Environmental Management

, 48:781

Trends of Forest Dynamics in Tiger Landscapes Across Asia

Authors

    • Department of GeographyUniversity of Florida
    • Department of Environmental ConservationUniversity of Massachusetts
  • Harini Nagendra
    • Ashoka Trust for Research in Ecology and the Environment
    • Centre for the Study of Institutions, Population and Environmental ChangeIndiana University
Article

DOI: 10.1007/s00267-011-9720-6

Cite this article as:
Mondal, P. & Nagendra, H. Environmental Management (2011) 48: 781. doi:10.1007/s00267-011-9720-6

Abstract

Protected areas (PAs) are cornerstones of biodiversity conservation, but small parks alone cannot support wide-ranging species, such as the tiger. Hence, forest dynamics in the surrounding landscapes of PAs are also important to tiger conservation. Tiger landscapes often support considerable human population in proximity of the PA, sometimes within the core itself, and thus are subject to various land use activities (such as agricultural expansion and road development) driving habitat loss and fragmentation. We synthesize information from 27 journal articles in 24 tiger landscapes to assess forest-cover dynamics in tiger-range countries. Although 29% of the PAs considered in this study have negligible change in overall forest cover, approximately 71% are undergoing deforestation and fragmentation. Approximately 58% of the total case studies have human settlements within the core area. Most changes—including agricultural expansion, plantation, and farming (52%), fuelwood and fodder collection (43%), logging (38%), grazing (38%), and tourism and development (10%)—can be attributed to human impacts largely linked to the nature of the management regime. This study highlights the need for incorporating new perspectives, ideas, and lessons learned locally and across borders into management plans to ensure tiger conservation in landscapes dominated by human activities. Given the increasing isolation of most parks due to agricultural, infrastructural, and commercial developments at the periphery, it is imperative to conduct planning and evaluation at the landscape level, as well as incorporate multiple actors and institutions in planning, instead of focusing solely on conservation within the PAs as is currently the case in most tiger parks.

Keywords

AsiaEcological studyForest dynamicsMeta-analysisRemote-sensingTiger landscape

Introduction

The geographical range of wild tigers has been greatly decreased during the last 50 years, with these large carnivores currently occupying a mere 7% of their historical range (Dinerstein and others 2007). Although tigers are still found in a wide variety of forest types, including the coniferous–deciduous forests of eastern Russia, the tall-grass habitats south of the Himalayas, the swamps and mangrove forests of Sundarbans, the moist/dry deciduous forests of Central Indian Highlands, and the tropical rainforests of Sumatra and Malaysia, reports of human–tiger conflicts are not uncommon in these heavily human-altered landscapes. One of the conservation strategies to save the wild tigers is to create tiger landscapes that are anchored on prey-rich protected areas (PAs), i.e., national parks, tiger reserves, and wildlife sanctuaries, and are linked to greater ecological systems through habitat corridors (Karanth and Stith 1999). However, in human-dominated landscapes, this strategy requires direct or indirect involvement of the local inhabitants to be effective. PAs are often surrounded by competing land uses and have been traditional sources of livelihoods for rural people. Hence, trends in forest dynamics in a protected landscape could be used to understand the prospects for sustainable human–nature coexistence, which is a prerequisite for long-term and effective tiger conservation. Here we review 24 case studies from 7 countries to summarize current knowledge on forest dynamics in and around tiger PAs. We also synthesize additional published works to better understand the multidirectionality of changes and differing socioecological settings at which the drivers act.

Forest dynamics are often used to measure PA effectiveness (DeFries and others 2005; Naughton-Treves and others 2005; Hayes 2006; Andam and others 2008; Nagendra 2008; Gaveau and others 2009). Avoiding forest-cover loss cannot be the ultimate test for PA effectiveness because components of biodiversity can be significantly compromised by other threats. For example, commercial hunting and/or poaching can significantly decrease PA effectiveness (Redford 1992), as can non–timber forest product harvesting and grazing in an otherwise ecologically sustainable PA landscape (Nagendra and others 2010). Nonetheless, intact or regenerated forest cover is a useful proxy for conservation success, probably because forest dynamics can be linked to many factors relevant to conservation evaluations. Forest composition, quality, and structure provide a fair representation of wild herbivore distribution because they show preferences for certain forest types (Kuijper and others 2009; Millington and others 2010). In addition, forest degradation through livestock grazing suggests community dependence on the forest products, thus indicating lack of alternative means for sustainable livelihood. Intense resource competition between grazing livestock and wild herbivores might result in shrinking prey base (Madhusudan 2004), which can have deleterious effects on tiger population (Karanth and others 2004). It is also possible to measure social sustainability of a conservation effort through documenting park–people relations, which is likely to reflect in the anthropogenic forest disturbance (e.g., destruction of forest, deliberate fire) of the region.

Forest cover, along with substantial prey occurrence and water, are the primary requirements for supporting viable tiger population and their prey base (Sunquist and Sunquist 2002). Although tigers can thrive in a wide variety of ecosystems, they have been reported to exhibit strong habitat preferences, at least for some of their activities. For example, studies have shown tiger preferences for better-covered mangrove woodlands and less-disturbed grasslands in Sundarbans, Bangladesh (Khan and Chivers 2007), less open moist-deciduous forest in Nagarahole, India (Karanth and Sunquist 2000), less-degraded or primary forest in Indonesia (Linkie and others 2008a), and mixed deciduous open forest and mixed tall grasslands in Nepal (Smith and others 1998). Evidently, tigers seem to prefer undisturbed, less-degraded forested habitat for better cover and greater prey density. However, studies have suggested the potential of selectively logged forests of Malaysia to support viable tiger populations when connected to a larger habitat patch (Rayan and Mohamad 2009) and documented potential benefits for tigers from forest-type conversion in Russia (Cushman and Wallin 2000). Nonetheless, lack of habitat corridors between isolated forest patches (or between two PAs), such as human settlements, extensive agricultural fields, and other land uses, breaks the continuity of genetic exchanges between two isolated populations and adversely affects demographic viability (Carroll and Miquelle 2006; Linkie and others 2006), eventually causing biodiversity loss (Heywood 1995). Most importantly, forested habitats are essential to support prey abundance and distribution because prey depletion is often considered the most important factor driving decline of tigers throughout their range (Karanth and Stith 1999). One of the reasons behind the extinction of the Javan tiger is the conversion of the forested habitats of the island to teak during the 1900s, resulting in rapid declination of the prey base and leading to the eventual extinction of Javan tigers (Seidensticker 1987). Hence, information on current status of forest-cover dynamics is an important component, not only in designing movement corridors and mixed-use transition zones surrounding PAs, but also in making timely decisions to protect forest-dependent prey species.

Identifying trends in forest dynamics requires time-series analysis of spatial patterns of forest cover. Remote-sensing provides a particularly useful tool in landscape-level spatial analysis by providing satellite snapshots at multiple spatial and temporal scales (DeFries and others 2005; Nagendra 2008). Satellite image analysis is probably the most widely used technique in global-level studies analyzing transformations in the biophysical and ecological attributes of the Earth’s surface (DeFries and others 2005; Nepstad and others 2006). However, such analyses are sometimes considered to be less capable of capturing degradation (i.e., changes within a single land cover), which also is an important indicator of forest dynamics. Field-based ecological study can be particularly useful in such cases to identify changes in forest structure, composition, and diversity (Kumar and Shahabuddin 2005; Yadav and Gupta 2006; Mehta and others 2008). Such studies are expensive and time-consuming compared to remote-sensing studies; however, these methods can complement each other to provide perspectives on both forest conversion and modification (Nagendra and others 2010).

Much attention has been given at national and international levels to establish a common platform for tiger conservation strategies in the 13 tiger-range countries (Global Tiger Initiative 2010). These 13 countries, i.e., Bangladesh, Bhutan, Cambodia, China, India, Indonesia, Laos, Malaysia, Myanmar, Nepal, Russia, Thailand, and Vietnam, represent a wide spectrum of socioecological factors that impact landscape dynamics. To propose feasible conservation strategies, the socioecological factors governing spatial landscape patterns must be understood and taken into account. One way to achieve this goal is by looking at landscape dynamics and drivers, especially within and outside PAs.

In this study, we synthesize results from published quantitative or qualitative data on forest dynamics from 24 PAs in 7 tiger-range countries (Fig. 1). The trends of forest dynamics were identified as temporal changes in forest covers, composition, and density, mostly within the entire landscape in which the PA is embedded, with occasional exclusion of neighboring landscapes. Additional data from these 7 countries were used to understand the nature of the institutional regime and the proximate factors governing landscape dynamics. This study uses both landscape- and species-level published data on forest dynamics in tiger landscapes to identify the dominant trends and most frequently discussed drivers of change in these landscapes, which can further be used to locate priority landscapes for habitat restoration. We expect to see a wide spectrum of forest-cover dynamics from deforestation to reforestation. This expectation is consistent with the wide variety in our database in terms of population density, location of the PA, and socioeconomic structure of the society.
https://static-content.springer.com/image/art%3A10.1007%2Fs00267-011-9720-6/MediaObjects/267_2011_9720_Fig1_HTML.gif
Fig. 1

Spatial distribution of the case studies within tiger-range countries

Methods

Two academic databases—Web of Science and Google Scholar—were used to locate peer-reviewed articles containing quantitative information about forest cover within and around PAs (national parks, tiger reserves, or wildlife sanctuaries) with reported tiger populations in 13 tiger countries. We did not include PAs that do not host wild tiger populations even though they are located within these countries. We considered the articles for analysis only if they used remote-sensing techniques or field-based ecological methods to report spatial or temporal changes in habitat quality or forest cover in general. Approximately 69% of the remote-sensing studies reported forest-cover change during decades. Other studies, however, only provided single-date measurements and/or maps of forest cover and used either satellite images or various field-based methods. These studies were included in our analysis only if they provided important information on spatial pattern of forest cover, composition, and density, e.g., within or outside PA boundaries. Information was primarily selected from 27 journal articles from 7 countries: Bangladesh, China, India, Indonesia, Myanmar, Nepal, and Russia. Although a considerable amount of research has been conducted in the rest of the tiger-range countries, most of them report tiger population density (commonly using camera traps) and do not provide information on changes in forest cover or habitat; hence, these reports were not included in this study.

Results

Summary Characteristics of the Case Studies

A total of 24 case studies were found that met all of the criteria for inclusion in this meta-analysis. A majority (n = 13) of the case studies are from India (approximately 54%), with 3 each from China and Indonesia (approximately 12.5%), 2 from Nepal (approximately 8%), and 1 each from Bangladesh, Myanmar, and Russia. Since socioeconomic factors often vary from one country to the other, there would be some bias in the findings due to overrepresentation of the Indian case studies. However, this also reflects the great number of tiger reserves in India and emphasizes the need for such research in the tiger-range countries not represented in our data set.

The size of the study area ranges from <500 to >2000 km2 (Fig. 2a), with the largest covering approximately 7000 km2. The institutional management regime for 79% of the cases is centralized, with PAs under government management except for those in China and Nepal. In China, collective forests total approximately 60% of the total forest area, and provide legal resource extraction and management rights to the communities, despite being government-controlled (Miao and West 2004). In Nepal, communities are partially involved in buffer-zone forests in some PAs (Nagendra and others 2005). Approximately 67% of the case studies mentioned participatory management or the lack thereof (Fig. 2b). Of the Indian case studies, 69% participated in ecodevelopment projects, which were expected to be participatory efforts to decrease extraction and grazing pressure from the core as well as the surrounding landscape by decreasing forest dependence for fuelwood (through supply of biogas) and controlled grazing on marked pasture. These efforts had mixed results in meeting social or ecological goals and are considered by many scholars to have been a “failure” to introduce a participatory management regime in inhabited landscapes (Shahabuddin 2010).
https://static-content.springer.com/image/art%3A10.1007%2Fs00267-011-9720-6/MediaObjects/267_2011_9720_Fig2_HTML.gif
Fig. 2

Summary characteristics of the case studies. a Size of the study area (n = 24), b participatory management (n = 16), c human settlements inside the PA (n = 14), d human settlements outside the PA (n = 17)

Approximately 58% of the case studies have human settlements within the core of the PA itself, with the number of settlements varying from 1 (Manas, Pench in India) to 27 (Sariska in India) (Fig. 2c). Approximately 69% of the Indian PAs host forest villages/settlements within the core, out of which 44% PAs have either relocated most of the forest villages or are in the process of relocation (Karanth and others 2006; Shahabuddin and others 2007; Mondal and Southworth 2010a; Nagendra and others 2010). However, the number of villages/settlements in the surrounding landscape (mostly within 10 km) is much greater (Fig. 2d). Approximately 35% of the case studies that reported outside PA settlements have >100 villages within 10 km from the park boundary, although only one has <5 villages in the vicinity of the PA (Namdapha in India).

Spectrum of Forest Dynamics: Deforestation to Reforestation

A major portion of the case studies (46%) reported forest dynamics based on multidate remote-sensing method, whereas 33% report forest-cover status using field-based ecological methods. Only 21% of the studies employed a single-date remote-sensing approach. As expected, these regions experienced different trajectories of forest change, from degradation and deforestation (71% of the case studies), leading to isolation of the PAs through changing land uses in the surroundings, to deforestation followed by forest growth, thus having negligible total change over decades in 29% of the case studies. A detailed description of the forest dynamics is provided in Table 1.
Table 1

Individual description of the case studies

Geographic area

Type of study

Study period

Forest dynamics

Sourcea

Bangladesh

Sundarbans National Park

Multitemporal remote-sensing

1977–1989–2000

Afforestation during the first half, followed by deforestation in the later half (net change within margin of error); degradation is common

Giri and others (2007)

China

Three nature reserves in Southeast Uplands: Meihuashan, Wuyishan, and Longxishan

Field-based ecological

1994–1995

Deforestation and degradation because most community-managed forests are being replaced by anthropogenic bamboo monoculture; evidence of high-level human and cattle disturbance within core of Meihuashan reserve

Coggins (2000)

India

Bandipur National Park, Karnataka

Field-based ecological

2005–2006

Differential degradation within park itself; evidence of change in species assemblages from herbaceous, large woody trees in less disturbed zones to small woody trees, more shrub species in more disturbed zones

Mehta and others (2008)

Bhadra Wildlife Sanctuary, Karnataka

Field-based ecological

2002

Evidence of degradation close to villages; degree of disturbance depends on size of village, i.e., human and cattle population

Karanth and others (2006)

Chandoli National Park, Maharashtra

Single-date remote-sensing

2005

High fragmentation within park and severe deforestation outside park; evidence of regeneration, especially in grasslands, after relocation of forest villages

Imam and others (2009)

Corbett Tiger Reserve, Uttarakhand

Single-date remote-sensing

2005–2006

Dense to moderate forest cover for major part of reserve

Singh and others (2009)

Kalakad-Mundanthurai Tiger Reserve, Tamil Nadu

Multitemporal remote-sensing

1973–1990–2004

Degradation in the evergreen forest; forest type continuously being changed to semievergreen forest and grassland, with most changes occurring during pre-establishment period

Giriraj and others (2008)

Kanha Tiger Reserve, Madhya Pradesh

Single-date remote-sensing

2007

Considerable amount of continuous surrounding forest cover, especially at higher elevation; core forest is in better condition than buffer forest; evidence of reforestation after village relocations

Ravan and others (2005), DeFries and others (2010)

Manas Tiger Reserve, Assam

Multitemporal remote-sensing

1977–1998–2006

Deforestation within the park even after establishment; drastic decrease in alluvial grassland

Sarma and others (2008)

Nagarhole Tiger Reserve, Karnataka

Single-date remote-sensing

2009

Good amount of intact forest cover within the reserve; evidence of forest corridors connecting reserve with surroundings; however, reserve is embedded within matrix of agricultural fields and tea and coffee plantations

Mahanty (2002), DeFries and others (2010)

Namdapha Tiger Reserve, Arunachal Pradesh

Field-based ecological

Degradation, especially in surrounding areas

Nath and others (2005)

Pench Tiger Reserve, Maharashtra

Multitemporal remote-sensing

1977–1989–2000–2007

Deforestation followed by reforestation both in park and surroundings

Mondal and Southworth (2010a)

Ranthambore Tiger Reserve, Rajasthan

Single-date remote-sensing

2008

Little forest cover outside park boundary; concrete wall surrounding park and agricultural mosaic isolate park from surroundings

DeFries and others (2010)

Sariska Tiger Reserve, Rajasthan

Field-based ecological

2004

Degradation, mainly changes in vegetation structure and composition in disturbed areas

Kumar and Shahabuddin (2005), Yadav and Gupta (2006), Shahabuddin and Kumar (2007)

Tadoba-Andhari Tiger Reserve, Maharashtra

Multitemporal remote-sensing

1989–2001

Reforestation and maintenance in the reserve; primarily deforestation and degradation in surroundings with reforestation in places

Nagendra and others (2006)

Indonesia

Bukit Barisan Selatan National Park

Multitemporal remote-sensing

1972–1985–2002

Deforestation both within park and within 10 km from park boundary, with higher rate of deforestation in buffer; however, total area of forest is higher in park; evidence of reforestation in inactive encroachments within park

Kinnaird and others (2003), Gaveau and others (2007)

Gunung Raya Wildlife Sanctuary

Multitemporal remote-sensing

1972–1985–2002

Severe deforestation within and outside sanctuary, with high fragmentation

Gaveau and others (2007)

Kerinci Seblat National Park

Multitemporal remote-sensing

1985–1992–1999

Deforestation

Linkie and others (2004, 2008b)

Myanmar

Chatthin Wildlife Sanctuary

Multitemporal remote-sensing

1973–1989–1992–2001–2005

Deforestation both within and outside the sanctuary until 2001, with greater rates outside; evidence of reforestation during 2001–2005 within and outside sanctuary

Songer and others (2009)

Nepal

Chitwan district covering Royal Chitwan National Park

Multitemporal remote-sensing

1976–1989–2001

Reforestation in buffer zone forests and deforestation within community forests; varying deforestation rate depending on forest types; deteriorating forest condition outside park

Nagendra and others (2004, 2005), Panta and others (2008)

Royal Bardia National Park

Field-based ecological

2005

Degradation in forests closer to villages; human disturbance impacted forest vegetation structure and diversity, specifically species richness, tree density, diversity, evenness, and basal area of trees; evidence of lack of alternative resource-collection areas and livelihood need for communities, leading to more disturbances

Thapa and Chapman (2010)

Russia

Sikhote-Alin range covering part of Sikhote-alinsly Biosphere Reserve

Multitemporal remote-sensing

1972–1992

Deforestation rate varied depending on forest type, with greater disturbance in accessible forests

Cushman and Wallin (2000), Kerley and others (2002)

aSource includes both primary and supplementary articles referred for each case study

India exhibits all of the forest-cover trajectories mentioned previously. While degradation is prevalent in Indian PAs (Kumar and Shahabuddin 2005; Nath and others 2005; Karanth and others 2006; Giriraj and others 2008; Mehta and others 2008), deforestation is often either dominant in the surroundings (Nagendra and others 2006; Imam and others 2009) or is followed by reforestation as a consequence of village relocation (Imam and others 2009; DeFries and others 2010) or change in degree of protection (Mondal and Southworth 2010a, b). What is alarming here is that some PAs experienced deforestation within the core itself, despite their strict protected status (Sarma and others 2008; Mondal and Southworth 2010a, b). On a positive note, PAs with maintained or regenerated forest cover are not uncommon (Singh and others 2009; DeFries and others 2010). However, PAs with dense to moderate forest cover in the core can be isolated if surrounded by other land uses (DeFries and others 2010), thus resulting in restricted animal movement.

The Bangladesh case study suggests forest disturbance generated by both natural and anthropogenic factors (Giri and others 2007). Soil erosion poses a major threat in the estuarine ecosystem of the Sundarbans, which is further aggravated by agricultural and aquacultural conversion of forested land. The study covers both the sides of Sundarbans (i.e., in India and Bangladesh) and suggests overall forest degradation, although the net area of mangrove forests during the study period remained within the error margin. All of the Chinese and Indonesian case studies suggest severe deforestation both within and outside the PAs (Coggins 2000; Kinnaird and others 2003; Linkie and others 2004; Gaveau and others 2007). The case study from Myanmar suggests deforestation both within and outside the PA, with greater rates outside the PA, followed by reforestation (Songer and others 2009). Interestingly, the degree of protection did not appear to contribute to the forest dynamics in these case studies as both protected and nonprotected forests experienced similar trajectories of forest change. In contrast, the role that the nature of management regime plays is quite evident in the forest dynamics of Nepal. Studies have reported reforestation in previously degraded forests in Nepal, which is often linked to institutional decentralization (Nagendra and others 2005; Panta and others 2008). The Russian case study suggests varied degrees of disturbance in nonprotected areas, with greater disturbance in easily accessible forests (Cushman and Wallin 2000). However, the reported change in forest type (conversion to hardwood forest) is thought to be beneficial for habitat generalist species, such as the Siberian tiger.

Multidirectionality of Proximate Drivers of Changes in Forested Landscapes

Given the biophysical, socioeconomic, and institutional differences in the case studies, a wide range in the proximate drivers of forest change that shape these landscapes was expected. Our analysis suggests that all of the factors cited as probable causes behind the changes can be grouped under seven broad categories (Table 2), consistent with other global-level studies of drivers of land-use and land-cover change (Lambin and others 2001; Geist and Lambin 2002). However, it is difficult to reach a firm conclusion on the factors responsible for forest-cover change based on the small number of case studies from Bangladesh, China, Myanmar, Nepal, and Russia. The list of the factors from these countries is likely to expand with inclusion of more case studies.
Table 2

List of proximate causes driving changes in forest structure and composition mentioned by the case studies

Proximate causes of deforestation and degradation

Citation by case studies (%)

Country location of case studies

Agricultural conversion, plantation, and farming

52

Bangladesh, China, India, Indonesia, Myanmar

Fuelwood and fodder extraction

43

India, Nepal

Illegal and legal logging

38

Bangladesh, India, Indonesia, Nepal, Russia

Grazing

38

India, Nepal

Flood and soil erosion

24

Bangladesh, India, Myanmar

Fire

19

India, Indonesia, Russia

Tourism and development

10

India

The most cited driver of forest change is agricultural conversion, plantation, and farming, mentioned by 52% of the case studies. These case studies are from five countries—Bangladesh, China, India, Indonesia, and Myanmar—suggesting that the drivers of changes cross beyond political boundaries, albeit following multiple trajectories. Although a land-use shift toward tea and coffee plantations can be identified as driving much of the habitat modification in India (Giriraj and others 2008), bamboo forestry in China (Coggins 2000) and conversion to paddy fields and shrimp farms in Bangladesh (Gopal and Chauhan 2006) also play pivotal roles in habitat destruction. Agricultural conversion is probably the most important driving factor in Indonesia because even the core forests are under direct threat of agricultural encroachments, especially along the boundaries (Kinnaird and others 2003; Gaveau and others 2007; Linkie and others 2008a, b).

Fuelwood and fodder extraction as well as grazing, problems mostly in India and Nepal, are identified by 43 and 38% of the case studies, respectively (Kumar and Shahabuddin 2005; Nagendra and others 2005; Nath and others 2005; Karanth and others 2006; Davidar and others 2007; Mehta and others 2008; Imam and others 2009; DeFries and others 2010). This finding is consistent with the Indian socioeconomic scenario, in which most of the PAs are surrounded by forest villages, and people are dependent on forest products for subsistent living and cattle-raising. Since most Indian PAs are exclusionary in terms of institutional regime, forest villages are often relocated outside the PA boundary, thus exerting more pressure on the surrounding landscape (Nagendra and others 2006).

Illegal and/or commercial timber collection through selective logging and/or clear felling is another common route to habitat loss and fragmentation, mentioned by 38% of the case studies from five countries (Table 2). Systematic logging, sometimes even from the core forest, has been practiced in India and Bangladesh as a part of forest management activities (Imam and others 2009; Mondal and Southworth 2010a). In addition, illegal logging of valuable timber contributes to deforestation and degradation, especially in easily accessed regions, in various countries (Cushman and Wallin 2000, 2002; Linkie and others 2004; Panta and others 2008). Road developments often act as a catalyst to deforestation and degradation by improving accessibility between remote areas and local markets, thereby increasing tourism pressure (Nagendra and others 2006; DeFries and others 2010).

In addition to the anthropogenic factors discussed previously, flooding and wildfire (both natural and anthropogenic origin) are also mentioned by 24 and 19% of the case studies, respectively (Table 2). Controlled flooding, along with natural floods, are mentioned as a common cause of deforestation in many PAs from Bangladesh, India, and Myanmar (Nath and others 2005; Giri and others 2007; Sarma and others 2008; Songer and others 2009; Mondal and Southworth 2010a). For example, post-independence India experienced rapid infrastructural growth, as a result of which considerable amounts of forested regions were cleared and later submerged for hydroelectric projects (Bhat and others 2001). Wildfire, often set by local farmers and/or poachers, is another common contributing factor of deforestation and/or degradation in India, Indonesia, and Russia (Cushman and Wallin 2000; Kinnaird and others 2003; Nagendra and others 2006; DeFries and others 2010).

Discussion

Forest Dynamics in Individual Countries: Trends, Challenges, and Implications for Tiger Conservation

Bangladesh

The mangrove swamps of the Sundarbans, Bangladesh, provide the last remaining refuge to the Royal Bengal tiger. This mangrove ecosystem, shared between Bangladesh (60%) and India (40%), is the world’s largest coastal wetland. Although this dynamic ecosystem frequently suffers from massive soil erosion due to its geographic location, the resulting forest loss is often offset by the accreted land and new mangrove plantations (Giri and others 2007). In addition to the natural disturbances, mangrove forests of Bangladesh have a long history of human exploitation as well (Seidensticker and others 1991). A vast proportion of mangrove forests was cleared and converted to agriculture during Turk and British colonization (Eaton 1990). Later, despite having a PA network, clear felling and timber extraction were allowed within the protected and reserved forests, leading to major mangrove forest loss in the three decades since independence was gained in 1947 (Richards 1990). Today’s densely populated Sundarbans is threatened not only by agriculture but also by the rapidly growing field of aquaculture. Local people depend on forest products for subsistence living and exert additional pressure on mangrove forests by harvesting fish as well as shrimp and prawn larvae (Gopal and Chauhan 2006). In addition to forest degradation, human activities (such as forest clearance and conversion to other land uses) also directly influence freshwater flow and sediment accretion in this ecosystem. Such changes in salinity can have deleterious effects on vegetation types, herbivorous species, and predators (Hussain and Acharya 1994).

To minimize forest loss, the Government of Bangladesh adopted its second forest policy in 1994 to expand the PA network to design and implement multiple usages of Sundarbans and to involve local people through afforestation of encroached forests and other activities (Government of Bangladesh 1994). However, none of the existing forest policy identified the importance of buffer zones surrounding the PAs, which are often embedded within an already degraded reserve forest (Sharma and others 2008). Although multiple commitments have been included in an amendment of the Bangladesh Wildlife (Preservation) Order to better protect and manage the forests and their species, challenges remain in terms of taking transboundary initiatives with India for Sundarbans management, limiting allowable resource extraction from the PAs, providing economically viable alternatives to shrimp farming, and implementing effective participatory management (Iftekhar 2006; Chowdhury and others 2009).

China

China is distinguished from the other countries included in this study by its formalized land tenure system, which started during collectivization and communization and resulted in collective forests. The management regime of collective forests is complex and has changed frequently (Miao and West 2004). During the post-revolution period (after 1949), China witnessed changes in the forest institutional regime, with many of its private, communal, and state-owned forested land changing to elementary cooperative-managed forest to promote sustainable forestry (Liu 2001). However, further amalgamation of cooperative lands into communes and associated corruption, along with the government-promoted program of iron and steel production, are believed to have resulted in large-scale deforestation (Liu 2001). The post-revolution era also witnessed a rapid increase in PAs, with >2000 nationally protected nature reserves by 2004 (Xu and Melick 2007). Park-people conflict has consequently increased because traditional forest management practices are affected by conservation restrictions. It is often argued that the PAs will be more effective in conserving flagship species under traditional community management rather than constantly changing and confusing institutional regimes (Xu and Melick 2007). However, the small size of many PAs, increasing PA isolation as a result of fragmentation, and high-level of human disturbance within the core forest, along with uncontrolled hunting during “anti-pest” campaigns in the 1950s, when tigers were considered pests that required large-scale eradication, are often held responsible for decline or complete absence of tiger populations and their prey species in South China tiger reserves (Tilson and others 2004).

India

India, hosting one third of the wild tiger population, is a major player in the global tiger-conservation scenario. With Project Tiger launched in 1973, the Indian government assigned maximum protection to tiger reserves in the country. However, conservation practices in India continued to be a reflection of the colonial legacy of commercial extraction for a long time even after independence (after 1947), resulting in alarming forest loss, especially in regions with limited protection (Bhat and others 2001). In addition, other factors, including frequent relocation of forest villages outside the PAs, national-level emphasis on agricultural expansion, and industrial activities (such as road network development, hydroelectric projects), contributed to postindependence forest degradation (Bhat and others 2001; Nagendra and others 2006; Mondal and Southworth 2010a). To restrict deforestation and degradation, various remedial measures, including a complete ban on tree felling within any national park and the implementation of plantation programs in the surrounding areas (Ministry of Environment and Forests 1988), have been implemented nationwide. Resource collection, grazing, and hunting activities are strictly banned in the core reserves, while these activities are often restricted in the surroundings as well. The World Bank-supported India Ecodevelopment Project (1996–2004) aimed at decreasing forest dependence among communities through livelihood-support activities (World Bank 1996). Although one of the main objectives of this project was to increase opportunities for local participation in PA management, it was never implemented as a participatory project. Given these flaws in implementation, many conservationists are skeptical about its impact on decreasing anthropogenic pressure on the PA landscapes (Gubbi and others 2009; Shahabuddin 2010). As a result, many of the reserves are becoming increasingly isolated because of rapidly changing land uses in the surroundings (DeFries and others 2010). With negligible participation from the forest-fringe communities in park management and little benefit received by the communities, human–tiger conflict is increasingly obvious (Madhusudan 2003; Ogra and Badola 2008).

Indonesia

The link between Asian financial crisis and massive deforestation in the biodiversity-rich ecosystems of Indonesia is highly contested and is often political (Sunderlin 1999; Robertson and van Schaik 2001). This rapid deforestation is especially alarming because Indonesia has an extensive system of PAs, detailed land-use plans for commercial logging, and abundant donor assistance (Kinnaird and others 2003). Multiple studies have documented legal and illegal logging, agricultural conversion by new settlers arriving through Indonesia’s transmigration program, and the development of oil palm and pulpwood plantations as driving factors influencing Indonesia’s massive deforestation (Barr 2001; Robertson and van Schaik 2001; Holmes 2002) and leading to increasing resource competition between tigers and local inhabitants. Although tigers show a tendency to move away from forest edges to avoid disturbances, such as hunting and other human activities (Kinnaird and others 2003), PAs with low elevation edges often have high prey densities, thus supporting tiger populations in lowland forests. As a result, human–tiger conflict is high along forest edges where the probability of human–tiger interaction is maximum (Nyhus and Tilson 2004a). Declaration of more PAs that have suitable habitat and prey base and thus serve as corridors between isolated forested patches, such as Batang Gadis National Park (Wibisono and others 2009), can have positive impacts on tiger conservation. However, to preserve the presumed dwindling Sumatran tiger population, the extensive agro-forestry matrix surrounding the PAs must be wisely used (Nyhus and Tilson 2004a, b).

Myanmar

Myanmar’s complex political history has shaped changes in protected landscapes (Aung and others 2004). Postindependence (after 1948) Myanmar had other priorities than wildlife protection, focusing instead on economic and institutional changes. Postwar social unrest aggravated hunting, and there was no provision for wildlife habitat protection in the 1954 amendment of the Burma Wildlife Protection Act (Gutter 2001). After the Asian economic crisis in the mid-1980s and increasing pressure for natural resources from neighboring countries, the severity of the threat to Myanmar’s forest and wildlife became apparent, and conservation strategies began to emerge in national policies. However, tiger populations continued to decline as a result of hunting, prey depletion, habitat fragmentation, inadequate PA coverage, lack of effective management, and the flourishing trade of tiger parts in global market (Lynam and others 2006). Myanmar’s first tiger reserve, also the world’s largest (Hukaung valley), was recently proposed to be created as part of a national tiger-conservation strategy (Lynam and others 2009). Although extensive surveys have confirmed tiger population in extensive and continuous forested tracks of farther north and farther south of Myanmar, it would be particularly challenging to establish a sound conservation plan, especially to stop tiger habitat loss in transborder areas and cross-border trafficking of tiger parts (Lynam and others 2006).

Nepal

Tigers used to range across most of lowland Nepal until settlement programs in the 1960s converted most of this region to agricultural land (Gurung 1983). Further decline in the tiger population was prevented through several government-initiated programs, such as creation of a lowland park system, community forestry in PA-surrounding landscapes, and the Terai Arc Landscape project (Gurung and others 2008). These projects together contributed significantly to improved habitat quality not only within the PA but also in the neighboring landscape, to decrease anthropogenic pressure on forest resources through benefit-sharing with the local communities, and to link distant lowland forests that can act as breeding habitat. As a result, the Terai arc landscape is now considered to be one of the best tiger habitats in the world (Barlow and others 2009), which underscores the need to re-establish the dispersal corridors across this landscape (Wikramanayake and others 2004). However, during the Maoist insurgency (2000–2007), conservation efforts were severely affected both in national parks and community-managed forests, resulting in a rapid increase in wildlife (especially rhino) poaching in and around Chitwan and Bardia national parks (Martin and others 2008). Although no study to our knowledge has identified the political instability as a direct factor affecting tiger conservation, Baral and Heinen (2006) have documented the adverse effects of such political turmoil on the wildlife population, habitat, and overall conservation efforts in Nepal.

Russia

Unlike other forested areas in temperate zone, the Russian Far East was able to maintain its primary forests until the last decades of the last century, when political and economic changes led to urbanization and industrialization of this area (Cushman and Wallin 2000; Kondrashov 2004). Although wildfire shapes most of the landscape-level changes in this landscape, increasing road density is rapidly becoming a severe threat to the tiger population through expanding tourism in PAs and increased hunting and logging in unprotected areas (Kerley and others 2002; Kondrashov 2004). Kerley and others (2002) found that road access severely hinders PAs from functioning as source populations of tigers and indicated that it would be difficult to sustain a viable tiger population with increased human access in unprotected areas. Moreover, it is not unusual for forest-fringe communities to turn to poaching of large mammals in PAs in the absence of traditional livelihood options and lack of federal support during political turmoil and economic decline (Shmatkov and Brigham 2003). Although some efforts have been made to engage indigenous communities in participatory management in PA surrounding landscapes (Shmatkov and Brigham 2003), such efforts should focus more on areas with suitable tiger habitats.

Forest Dynamics in Tiger Landscapes: Common Needs for Tiger Conservation

In theory, PAs are established to protect biodiversity; specifically in tiger landscapes, the first priority should be to provide adequate protection to this endangered charismatic species. However, across countries, conservation goals are often hindered by conflicting institutional regimes and management strategies, pressing need for alternate livelihood for forest-dependent communities, lack of infrastructure for effective PA management, and human–wildlife conflict. All countries that harbor viable in situ tiger populations have identified restoring habitats and establishing habitat corridors as one of the priority issues for tiger conservation (Global Tiger Initiative 2010). As the nucleus of a greater ecosystem, the core forest, along with the target species, must be strictly protected, both on paper and on the ground, to avoid local extermination as was seen in the case of the Sariska Tiger Reserve in India. In addition, the surrounding landscape warrants similar attention to maintain a greater ecosystem (DeFries and others 2007). An intact or minimally disturbed greater ecosystem is a prerequisite to host a thriving prey base, without which it is almost impossible to support a viable tiger population (Karanth and Stith 1999). As identified by Laurance (1999), it is not enough to conserve isolated and fragmented PAs; the quality of the peripheral forest matrix must also be protected and regenerated to decrease the risk of mortality for wide-ranging mammals. However, demands for maintaining such a landscape often severely underestimates the regular needs of the local communities who live in and around tiger-ranging PAs.

It has long been identified that development needs of the local communities that share the ecosystem with tiger and its prey species must be addressed to further conservation goals (MacKinnon and others 1999). However, from the case studies discussed here, these issues are evidently not being addressed sufficiently. According to our meta-analysis, tiger reserves are often subjected to agricultural, infrastructural, and commercial developments (Table 2), which are reflected in deforestation and degradation in the surrounding landscape if not in the core forest itself (Table 1). Thus, multiple-use buffer zones around the strictly protected PAs might be one option to address the ongoing debate on issues of conservation versus development. Designated buffer zones with sustainable extraction would not only allow the local communities to continue with their traditional livelihood, they would also decrease anthropogenic pressure in the core habitat and expand the available habitat to the tiger prey species.

Furthermore, none of the countries have effective participatory management regimes that balance the needs of the forest-fringe communities with conservation goals. Decentralized management, accused by conservationists of not being “conservation-centered” and by critics of coercive conservation as not being sufficiently “development-oriented,” has attracted much attention during the last decades. The balance between conservation and development is society and context specific and a complex task that is not easily achievable, requiring sustained financial, social, institutional, and political commitment at all levels from national to local. Of the countries discussed here, only Nepal has a community forest and park buffer-zone management program that incorporates community participation in the forest management, albeit with negligible power to introduce and/or modify rules of the government authority managing PAs. Although India has experimented with initiatives toward involving local communities in ecodevelopment programs, these have been widely critiqued as being far less participatory in practice and not having achieved their stated goals of providing local communities with a stake in conservation goals (Ostrom and Nagendra 2006; Shahabuddin 2010). Yet, much of the deforestation and forest fragmentation in and around tiger embedded PAs can be traced to human impact. There is a clear and imperative need to involve local forest-dependent communities as partners in conservation initiatives on the ground, not just on paper.

Conclusion

This meta-analysis synthesizes published quantitative data on forest dynamics from 24 tiger-hosting protected areas from 7 countries. Our findings indicate that the success of tiger-conservation efforts will depend on the overall management of the landscapes in which the PAs are embedded. There is also an imperative need for better livelihood options, along with active participation in PA management and conservation, for the forest-fringe communities, which appears critical given that most remaining tiger habitats are in areas inhabited by poor, rural, forest-dependent communities. New perspectives and lessons learned locally and across borders must be integrated to ensure tiger conservation in these human-dominated landscapes. Although the long-term persistence of large mammals, such as the tiger, is linked to the persistence of large, spatially connected habitats, most parks appear to be becoming increasingly isolated due to agricultural, infrastructural, and commercial developments outside the protected landscape. Thus, our findings indicate that it is imperative to conduct frequent revision of goals and performance evaluation at the landscape level, incorporating multiple actors and institutions into such planning, instead of focusing solely on conservation within the PAs as is currently the case in most tiger parks.

Acknowledgments

Harini Nagendra thanks the Department of Science and Technology, Government of India, for financial support. Pinki Mondal was supported by the University of Florida Alumni Fellowship. The manuscript was greatly improved by the constructive comments from three anonymous reviewers.

Copyright information

© Springer Science+Business Media, LLC 2011