range and its historical distribution map 5. Breeding behaviour of the species We fully conducted a survey in Cat Tien National Park using line transect method and distance sampling to estimate density of Germain s peacock pheasant (GPP) (Polyplectron germaini (Elliot, 1866)) and orange-necked partridge (ONP) (Arborophila davidi (Delacour, 1927)) in Cat Tien National Park (NP) From data of all line transects in different habitat types, we fully defined ecological variables associated with habitat use of the Germain s peacock pheasant. However, this work did not achieve this analysis for the orange-necked Partridge. This probably was attributed to small sample size as this species is quit rare. We plan to continue this part in the next programme. Based on results of habitat association analysis, a potential current distribution map for the GPP in Cat Tien NP. And the results of habitat association with analysis for GPP also were used with data of current forest cover in Southern Vietnam from LANDSAT8 (2015) to predict a potential distribution map of the GPP in Southern Vietnam. A historical range for the species was created based on historical detections from various previous observations. As the line transects were set in different areas quite far away from each others and difficulty of accessibility to these areas, taking us a lot of time for moving between areas. Moreover, the breeding season occurred in short period from February to May. Thus, we must focus on data collection for density estimation in all forest types in breeding season. This did not allow us to gather data of breeding behaviour. This activity will be carried out in the next study.

3 2. Please explain any unforeseen difficulties that arose during the project and how these were tackled (if relevant). Studying ecology and breeding behaviour of pheasants, especially rare and narrow geographical range species such as Germain s peacock pheasant (GPP) (Polyplectron germaini (Elliot, 1866)) and orange-necked partridge (ONP) (Arborophila davidi (Delacour, 1927)) are essentially important and interesting as they are in high risk due to high pressures from anthropogenic activities, especially habitat loss and hunting. However, their rare status is an obstacle on gathering their ecological information. In this study, we expected to evaluate their current status, distribution, and obtain their breeding behaviour in Cat Tien NP. However, due to some unanticipated reasons, we could not reach one of these objectives: breeding behaviour but instead we accomplished another one which was not in the proposal: evaluate current distribution of the GPP in its potential range in Southern Vietnam and its historical range. Breeding season of both GPP and ONP is considered dry season in Southern Vietnam, especially in South Annamite mountain, normally from December to May. In our plan for this study, we expected to have enough human resources, especially staff of the Cat Tien National Park as local partnership. However, the main survey time was at the same time with fruit season of a species named Sterculia lychnophora (family Sterculiaceae) which normally give yield every 4 years. As price of this kind fruit was very high in the market, about 20 25/ kg as it is not only considered a traditional medicine but also mainly exported to China. Cat Tien national park is one of a rare remain forest harbouring high community of Sterculia lychnophora attracting people to come to collect the fruits. Due to shortage of the rangers, the fruit collectors not only accessed many areas of the park but also cut the trees sometimes, partially affecting habitat of wildlife in the park. Because of this situation, the park did their best by deploying all the park s rangers to patrol the park much more intensively than as usual. This made us difficult to get some their staff as we planed before to deploy at the same time all these activities. However, we schedule to develop a long term research to gather this information in the next 2 years by using different ecological methods and modern equipment such as radio telemetry, camera trap etc. In addition, after working with some specialists in pheasant and we got some useful advices for our study. One of these comments was to evaluate status and current distribution of the species in their range. This is essentially important as it would contribute a concrete data for conservation of these species as most of the remaining forests in their distribution range are out of protected areas and have been declining at very high rate. In order to reach this goal, we need to have other surveys in some protected areas besides Cat Tien National park to get more data of the species and status of forest. This field data will help us to update forest cover created from LANDSAT 8 (2015) for species distribution analysis later. Thanks to helps from our colleagues, specialists in GIS, we overcame our difficulties of GIS working to define current status of forest in Southern Vietnam for our analysis. 3. Briefly describe the three most important outcomes of your project Estimation density of the species First of all, we stratified forest covers of Cat Tien National park to four main forest types supporting population of the species: evergreen, semi-evergreen, bamboo, and mosaic forests. The mosaic forest consists patches of previous forest types. Then Line transect method was used to collect data for density estimation. We set five 2-km transects in each forest types (Figure 1) (seven line transects in evergreen forest) and the data was collected during their breeding season in dry season from January to May as we expected to get highest number of detections of the species. The data

4 was collected twice daily, one in the morning (06:00 08:00) and the other in the late afternoon (16:00 18:00). The number of detections in each forest types was present in Table 1. The data was collected by walking along the transect at 1km/hr to collect all birds by sighting and calling. Before conducting the surveys, all surveyors had several surveys together in order to minimise the difference between surveyors when we detect birds by callings. To the GPP, most of detections were the male s callings so the density result was the number of male s calling/km 2. However, the ONP normally both male and female call at the same time so we can detect both individuals if they were together. Thus, the density of ONP will be individuals/km 2. To avoid double count, all detections based on bearing, distance, and time of detection of the calls were mapped using ArcGIS software version 10.3 (ESRI) in order to remove potential double detections. The density estimation was based on distance sampling (Buckland et al. 2001) and DISTANCE 6.0 (Thomas et al. 2010) was used to estimate detection probability and density of GPP and ONP in each habitat types and for whole study site. In this analysis, all key functions such uniform, half-normal, and hazard with cosine adjustments were used to run the analysis and the lowest Akaike s information criterion (AIC) (Akaike 1998) was used to select the best models from the analysis. We stratified detections by forest types in order to estimate density of GPP and ONP in each forest type. The density of the species of each model was compared using coefficient of variance (CV). The results of density estimation was present in Table 1. Table 1. Density estimates for Germain s Peacock Pheasant (GPP) and Orange-necked Partridge (ONP) in the Cat Tien National Park between Species Forest type Number Density Coefficient 95% confidence of (calling birds of variation detections km -2 intervals ) (%) GPP Evergreen Mixed-deciduous Mosaic Bamboo Pooled detections ONP Evergreen Mixed-deciduous Mosaic Bamboo Pooled detections The density of GPP and ONP were different in each habitat type. To the GPP, its density was highest in mosaic (8.8 male s callings/km 2 ), and following by evergreen forest (4.0 male s callings/km 2 ), mixed-deciduous forest (3.9 male s callings/km 2 ), and quite low in bamboo forest (0.7 male s callings/km 2 ). Similarly, the density of ONP was also high in evergreen forest (3.7 male s individuals/km 2 ), mosaic forest (2.5 individuals/km 2 ), in bamboo forest (0.6 individuals/km 2 ), and there was no any detection of ONP in the mixed-deciduous forest. This forest type is dominant in East region of south part of Cat Tien on a relative flat and low elevation from m which is within elevation range of the species (Vy et al. 2014, Birdlife International 2015). In the same region, previous observations also showed that the ONP was not detected but it was recorded on small hills,

5 adjacent to this region, named Dak Lua hills in the north, and Nui Tuong hill in the south. These hills are about 21 km apart (Robson et al. 1993, Atkins and Tentij 1998). However, it was interesting that in this region, scaly-breasted partridge (Arborophila chloropus) was commonly hearing but never on the hills which harbour the ONP, indicating different habitat preference between the orange-necked partridge and scaly-breasted partridge (Robson et al. 1993). This raised a question that if the orangenecked partridge is sympatric with the scaly-breasted partridge. We need to implement further researches about habitat preference and their micro habitat selection, giving us more details of factors affecting the orange-necked partridge s distribution. And this also will provide more scientific data about factors, especially biotic ones affecting distribution of animals besides physical ones. In this study, we also deployed several surveys in agriculture, mainly in cashew nut in the buffer zone of Cat Tien but this kind of habitat did not harbour both species. Both GPP and ONP were detected only in natural forest and in closed-canopy forests such as evergreen, semi-evergreen forests. Some research indicated that closed-canopy cover appears an essential to ground nest species, especially pheasants such as Hume s pheasant (Syrmaticus humiea) in Thailand (Iamsiri and Gale 2008), whiteeared pheasant (Crossoptilon crossoptilon) in Sichuan, China (Wang et al. 2006), or other research in China about Partridge like hill partridge (Arborophila torqueola) (Liao et al. 2007), Sichuan partridge (Arborophila rufipectus) (Liao et al.2008). This can be interpreted as a strategy to make the species avoid predators, especially in breeding season (Martin 1993). (A) Transect lines in each forest type in the Cat Tien National Park; (B) Predict distribution ranges of GPP in the Cat Tien NP; and (C) Predict current and historical ranges of GPP in Vietnam. Adult male - Photo: Nguyen Tran Vy

6 Adult female of Germain s Peacock Pheasant in the Cat Tien National Park - Photo: Nguyen Tran Vy An Orange-necked Partridge in the Cat Tien national Park - Photo: Nguyen Tran Vy Ground animal trap - Photo: Nguyen Tran Vy

7 A national park ranger removed snare traps set up by illegal hunters. Fruits of Sterculia lychnophora collected by local people from forest. Vegetation in Nghia Trung State Forest Enterprise (left) area adjacent to border of the Cat Tien NP and (right) Almost 80-90% land has been converted to rubber plantation.

8 (Left) 15 years ago (2000), this village was much smaller than now (2015), and this area used to support a significant population of both Germain s peacock Pheasant and Orange-necked Partridge. (Right) The forest, between the village and the Cat Tien s border has been cleared recently. Local people accessed the forest for fruit of Sterculia lychnophora Define ecological variables affecting distribution of the GPP In this study, we examine association between ecological variables and abundance of the GPP in study sites: elevation, slope, distance to water sources, and forest types. These variables were derived using ArcGIS 10.3 from the following source: 1) elevation and (2) slope were extracted from the ASTER GDEM (Global Digital Elevation Model) downloaded from the Earth Remote Sensing Data Analysis Center ( (3) river and stream system map sourced from Ministry of Science and Technology, (4) forest types were classified from LANDSAT 8 (2015) using supervised classification in ArcGIS All data was re-projected to the WGS84 datum before analysis. The association between habitats and the GPP was assessed using poisson regression. Independent variables were elevation, slope, forest types, and distance to water sources and the number of calling birds from each survey was treated as the response variable. There were total 132 surveys collected from 22 line transects which were surveyed consecutively 3 days. The models were performed as described below (see Regression models). The results were present in Table 2 and 3. Table 2. The best fitted regression models of habitat selection by Germain s Peacock Pheasant (GPP). K is number of parameters in the model: ΔAICcis the difference in AICc(model score) value, model with ΔAICc value 0 has most support, value between 0 and 2 have substantial support, values

9 greater than 2 have less support: wi= Akaike model weights. Forrest includes bamboo, evergreen, mixed-deciduous, and mosaic forest. Species Models and variables K AICc ΔAICc wi GPP (14 models tested including a null model) Slope + water + forest types Slope + forest types Table 3: Estimates of coefficients derived from the best models with standard errors (S.E.) and 95% confidence intervals. Species Variables Coefficient S.E. GPP 95 % CIs Lower Upper Slope < Water Bamboo (Intercept) Evergreen forest < Mixed-deciduous forest < Mosaic forest < The habitat use models showed that the GPP prefers to occur areas near water and covered by in closed forests like evergreen, semi-evergreen, and mosaic structure, and the bamboo forest was not significant association with the abundance of the GPP Potential distribution of the GPP in its range Potential distribution of the GPP in Cat Tien National Park and its range was analysed based on the results of habitat use analysis and current remain forest in its potential range. The remaining forest cover in Southern Vietnam was created from LANSAT 8 (2015) (Figure 10). Current remaining forest cover in Southern Vietnam from LNADSAT8 (2015) Current and historical records of the GPP were mostly from N to N. However, based on the results of habitat use analysis that the GPP appears to live in lowland and flat areas near water, the suitable habitat for GPP may extend to N. Beyond this range in the north the Annamite run from West to East, shaping a high elevation and steep slope areas that are not suitable for the GPP. Thus, the historical range of the species may from N to N and constrained by steep terrain of Annamite mountain in the north of N, mangrove forest and wetland in Mekong Delta in the south, Dry dipterocarp deciduous forest in the West (from P

10 central highland of Vietnam to Mekong River on Cambodia), and the coastal areas in the East. Total remaining natural forest in Southern Vietnam in 2015 from LANDSAT 8 (2015) was about 32,706 km 2 in which only 7,510 km 2 (23%) are in protected areas and 25,196 km 2 (77%) are out side protected areas. Most of natural forests loss were in lowland due to agricultural expansion (Sterling et al. 2006, Wege et al. 1999). Forest cover in 1994 was 98,630 km 2. Thus, total natural forest loss over 70 years is about 65,924 km 2 (67%). Based on results of habitat association analysis, a potential distribution map of GPP in southern Vietnam was created and the result is present in Figure 1C. Total areas of potential map is about 27,600 km 2 and only 5,840 km 2 is in protected areas and 21,760 km 2 may be in highly disturbed areas. I fact, from the current potential map we can see that the areas that can harbour significant population of GPP are quite small and mostly in lowland in South Annamite mountain Though, commercial logging in Vietnam was stopped since 1990, but the remain forests have been affecting in different ways such as selected-bamboo and rattan logging, conversion to rubber field, and hunting, etc., making remain forests getting more and more narrow and degraded. For example, forest cover of some areas such as Nghia Trung State Forest Enterpise (SFE), Da Teh SFE, and Dak Ơ SFE which harbour population of both GPP and ONP (Vy et al. 2014, L.M. Hung et al. 2006) have been significantly shrunk at least 50 60% or almost disappear due to rubber field expansion. Nghia Trung SFE, adjacent to Cat Tien National park, is a typical example for this impact as after 8 years, almost 80 90% of its forests was clear for rubber plantation (Figure 6 & 7). In addition to habitat loss, hunting is still occurring not merely in unprotected remaining forests where protected condition in these areas is not prior, but also in National park (Figure 4). 4. Briefly describe the involvement of local communities and how they have benefitted from the project (if relevant). The research was carried in different areas of the Cat Tien National Park so each survey site we always had at least one to two rangers and at least one local people to accompany with our field work. The aims of this cooperation were to take advantage of their indigenous knowledge for our works and conservation and to share our knowledge about ecology and biodiversity to local people and rangers. To the local people, they used to live in the forest and now they cooperate with the park to patrol the park. When they join our field work, we can share our knowledge about ecology, biodiversity to them, especially to help them understand more about conservation because they know well the forest, about biodiversity there but they really do know much about the values in terms of conservation. They can tell us exactly many wildlife species but they do not know conservation terms such as rare, endangered, or endemic species, etc. This can help them to realize the sense of their works in terms of conservation. Besides knowledge, they also can get some money from our program. To the park s staffs or rangers: they learned how to design a research, how to collect data in the field by detecting the species by sighting and callings. As they live there so I hope thanks to knowledge and experiences gained via our project, later they can work on their own or to cooperate with other scientists to do research for the park in the long run. In order to manage well the natural resources in their areas, knowing GIS is also important. Most of the ranger stations of the parks have equipped with computers and GPSs, so this is a good opportunity for them to use this equipment to improve their work when they gain GIS knowledge. During our field work, we share experiences of how to

11 use GPS for field works such as how to load all data from GPS into Mapinfo software and edit it. This is useful to them as they can see and manage their areas visually, making their work more effectively and efficiently. 5. Are there any plans to continue this work? The central ideas of this work were to define status of the GPP and ONP in Cat Tien in terms of density and distribution, and predict their distributions in their range in order to prepare a long term conservation action plan for these species. Moreover, it extremely urgent that we need to inform to the government which areas should be prior for the species conservation so that remain suitable habitat for the species should be protected as soon as possible. In order to do that we need to gather essential data about home range size, micro-habitat preference, interaction with other species, and breeding behaviour like mating, nesting behaviour, etc. by using radio telemetry, camera traps, etc. This ecological information will be used to indicate that the GPP and ONP requite specific habitats in their range and this make them more sensitivity to be vulnerable as these suitable habitats are getting smaller and smaller within their narrow geographical range. In addition, occupancy modelling is also a new method that can be applied to study habitat selection by species and to estimate abundance of the species in study site. Monitoring population and status of wildlife, especially rare endemic species is fundamental for conservation. Cat Tien National park is adjacent to Vinh Cuu nature reserve but their conservation history totally different. Cat Tien was protected sine 1978 while Vinh Cuu used to be a State Forest Enterprise until It is quite interesting to develop a long term permanent plots in both Cat Tien and Vinh Cuu and deploy different methodologies in order to carry out researches in terms of ecology and dynamic of tropical forest, then we can compare results between Vinh Cuu and Cat Tien so that we can see restoration of biodiversity in areas which were disturbed. In addition, we also can compare results of different methods so that we can evaluate the efficiency of each method for specific objectives. This long term monitoring programme will be beneficial not solely to conservation strategy of the parks but also to education as students from Universities can join the program and carry out their researches for their thesis in these areas. Besides doing research, I would like to get back to Cat Tien National Park, Vinh Cuu nature reserve and other protected areas surrounding Cat Tien to carry out a training programme about how to use ArcGIS programme for conservation activities. I have discussed with Cat Tien National Park and Vinh Cuu Nature Reserve about this plan and they agreed to help me to organize this training course. 6. How do you plan to share the results of your work with others? I am going to submit at least two papers based on the results of this study. So far, I am working on the first manuscript and will submit it by the end of this month. The result of the publications will be informed to RSG as long as they are published. They are also will be sent to other relevant organizations such as Birdlife International, IUCN in order to promote conservation plans to protect these species. The results also will be shared with Cat Tien national parks and other protected areas that potentially support population of the species for their conservation programmes.

12 Difference Actual Amount Budgeted Amount 7. Timescale: Over what period was the RSG used? How does this compare to the anticipated or actual length of the project? The RSGF was used from June 2014 through August 2015, which finished 2 months later than in the proposal because unanticipated works. We think that predicting the species distribution in their range is quite important but we need to have more data and information, especially current forest types in the species potential range, southern Vietnam. During the rainy season, we need to prepare for intensive field work in the dry season. First of all, we need to work with National Park to get official permit to carry the research in the National Park as well as to get the cooperation from their staffs. Secondly, in order to deploy line transect method for density estimation, we need to prepare line transects in each forest types. Because the study was stratified by forest types and locations covered by four forest types such as bamboo, evergreen, semi-deciduous, and mosaic that contain enough size for five 2 km transects are quite far from each other. And the terrain in some areas in the north of Cat Tien was quite steep and difficult to reach there. It cost us time to travel from sites to sites to make available transects for distance sampling method. During this preparation, we carried out most of training activities such as GIS, how to detect birds by calls, how to collect data, and to make familiar with distance sampling software. And finally we implemented the research intensively in dry season from January to May. A problem arising from our data analysis was lacking current forest cover in Southern Vietnam. Therefore, we need to visit some places in Southern Vietnam to check forest status in order to accomplish our forest classification from LANDSAT 8 (2015). Though this hard work consumed lots of our times, leading to extending the time to finish the project, we learned extremely skills and knowledge about promote sensing and other techniques related to GIS, especially species distribution modelling using ArcGIS. This is an essential benefit that I got from this project beside results of research. 8. Budget: Please provide a breakdown of budgeted versus actual expenditure and the reasons for any differences. All figures should be in sterling, indicating the local exchange rate used. Item Comments Camera trap: Scoutguard SG565 cam: units Shipping and import tax Metal blocks to protect camera trap: units Batteries for Camera trap and other equipment: 120 packs As we sued some rechargeable batteries so we saved this amount of budget for other activities. Garmin GPSMAP 62S Handheld GPS Navigator: 2 units We bought this equipment with promotion so the price was less than that in proposal Range finder: 2 units Clinometer: 2 units Field work within 1 year: 12 trips 3, (18days/ trip) Transportation to other sites besides Cat Tien This budget was used to visit some areas to gather information of status of current forest in order to up date forest cover from

13 Total 5,956 5,956 0 LNADSAT8 (2015) for potential distribution analysis. 9. Looking ahead, what do you feel are the important next steps? The orange-necked partridge has quite narrow geographical range (Birdlife International 2015). However, in this study, the rareness of the species was probably attributed to low detections in each forest types, leading to a difficulty to interpolate the distribution of the species in a large scale. In this study, though we also did not detect the species in the East region in the south sector of Cat Tien NP where the terrain is quite flat at elevation range of m and covered by Lagerstroemia-dominated semi-deciduous (or mixed-deciduous) though this elevation range is within the distribution range of the species. This observation was similar to previous observation (Robson et al. 1993, Atkins and Tentij 1998). However, this flat lowland dominated by Lagerstroemia-dominated semi-deciduous harbour other species of Arborophila genus scalybreasted partridge (Arborophila chloropus) and this species was commonly hearing in this area but never on the hills adjacent to this region. This suggested that there is a different habitat preference between two species and these species seem sympatric together though the evidence for this is still slender. This is quit interesting as so far ecology of these species are very poorly known. So, our ambition is to carry out ecology research of these species in their range, focusing on: (1) habitat preference of the species; (2) their ecology such as home range size, movement, breeding behaviour to see if their home range overlap each other. This helps us to understand not only about these species distribution, but also the way that animals can coexist in the same place. I used to work for national and local television to create a wildlife program for kids on Ho Chi Minh city Television and I also carried out outreach programs for biology teaches and children at secondary schools in Ho Chi Minh City. These programs equipped me fundamental experiences of how to get people involve in conservation activities. Therefore, I would like to develop an outreach program for teachers and children living in buffer zones of Cat Tien National parks and other protected areas near Cat Tien National park in order to improve their awareness on conservation. By doing this, I would like to link this outreach program with community-based programme of the park so that the park can get more supports from the community for conservation in the long run. Recently, Vinh Cuu nature reserve and Cat Tien national park have developed homestay ecotourism model. Thus, the outreach program also will focus on providing not merely knowledge of biodiversity, conservation but also skills of how to impart well and effectively their knowledge to their audiences. 10. Did you use the RSGF logo in any materials produced in relation to this project? Did the RSGF receive any publicity during the course of your work? As I have mentioned above, I am working on at least two manuscripts and one of these is going to be submitted soon. And of RSGF will be in the acknowledgment as an important sponsor to support budget for this study. And I also have plan to join some workshop next year and the RSGF logo will be appear in my presentation. 11. Any other comments? I would like to express sincere gratitude to Dr Pro. Le Cong Kiet, Professor Vo Quy, G. A. Gale, Dr Tommaso Savini for their support and advices for this research. I also gratefully thank the following

14 people and organizations for their support to this research: Cat Tien National Park, Vinh Cuu Nature reserve where I carried this research; Mr Tran Van Mui (Director of Vinh Cuu NR), Mr Nguyen Van Dien (Director of Cat Tien NP), Institute of Tropical Biology and the Vietnam Academy of Science and Technology, Dr Pham Huu Khanh, Mr Tran Van Binh, Mr Nguyen Luu Hao, Mr Nguyen Tan Hung (Cat Tien NP), Mr Nguyen Hoang Hao and Nguyen Van Diep (Vinh Cuu NR), Mr Nguyen Dang Hoang Vu, Le Van Toan. And I also would like to thank IDEA WILD for their support partial equipment for this research.