noig 37796 Recifi D i s c u s s i o n P a p e r s pac d an ai As ts Ea --tne mpolev M O N G O L I A Delaic Room to Roam? The Threat to Khulan (Wild Ass) So d from Human Intrusion an tne September 2006 mnoriv En M O N G O L I A Room to Roam? The Threat to Khulan (Wild Ass) from Human Intrusion September 2006 © 2006 e International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, NW Washington, DC 20433 USA September 2006 All rights reserved. is study was prepared by the Environment and Social Development Unit (EASES) of the East Asia and Pacific Region, and was funded by the World Bank's Netherlands-Mongolia Trust Fund for Environmental Reform. Environment and social development issues are an integral part of the development challenge in the East Asia and Pacific (EAP) Region. e World Bank's Environment and Social Development Strategy for the region provides the conceptual framework for setting priorities, strengthening the policy and institutional frameworks for sustainable development, and addressing key environmental and social development challenges through projects, programs, policy dialogue, non-lend- ing services, and partnerships. e EASES Discussion Paper series provides a forum for discussion on good practices and policy issues within the development community and with client countries. is publication is available online at www.worldbank.org/eapenvironment. Suggested citation: Kaczensky P., D.P. Sheehy, C. Walzer, D.E. Johnson, D. Lhkagvasuren and C.M. Sheehy. 2006. Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion. Mongolia Discussion Papers, East Asia and Pacific Environment and Social Development Department. Washington, D.C.: World Bank. Contact details for authors: Dr. Petra Kaczensky, University of Freiburg, Germany & University of Veterinary Medicine, Vienna, Petra. Kaczensky@wildlife.uni-freiburg.de Dr. Dennis P. Sheehy, dsheehy@eoni.com, www.ICAPSnet.com Dr. Chris Walzer, University of Veterinary Medicine, Vienna, www.wildvet.at www.fiwi.at Dr. Douglas E. Johnson, Oregon State University, Douglas.E.Johnson@oregonstate.edu D. Lkhagvasuren, National University of Mongolia, lkhagvasuren@nmum.edu.mn Cody Sheehy, Oregon State University, codysheehy@gmail.com, www.AutonomyProductions.com Cover design by the Word Express. is volume is a product of the staff of the International Bank for Reconstruction and Development / e World Bank. e findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Execu- tive Directors of e World Bank or the governments they represent. e World Bank does not guarantee the accuracy of the data included in this work. e boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of e World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. e material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. e International Bank for Reconstruction and Development / e World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978- 750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, e World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Contents Foreword vii Acronyms ix Acknowledgments xi Executive Summary xiii Introduction Herder and Livestock Situation 2 Khulan Situation 5 Study Area 6 Study Area Location 6 Climate Conditions 7 Vegetation 7 Water Occurrence 8 Chapter : Methods Study Limitations 11 Human and Livestock Trends 11 Herbivore Distribution 12 Transect Routes 12 Capture and Satellite Telemetry 12 Livestock Distribution 12 GIS Analysis 12 Impact of Different Habitat Parameters 12 Local Attitudes Toward Khulan 13 Questionnaire Survey 13 In-Depth Interviews 14 Media Presentations 14 iii Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Chapter : Results Human Population Trends 15 Livestock Trends 17 Khulan Capture 17 Collared Khulan Distribution 17 Sample Size 18 Distances Covered 20 Movement Barriers 20 Home Range Size 20 Khulan Seasonal Habitat Use 20 Khulan Distribution in the Southeast Gobi 20 Khulan Habitat Use 20 Vegetation Type 22 Khulan and Human/Livestock Interactions 25 Pasture and Water Use by Livestock 25 Water Use by Khulan 27 Khulan Spatial Relation to Human Intrusion 28 Impact of Well Construction and Rehabilitation 29 Human Attitudes Toward Khulan 30 Attitude Survey 30 Local Wisdom 32 Chapter : Discussion and Conclusions Spatial Dimension of Khulan Conservation 33 Habitat Fragmentation reats 34 Water Access 34 Poaching and Hunting for Sale 36 Khulan Monitoring 37 Conservation Management Planning 38 Lack of Institutional Capacity 38 Chapter : Recommendations Gobi-wide Initiatives 41 Develop a Gobi Region Infrastructure Strategy which Gives Attention to Conservation 41 Explore Potential for Sustainable Local Khulan/Wildlife Harvest 41 Develop Ecotourism Opportunities 42 Improve Khulan/Wildlife Access to Water 42 Enforce Existing Regulations 42 Improve Well Planning and Management 42 Enforce Hunting Regulations 42 Raise Awareness of Gobi Issues Relating to Large Herbivores 43 Database and Monitoring 43 Create a Habitat Database 43 Institutionalize a Herder/Livestock Monitoring System 43 Develop Khulan/Wildlife Monitoring Systems 43 Bibliography iv Contents Appendix A: Field Teams Involved in Data Collection in the SW Dornogovi Study Area Appendix B: Plant Community Composition in the Southeast Gobi Appendix C: Survey Instrument Used to Determine Local Attitudes Toward Khulan Appendix D: Interview Guidelines for the Six Semi-structured Interviews on Wildlife Values Appendix E: Daily Protocol of Herder in the SW Gobi (// to //) Appendix F: Selected Statements from Six In-depth Interviews About Human-Wildlife Relationships Figures Figure 1: Human Population Trends in Mongolia, 1950­2005 4 Figure 2: Livestock Numbers in Mongolia, 1961­2004 5 Figure 3: Study Area in the Core of the Khulan (Equus hemionus) Distribution Range in Mongolia 6 Figure 4: Average Annual Precipitation Based on Local Weather Stations in the Gobi 7 Figure 5: Vegetation Type Distribution in the Southeast Gobi 8 Figure 6: Population Development (since 1985) and Livestock Numbers (since 1970) in Southeast Gobi 15 Figure 7: Increase in Human Population in the Two Largest Aimag Centers in the Study Areas 16 Figure 8: Change in Livestock Composition in the 13 Soums of the Study Area in the Southeast Gobi, 1970­2004 16 Figure 9: Distances Traveled by Khulan Within 7-hour Time Steps 19 Figure 10: Movement Paths of Individual Khulan (July 2005 to February 2006) 19 Figure 11: Total Area Used by All Collared Khulan in the Southeast Gobi (July 2005 to February 2006) 21 Figure 12: Areas Covered by Individual Khulan over an Eight-Month Period in the Southeast Gobi Study Area 21 Figure 13: Seasonal Use of Habitat by Collared Khulan in the Southeast Gobi (July­August 2005) 22 Figure 14: Seasonal Use of Habitat by Collared Khulan in the Southeast Gobi (September­October 2005) 23 Figure 15: Seasonal Use of Habitat by Collared Khulan in the Southeast Gobi (November­December 2005) 23 Figure 16: Locations of Khulan Groups Encountered between June and October 2005 in the SW Dornogovi Study Area 24 Figure 17: Locations of Khulan Carcasses Encountered between June and October 2005 in the Southeast Gobi in Mongolia 24 Figure 18: Availability and Use of Different Plant Community Types by Khulan as Compared to Two Sets of Random Points 25 Figure 19: Location and Movements of GPS-equipped Herders in July and August 2005. 26 Figure 20: Distribution of Ulaankhuu's Small Livestock (16 July--27 August 2005). 26 Figure 21: Example of Additional Livestock Pasture Area Obtained by Well Rehabilitation in Dornogovi Aimag (5 km grazing buffer) 30 Figure 22: Local Attitudes Toward Khulan According to Perceived Population Trends 32 Figure 23: Khulan Locations and Movements Relative to Current and Planned Mining Operations 35 Tables Table 1: Summary of Environmental Constraints Affecting Large Herbivores in the Study Area 9 Table 2: Khulan Captured and Radio-Collared in the Southeast Gobi 17 v Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Table 3: Khulan Location Statistics, 7/7/2005­2/16/06 18 Table 4: Areas Covered by Individual Khulan During the Study Period 20 Table 5: Number of Khulan Counted During Five Trips in the Southeast Gobi between June and October 2005 22 Table 6: Distance to Open Water and Soum Centers. 29 Table 7: Water Source Availability and Grazeable Pasture 30 Table 8: Human Population in the Study Area and Sampled Population 31 vi Foreword M ongolia is facing a period of rapid habitat can be difficult and contentious, with percep- change, resulting from the need to tions of human-wildlife conflict often as important a strengthen its economy, provide factor in development decisions as the conflict itself. services to its population, and create e importance of rigorous, high-quality scientific new infrastructure. Meeting these needs will require research cannot be underestimated in informing sound planning and effective management if adverse development decisions, as the consequences can be environmental impacts--such as overgrazing and the far-reaching, and in some cases irreversible. is study degradation of pastures, pollution, and the fragmenta- makes an important contribution in advancing our tion and clearing of natural habitats for roads or understanding of ecosystem function and resource mining activities--are to be avoided. use in the Gobi region, and addresses key gaps in our knowledge of a threatened species, the khulan. e is research was initiated to improve our challenge is for the Government of Mongolia and understanding of human-wildlife conflict in the Gobi partners to take the results and recommendations of region, and provide recommendations on how negative this study and act appropriately to ensure that develop- impacts on wildlife from development projects could ment and livelihood improvement go together with the be avoided or mitigated. e report is the latest in a needs of Mongolia's wildlife. series of discussion papers from the Environment and Social Development Unit in the East Asia and Pacific Magda Lovei Region of the World Bank, and is based on research Environment Sector Manager supported by the Netherlands-Mongolia Trust Fund East Asia and Pacific Region for Environmental Reform (NEMO), a wide-reach- e World Bank ing initiative that has touched almost all aspects of environmental management in Mongolia in 2005­06. Arshad M. Sayed Country Manager Striking a balance between the needs of human Mongolia populations and the protection of wildlife and their e World Bank vii Acronyms ANOVA Analysis of Variance Dzud is the collective term for a range of winter weath- AWAC Asiatic Wild Ass Conference er-related conditions that prevent domestic animals EA Environmental Assessment from foraging in open grazing. Dzud is a fact of life for FAOSTAT Food and Agriculture Organization Corporate Database for Substantive Mongolian herders, who have developed strategies for Statistical Data coping with and adapting to their harsh environment. CITES Convention on the International Trade in Heavy accumulations of snow or ice crusts covering Endangered Species of Fauna and Flora pastures are the most common form of dzud (white CMS Convention on Migratory Species dzud). In situations where this precipitation is the GIS Geographic Information System GoM Government of Mongolia primary source of drinking water for livestock, the GPS Global Positioning System absence of snow or ice at winter pastures is also a ICAPS International Center for the Advancement type of dzud (black dzud). Since forage production of Pastoral Systems on natural pastures is almost entirely dependent on IUCN e World Conservation Union ( e rainfall during the short summer growing period, International Union for the Conservation of Nature and Natural Resources) dzud conditions are exacerbated by drought in the JICA Japan International Cooperation Agency preceding summer, so that there is less forage available MCP Minimum convex polygon for over-wintering animals. Historically, major dzud MNE Ministry of Nature and Environment have occurred roughly every seven years, but more NSO National Statistical Office of Mongolia experienced herders are not surprised when they occur sfu Sheep food unit SPA Strictly Protected Area in consecutive years, as has happened recently. SPSS Statistical Package for the Social Sciences Tg Mongolian Tugrik Ger simply means home in Mongolian. Gers are UNEP United Nations Environment Program traditional dwellings, well adapted to the nomadic USAID United States Agency for International life of Mongolia's herders, and consist of a wooden Development framework and a felt cover. e wooden framework Notes: includes the walls, long poles, a round smoke escape All dollars are U.S. dollars; all tons are metric tons. and its supports. Inside the ger, felt is laid either on a wooden floor or straight on the ground. e number Aimag (= province) is the largest sub-national admin- of walls and poles determines the size of the ger. Most istrative unit; below the aimag is the soum (= district), herders' gers have five walls, which make a living area which is divided into bag (= sub-district). In the of 16-18 m2. Each ger has a hearth, which is of great capital city, districts are called duureg and sub-districts symbolic and functional importance. In the winter khoroo. the hearth heats the ger and also serves as a stove for cooking. Gers weigh approximately 224 kg. It takes ix Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion about half an hour to collapse an average ger and Otor is a long-distance migration of herders and slightly longer to rebuild it (World Bank, 2004). livestock between pastures. x Acknowledgments T his report was prepared in partnership with Sheehy, along with a number of assistants. e project the Household Livelihood Support Project greatly benefited from the involvement and assistance Office by the International Center for the of people in the study areas, including a number of Advancement of Pastoral Systems (ICAPS) herders who recorded their movements with livestock as part of the Bank-supported Sustainable Livelihoods around water sources, and interviewees, who shared Project funded by the Government of the Netherlands their perceptions and knowledge on human-wildlife through the Netherlands-Mongolia Trust Fund for conflicts in the region. Enkhjargal Darambaza and Environmental Reform (NEMO). e work was man- Otgontsetseg Alexander helped with translations. aged by Tony Whitten, Senior Biodiversity Specialist, East Asia and Pacific Region of the World Bank. e publication benefited from the comments provided by the peer reviewers, Agi Kiss, Stephen We thank the researchers for their hard work in Ling, and Richard Reading. Bryony Morgan handled creating an excellent product. e lead researcher the complex final stages of report production. Bob was Dennis P. Sheehy. e project was carried out in Livernash provided editorial assistance. Desktop and collaboration with Petra Kaczensky, Chris Walzer, design of the publication was by the Word Express. Douglas E. Johnson, Davaa Lkhagvasuren, and Cody xi Executive Summary M ongolia, often called the "land without human and indigenous wildlife needs requires a fences," provides the last remaining greater understanding of ecosystem and resource use refuge for a number of migratory spe- in these areas. cies that require large areas of habitat. e Asiatic wild ass, or khulan, is one such species. A e overall objectives of the study were to (a) member of the horse family (Equidae), it has suffered improve our knowledge of large wild and domestic an immense reduction in range and numbers over the herbivore ecology, especially khulan ecology; (b) iden- last century. e population decline has left dwindling tify conflicts between human and khulan habitat use; and fragmented populations scattered across Eurasia. and (c) develop management recommendations that One large population remains in the sparsely would facilitate the protection of important khulan populated Gobi region of Mongolia. Protecting this habitat and mitigate khulan-human conflicts, enabling population is crucial for the long-term survival of the the long-term survival of this species in Mongolia. e species. study took place within Dornogovi aimag in southeast Gobi, which has the greatest concentration of khulan. is region also supports a growing human Several different approaches were used during the population, including a large number of livestock study. We determined their basic range, response to herders, who maintain a tenuous grip on survival human presence, and resource use by collaring seven after enduring the political and economic upheaval khulan with Global Positioning System (GPS) units wrought by the collapse of the socialist command and tracking their positions by satellite. ese data economy. With Mongolia's transition to a privatized were compared with similar data generated by GPS market economy, more people and exploitative units that were used to track the movement of herds economic activities--notably mining and road of sheep and goats in the study area. Examining the construction--could further impact environmental position points--in relation to vegetation community security and habitat needs of the khulan and associ- types, forage growth, water locations, and human ated wildlife in the region. infrastructure--provided many valuable insights into how wildlife and livestock interact with the changing For wildlife, people, and livestock alike, water environment. However, a truly analytical approach to is critical for survival in the Gobi. One method of the impact of well rehabilitation was inhibited by the improving livelihoods for rural people, especially herd- lack of a precise large-scale database that takes into ers, is to rehabilitate or construct new wells. is may account the temporal component of pasture and water increase the amount of pasture available for livestock, availability and tracks herder and livestock movements which in turn could adversely affect threatened large and numbers. In addition to tracking khulan and wild herbivores. Resolving such conflicts between livestock, surveys and interviews were conducted to xiii Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion shed light on resource conflicts between herders and decreasing. Poaching was identified as the main factor khulan, and the threat to khulan from illegal poaching. threatening khulan survival by local people. Illegal hunting seems to be undergoing a shift from local e study revealed that khulan are highly mobile resource use toward market hunting organized in the and cover wide areas. e individuals studied did capital city of Ulaanbaatar. However, without reliable make use of protected areas, but only as a small population estimates, the impact on khulan and other fraction of their total range. ey are highly sensitive wildlife populations in the Gobi is impossible to to fragmentation of the landscape--for example, quantify. the Trans-Mongolia railway acts as a barrier that the khulan are unable to cross, despite suitable habitat e results of the study demonstrate that address- on the eastern side of the railway. Khulan do not ing human-wildlife conflicts in the Gobi requires appear to select for a specific habitat type, but rather strategic planning on a landscape scale. Khulan are seem to seek areas of high biomass production. ese wide-ranging, and cover large areas in their search for high-quality foraging areas are usually spatially and areas of suitable vegetation. Development of fences temporally variable and result from differences in local and other linear or large-scale infrastructure would rainfall patterns. break the habitat into smaller, often non-contiguous patches, which would decrease the capacity of large Survival in the Gobi depends on access to water. wild herbivores to escape locally poor habitat condi- Khulan do not avoid wells or human settlements per tions. Rehabilitation of wells in the region is likely to se, but they can only drink from open water sources have considerable benefits for herders by increasing the or by digging potholes in dry river beds. Wells allow area of pasture available for grazing, and potentially human presence in areas where there is little surface by reducing the labor and time required for watering, water, but if there is an open water source nearby, depending on the type of well used. If well rehabilita- herders will generally utilize it in preference to wells. tion were to be done with mitigation measures in e presence of humans and larger livestock, such as place, it could also be beneficial to wildlife. For horses and camels, frighten khulan and prevent them example, by spreading livestock out over a larger area, from drinking at an open water source or pothole grazing impact and degradation of pasture could be at the same time, although they will approach a lessened--assuming that overall livestock numbers water source if un-herded small livestock are present. did not increase. e provision of wells could also Herders for the most part graze their livestock within provide an alternative to the use of open water sources five kilometers of a water source. If new water sources by livestock, potentially increasing access to water for are provided, the area of pasture they use will increase. khulan and other wildlife. Controls on the timing e pasture surrounding water points often becomes of well use could also allow maintenance of large overgrazed and degraded. Open water sources also movement corridors for migratory species. However, provide an obvious location for illegal poaching; signs if measures are not put into place, there is potential of poaching were seen at most natural water points for well rehabilitation schemes to be detrimental to visited during the study. wildlife populations. Interviews confirmed that traditionally, the Important steps that need to be taken to address khulan, like most other wildlife species, is an honored human-wildlife interactions in the Gobi areas of animal in Mongolian culture. Most local herders Mongolia include: felt the khulan should be protected from extinction. However, there is a widespread belief that the khulan Gobi-wide initiatives. A strategy should be prepared population is increasing, and consequently damage is for infrastructure development in the Gobi that occurring to range resources required by the herder to gives attention to conservation issues, in particular survive the harsh winters. is belief appears to have the barrier effect that fences, roads, and open-pit a significant influence upon local attitudes toward and strip mines can create. Developments should be khulan-- those people who believed that popula- preceded by an environmental assessment (EA), and tions were increasing felt less positive about khulan mitigation measures such as migration corridors and than those who believed populations were stable or wildlife-friendly designs should be used. e potential xiv Executive Summary for sustainable local khulan/wildlife harvests should valued. Both international and national media can be be explored as an incentive for wildlife conservation, used to inform the public about these issues; television tightly linked with sound monitoring and enforcement programs, and screenings of the khulan documentary schemes. Development of ecotourism opportunities film produced during the study should be effective. associated with khulan viewing should also be sup- Environmental education kits should also be devel- ported. Improvement of access to water for khulan and oped and disseminated. other wildlife is also important; one option could be to negotiate wildlife-only watering points, in return for Database and monitoring. To assess the impact of rehabilitation or development of wells. current and future infrastructure development, or changes in land use patterns on local pastoralists and Enforcement of existing regulations. Regulations wildlife, a common database on habitat, infrastruc- requiring prior assessment of development projects on ture, and land use needs to be assembled and made wildlife should be strictly enforced, with EAs carried freely accessible to personnel engaged in monitoring out. A wildlife-sensitive strategy for well rehabilita- and assessment. A herder and livestock monitoring tion should be developed and reinforced. Hunting system should also be established to facilitate efficient regulations must also be enforced, as illegal poaching EAs for rural development projects. For example, of khulan is becoming a key threat. Capacity building data should be collected and regularly updated on of the responsible staff and institutions will be key to seasonal herder camp locations and livestock numbers making effective enforcement a reality. and the composition of individual herds. Monitoring of large ungulate populations is required beyond the Raising awareness of wildlife issues in the Gobi. While local scale on which it is currently conducted. e most Mongolians continue to place a high value on only reliable method is by aerial survey. Data from wildlife, there is a lack of awareness of the ongoing these surveys should be transferred to the Mongolian decline in large herbivore populations--and also of Biodiversity Databank in the National University of the decline in traditional pastoralism, also highly Mongolia. xv Introduction W ater in the Gobi Desert region of and water; consequently, herders often take direct and Mongolia is a key resource for human indirect action against them (Kaczensky et al., 2006). occupation, livestock production, e increasing gap between rich and poor herders in and to sustain wildlife populations. Mongolia, and a succession of hard winters following roughout the Gobi region, most water for human summer drought, has apparently resulted in increased consumption and livestock is obtained from small, poaching pressure on khulan for meat. Khulan hand-drawn wells. Reportedly, approximately 30,000 are listed in Appendix I of the Convention on the small wells exist in the desert and desert steppe International Trade in Endangered Species of Fauna regions (Sheehy and Byambadorj, 2001). During the and Flora (CITES), and are regarded as threatened collective era, numerous mechanical wells were built with extinction by the World Conservation Union and maintained by the government, which greatly (IUCN). In 2002, they were added to Appendix II of expanded both the temporal and spatial scale of hu- the Convention on Migratory Species (CMS or Bonn man and livestock use. During the transition period Convention). Mongolia constitutes the main refuge for (> 1990), most of the mechanical wells (i.e., engine the species and consequently has a global responsibility and pump or Archimedes screw wells) have fallen to ensure conservation of the species. into disrepair (JICA, 2003 & 2004). Well construc- tion was considered a primary pastoral improvement e importance of the Gobi environment to the program facilitating pastoral livestock production conservation of khulan and other threatened wildlife in previously unused or only occasionally used Gobi and the future of pastoral livestock production is habitats. e present lack of deep, mechanical wells undeniable. With Mongolia's change to a free market has forced herders and livestock to abandon large economy, state support systems broke down and areas of Gobi pastureland. most herders became subsistence herders. At the same time regulatory mechanisms--including stocking e government, in part supported by the World rates, pasture use, and wildlife protection--broke Bank under the Sustainable Livelihoods Project, has down (World Bank, 2003). Without these regulatory been working toward rehabilitation or replacement of mechanisms, the anticipated commercialization of the these wells. An unanticipated impact is the possible agricultural sector could easily cause greater intrusion displacement of threatened wildlife, such as the Mon- of human activities, including more intensive livestock golian wild ass (Equus hemionus or khulan), which are production in the Gobi region. Development of other considered to be under great pressure from livestock sectors of the Mongolian economy, especially mining and hunting throughout their range (Wingard and and road construction, could further impact environ- Zahler, 2006). A common perception among herders mental security and habitat needs of the khulan and is that the khulan compete with livestock for food associated wildlife (Reading et al., 2006). Further 1 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion uncertainty is added by the potential impacts of global layer of control and management exerted by the herder climatic change, which is predicted to shift patterns on domesticated herbivores. Control and management and amounts of precipitation and evaporation, thereby of livestock are necessary to ensure herder livelihoods affecting the water resources and vegetation available in the Gobi region and elsewhere. ey are gener- to herders and wildlife alike (Batima, 2006). ally initiated in response to some current or future impending risk factor to mitigate potential negative e overall goal of this study was to (a) improve impacts on livestock directly and indirectly on the our knowledge of large wild and domestic herbivore livelihood of pastoralist's livelihood. Livestock are ecology, especially khulan ecology; (b) identify subject to many of the same environmental constraints conflicts between human and khulan habitat use; and limitations as wild herbivores. Until the mid-20th and (c) subsequently develop management recom- century, most livestock in Mongolia were derived from mendations that, if acted on, will facilitate protecting breeds native to Central Asia (Purev, 1990). important khulan habitat and mitigate khulan-human conflicts to guarantee the long-term survival of this Mongolia attempted to intensify and increase pro- species in Mongolia. ductivity of traditional pastoral livestock production in the 1940s (Bazargur et al., 1993; Mearns, 2004). Because the majority of the khulan population In the 1960s, state farms and the rural collectives is found in the southeast Gobi, this area was chosen were established in response to political and economic as the main study area. To determine the impact objectives of the socialist command economy. Both the of human intrusion on the khulan and other large state farm and the rural collective focused on expand- herbivores in the Gobi region, we: ing agricultural production by introducing centralized control, more intensive management, and more inputs Evaluated khulan behavior and movements to overcome environmental constraints affecting the relative to habitat parameters and artifacts of production system. State agricultural institutions human intrusion by collaring seven khulan with promoted grazing management strategies that focused satellite-linked Global Positioning System (GPS) on increasing livestock off-take through livestock collars. breed enhancement and systematic pasture use strate- Surveyed local people in the study area to gain gies. State dairy farms emphasized intensive, yearlong insight into their attitude toward the khulan and dairy production using Frisian and dual-purpose dairy associated wild herbivores. cattle and introduced livestock breeds. Evaluated livestock use of potential khulan habitat by providing selected livestock herders with e majority of Mongolian livestock production backpack GPS receivers to record spatial livestock during the socialist period continued to use traditional use of habitat around selected water sources. livestock and grazing management strategies, but with Evaluated forage habitat condition and use in the herder households organized into collectives. is area of Mongolia with highest known khulan was especially true of the Gobi region, where effective population (i.e., Southwest Dornogovi aimag). use of the desert environment and scarcity of natural water sources required larger spatial movements of Herder and Livestock Situation herders and livestock, and effectively limited increases in human population density. Except for occasional, For millennia, herders, livestock, and large wild more intensive "oasis agriculture," the Gobi region herbivores have used the Gobi environment to sustain was primarily an extensively managed livestock their livelihood. Prior to the 20th Century, traditional production system controlled by and operating within pastoral livestock production in the Gobi region the rural collective. e collective system attempted to approximated a naturally functioning wild herbivore increase livestock off-take by using livestock grazing system that employed seasonal movements, low inputs, management strategies that benefited from inputs into and primarily household consumption of livestock the livestock production system. Typical inputs in the products (Sheehy, 1996; Fernandez-Gimenez, 1999). Gobi region were livestock watering facilities, supple- e major difference between wild herbivore and mentary feed for winter season feeding of livestock, domestic livestock use of natural resources was the and specialized livestock breeding. 2 Introduction Opportunities to develop local sources of feed aimag, which has minimal surface water resources, supplements were severely limited by the Gobi region's were operating in 2003 (JICA, 2003 & 2004). high aridity. However, the region was the major Without access to wells, livestock herders have been recipient of hay and concentrate feed provided to the forced to withdraw from some of the collective Gobi livestock collectives through the State Emergency rangelands. is withdrawal, although possibly benefi- Fodder Fund. Provision of winter supplemental feed cial for large wild herbivores, causes concentration of and development of mechanized and shallow wells livestock around functioning wells and natural water (i.e., wells that lifted water through application of sources, which coupled with drought and increased the Archimedes screw principle) facilitated livestock livestock numbers, substantially increase the stocking production in the Gobi region and allowed for higher rate on rangeland close to water sources. Developing animal numbers. Under collective management, new wells or rehabilitating old wells is of considerable disturbance to habitat was mitigated by the extremely benefit to livestock herders in the aimag. high annual off-take of live animals and the provision of supplemental feed and nutrients into the production e costs associated with drilling and maintaining system. deep pipe wells--at an average cost of 6 million Tg/well--are too prohibitive for most individual During the collective period, well-drilling equip- households or groups to undertake. Households are ment and technology became available and subsequent most likely to select shallow, hand-dug and operated construction of wells greatly improved livestock access wells or shallow, wide-mouth pump wells if costs to pasturelands in the Gobi region. Water develop- have to be assumed. Construction of deep pipe wells ments included (a) hand-dug, unlined shallow wells and water distribution systems generally will require < 10 meters deep; (b) concrete-lined, wide-mouthed government or international development assistance shallow wells mechanically excavated < 20 meters because of the high cost associated with construction deep; (c) drilled, 16.4-cm pipe wells between 20 and and operation. 40 meters deep; and (d) drilled, 164-mm pipe wells 40 meters deep (JICA, 2003 & 2004). Access to Development of wells and water systems can have improved water sources also allowed herder livestock both positive and negative environmental impacts. to utilize rangelands that formerly were only occasion- Potential positive impacts, when overall livestock ally used following rain or snow events that provided numbers are kept at the present level, are: water for livestock. Reducing grazing pressure on local rangeland During the transition period (1990 to present), ecosystems by spreading livestock more evenly, both mitigating factors ceased. Livestock numbers which will improve ecological stability of range- increased in response to privatization and the shift to land communities. a market economy (World Bank, 2003; Reading et al., Improving the living conditions of herders and 2006; Johnson et al., 2006). Although longer-distance soum residents. seasonal migrations that mimic naturally functioning Potentially providing wildlife with access to water wild herbivore movements between seasonal pastures by reducing livestock concentrations at natural continue in the region, herders, for a number of water points. reasons, are becoming more sedentary (Fernandez- Potential negative impacts are: Gimenez, 1999; Fernandez-Gimenez and Batbuyan, An overall increase in livestock numbers, due to 2004). A primary reason for decreasing mobility access to more pastures. of herders is the critical need for access to water for Depletion of traditional winter pastures, where household and livestock use. the water requirement was traditionally derived from snow. Many of the wells developed during the collective Livestock-induced degradation of soils and era no longer function because pump and water vegetation around watering points. delivery systems have been destroyed or gravel filters Increased competition with wildlife for forage used in deep wells no longer function. For example, resources and habitat in areas formerly without only 1,000 of the 1,800 wells established in Dornogovi water. 3 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Degradation of native vegetation and soils in livestock production, numbers increased dramatically surrounding areas by livestock as they graze to during the 1990s to a peak of 33.5 million total and from watering points. animals in 2000. is increase was mainly due to Increased salinity of soils in irrigated cropping higher goat, cattle, and horse numbers, whereas camel areas if the water is used for irrigation. numbers have steadily decreased. Sheep numbers are also decreasing as herders replace sheep in the herd Water remains the critical need for both livestock with goats. Two consecutive harsh winters (Dzud and wild herbivores in the Gobi region, but other winters) in 2000/01 and 2001/02 resulted in high factors also affect sustainability of both domestic livestock mortalities, from which numbers have not livestock and wildlife. Mongolia's population more recovered (Figure 2). than tripled between 1950 and 2005, with the most rapid increase occurring during the socialist collective Loss of livelihood sustainability in rural areas forced period (Figure 1). In addition, the past 50 years have herders without family "safety nets" to abandon pastoral been characterized by a shift from a predominantly livestock production and move to soum and aimag rural to a predominantly urban population. A short centers that have rudimentary "social welfare" programs reversal in this trend was only observed in the early and some potential for alternative employment opportu- 1990s, when--due the breakdown of the socialist nities. Although subsistence-based livestock production system--many urban inhabitants lost their jobs and continues throughout Mongolia, and especially in the were forced to resume a rural herding existence. Since Gobi region, alternative livestock and crop production the mid-1990s, this pattern has again reversed and systems are being planned for, and initiated, both in the rural people have moved to the soum centers, aimag Gobi region and throughout Mongolia. centers, and the three major urban areas of Erdenet, Darkhan, and Ulaanbaatar. At the present time, a commercialized agricultural sector is gradually developing that could easily cause Livestock numbers remained rather constant from greater intrusion of human activities and more the 1960s to the late 1980s. With the breakdown of exploitative livestock production systems in the Gobi the collective system, subsequent privatization of the region. Development of other sectors of the Mongolian livestock herd, and the absorption of unemployed economy, especially mining and road construction, urban residents who had no other recourse than could further impact environmental security in general Figure 1: Human Population Trends in Mongolia, 1950­2005 Pre-collective Collective Market economy 3,000,000 Total Rural Urban 2,500,000 2,000,000 tion popula 1,500,000 Human 1,000,000 500,000 0 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Year (Source: FAOSTAT, 2005) 4 Introduction Figure 2: Livestock Numbers in Mongolia, 1961­2004 Pre-collective Collective Market economy 16,000,000 80,000,000 14,000,000 70,000,000 12,000,000 60,000,000 all livestock in sfu units 10,000,000 50,000,000 numbers 8,000,000 Sheep 40,000,000 units ock Goats sfu est Horses Dzud Liv 6,000,000 Cattle 30,000,000 Camels 2000/01 2001/02 4,000,000 20,000,000 2,000,000 10,000,000 0 0 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 Year (Source: FAOSTAT, 2005) and habitat needs of khulan and associated wildlife in eastwards. In the 1970s and 1980s, the south Gobi the Gobi environment in particular. seem to have been the key khulan area; by the 1990s, the southeast Gobi became the key distribution range, which still appears to be the case today (Reading et al., Khulan Situation 2001; MNE, 2003). In the IUCN Equid Action Plan, the status of Equus In Mongolia, the khulan has received full protec- hemionus1 is qualified as "insufficiently known" tion since 1953. It is listed in CITES Appendix I, and and the species is listed as vulnerable (Feh et al., in 2002 was included in Appendix II of the Conven- 2002). e distribution range of all the Asiatic wild tion of Migratory Species (Bonn Convention, CMS ass subspecies and/or populations has undergone a 2002). However, due to human population growth dramatic decline and the remnant populations are in conjunction with severe winters in the past years predominantly small. Probably no more than 5,000 (United Nations Disaster Management Team, 2000), individuals remain outside of Mongolia and China, the occurrences of herder-khulan conflicts appear to be and therefore Mongolia is a globally important on the increase. stronghold of the Asiatic wild ass (Feh et al., 2002). e most recent khulan population estimates Historical evidence from Mongolia suggests for Mongolia date from 1994­97 and 2003. e that a dramatic decrease in the distribution range 1994­97 survey, which used aerial surveys and a occurred during the 19th century, restricting khulan distribution to the Gobi areas in the south, southwest, 1 e systematic status of the different Asiatic wild ass populations and the depression of the great lakes in the Govi-Altai still remains controversial. e IUCN Equid Action Plan assigns and Khovd aimag (Bannikov, 1954, in Zevegmid and the Persian onagar (Equus hemionus onager), the Indian khur (E. h. khur), the Transcaspian khulan (E. h. khulan) and the Mongolian Dawaa, 1973). By the 1970s, khulan had disappeared wild ass (E. h. hemionus (North Mongolian dziggetai) and E. h. luteus from the depression of the great lakes and were (Gobi khulan) subspecies status, whereas the Chinese kiang (Equus restricted to the main Gobi region (Zevegmid and kiang) is considered a different species. However, recent molecular and morphological analysis does not support the status of the Dawaa, 1973). Simultaneously, the population either kiang as a separate species, but rather suggests subspecies status (A. expanded or shifted its main distribution area further Schreiber, University of Heidelberg, Germany, pers. comm. 2004). 5 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion distance sampling approach (Buckland et al., 2001), which were only of limited interest to pastoral people estimated 33,000 to 63,000 khulan populated a and their livestock, have acted as important refuge continuous distribution range encompassing southern areas for the species. Providing or repairing wells will Mongolia (Reading et al., 2001). e survey in 2003, bring herders and their livestock into areas that are which used ground surveys without standardized presently un-used. Political and societal changes have protocols, estimated 19,652 ± 600 animals over an disrupted traditional land use patterns, weakened law area of 177,563 km2 in southern Mongolia (MNE, enforcement, and changed attitudes toward the use of 2003). Although the two surveys are not comparable, natural resources (Reading et al., 2006)--for example, additional evidence, namely the abundance of khulan making wildlife an "open access" resource (Pratt et carcasses (Stubbe et al., 2005; P. Kaczensky, unpubl. al., 2004). e re-immigration of people and their data) and illegal hunting (Wingard and Zahler, 2006), livestock will potentially result in increased wildlife- suggest that the population is decreasing. human interactions and may well threaten the survival of rare wildlife species in the Gobi region. Competition with domestic livestock is believed to have pushed other wild equid species of central Asia--such as Przewalski's horse (Equus Study Area przewalskii)--into marginal desert-steppe habitats. Subsequently, hunting, capture of foals for zoological collections, military activities, and very harsh winters Study Area Location in 1945, 1948, and 1956 resulted in the eradication e study area covered parts of three Gobi provinces of the species in the wild by the 1960s (Bouman and (aimags): Omnogovi and Dornogovi, which are in the Bouman, 1994; Van Dierendonck and Wallies de core of the Mongolian khulan distribution range, and Vries, 1996). e khulan seems better adapted to the the southeast corner of Dundgovi. e area is believed harsh conditions of the desert-steppe and steppe areas to contain the majority of the Mongolian khulan of the Gobi (Kaczensky et al., 2006). ese areas, population (Reading et al., 2001; MNE, 2003). e Figure 3: Study Area in the Core of the Khulan (Equus hemionus) Distribution Range in Mongolia The study area encompasses 13 districts (soums) of the two southeast Gobi provinces (aimags) of Omnogovi and Dornogovi. 6 Introduction study area covered roughly 155,779 km² in 13 districts the amount and timing of precipitation, temperature, (soums; Figure 3). soils, as well as livestock grazing intensity. Other large ungulates in the study area include Monsoon winds are a major climatic influence Mongolian gazelle (Procapra gutturosa), goitered or in the region (Nuttonson, 1947). During May and black-tailed gazelle (Gazella subgutturosa), argali June, winds shift from the Pacific Ocean to Central wild sheep (Ovis ammon) and ibex (Capra sibirica). Asia and bring precipitation in the form of cyclonic Mammalian carnivores are the wolf (Canis lupus), lynx and convection storms. Approximately two-thirds of (Lynx lynx), Pallas' cat (Otocolobus manul), red fox total precipitation and almost all growth of vegeta- (Vulpes vulpes), corsac fox (Vulpes corsac), and Eurasian tion occur during the May to September period. In badgers (Meles meles). September, winds shift to the northeast and bring cold temperatures, dust, and little or no moisture from the Human population in the primary study area extremely cold regions of Siberia. In general, the cold numbers around 62,000 people (average density: 2.5 and dry winter; the dry, cold, and windy spring; and inhabitants / km²). However, about half of the popula- the occurrence of most precipitation between mid-June tion lives in the two aimag centers of Dalanzadgad and the end of August limit the potential productivity and Sainshand. e study area is also a major base for of Mongolian land resources. pastoral livestock production. Major large domestic herbivores are sheep, goats, horses, cattle, and camels. In the study area, the climate shows a gradient Environmental conditions relevant for large herbivores from west to east and north to south. Average annual in the study area are summarized in Table 1. precipitation is 150 mm in the southeast part of the study area, but considerably less (100 mm) toward the north and west (Figure 4). Average annual Climate Conditions temperatures vary from 6°C in the southwest to 2°C in e climate of Mongolia is cold and arid. Steppe the northeast corner of the study area. Average annual rangeland has been dry land since the Paleozoic (540 snow cover lasts for 50 days in the southwest and up to million years ago), arid since the Cretaceous Age (100 100 days in the west (Atlas of Mongolia, 2004). million years ago), and increasingly arid since the Neocene. Continental climatic conditions have favored Evidence exists that climate change is already development of extensive grass and shrub steppe occurring in Mongolia, with an increase in mean rangeland. Important environmental factors influenc- annual temperature of 1.8°C over the last 60 years ing stability and resilience of grazing ecosystems are (Batima, 2006). Summers are predicted to be hot Figure 4: Average Annual Precipitation Based on Local Weather Stations in the Gobi (courtesy H. Von Wehrden, based on data obtained from Hijmans et al., 2005). 7 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Figure 5: Vegetation Type Distribution in the Southeast Gobi (Original scale 1:500,000, Source: Russian-Mongolian Complex Ecological Survey 1999) and dry, and winters milder with increased snowfall. Gobi region. Extreme and unpredictable fluctuation ese predictions encompass a broad range of regional in forage quantity and quality between seasons, variation, and impacts are likely to be much less years, and places (often on a very local scale) is a pronounced in the arid Gobi regions than in moun- characteristic of forage growth. Although growth of tainous and steppe regions of Mongolia. e projected forage standing crops may be initiated for some plant increase in evaporation is higher than the projected species by April, both quantity and quality of forage increase in precipitation, and therefore water-stress is is insufficient to meet large herbivore needs until the likely to increase. Extreme climatic conditions appear rise in ambient temperature and precipitation creates to have become more frequent in recent years, with conditions suitable for a high rate of forage growth in several pronounced episodes of drought in the last late May and early June. decade. Over the past 60 years, the number of "heat wave" days, with abnormally high temperatures, has In the study area, vegetation was characterized by increased in the Gobi region by 6-8 days annually a mix of semi-desert steppe, desert steppe, and true (Batima, 2006). desert components (Figure 5, Appendix B). Relief energy is low and only a few small mountain ranges are found. Elevations in the study area range from Vegetation 750 to 1,900 meters. Rangeland is primarily arid or e composition of rangeland vegetation and the semi-arid steppe and accordingly is subject to drought, temporal and spatial distribution of annual forage wind, variable temperatures, and a short growing standing crops are important factors regulating season. To live here, both humans and animals must livestock production and wildlife distribution in the adapt to the harsh environment (Table 1). 8 Introduction Table 1: Summary of Environmental Constraints Affecting Large Herbivores in the Study Area Eco-Region Environmental factors Herbivore risk factors Gobi region (Aimags of Khovd, - Elevation between 700 and 3200 m; - Lack of water for grazing animals is a major factor limiting livestock Govi-Altai, Bayankhongor, - Mean annual temperature between distribution and production in the Gobi region. Ovorkhangai, Dundgovi, 0.0°C and >2.5°C with a low mean - Environmental in uences such as drought and/or dzud in combination Omnogovi, Gobi-Sumber, January temperature of -20°C and a with overstocking can change "steady state" conditions very quickly and and Dornogovi). Semi-arid and high mean July temperature of 23°C; decimate large herbivore populations. arid southern section of Mongolia - Between 90 to > 130 frost free days, - Di culty of access to major markets in the western Gobi region aimags is - Variable precipitation of 100 mm. a major constraint to sustainable livelihoods, but has safeguarded the area Wind speed between 2 and 8 m/sec and its wildlife against high human intrusion occurs. - Over-browsing of shrubs that dominate vegetation communities cannot be easily mitigated. - Hay, fodder, and grain production potential is low except in a few oases and developed irrigated areas. - Arid ecosystems are prone to both environmental and anthropomorphic deserti cation. Water Occurrence Unfortunately, well locations are generally known only on a local level; a Gobi-wide database is not In this environment, water for humans, livestock and available. Mapping of wells by the Japan International wildlife is a critical resource in the Gobi region. e Cooperation Agency (JICA) and the Ministry of availability of surface water--in the form of springs, Nature and Environment (MNE) in 2003 showed that lakes, seeps, or creeks--is highly variable and depends in Dornogovi aimag alone there are over 1,200 deep, on local rainfall patterns. e groundwater table is mechanical wells, with an averages of 1 well/92 km² rather high; in dry riverbeds, groundwater is often (JICA, 2003 & 2004). accessible at 0.5 meters below the surface. 9 1. Methods M ethods employed during the study e lack of fine-scaled habitat data (i.e., resolution reflected the need to collect all infor- of the plant community layer was at a scale of mation during one short field season 1:500,000, the resolution of the well layer was at a in the summer and early fall of 2005. scale of 1:2,000,000). Because of this limitation, data collection methods e high temporal variation in this ecosystem emphasized field observations, discussions with herders (i.e., availability of open water and pastures and local government officials, and collaboration with changes with rainfall patterns). other projects with an already established presence in Insufficient information on use of habitat by the region. e short duration of the study and the livestock and the degree of conflict caused by need to use available time to gain familiarity with livestock and khulan co-use of habitat . the study area effectively prohibited development of a A very small sample size of only three herders to more comprehensive study interaction with Mongolian assess livestock habitat use. government ministries and institutions responsible for wildlife, livestock, and land resources. Two teams were formed to implement the study. Team 1 focused on khulan behavioral ecology and the attitude of herders toward khulan. Most of the Study Limitations fieldwork by this team occurred in June and July 2005. Team 2 focused on habitat and nutritional aspects; is study on khulan habitat use and reaction to fieldwork occurred from June through October 2005 human intrusion should be treated as preliminary (Appendix A). evidence on what could and should be done in the future. At the present stage, in-depth analysis of habitat use was hindered by: Human and Livestock Trends A short study period of only eight months. Soum-level data on human population trends and A khulan sample population of only seven livestock numbers was obtained from the National animals, of which one produced data for only two Statistical Office of Mongolia (NSO) in Ulaanbaatar weeks. for 5-year intervals. Annually collected statistics on e unavailability of habitat parameters at human and animal population trends were available by landscape scale (i.e., location of wells, location of aimag and soum from the NSO. herder camps, etc.). 11 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Herbivore Distribution of 15 minutes. Only single animals or small groups without foals or pregnant mares were approached. Transect Routes All animals were collared with GPS / ARGOS2 Between May 1 and October 15, 2005, the teams collars (TGW-3580, Telonics, Mesa, USA). Collars made five field investigation trips to the southeast were programmed to attempt 1 GPS location every 7 Gobi (Dornogovi and Omnogovi aimags) study hours and transmit GPS data via ARGOS uplink every area. During each field trip, observations were made other day. With this schedule, collars were expected to of the number and location of (a) live and dead last for 19 months. For animal welfare reasons and khulan; (b) artifacts of human intrusion (wells, water to allow collar retrieval, all collars were equipped with sources, herder camps, roads, etc.); and (c) livestock pre-programmed drop-off devices (CR-2A, Telonis, encountered. Observations were mapped by recording Mesa, USA), which will release the collars on October GPS locations to allow overlay of locations with other 20, 2006. habitat parameters. No efforts were made to avoid double counts and no special effort was made to locate Livestock Distribution khulan carcasses. e transect route was chosen based on information concerning khulan presence obtained Backpack GPS units manufactured by Oregon State from the park administration at Dalanzadgad (Mr. University in the United States were used to sample Ravjev and Mrs. Dolgorma), local rangers (Mr. distribution of herder livestock in the study area. e Terbish), local soum administration staff, and local GPS units, which record a position every 15 minutes, herders/villagers. Although observational data was were provided to eight herding households in the insufficient for statistical analysis or to derive wildlife area of concentrated khulan use in the southwest population numbers, it does provide information on Dornogovi study area. Herders were asked to carry the presence or absence of large herbivores and human GPS units with them anytime they grazed and/or intrusion in the areas studied during the summer and moved livestock to and from camps/wells. Availability early fall of 2005. of collars restricted placement of GPS units to only eight herding households. Recording of proxy livestock GPS locations occurred over a period of 14 to 21 days Capture and Satellite Telemetry and was limited by battery function. Herders selected Capture attempts for khulan were concentrated at to carry GPS units were also requested to record or near water points where khulan had previously observations of khulan they encountered while herding been observed. Hidden behind bushes or rocks, we livestock on their pastureland. ambushed animals at the water with a CO2-powered dart gun (Daninject JM). All animals were darted at a range of 30 to 50 m with a single 3 ml dart (Telinject) GIS Analysis containing a combination of 4.4 mg Ethorphine (M99, C-Vet Veterinary Products, Lancs, UK), 10 mg Impact of Different Habitat Parameters Detomedine-HCl (Domosedan, Orion Corp. Farmos Finland) and 10 mg Buthorphanol (Torbugesic, Fort Digital information--on water points, aimag Dodge Animal Health, Iowa, USA). Anaesthesia was and soum boundaries, and soum centers--was reversed with an intravenous combination of 200 mg obtained from the National University of Mongolia. Naltrexone (Trexonil Wildlife Laboratories Inc., Fort e railroad layer was downloaded from an open Collins, Colorado, USA) and 20 mg Antipamazole access source (http://geoengine.nima.mil/geospatial/ (Antisedan, Orion Corp. Farmos Finland; for more details see Walzer and Kaczensky, 2004). 2 e ARGOS program is administered under a joint agreement between the National Oceanic and Atmospheric Administration An alternative capture method involved chasing (NOAA) and the French space agency, Centre National d'Etudes and darting animals from a 4x4 jeep (Russian UAZ) Spatiales. e system consists of in-situ data collection platforms equipped with sensors and transmitters, and the ARGOS on flat terrain. e chase time was closely monitored instrument located aboard the NOAA Polar-orbiting Operational and did not exceed our pre-established cut-off time Environmental Satellites. 12 Methods SW_TOOLS/NIMAMUSE/webinter/rast_roam.html). relative to a hypothetical case allowing a legal harvest; Because there is no central database available for the and (h) the respondent's own experience with khulan location of wells, all active and broken wells from (Appendix C). JICA maps were digitized at a scale of 1:2,000,000. Vegetation types from the Russian-Mongolian Most questions were asked using a 5-point Likert Complex Ecological Survey 1999 maps were digitized scale, allowing answers to range from (1) strongly at a scale of 1:500,000 (Appendix B). disagree, (2) disagree, (3) neutral, (4) agree, and (5) strongly agree. e respondent was given three GPS positions of collared khulan were analyzed in response options to knowledge questions: (1) correct, GIS (ArcView 3.1, Environmental Systems Research (2) not correct, or (3) don't know. In addition, the Institute, Inc., Redlands, California, USA) with the respondents were asked open questions about the Animal Movement (Hooge and Eichenlaub, 1997) estimated khulan population in Mongolia, the average and Home Range extensions (Rodgers and Carr, group size of khulan encountered by the respondent, 2001). e minimum convex polygon (MCP) of all and the key management issues of khulan in Mongolia. khulan locations (100 percent MCP) was used as "available habitat." To check for possible preferences Questionnaires were pre-tested/discussed with or avoidances of certain habitat structures, two sets of ~10 people in Ulaanbaatar prior to printing the final 880 random points were generated--one with normal version. Minor changes were applied after interviewing distribution and one with a uniform distribution--in the first 20 people in Dalanzadgad. the available habitat. All locations (i.e., the khulan data set and the random points) were overlaid with e survey targeted all local people 16 years. the habitat layer, and respective values were assigned Households were randomly selected; only one family to each location (e.g. nearest distance to water, wells, member was interviewed per household. soum centers, plant community type, elevation, slope, aspect etc.). All data analysis was done with SPSS version 10.0. For statistical tests we used p < 0.05 for significance Due to the short study duration and the difference level. To minimize random errors ( Zeller and in sampling intensity between individual khulan, Carmines 1980), the seven attitude questions were we did not attempt to identify outliers or calculate constructed in a way that enabled grouping of similar seasonal ranges. questions and calculation of a score. Answers were coded in a way that positive feelings were expressed by All statistical analysis was done in SPSS 10.0.7 high values on the Likert scale and negative feelings (Statistical Package for the Social Sciences; SPSS Inc., by low values. e seven questions to construct the Chicago, Illinois, USA). attitude score where: (1) Is having khulan in Mongolia important? Local Attitudes Toward Khulan (2) Do you like khulan? (3) Khulan are a necessary part of nature in Mongo- lia. Questionnaire Survey (4) It is important to maintain khulan for our Information on local attitudes toward khulan was children (future generations). obtained from a questionnaire that consisted of (5) Because large khulan populations exist elsewhere, 90 questions organized in eight sections, in which it is not necessary to have khulan in Mongolia. locals were asked to provide information on (a) (6) It is important for me that khulan exist in basic socioeconomic facts about the respondent; (b) Mongolia. their herding background; (c) attitude/value-related (7) I'm proud that Mongolia has khulan. questions about khulan; (d) knowledge / beliefs about khulan; (e) attitudes toward the present management e reliability estimate for the score was high of khulan; (f) the acceptance of possible management (Chrombach's Z=0.78) and supported our grouping scenarios for khulan; (g) support for khulan protection (Zeller and Carmines 1980). Score distribution did 13 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion not differ from a normal distribution (Kolmogorov- used. All interviews were conducted by H. Otgonbayar Smirnov test: Z=1.05, p=0.22) and therefore we used using the interview guide in Appendix D. All parametric tests (T-test or ANOVAs with post-hoc interviews were recorded on a small tape recorder and test) for subsequent comparisons. A more detailed subsequently translated from Mongolian into English analysis will follow, but was beyond the scope of this by Otgontsetseg Alexander. report. In-Depth Interviews Media Presentations Six semi-structured, in-depth interviews asking people Media presentations included preparation of a to tell stories about wildlife that made them happy, documentary film illustrating critical issues relating to sad, angry, or afraid were conducted. In a final ques- continued sustainability of khulan. Both English and tion, respondents were asked how they generally felt Mongolian language versions of the documentary film about wildlife (Appendix D). No questions were asked were prepared. Autonomyproductions, an independent specifically for khulan, but this species was frequently film company, was sub-contracted to prepare the film. mentioned. For this report only passages concerning Both versions of the film can be viewed at: www. khulan or the management of wild herbivores were khulan.org. 14 2. Results T he results of the GPS tracking of seven Human Population and collared khulan, studies of khulan behav- Livestock Trends ioral ecology, and surveys of local attitudes toward khulan are outlined below. e use In the southeast Gobi, human population and live- of complementary study methods provides basic data stock numbers followed the national trend. Between on khulan range, seasonal movements and barriers 1985 and 2004, the human population of the 13 study to movement, habitat and water resource use, and soums increased from 41,072 to 62,735--a 53 percent interactions with humans and livestock. increase (Figure 6). e most dramatic increase in human population occurred from 1985 to 1995, and was mostly due to population increases in Sainshand (Dornogovi aimag center) and Dalanzadgad (Omno- govi aimag center) (Figure 7). Figure 6: Population Development (since 1985) and Livestock Numbers (since 1970) in Southeast Gobi 1,400,000 70,000 livestock humans 1,200,000 60,000 1,000,000 50,000 800,000 40,000 tion numbers popula ock est 600,000 30,000 Liv Human 400,000 20,000 200,000 10,000 0 0 1970 1975 1980 1985 1990 1995 2000 2004 Year (Source: NSO, 2005) 15 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Figure 7: Increase in Human Population in the Two Largest Aimag Centers in the Study Area 20,000 18,000 16,000 Dalanzadgad 14,000 12,000 numbers Sainshand tion 10,000 popula 8,000 Human 6,000 4,000 2,000 0 1985 1990 1995 2000 2004 Year (Source: NSO, 2005) Figure 8: Change in Livestock Composition in the 13 Soums of the Study Area in the Southeast Gobi, 1970­2004. 100% Sheep 90% Goats Camels 80% Horses Cattle 70% 60% composition ock 50% est Liv 40% 30% 20% 10% 0 1970 1975 1980 1985 1990 1995 2000 2004 Year (Source: NSO, 2005) During the same period, total livestock numbers curred in goats and camels. e percentage of camels increased from 761,973 to 1,121,095-- a 47 percent decreased from 18 percent to 6 percent, whereas goats increase. e main increase occurred between 1990 and increased from 30 percent in 1970 to 58 percent in 2000 following the collapse of the collective (Figure 6). 2004 (Figure 8). Similar to the national situation, the increase in e goat-to-sheep ratio in the study area (1.6 goats livestock numbers corresponded with a change in to 1 sheep) has implications for sustainability of both livestock composition. e most dramatic shift oc- extensively managed pastoral livestock production and 16 Results khulan populations, especially during severe drought years, such as occurred between 1999 and 2002. During this period, grazing by all large herbivores had a severe impact on ecological condition of shrub steppe vegetation communities, especially in areas of appar- ently high livestock densities around natural water sources and wells. Although areas near water sources and camps are naturally high-impact grazing/browsing areas because of concentrated animal use, degradation from grazing/browsing was extensive throughout the shrub-steppe ecological zone in the eastern portion of the Gobi Region (Sheehy and Byambadorj, 2003). Following the high mortality of livestock numbers in 2000/1, grazing pressure from livestock is presently even lower than in 1970 (1970: 2,830,570 sfu3, 2004: 2,691,908 sfu). Average stocking densities from 1970­2004 have varied from 17 to 22 sfu/100 ha. Due to the change in livestock composition, overall grazing pressure has increased only slowly, but the component of the livestock herd in the Gobi region that can most effectively utilize vegetation comprising khulan habitat is increasing at a higher rate than other livestock. The collared khulan quickly recovered from anesthesia. Images: P. Kaczensky Khulan Capture Between July 3 and July 7, 2005, we captured and Collared Khulan Distribution radio-collared seven khulan (three mares and four stallions) at three different locations (Table 2). All Sample Size animals recovered smoothly from anesthesia, and two were seen back at the water point within a few hours of Between July 2005 and February 2006, we received the capture. 880 GPS locations of the collared khulan. Although Table 2: Khulan Captured and Radio-Collared in the Southeast Gobi Date Time ID Sex Age (years) Method Place type Place name 07/03/2005 15:20 58848 female 5 hide riverbed Chuluut 07/03/2005 11:20 58849 male 4 chased at steppe Shar Khat 07/03/2005 21:15 58850 female+foal 7 hide riverbed Chuluut 07/03/2005 9:10 58851 male 2­3 hide riverbed Chuluut 07/09/2005 13:50 58852 male 4 chased at steppe Shar Khat 07/05/2005 21:20 58853 male 5­6 hide canyon Shivnee 07/04/2005 21:00 58854 female+yearling 11 hide canyon Shivnee 3 One sheep food unit (sfu) is the amount of dry forage needed to feed an average Mongolian sheep for 1 year, which is approximately 365 kg (Fernandez-Gimenez, 1999). e equivalencies for the other species are: 1 camel = 5 sfu, 1 horse = 7 sfu, 1 cow/yak = 6 sfu, and 1 goat = 0.9 sfu. 17 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Collared khulan back with the main group drinking at pot holes only a few hours after the capture. Image: C. Walzer collars were programmed to attempt 1 GPS location Distances Covered every 7 hours, on average only 0.83 GPS locations Distances covered within 7 hours, 14 hours, 21 were actually realized per 24 hours (Table 3). hours, and 28 hours followed an almost perfect linear relationship and suggest that khulan move constantly Of the seven collars, one stopped transmitting after during any 24-hour period (Figure 9). Based on this two weeks (collar 58849, a young stallion)--the collar relationship, daily distances covered within 24 hours may have failed or the animal was killed and the collar averaged 11.9 km. destroyed. e remaining six collars still transmit data, but at fairly irregular intervals. Table 3: Khulan Location Statistics, 7/7/2005­2/16/06 No. days No. locations Mean Success Protected area ID Start date End date tracked obtained no. locations/day rate (%) No. locations Percentage of total 58848 07/03/05 02/18/06 226 254 1.12 0.37 0 0.00 58849 07/09/05 07/30/05 21 34 1.62 0.54 0 0.00 58850 07/09/05 02/16/06 222 136 0.61 0.20 37 0.27 58851 07/03/05 01/23/06 200 127 0.64 0.21 0 0.00 58852 07/09/05 01/23/06 198 140 0.71 0.24 38 0.27 58853 07/03/05 01/05/06 182 123 0.68 0.23 36 0.29 58854 07/03/05 12/02/05 148 66 0.45 0.15 19 0.29 Total / Mean 880 0.83 0.28 130 0.15 18 Results Figure 9: Distances Traveled by Khulan Within 7-hour Time Steps (Sampling Interval of GPS Locations) 16,000 y = 11856x R2 = 0.9809 14,000 12,000 10,000 (m) elledv 8,000 tra istanceD 6,000 4,000 2,000 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Time between successive locations (hours/24) Especially during the summer, individual khulan of green pastures. Two khulan moved over 200 km showed periods of directed movements, covering large straight-line distances from the capture site within less distances in short time periods possibly in search than one month (Figure 10). Figure 10: Movement Paths of Individual Khulan (July 2005 to February 2006) 19 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Group of nine khulan (including one foal) walking on the west side of railway along the railway fence (near Airag soum on 07/17/2005). Image: P. Kaczensky Movement Barriers Table 4: Areas Covered by Individual Khulan During the Study Period e movement pattern of the young stallion 58851 and observation of a group of khulan along the ID Sex Monitoring MCP Trans-Mongolia railway fence suggests that this traffic from to (km2) axis poses a serious barrier for khulan movements. 58848 female 07/07/2005 02/18/2006 18,258 Furthermore, clumped locations along the border fence during the winter support the impression that 58849 male 07/09/2005 07/30/2005 10,747 fences severely inhibit khulan movements. 58850 female+foal 07/09/2005 02/16/2006 23,160 58851 male 07/07/2005 01/23/2006 43,105 Home Range Size 58852 male 07/09/2005 01/23/2006 12,841 e area covered by all animals together, the 100 58853 male 07/07/2005 01/05/2006 18,044 percent MCP, was 92,200 km². Animals used the 58854 female+yearling 07/07/2005 12/02/2005 39,339 easternmost portion of Omnogovi Province and all parts of Dornogovi Province west of the railway track (Figure 11). Only 130 (15 percent) out of 880 locations Khulan Distribution in fell into a protected area, mainly in the Small Gobi B Strictly Protected Area (SPA). the Southeast Gobi Area covered by individual khulan within the On five field trips to the southwest Dornogovi study eight-month study period was huge and ranged from area between June and October 2005, a total of 4,769 10,747 km² to 43,105 km² (Figure 12, Table 4). khulan in 423 groups were observed (Figure 16). Khulan location along the vehicle transects during the five field trips varied considerably between June Khulan Seasonal Habitat Use and October. Areas that had high khulan use during Individual khulan roamed over the largest area during one trip would often have no, or very little, apparent the summer season (Figure 13), shifted toward the khulan use during a subsequent trip, indicating the south in the fall (Figure 14), and restricted most of high mobility of khulan and their capacity to move their movements to a 50-km wide strip parallel to the long distances in a short time period. Herd size varied Mongolian-Chinese border in the winter (Figure 15). between 1 and 200 (Table 5). 20 Results Figure 11: Total Area Used by All Collared Khulan in the Southeast Gobi (July 2005 to February 2006) Figure 12: Areas Covered by Individual Khulan over an Eight-Month Period in the Southeast Gobi Study Area 21 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Table 5: Number of Khulan Counted During Five Trips the carcasses were fresh (i.e., < 1-year old) and showed in the Southeast Gobi between June and October 2005 signs of human-induced mortality (e.g. knife cut marks, bullet holes, boned-out hindquarters). With Number of Khulan counted the exception of one or two carcasses, no evidence of Time period total groups min. max. predation or even scavenging by mammalian predators Team 1: was observed. 24 June­17 July 2005 3,729 338 1 200 Team 2: Khulan Habitat Use June 2005 381 37 1 76 July 2005 206 18 1 100 Vegetation Type August 2005 345 19 1 100 Vegetation types occurring in the southeast Gobi are October 2005 108 11 1 25 generally dominated by shrubs (or have a high shrub Total 4,769 423 component) and are classified as belonging to shrub desert steppe. e five dominant vegetation types comprised of 25 distinct communities found in the SW Dornogovi study area are (a) semi-desert steppe Additionally, 140 khulan carcasses along our (five communities), (b) north desert-semi-desert (four vehicle transects were documented (Figure 17). More communities), (c) middle desert steppified desert than half of these carcasses were found in a 50-km (four communities), (d) south desert-true desert (five radius around the soum centers of Undurshil (45), communities), and (e) desert (eight communities). Mandakh (23), and Baya-Ovoo (12). e majority of Khulan used a variety of plant communities found in Figure 13: Seasonal Use of Habitat by Collared Khulan in the Southeast Gobi (July­August 2005) 22 Results Figure 14: Seasonal Use of Habitat by Collared Khulan in the Southeast Gobi (September­October 2005) Figure 15: Seasonal Use of Habitat by Collared Khulan in the Southeast Gobi (November­December 2005) 23 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Figure 16: Locations of Khulan Groups Encountered between June and October 2005 in the SW Dornogovi Study Area Figure 17: Locations of Khulan Carcasses Encountered between June and October 2005 in the Southeast Gobi 24 Results Figure 18: Availability and Use of Different Plant Community Types by Khulan as Compared to Two Sets of Random Points 0.18 available in MCP use khulan 0.16 uniform random 0.14 normal random ypety 0.12 communit 0.10 plant of 0.08 tion opor Pr0.06 0.04 0.02 0.00 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 59 60 63 64 Plant Community composition ID See Appendix B for description of plant community types represented by the numbers on the X-axis. the SW Dornogovi study area and no obvious pattern occasionally observed at considerable distance from of selection or avoidance emerged from a first rough water sources, or if observed near a well, they were comparison of two different sets of random points usually unattended by a herder. versus khulan locations (Figure 18). Five of the eight livestock GPS units were recov- In addition, individual variation was high and ered from herders during the October field visit. Of would not allow for a lumped analysis. For in-depth the other three herder GPS units, one of the herders analysis (i.e., differences between seasons, sex and age had moved out of the SW Dornogovi study area to classes), the khulan database and the habitat database northeastern Dornogovi, one herder was known to be were not adequate to determine if khulan selected for in his normal grazing area but was moving and could specific plant communities or vegetation types. not be located at that time, and one herder's current location was unknown to soum officials. Khulan and Human/Livestock Downloaded herder GPS information from the five recovered GPS units indicated that one herder did not Interactions carry the backpack GPS unit when herding livestock (Orgunbold), one herder evidently carried the GPS Pasture and Water Use by Livestock unit with him during a relatively extensive trip along the border with China (Urgamal), and three herders Most observations of livestock, both large and small, deployed the units as requested (Yubaa, Haucpau and were associated with water sources and/or herder Ulaankhuu, Figure 19). camps. Often livestock were being watered from hand-drawn wells (i.e., shallow traditional wells or Of the remaining three herders, two made daily broken, shallow-depth screw wells). Livestock watering "log" entries of khulan encountered while herding appeared to be on-going throughout the day in areas small stock and collected fecal samples from all large occupied by several herding families, with livestock herbivores, including khulan (Appendix E). e animal waiting their turn to water. Large livestock were 25 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Figure 19: Location and Movements of GPS-equipped Herders in July and August 2005 Figure 20: Distribution of Ulaankhuu's Small Livestock (16 July ­ 27 August 2005; n=23,998) 26 Results distribution information collected for livestock belong- to represent habitat normally grazed on a daily basis. ing to Ulaankhuu (a herding household in Mandakh Small-stock grazing of habitat at distances greater than soum) by the GPS unit is shown in Figure 20. 6,000 m from the ger seems to represents movements of the entire camp and livestock to a different location, During the six-week period small livestock were or the herder grazing the livestock between two or more tracked, livestock spent over 85 percent of their time water sources over a period of several days (Mon. otor). within 500 meters of the herder's ger. Since most herder's gers are close to a water source, maintaining Water Use by Khulan access to drinking water seems a major factor limiting animal distribution in SW Dornogovi. Loitering time Opportunistic observations of wildlife and livestock at around ger/water locations seems also a primary cause water sources indicated the following interactions were of animal-induced degradation of forage resources ongoing between khulan and livestock: found in close proximity to ger/water sources. e remaining 14 percent of their time was spent at greater Wells do not provide water for wildlife. e water distances from ger/water source locations. table generally is one to several meters deep and needs to be pumped up by hand or mechanical We assume that location of small livestock at device. Only enough water to meet the water distances > 0.5 km from the ger/well position represents requirements of the livestock is pumped to the the portion of small livestock's time in which food surface at any given time. Troughs are mostly intake through grazing or browsing is maximized. leaking and any remaining water in the troughs Small stock are always herded when grazing at dis- quickly seeps into the ground or evaporates. tances from the ger. Daily grazing activities that occur Wells do allow human presence in areas with between 500 m and 4,000 m distance from a well seem very little or no surface water. However, if open Horses, camels and khulan at multiple pot-holes. Camel guarding pothole. Khulan at water near ger, running from jeep. Khulan coming to water despite human presence. Images: P. Kaczensky 27 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion water is nearby, livestock is actively herded there at multiple potholes, khulan can drink simultane- to drink. Additionally, the large, un-herded stock ously with livestock. such as camels, cattle, and horses go to the open In dry riverbeds, khulan dig their own drinking water points by themselves. For herders, open holes (about 50 cm deep and 1 m in diameter. water points have the advantage that they do not Livestock also use these potholes. Camels often have to pump up the water; for livestock, they dominate single potholes for extended periods can come and drink at any time during the day or (they actually rest in front of them). night. If disturbed by humans, vehicles, or while guarded Khulan approach water points during all 24 livestock (sheep and goats) is watered, khulan wait hours. In some locations, peak drinking times in the vicinity for access to water. If disturbances seem to be in the late afternoon and early morn- are frequent or large stock dominates the water, ing. However, we did not attempt any systematic khulan grazing time may be severely reduced due surveys over 24 hours. to long waiting periods. e combined grazing Khulan do approach open water even when a ger pressure of livestock and khulan may quickly is very close, but run from approaching people diminish available pasture around water points. or vehicles. Flight distance varies, but may be as At most natural water points we saw signs of short as 100 m. poaching (hides, ammunition shells, khulan Khulan do approach open water simultaneously carcasses, and one direct observation of poaching). with un-herded livestock. However, camels and At some water points, natural mortality occurs horses appear dominant over khulan and can due to khulan getting stuck in deep, mucous mud. inhibit khulan access to localized water (single Herders generally try to remove these carcasses in small open potholes). At streams, larger pools, and order to keep the water clean. Pothole dug by khulan. Khulan digging for water. Poachers hide at water point. Image: C. Walzer. Poached khulan ~50 m from water. 28 Results Khulan waiting for access to water. Dead khulan that were unable to escape from thick mud. Images: P. Kaczensky. Khulan Spatial Relation to Impact of Well Construction and Human Intrusion Rehabilitation Distance to the nearest open water source averaged A primary focus of this study was the direct and 13.7 km for khulan and was not significantly differ- indirect impact of well construction or rehabilitation ent from two sets of 880 random points (ANOVA, on khulan and associated wildlife. e direct impact F=0.142, d.f.=2, P=0.868). Interpretation is hindered of a well is obviously minimal since khulan have by the fact that we probably did not know the location no inclination or capacity to utilize water directly of all water points and additionally lacked information from a well. e most obvious indirect impacts will on duration of water availability at different water be the increase in human presence and livestock in points. pastureland most recently accessible only to wildlife, and especially the khulan. Distance to the nearest well averaged 6.3 km for khulan positions in Dornogovi Province, but again Assuming a 5 km grazing buffer around water there was no difference to the two sets of random sources, which is the maximum distance the three points (ANOVA, F=0.035, d.f.=2, P=0.966). Inter- herders who were given GPS units in this study moved pretation is hindered by the fact that the error of well (Figure 20) and which equals the distance described locations is in the range of 1 to 10km. In addition, by others (Mearns et al., 1994; P. Kaczensky and O. data on where and when herders and their livestock Ganbaatar, unpubl. data), the pasture available to were present is lacking altogether. We did not expect herders is a function of the number of wells in the area a well structure per se to be of any importance for (Table 7; Figure 21). khulan. If only natural springs and surface water sources e distance of khulan locations to soum centers are available to livestock, total pasture area during the showed a bimodal distribution with one peak around 40 km and a second peak around 90 km, whereas Table 6: Distance to Open Water and Soum Centers. random points showed a normal distribution with a mean of 39.9 and 37.8 km. We are not sure why Type of data Mean distance to (km) khulan show the bimodal distribution, but the first open water wells soum centers peak around 40 km equals the mean distance of the Khulan 13.7 6.3 bimodal with random points. However, the minimum distance peak at 40 and 90 of khulan locations from soum centers was 5.4 km, those of random points much shorter (1.3 km and Random (normal) 13.5 6.3 39.9 2.2 km). Random (uniform) 13.5 6.2 37.8 29 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Table 7: Water Source Availability and Grazeable e reconstruction of shaft and deep production Pasture wells increases the total livestock pasture area in the study area to 11.67 million ha (116,700 km²). By adding Additional area Water source Area (km2) gained with wells all former shaft and production wells at a relatively high cost for construction and maintenance, the grazeable Springs + lakes 55,903 area gained was another 0.73 million ha (7,300 km²). Springs + lakes + 109,461 53,558 traditional wells Springs + lakes + 114,991 5,530 Human Attitudes Toward Khulan traditional wells + shaft wells Attitude Survey Springs + lakes + 116,793 1,802 traditional wells + Study field teams surveyed residents of the southeast shaft wells + Gobi to determine their attitude toward the khulan. production wells In total, we interviewed 161 local families (0.3 percent Total area gained when 60,890 of the total population). Sixty-five of the respondents rehabilitating all broken wells lived in the countryside, 60 in a district center, two were interviewed at a railway station, and 34 were interviewed in the two provincial centers, Sainshand summer in Dornogovi aimag would be approximately and Dalanzadgad (Table 8). 5.59 million ha (55,590 km²). e addition of all formerly installed traditional hand wells increases Of the 161 people interviewed, 90 were women the total grazeable area by a further 5.35 million ha and 71 were men. e age of the respondents ranged (53,500 km²), which is almost double the area without from 16 to 76 years and averaged 40 years. Eleven hand wells. Figure 21: Example of Additional Livestock Pasture Area Obtained by Well Rehabilitation in Dornogovi Aimag (5 km buffer) Existing water resources Broken wells Overlay of existing water resources and broken wells Additional grazing areas after fixing the broken wells 30 Results Table 8: Human Population in the Study Area and Sampled Population Soum Total area (km2) Population in the area Population sampled Total Countryside Total Countryside Omnogovi province Khatanbulag 18,892 3,061 2,480 22 15 Bayan-Ovoo 11,113 1,589 1,166 20 17 Khankhongor 10,460 2,643 2,234 20 3 Manlai 12,268 2,393 2,484 9 3 Nomgon 18,395 3,027 1,862 8 3 Khanbogd 14,720 2,292 1,476 8 2 Tsogt-Tsetii 7,106 2,050 1,402 2 2 Dornogovi province Saihandolaan 9,699 1,289 935 21 3 Ulaanbadrakh 13,821 1,763 1,286 18 7 Khuvsgul 8,238 1,706 1,235 13 6 Erdene 10,179 2,442 1,418 8 3 Mandakh 12,436 1,904 1,406 7 1 Urgun 8,453 1,933 923 5 0 Total 155,779 28,092 20,307 161 65 respondents were between 16­20 years, 31 between remaining 15 respondents made between 166,000 and 21­30 years, 52 between 31­40 years, 35 between 750,000 Tg and can be considered wealthy. 41­50, 19 between 51­60 years, and 13 were older than 60 years. Reliability estimates were high (Chrombach's = 0.78) for the seven items in the attitude scale. Average Livestock raising was the main source of income attitude toward khulan (attitude score) was neutral for 67 (42 percent) of the respondents. For the other to slightly positive and averaged 3.7 (range: 3.2­4.1). respondents, current occupations were (a) other, but ere was no difference in the attitude score between not specified (38); (b) unemployed (10); (c) driver herders and people with other occupations (T-test: (5); (d) railway employee (4); (e) shop owners (3); (f) df=154, P=0.9), between sexes (T-test: df=154; government administration (3); (g) doctor (2); (h) P=0.22), or between age classes (ANOVA: df=5; cook (2); (i) student (2); (j) retired (2); (k) owned P=0.68). One of the key variables influencing the a company (2); (l) guard (2); (m) employees of an attitude score was the perceived trend of the khulan insurance company (2); (n) various other jobs (18); population (ANOVA: df=5; P<0.01; Figure 22). and (o) no answer (2). We grouped the most important items in response eir average monthly income was 77,300 Tg to the open-ended question "What are the most (~$65). Forty-six respondents claimed to earn less than important issues in respect to khulan?" into 218 35,000 Tg and thus can be considered poor, 47 had statements, which were grouped around 25 different a moderate income between 40,000­70,000 Tg, 35 topics. e most important--based on how often the had a good income of 71,000­150,000 Tg, and the statement came up (frequency in brackets)--were: 31 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Figure 22: Local Attitudes Toward Khulan According to Perceived Population Trends * * 5 4 e scor ttitudeA3 2 1 N = 57 15 43 Increase Stable Decrease The perceived trend of the khulan population was a key variable for the attitude score (1=very negative, 2=negative, 3=neutral, 4=positive, 5=very positive). The box indicates the median, 25% and 75% quartiles and circles mark outliers. We used ANOVA statistics and t-tests (with Bonferoni corrections) to check for differences in mean score values according to respondents perception of the khulan population trend. ** mark significant differences in the mean score value on the p < 0.001 level. Protection of khulan needed + combat poaching (57) incentives for conservation and involving local people Khulan impact on pastureland (43) in wildlife conservation. Outside of the khulan range, Controlled harvest wanted (23) people seem to know very little about the species and Increase khulan numbers (18) are largely unaware of any conservation concerns (for Aim for a stable population (15) details see Appendix F). Management of water points (12) Local Wisdom In-Depth Interviews ree of the herders selected to use the GPS backpack In all of the interviews, there was the notion that unit to determine livestock distribution and collect most wildlife populations are decreasing and nature fecal samples showed considerable interest in khulan. in general is becoming degraded. In addition, the One herder, based on his journal entries, had an interviewees felt that old values are fading away due exceptionally high interest (Appendix E). His interest to socioeconomic changes. All believe that wildlife are and cooperation shows the potential for involvement an important part of Mongolia's nature and all see a of selected local people to gather local information on need for better protection. All four interviewees from wildlife. During his four-week study period, herder the Gobi stated that poaching by locals, as well as by #7 saw over 250 khulan near or on the pasture his market hunters, is happening. Contrary to gazelles, livestock was using. His journal entries reflected his khulan are seen with some ambivalence. One the one perception of wildlife needs, especially the habitat hand, people want to ensure their survival, but on the needs of khulan, and their interaction with humans other hand, khulan are also seen as competing with and livestock. How representative his views are of what livestock for pasture. One possible solution offered other herders in the area believe--and how accurate to address human-khulan conflicts was providing his beliefs are--remains to be tested. 32 3. Discussion and Conclusions T he threat to khulan from human intrusion dynamics related to the availability of pasture, forage, is significant--particularly from those and water. Rather than select for a specific habitat developments which are affecting the type, observations and the data from this study suggest landscape of the region, as khulan range that khulan seek areas of high biomass production. over such large areas and are very sensitive to barriers ese high-quality foraging areas are usually spatially and habitat fragmentation. Access to water and grazing and temporarily distinct and result from differences is crucial for survival, and well rehabilitation has the in local rainfall patterns in the non-equilibrium potential to negatively impact khulan by increased environment of the Gobi region. However, the present competition for the resources unless appropriate man- data is not sufficient to prove this relationship. More agement strategies and mitigation actions are put in individuals need to be followed over an extended place. A key threat to khulan is from illegal hunting, period of time. increasingly for markets, rather than for subsistence. Effective law enforcement is needed to prevent this, It is well known to Mongolian herders that but there is also the potential for a controlled and availability of water and pasture varies among seasons sustainable subsistence harvest involving local people, and years. Herders, in order to capitalize on this alongside education and awareness efforts. Monitoring variation, have developed a form of pastoralism in the of populations is essential for wildlife management, Gobi region that provides access to critical resources and the responsible institutions require capacity over large tracks of land. eir form of traditional building to allow them to handle this role effectively. pastoralism mimics movements and foraging strategies of large wild herbivores such as the khulan, Mongolian gazelle, and black-tailed gazelle. However, due to their higher mobility, khulan can venture farther away Spatial Dimension of Khulan from water and can cover long distances more quickly, Conservation often arriving at forage green-up areas before herders and their livestock. e scale at which khulan use the One of the major finding of this project is that khulan landscape is therefore much larger than the scale local conservation needs to occur on a landscape scale. herders currently use the landscape, making it very Within less than one year, animals collared in the difficult for local people to understand the dynamics study ranged over an area in excess of 90,000 km². and associated needs of the khulan population. Khulan made use of protected areas, but protected areas alone are much too small to provide sufficient Large-scale movements and the observed annual water and pasture for a year-round khulan presence. and seasonal shifts in khulan distribution make We believe that the driving force behind observed localized counts almost meaningless. On the contrary, large-scale khulan movements are the strong temporal counts of the khulan population at soum or aimag 33 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion level can be highly misleading. In areas of low khulan other large herbivore wildlife to habitat in the Inner presence, people often have the impression that the Mongolian Autonomous Region of China. khulan population is decreasing. In areas of high khulan presence, people often have the impression Fragmentation of habitat into small, and often that the population is increasing. In reality, available non-contiguous patches, decreases capacity of large information on khulan population dynamics indicates wild herbivores to escape locally poor habitat condi- that the khulan population in the southeast Gobi is tions. e Gobi region is predisposed to large envi- one large population that moves freely over an area ronmental fluctuations and catastrophic events that of approximately 100,000 km². us, conservation can cause large fluctuations in wildlife and livestock and management decisions need to be made at the population numbers. Fencing, and other changes in landscape level, not at the local level. land use, reduces the capacity of wildlife populations to "outrun" droughts or harsh winters by moving to Mobility and their capacity to move long distances better areas. Without this escape option, intra- and in a relatively short time are the primary factors inter-specific competition will be high, resulting in influencing habitat use by khulan. Habitat use is at poor body condition, poor recruitment, and high landscape scales, but with short temporal duration. mortality. Numbers may regionally drop below a Having two components of scale not only increases the critical threshold and may eventually result in regional potential of conflict with human intrusion, but also extirpation. Reestablishment of regional populations adds a whole new dimension to monitoring khulan will be slow or impossible due to fragmentation. e populations and defining and mapping habitat needs. fact that khulan populations no longer exist east of Currently, data on important wildlife habitat features the Trans-Mongolia railroad--despite the presence are not available at a reasonable resolution at the of large areas of suitable habitat--is a warning that landscape scale. For a truly analytical assessment of the should not be ignored. impact of wells and other human intrusion on khulan and other rare wildlife species in the Gobi region, an e southeast Gobi is rich in mineral deposits, and adequate habitat database is needed. e limited dura- exploration and mining activities are increasing (World tion and funding of this study, while adding insight Bank, 2006). In the area around the Small Gobi SPA, to our understanding of khulan ecology and issues 117,000 km² of mineral concessions are held by a single threatening sustainability of khulan populations, company alone (after Ivanhoe Mines Mongolia 2005; basically only explored how these questions could be Figure 23). e Oyu Tolgoi mine holds major gold and addressed by a larger follow-up study. copper deposits, and further exploration in the conces- sion area will start soon. In order to allow transport, a fenced road and railway track is planned from the Habitat Fragmentation Threats Tavan Tolgoi coal mine to the Oyu Tolgoi gold/copper mine and further on into China. is track will cut e highly mobile and migratory lifestyle of khulan through prime khulan habitat and, without mitigation and other far-ranging large herbivore wildlife make measures, will provide a major migration barrier for them highly sensitive to fragmentation of the khulan, black-tailed gazelle, and Mongolian gazelle. landscape. Busy transportation routes and long fences ere is a high probability that the western part of impede movement of large herbivore wildlife by the current khulan range in Omnogovi Province will creating access barriers to important resources or by be separated from the khulan range in Dornogovi forcing them to make large detours to gain access. Province. Without imposition of appropriate mitigation e barrier effect of the fenced Trans-Mongolia measures, the planned traffic corridors from Oyu railroad between Ulaanbaatar and Bejing has already Tolgoi mine will result in the ecological separation been demonstrated for the Mongolian gazelle (Ito et of the Small Gobi A from the Small Gobi B SPA. al., 2005). Our data indicates that the railroad is a Additional mining concessions are held by smaller barrier to the khulan, as well as the gazelle. e fenced companies. Small scale miners engaged in placer border between Mongolia and China, which forms the gold mining--nicknamed "ninjas" for their habit of southern boundary of the Gobi region, including the carrying a green plastic bowl for panning strapped on Small Gobi SPA, also prevents access of khulan and their backs, thus resembling the ninja turtles of the 34 Discussion and Conclusions Figure 23: Khulan Locations and Movements Relative to Current and Planned Mining Operations The figure shows the area under mining concession, location of mining operations, and traffic corridors under construction or in the planning stage by Ivan- hoe Mines, the largest investor in the area. Additional, smaller concessions exist especially to the north. television series--also operate in and around the Small and by the availability of herder labor to provide Gobi SPAs (World Bank, 2006). water to livestock. If a natural water source is nearby, herders gener- Water Access ally prefer to water their animals there. e further a herder has to drive his flock to water, or if several In the arid Gobi region, access to water is a key factor herders have to use the same well or water source, the for the survival of wildlife, livestock, and humans. shorter is the time available for grazing. e staggered Wells do not provide a source of water for wildlife access to water results in khulan and other wildlife because the water table is generally one to several populations being the last ones in the line. Any time meters deep and water needs to be drawn by hand humans and their flocks arrive, wildlife populations or mechanical device. But wells clearly allow human get scared away and have to wait, which may greatly and livestock populations to have a presence in areas compromise grazing time and puts increased grazing that would otherwise have limited or no access. pressure on pastures close to water. Khulan do ap- Herder access only to shallow hand-drawn wells also proach open water simultaneously with un-herded live- limits livestock numbers because drawing water by stock. However, camels and horses appear dominant hand is labor intensive and time consuming. us, over khulan and can inhibit khulan access to localized livestock numbers are limited because their forage water (single small open potholes). At streams, larger needs must be met within the distance they can pools, and at multiple potholes, khulan can, and do, graze from water, by the time required for watering, drink water simultaneously with livestock. 35 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion A well itself appears to have little, if any, impact on the 10-km buffer zone, could have advantages for both khulan habitat use. is is of little surprise, as a well livestock and khulan, including: is a small structure in a large landscape and, without human presence, has little meaning for khulan or Spreading out livestock grazing pressure to a other wildlife. e use of a particular well by herders much larger area; i.e., lowers livestock density and their livestock not only depends on pasture condi- (number/unit area), which diminishes grazing tion, but also on socioeconomic constraints. Without impacts. near-real time information on herder presence and Increasing water and forage availability to khulan movements at and near wells, and the exact location and other wildlife by using water control at of all wells, an analytical assessment of the impact of secondary water sources to control livestock use wells on khulan habitat use was not possible. (i.e., timing of use by livestock can be regulated). Improving access of khulan and other wildlife to e distance that livestock can graze from a well habitat by installing float valves in newly devel- is not the only factor constraining herder use of forage oped water troughs that allow water from storage or khulan habitat. Traditional hand-drawn wells, tanks to keep water available for wildlife when which are the primary livestock water sources during livestock are not present at the water point. normal years, and especially during drought years in Maintaining large movement corridors available the Gobi region, can provide water for only a limited for khulan and other migratory large wild her- number of animals because of the time and labor bivores that are only temporarily or occasionally required to draw water by hand and the slow recharge used by livestock when water is made available. rate of shallow wells. Another factor related to time and labor is the need to water small stock separately Depending on mitigation measures, rehabilitation from large stock to avoid dominance behavior by of wells can be either beneficial or detrimental to camels, horses, and cattle. e slow rate at which wildlife. If wells reduce livestock presence at natural water can be provided to stock from traditional hand water points and allow for a more even distribution of wells requires that stock be separated by kind to avoid wild and domestic herbivores, the effect would be posi- interspecies conflict during watering. ese factors tive for the pastureland and associated wild herbivores. also limit distances traveled and pasture area grazed However, this would require that livestock numbers do by livestock. Reconstruction or new development of not increase in response to improved access to water, shaft- and deep-wells increases both the area of pasture livestock are kept largely away from open water, and grazed and the time livestock are able to spend grazing that people do not harass or kill wildlife. by reducing the time and labor involved in providing livestock water. Access to water and availability of water may in the future become even more complex as legal and Although rehabilitation of non-operating shaft and illegal mining has substantial impact on local or deep production wells will benefit herders and their regional groundwater tables. For example, the Ivanhoe livestock, conflict between large wild herbivores and Mine expects to have an average water demand of livestock for pasture area potentially could increase. 1,865 m³/hour. During its expected 40-year opera- Livestock will have the highest access to water and tion, the mine theoretically will consume 650 Mm² associated forage around wells, while wildlife will have of water (Ivanhoe Mines Mongolia, 2005). e mine diminished access to habitat in the area and no access will not use surface water. It intends to meet its water to the water, i.e., livestock and humans benefit poten- needs from groundwater stored in sedimentary basins. tially at the expense of wildlife. However, alternatives However, the long-term effects of gold mining on the to construction of solitary new or rehabilitated deep water regime in the Gobi region are unknown, and wells as the sole approach to resolving water shortages are subject to very controversial and widely divergent exist, and should be considered. For example, develop- predictions (Heaps, 2004; acker, 2004; Farrington, ing a water distribution system whereby water from 2005). an existing high volume deep-well is piped to storage tanks positioned at a higher elevation, and then is Further uncertainty is provided by the impacts gravity-fed to livestock-watering facilities beyond even of global climatic change. Mongolia is predicted 36 Discussion and Conclusions to become hotter and drier in the future, particu- Illegal subsistence hunting of khulan is done by larly in the summer. ere may be some increase local people as well as by outsiders engaged in market in snowfall in winter, but the predicted increase in hunting. Multiple people, including several persons evapotranspiration is much higher than the predicted interviewed during the study, claimed that khulan increase in precipitation, and water-stress is therefore meat is sold to meat markets and factories in Ulaan- likely to increase (Batima, 2006). Although climate baatar and is served in ger restaurants along the border changes predicted for the Gobi region are much less and the main transportation routes in the Gobi region. pronounced than those for other parts of Mongolia, Hunting stands littered with rifle shell casings and the extreme climate and limited water resources in this khulan carcasses at, or near, water points provide clear region mean that its inhabitants--human, livestock evidence that people take advantage of the khulans' and wildlife--are highly vulnerable to change. To need to regularly access drinking water. Wells allow date, there have been many reports of surface water human presence in areas that otherwise would be of sources drying up (Davaa, 2004, in Batima, 2006; little interest for human occupation. As a side effect, Fernandez-Gimenez, in prep.) and although to date the this brings people closer to wildlife and allows for easy evidence is largely anecdotal, it is consistent. hunting opportunities, far away from understaffed and underpaid law enforcement personnel in the soum or aimag centers. Poaching and Hunting for Sale Opportunistic subsistence hunting by the Illegal poaching has been identified as the primary occasional herder is not the main problem. Rather, threat to Mongolian wildlife (Scharf and Enkhbold herder camps may be sources of information and 2002; Wingard and Zahler, 2006). Our data sup- support for poachers and market hunters--a ports this view. During our five trips through the place to stop, refresh, and ask the whereabouts of southeast Gobi, and without much of a search effort, wildlife. New markets may also open up with the we encountered 140 khulan carcasses, most of which development of the mining industry. e mine showed signs of human-induced mortality. We even will provide much needed employment and can be observed one well-equipped hunting party shooting expected to provide increased income and improved and boning a khulan. Local herders, rangers, and infrastructure for the whole region (Ivanhoe Mines soum governors repeatedly complained about poach- Mongolia, 2005). As the population increases and ing. In addition, controlling poaching was the key changes, the demand for game meat ("steppe meat") issue that came up during our interviews with local and opportunities for recreational hunting by mining herders. personnel and their families may increase. Under these circumstances, and in the absence of monitor- e relaxation of gun ownership laws in 1995, and ing and conservation activities, mining and other the availability of cheap ammunition, have allowed forms of human intrusion associated with economic more individuals to harvest wildlife in larger quantities development will likely have adverse impacts on local than were ever imagined under the socialist system. wildlife populations. A nationwide survey (Wingard and Zahler, 2006) extrapolated that an estimated 245,000 Mongolians actively hunt today as compared to 25,000 hunters Khulan Monitoring under the Soviet-style hunting brigade system. is translates to a 10-fold increase in the number of Reliable population estimates of khulan and other hunters and ranks Mongolia as one of the countries steppe- and desert-steppe ungulates in Mongolia are with the highest hunter/non-hunter ratios in the hindered by the large size and remoteness of their world (Wingard and Zahler, 2006). From the same distribution range, the lack of suitable fixed winged survey, the illegal take of khulan was estimated at aircraft, lack of proven ground survey protocols, 4,500 individuals annually. Since both population marked seasonal movements, a locally clumped estimates and population takes are presently only "soft occurrence, and the large variation in possible group numbers," the actual threat potential is impossible to sizes (Buckland et al., 2001; Kaczensky and Walzer, quantify. 2002b-2003b). is uncertainty makes the assessment 37 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion of the impacts of human intrusion on large wild forage growth modeler and fecal profiling (nutritional herbivores impossible and hinders or delays conserva- management system) are currently being developed tion actions. is may prove detrimental given the and adapted to Mongolian conditions (Gobi Forage precarious state of Mongolia's endangered large Project-USAID/Mercy Corps). Decision support herbivore wildlife in the Gobi region. e long-dis- systems are available to help managers and users tance migratory movements of large wild herbivores make appropriate decisions relative to sustainable and the size of the southeast Gobi make aerial surveys, use of natural resources (Grazing Land Applications, similar to those conducted in the past (e.g. Reading Nutritional Balance Analyzer-Texas A&M Univer- et al., 2001), the preferred monitoring method. e sity). Integration of these, and other support systems urgent need for a sound monitoring scheme was also a with more traditional methods of data collection, recommendation of the 1st Asiatic Wild Ass Confer- would facilitate development of current and future ence held in Hustain Nuruu, Mongolia, in July 2005 natural resource management and conservation (AWAC, 2005). strategies in the Gobi region and ultimately through- out Mongolia. Simultaneous to monitoring the ungulate popula- tions, it is necessary to raise awareness about the fate of Mongolia's wildlife. Public education and law Lack of Institutional Capacity enforcement need to go hand-in-hand, and ideally local people need to be involved in local monitoring e best monitoring and decision-making tools are activities. useless when not integrated into clear management plans with control and feedback loops. At the present time, conservation of natural resources, including Conservation Management Planning large wild herbivores and pasture resources, receives little attention from either the public or private sectors. Understanding the range, habitat, and ecology of large During our study, it became obvious that, while inter- wild and domestic herbivores has long-term implica- est in management of wildlife and natural resources tions for the sustainability of Mongolia's wildlife, was evident at all administrative levels of government pastureland resources, and extensively managed and among all segments of the Mongolian population, pastoral livestock production systems. Mongolia needs the capacity to actually manage wildlife and natural to put into place, activate, and support a natural resources was almost nonexistent. Although laws and resource management system that facilitates formation regulations protecting khulan from poaching and il- of rational policies and conservation programs that can legal market hunting exist, enforcement is constrained mitigate adverse impacts of privatization, natural re- by the lack of personnel, equipment, and financial source exploitation, and human intrusion in the face of support needed to actually control such activities economic development. Such a system should be able or have a "presence" in areas where most poaching to assess and monitor resource information on both occurs. a landscape and site-specific scale; develop and assess resource data, and generate near-real time output that In our discussions with aimag and soum govern- can be immediately used by natural resource managers ment staff and residents, and with herders throughout and users throughout Mongolia. the southeast Gobi, their recognition of problems and issues threatening khulan and other large herbivore e basis for such a system already exists. e wildlife was obvious. Many of the people we met technology platforms on which the system operates suggested management and habitat improvement are developed and have been tested in other areas as alternatives, which if implemented would improve resource conservation tools. GIS systems are becom- the sustainability of khulan and other large herbivore ing widely used in Mongolia for a variety of purposes. wildlife. Many of these recommendations are included Other systems are becoming available that, when in this report. However, without a conservation man- combined with enhanced institutional management agement plan and an enhanced institutional manage- capacity at local levels, could make conservation ment capacity, efforts to ensure survivability will have management planning a reality. e PHYGROW little chance of success. Although herder involvement 38 Discussion and Conclusions in khulan conservation is critical, expecting them to Most international development projects are assume responsibility for khulan welfare in the absence required by their own environmental regulatory of a working conservation management plan and agencies to ensure that project development will adequate government involvement is unrealistic. not adversely impact threatened species directly or indirectly through habitat modification. e presence International development agencies and the private of the khulan in a project area should automatically sector need to become more involved in local-level trigger environmental safeguards designed to mitigate conservation management planning and developing lo- potential or actual negative impacts occasioned by cal-level institutional capacity. For example, the World project implementation. ese safeguards should be Bank's Sustainable Livelihood Project is focused on identified and addressed during project design, and improving the livelihood potential of rural livestock upgraded as necessary during project implementa- herders and residents of rural soums. e pastoral tion. Well rehabilitation and most other forms of risk management component of the project considers human intrusion require adherence to environmental rehabilitation of mechanical wells constructed during safeguard policies to ensure that habitat needs of the the collective period (i.e., < 1990) as an important and khulan and other endangered species are addressed effective method of improving livelihood sustainability and not negatively impacted by the project or through of rural residents in the Gobi region. Mechanical project implementation. National and international wells, which were designed to be used exclusively by development agencies need to ensure that conserva- humans (especially livestock herders) to water livestock tion management planning and building local-level and as sources of water for human consumption, institutional capacity are priorities of the project before can cause degradation of khulan habitat near wells actually implementing development activities of the by concentration of livestock and humans. is project. Without prior development of local conserva- was recognized during implementation and was the tion management plans and sufficient institutional genesis for this work. However, good EAs can only be capacity to enforce laws and regulations, private conducted when sufficient baseline data are available. sector companies such as large foreign companies Other international organizations are also involved developing mineral resources in the Gobi region will in well rehabilitation, and it would be good if the have little incentive to ensure that khulan and other awareness of the impacts on threatened wildlife were large herbivore wildlife are not adversely impacted by broadly recognized. For example, in the JICA (2003, mining activities. 2004) reports well rehabilitation is only discussed in respect to optimizing livestock use of pastures. 39 4. Recommendations I mportant steps which need to be taken to protect newly acquired land resources or infrastructure address human-wildlife interactions in the Gobi developments from infringement by others. Since long areas of Mongolia include preparing appropriate fences are barriers to the movement of free ranging Gobi-wide strategies addressing infrastructure large herbivores, extensive fencing constructed without developments, sustainable harvesting, ecotourism, and sufficient foresight and planning could significantly water supply; the enforcement of existing regulations affect sustainability of endangered khulan populations regarding hunting and increase in capacity; raising and other highly mobile large herbivores. awareness of wildlife issues in the Gobi; and develop- ing the means for developing accessible databases GoM should initiate a dialogue on the impacts of and monitoring to measure impacts of infrastructure large-scale privatization of natural resources, espe- development, changes in land use patterns on local cially the impact that artificial barriers such as long pastoralists and wildlife, and climate change. ese are fences, roads, and open-pit and strip mines will have elaborated below. on movement of migratory large herbivore wildlife. Where fences and transportation routes are to be developed, EA studies need to be conducted Gobi-wide Initiatives beforehand and wildlife corridors incorporated in appropriate places. GoM and appropriate donors should initiate a Develop a Gobi Region Infrastructure Strategy which technical and financial feasibility study of wildlife- Gives Attention to Conservation, Especially Habitat friendly road and railroad underpasses/overpasses, Fragmentation and Water-Wildlife Issues and provide those where likely to be most effective. Mongolia remains largely a nation without fences, is measure will also restore connectivity of the except for a few specific cases such as the border fence, Mongolian gazelle population. animal exclusion fences along the Trans-Mongolia GoM should initiate a project to track movements of railroad, and fences around military and police instal- migratory large herbivore wildlife throughout their lations. Other existing barriers to wildlife and livestock current range to provide baseline information for movement include the Trans-Mongolia railroad itself, planning wildlife movement corridors and ensuring the increase in the length and number of surfaced continuity of habitats. roads, and the increase in mining activities, especially strip mines and open pit mines. However, there is a Explore Potential for Sustainable Local Khulan/Wild- high probability that ongoing privatization of land life Harvest and other natural resources will increase the perceived Subsistence hunting by local people is still a major need for fences to define ownership boundaries and economic factor in rural Mongolia. Income generated 41 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion from the trade in fur alone is in the region of $100 Negotiate and enforce a 500 m minimum distance of million annually (Wingard and Zahler 2006). A large herder camps from open water sources. portion of wildlife use and trade is presently illegal, and natural wildlife resources are rapidly decreasing. Enforce Existing Regulations e option for the legal, sustainable use of wildlife should be used as an incentive for wildlife conserva- Improve Well Planning and Management tion. Rehabilitation and construction of wells in the Gobi To guarantee long-term sustainability, use of natural region is a priority for national government and resources needs to be tightly linked with sound international development agencies. Many of the monitoring schemes. mechanical wells built during the collective period Clear management goals need to be negotiated. no longer operate, reducing accessible pasture area for e negotiations should include all stakeholders; livestock production and creating hardship for herders management plans should be developed based on whose water access is limited to hand-drawn wells with these negotiated goals. high labor requirements. Establish a control and law enforcement unit. Develop Ecotourism Opportunities Strictly enforce regulations requiring prior assessment of impacts on wildlife as well as benefits to livestock Ecotourism opportunities associated with the khulan production in at least all donor-funded projects. are limited in Dornogovi, even though the six soums Include an EA of well rehabilitation impacts on in the SW include the Special Protected Area and have seasonal access of khulan to critical resources before the highest population of khulan. e presence of the any well rehabilitation commences. As indicated railroad in Dornogovi aimag and the improvement of elsewhere, this is not easy and so those contracted to the highway between Ulaanbaatar and the Chinese conduct EAs must seek specialist advice to make the border substantially improve access of tourists to the investigation, description and mitigation as mean- Small Gobi SPA. Ecotourism could provide some ad- ingful as possible. See also Database and Monitoring ditional income to local herders and would give them below. an incentive to conserve or at least tolerate khulan Include an EA of new well impacts on seasonal presence on their pastures. access of khulan to critical resources before any well development commences. GoM and relevant donors should support develop- Improve capacity of soum and aimag government ment of private sector ecotourism focused on viewing staff to prepare and update resource development and khulan and other wildlife and interacting with management plans. families engaged in traditional pastoral livestock Improve cooperation between the different interna- production. Herders living in khulan areas should tional organizations involved in well re-habilitation be integrated into the ecotourism project and receive projects through the new National Coordinating compensation for their time and effort. Council for Pastureland Management. Link wildlife-based ecotourism to other tourism to GoM should initiate a dialogue among the ap- the region, such as visiting monasteries, petrified propriate agencies on organizing access to, and the forests, etc. use of, water resources and defining "water rights" of Improve Khulan/Wildlife Access to Water persons, entities, and users. Drinking water is a critical resource for khulan and Enforce Hunting Regulations other large herbivores in the Gobi region. Human Illegal take and sale of khulan meat and parts is presence and their domestic livestock can inhibit becoming organized and increasingly effective. A wildlife use of available spring and surface water. major stimulus to "market hunting" is the relatively high price of meat from domestic livestock in the Negotiate and enforce wildlife-only (i.e. khulan, marketplace and the lack of non-livestock income gazelle, etc.) watering points in return for well generating activities in rural areas. rehabilitation or newly developed wells. 42 Recommendations Train and support a wildlife law enforcement unit be presented to herders and soum residents during in the national police force or establish a wildlife annual soum and aimag "naadam" fairs. agency to investigate and prosecute wildlife poachers, Environmental education kits should be developed especially poachers engaged in organized illegal take and disseminated to kindergartens and schools and marketing of wildlife meat and body parts throughout Mongolia. rather than poaching for local self-consumption. Recruit local residents to the wildlife enforcement Database and Monitoring unit whenever and wherever possible. Develop and apply "quick-test" technology on meat Create a Habitat Database markets to identify illegally offered wildlife meat. Develop and apply enforcement protocols and impose To assess the impact of current and future infrastruc- substantial fines on businesses and entities profiting ture development, or changes in land use patterns on from illegal take of khulan and associated wildlife. local pastoralists and wildlife, a common database Initiate a reward system to compensate herders needs to be assembled and freely accessible to person- and soum residents who provide information and nel engaged in monitoring and assessment. assistance to wildlife enforcement officers. Develop aimag and soum wildlife enhancement Permanent structures like wells, winter camps, bag and protection programs that engage local herders to centers, mining camps, roads and fences outside of obtain wildlife biological data, provide compensation urban areas need to be mapped and updated on a for wildlife use of pastureland habitat, and encour- regular basis. age herder protection of the wildlife resource. Plant communities need to be re-mapped and extrapolated via remote sensing on a scale of Raise Awareness of Gobi Issues Relating 1:100,000 or less in order to allow the identification of key habitats (e.g. oasis communities and areas to Large Herbivores having a high occurrence of foraging patches). Climate- and weather-related data could be obtained Even though Mongolia has a long history of coex- by installing a network of automated weather stations istence and identification with wildlife, the current (e.g. in each soum center) to record temperature, generation of Mongolians appears to be losing that precipitation, soil moisture, and wind speed. connection. Mongolia is fast becoming an urban na- Availability of open water needs to be mapped tion. As a result, most Mongolians have little contact over several seasons and linked with local rainfall with wildlife or pastoral livestock production. While patterns to be able to predict water availability on a most Mongolians continue to place high value on landscape scale over all seasons. wildlife and traditional pastoral livestock production, Monitoring of pasture productivity and dynamics there is a general lack of awareness of the ongoing using remote sensing techniques needs to be institu- decline in large herbivore populations and traditional tionalized. pastoralism caused by pressures emanating from the developing market economy. e developing media Institutionalize a Herder/Livestock Monitoring system and communication networks create opportunities To assess the impact and efficiency of pasture develop- to use media in near-real time to address pertinent ment projects and allow for efficient EAs, the National issues. Coordinating Council for Pastureland Management should institutionalize the mapping and data collec- Media of all types, both national and international, tion of herder camps and movements piloted under the should be engaged to inform the general public about Sustainable Livelihoods Project, together with soum- issues threatening wildlife and the Gobi region. level livestock data, and maintain these in a common Television should be used to examine important issues format for easy access. affecting wildlife and herders and to reconnect urban and rural residents. Information is needed on the location of herder e khulan documentary film (produced under this camps throughout the year, including arrival and project) and other wildlife-focused media should departure dates. 43 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Information is needed on livestock numbers and tions in similar habitats in Africa, Australia, and composition on the basis of individual herders. North America and have been conducted in Mongolia Data should be collected as map coordinates and in past times (Reading et al. 2001). Presently, suitable lists at the soum level and subsequently sent to small planes are not available. Ulaanbaatar for input into a central database. Initiate aerial surveys of khulan and other large Develop Khulan/Wildlife Monitoring Systems ungulates. Monitoring--of khulan and other wildlife popula- Train pilots and selected staff from the aimags and tions--is conducted mainly on a local scale and does relevant central agencies on how to conduct and not take into account the huge distribution range or adequately document wildlife surveys. the seasonal and annual shifts in the animal's distribu- Train wildlife monitoring staff in the underlying tion. At present, no commonly agreed and statistically statistics, analysis, and visualization of wildlife sound methods are applied and no common database monitoring data. exists. Aerial census techniques have proven to be the Input the data into the Mongolian Biodiversity only reliable means to monitor large ungulate popula- Databank in the National University Department of Biology. 44 Bibliography Atlas of Mongolia. 2004. Institute of Geology, Buckland, S.T., D.R. Anderson, K.P. Burnham, Ulaanbaatar, Mongolia.[In Mongolian] J.L. Laake, D.L. Borchers, and L. omas. 2001. Introduction to Distance Sampling. Oxford, UK and AWAC. 2005. Recommendations of the First New York: Oxford University Press. Asiatic Wild Ass Conference in Hustain Nuuru, Mongolia. Exploration into the Biological Resources of CMS. 2002. Convention on Migratory Species. Mongolia (Halle/Saale, Germany) 10, 00­00. Appendix II. [Available from: http://www.cms.int/docu- ments/appendix/cms_app2.htm; last accessed September Bannikov, A.G. 1954. e Mammals of the Mongo- 2006] lian People's Republic. Moscow: Academy of Sciences. [In Russian] Davaa, G. 2003. Annual report, Institute of Meteorology and Hydrology, Ulaanbaatar. (unpub- Batima, P. 2006. "Climate Change Vulnerability lished material). In Batima, P. 2006. "Climate Change and Adaptation in the Livestock Sector of Mongolia." Vulnerability and Adaptation in the Livestock Sector Final Report submitted to AIACC. Washington, D.C.: of Mongolia." Final Report submitted to AIACC. International START Secretariat Washington, D.C.: International START Secretariat Bazargur, D., C. Shiirevadja, and B. Chinbat. Food and Agricultural Organization of the 1993. "Territorial organization of Mongolian pastoral United Nations (FAOSTAT) 2005. Online Statistics livestock husbandry in the transition to a market [Available at: http://www.fao.org/waicent/portal/statis- economy." PALD Research Report, No 1, Institute of tics_en.asp; last accessed September 2006] Development Studies. Farrington, J. D. 2005. " e Impact of Min- Bedunah, D. J., and S. M. Schmidt. 2004. "Pasto- ing Activities on Mongolia's Protected Areas: A ralism and protected area management in Mongolia's Status Report with Policy Recommendations." Gobi Gurvansaikhan National Park." Development and Integrated Environmental Assessment and Management Change 35(1):167­191. 1(3):283­289. Bouman, I., and J. Bouman. 1994. " e history of Feh, C., N. Shah, M. Rowen, R. P. Reading, and the Przewalski´s horse." In L. Boyed and K.A. Houpt, S. P. Goyal. 2002. "Status and action plan for the eds. Przewalski´s horse. Albany, New York: State Asiatic wild ass (Equus hemionus)." In P.D. Moehlman, University of New York Press. ed. Equids: Zebras, Asses and Horses. Cambridge, UK: IUCN Publication Services Unit. [Available from: http://iucn.org; last accessed September 2006] 45 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Fernandez-Gimenez, M. E. 1999. "Sustaining the Kaczensky, P., N. Enkhsaihan, O. Ganbaatar, and Steppes: A Geographical History of Pastoral Land Use C. Walzer. 2006 (in press). Identification of herder- in Mongolia." Geographical Revue 89(3):315­342. wild equid conflicts in the Great Gobi B Strictly Protected Area in SW Mongolia. Exploration into Fernandez-Gimenez, M. E., and B. Batbuyan. the Biological Resources of Mongolia (Halle/Saale, 2004. "Law and Disorder: Local Implementation of Germany) 10, 00­00. Mongolia's Land Law." Development and Change 35(1):141­165. Mearns, R. 2004. "Sustaining livelihoods on Mongolia's pastoral commons: Insights from a Heaps, T. A. A. 2004. "Canadians at the gate." participatory poverty assessment." Development and Corporate Knights, 24­29. [available at: http://www. Change 35(1):107­139. corporateknights.ca/content/page.asp?name=Mongolia; last accessed September 2006] MNE, 2003. Status and distribution of the khulan in Mongolia in 2003. Unpublished report. Ulaanbaatar, Hijmans, R. J., S.E. Cameron, J.L. Parra, P.G. Mongolia: Ministry of Nature and Environment. Jones, and A. Jarvis. 2005. "Very high resolution interpolated climate surfaces for global land areas." Nuttonson, M.Y. 1947. "Ecological crop geography International Journal of Climatology 25: 1965­1978. of China and its agro-climatic analogues." American Institute of Crop Ecology. Hooge, P. N., and B. Eichenlaub. 1997. "Animal movement extension to arcview, version 1.1." Anchor- Pratt, D. G., D. C. MacMillan, and I. J. Gordon. age, Alaska: Alaska Biological Science Center, U.S. 2004. "Local community attitudes to wildlife utilisa- Geological Survey. tion in the changing economic and social context of Mongolia." Biodiversity and Conservation 13:591­613. Ito, T. Y., N. Mura, B. Lhagvasuren, D. Enkhbileg, S. Takasuki, A. Tsunekawa, and Z. Jiang. 2005. Purev, B. 1990. "Traditional pastoral livestock "Preliminary evidence of a barrier effect of a railroad management in Mongolia." In Proceedings of the on the migration of Mongolian gazelles." Conservation International Workshop on Pastoralism and Socio-Eco- Biology 19(3):945­948. nomic Development. Rome: FAO. Ivanhoe Mines Mongolia. 2005. "Oyu Tolgoi Reading, R. P., D. J. Bedunah, and S. Am- Project, Mongolia. Integrated Development Plan." galanbaatar. 2006. "Conserving biodiversity on Executive Summary. [Available from: http://www. Mongolian rangeland: Implications for protected area ivanhoe-mines.com/i/pdf/IDP_ExecutiveSummary.pdf; development and pastoral use." USDA Forest Service last accessed September 2006] Proceedings RMRS-P-39 [Available at: http://www. fs.fed.us/rm/pubs/rmrs_p039/rmrs_p039_001_017.pdf; Japan International Cooperation Agency (JICA). last accessed September 2006] 2003 & 2004. " e study for improvement plan of livestock farming system in rural area." Interim Reading, R. P., H. M. Mix, B. Lhagvasuren, Report. Pacific Consultants International, Mitsui C. Feh, D. P. Kane, S. Dulamtseren, and S. Mineral Development Engineering Co., Ltd. Enkhbold. 2001. "Status and distribution of khulan (Equus hemionus) in Mongolia." Journal of Zoology Johnson, D.A., D.P. Sheehy, D. Miller, and D. 254:381­389. Daalkhaijav. 2006. Mongolian rangelands in transition. (In print). Secheresse. Rodgers, A.R., and A.P. Carr. 2001. "Tracking Animals with GPS." In A. Sibbald and I. J. Gordon, Kaczensky, P. and C. Walzer. 2002b-2003b. Bi-an- eds. Tracking Animals with GPS. Aberdeen, UK. nual progress reports. On-line report [Available from: http://www.takhi.org; last accessed September 2006] Scharf, K., and S. Enkhbold. 2002. "Hunting in eastern Mongolia: e challange of wildlife 46 Recommendations management in a post-socialist country." Case Study. ass (Equus hemionus)." In A.H.M. Erken and G.MM. Ulaanbaatar, Mongolia: UNDP-GEF Eastern Steppe Dorrestain, eds. Proceedings of the European Association Biodiversity Project. of Zoo and Wildlife Veterinarians (EAZWV), 5th scientific meeting, Ebeltoft, Denmark, May 19­23. Sheehy, D.P. 1996. "Sustainable livestock use of pastoral resources." In Ole Bruun and Ole Odgaard, Wingard, J.R. and P. Zahler. 2006. Silent Steppe: eds. Mongolia in Transition: Old Patterns, New Chal- e Illegal Wildlife Trade Crisis in Mongolia. Mongolia lenges. Nordic Institute of Asian Studies. pp. 42­64. Discussion Papers, East Asia and Pacific Environment and Social Development Department. Washington, Sheehy, D.P. and J. Byambadorj. 2001. "Envi- D.C.: World Bank. [Available at: www.worldbank. ronmental Assessment: Pastoral Risk Management." org/eapenvironment; last accessed September 2006] Unpublished report, Mongolia Sustainable Livelihoods Project. Washington, DC: World Bank. World Bank. 2003. From Goats to Coats: Institu- tional Reform in Mongolia's Cashmere Sector. Mongolia acker, P. D. 2004. "Mongolia's environment Discussion Papers, East Asia and Pacific Environment undermined by gold fever." Environmental Science and and Social Development Department. Washington, Technology 1:192A­193A. D.C.: World Bank. [Available at: www.worldbank. org/mongolia; last accessed September 2006] United Nations Disaster Management Team (UNDMT) and National Civil Defence and State World Bank. 2006. A Review of Environmental Emergency Commission Ulaanbaatar. 2000. DZUD and Social Impacts in the Mining Sector. Mongolia 2000-Mongolia: An evolving ecological, social and Discussion Papers, East Asia and Pacific Environment economic disaster: A rapid needs assessment report. and Social Development Department. Washington, Ulaanbaatar: United Nations Disaster Management D.C.: World Bank. [Available at: www.worldbank. Team (UNDMT) and National Civil Defence and org/eapenvironment; last accessed September 2006] State Emergency Commission Zeller, R. and E. Carmines. 1980. Measurement Van Dierendonck, M. C. and M. F. Wallis de Vr- in social sciences: e link between theory and data. ies. 1996. "Ungulate reintroductions: Experiences with Chelsea, Michigan: Cambridge University Press. the takhi or Przewalski horse (Equus ferus przewalskii) in Mongolia." Conservation Biology 10(5):728­740. Zevegmid, D. and N. Dawaa. 1973. Die seltenen Großsäuger der Mongolischen Volksrepublik und ihr Walzer, C. and P. Kaczensky. 2004. "Capture and Schutz. Archiv für Naturschutz und Landschaftsforsc- field anaesthesia of a fast runner--the Mongolian wild hung 13(2):87­106. [In German] 47 Appendix A: Field Teams Involved in Data Collection in the SW Dornogovi Study Area Team 1 A liation Dr. P. Kaczensky Department of Wildlife Ecology and Management, University of Freiburg, Germany Dr. C. Walzer Research Institute of Wildlife Ecology, University of Veterinary Medicine, Austria D. Lkhagvasuren Research Assistant at the Department of Zoology, Faculty of Biology, National University of Mongolia Ch. Myagmasuren Graduate student of ecology, National University of Mongolia B. Ulziijargal Graduate student of biology, National University of Mongolia H. Otgonbayar Undergraduate student of biology, National University of Mongolia J. Khurelbaatar Minibus driver J. Badraljargal Jeep diver Guests Dr. V. Galanti OIKOS Institute, Italy G. Gandolfo Media 99, Italy Team 2 Dr. D.P. Sheehy Executive Director, International Center for the Advancement of Pastoral Systems (ICAPS), Wallowa, USA Dr. D. E. Johnson Professor, Dept. of Rangeland Ecology and Management, Oregon State University, USA Mr. C. M. Sheehy Graduate Student, Dept. of Rangeland Management and Ecology, Oregon State University, USA Dr. M. Louhaichi Postdoctoral Student from Tunesia, Oregon State University, USA Dr. D. Tsogoo Plant Taxonomist, USAID Gobi Forage Project, Mongolia Ms. G. Udval Animal Nutritionist, USAID Gobi Forage Project, Mongolia Mr. Urgamal, Graduate Student, Mongolian Academy of Science & USAID Gobi Forage Project Mr. Byambadorj, Director, Monconsult, Ulaanbaatar, Mongolia Ms. U. Sedbazar Translator and Interpreter, Monconsult, Ulaanbaatar, Mongolia Mr. Bogey Graduate student, National Agriculture University, Mongolia Guests Mr. J. McCormack Nez Pierce Tribal Fisheries, USA Mr. M. Hale ICAPS, USA 49 Appendix B: Plant Community Composition in the Southeast Gobi ID Description Semidesert Steppe 25 Artemisia-bunchgrass, bunchgrass (Stipa spp., Cleistogenes spp., Agropyron spp.) steppes with Caragana spp. on light chestnut soils. 26 Petrophytic forbs-Artemisia-bunchgrass (Agropyron, Stipa) steppes on light chestnut and mountain chestnut soils. 27 Psammophytic and hemipsammophitic bunchgrass (Agropyron, Stipa glareosa and Stipa gobica, Cleistogenes) steppes with shrubs on light chestnut sandy loamy and sandy soils. 28 Hemihalophytic Nanophyton-Artemisia-bunchgrass, Allium-Stipa glareosa steppes on light chestnut solonetz soils and solonetzes. DESERT North Desert ( Semi-Desert) 29 Bunchgrass (Stipa gobica, Sipa glareosa) with Anabasis, Allium, Ajania, Artemisia, Nanophyton on brown desert- steppe, locally calcareous soils. 30 Petrophytic bunchgrass (Stipa gobica, Stipa glareosa) with Ajania, Salsola laricifolia, Ceratoides papposa, Caragana on brown soils, locally in combination with perennial saltworts on solonetz brown soils. 31 Psammophytic bunchgrass (Stipa gobica, Stipa glareosa) with Caragana, Ceratoides papposa, and Stipa- Cleistogenes communities on brown loose-sandy soils and sands. 32 Halophytic bunchgrass (Stipa gobica, Stipa glareousa) with perennial saltworts, Salsola passerina with Stipa and Allium; Reaumuria songarica with Stipa and Allium communities on solonetz brown soils and their complexes with solonetzes. Middle-Desert( Steppi cated Desert) 33 Anabasis brevifolia with Stipa gobica, Stipa glareosa, Allium; Nanophyton erinaceum with Stipa, Artemisia, Ajania with Stipa deserts on pale-brown locally weakly solonetz soils. 34 Petrophytic Anabasis brevifolia, Sympegma, Ajania, Salsola laricifolia with Stipa glareosa on deserts with pale- brown soils. 35 Psammophytic Artemisia with grasses, Ceratoides papposa, Caragana, Potaninia on deserts with pale-brown sandy soils. 36 Halophytic perennial saltworts with Stipa glareosa in combination with Kaldium on deserts with solonchaks and Haloxylon stands on pale solonetz-solonchak. cont. next page 51 from previous page ID Description South-Desert (True) 37 Anabasis, Nanophyton, Sympegma, Ephedra, low Haloxylon stands on grey-brown desert, locally solonetz soils, often in combination with Sympegma-Potaninia or Artemisia terrae-abbae-Ceratoides papposa communities on sands. 38 Petrophytic Anabasis, Salsola Iaricifolia, Sympegma, Amygdalus, perennial saltwort deserts on grey-brown skeleton and grey brown raw soils. 39 Psammophytic Psammochloa, Artemisia, Caragana, Potaninia, Zygophyllum deserts, high Haloxylon stands on grey-brown, locally gypsic, sandy, weakly di erentiated soils and sands. 40 Halophytic, Reamuria, Salsola passerina, Anabasis brevifolia, Brachanthemum on deserts with grey-brown solonetz soils and solonchak soils. 41 Gypsum-halophytic Nitraria, Haloxylon with Nitraria on perennial saltworts deserts on grey-brown solonchak strongly gypsic soils. Desert 56 Sedge halophytic grass (Puccinellia, Hordeum) meadows on saline meadow soils, Iris-Carex duriuscula meadows on saline soddy soils, Puccinellia-Achnatherum and Suaeda-Achnatherum meadows on meadow solonchacks and saline meadow-chestnut soils with participation of Trisetum-Carex meadows, locally with Phragmites, halophytic forb-grass, Puccinella-Achnatherum meadows on saline meadow-chestnut soils. 57 Carex duriuscula-Iris and Anchnatherum communities on saline soddy soils, halophytic grass communities on saline meadow soils in combination with: a) Artemisia frigida-Cleistogenes communities on soddy and chestnut soils, b) Allium and Leymus communities on soddy deserti cating calcareous soils. 58 Puccenilia, Calamagrostis communities on saline meadow soils, Juncus, Eleocharis-Carex communities on swampy clay mucky-gley soils, Achnatherum and Iris communities with Caragana on soddy deserti cating calcareous soils, locally with Phragmites on meadow-swampy soils in combination with poplar stands with shrubs on soddy primitive soils. 59 Combination of halophitic meadow communities (Phragmites, Carex, Achnatherum) and shrub tugals (Tamarix, Halimodendron halodendron), locally with Populus on saline meadow and meadow -desert soils. 60 Shrub (Caragana, Halimodendron, Tamarix) Achnatherum splendens communities with Artemisia and halophytic forbs locally with Stipa on soddy deserti cating calcareous soils. 61 Phragmites, Eleocharis-Phragmites communities on meadow-swampy soils in combination with: a) Elymus- Carex communities on saline swampy clay-mucky gley soils and forb-Puccinella communities with Achnatherum on saline meadow soils; b) Eleocharis-Juncus communities on swampy peaty soils, Leymus communities with Limonium and Achnatherum, locally with shrubs (Tamarix, Caragana) on saline meadow soils; c) Phragmites, Carex-Phragmites communities, locally on peaty gley soils. 62 Achnatherum communities (with Carex spp, Carex -Agropyron, Potentilla-Artemisia -Stipa krylovii, Allium-Carex -Stipa krylovii) on meadow -chestnut, locally solonetz soils. 63 Combinations of halophitic (perenial saltwort Reaumuria, Kalidum, Nitraria, Haloxylon) communities on meadow and u y solonchaks. 64 Haloxylon (Reaumuria, Nitraria) with shrubs, sometimes in combination with Tamarix tugals and psammophytic communities on primitive sair soils. (Source: UNEP Vegetation Type Maps compiled by the Russian-Mongolian Complex Ecological Study 1995). Appendix C: Survey Instrument Used to Determine Local Attitudes Toward Khulan Khulan Questionnaire for Mongolia agencies responsible for khulan management and will help to design a research program about khulan in e student should introduce himself/herself as a Mongolia. student and explain the purpose of the study. It will be essential to set up an atmosphere of trust and explain ere are no right or wrong answers in the that they will not have any negative consequences questionnaire, we rather want to hear your opinion from the interview--rather for park management it about the khulan situation. Because your opinions will be important to know their wishes/visions. are important we encourage you to answer all the questions. Your answers will be grouped together with e National University in Ulaanbaatar and the those of other respondents, and individual responses International Takhi Group (ITG) are conducting a will be kept strictly confidential. If you have any study of people's attitudes, knowledge and problems questions please feel free to ask anything during this with khulan in Mongolia. e results of the study will interview. help in the decision making process of the various No name--the questionnaire should be anonymous! O. General info about the interview situation Date of interview: . . Location of interview: Type of camp: winter camp spring camp summer camp autumn camp village How long have you lived in the area? Gender: male female Age: Number of people in household: 53 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion Occupation of interview partner: herder other: Number of livestock owned (in total): How much do you earn per month? Do you have: jeep truck motorbike gun I. Herding background (herders only) 1. For how many years have you been a herder? 2. Is livestock raising your main income? yes no 3. Who owns the livestock?: own: % other families: % company: % 4. Number of livestock and composition: sheep: goats: horse: camel: yak/cattle: 5. Has the total number of your livestock within the last 5 years: increased decreased remained the same II. Attitude towards khulan In this section we would like to learn more about how you feel about khulan in general. 1. Is having khulan in Mongolia? very good good neither good nor bad bad very bad Strongly dislike Dislike Neutral Like Strongly like 2. Do you like khulan? 1 2 3 4 5 - How do you feel about the following statements? - Strongly disagree Disagree Neutral Agree Strongly agree 3. Khulan are a necessary part of nature in Mongolia 1 2 3 4 5 4. It is important to maintain khulan for our children (future generations). 1 2 3 4 5 5. Because large khulan populations exist elsewhere, it is not necessary to have khulan in Mongolia. 1 2 3 4 5 6. It is important for me that khulan exist in Mongolia 1 2 3 4 5 7. I'm proud that Mongolia has khulan 1 2 3 4 5 54 Appendix C: Survey Instrument Used to Determine Local Attitudes Toward Khulan III. Knowledge/beliefs about khulan In this section we would like to get your judgment/estimate of the khulan situation in the Mongolia: 1. How many khulan do you think live in Mongolia? 2. Do you think khulan numbers are? increasing remain the same decreasing do not know 3. If you see khulan, how many in a group you see most often in your area: 4. What do you think has the strongest influence on khulan numbers? weather wolf poaching pasture condition don't know other: Correct Not correct Do not know 5. Khulan can run over 100 km/hour 1 2 3 6. Khulan do not live in family groups (harems) like takhis 1 2 3 7. Khulan mares often have 2 foals 1 2 3 8. Khulan need to drink only once a week 1 2 3 9. Khulan live in all parts of Mongolia 1 2 3 10. Only small numbers of khulan live outside of Mongolia and China 1 2 3 11. e Gobi is the most important habitat for khulan in Mongolia 1 2 3 12. Snow leopards are important predators of khulan 1 2 3 13. Khulan are fully protected throughout Mongolia 1 2 3 Very much Much Some what Not much Not at all 14. How much do you know about khulan? 1 2 3 4 5 15. How interested are you in learning more about khulan? 1 2 3 4 5 16. How important is the topic of khulan manage- ment for you? 1 2 3 4 5 17. How important is the topic of nature protection for you? 1 2 3 4 5 18. From where did you get your knowledge about khulan? own observation family school books newspaper radio TV other: 55 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion IV. Attitudes toward khulan management in the Gobi regions In this section we would like to get your opinion about khulan management in the Mongolia. Strongly disagree Disagree Neutral Agree Strongly agree 1. ere are too many khulan in the Mongolia 1 2 3 4 5 2. ere are too many khulan in your home area 1 2 3 4 5 3. Khulan are important for the ecosystem 1 2 3 4 5 4. Khulan should be managed so that local people benefit from them. 1 2 3 4 5 5. Khulan eat mostly plants that are of no use for livestock 1 2 3 4 5 6. Khulan expand into many new areas 1 2 3 4 5 7. Herder presence at water points makes it difficult for khulan to drink 1 2 3 4 5 8. Khulan often deplete the winter pastures for livestock 1 2 3 4 5 9. Khulan cause a lot of damage to agricultural crops 1 2 3 4 5 10. If there are khulan, wolves kill less livestock 1 2 3 4 5 11. It is great fun to chase khulan by jeep 1 2 3 4 5 12. Khulan have no value for people 1 2 3 4 5 13. Khulan attract tourists to the Gobi regions 1 2 3 4 5 14. I believe there would be a market for khulan meat 1 2 3 4 5 15. I would not like to eat khulan meat 1 2 3 4 5 56 Appendix C: Survey Instrument Used to Determine Local Attitudes Toward Khulan In this section we would like to get your opinion what management options for khulan in Mongolia you would support. Strongly disagree Disagree Neutral Agree Strongly agree 1. Khulan should not be hunted at all in Mongolia 1 2 3 4 5 2. Khulan should only be protected in nature reserves 1 2 3 4 5 3. Khulan numbers should be increased 1 2 3 4 5 4. Khulan numbers should be decreased 1 2 3 4 5 5. e number of nature reserves should be increased 1 2 3 4 5 6. e number of nature reserves should be decreased 1 2 3 4 5 7. Local people should be allowed to hunt khulan outside of nature reserves 1 2 3 4 5 8. Local people should be allowed to hunt khulan within nature reserves 1 2 3 4 5 9. Wherever local people are allowed to hunt, anyone should be allowed to hunt khulan 1 2 3 4 5 10. Hunting should not reduce khulan too much 1 2 3 4 5 11. For hunting, it should not be allowed to chase khulan by jeep 1 2 3 4 5 12. Khulan should not be hunted at water points 1 2 3 4 5 13. Mares with small foals should be always protected 1 2 3 4 5 14. Khulan should only be hunted when the meat is used (eaten or sold) 1 2 3 4 5 15. Mongolia should allow khulan meat on domestic markets 1 2 3 4 5 17. ere should be stricter rules to protect wildlife in Mongolia 1 2 3 4 5 18. More rangers are needed to protect wildlife 1 2 3 4 5 19. ere are already too many rangers 1 2 3 4 5 20. I would be more interested in helping to protect khulan if I would be allowed to hunt them 1 2 3 4 5 57 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion 21. Some water points in protected areas should be reserved for wildlife only 1 2 3 4 5 22. In protected areas people should not be allowed to set up a camp right next to a waterpoint 1 2 3 4 5 23. Khulan should be brought back to areas where they have lived in the past (e.g Eastern steppe) 1 2 3 4 5 Khulan are presently fully protected by law in and outside of nature reserves. However, some people go and hunt khulan, which under the present legal status is an illegal action. Let's assume the ministry decides to allow a certain harvest of khulan--e.g. in your area, local people were allowed to harvest a certain number of khulan each fall. e number that can be harvested would depend on the number of khulan in the area; that is, the bigger the herd of khulan; the more khulan local people would be allowed to hunt--a situation similar to a flock of sheep. If the present status of full protection would be changed to allow a certain use by local people, how likely do you think that people would support the following statements. For sure not Probably not Maybe Probably yes Certainly 1. People would stop hunting khulan at other times of the year 1 2 3 4 5 2. People would be careful so that khulan numbers do not decline 1 2 3 4 5 3. People would not hunt more khulan than the allowed quota 1 2 3 4 5 4. People would be willing to help with counting khulan at water points 1 2 3 4 5 5. People would write down the location of khulan carcasses 1 2 3 4 5 6. People would report about local people poaching khulan (e.g. hunting in spring or more than allowed) 1 2 3 4 5 7. People would report about people from outside poaching khulan 1 2 3 4 5 58 Appendix C: Survey Instrument Used to Determine Local Attitudes Toward Khulan In your opinion, what are the most important issues about khulan in Mongolia? VII. Own experience with khulan In this section, we would like to get some information on khulan presence and distribution in the Gobi region. 1. How often do you see khulan per year? never 1-2 times seldom (3-10 times) often (>10 times) very often (>100 times) 2. How were most of your experiences with khulan? very positive positive neutral negative very negative What was the reason: 4. Are you a hunter? yes, I hunt I no longer hunt I never hunt 3. Have you ever hunted khulan? yes no Why? ank you very much!!! 59 Appendix D: Interview Guidelines for the Six Semi- structured Interviews on Wildlife Values Guidelines for the Interview excited or happy? You can emphasize as many cases as you can. No need to mention year by year. You can Introduction: Hello. My name is .............. I am start from the case you remember. And please tell me a student of Mongolian National University. Our all of those after thinking well. I will not make you University is implementing a project on studying hurry or interrupt your conversation. relationship between human and animals in the cooperation with international experts of USA and If you need to clarify during the conversation European countries. In the project it is aimed to know please use below questions: what kind of beliefs people of different countries have about animals and what they like or dislike about the Can you please tell me about this more in details? wild animals. A wild animal is not a pet. Do you have any other samples as same as this? erefore, we would like ask you to tell us from If that person doesn't have such experiences to your experiences and examples about this. It will be share, you can ask below questions: interesting if you share with us about your story or experience related to animals and your perception Well...What kind of occurrences related to animal about the animals. en, fill this questionnaire please. can make you feel exited or happy? What do you It will take approximately 10 minutes. think? How do you imagine? (Variant 1 for the question) To get above information I will ask from you 5 basic questions. Every question is divided into sub If he/she doesn't have definite answer you can ask: questions. It will help us a lot if you spend time to answer. ank you for your voluntary participation Do you remember such occurrences happened to in this questionnaire. We will be confidential about someone? Have you heard such occurrences from your answers. We will not transfer your information to someone else before? (Variant 2 for the question) others. We will use your answers only for the survey. Would you mind if we record the interview in a tape. If it doesn't work you can tell: We will also make brief notes. You can mention the occurrences that you have /....record..../ seen on TV or read from book. (Variant 3) Why did you feel exited and happy in that Question 1. Do you have any experiences, cases or situation? occurrences related to animal that make you feel 61 Question 2: Do you have any occurrences related to If it doesn't work you can tell: animal that make you feel sad or distressed? You can mention as many as you can. You can mention the occurrences that you have seen on TV or read from book. (Variant 3) Can you clarify that please? Why did you feel upset in that situation? Do you have any examples as same as this? Question 4: Do you have any occurrences related If that person doesn't have such experiences to to animal that make you feel frightened? You can share, you can ask below questions: mention as many examples as you can. Well...What kind of occurrences related to animal If you need to clarify during the conversation can make you feel upset? What do you think? please use below questions: How do you imagine? (Variant 1 for the question) Do you have any other samples as same as this? If he/she doesn't have definite answer you can ask: If that person doesn't have such experiences to Do you remember such occurrences happened to share, you can ask below questions: someone? Have you heard such occurrences from someone else before? (Variant 2 for the question) Well...What kind of occurrences related to animal can make you feel frightened? What do you If it doesn't work you can tell: think? How do you imagine? (Variant 1 for the question) You can mention the occurrences that you have seen on TV or read from book. (Variant 3) If he/she doesn't have definite answer you can ask: Why did you feel sad or distressed in that situa- tion? Do you remember such occurrences happened to someone? Have you heard such occurrences from Question 3: Do you have any occurrences related to someone else before? (Variant 2 for the question) animal that make you feel upset? You can mention as many examples as you can. If it doesn't work you can tell: Can you clarify that please? You can mention the occurrences that you have Do you have any other examples as same as this? seen on TV or read from book. (Variant 3) Why did you feel frightened in that situation? If that person doesn't have such experiences to share, you can ask below questions: Question 5: Tell me what you think about animals and how you treat them please. Well...What kind of occurrences related to animal can make you feel sad or distressed? What do you I have no more questions now. Do you have think? How do you imagine? (Variant 1 for the something to add? anks for your time for the question) interview. It will be very helpful if you fill this questionnaire now. is questionnaire also reflects If he/she doesn't have definite answer you can ask: about animals. You can fill it alone if you want or if you have something to clarify from me we can discuss Do you remember such occurrences happened to and fill it together. Please don't hurry while filling this someone? Have you heard such occurrences from questionnaire. someone else before? (Variant 2 for the question) Appendix E: Daily Protocol of Herder #7 in the SW Gobi (08/19/2005 to 09/13/2005) Translation by Enkhjargal Darambaza traveled approximately 16 km we've seen two separate groups of khulan with fewer khulan in each. Couldn't 08.19.05 Tuesday get their numbers though. Today going for my sheep herding I took the equip- ment in the backpack for the first time to the range- 08.26.05 Tuesday land. It was very hot day. Four khulan were sighted. e place we moved into seems to have plenty of water sources like ponds and streams. So khulan could be 08.20.05 Wednesday seen quite a lot, almost everywhere. Again couldn't It was a hot day, too. Saw two khulan while I was count them though. It seems there may be over a walking with the herd. hundred of them around here. 08.21.05 ursday 08.27.05 Wednesday e sky was clear and sunny. It was hot, although has Sunny and clear sky and a pretty hot day. Khulan become cloudy in the afternoon. herds were seen quite a bit and located pretty close to each other. Tried to count their number. ere were 08.22.05 Friday around 40-43 khulan. Found khulan pellets (fresh), Since the afternoon it was raining, so I didn't go much put them in a sunny spot to get them dried, marked far with the sheep. No khulan was sighted. the spot with a small pile of rocks. 08.23.05 Saturday 08.28.05 ursday It was sunny earlier in the morning. Saw three khulan. e day has been little breezy but still hot. Some But clouds came up in the afternoon and there was a khulan could be seen on the hilltops or around brief rain shower which lasted about 25 min and now in separate groups. It was not possible to get their it has become quite wet everywhere. ink khulan numbers. But anyway, there may have been somewhere movement may benefit from this. around 50-60 of them. Mongol horses have been spotted occasionally, their pellets were found and left 08.24.05 Sunday to dry. Today was a very hot day with little clouds. No khulan was sighted. 08.29.05 Friday No khulan was seen today, although their screaming 08.25.05 Monday noise could be heard from distance. Apparently they Today we are moving out of the place near Khaliv were at the pond having water. Bagg into a place called West Toirom. While we have 63 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion 08.30.05 Saturday 09.06.05 Tuesday Haven't seen any khulan today. Weather has been bad: I still keep trying to identify the pellets of khulan foggy and therefore poor visibility. and mongol horses. Collected some fresh and quite wet pellets, crushed them pressing in between of two 08.31.05 Sunday rocks, and looked inside. Some were quite wet with Today I felt tired, so stayed at home. A kid sent to liquid coming out when squeezed, some were not so herd the sheep and was carrying the equipment, but much. Still was not successful to recognize them, when the kid got back home didn't say anything about although it seems they may be differentiated by the khulan, I didn't ask, either. shapes. 09.01.05 ursday 09.07.05 Wednesday Today there was a clear sky and really hot. ere were As autumn is coming, water is getting scarce around some errors in putting days for the previous dates, so here, at least around our camp location and may be I made corrections to them. Quite a few khulan were because of that khulan have been sighted fewer and seen running along the hills. Tried to count them; it fewer. en I saw two khulan first, and then four, in seems more than 110 of them there. Didn't go much total six of them for today. far away this afternoon; nevertheless, 17 more khulan were sighted. 09.08.05 ursday Again tried to see how khulan and mongol horses 09.02.05 Friday pellets are different; crushed and split them. Again It was much cooler day today, so I was able to go couldn't see any difference. Anyway, it seems that I pretty farther than usual but was not able to sight any usually travel from 13 to 18 km a day herding my khulan. No khulan pellets was found either. I was sheep, I estimated. thinking maybe this equipment can detect some signal even if khulan may have escaped somewhere, because 09.09.05 Friday it was making some buzzing noise in the backpack. Made an attempt to separate sheep pellets from those Hope it doesn't read mind. of goats. Not much difference visually, it seems. But goats prefer plant tops flowers leaves and seeds more 09.03.05 Saturday than sheep. erefore, composition of the pellets inside Distant visibility was poor because of the foggy and might be a bit different, I guess. Today I traveled windy condition. Haven't sighted khulan yet, but fresh farther than I had the previous days and as a result saw traces could be seen quite often. ey might have been 13 khulan. run away from me, since khulan don't usually avoid livestock on rangelands. 09.10.05 Saturday Didn't see any khulan today. Although I saw some 09.04.05 Sunday signs of their fighting with each other; a khulan's tail Although I didn't see khulan today, I could hear cut off by its end was laying on the ground. Looks like clearly that they have come to the lake by their scream- these animals have got very aggressive behavior. ing and fighting with each other noises. 09.11.05 Sunday 09.05.05 Monday For days I have been trying to identify khulan and Today I brought my sheep to the lake for water. Tried mongol horse pellets. Asked about it the people I've to collect khulan pellets but couldn't identify them. seen on the way, but still didn't find anything that Collected few bunches tried to separate them from could differentiate them. mongol horse pellets but wasn't successful. So I think maybe khulan and mongol horses eat very much 09.12.05 Monday similar diet when there is lush vegetation which is the Collected a sample from camel feces. e feces were of case for this rangeland around the lake area. pretty liquid consistency. Probably camels that are on a green, lush rangeland and gaining weight, have got this liquid kind of feces. Meanwhile, during the heat 64 Appendix E: Daily Protocol of Herder #7 in the SW Gobi (08/19/2005 to 09/13/2005) of a day and when flies come out, goats seem to move 09.13.05 Tuesday grazing along the wind direction following after each Today I brought the sheep herd pretty far hoping to other and/or they lay down near water or ponds and get khulan sight. But failed to see any khulan for the would get up to graze only after it cools down in the last couple of days. is is probably due to that the mornings and evenings. e forage plants that camels nearest water source--ponds and streams--have been like to graze and seem to gain weight from them shrinking or dried out, and khulan maybe moved would be: Anabasis spp., Berberis sibirica, Reaumuria elsewhere relocating to the other ranges. So, from this soongorica, Salsola passerine, Caragana pygmaea, Allium day on I decided to quit chasing khulan, although I polyrrhizum, Allium mongolicum, Bassia spp., Artemisia will be carrying that backpack with me anyway. spp., and Artemisia anethifolia. ese are the most nutritionally valued forages in Gobi desert rangelands. 65 Appendix F: Selected Statements from Six In-depth Interviews About Human-Wildlife Relationships (1) 48-year old woman--former herder, now works encourage those people. We have many of such people for public service in a soum center: who can talk about earlier times. Especially elders, they talk that those animals were abundant before but ...Well, of course it is right that wildlife should today's young people are scaring and killing them. exist and stay together with people. It is a natural law. So it would be good if we use such people who have a From the early time, wildlife was together with people. heart towards wildlife and who live in the area, in the Well, many years ago, the time was good. We used to conservation work... have many livestock and nature was in good condition. But now the natural situation has totally changed... ...Broad awareness of nature conservation is needed for local people. Only herders know the ...People's negative treatment to nature and wildlife in the mountains. To protect or to kill should poverty all these issues are reasons to the change in the be under the herders' responsibility and management. behavior of wildlife. When I was a child, time used In that case, if someone from outside the area come to be very nice. All things were in good condition. and tries to kill the wildlife, local people would resist ere used to be rich green vegetation by mid May them and protect the wildlife. erefore it would be and livestock already gained weight by that time. right to talk to local people and cooperate with them. But today, almost July is coming, yet we have had no For example, you can research what families live near rain. So, I think as people miss that nice time, so do what kind of wildlife and talk to the families... wildlife... (2) 62-year old man--retired herder: ...Now, because of poverty and natural disaster, number of wildlife extremely decreased. People who ...In the earlier times, hunters hunted with horse lost their livestock in the disaster, I heard that (it is and camel. But today, people are hunting with vehicles not my experience) they kill or run after or scare the and machines!... animals like khulan, gazelle that live in the low areas. Not many such cases for mountain animals. So, I don't ...Khulan destroy the nature a lot. ey really know how we need to combat it and protect wildlife. I influence negatively on the yielding of vegetation. think local people, those who live in the area and the eir growth is actually slow. ey give a birth to pasture should protect... young ones not often... ... ere are many people who think and wish ...Today, number of khulan is increasing and that wildlife live together with livestock and worried people's livelihood is deteriorating. It seems that people about their leave and decrease. It would be good if we who are poor sometimes use khulan for their own 67 Mongolia--Room to Roam? e reat to Khulan (Wild Ass) from Human Intrusion use. And the state is trying to control this through under the name of `horse meat'. And there is a rumor a law and regulation and government resolution was that many of them loaded on the train wagon to made on this, I heard it recently. It would be good Ulaanbaatar. Khulan meat from our area, supplies the if a government, MNE takes an organized measure needs of many of the canteens in Ulaanbaatar that sell on hunting khulan and give permissions to local `huushuur'. I believe it's true. So I feel very sad about people to hunt some number of wild khulan to use it. Even though a khulan is considered as rare animal for their livelihood.--Well, in the end, the state and and written in the Red Book, nowadays, herders say government knows it. Khulan is animal in the Red that they destroy the pasture and especially in today's book.--But nowadays argali (wild sheep) is extremely unfavorable natural condition, khulan is known as the decreasing. Argali is a nice animal. Also ibex became main animal that destroy the pasture. Well, maybe very rare... it's true. Khulan are animals that go in big herds. Especially, today many of them are in the south west ..In earlier times, we used to kill the injured ones part of our aimag. Sometimes few of them can be seen first. We used to tell that it is not good to leave a in the west and east soums. So, there is a need of state gazelle injured, it closes the hunting blessing. Today, regulation on this issue. If it is really true that they are there is no such kind of beliefs and rules. Today, increased in number, then it is okay to hunt them in people just let them leave with bleeding and injured. certain number for food. Maybe they were very rare If they shoot baby gazelles or female gazelles or weak in the time of being included in the Red Book. And ones, they will just leave them. is is very wrong. In maybe they increased in number since then. If so, then earlier times, we used to get whatever is shot down by they can be excluded from the Red Book list. Today, the bullet whether it is to be a baby or an adult. And even though it is considered as Red Book animal, we used everything from the gazelle. Even we put the many of them are killed and eaten. So the state should blood into the intestine and eat... do something about this. e state should research whether it is really increased in number or if they are (3) 55-year old man--working in public service at a still need to be strictly protected then, the state should soum center: take a policy that ensures their peace and safety. Otherwise, they cannot be left behind as it is today... ...For example, when you are traveling in Gobi steppe and you see a camel, which is a pride of the ... Environmental inspectors who are in charge of Gobi people or khulan, or gazelle herds, you feel very this matter are not really productive in their work and nice, don't you? e nature can be beautiful itself, don't do much on public awareness, inspection and but it's the wildlife that makes it more beautiful and control. Nothing such kind of activities today! ey lively and people can see it and feel happy. People feel don't really do their basic job telling the excuses like proud of their country and talk the beauty of nature in no car and no fuel. So basically, nature is left with no relation to the existence of wildlife... owners and caretakers. And it encourages poachers and bad people to enjoy their freedom on nature with ...So, wildlife purely depends on nature and an idea that they can do and kill anything and make natural survival. Besides, many of them die due to a profit out of it. at kind of attitude is there today human negative influence. ey are extremely reduced in our aimag. erefore, there is a necessity that the in number. For example, black tailed gazelles. ey top (state) should do something about this through were abundant in earlier times. But today, you can control, policy and manpower... hardly see them. Now in this market system, people treat to wildlife (even to their livestock) with violence ...But in today's market economy society, there is and without love. erefore, number of wildlife is a harsh or unmerciful law that someone has to make decreasing significantly. Because of degrading living a living with what he/she can find- If you can make a standards and high price of goods and high expenses living, then you'll live and if you can't, then nobody to prepare food, people tend to kill more wildlife will help you! Such a harsh law that you yourself have and use its meat for their food. In recent years, in to work in order to live...And some kill and steal our aimag, besides killing the gazelles, people are people's livestock and sell. Every one is thinking of killing khulan a lot. ey slaughter them and sell it some ways to make a living. In such wavy situation, 68 Appendix F: Selected Statements from Six In-depth Interviews About Human-Wildlife Relationships wildlife is of course under the great threat and risk. If (5) 75-year old women--retired Ulaanbaatar the condition continues this way, then, maybe one day resident: we will be left with the photos of those animals once were abundant before... ...Wildlife is rare in our Gobi area. We had a lot of livestock from Negdel in earlier time. Nature was nice (4) 69-year old man--herder: at that time. But now, we face drought and it's very difficult... ...We used only horse and camel in early times. Unfortunately, winter disaster has been around since ...Well, we used to have many gazelles. But now 1997- 1998 and the white gazelles that used to be people say they migrate to other better areas. We used a lot grazing here have all disappeared. Now, you to have many of them grazing around... can hardly see them. Well, occasionally you can see khulan--two or three of them. Manlai area has a lot ...I don't know about them [referring to khulan] of khulan. And there were khulan in Zalaat area. But much. I just heard from people that they were grazing now, people from there are saying that there is no here and there occasionally. And I saw them on TV. grass... Otherwise, I even haven't seen them with my eyes in reality... ... ere are many wolves. No wildlife, no gazelles in our area today. Well, when it rains, many white ...Also, on TV, I saw Przewalski's horse running gazelles come to our area... around. Sometimes a view that a snow-leopard chasing after them. Only saw them on TV. I have never ...We really need to increase number of wildlife. traveled far. I lived in a city since the age of 23. When Generally, they have gone away and disappeared I was in a city, I used to go only to resort places... because of winter disaster and drought. Frankly saying, people chase them with motorbike and car and (6) 37-year old man--unemployed Ulaanbaatar kill them. So wildlife is afraid and move further away. resident, University degree: Now, in Mongolia, black-tailed gazelles are almost reduced to the number to be written in the Red Book, ...I am a young person so I don't have much life white gazelles are very rare, ground squirrel gets also experiences with wildlife. But I would like to share rare... about nature. I grew up in countryside so I can share my little knowledge about what I know about ...We used to have very rare white gazelles. But wildlife... for some time, we had many of them from Dornod aimag and Inner Mongolia. We exactly don't know ...Wildlife is getting rare these years. For instance, where they are from. Many of them! In the morning marmots. ey were very abundant before. And that steppe looks yellow. You can see them in herds by gazelles as well. Well, gazelles are said to be animals herds. In the morning, I used to look through bin- that don't have permanent habitat. Our area and oculars. But now, none of them is here. Well, time has environment also degraded and have little grass. So changed. Wildlife has finished. We need to love them maybe because of change in nature, they are moving and protect them. We kill them a lot. I would like to around. I think wildlife is connected to each other... tell that we can kill a wolf but not other wildlife. If there is a reporter, I would like to send a message to ...But I would like to say to young people that the state authorities that please protect wildlife... let us love nature and not kill wildlife. Please love wildlife... 69 Environment and Social Development East Asia and Pacific Region THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, USA Telephone: 202 473 1000 Facsimile: 202 522 1666 E-mail: worldbank.org/eapenvironment worldbank.org/eapsocial