Eld’s deer is an endangered species confined to dry forests of Southeast Asia. With the largest populations occurring in Myanmar, the Smithsonian Institution engaged the government and local organizations in conservation efforts. Nationwide, there has been a decline in both dry forest and deer distribution since the 1980s, despite the deer having national and international protected status. The deer persist in two national protected areas, Chatthin and Shwesettaw Wildlife Sanctuaries. From 1995 to 2005, the Smithsonian intensively upgraded facilities, staff capacity, and local education at Chatthin Wildlife Sanctuary and documented a marked rise in deer numbers. Following a 2005 pullout by the Smithsonian and a steep decline in international funds, deer populations at this reserve declined rapidly to below 1995 levels. This period coincided with a near complete loss of suitable habitat outside the reserve, repeated changes of reserve leadership, and emphasis on sustainable livelihoods for the surrounding communities rather than enforcement. At Shwesettaw Wildlife Sanctuary, chronically low deer densities and lack of funding were dramatically reversed with the influx of military personnel around a new military base established inside the reserve. The military presence resulted in extensive forest clearing and was protested heavily by the Smithsonian and other international organizations as detrimental to the deer. However, the immediate response in Shwesettaw Wildlife Sanctuary has been an increase in deer numbers. Whether examining the “soft diplomacy” of public engagement and increased staff capacity or the “hard enforcement” by the military, both demonstrated that the species can respond rapidly to management actions; but the sustainability of these population responses without further government agency investment is uncertain.

KEY MESSAGE

Deer conservation in Myanmar depends on protecting habitat and animals. The extent to which government agencies in developing countries can effectively sustain conservation efforts without outside (NGO or military) support remains an open question.

INTRODUCTION

Eld’s deer (Rucervus eldii) is an endangered species that was historically common throughout Southeast (SE) Asia and eastern India [1, 2]. They are a large cervid (90–150 kg) that occupies open dipterocarp forests with abundant grass and forbs in the understory [2, 3]. These forests are found throughout SE Asia where there are monsoon rains and silty soils. In the twentieth century, much of this forest was converted to agriculture, primarily for rice production [4, 5]. Exacerbating the plight of the deer was the influx of modern weapons during the Second World War and the conflicts of the 1960s. There are three recognized subspecies of Eld’s deer [6], and by the 1990s, one subspecies (Rucervus eldii eldii) was confined to a single wetland in India, a second (Rucervus eldii siamensis) was extirpated from Thailand and Vietnam with scattered reports in Lao PDR and Cambodia, and the third (Rucervus eldii thamin) was extirpated from Thailand, with isolated populations still occurring in Myanmar [2]. The wild populations of R. eldii siamensis in eastern Cambodia are the largest for that subspecies. The largest remaining populations of R. eldii thamin are in Myanmar, and regional estimates estimate >90% of these animals reside along a crescent of forests between the Central Dry Zone and the Chin Hills to the west [7].

The Smithsonian Institution holds Eld’s deer at its facility in Front Royal, Virginia, and assists with ex situ conservation for this species by increasing knowledge on reproductive physiology and captive breeding [8]. In the early 1990s, Smithsonian was approached by the Myanmar government about assisting with a field survey and ecological study of wild populations of Eld’s deer. This outreach resulted in the first field team arriving in Myanmar in 1995, and the Smithsonian effort has continued to evolve to the present day with varying success. Published scientific papers on the deer’s social organization [9], ecology [3], and distribution [1, 7], as well as the history of wildlife management in the region [10] and the socioeconomics of local communities [11, 12] have greatly expanded our knowledge of the species and their management needs. A country-wide Conservation Plan was created in 2015 based on these advances in our understanding of the species [13]. The current paper is a case study highlighting the efforts of the Myanmar government and the Smithsonian Institution to save a species whose ecology has been well studied, but whose conservation has proven problematic.

CASE EXAMINATION

Ecology and Habitat Needs

There are three critical requirements for effective Eld’s deer conservation. The first is the presence of dry dipterocarp forest, which is a fire-adapted forest type that is dominated by a few tree species but has an abundant grass and forb groundcover [3, 14]. We have found no deer populations outside of this general forest type [1]. The second requirement is access to water during the extended dry season; this access can be either through fruits produced by the dominant canopy trees [3] or open waterholes [15]. The last critical factor, in our opinion, is protection from dogs and poaching, as the influx of either into a protected area has repeatedly resulted in severe reductions in deer numbers.

dry forest loss Tropical dry forests were historically widespread across much of mainland SE Asia, primarily in lowland areas with limited annual rainfall and a distinct, prolonged dry season [5]. Many of these forest areas are dominated by deciduous dipterocarp tree species (family Dipterocarpaceae) and have open canopies with grassy understories that historically supported high abundances of herbivores and grazers such as rhino, Asian elephant, gaur, banteng, and Eld’s deer [5, 13]. Due to their accessibility and proximity to areas of high human population densities, dry forests are among the most threatened ecosystems in SE Asia. In 2011, greater than 50% of the region’s remaining dry dipterocarp forest occurred in Myanmar [5]. Despite having a relatively large extent of forest remaining, just 2% of this total area is included in Myanmar’s protected area system [1, 5].

Several recent studies have reported that Myanmar’s poorly protected dry forests have experienced high rates of deforestation in recent decades. The country’s deciduous forests have experienced the second highest annual rate of forest loss out of five major forest types [16]. According to this study, estimated deciduous forest loss between 2001 and 2010 proceeded at an annual rate of 2.3% and only mangroves showed a faster rate of decline [16]. A comparison of national forest cover maps based on Landsat imagery between 1990 [17] and 2014 [18] also shows a high overall reduction in dry forest area, with 15% of forests being lost within an area generally approximating central Myanmar’s “dry zone” (Figure 1). Importantly, these high rates of dry forest loss provide a conservative estimate of habitat loss because many small, fragmented patches will not support wide-ranging species like Eld’s deer [19].

FIGURE 1.

Map of Myanmar showing 2014 forest cover and the extent of forest loss from 1990 to 2014. The highlighted area indicates the combined extent of the Irrawaddy dry deciduous and Irrawaddy moist deciduous ecoregions of Olson et al. [20] that approximate the area of Myanmar’s dry zone.

FIGURE 1.

Map of Myanmar showing 2014 forest cover and the extent of forest loss from 1990 to 2014. The highlighted area indicates the combined extent of the Irrawaddy dry deciduous and Irrawaddy moist deciduous ecoregions of Olson et al. [20] that approximate the area of Myanmar’s dry zone.

National Eld’s Deer Township Survey

In conjunction with the widespread loss of dry forest across Myanmar, there has been an alarming reduction in the Eld’s deer distribution [7]. A basic Eld’s deer survey has been repeatedly conducted across the central dry zone of Myanmar. This survey involves government or NGO staff traveling to each township that is reported to have Eld’s deer in the previous survey, as well as any additional townships with reports of deer. Expert interviews are then conducted with local villagers or forestry staff, and visits are made to specific forest patches reported to contain deer. This survey protocol was followed during six surveys (1960, 1992, 1997, 2002, 2007, and 2015). These surveys show that the distribution of Eld’s deer in Myanmar has declined from 60 townships in the 1960s to 20 townships in 2015 (Figure 2).

FIGURE 2.

The extent of Eld’s deer in Myanmar, as determined by national surveys conducted by government or NGO staff. The original 1960 extent is based on government records, and the townships with deer are indicated for each survey. No township regained deer following their extirpation during a previous survey.

FIGURE 2.

The extent of Eld’s deer in Myanmar, as determined by national surveys conducted by government or NGO staff. The original 1960 extent is based on government records, and the townships with deer are indicated for each survey. No township regained deer following their extirpation during a previous survey.

Protected Area Management

There are two wildlife sanctuaries that were found to include Eld’s deer during the National Survey: Chatthin Wildlife Sanctuary (CWS), which is 362 km2 in Kanbalu and Kawlin Townships, and Shwesettaw Wildlife Sanctuary (SWS), which is 500 km2 and spread across Minbu, Pwint Phyu, Ngape, and Saytotetaya Townships. Smithsonian Institution staff visited both sanctuaries in 1995 and established a research station in CWS. We have extensive data on CWS and sparse data on SWS, but informal inquiries gave us some knowledge of management and staffing situations.

chatthin wildlife sanctuary CWS is one of Myanmar’s oldest protected areas, first created as a fuel reserve in 1919 and converted to protected status in 1941 to conserve Eld’s deer [10]. CWS is under high pressure from extraction of nontimber forest products from over 25,000 people living within 10 km of CWS and three villages maintained within the protected area boundaries [10]. Land cover change analysis between 1973 and 2005 shows that 62% of the forest surrounding the sanctuary was lost (1.9% annually); however, inside CWS, it was only 16% (0.45% annually) (Figure 3). When Smithsonian scientists began work in the 1990s, there was infrequent patrolling or wildlife monitoring activities underway and most staff were based at the headquarters 35 km away. Smithsonian began training and provided funds for surveys, patrolling, and basic infrastructure including lodging, ranger stations, and boundary demarcation (approximately 15 training courses and $250,000 in aid). With the opportunities for supplemental income and training, the majority of field staff moved full time to CWS. Work expanded to include biodiversity surveys, outreach to communities, and environmental education efforts that increased conservation awareness of local people and raised the profile of CWS.

FIGURE 3.

Forest cover (green) and its loss (red) at Chatthin Wildlife Sanctuary. Panels A (1973–1992), B (1992–2005), and C (2005–2016) show significant periods, and Panel D (1973–2016) is a composite of all loss over the period. Note the creation of a large water impoundment to the southeast of the sanctuary in 2005.

FIGURE 3.

Forest cover (green) and its loss (red) at Chatthin Wildlife Sanctuary. Panels A (1973–1992), B (1992–2005), and C (2005–2016) show significant periods, and Panel D (1973–2016) is a composite of all loss over the period. Note the creation of a large water impoundment to the southeast of the sanctuary in 2005.

One central training activity was an annual survey of the deer. Fortunately one of the original staff trained at Chatthin (Daw Thin Thin Yu) was subsequently transferred to SWS and initiated the same survey technique at that site. The CWS staff had been annually using the King Method to survey the deer; a form of strip sampling with marked transects walked by staff who recorded the number of deer as well as their perpendicular distance to the transect [21]. We modified the survey to more easily adapt it to Distance Sampling software, expanded the survey effort, and standardized the protocol. Annual surveys were conducted in the last week of March when the deer were in larger groups and visibility was highest in the open forests. Unfortunately, the marked transects were concentrated in the best forest habitat; yet, the density estimates were extrapolated across the entire reserve. We consider the estimates a reliable index of deer numbers rather than an absolute number of deer contained in either site [9].

An analysis of forest cover change in the region shows that when Smithsonian efforts began in the early 1990s the deforestation rate inside CWS was 1.4% annually (prior to 1992), then dropped to 0.6% annually over the next decade, and then transitioned to a period of net forest gain between 2001 and 2005 (+0.4 annually). This forest gain inside CWS provides evidence that the outside support and increased capacity of protected area staff were able to reverse the trend in forest loss, which corresponded with an increase in the deer population (Figure 4). This trend reversed after 2005 (Figure 3); between 2005 and 2016, the annual deforestation rate within the sanctuary boundaries exceeded the pre-1992 rate and the reserve experienced a steady decline of the Eld’s deer population.

FIGURE 4.

Population estimates of Eld’s deer at Chatthin Wildlife Sanctuary (CWS, 1983–2016) and Shwesettaw Wildlife Sanctuary (SWS, 1989–2016). Estimates are based on the King Method, which is a derivation of a strip transect survey and suitable for examining population trend [21]. Significant local events are indicated along the top of the plot area; * indicates a new warden, ∆ indicates nearby dam construction at CWS, and the horizontal bar indicates the period of the Smithsonian support at CWS and the military base at SWS.

FIGURE 4.

Population estimates of Eld’s deer at Chatthin Wildlife Sanctuary (CWS, 1983–2016) and Shwesettaw Wildlife Sanctuary (SWS, 1989–2016). Estimates are based on the King Method, which is a derivation of a strip transect survey and suitable for examining population trend [21]. Significant local events are indicated along the top of the plot area; * indicates a new warden, ∆ indicates nearby dam construction at CWS, and the horizontal bar indicates the period of the Smithsonian support at CWS and the military base at SWS.

The Smithsonian’s main mission in CWS was increasing protected area staff capacity through capacity building. Training was reinforced through creation of specialized ecology teams that focused on important aspects of the sanctuary (i.e., Eld’s deer, birds, trees, insects, reptiles, nonforest products, and education). These actions created a cadre of >20 effective park staff and culminated with the head of the Eld’s deer team (U Myint Aung) being promoted to Park Warden in 1996. Many highly trained staff were promoted and transferred to other reserves (most other reserves lacked sufficient trained personnel). Due to increasing conflicts with the National and Provincial government over reserve management and finances, the Smithsonian withdrew from active support of CWS at the end of 2004. These conflicts involved both the increasing reluctance of the US Government to support a military government and the Myanmar Government’s desire to control more of the fund allocation. In addition, forest protection efforts progressed from issues with individual farmers or villages to conflicts around large-scale agricultural projects (i.e., dam construction, village relocation, and sugar cane production) where the primary antagonists were one branch of government against another. Partially in response to these conflicts, U Myint Aung resigned as Park Warden in 2005. By 2010, only two to three of the original staff trained by Smithsonian remained in the reserve.

shwesettaw wildlife sanctuary SWS contains a mix of forest types, including areas of dry dipterocarp forest, although it experiences less annual rainfall and has poorer forest quality than CWS. Its overall deer density was lower in the 1990s, possibly due to lower forest productivity. The staff turnover at the reserve mimicked that of CWS, with the arrival of new park wardens on average every 2 years over the past 2 decades. In 2010, the Myanmar military confiscated 88 km2 of the reserve for a troop base and cotton plantation, which involved extensive forest clearing and construction of a perimeter fence (Figure 5). Large areas inside the northern and eastern boundaries of the sanctuary have been converted to agriculture, including within the military area. Smaller-scale forest clearing has also been taking place along a number of stream valleys that penetrate the sanctuary from the southeast (Figure 5). SWS is currently 77.7% forested, with 21.8% deforested (much is cropland) and 0.4% in human settlement (Figure 5). Although SWS experienced no growth in the deer population during the decades prior to the military intervention, the deer population increased by at least 30% since they assumed management control of their area. Reserve staff do not manage the natural resources in the military area, but are allowed to annually survey the deer. The recovery of deer within the military area was a surprise to the conservation community, but reinforced our view of the changing situation at CWS, that effective law enforcement to counter poaching and forest encroachment is of primary importance to the successful conservation of Eld’s deer.

FIGURE 5.

Forest cover at Shwesettaw Wildlife Sanctuary in December 2016. We performed a basic land cover classification from a single Sentinel-2 satellite image acquired on December 27, 2016. We used a mean shift algorithm to perform image segmentation and manually labeled each segment as forest, deforested, settlement, or water. The location of a military base and an important pagoda are indicated.

FIGURE 5.

Forest cover at Shwesettaw Wildlife Sanctuary in December 2016. We performed a basic land cover classification from a single Sentinel-2 satellite image acquired on December 27, 2016. We used a mean shift algorithm to perform image segmentation and manually labeled each segment as forest, deforested, settlement, or water. The location of a military base and an important pagoda are indicated.

Synthesis Section

The 1990s brought the arrival of international conservation NGOs to Myanmar, including the Wildlife Conservation Society and the Smithsonian Institution. The government wildlife staff were generally ecstatic with the arrival of funds, expertise and training for critical species’ conservation. At CWS, the emphasis on Eld’s deer and dry forest ecology was welcome throughout the staff. Training and increased capacity resulted in rapid promotions, transfer of key staff, and increased conflict between CWS Warden (Myint Aung) and the Wildlife Department as the park tried to expand its conservation message outside the boundaries of the park with mixed success due to competing agricultural interests. These conflicts resulted in the Smithsonian removing itself from CWS in 2004 and Myint Aung resigning in 2005. Leadership is critically important in Myanmar society, and the chronic turnover of Park Wardens at both sanctuaries prevented any effective Eld’s deer management. At the country level, deer populations have declined throughout the modern period and, while CWS deer population declined after 2005, the SWS populations increased following military establishment in 2010. Regardless of the deer’s status in the individual reserves, country-wide distributions and suitable habitat have continued to decline. Our conclusions were that capacity building was not sustainable at the local level, without a plan at the national level to maintain such initiatives. We worry that any reduction in military protection of deer populations at SWS will not be offset by increased effort by the reserve staff.

The reengagement of international NGOs around Eld’s deer conservation has started with the creation of a country-wide Conservation Plan and the creation of new management plans for both SWS and CWS. These government-derived plans may generate commitment by Myanmar’s Ministry of Natural Resources and Environmental Conservation to train and retain staff at these critical reserves and bring attention to deer outside of the two reserves. If these plans rely on outside funds to accomplish goals, they may reflect a continuation of the status quo that has resulted in decades of rapid dry forest loss and shrinking deer distributions.

CASE STUDY QUESTIONS

  1. 1.

    Conservation happens at both the national and the local scale. Have the distribution changes at the national scale been reflected in population numbers within the two focal reserves?

  2. 2.

    What evidence is there that international support improved conservation outcomes over the short (<5 years) and long term (>10 years) at CWS and SWS?

  3. 3.

    What was the positive and negative role of military intervention in SWS? Are there similar examples in other systems? How can the future concerns of the conservation community around military control be alleviated?

  4. 4.

    Habitat loss and protection are critical factors for this species, but what is the relative importance of each factor?

  5. 5.

    At what level should international agencies focus their work in a developing country? At the reserve level to upgrade capacity and facilities or at the national level to shape policy?

  6. 6.

    Monitoring of Eld’s deer has been supported throughout the case study period; has it been linked to conservation actions?

AUTHOR CONTRIBUTIONS

Authors McShea, Aung, and Songer contributed much of the original work at Chatthin Wildlife Sanctuary; McShea and Aung conducted the National Surveys and helped produce the Conservation Plan. Connette produced the Forest Loss Analysis. All authors contributed to the writing of the manuscript.

The staff of CWS and SWS have been dedicated Eld’s deer conservation throughout the period. The staff of Friends of Wildlife have assisted the authors with much of the field work and data analysis. We thank Haydee Hernandez for analysis of land cover change in and around CWS between 2005 and 2016. Two anonymous reviewers provided valuable suggestions to improve the manuscript.

FUNDING

Over the years of this field work funding was provided by Smithsonian Institution, National Geographic Society, and Conservation Force.

COMPETING INTERESTS

The authors have declared that no competing interests exist.

DATA ACCESSIBILITY STATEMENT

All data from cited references or contained in figures provided.

REFERENCES

REFERENCES
1.
McShea WJ, Koy K, Clements T, Johnson A, Vongkhamheng C, Aung M.
Finding a needle in the haystack: regional analysis of suitable Eld’s deer (Cervus eldi) forest in Southeast Asia
.
Biol Conserv
.
2005
;
125
:
101
111
.
2.
Gray TNE, Brook SM, McShea WJ et al.
Rucervus eldii. The IUCN Red List of Threatened Species 2015:
2015
; e.T4265A22166803.
3.
McShea WJ, Aung M, Poszig D, Wemmer C, Monfort S.
Forage, habitat use and sexual segregation by a tropical deer species (Cervus eldi thamin) in a dipterocarp forest
.
J Mammal
.
2001
;
82
:
848
857
.
4.
Rundel PW, Boonpragob K.
Dry Forest Ecosystems of Thailand
. In: Bullock SH, Mooney HA, Medina E, editors.
Seasonally Dry Tropical Forests
.
Cambridge
:
Cambridge University Press
;
1995
. pp.
93
124
.
5.
Wohlfart C, Wegmann M, Leimgruber P.
Mapping threatened dry deciduous dipterocarp forest in South-east Asia for conservation management
.
Trop Conserv Sci
.
2014
;
7
:
597
613
.
6.
Balakrishnan CN, Monfort SL, Gaur A, Singh L, Sorenson MD.
Phylogeography and conservation genetics of Eld’s deer (Cervus eldi)
.
Mol Ecol
.
2003
;
12
:
1
10
.
7.
McShea WJ, Leimgruber P, Aung M, Monfort S, Wemmer C.
Range collapse of a tropical cervid (Cervus eldi) and determining the extent of remaining habitat
.
Anim Conserv
.
1999
;
2
:
173
183
.
8.
Monfort SL, Brown JL, Bush M et al.
Circannual inter-relationships among reproductive hormones, gross morphometry, behaviour, ejaculate characteristics and testicular histology in Eld’s deer stags (Cervus eldi thamin)
.
J Reprod Fertil
.
1993
;
98
:
471
480
.
9.
Aung M, McShea WJ, Htung S et al.
Ecology and social organization of a tropical deer (Cervus eldi thamin)
.
J Mammal
.
2001
;
82
:
836
847
.
10.
Aung M, Swe KK, Oo T et al.
The environmental history of Chatthin Wildlife Sanctuary, a protected area in Myanmar (Burma)
.
J Environ Manage
.
2004
;
72
(
4
):
205
216
.
11.
Allendorf TD, Aung M, Songer M.
Using residents’ perceptions to improve park-people relationships in Chatthin Wildlife Sanctuary, Myanmar
.
J Environ Manage
.
2012
;
99
:
36
43
.
12.
Allendorf T, Swe KK, Oo T et al.
Community attitudes toward three protected areas in Upper Myanmar (Burma)
.
Environ Conserv
.
2006
;
33
:
344
352
.
13.
MOECAF.
Conservation Strategy and Action Plan for Eld’s Deer in Myanmar
.
Nay Pyi Daw, Myanmar
:
Ministry of Environment
;
2014
.
14.
Koy K, McShea WJ, Leimgruber P, Haack BN, Aung M.
Percent canopy-cover – using Landsat imagery to delineate habitat for Myanmar’s endangered Eld’s deer (Cervus eldi)
.
Anim Conserv
.
2005
;
8
:
289
296
.
15.
Gray TNE, McShea WJ, Koehncke A, Sovanna P, Wright M.
Artificially deepening natural seasonal waterholes in eastern Cambodia: impact on water retention and use by globally threatened large ungulates and waterbirds
.
J Threatened Taxa
.
2015
;
7
:
7189
7195
.
16.
Wang C, Myint S.
Environmental concerns of deforestation in Myanmar 2001–2010
.
Remote Sens
.
2016
;
8
:
728
.
17.
Leimgruber P, Kelly DS, Steininger MK, Brunner J, Müller T, Songer M.
Forest cover change patterns in Myanmar (Burma) 1990–2000
.
Environ Conserv
.
2005
;
32
:
356
364
.
18.
Bhagwat T, Hess A, Horning N et al.
Losing a jewel – Rapid declines in Myanmar’s intact forests from 2002-2014
.
PLoS One
.
2017
;
12
(
5
):
e0176364
.
19.
Songer M, Aung M, Senior B, DeFries R, Leimgruber P.
Spatial and temporal deforestation dynamics in protected and unprotected dry forests: a case study from Myanmar (Burma)
.
Biodivers Conserv
.
2009
;
18
(
4
):
1001
1018
.
20.
Olson DM, Dinerstein E, Wikramanayake ED et al.
Terrestrial ecoregions of the world: a new map of life on earth: a new global map of terrestrial ecoregions provides an innovative tool for conserving biodiversity
.
Bioscience
.
2001
;
51
:
933
938
.
21.
Anderson DR, Laake JL, Crain BR, Burnham KP.
Guidelines for line transect sampling of biological populations
.
J Wildl Manage
.
1979
;
43
:
70
78
.