42916-014: Sarulla Geothermal Power

Home , Banded Linsang, Banded palm civet, Common treeshrew, Exilisciurus, Large treeshrew, Lariscus

Environmental and Social Monitoring Report

Safeguard and Social Monitoring Report Appendix G – Biodiversity Monitoring 2016 (Monitoring period: July – December 2016) June 2018

INO: Sarulla Geothermal Power

Prepared by Sarulla Operations Limited for the Asian Development Bank.

This safeguard and social monitoring report is a document of the borrower. The views expressed herein do not necessarily represent those of ADB's Board of Directors, Management, or staff, and may be preliminary in nature.

In preparing any country program or strategy, financing any project, or by making any designation of or reference to a particular territory or geographic area in this document, the Asian Development Bank does not intend to make any judgments as to the legal or other status of any territory or area.

Biodiversity Monitoring 2016

SOL Geothermal Project Area, North Tapanuli, North Sumatra

Camera trapping and Primate Surveys

A picture of a female sambar deer (Rusa unicolor), listed as Vulnerable on the IUCN Red List and protected species under Indonesian law, but still heavily hunted throughout Sumatra, caught on camera trap at SOL.

27th of November 2016 by PanEco/Yayasan Ekosistem Lestari

requested by

Sarulla Operations Ltd.

Executive Summary

Biodiversity monitoring is an integral part of the Biodiversity Action Plan (BAP) that was developed as part of a large-scale geothermal development project in North Tapanuli, North Sumatra, Indonesia. Baseline surveys were carried out between November 2013-June 2014 by Yayasan Ekosistem Lestari [camera trapping, primate and wildlife surveys, reptile and amphibian surveys] and a team of experts looking at vegetation and bird diversity. This report discusses the results of repeat biodiversity surveys focusing on camera trapping [3rd March-14 August 2016] and specific primate surveys [25 July-16 August 2016] in the forest area where Sarulla Operations Limited (SOL) is developing a large-scale geothermal project in/adjacent to the protected Batang Toru forest complex. These repeat biodiversity surveys are intended as a monitoring tool to investigate temporal changes in wildlife populations in relation to human disturbance in the SOL project area. The aim is to balance SOL geothermal activities with minimal impact on wildlife populations and to provide recommendations what is needed to maintain healthy wildlife populations in the Batang Toru forest surrounding the SOL geothermal area. The biodiversity data compiled in 2016, through camera trapping and primate surveys, have been compared with baseline data that was collected in the SOL project area between November 2013-June 2014. A serious decline in species diversity as well as relative abundance of the more common species was encountered between the two sampling periods. Land clearing activities have proceeded at a serious pace, with very wide road construction through the forest to the WJP area, with large amount of traffic and infrastructure development, which have obviously had a negative effect on the wildlife living in the SOL area.

Key findings

¡ A total of 17 mammal species were encountered during the biodiversity surveys [combined direct observations, camera trapping and sign] in 2016 compared to 27 mammal species during the surveys in 2014;

¡ 11 mammal species were recorded on camera trap in 2016 [664 trapnights, 117 events] compared to 19 species in 2014 [984 trapnights, 527 events];

¡ Of the mammal species recorded in 2016, 8 mammal species are listed under the IUCN Red List as Critically Endangered, Endangered or Vulnerable, and 8 species are protected under Indonesian law;

¡ A significant reduction was found in the number of events/trapnights and the Relative Abundance Index (RAI) of species caught on camera trap between 2016 and 2014;

¡ Of the Critically Endangered species, only one pangolin (Manis javanica, CR) was caught on camera trap in 2016, compared to 3 independent events during the 2014 monitoring session;

¡ No evidence was found of the other two species listed as Critically Endangered on the IUCN Red List, the Sumatran orangutans (Pongo abelii, CR) nor Sumatran tiger (Panthera tigris sumatrae, CR), evidence of both species still encountered during the 2014 surveys;

¡ Density of resident agile gibbon (Hylobates agilis, EN) groups was found to be 2.23 groups/km2 with no significant difference found between 2016 and 2014 density calculations;

¡ Density of resident Siamangs (Symphalangus syndactylus, EN) was found to be 0.7 groups/km2 with no significant difference found between 2016 and 2014 density calculations;

¡ Mitred leaf monkeys (Presbytis melalophos, EN) were directly observed once during the surveys;

¡ Pig-tailed macaques (Macaca nemestrina, VU) were the most common species caught on camera traps together with the Barking deer (Muntiacus muntjac) but their Relative Abundance Index was significantly reduced from 14.4 in 2014 to 4.5 in 2016;

¡ Sun bear (Helarctos malayanus, VU) sign (claw marks and feeding sign) were still encountered along 4 of the primate transects and a small sun bear cub [est. 3 months old] was encountered in a den site along one of the primate transects;

¡ Sambar deer (Rusa unicolor, VU) were caught on camera trap on 16 independent events, and although their Relative Abundance Index was reduced from 3.2 in 2014 to 2.4 in 2016, this was the least drastic reduction of the three most frequently camera trapped species;

Recommendations Due to the high occurrence of a significant number of IUCN Red Listed species as well as wildlife protected under Indonesian law in the SOL project area, we provide the following recommendations:

Connectivity: The new road constructed in the forested areas to WJP 1 is very wide [estimated 20 meters] and no arboreal connectivity has been maintained. This will seriously affect arboreal primates (gibbons, siamang, orangutans, mitred leaf monkeys) to continue to use their home ranges that have been dissected by this road development. A professional assessment is needed to evaluate and plan how to restore connectivity in the most likely places along this road to avoid further decline in movement patters of various arboreal and terrestrial species.

Hunting: Hunting of wildlife was still found to occur in the SOL project area, with a fresh snare encountered along one of the transects. The following recommendations are made to halt illegal hunting in the SOL area:

¡ Develop conservation/hunting agreements focusing on protected species with the main villages/hamlets close to the project area, and in return provide support for developing small enterprises that will provide sustainable protein alternatives (such as fish ponds, chicken farming, pig farming);

¡ Strict 24-hour control on free access using project roads;

¡ Establish a SOL wildlife protection team with a clear mandate to patrol forest area surrounding the SOL project area.

Illegal Logging: Illegal logging within the SOL project area was still encountered on several occasions with piles of sawn timber encountered on three occasions.

General recommendations:

¡ Human disturbance (i.e. noise, motorbikes, music), as well as dogs, cats, and invasive species should be minimized as much as possible, as well as increased garbage management;

¡ Strict regulations should be put in place regarding the above, both for SOL workers, contractors, visitors and local communities passing through the area; Biodiversity monitoring:

¡ Relating to the steep decline in Relative Abundance Index of most of the more common species, and decline in species diversity observed between 2016 and 2014, we recommend more intensive camera trapping to be carried out with similar amount of trapnights [and time period] as in 2014 so that comparison of biodiversity data between years will be more solid;

¡ We recommend repeat bird and reptile/amphibians surveys as important indicators of ecological stability, as well as abundance measures regarding illegal hunting for song birds;

General Comments: Because of its scale and importance to the local economy, SOL is perfectly placed to introduce improved environmental practices by implementing the above recommendations for reduced disturbance, reconstruction of connectivity, and prevention of hunting of protected wildlife, illegal logging, and garbage management. While the company may feel it maintains the support of local communities by bowing to pressure to allow open access to the forests surrounding its worksites, in the long term this policy can rebound on the company as its practices become more widely known. On the other hand, rigorously imposing best practices on its own workforce and dissemination of these amongst local communities will enhance the company's international reputation, provide a strong case study for other companies operating in Indonesia, and modify environmentally damaging practices by local communities that appear to currently be accelerating around the work sites.

iii

Furthermore, the company is building a relatively environmentally-friendly energy power plant that have local impacts on the highly biodiverse and unique Batang Toru forests. But it has the opportunity to play a very important role in the long-term conservation of the wider Batang Toru forest complex (both East and West blocks), thus further demonstrating a wider commitment to environmental best practices and conservation.

Credits

Report prepared by PanEco/Yayasan Ekosistem Lestari:

• Gabriella M. Fredriksson, • Graham Usher, • Matthew G. Nowak;

Field data collection & survey teams:

• Eka Siswiyati (Primate surveys) • Dewi Kurnia (Primate surveys) • Nursaniah Nasution (Primate surveys) • Pardi Sitompul (Local assistant provided by SOL) • Hanjandri Matondang (Local assistant provided by SOL) • Parel Sitompul (Local assistant provided by SOL)

Acknowledgements We would like to express our gratitude to Alan Pierce for help with organizing the camera trapping database and all the SOL field workers who joined the primate surveys and camera trapping efforts.

Contents Executive Summary ...... i Key findings ...... i Recommendations ...... ii Credits ...... v Acknowledgements ...... v 1. Introduction ...... 1 1.1 Conservation Importance of the Batang Toru forest complex ...... 1 1.2 Background to Surveys ...... 4 1.2.1 SOL Site description ...... 4 1.2.2 Geographical scope ...... 5 1.2.3 Aims of the biodiversity monitoring field surveys ...... 6 1.2.4 Target species ...... 6 2. Methodology ...... 7 2.1 Camera Trapping ...... 7 2.1.1 Equipment& time frame ...... 7 2.1.2 Locations of camera traps ...... 7 2.1.3 Monthly Camera Trap Checking ...... 7 2.2 Primate surveys ...... 9 2.2.1 Orangutan Nest Surveys ...... 9 2.2.2 Gibbon/Siamang Fixed Call Count Surveys ...... 10 3. Results ...... 11 3.1 Results - Camera Trapping ...... 11 3.1.1 Time-frame and number of independent events ...... 11 3.1.2 Description of Camera Trap Locations ...... 11 3.1.3 Species encountered on camera traps ...... 13 3.1.4 Relative abundance and species encounters [2014 versus 2016] ...... 15 3.2 Results - Line Transect Surveys ...... 17 3.3 Results – Orangutan Nests ...... 19 3.4 Results – Gibbon & Siamang Vocal Surveys ...... 19 4. Discussion ...... 25 4.1 Discussion wildlife monitoring through Primate Surveys and Camera Trapping ... 25 5. REFERENCES: ...... 28 APPENDIX I ...... 30

1. Introduction

1.1 Conservation Importance of the Batang Toru forest complex The East and West Batang Toru Forest Blocks have received significant conservation attention over the last decade, after the 'discovery' of a viable orangutan population in the late 1990s (Wich et al. 2003). This, and follow-up surveys in the East and West Batang Toru Forest Blocks have resulted in the current population estimate of close to 800 individuals (PHVA 2016). This area harbours the only viable orangutan population south of Lake Toba, which is now known to be genetically unique relative to other orangutan populations (Nater et al. 2011, 2012), making it the most endangered Sumatran orangutan population. The West Batang Toru Forest Block (WBTFB) consists of medium elevation hill and submontane forest primarily covering extremely rugged terrain (400-1300m asl.), and contains orangutans throughout its c. 810 km2 (81.344 ha) of forest. A smaller orangutan population has also been encountered in the East Batang Toru forest block (forest area: 54.940 ha), east of the Tarutung-Sipirok road. The Batang Toru forests are also home to the critically endangered Sumatran tiger. The most recent published Sumatran tiger population, estimated at 300 individuals in the wild, is scattered throughout the island of Sumatra (Ministry of Forestry, 2007), although this number is considered to be an underestimate (Wibisono et al., 2011). Sumatran tigers have been camera trapped in various locations and observed directly on rare occasions throughout the West and East Batang Toru Forest Blocks over the last decade. In addition, orangutans, tigers, tapirs, serow, pangolins and a variety of other Critically Endangered, Endangered and Vulnerable mammal species have been caught regularly on camera traps in an ongoing monitoring programme carried out by PanEco/YEL in the Batang Toru Forest Complex. Extensive biodiversity surveys have been carried out in the West Batang Toru Forest Block (PT.Newmont Horas Nauli/Hatfindo 2003, PanEco/YEL 2006-onwards, Conservation International 2007/2008) and the East Batang Toru Forest Block (PanEco/Yayasan Ekosistem Lestari 2007-2015). Since 2006, PanEco/Yayasan Ekosistem Lestari (YEL), who jointly run the Sumatran Orangutan Conservation Project (SOCP) have been operating a flora and fauna monitoring station in the West Batang Toru Forest Block, located some 20 km from the SOL project area (see Fig. 1). The forest area in between the two locations is contiguous primary forest, at elevations around 1000 m asl.

Fig 1: Location of the SOL project area in the West Batang Toru Forest Block and the Sarulla valley [pink outline], and location of the SOCP research station [red circle].

To date, through a combined effort of camera trapping, direct observations and preliminary studies of small mammals a total of 60 mammal species (non-Chiropteran) have been documented in the Batang Toru Forest complex (Table 1).

Order Latin English IUCN Red List Protected 1 Primata-Pongidae Pongo abelii Sumtatran orangutan CR 1 2 Carnivora-Felidae Panthera tigris sumatrae Sumatran tiger CR 1 3 Pholidota-Manidae Manis javanica Pangolin CR 1 4 Perissodactyla_Tapiridae Tapirus indicus Asian Tapir EN 1 5 Primata-Cercopithecidae Presbytis melalophos Mitred leaf monkey EN 6 Primata-Hylobtaidae Hylobates agilis Agile gibbon EN 1 7 Primata-Hylobtaidae Symphalangus syndactylus Siamang EN 1 8 Artiodactyla_Cervidae Capricornis sumatrensis Serow VU 1 9 Artiodactyla_Cervidae Cervus unicolor Sambar deer VU 1 10 Carnivora_Mustelidae Aonyx cinerea Oriental small-clawed otter VU 1 11 Carnivora_Viverridae Arctitis binturong Binturong VU 1 12 Carnivora_Viverridae Hemigalus derbyanus Banded Palm Civet VU 13 Carnivora-Felidae Pardofelis marmorata Marbled cat VU 1 14 Carnivora-Ursidae Helarctos malayanus Malayan sun bear VU 1 15 Primata-Cercopithecidae Macaca nemistrina Pig tailed macaque VU 16 Primata-Lorisidae Nycticebus coucang Slow loris VU 1 17 Rodentia_Muridae Maxomys whiteheadi Whitehead’s spiny rat VU 18 Rodentia_Muridae Niviventer cremoriventer Dark-tailed tree rat VU 19 Carnivora-Felidae Pardofelis teminckii Golden cat NT 1 20 Rodentia_Sciuridae Callosciurus nigrovittatus Black banded squirrel NT 21 Rodentia_Sciuridae Ratufa affinis hypoleucos Cream-coloured giant squirrel NT 22 Rodentia_Sciuridae Ratufa bicolor palliata Black giant squirrel NT 1 23 Rodentia_Sciuridae Rhinosciurus laticaudatus Shrew-faced ground squirrel NT 24 Rodentia_Sciuridae Sundasciurus hippurus Horse-tailed squirrel NT 25 Artiodactyla_Cervidae Muntiacus muntjac Common Barking deer LC 1 26 Artiodactyla_Suidae Sus scrofa Wild pig LC 27 Artiodactyla_Tragulidae Tragulus javanicus Lesser mouse deer LC 1 28 Artiodactyla_Tragulidae Tragulus napu Greater mouse deer LC 1 29 Carnivora_Mustelidae Martes flavigula Yellow-throated marten LC 30 Carnivora_Mustelidae Mustela nudipes Malay weasel LC 31 Carnivora_Viverridae Arctogalidia trivirgata Small-toothed palm civet LC 32 Carnivora_Viverridae Paguma larvata Masked palm civet LC 33 Carnivora_Viverridae Paradoxurus hermaphroditus Common palm civet LC 34 Carnivora_Viverridae Prionodon linsang Banded Linsang LC 1 35 Carnivora-Felidae Prionailurus bengalensis Leopard cat LC 1 36 Dermoptera-Cynocephalidae Cynocephalus variegatus Colugo LC 1 37 Insectivora-Erinaceidae Echinosorex gymnura Moonrat LC 38 Primata-Cercopithecidae Macaca fascicularis Long-tailed macaque LC 39 Rodentia_Hystricidae Hystrix brachyura Asian porcupine LC 1 40 Rodentia_Hystricidae Trichys fasciculata Long-tailed porcupine LC 41 Rodentia_Muridae Leopoldamys sabanus Long-tailed giant rat LC 42 Rodentia_Muridae Maxomys surifer Red spiny rat LC 43 Rodentia_Muridae Niviventer fulvescens Chestnut rat LC 44 Rodentia_Muridae Rattus exulans Polynesian rat LC 45 Rodentia_Muridae Rattus tanezumi Japan’s house rat LC 46 Rodentia_Muridae Rattus tiomanicus Malaysian field rat LC 47 Rodentia_Muridae Sundamys muelleri Muller’s rat LC 48 Rodentia_Rhizomyinae Rhizomys sumatrensis Sumatran Bamboo Rat LC 49 Rodentia_Sciuridae Callosciurus notatus Plantain squirrel LC 50 Rodentia_Sciuridae Callosciurus prevostii piceaus Prevosts squirrel LC 51 Rodentia_Sciuridae Iomys horsfieldii horsfieldii Javanese flying squirrel LC 1 52 Rodentia_Sciuridae Lariscus insignis Three-striped ground squirrel LC 1 53 Rodentia_Sciuridae Nannosciurus melanotis Black-eared pygmy squirrel LC 54 Rodentia_Sciuridae Petaurista petaurista Red Giant Flying squirrel LC 55 Rodentia_Sciuridae Sundasciurus lowii Low's squirrel LC 56 Rodentia_Sciuridae Sundasciurus tenuis Slender squirrel LC 57 Scandentia-Tupaiidae Tupaia glis Common treeshrew LC 58 Scandentia-Tupaiidae Tupaia tana Large treeshrew LC 59 Carnivora_Viverridae Herpestes semitorquatus Collared Mongoose DD 60 Rodentia_Sciuridae Exilisciurus exilis Plain pygmy squirrel DD Table 1: Mammal species (non-Chiropteran) encountered in the West Batang Toru Forest Block (PanEco/YEL data 2006-2015) and their IUCN/Indonesian Protected status.

1.2 Background to Surveys In September 2013, PanEco/YEL were approached by Mott MacDonald, who had been commissioned by Sarulla Operations Ltd (SOL) to coordinate biodiversity monitoring field surveys at their Sarulla geothermal field, to assist with these surveys. An initial reconnaissance (recce) was carried out in November 2013, followed by more intensive line transect and primate call monitoring surveys in June 2014. Between November 2013 and June 2014 a camera trapping programme was run. In 2016 PanEco/YEL was asked by SOL to assist with further biodiversity monitoring, where it was decided that SOL's environmental department would carry out camera trapping efforts themselves in the field, with PanEco/YEL assisting with identification and analyses of camera trapping data, and a PanEco/YEL survey team focusing on repeat primate surveys.

1.2.1 SOL Site description The following description of the site, geographical location, the aims of the monitoring surveys, and list of species of interest are taken from the March 2014 Mott MacDonald proposal on the Biodiversity Monitoring Metholodogy (in italics). The overall Sarulla Geothermal Project comprises two geothermal fields (Namora I Langit or "NIL" field, and Silangkitang or "SIL" field), each consisting of a power station (two at NIL), production and re-injection wells, connecting pipeline, and a transmission line to the Perusahaan Listrik Negara (PLN) power sub-station. The Project is located between the East and West Batang Toru Forest Blocks in the Sarulla valley (Figs. 1 and 2). The majority (approximately 90%) of the infrastructure will be located within existing areas of Modified habitats (as defined under IFC Performance Standard 6) and are represented by four vegetation communities: yard vegetation, field vegetation, rubber plantation and mixed plantation vegetation. These habitats have been assessed as not being of significant conservation importance and the impact upon them is considered to be negligible (Biodiversity Impact Assessment, ERM, 2013). These areas will not be surveyed as part of this biodiversity monitoring program. Approximately 10% of the Project will be located within an area identified as Natural/ Critical Natural Habitat (IFC PS6). This area is situated around the infrastructure of NIL1 and WJP1. The habitats within the Project footprint comprise secondary lower montane (dipterocarp) forest, natural scrub (padang vegetation), heath forest, non-natural scrub and mixed forest (Habitat Condition Assessment, Mott MacDonald, 2013). Collectively this area covers approximately 8ha.The biodiversity monitoring program will target the habitats within these areas. Approximately 92ha of the Project is located within the Western Batang Toru Forest Block (WBTFB). This area is situated around the infrastructure of the NIL Geothermal Field. The habitats within and adjacent to the Project comprise secondary (dipterocarp) forest, natural scrub (padang vegetation), scrub forest (heath forest), non-natural scrub, and mixed forest (Mott MacDonald, 2013). The biodiversity monitoring program will target these habitats within this area.

1.2.2 Geographical scope The geographical scope of the surveys comprised three zones as follows: Zone 1: Habitats within the footprint of the infrastructure identified as Natural/Critical Natural Habitat (NIL1 and WJP1). Zone 2: Habitats within 500m of the footprint of the infrastructure identified as Natural/Critical Natural Habitat. Zone 3: Habitats over 500m from the footprint of the infrastructure identified as Natural/Critical Natural Habitat. These comprise the wider landscape within the Batang Toru Forest. The biodiversity monitoring field surveys targeted the habitats located within Zones 1 & 2 (collectively hereafter the ‘Project Area’), as well as Zone 3 as these areas are considered part of the ranges of various species occurring in the 'Project Area'.

Fig 2: Detailed map of the SOL project area in the West Batang Toru Forest Block and the Sarulla valley.

1.2.3 Aims of the biodiversity monitoring field surveys The aims of the biodiversity monitoring field surveys are as follows: To provide further baseline data on the likely presence and habitat usage of species of conservation importance (‘Target Species’), which are known or have the potential to occur in the Project Area. This includes:

− Creating habitat and species distribution maps. − Interpreting data collected from the field surveys in the context of the Project and the wider conservation objectives of the Batang Toru Forest.

To monitor the likely effects of the Project on Target Species and to inform changes to the Environmental Management Plan as well as requirements within the Biodiversity Action Plan and Biodiversity Off-set Management Plan.

1.2.4 Target species Seven species were initially identified as occurring, or having the potential to occur, within the Project Area and were said to be listed by the IUCN Red List as Critically Endangered or Endangered. This initial list of target species was as follows:

1. Sumatran orang-utan (Pongo abelii) [Critically Endangered, IUCN Red List]; 2. Sumatran tiger (Panthera tigris sumatrae) [Critically Endangered, IUCN Red List]; 3. Malayan pangolin (Manis javanica) [Critically Endangered, IUCN Red List]; 4. Asian tapir (Tapirus indicus) [Endangered, IUCN Red List]; 5. Mitred leaf-monkey (Presbytis melalophos) [Endangered, IUCN Red List]; 6. Siamang (Symphalangus syndactylus) [Endangered, IUCN Red List]; 7. Agile gibbon (Hylobates agilis) [Endangered, IUCN Red List].

A list of other species identified within the Project area was put forward, most recognized as species of conservation importance and listed as Vulnerable on the IUCN Red List, except for the wild pig, an important prey species for Sumatran tiger. These were to be considered secondary species, to be recorded during the biodiversity monitoring, but not be subject to species-specific surveys:

1. Marbled cat (Pardofelis marmorata) [Vulnerable, IUCN Red List]; 2. Sumatran serow (Capricornis sumatraensis) [Vulnerable, IUCN Red List]; 3. Sambar deer (Rusa unicolor) [Vulnerable, IUCN Red List]; 4. Malayan sun bear (Helarctos malayanus) [Vulnerable, IUCN Red List]; 5. Binturong (Arctictis binturong) [Vulnerable, IUCN Red List]; 6. Slow loris (Nycticebus coucang) [Vulnerable, IUCN Red List]. 7. Wild pig (Sus scrofa). A number of other species listed as Vulnerable under the IUCN red-list are also known to occur in the Batang Toru forest (see Table 1).

2. Methodology

2.1 Camera Trapping Camera traps are the preferred survey technique for some of the more cryptic fauna (e.g., tiger, marbled cat, pangolins, tapirs). PanEco/YEL have been operating camera traps in the Batang Toru Forest Complex starting in 2008. Camera traps were utilized for the biodiversity surveys in the SOL project area to determine the presence, distribution, and relative abundance of various cryptic faunal species. Techniques followed here are similar to those described by Ancrenaz et al. (2012).

2.1.1 Equipment& time frame A total of 5 Bushnell Trail Cam camera traps were used for the camera trapping efforts in 2016. These camera's all use a motion sensor and hyper passive infrared sensor (PIR) to detect heat to trigger the camera to take an image. These cameras are in theory less prone to false captures. All camera's used high speed SD cards [8 or 16GB]. Camera trapping commenced on 3rd of March 2016 and continued till 14 of August 2016.

2.1.2 Locations of camera traps Camera trapping survey aimed to cover a combination of habitat types and locations within the forest areas of NIL1 and WJP1. The 5 camera traps were deployed within the ‘Project Area’, placed at least 100 meters from one another.

2.1.3 Monthly Camera Trap Checking Following placement, the camera-traps were run between 2-3 months at a single location, depending on previous results. Camera-traps were checked approximately once a month, and the following actions were taken:

¡ Check and replace batteries; ¡ Change SD card; ¡ Ensure that the camera is still functioning and that there is nothing obstructing the camera’s lens. After each camera trap session the following data was be recorded for each camera trap night:

¡ Species captured; ¡ The number of capture events per species; ¡ The number of individuals per event. Each period that one camera trap was left in the field for approximately 1 month was labelled as a separate a "session". A camera trap could be left for another session at the same location if several or interesting animals were photographed during the previous session.

Locations for placement of the camera traps were chosen to maximize results, thus traps were placed in locations with tracks, scats or animal trails. The camera trap was generally placed 50cm above the ground to try to detect animals of various size, although this was partly dependent on slopes in the different locations. Camera traps were set to take three picture each time the sensor was triggered to maximize chances of identification. Identification of species was assisted with the help of several mammal identification books:

− A field guide to the Mammals of Borneo by Q. Phillips and K. Phillips (2016);

− A field guide to the Mammals of Borneo by J. Payne, C. Francis, K. Phillips (1998);

− A field guide to the Mammals of South-East Asia by C. Francis (2008). Some pictures were sent to specialist to assist with identification [mainly Tupaiidae and Muridae]. We defined a species recording as an ‘event’ when a photograph (or series of photographs) of an individual species (counting number of individuals of that species) was recorded by the same camera at one location, including all other pictures of the same species taken within a time period of 30 minutes at the same camera placement (O'Brien et al. 2003, Linkie and Riboud 2011). Photographs that were recorded within 30 min of a previous photograph of the same species, and at the same camera placement, were not recorded as an event, because they were not considered to be independent species events.

We report number of animal detections [events] and calculate Relative Abundance Indices (RAI) (Carbone et al. 2001, O’Brien et al. 2003) for the most frequently recorded species, as a measure to compare frequencies/ relative abundances between 2014 and 2016. To compute the RAI for each species, all detections for each species are summed for all camera traps over all days, multiplied by 100, and divided by the total number of camera trap nights.

2.2 Primate surveys Primate surveys were to be undertaken between the 25th of July and 16th of August 2016. The following materials and methods were to be used:

2.2.1 Orangutan Nest Surveys Using orangutan nest surveys, relative orangutan densities were to be calculated for the Project area following a previously published protocol (e.g. van Schaik et al., 1995, Buij et al., 2002, Buij et al., 2003, Wich et al., 2004). This methodology relies on the fact that orangutans construct nests daily, which are used at night and also sometimes during the day for resting (van Schaik et al., 1995, Singleton, 2000). Using nest counts instead of live encounters with orangutans is preferred here due to the low density of orangutans, which makes density estimates based on live encounters a very time consuming exercise.

Nest counts were to be conducted along 22 randomly placed transects in the Project area, following methods of Marshall et al. (2008) and Buckland et al. (2010), with each transect being 500 m in length, with a minimum of 200 meters between transects. All line transects should be only large enough for observers to walk without disruption. It is recommended that the surveyors wait a week after the line transects are cut before they start the line transect surveys. This allows the animal population that may have been disturbed during the making of the line transects, to come back into the area.

During nest counts, two experienced observers slowly walked on transects and record the perpendicular distance of all identified nests. Perpendicular distances are measured with a hip chain. Each transect was walked twice, once in each direction.

Orangutan density was estimated from nest counts using the following formula:

d = N/(L * 2 * w * p * r * t) in which :

d = orangutan density (individuals/km2), N = number of nests observed along the transect, L = length of the transect covered (km), w = estimated width of the strip of habitat actually visually covered (km), p = proportion of nest builders in the population, r = rate at which nests are produced (n/day/individual), and t = time during which a nest remains visible (in days).

Estimating strip width (w): Using the measured perpendicular distance of nests from the transect line, (w) was estimated using the computer programme ''Distance 6.0'' (Thomas et al., 2001).

Parameters (p) and (r): In two long-term studies of Sumatran orangutans, roughly 10% of the population was comprised of non-nest building infants, and as such, (p) was set at 0.9 (van

Schaik et al., 1995). Conversely, the number of nests built per day (r) was found to be approximately 1.25 at SOCP’s long-term monitoring station (SOCP, unpublished data).

Nest decay rate (t): A study by Wich et al. (2016) found that nest decay rate in Batang Toru was 502 days.

2.2.2 Gibbon/Siamang Fixed Call Count Surveys While gibbons and siamangs can also be surveyed via line transects, the abundance estimates that are obtained for direct contact surveys are considered a minimum estimate, as these species are extremely cryptic. However, both these species have distinctive calls and thus the density of gibbon and siamang groups was to be calculated using the fixed call count methodology established by Brockelman & Ali (1987). In particular, a series of three listening stations (all consisting of a set of three listening posts each), were each to be monitored for five consecutive days. Each survey day, between 05:00 – 12:00, surveyors from each listening post recorded the time of a group’s call, the species (i.e., gibbon or siamang), the compass bearing, and the estimated distance. Via triangulation, group calls were then marked on a map in GIS, and the total number of gibbon and siamang groups heard was estimated for each listening location.

In order to control for call point location estimation errors, vocalization points were restricted to within a 1 km radius from any given listening post. The effective listening area (E) for both gibbons and siamangs was calculated as 3.14 km2 for a given array, which represents the area within 1 km of the centroid of a given vocal array.

Group density estimates (D) was calculated using the equation (Brockelman & Ali, 1987):

D = n/[p(m) x E] where:

n = the number of groups heard, p(m) = the estimated proportion of groups expected to sing at a survey location during a sampling period of m days, E = Effective listening area.

Also following Brockelman & Ali (1987), p(m) was determined using the equation:

p(m)=1-[1-p(1)]m where:

p(1) = the average probability of singing for any given day, m = the number of survey days.

3. Results

3.1 Results - Camera Trapping

3.1.1 Time-frame and number of independent events Between 3rd of March and 14th of August 2016, a total of 664 trap nights were carried out. During these, a total of 701 pictures were of mammal/bird species. In total 117 independent events were recorded.

3.1.2 Description of Camera Trap Locations A total of 6 different locations were used for placement of the camera traps. Table 2 describes the beginning and end date of placement at that location, and altitude. Unfortunately no vegetation description of each location was provided. Fig. 3 shows the locations of camera trap placement in relation to NIL 1 and WJP.

Loc Sess. Date start Date finish Total Days NORTHING EASTING LAT_2 LONG_2 Alt Vegetation L3 1 04-Mar-16 26-Mar-16 23 501246 207932 1.881215 99.011202 1065 L3 5 26-Mar-16 15-Apr-16 22 501246 207932 1.881215 99.011202 1065 L3 19 16-May-16 09-Jun-16 23 501246 207932 1.881215 99.011202 1065 L4 2 04-Mar-16 26-Mar-16 23 501450 207954 1.881416 99.013041 1078 L4 10 15-Apr-16 16-May-16 30 501450 207954 1.881416 99.013041 1078 L4 28 11-Jul-16 14-Aug-16 33 501450 207954 1.881416 99.013041 1078 L7 4 03-Mar-16 24-Mar-16 21 502295 207152 1.87416 99.02060 916 L7 6 13-Apr-16 23-Apr-16 10 502295 207152 1.87416 99.02060 916 L7 11 23-Apr-16 13-May-16 20 502295 207152 1.87416 99.02060 916 L7 17 14-May-16 08-Jun-16 24 502295 207152 1.87416 99.02060 916 L7 23 09-Jun-16 10-Jul-16 30 502295 207152 1.87416 99.02060 916 L7 25 11-Jul-16 14-Aug-16 33 502295 207152 1.87416 99.02060 916 L8 3 03-Mar-16 24-Mar-16 21 503315 207755 1.87962 99.02980 945 L8 7 13-Apr-16 23-Apr-16 10 503315 207755 1.87962 99.02980 945 L8 12 23-Apr-16 13-May-16 20 503315 207755 1.87962 99.02980 945 L8 16 14-May-16 08-Jun-16 24 503315 207755 1.87962 99.02980 945 L8 21 09-Jun-16 10-Jul-16 30 503315 207755 1.87962 99.02980 945 L8 26 11-Jul-16 14-Aug-16 33 503315 207755 1.87962 99.02980 945 L11 8 13-Apr-16 23-Apr-16 10 502691 207198 L11 13 23-Apr-16 13-May-16 20 502691 207198 L11 15 14-May-16 08-Jun-16 24 502691 207198 L11 22 09-Jun-16 10-Jul-16 30 502691 207198 L11 27 11-Jul-16 14-Aug-16 33 502691 207198 L12 9 13-Apr-16 23-Apr-16 10 502441 207214 L12 14 23-Apr-16 13-May-16 20 502441 207214 L12 18 14-May-16 08-Jun-16 24 502441 207214 L12 20 09-Jun-16 10-Jul-16 30 502441 207214 L12 24 11-Jul-16 14-Aug-16 33 502441 207214 Table 2: Location, number of camera at that location, begin and end date of camera trapping at that location, coordinates, altitude of each camera trapping location. No vegetation description was provided by the SOL team for the camera trapping locations.

Fig. 3. Location of camera trap placements. Numbers (L3-L12) refer to Table 2.

3.1.3 Species encountered on camera traps A total of 11 mammal species were photographed and identified to species level (table 3). The three most common species photographed made up 65% of all events. These were the Red Muntjac/Barking deer Muntiacus muntjac (n=30 events, 25.6%) and Pig-tailed macaques, Macaca nemestrina (n=30 events, 25.6%), followed by Sambar deer Rusa unicolor (n=16 events, 13.7%).

no Species Scientific name No of events % IUCN status Protected 1 Red Muntjac/Barking deer Muntiacus muntjac 30 25.6 LC X 2 Pig-tailed macaque Macaca nemestrina 30 25.6 VU 3 Sambar deer Rusa unicolor 16 13.7 VU X 4 Greater mouse-deer Tragulus napu 5 4.3 LC X 5 Wild pig Sus scrofa 4 3.4 LC 6 Three-striped ground squirrel Lariscus insignis 3 2.6 LC X 7 Banded Linsang Prionodon linsang 3 2.6 LC X 8 Shrew-faced ground squirrel Rhinosciurus laticaudatus 2 1.7 NT 9 Large Treeshrew Tupaia tana 2 1.7 LC 10 Pangolin Manis javanica 1 0.9 CR X 11 Malayan Porcupine Hystrix brachyura 1 0.9 LC Rat/mice_unidentified Muridae 4 3.4 Mammal_Unidentified 3 2.6 Squirrels_unidentified Sciuridae 2 1.7 Treeshrews_unidentified Tupaiidae 2 1.7

1 Emerald dove Chalcophaps indica 4 3.4 2 Red Junglefowl Gallus gallus 2 1.7 LC 3 Sumatran peacock pheasant Polyplectron chalcurum 1 0.9 LC 4 Ferruginous Partridge Caloperdix oculea 1 0.9 NT Birds_unidentified 1 0.9

TOTAL EVENTS 117 100.0 Table 3: Species caught on camera trap at SOL (3rd of March and 14th of August 2016), ranked by number of events.

Ungulates made up 47% of all camera trap events, whereas carnivores only made up 2.6% of events [one species caught on camera trap, the Banded linsang with 3 events], and birds 7.7%. Several interesting birds from the Family Phasianidae were caught on camera trap, namely the Red Junglefowl, the Sumatran peacock pheasant and the Ferruginous Partridge. A total of 11 camera trap events (9.4%) could not be identified to species level. These were primarily pictures of rat, mice, tree shrews and squirrels.

A total of 40.2 % of events were photographs of species listed as Critically Endangered, Endangered or Vulnerable under the IUCN Red List. Species protected under Indonesian law made up 49.6% of all events (Table 4).

Species Scientific name IUCN status Protected No of events % Pangolin Manis javanica EN X 1 0.9 Pig-tailed macaque Macaca nemestrina VU 30 25.6 Sambar deer Rusa unicolor VU X 16 13.7 Red Muntjac/Barking deer Muntiacus muntjac LC X 30 25.6 Greater mouse-deer Tragulus napu LC X 5 4.3 Three-striped ground squirrel Lariscus insignis LC X 3 2.6 Banded Linsang Prionodon linsang LC X 3 2.6 Malay Porcupine Hystrix brachyura LC X 1 0.9 Table 4: IUCN listing and protected status of species camera trapped at SOL in 2016.

No pictures were obtained of Sumatran tiger, sun bears, serow, or tapir, despite the fact that indirect evidence of the first three of these species was found in November 2013 and during surveys in June 2014 (Fredriksson et. al. 2014}

A total of 17 mammal species were encountered in the SOL area between March-August 2016 from camera trapping results and during the primate surveys (see table 5).

# English Latin IUCN Protect CT Obs Sign CT Obs Sign 2014 2016 1 Sumatran tiger Panthera tigris sumatrae CE 1 1 2 Sumtatran orangutan Pongo abelii CE 1 1 3 Pangolin Manis javanica CE 1 1 1 1 4 Mitred leaf monkey Presbytis melalophos EN 1 1 5 Siamang Symphalangus syndactylus EN 1 1 1 6 Agile gibbon Hylobates agilis EN 1 1 1 7 Marbled cat Pardofelis marmorata VU 1 1 8 Malayan sun bear Helarctos malayanus VU 1 1 1 1 9 Sambar deer Cervus unicolor VU 1 1 1 1 1 10 Serow Capricornis sumatrensis VU 1 1 1 1 11 Pig-tailed macaque Macaca nemistrina VU 1 1 1 1 12 Long-tailed macaque Macaca fascicularis LC 1 13 Banded Linsang Prionodon linsang LC 1 1 1 14 Common Barking deer Muntiacus muntjac LC 1 1 1 1 1 15 Lesser mouse deer Tragulus javanicus LC 1 1 16 Greater mouse deer Tragulus napu LC 1 1 1 17 Three-striped ground squirrel Lariscus insignis LC 1 1 1 18 Shrew-faced ground squirrel Rhinosciurus laticaudatus LC 1 1 1 19 Yellow-throated marten Martes flavigula LC 1 20 Malay weasel Mustela nudipes LC 1 21 Masked palm civet Paguma larvata LC 1 22 Common palm civet Paradoxurus hermaphroditus LC 1 23 Small-toothed palm civet Arctogalidia trivirgata LC 1 24 Wild pig Sus scrofa LC 1 1 1 1 25 Malayan Porcupine Hystrix brachyura LC 1 26 Low's squirrel Sundasciurus lowii LC 1 27 Common treeshrew Tupaia glis LC 1 28 Large treeshrew Tupaia tana LC 1 1 29 Black banded squirrel Callosciurus nigrovittatus NT 1 TOTAL SPECIES 19 6 8 11 6 5 Table 5: Mammal species encountered in the SOL area in 2014 and during the current surveys in 2016, with their IUCN/Protected status and whether species were directly observed, caught on camera trap (CT) or their sign encountered.

3.1.4 Relative abundance and species encounters [2014 versus 2016] Significantly less species were caught on camera trap in 2016 than in 2014 [11 versus 19 species, 42% fewer species]. There specifically seemed to be a lack of species in the Ordo Carnivora [Felidae, Viverridae, Mustelidae] documented through camera trapping. The only carnivore species caught on camera both in 2014 and 2016 was the Banded Linsang (Prionodon linsang). Species caught on camera trap in 2014 but not in 2016 included the Masked palm civet (Paguma larvata), the Common palm civet (Paradoxurus hermaphroditus), the Small-toothed palm civet (Arctogalidia trivirgata), the Yellow-throated marten (Martes flavigula) and the Malay weasel (Mustela nudipes).

The Relative Abundance Index (RAI) of most species was also significantly lower in 2016 than in 2014 (table 6). Differences in RAIs may be a result of a number of factors including differences in detection probabilities between surveys, hence we report only the strongest differences in RAI as an indication of true differences in species abundance.

No of No of IUCN Species Scientific name events events % % RAI RAI % ê status Protected 2016 2014 2016 2014 2016 2014 Red Muntjac/Barking deer Muntiacus muntjac 30 154 25.6 29.2 4.5 15.7 71.1 LC X Pig-tailed macaque Macaca nemestrina 30 142 25.6 26.9 4.5 14.4 68.7 VU Sambar deer Rusa unicolor 16 31 13.7 5.9 2.4 3.2 23.5 VU X Greater mouse-deer Tragulus napu 5 24 4.3 4.6 0.8 2.4 69.1 LC X Wild pig Sus scrofa 4 6 3.4 1.1 0.6 0.6 LC Three-striped ground squirrel Lariscus insignis 3 9 2.6 1.7 0.5 0.9 LC X Banded Linsang Prionodon linsang 3 6 2.6 1.1 0.5 0.6 LC X Shrew-faced ground squirrel Rhinosciurus laticaudatus 2 3 1.7 0.6 0.3 0.3 NT Large Treeshrew Tupaia tana 2 1.7 0.3 LC Pangolin Manis javanica 1 3 0.9 0.6 0.2 0.3 EN X Malayan Porcupine Hystrix brachyura 1 0.9 0.2 LC Marbled cat Pardofelis marmorata 2 0.4 0.2 VU X Masked palm civet Paguma larvata 3 0.6 0.3 LC Common palm civet Paradoxurus hermaphroditus 4 0.8 0.4 LC Malay weasel Mustela nudipes 1 0.2 0.1 LC Small-toothed palm civet Arctogalidia trivirgata 1 0.2 0.1 LC Rat/mice_unidentified Muridae 4 12 3.4 2.3 0.6 1.2 Mammal_Unidentified Mammalia 3 15 2.6 2.8 0.5 1.5 Squirrels_unidentified Sciuridae 2 1.7 0.3 Treeshrews_unidentified Tupaiidae 2 69 1.7 13.1 0.3 7.0 Bats_unidentified Chiroptera 2 0.4 0.2 Man & dogs Homo & Canis 23 4.4 2.3

Emerald dove Chalcophaps indica 4 4 3.4 0.8 0.6 0.4 Red Junglefowl Gallus gallus 2 1.7 0.3 LC Sumatran peacock pheasant Polyplectron chalcurum 1 1 0.9 0.2 0.2 0.1 LC Ferruginous Partridge Caloperdix oculea 1 0.9 0.2 NT Eyebrowed thrush Turdus obscurus 1 0.2 0.1 Forest wagtail Deonanthus indicus 1 0.2 0.1 Siberian thrush Zoothera siberica 1 0.2 0.1 Birds_unidentified Aves 1 7 0.9 1.3 0.2 0.7 Snake_unidentified Serpentes 1 0.2 0.1 Skink/Lizard_unidentified Reptilia 1 0.2 0.1 TOTAL EVENTS 117 527 100 100 Table 6: Mammal species encountered in the SOL area in 2014 and 2016, their number and percentage of events, and their Relative Abundance Index [no of events/100 trapnights].

The RAI of barking deer declined from 15.7 barking deer/100 trapnights in 2016 to 4.5 barking deer events/100 trapnights in 2016 (a 71% decline in number of events/100

trapnights). Similarly, the RAI of the greater-mouse deer [protected species by Indonesian law] and pig-tailed macaques [listed as VU] declined by 69.1 and 68.7% respectively.

Overall there was a decline in catch/effort index where in 2016 it took 5.7 days to get an event on camera trap, versus 1.8 days in 2014. In terms of percentage of events, more protected species and IUCN red listed species were caught on camera trap in 2016 compared to 2014 [see table 7].

2013-2014 2016 Trend Period 16 Nov 2013-21st June 2014 3rd March 2016-14 August 2016 No of months 7.5 5.5 Trapnights 984 664 ê Events 527 117 ê Catch/effort index 1.9 5.7 ê Mammal species IDs 19 11 ê

Top 3 species 69.3 65.0 IUCN Red Listed species 33.8 40.2 é Indonesian Protected species 43.5 49.6 é Table 7: Recapitulation of camera trapping effort in 2013-2014 and the current efforts and results in terms of catch/effort index, number of species and percentage of top-3 species, percentage of IUCN red listed species and Indonesian protected species.

3.2 Results - Line Transect Surveys

A total of 22 transects (each 500 m in length) were systematically walked in each direction, for a total survey effort of 22 km. Four primary habitat types were surveyed while on transect, including 1) primary forest, 2) secondary forest, 3) arang forest, and 4) kemenyan gardens (Fig. 4). For descriptions of arang forest and kemenyan gardens, see the SOL Biodiversity Report (2014).

Fig. 4. Location line transects and primate listening posts

During the transect walks, efforts were made to collect data on all vertebrate direct contacts and indirect signs (e.g., feces, feeding remains, footprints, scratch marks, nests, and vocalizations). From the 22 km walked, there were a total of 63 direct and indirect vertebrate signs recorded (Table 8). This represented a slight increase from the 53 signs recorded in the 2014 survey. More signs were detected on Transects 2, 11, 14, 15, 22, and 23 in 2016 than in the previous 2014 survey, whereas fewer signs were found on Transects 3,

4, 6, 7, 9, 10, 12, 13, 16, 17, 18, 21, and 24 in 2016 that that of 2014. There were no differences in the number of animal signs detected among Transects 1, 5, and 25.

Total Total Transect # T1 T2 T3 T4 T5 T6 T7 T9 T10 T11 T12 T13 T14 T15 T16 T17 T18 T21 T22 T23 T24 T25 Trend (2016) (2014) Pheasant 1 1 2 1 ↑ Wild Boar 1 4 3 10 1 1 4 1 25 2 ↑ Pig-tailed Macaque 0 1 ↓ Agile Gibbon 0 1 ↓ Muntjac 5 5 10 7 ↑ Hornbill 1 1 1 3 1 ↑ Sambar Deer 2 1 1 4 15 ↓ Siamang 0 1 ↓ Mitred Leaf Monkey 0 1 ↓ Sun Bear 3 1 1 1 6 23 ↓ Serow 1 1 1 1 1 1 6 0 ↑ Mouse Deer (sp.?) 1 3 4 0 ↑ Squirrel (sp.?) 1 1 2 0 ↑ Bird (sp?) 1 1 0 ↑ Total (2016) 0 1 0 0 1 3 1 0 0 14 0 0 5 18 1 0 5 0 3 10 0 1 63 - ↑ Total (2014) 0 0 1 2 1 9 4 9 2 2 3 1 1 0 3 2 6 2 1 1 2 1 - 53 Trend = ↑ ↓ ↓ = ↓ ↓ ↓ ↓ ↑ ↓ ↓ ↑ ↑ ↓ ↓ ↓ ↓ ↑ ↑ ↓ = ↑ -

Table 8. Vertebrate species encountered (both directly and indirectly) during the July-August 2016 line transect surveys [note that gibbons, siamang, pig-tailed macaques and mitred leaf monkeys were not directly observed along the transects in 2016 but are listed in the table as they were observed in 2014].

The most frequently contacted mammal, either directly and/or indirectly, was the wild boar (39.7% of all contacts), followed by muntjac (15.9% of all contacts), sun bear (9.5% of all contacts), serow (9.5% of all contacts), sambar dear (6.3% of all contacts), mouse deer (6.3% of all contacts), hornbill (4.8% of all contacts), pheasant (3.2% of all contacts), an unidentified squirrel species (3.2% of all contacts), and an unidentified bird species (1.6% of all contacts) (Table 9). The aforementioned vertebrates were found throughout the transect blocks, with the signs found as close as 325 m from NIL1 and 400 m from WJP1.

Species n-2014 n - 2016 % Total -2014 % Total -2016 n/km - 2014 n/km - 2016 Pheasant 1 2 1.9 3.2 0.05 0.09 Wild Boar 2 25 3.8 39.7 0.09 1.14 Pig-tailed Macaque 1 0 1.9 0 0.05 0 Agile Gibbon 1 0 1.9 0 0.05 0 Muntjac 7 10 13.2 15.9 0.32 0.45 Hornbill 1 3 1.9 4.8 0.05 0.14 Sambar Deer 15 4 28.3 6.3 0.68 0.18 Siamang 1 0 1.9 0 0.05 0 Mitred Leaf Monkey 1 0 1.9 0 0.05 0 Sun Bear 23 6 43.4 9.5 1.05 0.27 Serow 0 6 0 9.5 0 0.27 Mouse Deer (sp.?) 0 4 0 6.3 0 0.18 Squirrel (sp.?) 0 2 0 3.2 0 0.09 Bird (sp?) 0 1 0 1.6 0 0.05 Total 53 63 100 100 2.41 2.86 Table 9. Line transect survey summary statistics [2014 and 2016].

50 40 30 20

2014 10 2016 0 Encounter Rate - Encounters/km2 Walked

Vertebrate Species

Fig. 5. Vertebrate species direct and indirect encounter rates during the 2014 and 2016 systematic transect surveys. See Tables 7 and 8 above for the summary figures.

There was a number of differences in the species encountered during the systematic transect surveys conducted in 2014 and 2016 (Tables 8-9). The most commonly encountered sign in 2014 was that of the sun bear, which drastically declined in 2014. Conversely, no primates were encountered in the most recent 2016 survey period, whereas pig-tailed macaques, agile gibbons, siamangs, and mitred lead monkeys were all encountered during the 2014 survey period. Pheasant, wild boar, muntjac, and hornbill signs were more abundant during the 2016 surveys than that of 2014, whereas sambar deer and sun bear signs were less abundant.

3.3 Results – Orangutan Nests

No orangutan nests were encountered during the July-August 2016 surveys, including both systematic transect surveys and also recce walks between transects.

3.4 Results – Gibbon & Siamang Vocal Surveys

Following the fixed call count methodology established by Brockelman and Ali (1987), a series of 3 listening stations (all consisting of a set of three listening posts each) were monitored for a total of 12 surveys days from the hours of 0500-1200 during the months of July-August 2016. Data recorded for each listening post included the time of the group call, the species (i.e. agile gibbon or siamang), the compass bearing, and the estimated distance.

During the 2016 survey period, an average of 72 (range: 58-86 calls) agile gibbon vocalizations were heard per listening post [fig 6], while an average of 3 (range: 2-3 calls) siamang vocalizations were heard per listening post. Agile gibbon groups were heard to call throughout the three surveyed areas, with a number of calls being heard <100 m from NIL1 and WJP1. Siamangs were only heard calling at the edges of the SOL project area, with calls being heard only as close as 600 m from NIL1 and 300 m from WJP1. For the moment, it appears that agile gibbon and siamang calling locations have not yet been impacted by infrastructure development.

Fig. 6. Map of gibbon calling locations as mapped from the listening post surveys.

On average, gibbons started calling at 07:06 (range - 05:05-11:36), whereas siamangs started calling at 09:03 (range - 06:54-11:20). This is very similar to the averages reported for the 2014 field season, where gibbons were started calling at 06:44 (range - 05:20-11:44), and siamangs started calling at 09:01 (range - 07:18-10:39). When compared to call times from SOCP’s long-term monitoring station (Figure 7), vocalization times from both gibbons and siamangs from the SOL 2016 survey fell within the range of their SOCP counterparts.

In contrast to the 2014 survey period, where the survey closest to the SOL project area (array 2) displayed earlier calling times, little variation in agile gibbon calling times was observed between each array (Figure 7). Siamang calling times displayed an opposite pattern from that observed in the 2014 survey period, with the array closest to the SOL project area (array 2) showing later calling times than those arrays father from the SOL project area (arrays 1 and 3) (Figure 8). These differences may be associated with a number of non-mutually exclusive factors including climate (e.g. rain), sunrise during the period of survey, past and current disturbances (e.g. machinery), and/or audible distance.

12:00

10:48

9:36

8:24

Time of Call 7:12

6:00

4:48

Fig. 7. Results of agile gibbon and siamang call times. SOCP – Sumatran Orangutan Conservation Programme; SOL 2014 – Sarulla Operations Limited during the 2014 field season; SOL 2016 – Sarulla Operations Limited during the 2016 field season. The SOCP data is unpublished data from the SOCP long-term monitoring station in the West Batang Forest Block. For each boxplot, the middle line is the median value of the distribution, with the box representing the first (lower extreme) and third (upper extreme) quartile values (i.e., the interquartile range [IQR]), and the whiskers representing the minimum and maximum values.

12:00

10:48

9:36

8:24 Time of Call 7:12

6:00

4:48

Fig. 8. Results of agile gibbon call times, displayed by vocal survey array (i.e., a group of three listening posts). For each boxplot, the middle line is the median value of the distribution, with the box representing the first (lower extreme) and third (upper extreme) quartile values (i.e., the interquartile range [IQR]), and the whiskers representing the minimum and maximum values.

12:00

10:48

9:36

8:24 Time of Call 7:12

6:00

4:48

Fig. 9 Results of siamang call times, displayed by vocal survey array (i.e., a group of three listening posts). For each boxplot, the middle line is the median value of the distribution, with the box representing the first (lower extreme) and third (upper extreme) quartile values (i.e., the interquartile range [IQR]), and the whiskers representing the minimum and maximum values.

Via triangulation, group calls were marked on to a map in a GIS and the total number of agile gibbon and siamang groups heard was estimated for each listening location. We estimate

that the density of agile gibbons groups is 2.23 groups/km2 (range: 2.23-2.24 groups/km2) (Table 10). Using an average agile gibbon group size of 3.3 individuals (i.e., calculated from direct observation with 12 gibbon groups within the SOCP Batang Toru research area), it is estimated that there are an average of 7.37 individuals/km2 adjacent/within the SOL project area. Similarly, the average density of siamang groups is estimated at 0.7 groups/km2 (range: 0.68-0.75 groups/km2) (Table 11). Using an average siamang group size of 2.8 individuals (i.e., calculated from direct observation with 9 siamang groups within the SOCP Batang Toru research area), it is estimated that there are an average of 1.97 individuals/ km2 adjacent/within the SOL project area.

These density figures differ slightly from that presented previously for the 2014 field season. However, this is related to the fact that both agile gibbon and siamang vocalizations were consistently heard up to 1 km from all vocal arrays during the 2016 field survey, which increased the possible area of survey. Thus, the analyses of the 2014 survey underestimated the possible area surveyed. As such, we have recalculated the agile gibbon and siamang figures from the 2014 survey period and show that with a standard effective listening area of 3.14 km2 from each array, density estimates of both agile gibbons and siamangs do not significantly differ between the 2014 and 2016 survey periods (Tables 10 and 11).

Agile Gibbon - Recalculated 2014 Effective Listening Groups p(1) m (days) p(m) Groups/km2 Individuals/km2 Area Heard (n) (km2) Listening Array 1 3.14 7 0.71 4 0.99 2.25 7.41 Listening Array 2 3.14 7 0.71 4 0.99 2.25 7.41 Listening Array 3 3.14 7 0.79 4 1.00 2.23 7.37

Average 3.14 7 0.74 4 0.99 2.24 7.40 Agile Gibbon - 2016 Effective Listening Groups p(1) m (days) p(m) Groups/km2 Individuals/km2 Area Heard (n) (km2) Listening Array 1 3.14 7 0.75 4 1.00 2.24 7.39 Listening Array 2 3.14 7 0.82 4 1.00 2.23 7.36 Listening Array 3 3.14 7 0.89 4 1.00 2.23 7.36

Average 3.14 7 0.82 4 1.00 2.23 7.37

Table 10. Summary statistics for the agile gibbon vocal surveys.

Siamang - Recalculated 2014

Effective Listening Groups p(1) m (days) p(m) Groups/km2 Individuals/km2 Area Heard (n) (km2) Listening Array 1 3.14 1 0.25 4 0.68 0.47 1.30 Listening Array 2 3.14 2 0.25 4 0.68 0.93 2.61 Listening Array 3 3.14 1 0.50 4 0.94 0.34 0.95

Average 3.14 1 0.33 4 0.77 0.58 1.62 Siamang - 2016 Effective Listening Groups p(1) m (days) p(m) Groups/km2 Individuals/km2 Area Heard (n) (km2) Listening Array 1 3.14 2 0.38 4 0.85 0.75 2.09 Listening Array 2 3.14 2 0.50 4 0.94 0.68 1.90 Listening Array 3 3.14 2 0.50 4 0.94 0.68 1.90

Average 3.14 2 0.46 4 0.91 0.70 1.97

Table 11. Summary statistics for the siamang vocal surveys.

4. Discussion

4.1 Discussion wildlife monitoring through Primate Surveys and Camera Trapping Between 2013 and 2016 large scale development has taken place at the SOL Geothermal Area. Road development has taken place to the WJP area through forest habitat, with this road being very wide [estimated 20 meters rather than the proposed 6 meters in critical species habitat], where canopy connectivity would be maintained at least in several locations to allow for arboreal crossings of primates and other mammals. Significant daily traffic consisting of cars, motorbikes, trucks is using this road, causing high noise levels. All of the above impact wildlife both directly and indirectly.

Both species diversity and relative abundance of mammal species monitored through camera trapping, has declined significantly in the target area. For several of the most common species [muntjac, mousedeer, pig-tailed macaques] these declines are close to 70% compared to data from 2014. Using the Relative Abundance Index [RAI] can be useful for monitoring changes of the more 'common' species in the SOL area, to look at trends of these species’ abundance over time. It is important though to maintain the same study setup (i.e. number of cameras, sites where cameras are deployed, way in which cameras are deployed, and season in which sampling is taking place) throughout the entire study, in order for changes in the RAI to reflect changes in the focal species’ abundance, rather than changes in detection due to other factors.

Several of the Critically Endangered target species [orangutan, tiger] that were encountered [through their sign] in 2013-2014 have now not been recorded in 2016. Although in 2014 only a small number of orangutan nests (n=5) were encountered in the SOL area, orangutans are known to occur throughout the West Batang Toru forest block (Wich et al 2003, Nater et al. 2012, Wich et al. 2012). The overall orangutan population in the West Batang Toru forest block has been estimated at some 500-600 individuals (PHVA 2016), and is considered of highest conservation importance due to the fact that the population is considered viable and is now known to be genetically unique (Nater et al. 2011, 2012). SOL project staff that had worked in the area some years prior to the current construction activities reported a direct observation of an orangutan with offspring near the NIL 1 road (A. Samuel, pers.comm.).

One camera trap recording was made of the Critically Endangered pangolin, although its RAI also declined by 50%, although sample size for this species is [in general] limited. Another species, the marbled cat [VU], which was caught on camera trap in 2014 did not get photographed in 2016.

There was a significant decline in the number of Carnivore species recorded from the combined surveys [down from 8 species recorded in 2014 to only 2 species recorded in 2016]. The sun bear [VU] was recorded in 2016 through sign [claw marks] as well as a direct observation of a small cub at a den site. Large carnivores such as the sun bear, have well defined home ranges and cannot easily shift the boundaries of their home ranges as they

will 'encroach' into the home range of another sun bears whilst food resources are spatially limited [Fredriksson 2012].

Some other mobile species might have moved out of the SOL project area, away from the disturbance [heavy machinery, felling of trees with chainsaws in various locations, and significant human activity in the area including illegal logging and hunting], although resources are limited in the forest surrounding SOL and will already be occupied by those species. Other species might have been affected by the increase in human activity in the area through increased hunting pressures.

Of the original lists of target species put forward in the BAP, known to occur in the Batang Toru forest, several species have not been encountered in the SOL area. These include Asian tapir (VU) and binturong (VU). These species might need more extensive camera trapping efforts, but it might also be that they are 'extinct' in the SOL area. One observation was made of the nocturnal slow loris, Nycticebus coucang, in 2014 and was not recorded during the 2016 surveys, but this species would need night spotting surveys.

For transect surveys, 59% of the 22 transects surveyed showed a decline in animal signs, both direct and indirect. No animal signs were found on those transects closest to the WJP1 road area, indicating that infrastructure development may be impacting animal presence. Sambar deer and sun bears showed declines in presence comparing the 2014 and 2016 survey periods and no primate species were directly encountered while conducting the line transect surveys. Agile gibbons (n=1), siamangs (n=1), mitred leaf monkeys (n=1), and long- tailed macaques (n=2) were encountered during recce surveys and while moving between transects; however, all direct primate contacts were in areas >500 m away from any SOL infrastructure.

Agile gibbon groups still appear to saturate the entire SOL project area, with a number of group calls falling <100 m from the NIL1 and WJP1. Siamang groups were detected a bit farther from the SOL Project area, relative to agile gibbons, with estimated positions as close as 600 m from NIL1 and 300 m from WJP1. It is clear that agile gibbons still directly overlap the SOL Project area in many locations, with siamangs maintaining more peripheral positions relative to NIL1 and WJP1. For the moment, it does not appear that that the development of the SOL Project area is directly impacting the agile gibbon and siamangs groups surveyed in 2014 and 2016 (i.e., group calls were heard throughout the survey areas during both periods); however, given their ability to subsist on both preferred fruits and lower quality food items (e.g., leaves), impacts related to disturbance might only be able to detected with longer-term monitoring.

During a large socio-economic interview survey carried out by YEL between 2007-2009, which also covered some 49 villages/hamlets near the SOL project area (out of a total of 378 villages/hamlets surveyed around the Batang Toru forests), data was also collected on forest use, wildlife sightings, and hunting. Hunting levels have been found to be consistently high around the Batang Toru forest area (Wich et al. 2012). Respondents from nearly all villages/hamlets interviewed near the SOL project area reported to hunt (44 out of 49

villages/hamlets). This included hunting of orangutans, siamang and gibbons (3 out of 49 villages/hamlets), but also tapir, serow, sambar, sun bear and a variety of other species (YEL unpublished data). In 2015 an intensive Biodiversity Socialization Programme was carried out in all villages surrounding the SOL project area, which focused on increasing awareness and knowledge regarding the forest and its protected species. No people/dogs were caught on camera trap in 2016 [with 23 independent events of these in 2014 on camera traps], although evidence of illegal logging was encountered in the forest during the primate surveys.

5. REFERENCES:

Ancrenaz M., Hearn, A.J., Ross, J., Sollman, R., Wilting A. (2012). Handbook for wildlife monitoring using camera traps. BBEC, Sabah, Malaysia. Buckland ST, Plumptre AJ, Thomas L, Rexstad EA. 2010. Design and analysis of line transect surveys for primates. International Journal of Primatology, 31, 833-847.

Buij R, Wich SA, Lubis AH, Sterck EHM. 2002. Seasonal movements in the Sumatran orangutan (Pongo pygmaeus abelii) and consequences for conservation. Biological Conservation, 107, 83-87.

Buij R, Singleton I, Krakauer E, van Schaik CP. 2003. Rapid population assessment of orangutan density. Biological Conservation, 114, 103-113. Brockelman WY, Ali R. 1987. Methods of surveying and sampling forest primate populations. In: Marsh CW, Mittermeier RA, eds. Primate conservation in the tropical rain forest. New York, Alan R. Liss, p 23-62. Francis, C. M. 2008.A field guide to the mammals of South-East Asia.New Holland Publishers (UK) Ltd. Fredriksson, G.M., Nowak, M.G., and G. Usher (2014). Biodiversity Monitoring: Stage 1. Camera trapping and Wildlife Surveys. PanEco/Yayasan Ekosistem Lestari. Medan, Sumatra, Indonesia. Report requested by Sarulla Operations Ltd. Marshall A. R, Lovett JC, White PCL. 2008. Selection of line-transect for estimating the density of group-living animals: Lessons from the primates. American Journal of Primatology, 70, 1-11. Mott MacDonald (2013) Habitat Condition Assessment. Mott MacDonald, Singapore. O'Brien, T.G., Kinnaird, M.F., and Wibisono, H.T. 2003. Crouching tigers, hidden prey: Sumatran tiger and prey populations in a tropical forest landscape. Animal Conservation 6:131–139. Payne, J., C. M. Francis, K. Phillips. 1985. A Field Guide to the Mammals of Borneo. Sabah Society; World Wildlife Fund Malaysia, Malaysia. Phillips Q. and K. Phillips (2016), A field guide to the Mammals of Borneo and their Ecology. John Beaufory Publishing. Singleton, I. (2000) Ranging behaviour and seasonal movements of Sumatran orangutans (Pongo pygmaeus abelii) in swamp forests. Ph.D. Thesis, Durrell Institute of Conservation and Ecology, University of Kent, Canterbury, UK. Singleton I, van Schaik CP. 2001. Orangutan home range size and its determinants in a Sumatran swamp forest. International Journal of Primatology. 22(6):877-911.

SOL (2013) Volume II: Environmental, Social & Health Impact Assessment (ESIA) Addendum. Environmental Resources Management on behalf of Sarulla Operations Ltd.

Sopyan, E. (2008), Malayan Pangolin Manis javanica Trade in Sumatra, Indonesia. pp 134- 142 In Proceedings of the Workshop on Trade and Conservation of Pangolins Native to South and Southeast Asia Singapore Zoo. (eds. Pantel S. and S.Y. Chin).

Thomas L, Buckland ST, Rexstad EA, Laake JL, Strindberg S, Hedley SL, Bishop JRB, Margues TA, Burnham KP. 2010. Distance software: design and analysis of distance sampling surveys for estimating population size. Journal of Applied Ecology, 47, 5-14. van Schaik CP, Azwar, Priatna D. 1995. Population estimates and habitat preferences of orangutans based on line transects of nests. In: Nadler RD, Galdikas BMF, Sheeran LK, Rosen N, eds. The Neglected Ape. Plenum Press, New York, p. 129-147. van Schaik CP, Mirmanto E. 1985. Spatial variation in the structure and litterfall of a Sumatran rain forest. Biotropica, 17, 196-205. Wich S, Buij R, van Schaik CP. 2004. Determinants of orangutan density in the dryland forests of the Leuser Ecosystem. Primates, 45, 177-182. Wich, S. A., G. M. Fredriksson, G. Usher, H. H. Peters, D. Priatna, F. Basalamah, W. Susanto, and H. Kuhl. 2012. Hunting of Sumatran orang-utans and its importance in determining distribution and density. Biological Conservation 146:163-169. Wich, S. A., G. Usher, H. H. Peters, M. F. R. Khakim, M. G. Nowak, and G. M. Fredriksson. 2014. Preliminary data on the highland Sumatran Orangutans (Pongo abelii) of Batang Toru. Pages 265-284 in N. B. Grow, S. Gursky-Doyen, and A. Krzton, editors. High Altitude Primates. Springer, New York, Heidelberg, Dordrecht, London.

APPENDIX I Camera Trap Photographs of Interest _SOL 2016

The pangolin (Manis javanica), listed as Critically Endangered on the IUCN Red List and protected

in Indonesia, was once caught on camera trap during the 2016 camera trapping in SOL.

This female sambar deer (Cervus unicolor), listed as Vulnerable on the IUCN Red List and protected in Indonesia, was caught on camera trap during the 2016 camera trapping in SOL.

The pig-tailed macaque (Macaca nemestrina) listed as Vulnerable on the IUCN Red List but considered a 'pest' locally as it frequently enters gardens at the edge of the forest, was the most common primate species encountered. This social primate can occur in fairly large groups. Solitary adult males [picture above] are also encountered.

Appendix I - 31

The ferruginous partridge (Caloperdix oculea) is a shy, rarely observed bird of the forest floor.

The red junglefowl (Gallus gallus) ancestor of the domesticated chicken, is a very shy and rarely observed/photographed bird of the forest floor.

Appendix I - 32

Other Photographs of Interest _SOL 2016

During the transect surveys a sun bear cub (Helarctos malayanus) was encountered in a denning site. Sun bears are listed as Vulnerable on the IUCN Red List and protected under Indonesian law. Sun bear sign [clawmarks] were much less commonly encountered during the 2016 surveys than in 2014.

Appendix I - 33

The highly venomous pit viper Popea barati was encountered during the surveys. Though fairly common in Batang Toru, pit vipers rarely bother humans. Below an example of mammal sign encountered during the surveys [here a Sus scrofa scat].

Appendix I - 34

Development of WJP area, as seen in between the forest.

Below a picture of the 2016 survey team [Dewi, Parel and Pardi, Eka and Saniah].

Appendix I - 35