Ecological Dynamics of Vultures, Blackbuck Antelope, Khejeri Trees, and the Bishnoi People in Western Rajasthan, India

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Jonathan Clarence Hall

Graduate Program in Evolution, Ecology and Organismal Biology

The Ohio State University

2011

Dissertation Committee:

Ian Hamilton, Advisor

Elizabeth Marschall

Mark Moritz

Maria Miriti

Kendra McSweeney

Jonathan Clarence Hall

2011

Dissertation Abstract

In the last half century Western Rajasthan has experienced dramatic changes in resource availability, climate, species abundance, livestock populations, and human development. Human population growth and development drives these changes as an increase percentage of land area is put to human use. Understanding the dynamics of this human dominated landscape are important if other species are to survive and coexist with human populations. I pursue a novel approach in this research and examine the interplay between human beliefs and practices and the abundance of keystone species that share this landscape. I performed a comparative study of two set of villages in Western Rajasthan to investigate the dynamics that drive abundance of Indian vultures (Gyps indicus), blackbuck antelope (Antilope cervicapra), and Khejeri trees (Prosopis cineraria) incorporating the presence of the

Bishnoi people. The Bishnoi are a caste of Indians who’s religion mandates that they protect Khejeri trees and antelope with their lives. The results of this study show that Bishnoi populations are positively associated with all study species. Indian vulture dynamics are driven by La Niña induced drought, synchronizing population dynamics across a wide region, but may be partially mitigated in one particular village by the presence of Bishnoi. Khejeri trees were found in greater abundance in villages with Bishnoi and their abundance appears to be influenced differently by

ii household income and livestock populations in Bishnoi areas than in other areas.

Blackbuck antelope were also found in greater abundance in areas with Bishnoi populations despite being present in less than half of the villages in the study.

Finally, this study highlights important influences on household income, namely livestock populations, that captures the ecological complexity of human capitalization and development in this region. This research is the first to systematically investigate the ecological impact of the Bishnoi people on keystone species and one of the few studies that seeks to understand the intersection among human cultural practices and the coexistence of humans and other species in a human dominated landscape.

iii

Dedication

For my wonderful parents, Clarence and Ester Hall, who have always encouraged

me to follow my curiosity and pushed me to achieve more than I thought I was

capable of.

In loving memory of my sister Alicia who left us too soon and was one of my best

friends.

Acknowledgments

I’d like to thank my advisor, Ian Hamilton, for his sharp insight, extensive comments, blazing fast turn-around times, patience, and enthusiasm during my research. I have learned so much about how to think as scientist from him. I am grateful to my committee, Libby Marschall, Mark Moritz, Maria Miriti, and Kendra McSweeney, for their diverse perspectives that have made this an interdisciplinary body of work. I am especially grateful to Libby Marschall for her support, suggestions, patience, and perspective during my transition to another lab. Libby is the perfect example of what a graduate advisor should be. I’d also like to thank Tom Waite for facilitating the start of this work and for being one of the most interesting and awesome people I know. I am beyond grateful to Anil Chhangani for sharing his knowledge, giving his time and resources, opening up his home, and welcoming me into his family. He is both a scientific and a humanitarian role model. Rajuji and Bhughji, my partners in the field, made work in villages possible and fun. I’d like to thank Ramesh, Bunti, and the Rathore brothers, particularly Bharat Singh, for teaching me some Hindi and for the enthusiastic exchange of culture and ideas. From the bottom of my heart I thank my mother and father, Clarence and Ester Hall, for always supporting and believing in me. I could not have asked for better parents. I thank my dear friends, Anthony, Nichole, Nadine, and Bennett for also being there, protecting my sanity, challenging my thinking, and always making me laugh. I’d like to thank The Ohio State University, the College of Biological Sciences, the department of EEOB, James Moore and the Bell Resource Center, Larry Williamson and the Hale Black Cultural Center, and the Office of International Affairs for their support and guidance. Thank you all!

Vita

June 2001 ...... Howard High, Ellicott City, MD

2005 ...... B.S. Biology, Morehouse College

2006 to present ...... Graduate Teaching Associate, Department

of Evolution, Ecology, and Organismal

Biology, The Ohio State University

Publications

Hall JC, Chhangani AK, Waite TA, and Hamilton IM (in press) The impacts of La Niña induced drought on Indian vulture Gyps indicus populations in Western Rajasthan. Bird Conservation International.

Fields of Study

Major Field: Evolution, Ecology and Organizmal Biology

Table of Contents

Chapter 1: General Introduction ...... 1

Chapter 2: The impacts of La Niña induced drought on Indian vulture Gyps indicus populations in Western Rajasthan ...... 4

Abstract ...... 4

Introduction ...... 4

Methods ...... 7

Results ...... 11

Discussion ...... 13

Conclusions ...... 17

Chapter 3: Distribution of Prosopis cineraria on agricultural farmland in Western

Rajasthan: influence of the Bishnoi people on tree abundance ...... 18

Abstract ...... 18

Introduction ...... 19

Methods ...... 21

Results ...... 25

Discussion ...... 27

Conclusions ...... 30

vii

Chapter 4: Patterns of blackbuck distribution on village lands in Western Rajasthan

...... 31

Abstract ...... 31

Introduction ...... 32

Methods ...... 34

Results ...... 40

Discussion ...... 40

Conclusions ...... 42

Chapter 5: Patterns of economic variation among villages in Western Rajasthan .... 44

Abstract ...... 44

Introduction ...... 45

Methods ...... 48

Results ...... 54

Discussion ...... 55

Conclusions ...... 58

Chapter 6: General Conclusions ...... 59

Literature Cited ...... 62

Appendix A: Chapter 3 ...... 94

Appendix B: Chapter 3 ...... 95 viii

Appendix C: Chapter 5 ...... 96

Appendix D: Chapter 5 ...... 97

Appendix E: Chapter 5 ...... 98

List of Tables

Table 1. Vulture counts from each of the 11 villages from 1996 to 2005...... 59

Table 2. Annual population growth rates of Indian vultures in each surveyed village from 1996 to 2005...... 70

Table 3. Cross-correlation matrix for time series of annual population growth rate for 11 local populations (named after villages) of the Indian vulture. All 55 pairwise combinations of time series were positively correlated, indicating region-wide synchrony...... 71

Table 4. Local populations of Indian vulture ordered from highest to lowest mean population growth rate across the time series, 1996-2005. Also shown are predictor variables included in the best AIC-based ARIMA models...... 72

Table 5. Goodness-of-fit as measured by stationary r2 for first-order ARIMA models.

Villages (local breeding populations of Indian vultures) ordered from best to worst for the model with predictors AR1+MEIt...... 73

Villages (l local breeding populations of Indian vultures) ordered based on ascending values of this statistic. Local breeding populations of Indian vulture in villages nearer the bottom of the list were apparently less influenced by ENSO...... 74

Table 7. Results from two non-parametric test comparing Bishnoi and non-Bishnoi villages for each village characteristic. Khejeri tree numbers are significantly different between two village types...... 75

Table 8. R2 of village characteristics with each axis. Average annual income is the most closely association with both axis 1 and 2. Axis 1 is also influenced by total population and farmland area. Axis 2 is influenced by total livestock. Axes are unrotated...... 76

Table 9. Results of ANCOVA analysis. The effect of Bishnoi presence and the interaction between Bishnoi and Axis 2 were significant. The effect of the interaction between Bishnoi presence and Axis 1 was non-significant (F=2.867, df=1, p=0.120) and was removed...... 77

Table 10. List of villages categorized by presence of Bishnoi and blackbuck with the total number and number per sampling occasion for each observation year. Note that villages were unequally sampled in 2009 and equally sampled in 2010...... 78

Table 11. Model output values for 2009, 2010, and the combine year’s analysis. Two models were run for each of the three test, one with missing observations (actual) and one where missing observations were considered absences (simulated). Ψ stands for occupancy probability, p is detection probability, ε and is extinction probability across years. Standard errors for each model output are included. ε values are not an output for individual year analyses and are thus blocked out...... 79

Table 12. R2 values of village characteristics associated with both NMS ordination axes. Axis 1 explains 62.9% of variation while axis 2 explains 33.4%. Number of

Khejeri trees is the only characteristic not noticeably associated with either axis. .. 80

Table 13. A) ANCOVA models ranked based on AIC values. All models are within 2 

AIC values of each other and are thus appropriate models. SSE = standard sum of squares; k = number of model parameters. B) p-values for each model effect in for the final three models. No model effect is significant for all models suggesting no model effect influences income. Interaction effects were removed from each model in order of greatest p-value...... 81

Table 14. R2 values of village characteristics associated with both NMS ordination axes for all villages excluding outlier village Guda Bishnoi. Axis 1 explains 57.9% of variation while axis 2 exaplins 38.4%. Axis 1 is primarily associated with total livestock population and axis 2 is primarily associated with population, agricultural area, and livestock...... 82

= number of model parameters. B) p-values for each model effect in for the final three models. Axis 1 is shows a significant effect on income for both models.

Interaction effects were removed from each model in order of greatest p-value...... 83

xii

Table 16. Effects of village characteristics on average annual income (ANCOVA model 1). Axes refer to axis scores from NMS analysis (see Table 4 and Appendix 3).

...... 84

Table 17. Effects of village characteristics on average annual income (ANCOVA model 2). Axes refer to axis scores from NMS analysis (see Table 4 and Appendix 3).

...... 85

xiii

List of Figures

Figure 1. A) Map of Rajasthan. B) Map of village locations in relation to the major cities Jodhpur and Pali (pictured). Push pin icons mark village sites...... 86

Figure 2. Time series of Multivariate ENSO Index (MEI). Positive MEI values indicate El Niño events and negative values indicate La Niña events...... 87

Figure 3. Map of Rajasthan (above) and the study villages (below). Black arrow indicates Jodhpur city...... 88

Figure 4. A) Schematic representation of one transect run in Khejarli Kalla village.

The vertical white arrow points to the village center. The black arrow with the white border represents the “buffer transect” 500m from the village center. The white arrow with the black border represents the transect where Khejeri trees were counted. Each transect heading has a starting point 500m from the village center and an end point 1km from the village center. KKL stands for Khejarli Kalla. B)The solid black barbell line is a representation of a supplemental transect run equidistant from two original transect starting points. SSW stands for south of southwest...... 89

Figure 5. Plot of Khejeri tree abundance against NMS axis 2 scores for two villages groups in the study. Trend lines for each village type included. Lack of similarity in the data around the intersection point indicates difference in trend between village groups is most likely not due to a non-linear interaction, but the result of a differential influence of axis 2...... 90

xiv

Figure 6. Aerial photo of one village in the study area, Khejarli Kalla village. The various land types within the village are labeled...... 91

Figure 7. Photo of male blackbuck antelope crossing the road in Baniawas village. 92

Figure 8. Graph of NMS axis scores vs. Average Annual Household Income (AAI) with each of the nineteen villages represented by a diamond. Guda Bishnoi village is located in the upper right portion of the graph...... 93

Chapter 1: General Introduction

Humans have colonized the overwhelming majority of the terrestrial land area and transformed environments and ecosystems through our activities. As human populations have increased, we have influenced the population dynamics and the evolution of many species and caused others to go extinct (Pimm et al. 1995, Wackernagel and Rees 1998).

Western philosophy separates humanity from nature describing our role in the biosphere as stewards and managers that live somewhat apart from our environments. Undoubtedly, this outlook has shaped the way many nations and peoples manage their natural resources and interact with other species. However, rather than non-human species occupying a space apart from humanity, these species find themselves in an increasing number of human dominated landscapes (Wackernagel and Rees 1998, Anand et al. 2010).

Given that human dominated landscapes are increasing it is becoming increasingly important to understand how humans and other species are able to coexist in these landscapes and to devise strategies that mitigate conflict and support coexistence with other species (Brockmann and Pichler 2004, Sanginga et al 2007, Anand et al. 2010, Abson and Termansen 2010).

Non-western philosophies are more integrative, placing humans within the context of their environment as active, albeit uniquely skilled, participants and contributors to ecosystems (Brockmann and Pichler 2004). Living with, managing and preserving the 1 components of an ecosystem is a general principle readily practiced in communities and cultures in India (2004). The Bishnoi people of western Rajasthan, India are one such culture where the principles of species conservation and protection are components of their religion. The historical and traditional implications of the environmental principles of the Bishnoi religion serve as a platform for investigating the impact human culture can have on non-human species in human-dominated environments.

The Bishnoi people are a caste of agro-pastoralists found in Western Rajasthan that practice a specific religion that is a sect of Hinduism (Fisher 1997; Brockmann and Pichler

2004). Among their chief tenets is the protection and maintenance of the Khejeri tree, which they deem to be sacred. According to Bishnoi history, the founder of their faith,

Jambeshwar, received inspiration for the twenty-nine principles that would become the

Bishnoi faith while sitting underneath a Khejeri tree (Brockmann and Pichler 2004).

Because of Prosopis cineraria’s inspirational power and ecological significance, Bishnoi people are never to lop the branches or otherwise harm any Khejeri tree (Fisher 1997;

Brockmann and Pichler 2004). Moreover, Bishnoi people are to protect their sacred tree with their lives, and did so during an extraordinary demonstration in 1730 where 363 individuals sacrificed their lives in protest of Khejeri tree harvesting by the Maharaja

Abhay Singh (Fisher 1997; Brockmann and Pichler 2004). In addition to protecting trees the Bishnoi people detest hunting of antelope and have been known to beat poachers if they are caught hunting on their lands (Fisher 1997, personal observation). The Bishnoi believe that their ancestors are reincarnated as antelope and thus consider them family members (Brockmann and Pichler 2004).

Aside from their religious prominence the Bishnoi are politically and economically active. They are one of the wealthier castes groups within the research area (Chapter 5) and hold several political offices in Jodhpur city including police chief (personal observation). In 1998 famous Bollywood actor Salman Khan was caught poaching blackbuck antelope in Khejarli village (a village with a prominent Bishnoi population) (The

Hindu 2007). It was protests by Bishnoi citizens that pressured local authorities to charge and arrest Mr. Khan. The actor was sentenced to five years in prison for his crimes (he served less than two months). The Bishnoi are a unique group of citizens that potentially have an effect on their economic, political, and natural environment.

The following research is a comparative study examining the differences in abundance of three key species; Indian vultures, blackbuck antelope, and the Khejeri tree, and differences in income and human practices between villages with Bishnoi populations and villages without Bishnoi. The overall goal of this research is to measure the potential differential ecological and economic impacts people of varying cultural groups subcultures have on their respective environments. Describing the dynamics and principles of this system, where humans and other species directly share a common landscape, has the power to inform the management and philosophy that governs other similar landscapes where humans and other species live in close proximity.

Chapter 2: The impacts of La Niña induced drought on Indian vulture Gyps indicus populations in Western Rajasthan

ABSTRACT

Previous research on the catastrophic decline of the Gyps species complex has identified diclofenac, an anti-inflammatory drug administered to livestock, as the primary cause.

Large-scale climatic phenomena, such as ENSO-induced drought, however have not been examined. Based on time series analysis of annual count data, 1996-2005, we provide evidence that ENSO synchronized population dynamics throughout western Rajasthan.

Here, we ask whether El Niño Southern Oscillation (ENSO) can also explain the population dynamics of the critically endangered Indian vulture (Gyps indicus). We attribute this impact largely to two La Niña events, including the major event spanning 1999. Although these climatic events apparently affected local populations in a parallel way across the region, we explore whether one particular local population might have been partially buffered by the religion-based conservation practices of the Bishnoi people. Our results show that the Indian vulture population in the Bishnoi village of Khejerli was buffered from drought events. Finally, we discuss how the Bishnoi people may have contributed to the buffering that occurred.

INTRODUCTION

Vulture species across the globe face numerous threats to their survival. Many of these challenges are the result of human persecution, development, and unintentional poisoning 4

(Green et al. 2004, Oaks et al. 2004, Herna ndez and Margalida 2008, Margalida et al. 2008,

Herna ndez and Margalida 2009 . In western sia catastrophic declines of vultures belonging to the genus Gyps have been documented throughout India and Pakistan in recent years (Prakash 1999, Prakash et al. 2003, Gilbert et al. 2002, Oaks et al. 2004). The first early warning sign was detected in India’s Keoladeo National Park in the 1990s, when white-backed vultures (Gyps bengalensis), then one of the most common raptors on the

Indian subcontinent, began a massive decline (Prakash 1999). Since then, catastrophic declines, also involving the Indian (Gyps indicus, formerly G. indicus indicus) and slender- billed vulture (Gyps tenuirostris, formerly G. indicus tenuirostris), have been reported across the subcontinent (Prakash et al. 2003). Not surprisingly, these species are now listed as critically endangered (BirdLife International 2009) and a flurry of research activity has attempted to identify the underlying cause of these population crashes.

These crashes have been largely attributed to poisoning by a non-steroidal anti- inflammatory drug, diclofenac, used widely to treat sick and arthritic livestock (Green et al.

2004, Oaks et al. 2004, Shultz et al. 2004, Green et al. 2006, Taggart et al. 2006, Taggart et al. 2007). Vultures that feed on diclofenac-laced carcasses can suffer renal failure and visceral gout (Oaks et al., 2004, Shultz et al. 2004, Swan et al. 2006), which caused high mortality and led to dramatic population declines (34-95%) of white-backed vultures in

Pakistan between 2000 and 2003 (Oaks et al. 2004). A follow-up study purported to show that diclofenac concentrations in livestock carcasses were sufficient to explain documented declines of all three abovementioned Gyps species across the Indian subcontinent (Green et al. 2006). The evidence that diclofenac played a key causal role in the vulture die-off is compelling.

However, the potential role of climate in connection with vulture declines has not been thoroughly explored thus far. There is a growing body of evidence that climatic factors including El Niño Southern Oscillation (ENSO) events can have dramatic negative effects on populations (Chhangani 2004, Chhangani and Mohnot 2004, Chhangani 2005, Waite 2007a,

Waite 2007b, Chhangani and Sivaperuman 2009). In arid regions, ENSO alternates between El Niño events leading to rainfall and La Niña events leading to drought. By suppressing plant growth and recruitment, drought can have bottom-up trophic effects

(Holmgren and Scheffer 2001), leading to vertebrate population crashes (Harrison 2000,

Cleary et al. 2006). In the case of the recent crash of Gyps spp., the role of climate seems worth considering because the crash coincided with the major ENSO-induced drought of

2000, which apparently caused a simultaneous decline of the mammal community in a nearby wildlife sanctuary in the Aravalli Hills (Waite et al. 2007a). Given the widespread use of diclofenac among livestock (Green et al. 2004) it seems plausible that the coincidental crash of vultures could have been exacerbated by the same ENSO-induced drought.

Here, we examine the impact of ENSO on population dynamics of the Indian vulture at11 village sites scattered throughout western Rajasthan, India. We ask whether ENSO- induced drought synchronised population dynamics across the region. We also model village specific vulture population growth rates in order to discover what variables contributed to respective trends in vulture populations over time.

METHODS

Study area

The study area was situated within 175 km of Jodhpur (26o18N, 73o08E), a major city

(population: ~1.5 million) in western Rajasthan (Figure 1), at the eastern edge of the Great

Indian (Thar) Desert. The region is covered with open scrub forest. Dominant tree and shrub species include: Acacia senegal, A. nilotica, Euphorbia caducifoliya, Anogeisus pendula,

Mytenus emarginata, Greioia tenex, Ziziphus numularia, Prosopis cineraria, Capparis deciduas, Tecomella undullata and the exotic invasive Prosopis juliflora. Temperatures can reach highs of ~48oC in May and June, and lows of ~1oC during winter. Annual rainfall averages 360 mm, with 90% occurring during the monsoon season (July – September).

Monsoon failure can lead to hydrologic and vegetative drought.

Villages are composed of a central housing complex (multiple homes all within a

1km area) and surrounding agricultural and community lands. Boundaries between villages are established via landmarks on individual or village lands.

Study species

The Indian vulture is a large (5.5-6.3 kg) scavenging raptor found almost throughout the

Indian subcontinent (for a detailed description see Ferguson-Lees et al. 2001, BirdLife

International 2009). The Indian vulture feeds almost exclusively on carrion, mostly livestock carcasses. It typically nests in colonies on cliffs or crags, but also nests in trees where rocky habitat is unavailable (Chhangani and Monot 2004, Chhangani 2005,

Chhangani and Sivaperuman 2009). Nesting begins in November and lasts until May with successful pairs raising one offspring per breeding season.

The Indian vulture was recently granted species status, when two subspecies of the long-billed vulture were split into Indian and slender-billed vultures (Rasmussen and Parry

2001). Despite morphological similarities and some overlap in geographic range between these species, we can state confidently that we collected data on Indian vultures only. None of our vultures showed the diagnostic field marks of slender-billed vultures (i.e., slender bill and dark brown plumage on body and head [Ferguson-Lees et al. 2001]) and none of the local populations we monitored occurred in the area of known overlap in geographic ranges (IUCN 2007).

Field methods

Monitoring of vultures in and around 11 villages within 175 km of the city of Jodhpur

(Figure 1) began in the winter of 1996-1997. Annual counts were conducted until 2004-

2005. Vultures were counted at known nesting sites within each village in trees and rock formations. Local villagers were often consulted as to where they had observed vulture nesting sites. Observations were also made from vehicles during road transects and opportunistically during periodic visits to villages in the region. The known local breeding populations included in our analysis were surveyed extensively during each nesting season and are an accurate representation of local populations.

These data provide an index of local abundance of Indian vultures during the nesting season. Because vultures were not individually identifiable and because our monitoring efforts were periodic rather than continuous, we do not know the precise relationship between observed and actual abundance. Therefore, our analysis is based on the assumption that the index of abundance is proportional to actual abundance and that this proportionality holds steady across years and locations.

Climate data

We used an El Niño Southern Oscillation (ENSO) index as a predictor of Indian vulture population dynamics. ENSO is the most important oceanic-atmospheric driver of year-to- year variability in global climate. We used the Multivariate ENSO Index (MEI)

(http://www.cdc.noaa.gov/people/klaus.wolter/MEI/mei.html#ref_wt1), which is based on six variables observed over the tropical Pacific Ocean: sea-level pressure, zonal and meridional components of surface wind, sea surface temperature, surface air temperature, and total fractional cloudiness. MEI is computed for sliding bimonthly periods, and standardized with respect to season and the reference interval 1950-1993. We used mean

MEI values during winter, December-February. Negative values represent the cold ENSO phase (La Niña), while positive values represent the warm ENSO phase (El Niño).

Data analysis

All of the time-series analyses described below used SPSS routines (2007). All statistical tests were two-tailed.

Cross-correlation, ‘Leave-one-out’ analysis of synchrony

To evaluate whether ENSO might have synchronized Indian vulture population dynamics across the region, we began by cross-correlating (lag = 0) the time series of the local populations. We did so after first transforming (natural logarithm) and differing (1 year) the data, thereby converting the time series of raw count data into time series of annual population growth rates (i.e., ln[Nt]-ln[Nt-1]) (N= number of vultures). We asked whether local population growth rate from year to year covaried across the region, as expected if climate had an overwhelming influence.

Finding considerable evidence for synchronous population dynamics, we next evaluated whether this synchrony might have been driven by two La Niña events, one ending in 1997 and another spanning 1999 (Figure 2). We reasoned that these events might have triggered downturns in local populations throughout the region and so might have had a major synchronizing influence. To evaluate this possibility, we repeated the cross-correlation analysis described above while leaving out (without replacement) the interannual growth for each year. We reasoned that leaving out growth rate estimates for years coinciding with drought would yield a suppressed mean correlation coefficient. Any such cases would indicate that drought contributed to the overall synchrony in population dynamics across the region.

Time series analysis of ENSO impact

We used Auto Regressive Integrated Moving Average (ARIMA) models (Carroll and Pearson

1999, Becerra-Muñoz et al. 2003) to evaluate more directly whether ENSO influenced

Indian vulture population dynamics. Specifically, we asked whether annual population growth rate could be adequately explained by MEI during year t alone or together with MEI during year t-1. We used first-order models (i.e., with autoregressive lag term for 1 year

[AR1]). Parameters were set as follows: p = 1 (lag of 1 year), d = 1 (time series differenced once to make it stationary, following natural logarithm transformation), and q = 0 (order of moving average set to zero).

We used the Akaike Information Criterion (AIC) (Bozdogan 1987) for model selection to evaluate whether the dynamics of each local population were best described by AR1 alone, by AR1 and MEIt, or by AR1, MEIt and MEIt-1. Best models were identified as those with minimal AIC values. We used stationary r2 as the goodness-of-fit measure.

Time series analysis for individual villages

We first determined whether the best AIC-based RIM model for each village’s local population included neither MEIt nor MEIt-1. If the best model did not include either metric for ENSO then we assumed the local population was somewhat buffered by the negative effects of drought. We next evaluated whether each model including one or both of these

MEI predictors had especially poor fit for this local population. Finally, we asked whether each local population’s time series showed low autocorrelation (transformation: natural logarithm; differencing: 1 year), particularly when MEI predictors were included. We thus reasoned that Indian vulture population dynamics in any village might have been largely independent of MEI and representative of an underlying white noise process.

RESULTS

ENSO and synchronous population dynamics throughout the region

The MEI time series (Figure 2) includes two La Niña events, as shown by the prolonged series of negative MEI values extending into 1997 and the subsequent series spanning

1999. Such events can lead to monsoon failure. The major event spanning 1999 caused consecutive monsoon failures leading to a severe vegetative drought throughout Rajasthan in 2000. Our time series of Indian vulture population growth rates suggests that both of these La Niña events had a latent impact. Table 2 reveals that all 11 local populations shrank from 1997 to 1998 and again from 1999 to 2000. These universal downturns suggest that ENSO-induced drought might have synchronized population dynamics across the region.

Table 3 confirms a high degree of region-wide synchrony. The time series of population growth rates were universally positively correlated across the 55 pairwize combinations of local populations, with 18 of these correlations reaching nominal significance (i.e., P < 0.05, unadjusted for multiplicity). Our leave-one-out analysis revealed three interannual growth rates that most strongly influenced this synchrony. Specifically, leaving out the growth rates for these interannual intervals (1997-1998, 1999-2000, and

1996-1997) suppressed the mean correlation coefficient (0.309, 0.534, and 0.519, respectively) below the original value of 0.564 (i.e., the mean of values in Table 2). The strong region-wide synchrony was driven by the universal downturns in 1998 and 2000, following the two La Niña events (and by the universal upturn from 1996 to 1997) (Table

2).

Time series analysis using ARIMA models further suggests that ENSO influenced population dynamics. Table 4 shows the best AIC-based model included at least one MEI predictor for all eleven villages in the study. Table 5 shows relatively good fit for models including one or both MEI predictors.

Individual village variation of vulture populations under drought conditions

We report several lines of preliminary evidence that Indian vultures in several villages were buffered against ENSO impacts. Dugar, Khejarli, and Chotila’s respective local populations had the most favorable population growth rates with Dugar having a net population change of zero (Table 4 . Second, for all three of these villages’ local populations, the fit of ARIMA model was the poorest of all when one or both of the MEI predictors were included (Table 5). Lastly, autocorrelation analysis revealed that annual population growth rates for each of the three villages’ respective populations were serially

12 independent (Table 6), as expected for an underlying white noise process. Overall, the

Indian vulture populations in Khejarli, Dawra, and Sumer appeared to be less influenced by the vagaries of ENSO.

DISCUSSION

We offer several insights regarding population dynamics of the Indian vulture, a critically endangered species of the Indian subcontinent. Population dynamics were synchronous across western Rajasthan, as if a large-scale abiotic driver overwhelmed intrinsic regulatory factors. Our analysis suggests ENSO was the major synchronizing factor. In part, a major La Niña-induced drought in 2000 apparently triggered a universal downturn in local populations throughout the region. Despite this widespread synchrony, several local populations appeared to be less influenced by ENSO, providing the impetus to further investigate other possible influences on Indian vultures nesting in these villages.

ENSO and Indian vulture population dynamics

Our results indicate that ENSO events synchronized Indian vulture population dynamics throughout western Rajasthan. We suggest that the major La Nina event spanning 1999 might have played a role in triggering the downturn in Indian vultures throughout western

Rajasthan (Chhangani 2005). Until now, diclofenac has been implicated as the major, perhaps only, cause of the widespread decline of this species (Green et al. 2004). Our intent is not to discredit evidence that diclofenac poisoning has played a major causal role or to question IUCN’s diclofenac-related rationale for recently listing the Indian, white- backed, and slender-billed vulture as being critically endangered. We simply emphasize that the potentially additive or synergistic role of climate should not be ignored, especially

13 in an arid region where ENSO-induced drought is now known to cause catastrophic die-offs of vertebrates, even in a protected area and presumably in the absence of dicofenac poisoning (Waite et al. 2007a). Future attempts to model causes of widespread population declines should begin by incorporating the climate cycle as the primary abiotic driver.

Buffering from drought in villages

While any attempt to understand population dynamics should consider the role of climate, there are several contributing factors, including human activity, that can potentially buffer vertebrate populations against climatic vagaries, even protecting them against catastrophic die-off during drought (Waite et al. 2007b). In keeping with the growing emphasis on community-based conservation in developing countries (Agrawal and Gibson 1999, Lepp and Holland 2006, van Eden et al. 2006, Sommerville et al. 2010,

López-Arévalo et al. 2011, Waylen et al. 2010) we would like to specifically explore the potential impact humans living in the region may have on Indian vulture populations.

Permanent food and water supply, and nesting habitat are factors that are positively associated with sustained vulture populations (Chhangani 2005, Monadjem and Garcelon

2008, Mateo-Tomás and Olea 2010). Future research will further investigate village specific characteristics such as small cliff formations (Dawra village), large temples (Dugar village), and permanent water bodies (Khejarli and Chotila village) and the potential effects on Gyps vulture populations.

Khejarli village is of particular interest however, because of the presence of a unique caste of local people that may have a positive influence on vulture nesting habitat. The

Bishnoi people of western Rajasthan could be an exemplar of built-in community-based conservation of vultures and other species of conservation concern. The Bishnoi faith,

14 founded in 1485 by Jambeshwar, is based on twenty-nine principles (“Bish-noi” translates to “twenty-nine” , one of which specifies doing no harm to any living creature (Fisher 1997,

Brockmann and Pichler 2004). In particular, they hold sacred a key tree species, Prosopis cineraria (locally known as the Khejeri tree) (Brockmann and Pichler 2004). Bishnoi do not lop off branches of Khejeri to feed livestock presumably providing more favorable nesting habitat for vultures (Brockmann and Pichler 2004). Vulture populations in areas with proper nesting habitat may better maintain village specific birth rates during drought years where adults may emigrate to other areas or experience greater juvenile mortality.

Presumably vultures would avoid areas where Khejeri trees are heavily lopped and choose to nest in areas where trees are unlopped, specifically areas with relatively high Bishnoi populations.

Though more research is on the subject of a “Bishnoi effect” is necessary we speculate that a portion of the buffering that occurred in Khejarli village is attributable to the protection practices of the Bishnoi people. The Bishnoi have a long history of protecting this tree. In the year 1730 three hundred sixty three Bishnoi men, women, and children sacrificed their lives in protest of Maharaja Abhay Singh felling of Khereji trees in Khejarli village for renovations of his palace (Brockmann and Pichler 2004). The Bishnoi, who make up 41% of the ~1200 residents of Khejarli, still actively protect this tree (personal observation) and may provide an oasis of suitable, undisturbed nesting habitat for vultures.

Any strong inferences about whether and how Bishnoi activities benefit species of conservation concern cannot, however, be solely based on one species in one village.

Future work will expand the geographic scope of this preliminary study and attempt to

15 uncover the underlying basis for how Bishnoi provide a safe haven for endangered wildlife species.

Implications for conservation-breeding strategies

In response to the vulture crisis, a captive breeding program was initiated eight years ago for Indian, white-back, and slender-billed vultures in Pinjore, India (Vulture Rescue 2008).

This program aims to protect remaining populations while giving the country’s ban on diclofenac time to take effect, and then to release captive-reared vultures into the wild. Our results prompt two suggestions for those devising strategies for reintroduction. First, given our evidence for region-wide universal downturns of Indian vulture populations following La Niña events, we suggest that releases should not be scheduled to coincide with ongoing or imminent La Niña events. Straightforward monitoring of MEI data could help optimize the timing of release.

Second, while releasing captive-reared vultures in protected areas (i.e., national parks or wildlife sanctuaries) seems like the obvious preferred approach it is important that the potentially positive effects of human activity on wildlife are also considered. The miraculous recent recovery of spot-billed pelicans (Pelecanus philippensis) in Cambodia

(400% increase in population size in 7 years) provides a striking example of successful local community involvement with wildlife conservation (Wildlife Conservation Society

2008). Beyond the speculative benefits provided by their lifestyle the Bishnoi also opportunistically perform wildlife rescue and rehabilitation (e.g., of injured black buck antelope [personal observation]). Whether recovery efforts succeed depends on the fate of wild populations after reintroduction. Ideally, the ongoing captive-breeding efforts will not be squandered by the failure of wild populations in the future. Well-timed release of

16 captive-raised vultures into the wild, combined with local community participation, could improve the prospects for successful recovery of vulture populations.

CONCLUSIONS

We found that Indian vulture population dynamics were synchronized across a broad region of western Rajasthan, apparently by ENSO events. Although ENSO apparently impacted local populations in a parallel way across this region, some local populations were partially buffered from the effects of drought. Factors contributing to drought buffering, including human activity, are currently being investigated. Future attempts to model the effects of putative causes of widespread vulture declines (e.g., via diclofenac poisoning) should incorporate the potentially significant impact of climate. Ongoing research aims to explore whether and how human activity may directly impact the conservation status of the Indian vulture and other endangered wildlife species, within officially unprotected areas.

Chapter 3: Distribution of Prosopis cineraria on agricultural farmland in Western Rajasthan: influence of the Bishnoi people on tree abundance

ABSTRACT

Prosopis cineraria, locally known as the Khejeri tree, is an ecologically and economically important tree species in Western Rajasthan. Khejeri trees provide food and building material as well as improve agricultural productivity for human populations. Additionally

Khejeri trees provide shelter and food for a variety of wildlife species in the area. The

Bishnoi people of Rajasthan practice a religion in which Khejeri trees are considered sacred and protected. Local researchers and villagers claim that there are more Khejeri trees in villages with Bishnoi people than in villages without, although this has never been systematically investigated. In this study I conduct a comparative study of Khejeri tree populations among nineteen villages, eight with Bishnoi populations and eleven without.

Analysis of transects reveals that villages with Bishnoi contain significantly more Khejeri trees than villages without Bishnoi, as predicted. Further, I found that an interaction between the presence of Bishnoi and household income is associated with Khejeri tree abundance. In Bishnoi areas, tree abundance is positively associated with income. This relationship, however, does not exist for villages without Bishnoi. These results are consistent with the suggestion that Bishnoi practices and/or beliefs have an influence on the abundance of this keystone species, a novel example of the influence of human beliefs on the conservation of plant populations. 18

INTRODUCTION

India, and especially arid rural India, faces growing ecological problems (Tewari and Arya

2005, Green et al. 2006, Khanna et al. 2008, Singh et al. 2008, Chaudhary and Aggrawal

2009). Increased drought and drought severity loom as one of the biggest challenges people of India will have to confront in the next ten years (Waite et al. 2007a, Khanna et al.

2008, Chaudhary and Aggrawal 2009, Francis and Gadgil 2009). Monsoon failures that lead to increased drought threaten agricultural productivity (Challinor 2007, Khanna et al.

2008, Singh et al. 2008, Chaudhary and Aggrawal 2009). Agricultural productivity is vital to the people of this region and must be maintained in order to preserve the primary sector of the local economy (Fisher 1997, Brockmann and Pichler 2004, Jones 2008, Singh et al.

2008, Chaudhary and Aggrawal 2009).

Prosopis cineraria is a keystone tree species in arid northwest India (Fisher 1997,

Brockmann and Pichler 2004, Sharma et al 2008). For wildlife Prosopis cineraria provides food and shelter, especially during the dry season’s extreme temperatures and drought

(Goyal et al. 1988; Fisher 1997). Villagers use Prosopis cineraria, locally known as the

Khejeri tree for fuel wood, livestock fodder, building material, and as a dietary supplement

(Fisher 1997, Singh et al. 2007). The fruit of the Khejeri are used in several local dishes

(Fisher 1997) and the young, leafy branches are routinely lopped off and feed to goats and sheep (Goyal et al. 1988, Pasiecznik et al. 2001, Vaithiyanathan 2007). Several studies have also demonstrated the benefits of the presence of Prosopis cineraria on agricultural lands

(Puri et al. 1994, Kaushik and Kumar 2003, Singh et al. 2007, Yadav et al. 2008, Singh

2009). These benefits include increased crop yield, retention of topsoil, increase soil

19 moisture, and increased soil nutrient decomposition (Singh et al. 2007, Yadav et al. 2008,

Singh 2009).

The Bishnoi people are a caste of agro-pastoralists found in Western Rajasthan that practice a specific religion that is a sect of Hinduism (Fisher 1997; Brockmann and Pichler

2004). Among their chief tenets is the protection and maintenance of the Khejeri tree, which they deem to be sacred. According to Bishnoi history, the founder of their faith,

Jambeshwar, received inspiration for the twenty-nine principles that would later become the Bishnoi faith while sitting underneath a Khejeri tree (Brockmann and Pichler 2004).

Because of Prosopis cineraria’s inspirational power and ecological significance, Bishnoi people are never to lop the branches or otherwise harm any Khejeri tree (Fisher 1997;

Rajasthan (Fisher 1997; Brockmann and Pichler 2004).

For the past three decades villagers, local researchers, and government officials have associated Bishnoi with higher numbers of Khejeri tree when compared to non-

Bishnoi areas (personal observation). The association, while widely regarded as fact, has never been systematically examined. Determining whether the widely known ‘fact’ that

Bishnoi are associated with more Khejeri trees has a basis in science holds potential value for agricultural management in this region of India. The first objective of this study is to determine if there are in fact differences in Khejeri tree abundance between villages with

Bishnoi versus villages without Bishnoi. The second objective of this study is to determine what village characteristics best explain the abundance of Khejeri trees among villages.

METHODS

Study Area

The rural area of Rajasthan surrounding Jodhpur, Rajasthan’s third largest city

(approximately 1.5 million), is characterized by extreme temperatures and frequent drought (Fisher 1997). This area is the eastern border of the Thar Desert, the world’s most populated desert (Rahmani and Sankaran 1991, Naewboonnien 2007). Most of the non- rural population practice some form of agropastoralism (Tewari and Arya 2005, Fisher

1997). People in this area depend heavily on monsoon rains for growing crops, drinking water, and feeding livestock (Fisher 1997). Survival and income depend heavily on the land and resource management practices of villagers (Fisher 1997, Tewari 2004).

The study area is situated within 50 km of Jodhpur (26o18N, 73o08E) (Figure 3).

The region is covered with open scrub forest. Dominant tree and shrub species include:

Acacia senegal, A. nilotica, Euphorbia caducifoliya, Anogeisus pendula, Mytenus emarginata,

Greioia tenex, Ziziphus numularia, Prosopis cineria, and the exotic invasive P. juliflora. The climate is hot and dry. Temperatures can reach highs of ~48oC in May and June, and lows of ~1oC during winter. Annual rainfall averages 360 mm, with 90% occurring during the monsoon season (July – September). Monsoon failure can lead to hydrologic and vegetative drought.

All villages in the study area are organized more or less the same way with centralized housing structures; agricultural, fallow, and sacred land on the outskirts; water catchment land (non-arable land used to funnel water into the water body); manmade water body (small pond area for collecting monsoon rain water); and wells (usually located

21 near the water body). The near majority of the land in the study area is under human use or has been transformed by humans with very little, if any, wilderness. Village boundaries are established by land ownership (land owned by members of different villages) and/or by roads.

Study Species

Prosopis cineraria, commonly known as the Khejeri tree, is a native tree species of northwest India and the Thar Desert. Thin leaves, thorny branches, and yellow flowers characterize the Khejeri tree. Khejeri are well adapted to arid environments (annual rainfall of less than 500mm) and thrive on cultivated land (Singh et al. 2007). Trees in the study area can be as tall as 12m, but were usually less than 8m (personal observation). In addition to providing many benefits to humans and livestock, the Khejeri tree is also an important source of food, shelter, and nesting habitat for various wildlife spieces in

Rajasthan (Chhangani 2005, Hall et al. in press, personal observation).

Khejeri Tree Transects

Nineteen villages in the study area were chosen for Khejeri Tree transects. Villages were selected based on the fact that the partnering NGO, The School of Desert Sciences (SDS), had carried out prior social and ecological studies.

Each village center was marked using GPS. Village centers were chosen based on their central location within the village residential area and the self reported social and economic center of villager activity. Village centers were often located in front of or near the Tako’s (the head family of the village) residence. I used Garmin mapping program

(MapSourcetm) to establish eight transect starting positions 500m from each village center.

The eight starting point compass headings were designated north (0o), northeast (45o), east

(90o), southeast (125o), south (180o), southwest (225o), west (270o), northwest (315o). I then marked transect end points 500m from each of the eight compass heading points using the same Garmin software. I then loaded each drawn transect map onto the GPS unit and returned to the village to walk each of the eight transects. I marked the location of each

Khejeri tree within 10m of the transect line along each transect line in every village (Figure

4a).

In several villages one or more transects crossed the border of a neighboring village before the end of 500m. In such cases the requisite number of ‘substitute’ transects were mapped along compass headings equidistant from two established compass headings. For example, a substitute transect along heading north of northeast (SSW; 193.5o) would be drawn and then walked if any other heading proved insufficient (Figure 4b).

Each village yielded different numbers of transects from farmland vs. transects from community land. Because this paper is primarily concerned with the ecological implications of Khejeri tree numbers on farmland rather than total village Khejeri tree numbers I used the five transects with the highest number of Khejeri trees on farmland rather than total Khejeri trees for each village transect. I selected transects with the highest number of Khejeri trees rather than randomly selecting five transects in order to evaluate villages at their highest population of trees rather than potentially underestimating tree populations in some villages.

Village Demographic Data

In each of the nineteen villages I conducted a semi-structured interview (Bernard 2001) to collect data on village characteristics. A focus group of 3-18 adult male (18 years or older) village residents voluntarily gathered to be interviewed. In each of the focus groups one

23 focus group member served as the spokesperson for the group. Other members provided input and comments, but in each case there was a clear final authority on the village information that was reported and recorded. Interviews lasted from half an hour to one and a half hours. I visited each village once for interviews.

In addition to focus group interviews I collected and otherwise verified interview data from data collected by the School of Desert Sciences (SDS) during village surveys in

2007 and from the 2001 Rajasthan census data available online (Sharma 2011). Though I collected a variety of demographic, ecological, and ethnographic data I was specifically interested in four village characteristics that could have an impact on Khejeri tree numbers

(Goyal et al. 1988, Lal 1991, Fisher 1997, Brockmann and Pichler 2004): average annual household income, total population, total farmland area, and total livestock population

(Appendix A).

Data Analysis

Bishnoi vs. Non-Bishnoi villages

In order to determine if there is a significant difference between the two classes of villages

(with Bishnoi and without Bishnoi) in the number of Khejeri trees counted along village transects as well as the aforementioned village characteristics of interest I analyzed the transect data using two non-parametric tests, the Mann-Whitney U-test and Multiple

Response Permutation Procedure (MRPP) (McCune and Grace 2002). MRPP provides values of within group heterogeneity (A) and separation between groups (T statistic). An A value of 0 indicates within group heterogeneity to be equal to that expected by chance while an A value of 1 indicates perfect within group agreement or complete homogeneity

(McCune and Grace 2002). MRPP tests that evaluate ecological data often yield an A value

24 of 0.1 or below (2002). Finally, the more negative the T value statistic the greater the separation between the groups (2002).

Ordination

To investigate whether a pattern among the villages exists, given the four demographic characteristics, I used the Non-metric Multidimensional Scaling (NMS) (Clarke 1993,

McCune and Grace 2002). NMS is a nonparametric ordination technique similar to Principle

Components Analysis (PCA) (McCune and Grace 2002) and is becoming increasingly popular as an ordination technique for analyzing ecological community data (Clarke 1993,

McCune and Grace 2002, Daley 2008, Bruelheide et al. 2011, Shipley et al. 2011). I used the program PC-ORD (McCune and Mefford 2009) to conduct the NMS analysis.

Covariate Analysis

I used analysis of covariance (ANCOVA) to model the abundance of Khejeri trees based on the presence of Bishnoi and the axis values from the NMS analysis. The purpose of this analysis is to determine what variables, axis values from the NMS analysis and/or presence of Bishnoi, can explain the distribution of Khejeri trees. Along with evaluating a model of these individual variables I also evaluated whether there was a significant effect of the interaction among variables for explaining Khejeri tree distribution. The final reduced model is the result of sequential removal of non-significant interactions. I performed the

ANCOVA analysis using the program SPSS (v.19.0).

RESULTS

There was a significant difference between the number of Khejeri trees within the transect area between villages with Bishnoi and villages without Bishnoi (Table 7). Villages with

Bishnoi populations have a higher average number of Khejeri trees than villages without

Bishnoi. Results from the Mann-Whitney U-Test indicate a significant difference between village types in average annual household income while results from MRPP analysis reveal a nearly significant difference (Table 7).

The results from NMS analysis revealed an appropriate 2-dimensional solution.

Final solutions of two or three dimensions are considered appropriate solutions for ecological data (McCune and Grace 2002). Axis 1 captured 82.3% of the data while axis 2 captured an additional 16.0%. Overall the ordination explained 98.3% of the data with a final stress value of 3.90, well below the threshold of 20 indicating an ordination that effectively describes the data (McCune and Grace 2002). Axis 1 is most closely associated with average annual household income with additional notable associations with total population and total farmland area (Table 8). Axis 2 is again most closely associated with average annual household income with a notable contribution from total livestock (Table

8). The axis scores suggest that average annual household income is the most influential village characteristic for describing the nineteen study villages (Table 8).

Using the axis scores from the NMS analysis (Appendix B) ANCOVA results indicate that Bishnoi and the variation in the interaction between Bishnoi and axis 2 are non- randomly associated with Khejeri tree abundance. In the full model the interaction between Bishnoi and axis 1 was nonsignificant (F=2.867, df=1, p=0.120) and was removed.

Axis 1 also did not have an effect on Khejeri tree abundance (Table 9). A scatter plot of the axis 2 versus Khejeri tree abundance (Figure 5) shows a noticeable difference in the trend lines between villages with Bishnoi and villages without Bishnoi. Pearson’s correlation analysis of axis 2 values and Khejeri tree numbers for the two village classes reveals a

26 significant correlation between the two factors in villages with Bishnoi (-0.708, N=8, p=0.049) as opposed to a highly non-significant correlation in villages without Bishnoi

(0.147, N=11, p=0.676). The combined results from the scatterplot and correlation analysis suggest the abundance of Khejeri trees in villages with Bishnoi appears to be influenced by income and livestock.

DISCUSSION

Village transects and subsequent analysis reveal that there is a significant difference in the number of Khejeri trees surveyed in villages with Bishnoi versus villages without Bishnoi.

Average annual household income (AAI) in villages with Bishnoi is also significantly greater than in villages without. NMS analysis indicates that AAI is the highest contributing factor explaining the variation among villages while total population, farmland area, and livestock are noticeable contributors. Subsequent covariate analysis shows that there is a significant influence of the presence of Bishnoi when controlling for both axes and while AAI by itself does not appear to effect Khejeri tree abundance (Tabel 3), the interaction between Bishnoi and axis 2 (AAI and livestock population) have a significant effect on Khejeri tree abundance in villages with Bishnoi populations.

There may be some concern that difference in influence of axis 2 in villages with

Bishnoi versus villages without Bishnoi is due to a non-linear relationship between the two village classes. Figure 5 shows a slight overlap between the best-fit lines of the two village classes, but little similarity in y-values at the intersection point. Because of the lack of similarity in the data around the intersection point I am confident that difference between

27 villages classes in relation to axis 2 is not due to confounding factors. This difference in correlation further supports a difference in influence of axis 2 between village groups.

Axis 1 is composed of village characteristics most likely associated with agricultural productivity (Table 9). Khejeri trees are known to provide benefits to agricultural productivity (Puri et al. 1994, Kaushik and Kumar 2003, Singh et al. 2007, Yadav et al.

2008, Singh 2009) yet axis 1 has no significant effect on Khejeri tree abundance in either village type. The significant effect of axis 2 on Khejeri tree abundance suggests that the income associated with livestock is a greater contributing factor to tree abundance than income associated with agriculture. This is a plausible dynamic given the fact that water is the primary limiting resource for agriculture (Fisher 1997) while fodder plays a more important role in sustaining livestock populations (Fisher 1997) with much of the livestock fodder coming from Khejeri trees (Fisher 1997, Pasiecznik et al. 2001, Brockmann and

Pichler 2004). Given the value Khejeri trees hold for livestock (Fisher 1997) it might be expected that wealthier individuals would put some of those resources towards preserving or maintaining a valuable natural resource. The collective lack of influence of total population, farmland area, and total livestock on Khejeri tree abundance is also not surprising given the fact that these characteristics do not differ between the two village types.

While the four village characteristics measured had little effect on Khejeri tree abundance (or interacted with Bishnoi presence to influence Khejeri tree abundance) there are additional demographic and ethnographic characteristics that could potentially explain tree abundance. The impact of water availability is an important factor to consider when examining the ecology of this region (Fisher 1997) that could potentially influence Khejeri

28 tree abundance. It is, however, worth noting that despite the fact that farmland area is statistically similar between both village types there is a significant difference in the number of Khejeri trees in the proportion of farmland area sampled. This suggests that farmland areas in villages with Bishnoi have higher Khejeri tree densities than villages without Bishnoi within 1km of the village center.

The difference in the effect of axis 2 on Khejeri tree abundance between the two village types suggests that monetary resources have a greater influence on Khejeri tree abundance in villages with Bishnoi than in villages without Bishnoi. It follows that a group of people dedicated to protecting Khejeri trees would allocate more resources towards a species they deem sacred, but it is worth investigating whether or not the influence of

Bishnoi principles extends to members of a different caste. As a collective the Bishnoi hold considerable political and economic influence within and among villages they inhabit

(Fisher 1997, Brockmann and Pichler 2004, personal observation).

The next step in investigating the dynamics of Khejeri tree abundance and the

Bishnoi people is determining what effects differential abundance has on human populations, specifically through agricultural and livestock productivity. These factors are the primary sources of income for villagers in this region (Fisher 1997) and it is worth investigating whether or not having more or less Khejeri trees impacts village and/or household economics. Aside from investigating the potential benefits to human populations the impact of differential Khejeri tree abundance on wildlife species should also be explored. Gyps vultures are known to nest in Khejeri trees on village lands (Hall et al. in press) and provide an essential ecological service of livestock carcass disposal (Prakash et al. 2003, Green et al. 2004, 2006). Over the past two decades however, Gyps vultures have

29 experienced a catastrophic decline (over 95%) in population (Prakash et al. 2003). The successful reintroduction of Gyps vultures depends on suitable nesting habitat for reintroduced individuals, habitat that is in greater abundance in areas with Bishnoi populations. There are several plausible benefits to both human and wildlife populations that bear investigating given the results of this study. As such, these supposed benefits represent a “win-win” scenario for humans and wildlife species in an area threatened by continued human development.

CONCLUSIONS

I have conducted the first standardized comparative survey of Khejeri tree abundance between villages with Bishnoi populations and villages without Bishnoi, to test the widely held assumption that presence of Bishnoi is associated with larger populations of Khejeri trees. There are significantly more Khejeri trees, within the transect area, in villages with

Bishnoi populations than in villages without Bishnoi. Moreover, the presence of Bishnoi along with the interaction between Bishnoi presence and axis values representative of average annual household income and livestock population influence Khejeri tree abundance in villages with Bishnoi populations. The lack of any influence of the village characteristics examined on villages without Bishnoi point to a positive “Bishnoi effect” on

Khejeri tree abundance. Further investigation into the potential influence of the Bishnoi effect on agricultural productivity and endangered wildlife species is recommended.

Chapter 4: Patterns of blackbuck distribution on village lands in Western Rajasthan

ABSTRACT

The blackbuck antelope is an important ecological and symbolic species in India. Over the last three decades populations have rebounded from sharp declines brought on by a variety of human activities. Continued human population growth and development currently threaten populations despite the steps taken by the Indian government to protect blackbuck. Here I examine differences in populations of blackbuck in two sets of villages in

Western Rajasthan. One group of villages is home to the Bishnoi people, a local caste whose religion mandates they protect blackbuck. Local researchers and villagers claim that there are more blackbuck in villages with Bishnoi people than villages without, although this has never been systematically investigated. Here I test this prediction. Of the nineteen villages surveyed over a period of two years, villages with Bishnoi people differ significantly from non-Bishnoi areas in their respective populations of blackbuck. These results support the locally held belief that blackbuck are positively associated with Bishnoi and provide a foundation for further investigation into the potential positive impacts of human beliefs/practices on threatened species in a human dominated landscape.

INTRODUCTION

Human population growth and development threaten many ungulate species around the globe, particularly in India, as further habitat is put to human use, leaving less wilderness habitat for other species (Dzialak et al. 2011, Singh et al. 2010, Karanth 2009, Krishna et al.

2009, Mallon and Jiang 2009, Madhusudan & Mishra 2003,). Ungulate species play an important role in local ecology and economy across the world (Ismail et al. 2011, Singh et al. 2010, Karanth 2009, Fisher 1997). In India ungulates serve as prey animals for various species including humans as well acting as seed dispersers for both wild and domesticated plants (Karanth 2009, Madhusudan and Mishra 2003, Fisher 1997, Jhala 1997, Goyal 1988).

This problem is especially prevalent where human population growth and expansion threaten many native species (Naewboonnien 2007). The blackbuck antelope (Antilope cervricapra) is one such threatened species that occupies a unique cultural space that may enhance its conservation prospects (Brockmann & Pichler 2004, Fisher 1997).

Blackbuck antelope are a medium sized (25-40kg) ungulate species historically found across India and in parts of Pakistan and Nepal (Karanth 2009, Mallon 2008, Isvaran

2001, Jhala 1997). Blackbuck populations declined during the 20th century such that the species is now only found in India (Mallon 2008). Efforts to protect and conserve blackbuck began in the 1970s and effectively stimulated the rebound of populations from approximately 25,000 to now an estimated 50,000+ animals (2008). These numbers are predicted to decline over the coming decades as livestock and human population growth threaten the habitat of blackbuck populations (2008).

A possible refuge for this threatened species exists among a caste of local people, called the Bishnoi, living in the northwestern state of Rajasthan (Mallon 2008, Brockmann

& Pichler 2004, Fisher 1997, Goyal 1988). Their religion is a sect of Hinduism and was established in 1495 by a man named Jambeshwar. Jambeshwar established the Bishnoi faith based on twenty-nine principles (Bishnoi means “twenty-nine” that were to help followers live a more healthy and prosperous life (Brockmann and Pichler 2004). Among these principles Bishnoi are never to cut down living trees nor harm or consume wildlife or livestock. Blackbuck are particularly sacred as Bishnoi believe the antelope are the manifestation of their guru Jambeshwar(2004). This specific Bishnoi principle has gained traction in the Rajasthani legal system such that it is illegal to hunt antelope on Bishnoi lands. In 1998 famous Bollywood actor Salman Khan was caught hunting blackbuck on

Bishnoi lands in Rajasthan and sentenced to 3 years in prison for the offense (he served only 3 months) (The Hindu 2007). While it is widely assumed among local people and wildlife biologists that Bishnoi areas hold comparatively larger populations of blackbuck

(personal observation) there have been no scientific studies to confirm or deny these claims.

This study is the first to comparatively investigate populations of blackbuck antelope in village areas with and without Bishnoi. The purpose of this study is to test the prediction that there are measurable differences between blackbuck populations in these two village types. Describing potential factors that influence blackbuck species abundance is potentially valuable data to help inform conservation efforts already in place to preserve this species. To date any possible influence of the Bishnoi people’s religious practices on blackbuck distribution and abundance has not been evaluated, but yet may yield novel strategies for conserving and protecting an important species of India.

METHODS

Study Area

The study area is situated within 50 km of Jodhpur (26o18N, 73o08E) bordering the Thar

Desert. The landscape is mostly semi-arid open scrubland forest (Rahmani 1991;

Naewboonnien 2007). Dominant tree and shrub species include: Acacia senegal, A. nilotica,

Euphorbia caducifoliya, Anogeisus pendula, Mytenus emarginata, Greioia tenex, Ziziphus numularia, Prosopis cineria, and the exotic invasive P. juliflora. The climate is hot and dry.

Temperatures can reach highs of ~48oC in May and June, and lows of ~1oC during winter.

Annual rainfall averages 360 mm, with 90% occurring during the monsoon season (July –

September). Monsoon failure can lead to hydrologic and vegetative drought. Most of the local population practice some form of agropastoralism (Tewari 2005; Fisher 1997) People in this area depend heavily on monsoon rains for growing crops, drinking water, and feeding livestock (Fisher 1997). Survival and income depend heavily on the land and resource management practices of villagers (Fisher 1997; Tewari 2004).

Nineteen villages were selected in this area for the study. Village selection was based on a previously established research relationship between villages and the local

Non-Governmental Organization (NGO) the School of Desert Sciences (SDS). SDS, for the past fifteen years, has conducted various wildlife and socio-economic surveys in this study’s village areas as well as surrounding areas in Rajasthan. All villages in the study area are organized in the same way (Figure 6); a centralized housing area with agricultural and community lands located on the periphery. The near majority of the land in the study area is under human use or has been transformed by humans with very little, if any, wilderness.

Village boundaries are established by land ownership (land owned by members of different villages) or delineated by roads.

Study Species

The blackbuck antelope is a species native to the Indian subcontinent. They can most commonly be found in open plain or scrubland forest, but can also be found in marshy coastal plains (Isvaran 2003). Blackbuck are herd animals traveling in groups with herd sizes varying from as little as two to as many as several hundred (Mungall 1978, Ranjitsinh

1989). This species of antelope exhibits sexual dimorphism; males are larger in size (~5-

15kg heavier than females), possess two long spiraling horns (as long as 75cm), and have a distinctive dark brown or black coloration for which the species is named (Figure 7)

(Mungall 1978, Ranjitsinh 1989, Jhala 1992).

Blackbuck populations were once widely distributed across India and in parts of

Pakistan and Nepal (Mallon 2008). Due to habitat destruction and hunting populations have shrunk over the last century (2008). Despite the symbolic value attached to blackbuck across many regions of India they are looked at as a pest species that feed on crops

(Naewboonnien 2007). In Rajasthan some non-Bishnoi castes routinely hunt blackbuck for food (Fisher 1997).

Survey of blackbuck populations

The survey period for this study began in 2007 during focus group interviews with villagers. During these interviews I asked the focus group if they had seen or heard of others who had seen certain wildlife species, blackbuck among them, in their respective village during the last five years. Before continuing surveys in the summer of 2009 local wildlife officials for SDS conducted a second round of interviews in May of 2009 to

35 determine if any village in the study had changed its status of blackbuck presence or absence. Local hunters were also consulted to verify the location of blackbuck populations due to the fact that they tended to travel through more than one village. I then performed a systematic one day survey of each village collecting data on the presence of blackbuck while running vegetation transects for an additional study. During these surveys I covered multiple land cover types (farmland, community land, livestock dumping ground, and village center) within 1.5km of each village center. Each sighting of blackbuck was recorded and marked using a handheld GPS unit. Once each village was systematically surveyed I returned to all but one of the villages where I had observed blackbuck an additional time to record numbers of blackbuck. Due to logistical constraints I was not able to return to each

“blackbuck village” (villages observed and reported having blackbuck an equal number of times. I was able to survey two villages within 5km of my field station a total of three and four times (including my initial systematic survey) respectively.

Before beginning surveys for the 2010 field season SDS wildlife personal in March once again conducted preliminary interviews with villagers from each village and local hunters. Due to logistical constraints I chose to survey the same villages from 2009 that had both reported and where I observed blackbuck on village lands (Note: none of the villages, according to SDS interviews, changed its status of having blackbuck throughout the study period). Each of the villages with blackbuck were surveyed a total of four times, twice in the morning (from 0700-0900) and twice in the late afternoon/evening (1700-

1900).

During both years of blackbuck surveys (2009 and 2010) total number of blackbuck, number of each sex (male, female, or juvenile), time of day, and the occupied habitat type were recorded. Villages that did not have blackbuck were surveyed once in each year.

Data Analysis

Basic Statistics

The observation data were split into two groups, data collected in 2009 and data collected in 2010. To standardize blackbuck counts in both years I calculated sampling rates for each village by dividing the total number of blackbuck observed by the number of sampling occasions. For both years I first categorized villages based on whether or not Bishnoi were present. I then evaluated differences in total number of blackbuck and number of blackbuck per sampling occasion using multiple response parametric permutation (MRPP) on the computer program PC-ORD (MjM Software, version 5.10). MRPP is a rank-based non-parametric multivariate test that measures differences between a given set of samples

(in this case villages) in ordinate space using a constructed distance matrix (McCune and

Grace 2002). I used a Euclidean distance measure for this analysis (Lesica et al 1991).

There are three important output values from an MRPP test, A, p-value, and a T value test statistic. A indicates “chance-corrected within-group agreement”, which is a measure of within group heterogeneity (how different members of a group are to other members of the group). An A value of 0 indicates within group heterogeneity to be equal to that expected by chance while an A value of 1 indicates perfect within group agreement or complete homogeneity (McCune and Grace 2002). MRPP tests that evaluate ecological data often yield A values of 0.1 or below (2002). Finally, the T value statistic indicates separation

37 between groups (2002). The more negative the T value the greater the separation between the groups (2002).

Due to the unequal sampling of villages in 2009 I analyzed that year’s data using sample rates and not the total numbers. For data from 2010 I used the total number of blackbuck sampled in my analysis and assumed villages without blackbuck would have reported an absence of blackbuck had I surveyed the village three more times. The validity of this assumption is discussed below.

Model Analysis

Occupancy modeling techniques estimate species abundance in a given area under the assumption that species may be present without being detected (MacKenzie et al. 2002).

This technique is especially useful when only a proportion of each given area can be monitored and when selected sample areas are differentially surveyed, e.g. site one surveyed on four occasions while site two is surveyed twice (2002). Rather than estimating species abundance, occupancy modeling estimates the proportion of sites occupied

(MacKenzie et al. 2002, 2003, 2005, Karanth 2009).

There are three assumptions of this model: 1) the system is closed to changes in occupancy in a given season, 2) species are not falsely detected, and 3) detection of a species at one site is independent of detection at another site (MacKenzie et al. 2002).

Separate models were developed for both years’ blackbuck surveys in 2009 and 2010. The model output values are , the probability the given land area is occupied, and p, the probability that species are detected when present in the observed area.

In addition to modeling occupancy and detection probabilities for each year I used a multiple-season modeling technique (MacKenzie et al. 2003) to estimate local extinction ()

38 and colonization () probabilities, as well as  and p, across both survey years. I used the program PRESENCE (v 2.0; Hines 2006) to model individual years and across both observation years’ data. I used kaike’s Information Criterion ( IC (Bozdogan 1987,

MacKenzie et al. 2002, 2003, Karanth 2009) to determine whether the best model to estimate  and p assumes a constant value for  and p across sites or a site-specific value for p.

An additional advantage of occupancy modeling is one can model a population in a given study area with missing observations (MacKenzie et al. 2002). This technique makes it ideal to analyze the data in this study given the fact that not every village was surveyed an equal number of times. I ran two sets of analysis for each of the three modeling scenarios. The first set (“actual data”) is simply the raw data with missing observations. For the second set (“estimated data” I filled in missing observations as either absences for the eleven villages that reported absences or presence for the villages that had blackbuck but were not surveyed a total maximum number of four times. The purpose of running a second analysis for each year’s scenario is to see if there is any large difference in model parameter outputs resulting from the assumption that missing observations were indeed absences. Villages that reported no blackbuck and that were surveyed once in a given year retained their status across four years of demographic data collection (2007-2010) and were extensively and systematically surveyed during a vegetation study in 2009.

Therefore, the assumption that ‘missing’ data represents absence of blackbuck is likely met.

RESULTS

Basic statistics

Four of the eight villages with Bishnoi populations also had blackbuck populations while three out of the eleven villages without Bishnoi had blackbuck (Table 10). There is a noticeable difference in number of blackbuck observed per sampling occasion between village types for both years (Table 10). However, MRPP analysis of number of blackbuck per sample for either year did not yield a significant difference between village groups in either year. Analysis of the total number of blackbuck surveyed in 2010, however, revealed a significant difference between blackbuck populations between villages with Bishnoi and villages without Bishnoi (T=-2.396, p=0.024; Table 10). A was equal to 0.107, a value within the expected range for ecological data (McCune and Grace 2002).

Modeling Analysis

Modeling analysis for all three scenarios yielded a total land occupancy value between 32-

39% and a detection probability value between 79%-100% (Table 11). For the combined years’ analysis the estimated extinction probability was between 9.31%-9.44% (Table 11).

An appropriate estimation of colonization probability could not be reached. There was little difference between the actual and estimated model outputs (Table 11). The best fit model for all year scenarios held  and p constant across all sites.

DISCUSSION

Analysis of data for 2010 revealed a significant difference between the respective populations of blackbuck in villages without Bishnoi versus villages with Bishnoi people.

Half of the eight “Bishnoi villages” contain populations of blackbuck while only three out of

40 the eleven “Non-Bishnoi villages” reported blackbuck populations. Occupancy modeling analysis estimates that 32-39% of the total land surveyed is occupied by blackbuck and that across a two year period the local extinction probability remained low at roughly 9%.

Because each village that had blackbuck was not equally sampled I could not analyze

2009’s data using MRPP. However, total numbers of blackbuck from 2010 were roughly similar to those from 2009 (Table 10) with one exception, Baniwas village. The area blackbuck were observed in this village is a community land space that borders a neighboring village with Bishnoi residents that was not part of the nineteen villages in this study. It is important to note that even with this large influx of blackbuck, possibly from a village with Bishnoi, there is still a significant difference between blackbuck populations in the two village types in 2010.

The similarity in model output values for actual and estimated data support the assumption that classifying ‘missing’ observations as absence of blackbuck does not strongly affect the results. Interview data further supports the validity of this assumption.

Occupancy modeling also allows for the implementation of site-specific covariates as a function of  and/or p. The beta values of site covariates, as determined by the model output, indicate which covariates are positively or negatively occupancy and detection probabilities (Donovan and Hines 2007). Given the size of the data set, however, I could not incorporate covariates into the model. Increasing the number of sites would allow for the investigation of the effect of village-specific characteristics (e.g. land area, total human population, livestock population, water catchment area, average annual income, etc) on the patterns of blackbuck distribution in this area of Rajasthan.

Because blackbuck are relatively scarce in this area of Rajasthan and are further threatened by human and livestock population growth (Mallon 2008, Lal 1991) management of blackbuck population requires the identification of factors that both positively and negatively influence species abundance. Despite the efforts of the local government local people still hunt blackbuck for food and sport (JCH personal observation). Due to the nature of the landscape setting aside protected areas within this rural environment would be difficult to implement and justify. Therefore, identifying areas where blackbuck can exist ultimately becomes a question of where can blackbuck coexist with human populations. Although this study does not identify a causal relationship between presence and abundance of blackbuck and Bishnoi people there is certainly a positive association between the two populations. There are several examples of the benefits of community-based conservation (Agrawal and Gibson 1999, Lepp and Holland

2006, van Eden et al. 2006, Sommerville et al. 2010, López-Arévalo et al. 2011, Waylen et al.

2010) and further research should further explore the role Bishnoi practices may play in blackbuck species abundance and conservation.

CONCLUSIONS

I have conducted the first standardized survey of populations of blackbuck antelope living in two distinct human environments; one including a group of people following a conservation-focused religion (the Bishnoi) and the other area devoid of such people. I have found a measureable difference in blackbuck populations between these environments; the areas that do have Bishnoi populations have significantly more blackbuck antelope despite overall blackbuck occupancy estimate of less than 40%.

Examining how human beliefs influence practices is an important and traditionally neglected piece of the puzzle in solving issues of species conservation. While the scope of this research needs to be expanded to elucidate whether the relationship between blackbuck abundance and the presence and practices of the Bishnoi is causal these results suggest that the conservation-focus practices of the Bishnoi people may have a measureable impact on wildlife living in village areas. This novel approach to conservation research allows for the incorporation of information that is often neglected; the intersection of human religion and beliefs and its influence on species abundance and environmental stewardship.

Chapter 5: Patterns of economic variation among villages in Western Rajasthan

ABSTRACT

Rural community economic dynamics are often closely tied to agriculture. Recent trends in research suggest, however, that rural community income is the result of a diverse network of influences that include, but are not limited to agriculture. Native species and the differential practices of local people can also influence economic variation among human populations. This study investigates the influence of land area use, human and livestock population, the ecologically important Khejeri tree (Prosopis cineraria), the practice of land fallowing, and the presence of the conservation conscious religion of the Bishnoi people on household income in nineteen villages in western Rajasthan. Here I test two hypotheses: 1)

Households in areas with Bishnoi populations have higher income and 2) Agricultural land area is primarily associated with income. Results show households in villages with Bishnoi have significantly greater annual household income when compared to villages without

Bishnoi. Results also indicate that total livestock population is primarily associated with household income. Agricultural land area and total human population are also significantly associated with income. These results provide the foundation for testing the hypothesis that wealth disparities among village households is due to Bishnoi religious practices.

Finally, this study highlights the importance of livestock in framing household economics and suggests these dynamics are not adequately explained by agricultural land area alone.

INTRODUCTION

Western Rajasthan located in the northwest portion of India, borders the Thar Desert and is a representative of an environment with limited resources (Fisher 1997, Khanna et al.

2008, Singh et al 2008). Frequent droughts and subsequent famine have characterized the ecological and economic dynamics of this region for centuries (Fisher 1997, Singh et al

2008, Francis and Gadgil 2009, Rajput and Tripathi 2009, Chand et al. 2010). Despite these challenging conditions western Rajasthan has sustained human life and remains one of the most heavily populated desert regions in the world (Fisher 1997). The majority of

Rajasthan’s human population lives in rural communities and practice some form of agropastoralism (Fisher 1997, Brockmann and Pichler 2004, Rajput and Tripathi 2009,

Chand et al. 2010 . Rajasthan is India’s largest state by area and also one of India’s most impoverished states (Khanna et al. 2008). Over the last three and a half decades local and national authorities have legislated multiple initiatives to increase the capitalization of agriculture (Robbins 1998a, Sidhamend 1999 in an effort to boost Rajasthan’s economy.

This intensification in agriculture has led to the appropriation of Rajasthan’s land for the primary purpose of growing crops. In some districts as much as 84% of the land surface is used for agriculture (Ram and Chauhan 2009). The cultural relevance of pastoralism and its role in generating capital, however, has largely been underrepresented and potentially underappreciated in the capitalization initiatives set forth by local authorities and academics (Robbins 1998a, 2004, Jones 2008). In general, rural economies are complex and require a thorough understanding of income generating networks in order to maximize capital (Jones 2008). Many rural landscapes in western Rajasthan serve a dual purpose, sustaining agriculture while also supporting livestock populations (Fisher 1997, Robbins

1998b), which forms the backbone of rural economies (Fisher 1997). Community grazing pastures, sites for growing and cultivating fodder trees, and water catchment areas are some of the alternate uses of land area besides cultivation land that exist in this semi-arid rural landscape (Fisher 1997, Robbins 1998b, Brockman and Pichler 2004, Robbins 2004).

Collectively these land areas directly or indirectly support rural economies (Lal 1991,

Fisher 1997, Robbins 1998b, Sidahmend 1999, Brockmann and Pichler 2004, Robbins

2004).

Currently inhabitants of western Rajasthan face acute challenges to sustain livelihoods and subsistence. Increased frequency of droughts caused by climatic fluctuations such as La Niña events, rapidly increasing human and livestock populations, loss of keystone species, and increased incidence of disease represent some of the challenges to be met by current and future generations (Lal 1991, Fisher 1997, Robbins

1998a, Sidahmend 1999, Prakash 2003, Khan et al. 2004, Robbins 2004, Naewboonnien

2007, Singh et al. 2008, Hall et al. in press). Poverty alleviation is seen as one of the key factors in controlling and reversing the adverse trends faced by many rural inhabitants across the world (Sidahmend 1999, Khan et al. 2004, Singh et al. 2008). In order to mitigate these issues however, a clear understanding of how natural resources and the use of area landscapes impacts individuals is needed. Such an understanding can potentially add to and enhance the current strategies in place for raising capital.

An often-overlooked source of ideas and strategies for understanding and improving resource management and production in rural communities is the traditional strategies used by inhabitants of the region (Kohler-Rollefson and Rathore 2004, Gilchrist and Mallory 2007, Steele and Shackleton 2010, López-Arévalo et al. 2011). Communities

46 with generations of experience and traditional knowledge of their environment are in prime position to plan, implement, and sustain resource management and economic improvement (Gilchrist and Mallory 2007, Steele and Shackleton 2010, López-Arévalo et al.

2011). In western Rajasthan a particular caste of local people practice a religion where environmental stewardship may impact resource availability. The Bishnoi people hold the tree species Prosopis cineraria, locally known as the Khejeri tree, sacred due to the fact that the founder of their faith composed the twenty-nine principles of faith while sitting underneath one of these Khejeri (Brockmann and Pichler 2004). Because the Bishnoi consider the Khejeri tree sacred they do not harm the tree by lopping off branches or cutting them down and are famous throughout the region for protecting Khejeri with their lives (Fisher 1997, Brockmann and Pichler 2004). In 1730 a Bishnoi woman by the name of

Amrita Devi led a protest against the felling of trees in her village that resulted in the sacrificial suicides of 363 Bishnoi (Brockmann and Pichler 2004). This extreme devotion to environmental protection combined with the benefits to the agropastoral lifestyle of the majority of the inhabitants of this region warrant investigation and consideration for plans to improve the economic state of rural Rajasthan. Along with being sacred to the Bishnoi

Khejeri trees are important for other groups of people and wildlife species living in the area

(Puri et al. 1994, Kaushik and Kumar 2003, Singh et al. 2007, Yadav et al. 2008).

Specifically, Khejeri trees support both agriculture, through nutrient recycling and topsoil retention and pastoralism by providing fodder for livestock (Lal 1991, Fisher 1997,

Brockmann and Pichler 2004, Hall et al in press).

This study examines the influence of different village land area types, the number of

Khejeri trees, and the presence Bishnoi people on household income in nineteen villages in

47 western Rajasthan. The goal of this study is to determine if household income can be solely explained by land area usage or if describing household economics includes alternative metrics.

METHODS

Study area

The research area includes nineteen villages situated within 50 km of Jodhpur

(26o18N, 73o08E) (Figure 1). The region is covered with open scrub forest. Dominant tree and shrub species include: Acacia senegal, A. nilotica, Euphorbia caducifoliya, Anogeisus pendula, Mytenus emarginata, Greioia tenex, Ziziphus numularia, Prosopis cineria, and the exotic invasive P. juliflora. The climate is hot and dry with temperatures ranging from 50oC in May and June to a low as 1oC during winter. Annual rainfall averages 360 mm, with 90% occurring during the monsoon season (July – September).

The village areas in this study are all organized in the same fashion. Housing structures are clustered at the center of village areas with agricultural and community land adjacent radiating outwards (Figure 2). The near majority of the land in the study area is under human use or has been transformed by humans with very little, if any, wilderness land areas. Village boundaries are established by land ownership (land owned by members of different villages) and/or by roads.

Villagers

The people of this area practice some form of agropastoralism (Fisher 1997, Robbins

1998). The majority of households in this study own or rent farmland and cultivate some portion of said land. Additionally, the majority of households keep livestock for various

48 purposes (Lal 1991, Fisher 1997). Cows and buffalos are kept for their milk, an essential ingredient in Rajasthan food and drink (Fisher 1997). Camel and buffalo are primarily work animals that are used in farming and transportation of produce and construction materials (1997). Sheep and goats are kept for their wool, milk, and consumed by certain castes, particularly Muslim populations (1997). Wealthier families primarily keep cows, buffalos, and camels while households with less capital tend to primarily keep sheep and goats.

Every village has populations from cultural groups, or caste, living within the borders. There is a clear hierarchy among castes that represents a traditional division of roles and responsibilities prescribed to each group. In every village there is upper class that oversees land ownership (Rajputs) and religion (Brahmin), a middle class primarily composed of merchants and manufacturers, and a lower class of laborers. The Bishnoi people belong to the middle class and are somewhat unique in that their religious practices are not under the perview of the Brahmin. The Bishnoi faith is a sect of Hinduism as it shares many common principles with Hinduism such as vegetarianism (Brockmann and

Pichler 2004). The founder of the Bishnoi faith, Jambeshwar, was in fact Hindu before devising the twenty-nine principles of Bishnoi religion (Bishnoi translates to “twenty- nine” (Brockmann and Pichler 2004 .

Khejeri Trees

The Khejeri tree is a native tree species of northwest India and the Thar Desert and is characterized by thin leaves, thorny branches, and yellow flowers. Khejeri are well adapted to arid environments and are the dominant and keystone tree species within the study area (Fisher 1997, Brockmann and Pichler 2004, Sharma et al 2008, Hall et al. in

49 press). Khejeri trees provide food and shelter for wildlife especially during the dry season’s extreme temperatures and frequent drought (Goyal et al. 1988; Fisher 1997). Villagers use the Khejeri tree for fuel wood, livestock fodder (primarily for sheep and goats), building material, and in several local delicacies (Goyal et al. 1988, Fisher 1997, Pasiecznik et al.

2001, Singh et al. 2007, Vaithiyanathan 2007). Several studies have also demonstrated the benefits of the presence of Prosopis cineraria on agricultural lands (Puri et al. 1994,

Kaushik and Kumar 2003, Singh et al. 2007, Yadav et al. 2008, Singh 2009). These benefits include increased crop yield, retention of topsoil, increase soil moisture, and increased soil nutrient decomposition (Singh et al. 2007, Yadav et al. 2008, Singh 2009).

Demographic Data Collection

During three field seasons in 2007, 2009, and 2010 I conducted semi-structured interviews

(Bernard 2001) in each of the nineteen villages. A focus group of 3-18 adult male (18 years or older) village residents voluntarily gathered to be interviewed where I collected data on village demographics. Cultural norms prevented me from having discussions with female residents. The topics of questions included number of livestock in the village, average household income of each caste, attitudes toward Khejeri trees, and fallowing practices

(Appendix 1). In each of the focus groups one focus group member served as the spokesperson. Other members provided input and comments, but in each case there was a clear final authority on the village information that was reported. Interviews lasted from half an hour to one and a half hours. I visited each village in the study twice for interviews, once in 2007 and again in 2009 and 2010.

In addition to focus group interviews I collected and otherwise verified interview data from data collected by the non-governmental organization (NGO) the School of Desert

Sciences (SDS) during a survey they conducted of seven of the nineteen villages in this study in 2007. I also collected and/or verified interview data from the 2001 Rajasthan census available online (Sharma 2011). Although I collected a variety of demographic, ecological, and ethnographic data I was primarily interested in five village demographic characteristics for this study: total human population, agricultural land area, community land area, total livestock population, and the presence or absence of fallow land practices.

Agricultural land was defined as land under repeated and sustained cultivation over the course of the study. Community lands are composed of several different land types; goucher (state controlled pasture land), community fallow land (locally controlled pasture land), water catchment area, and oran (locally controlled forested area).

In addition to the land area data I collected during interviews I report and discuss the reported importance of Khejeri trees and where they rank among other tree species in the area.

Khejeri Tree Transects

Each village center was marked using GPS. Village centers were chosen based on their central location within the village residential area and the self-reported social and economic center of villager activity. Village centers were often located in front of or near the Tako’s (the head family of the village) residence. I used Garmin mapping program

(MapSourcetm) to establish eight transect starting positions 500m from each village center.

The eight starting point compass headings were designated north (0o), northeast (45o), east

51 and returned to the village to walk each of the eight transects. I marked the location of each

Khejeri tree within 10m of the transect line along each transect line in every village.

Data Analysis

I first calculated average annual household income (AAI) for each village by summing the individual household incomes and dividing by the total number of households. In order to investigate whether or not there are differences in AAI between villages with Bishnoi populations and villages without Bishnoi I used a Mann-Whitney U-Test. I used Fisher’s exact test to evaluate whether or not there is a difference in the incidence of fallowing between villages with Bishnoi and villages without Bishnoi.

I used non-metric multidimensional scaling (NMS) (McCune and Grace 2002) ordination, excluding AAI, to investigate the pattern of variation in villages using the other five continuously variable village characteristics (total human and livestock population, numbers of Khejeri trees from transects, and total agricultural land area). NMS is a non- parametric ordination technique for analyzing ecological community data (Clarke 1993,

McCune and Grace 2002, Daley 2008, Bruelheide et al. 2011, Shipley et al. 2011). Similar to principle components analysis NMS describes the pattern among sites by fitting the data to a certain number of axes that are variably associated with site characteristics (McCune and

Grace 2002). I excluded AAI from the NMS analysis in order to perform the subsequent covariate analysis. I used the program PC-ORD (McCune and Mefford 2009) for the NMS ordination.

I performed an analysis of covariance (ANCOVA) to investigate the influence of each

NMS axis on village AAI. In addition to using the NMS axis scores I included the fixed factors of presence of fallowing practices and the presence of Bishnoi. I evaluated each individual factor’s influence on I as well as the various interactions between factors. Interactions between factors that were highly non-significant were sequentially removed and the analysis was rerun. I performed ANCOVA analysis using the program SPSS (v.19.0).

Outlier Analysis

Guda Bishnoi is a home to the largest population of Bishnoi (8024) among villages in the study. This population of Bishnoi is greater than any other total village population in the study ( ppendix 1 . dditionally Guda Bishnoi’s I is approximately 30,000Rs

(~$675.00 more than the next highest village’s I ( ppendix 1 . I reran both the NMS

53 ordination and the ANCOVA to see what effect Guda Bishnoi might have on the pattern and what said pattern’s influence would be on income.

RESULTS

The results from the Mann-Whiney U-Test indicate a significant difference in AAI between villages with Bishnoi and villages without Bishnoi (p=0.032 . The Fisher’s exact test shows that there is a significant difference in incidence of fallowing between villages with Bishnoi and villages without Bishnoi (p=0.0371). The NMS analysis including all villages resulted in a two dimensional, or two axis, solution with a cumulative r2 value of 0.962 and a final stress value of 3.217. The proportion of data described by axis one is 0.629 while the remaining proportion, 0.334, is describes axis two. Table 1 outlines the association of each village characteristic with each axis. All village characteristics with the exception of Khejeri trees have a notable contribution to both axes.

Using the NMS axis scores (Appendix 2) and incorporating the fixed factors of both presence of Bishnoi and presence of fallowing practices ANCOVA results indicate a model with axis 1 as the only significant model parameter (r2=0.426) (Table 2). This result suggests that AAI is influenced by axis 1, which is strongly correlated with agricultural land area, total population, community land area, and total livestock population.

Figure 3 suggests that Guda Bishnoi may be an outlier. The results from the NMS ordination without Guda Bishnoi resulted in a 2-dimensional solution with a cumulative r2 value of 0.963 and a final stress value of 3.7520. The proportion of data described by axis one is 0.579 and the proportion of data described by axis two is 0.384. Table 3 outlines the association of each village characteristic with each axis. In this analysis total livestock

54 population is primarily associated with axis 1. Total livestock, total population, and agricultural area are associated with axis 2. Overall the pattern among villages is more clearly described when Guda Bishnoi village is excluded from the analysis, as total livestock is most closely associated with the variation among villages. The subsequent ANCOVA resulted in several models that described the influence of axis scores and fixed factors on

AAI. I used Akaike Information Criterion (AIC) to select the model that described the data

(Table 4). The model with the lowest AIC value was chosen (Table 5) and indicates that both axis 1 and axis 2 significantly influence AAI.

Despite the lack of influence Khejeri trees have on AAI 15 out of the 19 total villages reported Khejeri trees as the most important tree for sustaining life.

DISCUSSION

Mann-Whitney U-Test of differences in average annual household income (AAI) indicates a significant difference between villages with Bishnoi and villages without Bishnoi. A Fisher’s exact test also indicates a significant difference between village classes and the incidence of fallowing. Initial ordination and covariate analysis indicate that AAI is influenced by the interaction between population, community and agricultural land area, and total livestock.

A second round of ordination and covariate analyses excluding probable outlier village

Guda Bishnoi indicates AAI is primarily influenced by total livestock population and additionally by total livestock, agricultural land area, and total population. These combined results suggest income is primarily associated with livestock with influences from total population and agricultural land area, but not community land area, presence of Bishnoi or

55 fallowing. These results are supported by literature that suggests western Rajasthan economic dynamics are multifaceted and not simply the result of agricultural production.

Guda Bishnoi appears to be driving the importance of community land and the relative importance of agricultural land and total human population in describing variation among villages and their influence on AAI. Without Guda Bishnoi village in the analysis the pattern among villages and the influence of said pattern on AAI is simplified and is primarily associated with livestock populations. Therefore I consider the analysis that excludes Guda Bishnoi to more accurate because of the singular influence of this one particularly large and wealthy village. Because Guda Bishnoi has by far the largest human population and community land area (Appendix 1) it follows that inclusion of this village would over represent the importance of these factors in the analysis. Axis 2 for this analysis (Table 2) is represented by three village characteristics and describes nearly 40% of the variation between villages. The fact that axis 2 significantly influences AAI highlights the importance of total population and agricultural area for income. The most important result, however, is the fact that nearly 60% of the variation can be primarily described by axis 1 and that axis is most closely associated with total livestock. Model 2 (Table 6) is within 2  AIC values of model 1 and only considers axis 1 as a significant influence on AAI.

Both models highlight the primary importance of livestock’s influence on I. These results should concern policy makers that have created and enacted policy to increase agricultural productivity by appropriating more land (Robbins 1998, 2004, Ram 2009) as these initiatives alone clearly neglect a significant segment of the local economy. Moreover, the appropriation of land for agriculture leaves less grazing land for livestock, which may initially reduce household income. I propose that current strategies to increase capital

56 through agricultural land (Robbins 1998, 2004, Ram 2009) may be misguided and prove ultimately ineffective if efforts to mitigate income lost through reduced livestock productivity are not explored.

The lack of association between Khejeri tree numbers and either axis indicates that these trees are not directly important for describing variation among villages. This result runs counter to the literature implicating Khejeri trees as having a positive influence on agricultural productivity (Puri et al. 1994, Kaushik and Kumar 2003, Singh et al. 2007,

Yadav et al. 2008, Singh 2009) and presumably household income. The fact that the majority of villages (79%) indicated Khejeri trees are the most important species for sustaining life suggests the relationship between Khejeri and income may be indirect and not necessarily captured by transect numbers. The results from this study also provide important insight into the results from chapter 3, which determined that AAI influences

Khejeri tree abundance in Bishnoi villages. Given the combined results from both chapters it appears that the income generated through livestock, agriculture, and total population is only being used for/on Khejeri trees in villages with Bishnoi. There are additional reasons for exploring Khejeri’s potential influence on AAI that include: 1) This study includes

Khejeri tree numbers only within 1km of the village center and not the entire village area

2) The abundance of Khejeri trees recorded does not account for differences in individual size which may affect nutrient cycling efficiency, fodder and fruit yields, and amount of soil retained. Villages with the same number of Khejeri trees may differ significantly in the benefits they afford villagers.

One factor that was not directly examined in this study is water availability. Water is an important resource for villagers living in this area (Fisher 1997), but is somewhat

57 difficult to measure given the variability of rainfall and differential access to city water sources from village to village. Certain villages have access to water pipelines from Jodhpur

City while others do not while interruptions in service are frequent and highly variable.

Moreover the influence of water availability and it’s relationship with I is a two sided question: do villages have more water because they have more money or do villages have more money because they have more water? Nevertheless, determining the relationship between water availability and income will most likely contribute to the overall picture of characterizing household income.

CONCLUSIONS

Variation between villages is primarily associated with livestock populations with notable associations with agricultural land area and total population. Livestock population, agricultural land area, and human population significantly influence average annual household income (AAI) with livestock population as the most influential of the three factors. These results suggest that policy makers may overestimate the relative importance of agriculture on household income while the influence of pastoralism may be underestimated. The significant difference in number of Khejeri trees and incidence of fallowing between the two village classes, though not a direct influence on AAI, may indirectly influence income by supporting livestock populations. Finally, I contend that current initiatives to further capitalize agriculture in rural Rajasthan through land appropriation does not fully capture the complexity of household economics and may in fact undermine its intended objectives.

Chapter 6: General Conclusions

All three study species, Indian vultures, blackbuck antelope, and Khejeri trees are positively associated with the presence of Bishnoi people. While causality of this relationship has yet to be investigated this study has provided the foundational evidence for answering the question of why species are associated with Bishnoi.

In Chapter 2, I found that Indian vulture dynamics are largely governed by the synergy between La Niña induced drought and the use of the veterinary drug diclofenac.

Khejarli Kalla village, the same village where the Bishnoi faith was founded, may provide a safe haven for vultures suffering from drought. The fact that Bishnoi protect nesting trees by not lopping branches may help sustain breeding populations. In Chapter 3, I found that

Khejeri trees are found in greater numbers in Bishnoi villages and their populations are associated with household income in villages with Bishnoi, but not in villages without

Bishnoi. In Chapter 4, I found that Blackbuck antelope occur in greater numbers in areas with Bishnoi despite being a relatively rare species. Bishnoi openly oppose poaching and care for orphan blackbuck on a routine basis and may provide safe habitat for these animals. In Chapter 5, I found that variation in village economies is associated with variation in livestock populations, total population, and agricultural farmland area.

Athough caution must be taken when interpreting these patterns, my results suggest that Indian vultures and blackbuck antelope, both rare species, may benefit from the presence of Bishnoi people in village areas. The religious beliefs of the Bishnoi people

59 may impact both species directly and/or indirectly through habitat preservation (Khejeri trees) and protection from poachers. As such, conservation efforts to preserve and restore populations of Indian vultures and blackbuck may be informed by the results of this study and the potential importance of Bishnoi presence. The effects of Bishnoi on their landscape may also impact local production and wealth. Khejeri trees are important for agriculture and for maintaining livestock populations. Although Khejeri do not appear to have a direct impact on income they do support two of the three factors that do influence income variation.

Each of the individual studies suggests a relationship between a human community, the Bishnoi, and populations of other species. Looked at together these individual studies suggest the following scenario: Bishnoi protection of Khejeri trees helps support vulture nesting habitat, which presumable supports a health ecosystem where livestock are disposed of. Khejeri tree preservation may also support increased agricultural productivity and provide increased fodder for livestock. Although there are likely alternative sources of income for villagers this research shows that a portion of said income in villages with

Bishnoi is associated with Khejeri trees, whether as a cause or consequence of Khejeri preservation. One possible explanation is that this money is used to buy fodder from outside the village area thus preserving local tree populations. Higher numbers of antelope in Bishnoi areas may be related to higher numbers of Khejeri trees and the protection

Bishnoi provide for the animals. Bishnoi attitudes towards hunting are a likely deterrent to local hunters and may prevent local extinction of blackbuck.

Maintaining a healthy ecosystem in the face of human development is often juxtaposed in the minds of stakeholders, managers, and scientists. This work challenges

60 that notion and begins to provide evidence that a strong commitment to conservation, in this case, a commitment enforced through a belief system, can potentially provide benefits to non-human species while simultaneously maintaining the lifestyle of humans that share the same space. Few, if any, habitats are devoid of human influence; thus, understanding systems where humans have apparently successfully shared their living space and resources with other species are important to investigate.

Literature Cited

Abson DJ and Termansen T (2010) Valuing ecosystem services in terms of ecological risks and returns. Conservation Biology DOI: 10.1111/j.1523-1739.2010.01623.x Agrawal A and Gibson C (1999) Enchantment and disenchantment: The role of community in natural resource conservation. World Development 27: 629-649. Anand MO, Krishnaswamy J, Kumar A, and Archana B (2010) Sustaining biodiversity conservation in human-modified landscapes in the Western Ghats: remnant forests matter. Biological Conservation 143: 2363-2374. Becerra-Muñoz S, Buelna-Osben HR, and Catalán-Romero JM (2003) Spatial patterns of ARIMA modeled rates of change of atherinids (Chirostoma spp.) and goodeid Chapalichthys encaustus from Lake Chapala, México. Ecological Modelling 165: 237-250. Bernard HR (2001) Research Methods in Anthropology: Qualitative and Quantitative Approaches. AltaMira: Oxford, UK. BirdLife International (2009) Gyps bengalensis. In: IUCN 2010. IUCN Red List of Threatened Species. Version 2010.2. . Downloaded on 25 July 2010. Bozdogan H (1987 Model selection and kaike’s information criterion ( IC : the general theory and its analytical extensions. Psychometrika 52, 345–370. Brockmann H and Pichler R (2004) Paving the way for peace: The living philosophies of Bishnois and Jains. D.K. Fine Art Press, Delhi, India. Brooks J, Franzen MA, Molmes CM, Grote MN and Borgerhoff Mulder M (2006) Testing hypotheses for the success of different conservation strategies. Conservation Biology DOI: 10.1111/j.1523-1739.2006.00506.x Bruelheide H et al (2011) Community assembly during secondary forest succession in a Chinese subtropical forest. Ecological Mongraphs 81: 25-41. Carroll SS and Pearson DL (1999) Detecting and modeling spatial and temporal variance in conservation biology. Conservation Biology 14: 1893-1897. Challinor AJ, Weeler TR, Craufurd PQ, Ferro CAT, and Stephenson DB (2007) Adaptation of crops to climate change through genotypic responses to mean and extreme temperatures. Agriculture, Ecosystems, and Environment 119: 190-204. Chand K, Jangid BL, Rohilla PP, Kachhawaha S (2010) Economics of camel production in Rajasthan. Journal of Camel Practice and Research 17: 15-20. Chaudhary A and Aggrawal PK (2009) Impact of climate change on food security of India. Proceedings of World Academy of Science: Engineering & Technology 39: 22-24. Chhangani AK (2004) Status of a breeding population of long-billed vulture (Gyps indicus) in and around Jodhpur (Rajasthan), India. Vulture News 50: 15-22. Chhangani AK (2005) Population ecology of vultures in western Rajasthan. Indian Forester 131: 1373-1383.

Chhangani AK and Sivaperuman C (2009) Present Status of vultures in the Great Indian Thar Desert. Faunal Ecology and Conservation of the Great Indian Desert, pp. 65-82. Chhangani AK and Mohnot SM (2004) Is Diclofenac the only cause of vulture decline. Current Science 87: 1496-1497. Clarke KR (1993) Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18: 117-143. Cleary DFR, Fauvelot C, Genner MJ, Menken SBJ, and Mooers AO (2006) Parallel responses of species and genetic diversity to El Niño Southern Oscillation- induced environmental destruction. Ecology Letters 9: 301–307. Daley AC (2008) Statistical analysis of mixed-motive shell borings in Ordovician, Silurian, and Devonian brachiopods from northern and eastern Canada. Candian Journal of Earth Science 45: 213-229. Donovan, T. M. and J. Hines. 2007. Exercises in occupancy modeling and estimation. http://www.uvm.edu/envnr/vtcfwru/spreadsheets/occupancy/occupancy.htm. Dzialak MR, Harju SM, Osborn RG, Wondzell JJ, Hayden-Wing LD, Winstead JB, and Webb SL (2011). Prioritizing conservation of ungulate calving resources in multiple-use landscapes. PLoS One 6: 14597.doi:10.1371/journal.pone.0014597. Fergusons-Lees J, Christie DA, Franklin K, Philip B, and Mead D (2001) Raptors of the World. Houghton Mifflin: London. Fisher RJ (1997) If the rain doesn’t come: An anthropological study of drought and human ecology in western Rajasthan. Monohar: New Delhi, India. Francis PA and Gadgil S (2009) The aberrant behaviour of the Indian monsoon in June 2009. Current Science (00113891) 97: 1291-1295. Gilbert M, Virani MZ, Watson RT, Oaks JL, Benson PC, Khan AA, Ahmed S, Chaudhry J, Arshad M, Mahmood S, and Shah QA (2002) Breeding and mortality of oriental white-backed vulture Gyps bengalensis in Punjab Province, Pakistan. Bird Conservation International 12: 311–326. Gilchrist G and Mallory ML (2007) Comparing expert-based science with local ecological knowledge: what are we afraid of? Ecology and Society 12: r1. [online] URL: http://www.ecologyandsociety.org/vol12/ iss1/resp1/ Google Earth (version 5.2.1.1329 beta) (2010) Village area map, Rajasthan, India. www.googleearth.com. Mountain View, California. Goyal SP, Bohra HC, Ghosh PK, and Prakash I (1988) Role of Prosopis cineraria pods in the diet of two Indian desert antelope. Journal of Arid Environments 14: 285- 290. Green RE, Newton I, Shultz S, Cunningham AA, Gilbert M, Pain DJ, and Prakash V (2004) Diclofenac poisoning as a cause of vulture population declines across the Indian subcontinent. Journal of Applied Ecology 41: 793–800. Green RE, Taggart MA, Das D, Pain DJ, Kumar CS, Cunningham AA, and Cuthbert RA (2006) Collapse of Asian vulture populations: risk of mortality from residues of the veterinary drug diclofenac in carcasses of treated cattle. Journal of Applied Ecology 43: 949-956. Hall JC, Chhangani AK, Waite TA, and Hamilton IM (in press) The impacts of La Niña induced drought on Indian vulture Gyps indicus populations in Western Rajasthan. Bird Conservation International. 63

Harrison RD (2000) Repercussions of El Niño: drought causes extinction and the breakdown of mutualism in Borneo. Proceedings of the Royal Society B 267: 911– 915. Hernandez M and Margalida A (2008) Pesticide abuse in Europe: effects on the Cinereous vulture (Aegypius monachus) population in Spain. Ecotoxicology 17: 264-272. Hernandez M and Margalida A (2009) Poison-related mortality effects in the endangered Egyptian vulture (Neophron percnopterus) population in Spain. European Journal of Wildlife Research 55: 415-423. Hines, J.E. (2006). Program PRESENCE (Version 3.1). http://www.mbr- pwrc.usgs.gov/software/doc/presence/presence.html. Holmgren, M. & Scheffer, M. (2001) El Nino as a window of opportunity for the restoration of degraded arid ecosystems. Ecosystems 4: 151–159. IUCN (2004) IUCN Red list of Threatened Species. http://www.iucnredlist.org, accessed January 30, 2008. IUCN, Gland Switzerland. IUCN (2007) IUCN Red list of Threatened Species. http://www.iucnredlist.org, accessed March 29, 2008. IUCN, Gland Switzerland. Ismail K, Kamal K, Plath M, Wronski T (2011). Effects of an exceptional drought on daily activity patterns, reproductive behavior and reproductive success of reintroduced Arabian oryx (Oryx leucoryx). Journal of Arid Environments 75, 125- 131. Isvaran K (2003). The evolution of lekking: insights from a species with a flexible mating system. Ph.D. Dissertation. University of Florida, Gainesville, FL. Jackson R and Wangchuk R (2004) A community-based approach to mitigating livestock depredation by snow leopards. Human Dimensions of Wildlife 9: 307- 315. Jhala Y V, et al. (1992) Water in the ecophysiology of blackbuck. Journal of the Arid Environments, 22, 261-269. Jhala Y V (1997) Seasonal effects on the nutritional ecology of blackbuck Antelope cervicapra. Journal of Applied Ecology, 34, 1348-1358. Jones HM (2008) Livelihood diversification and moneylending in a Rajasthan village: what lessons for rural financial services? The European Journal of Development Research 20: 507-518. Karanth KK, Nichols JD, Hines JE, Karanth KU, and Christensen NL (2009). Patterns and determinants of mammal species occurrence in India. Journal of Applied Ecology doi: 10.1111/j.1365-2664.2009.01710.x Kaushik N and Kumar V (2003) Khejri (Prosopis cineraria)-based agroforestry system of arid Haryana, India. Journal of Arid Environments 55: 433-440. Khanna RK, Rathore RS, and Sharma C (2008) Solar still an appropriate technology for potable water need of remote villages of desert state of India-Rajasthan. Desalination 220: 645-653. Kohler-Rollefson I and Rathore HS (2004) Indigenous versus official knowledge, concepts and institutions: Raika pastoralists and the outside world. Nomadic Peoples 8: 150-167. Krishna YC, Clyne PJ, Krishaswamy J, and Kimar NS (2009). Distribution and ecological review of the four horned antelope, Tetracerus quadricornis. 64

Mammalia 73, 1-6. Lal S (1991) Development of underdeveloped regions: A case of arid regions. Mittal Publications, New Delhi, India. Lepp A and Holland S (2006) A comparison of attitudes toward state-led conservation and community-based conservation in the village of Bigodi, Uganda. Society and Natural Resources 19: 609-623. Lesica P, McCune B, Cooper S, and Hong WS (1991). Differences in lichen and bryophyte communities between old-growth and managed second-growth forests. Canadian Journal of Botany 69, 1745-1755. López-Arévalo HF et al. (2011 Local knowledge and species distribution models’ contribution towards mammalian conservation. Biological Conservation doi:10.1016/j.biocon.2011.01.014 MacKenzie, D.I., Nichols, J.D., Lachman, G.B., Droege, S., Royle, J.A. & Langtimm, C.A. (2002) Estimating site occupancy rates when detection probabilities are less than one. Ecology 83, 2248–2255. MacKenzie, D.I., Nichols, J.D., Hines, J.E., Knutson, M.G. & Franklin, A.D. (2003) Estimating site occupancy, colonization and local extinction when a species is detected imperfectly. Ecology 84 2200–2207. MacKenzie DI, Nichols JD, Royle JA, Pollock KH, Bailey LA and Hines, JE (2005). Occupancy modeling and estimation: inferring patterns and dynamics of species occurance. Academic Press, San Diego, CA. Madhusudan MD and Mishra C (2003). Why big, fierce animals are threatened: conserving large mammals in densely populated landscapes. Battles Over Nature: Science and Politics of Conservation (eds V. Saberwal & M. Rangarajan), pp. 31–55. Permanent Black, New Delhi. Mallon DP and Jiang Z (2009). Grazers on the plains: challenges and prospects for large herbivores in Central Asia. Journal of Applied Ecology 46, 516-519. Mallon DP (2008). Antilope cervicapra. In: IUCN 2010. IUCN Red List of Threatened Species. Version 2010.4. . Margalida A, Heredia R, Razin M, and Hernandez M (2008) Sources of variation in mortality of the Bearded Vulture Gypaetus barbatus in Europe. Bird Conservation International 18: 1-10. Mateo-Tomás P and Olea PP (2010) When hunting benefits raptors: a case study of game species and vultures. European Journal of Wildlife Research 56: 519-528. McCune B and Grace JB (2002) Analysis of Ecological Communities. MjM Software Designs: Oregon, USA. McCune B and Mefford MJ (2009) PC-ORD. Multivariate Analysis of Ecological Data. Version 5. MjM Software: Gleneden Beach, Oregon, USA. Monadjem A and Garcelon DK (2008) Nesting distribution of vultures in relation to land use in Swaziland. Biodiversity and Conservation 14: 2079–2093. Moon K and Cocklin C (2010) Landholder-based approach to the design of private- land conservation programs. Conservation Biology DOI: 10.1111/j.1523- 1739.2010.01639.x Mungall E C (1978) The Indian Blackbuck Antelope: A Texas View. Cesar Kleberg Studies in Natural Resources, No. 3 Naewboonnien J (2007) TigerPaper. Regional Quarterly Bulletin on Wildlife and 65

National Parks Management 21: 1-32. Oaks JL, Gilbert M, Virani MZ, Watson RT, Meteyer CU, Rideout BA, Shivaprasad HL, Ahmed S, Chaudhry MJI, Arshad M, Mahmood S, Ali A, and Khan AA (2004) Diclofenac residues as the cause of vulture population decline in Pakistan. Nature 427: 630-633. Pimm SL, Russell GJ, Gittleman JL, and Brooks TM (1995) The future of biodiversity. Science, New Series 269: 347-350. Prakash V (1999) Status of vultures in Keoladeo National Park, Bharatpur, Rajasthan, with special reference to population crash in Gyps species. Journal of the Bombay Natural History Society 96: 365–378. Prakash V, Pain DJ, Cunningham AA, Donald PF, Prakash N, Verma A, Gargi R, Sivakumar S, and Rahmani AR (2003) Catastrophic collapse of white-backed Gyps bengalensis and long-billed Gyps indicus vulture populations. Biological Conservation 109: 381-390. Puri S, Kumar A, and Singh S (1994) Productivity of Cicer arietinum (chickpea) under a Prosopis cineraria agroforestry system in the arid regions of India. Journal of Arid Environments 27: 85-98. Rahmani A and Sankaran R (1991) Blackbuck and chinkara in the Thar Desert: a changing scenario. Journal of the Arid Environments 20: 379-391. Rajasthan Map (2008). Available at: http://www.tour2rajasthan.com/rajasthan- map.html. Accessed March 5, 2009. Rajput DS and Tripathi H (2009) Socio-economic profile of Raika pastoralists of Rajasthan. Journal of Camel Practice and Research 16: 195-200. Ram and Chauhan (2009) Application of remote sensing and GIS to asses land use changes in Jhunjhunun district in arid Rajasthan. Journal of Indian Society Remote Sensing 37: 671-680. Ranjitsinh M K (1989) The Indian Blackbuck. Natraj Publishers: Dehradun, India Rasmussen PC and Parry SJ (2001) The taxonomic status of the "longbilled" vulture Gyps indicus. Vulture News 44: 18-21. Roach CM, Hollis TI, McLaren BE, and Bavington D (2006) Ducks, bogs, and Guns. Ethics & the Environment 11: 43-70. Robbins P (1998a) Authority and environment: institutional landscapes in Rajasthan, India. Annals of the Association of American Geographers 88: 410-435. Robbins P (1998b) Nomadization in Rajasthan‚ India: migration‚ institutions‚ and economy. Human Ecology 26: 87-112. Robbins P (2004) Pastoralists inside and out: the contradictory conceptual geography of Rajasthan’s Raika. Nomadic Peoples 8: 136-149. Sanginga PC, Kamugisha RN, and Martin AM (2007) The dynamics of social capital and conflict management in multiple resource regimes: a case of the soughtwestern highlands of Uganda. Ecology and Society 12: 6[online] http://www.ecologyandsociety.org/vol12/iss1/art6/. Sharma, A (2011) VikasDarpan: Government of Rajasthan. http://www.gis.rajasthan.gov.in/. Sharma SK, Kumar S, Rawat D, Kumaria S, Kumar A, and Rao SR (2011) Genetic diversity and gene Flow estimation in Prosopis cineraria (L.) Druce: A key stone tree species of Indian Thar Desert. Biological Systematics and Ecology 39: 9-13. 66

Shipley B, Laughlin DC, Sonnier G, and Otfinowski R (2011) A strong test of a maximum entropy model of trait-based community assembly. Ecology 92: 507- 517. Shultz S, Baral HS, Charman S, Cunningham AA, Das D, Ghalsasi GR, Goudar MS, Green RE, Jones A, Nighot P, Pain DJ, and Prakash V (2004) Diclofenac poisoning is widespread in declining vulture populations across the Indian subcontinent. Proceedings of the Royal Society of London, B (Supplement) 271: S458–S460. Sidahmed (1999) Rangeland development for the rural poor in developing countries: The experience of IFAD. Proceedings Paper: 2nd International Conference on Land Degradation (ICLD). Enfield, NH: Science Publishers Inc. Singh G, Mutha S, and Bala N (2007) Effect of tree density on productivity of a Prosopis cineraria agroforestry system in North Western India. Journal of Arid Environments 70: 152-163. Singh G (2009) Comparitive productivity of Prosopis cineraria and Tecomella undulata based agroforestry systems in degraded lands of Indian Desert. Journal of Forestry Research 20: 144-150. Singh M, Lakshminarayana J, and Fotedar R (2008) Chronic energy deficiency and its association with dietary factors in adults of drought affected desert areas of Western Rajasthan, India. Asia Pacific Journal of Clinical Nutrition 17: 580-585. Singh NJ, Grachev IA, Bekenov AB, and Milner-Gulland EJ (2010). Saiga antelope calving site selection is increasingly driven by human disturbance. Biological Conservation 143, 1770-1779. Sommerville M, Milner-Gulland EJ, Rahajaharison M, and Jones JPG (2010) Impact of a community-based payment for environmental services intervention on forest use in Menabe, Madagascar. Conservation Biology 24: 1488-1498. SPSS (2007). Version 15.0, SPSS Inc., Chicago, USA. SPSS Statistics v.19.0 (SPSS: An IBM Company). Steele MZ and Shackleton CM (2010) Using local experts as benchmarks for household local ecological knowledge: scoring in South African savannas. Journal of Environmental Management 91: 1641-1646. Swan GE, Cuthbert R, Quevedo M, Green RE, Pain DJ, Bartels P, Cunningham AA, Duncan N, Meharg AA, Oaks JL, Parry-Jones J, Shultz S, Taggart MA, Verdoorn G, and Wolter K (2006) Toxicity of diclofenac to Gyps vultures. Biology Letters 2: 279–282. Taggart MA, Cuthbert R, Das D, Sashikumar C, Pain D, and Green RE (2006) Disposition of diclofenac in Indian cow and goat with reference to Gyps vulture population declines. Environmental Pollution doi:10.1016/j.envpol.2006.08.017. Taggart MA, Senacha KR, Green RS, Jhala YV, Raghavan B, Rahmani AR, Cuthbert R, Pain DJ, and Meharg AA (2007) Diclofenac residues in domestic ungulates available to vultures in India. Environmental International 33: 759-765. Tewari VP and Arya R (2005) Degradation of arid rangelands in Thar Desert, India: a review. Arid Land Research and Management 19: 1-12. The Hindu (2007). Salman Khan poaching case verdict today. The Hindu: New Delhi national newspaper, e-paper. http://www.thehindu.com/2007/08/24/stories/2007082452090400.htm

67 van Eeden DG, Van Rensburg BJ, De Wijn M, and Bothma J (2006) The value of community-based conservation in a hetergenous landscape: an avian case study from sand forest in Maputaland, South Africa. South African Journal of Wildlife Research 36: 153-157. Viathiyanathan S, Raghavendra B, Mishra AS, Prasad R, Verma DL, and Singh NP (2007) Effect of feeding graded levels of Prosopis cineraria leaves on rumen ciliate protozoa, nitrogen balance and microbial protein supply in lambs and kids. Animal Feed Science and Technology 133: 177-191. Vulture Rescue (2008) Vulture Conservation Breeding Centres. Vulture Rescue, Pinjore, India. Available online at: http://www.vulturerescue.org/page8.html. Accessed 12 April 2008. Wackernagel M and Rees WE (1998) Our ecological footprint: reducing human impact on the earth. New Society Publishers: Waite TA, Campbell LG, Robbins P, and Chhangani AK (2007a La Niña’s signature: synchronous decline of the mammal community in a ‘protected’ area in India. Diversity Distrbutions 13: 752-760. Waite TA, Chhangani AK, Campbell LG, Rajpurohit LS, and Monot SM (2007b) Sanctuary in the city: urban monkeys buffered against catastrophic die-off from ENSO induced drought. EcoHealth 4: 278-286. Waylen, K.A., Fischer, A., McGowan, P.J.K., Thirgood, S.J. and Milner-Gullad, E.J. (2010) Effect of local cultural context on the success of community-based conservation interventions. Conservation Biology 24: 1119-1129. Wildlife Conservation Society (2008) Asian water birds stage remarkable comeback. Science Daily. Available online at: http://www.sciencedaily.com/releases/2008/04/080403125426.htm. Accessed 12 April 2008. Yadav RS, Yadav BL, and Chhipa BR (2008) Litter dynamics and soil properties under different tree species in a semi-arid region of Rajasthan, India. Agroforest Systems 73: 1-12.

Table 1. Vulture counts from each of the 11 villages from 1996 to 2005.

Sardar Jhala Year Khajarli Samand Chotila Dugar Nakora Dawra Bijapur Chhatri Parshuramji Sumer Dantiwara

1996 22 30 18 16 21 17 38 16 24 42 39

1997 25 39 21 19 25 21 45 19 31 51 48

1998 20 29 12 12 12 13 23 11 16 27 21

69 1999 15 22 10 15 14 10 20 12 19 20 20

2000 10 15 8 11 12 6 14 10 15 15 15

2001 12 18 10 13 10 8 16 7 12 17 10

2002 14 16 9 11 14 13 23 8 10 16 12

2003 13 17 12 14 12 10 19 5 14 21 16

2004 12 15 10 15 14 9 25 6 11 20 14

2005 16 17 13 16 13 12 22 6 12 26 14

Table 2. Annual population growth rates of Indian vultures in each surveyed village from 1996 to 2005.

Sardar Jhala Year Khejarli Samand Chotila Dugar Nakora Dawra Bijapur Chhatri Parshuramji Sumer Dantiwara 1996-97 0.128 0.262 0.154 0.172 0.174 0.211 0.169 0.172 0.256 0.194 0.208 1997-98 -0.223 -0.296 -0.560 -0.460 -0.734 -0.480 -0.671 -0.547 -0.661 -0.636 -0.827 1998-99 -0.288 -0.276 -0.182 0.223 0.154 -0.262 -0.140 0.087 0.172 -0.300 -0.049 1999-00 -0.405 -0.383 -0.223 -0.310 -0.154 -0.511 -0.357 -0.182 -0.236 -0.288 -0.288

2000-01 0.182 0.182 0.223 0.167 -0.182 0.288 0.134 -0.357 -0.223 0.125 -0.405 2001-02 0.154 -0.118 -0.105 -0.167 0.336 0.486 0.363 0.134 -0.182 -0.061 0.182 2002-03 -0.074 0.061 0.288 0.241 -0.154 -0.262 -0.191 -0.470 0.336 0.272 0.288 2003-04 -0.080 -0.125 -0.182 0.069 0.154 -0.105 0.274 0.182 -0.241 -0.049 -0.134 2004-05 0.288 0.125 0.262 0.065 -0.074 0.288 -0.128 0.000 0.087 0.262 0.000

Mean -0.035 -0.063 -0.036 0.000 -0.053 -0.039 -0.061 -0.109 -0.077 -0.053 -0.114 Variance 0.057 0.053 0.082 0.064 0.099 0.134 0.109 0.083 0.100 0.094 0.124

Sardar Jhala Village Khejarli Samand Chotila Dugar Nakora Dawra Bijapur Chhatri Parshuramji Sumer Dantiwara

Khejarli Sardar Samand 0.844

Chotila 0.678 0.848 71 Dugar 0.387 0.649 0.764

Nakora 0.307 0.239 0.360 0.522

Dawra 0.921 0.699 0.576 0.367 0.567

Bijapur 0.593 0.499 0.451 0.515 0.850 0.781

Jhala Chhatri 0.244 0.111 0.091 0.300 0.872 0.472 0.698

Parshuramji 0.254 0.523 0.747 0.820 0.554 0.248 0.322 0.332

Sumer 0.727 0.860 0.965 0.739 0.468 0.633 0.581 0.235 0.721

Dantiwara 0.371 0.457 0.655 0.641 0.785 0.456 0.600 0.535 0.862 0.730

Village Pop. Growth Rate (R) Best

Dugar 0 AR1; AR1+MEIt

Khajarli -0.035 AR1; AR1+MEIt

Chotila -0.036 AR1; AR1+MEIt

Dawra -0.039 AR1; AR1+MEIt

Nakora -0.053 AR1+MEIt; AR1+MEIt+MEIt-1

Sumer -0.053 AR1; AR1+MEIt

Bijapur -0.061 AR1+MEIt

Sardar Samand -0.063 AR1; AR1+MEIt+MEIt-1

Parshuramji -0.077 AR1; AR1+MEIt+MEIt-1

Jhala Chhatri -0.109 AR1+MEIt; AR1+MEIt+MEIt-1

Dantiwara -0.114 AR1; AR1+MEIt

Table 5. Goodness-of-fit as measured by stationary r2 for first-order ARIMA models. Villages (local breeding populations of Indian vultures) ordered from best to worst for the model with predictors AR1+MEIt.

Model parameters

Local population AR1 AR1+MEIt AR1+MEIt+MEIt-1

Sardar Samand 0.035 0.729 0.730

Dugar 0.477 0.574 0.591

Parshuramji 0.380 0.515 0.540

Jhala Chhatri 0.427 0.486 0.528

Bijapur 0.036 0.407 0.567

Nakora 0.281 0.405 0.651

Dantiwara 0.048 0.389 0.467

Chotila 0.087 0.338 0.447

Sumer 4.7×10-4 0.329 0.363

Dawra 1.7×10-4 0.287 0.318

Khejarli 0.018 0.272 0.280

Table 6. Results of autocorrelation analysis on time series of Indian vulture count data (transformation: natural logarithm; differencing: 1 year). Smaller Box-Ljung statistic values indicate stronger support for an underlying white noise process. Villages (l local breeding populations of Indian vultures) ordered based on ascending values of this statistic. Local breeding populations of Indian vulture in villages nearer the bottom of the list were apparently less influenced by ENSO.

Box-Ljung statistic

Village Autocorrelation Value Pa

Dugar -0.686 5.826 0.016

Jhala -0.644 5.132 0.023 Chhatri Parshuramji -0.601 4.477 0.034

Nakora -0.53 3.475 0.062

Chotila -0.269 0.892 0.345

Dantiwara -0.215 0.573 0.449

Bijapur -0.19 0.446 0.504

Sardar -0.179 0.395 0.530 Samand Khejarli 0.112 0.155 0.694

Sumer 0.02 0.005 0.945

Dawra -0.012 0.002 0.966

aTwo-tailed, based on chi-square distribution (df =1).

Table 7. Results from two non-parametric test comparing Bishnoi and non-Bishnoi villages for each village characteristic. Khejeri tree numbers are significantly different between two village types.

Mann-Whitney U- Test MRPP Village Characteristic p-value p-value A values T statistic

Khejeri Trees 0.017* 0.018* 0.147 -3.065 Average Annual Household Income 0.032* 0.061 0.059 -1.764

Total Population 0.620 0.612 -0.018 0.488

Farmland Area 0.283 0.605 -0.017 0.475

Total Livestock 0.509 0.467 -0.008 0.206

Axis 1 Axis 2

Village Characteristic r2 r2

Average Annual Household Income 0.852 0.459

Total Population 0.647 0.004

Farmland Area 0.516 0.002

Total Livestock 0.188 0.243

Effect F df p

Bishnoi 5.499 1 0.034*

Bishnoi * Axis2 5.539 1 0.034*

Axis1 1.038 1 0.326

Axis2 3.826 1 0.071

Bishnoi Blackbuck 2009 Total 2009 2010 Total 2010 Village Present? Present? Blackbuck #/Sample Blackbuck #/Sample Golia Yes Yes 12.00 6.00 10.00 2.50 Guda-Bishnoi Yes Yes 89.00 7.42 94.00 24.75 Khejarli Kallan Yes Yes 64.00 10.67 47.00 11.75 Khejarli Kurd Yes Yes 12.00 4.00 38.00 9.50 Bidasni Yes No 0.00 0.00 0.00 0.00 Birami Yes No 0.00 0.00 0.00 0.00 Kakelao Yes No 0.00 0.00 0.00 0.00 Sangahasni Yes No 0.00 0.00 0.00 0.00 Baniawas No Yes 6.00 3.00 57.00 14.25 Mortuka No Yes 9.00 4.50 0.00 0.00

78 Phitasani No Yes 6.00 6.00 6.00 1.50

Birdawas No No 0.00 0.00 0.00 0.00 Gujarawas No No 0.00 0.00 0.00 0.00 Lolawas No No 0.00 0.00 0.00 0.00 Miyasani No No 0.00 0.00 0.00 0.00 Modi Sutara No No 0.00 0.00 0.00 0.00 Palasni No No 0.00 0.00 0.00 0.00 Pisawas No No 0.00 0.00 0.00 0.00 Singahasni No No 0.00 0.00 0.00 0.00

2009 Total 2010 Total Blackbuck 2009 #/Sample Blackbuck 2010 #/Sample Bishnoi Villages 22.13 3.51 23.63 6.06 Non-Bishnoi Villages 1.91 1.23 5.73 1.43

Year-Form Ψ SE(Ψ) p SE(p) ε SE(ε) 2009-Actual 0.3684 0.1107 1.0000 0.0000 2009-Estimate 0.3160 0.1067 0.8327 0.0768

2010-Actual 0.3929 0.1389 0.7852 0.0947 2010-Estimate 0.3160 0.1067 0.8327 0.0768

Combined-Actual 0.3929 0.1389 0.8832 0.0577 0.0931 0.0716 Combined- Estimate 0.3689 0.1108 0.8879 0.0543 0.0944 0.0716

Table 12. R2 values of village characteristics associated with both NMS ordination axes. Axis 1 explains 62.9% of variation while axis 2 explains 33.4%. Number of Khejeri trees is the only characteristic not noticeably associated with either axis.

Axis 1 Axis 2 Village Characteristic r2 r2 Total Population 0.689 0.499 Agriculture Land Area 0.767 0.654 Community Land Area 0.587 0.402 Khejeri Trees 0.001 0.046 Total Livestock 0.482 0.812

Table 13. A) ANCOVA models ranked based on AIC values. All models are within 2  AIC values of each other and are thus appropriate models. SSE = standard sum of squares; k = number of model parameters. B) p-values for each model effect in for the final three models. No model effect is significant for all models suggesting no model effect influences income. Interaction effects were removed from each model in order of greatest p-value.

Model SSE k AIC delta AIC 1 0.134 7 -78.1327 -0.6307 2 0.154 6 -77.4896 -1.2738 3 0.160 5 -78.7634 0.0000

p value Effect Model 1 Model 2 Model 3 Bishnoi 0.592 0.562 0.633 Fallowing 0.162 0.112 0.067 Axis 1 0.036* 0.075 0.082 Axis 2 0.031* 0.060 0.066 Axis 1*Axis 2 0.034* 0.05* 0.051 Fallowing*Axis 1 0.169 0.508 Not Included Fallowing*Axis 2 0.222 Not Included Not Included

Axis 1 Axis 2

Village Characteristic r2 r2 Total Population 0.371 0.696 Agriculture Land Area 0.368 0.674 Community Land Area 0.149 0.326 Khejeri Trees 0.151 0.038 Total Livestock 0.702 0.728

Table 15. A) ANCOVA models ranked based on AIC values. Models 1 and 2 differ by less than 2 and are thus appropriate models to describe the influence of model parameters on average annual household income. SSE = standard sum of squares; k = number of model parameters. B) p-values for each model effect in for the final three models. Axis 1 is shows a significant effect on income for both models. Interaction effects were removed from each model in order of greatest p-value.

Model SSE k AIC  AIC 1 0.089 7 -79.5708 0 2 0.106 6 -78.4244 -1.1464 3 0.130 5 -76.7507 -2.8202 4 0.147 4 -76.5385 -3.0323

p value Effect Model 1 Model 2 Model 3 Model 4 Bishnoi 0.509 0.220 0.266 0.369 Fallowing 0.117 0.215 0.166 0.139 Axis 1 0.011* 0.021* 0.062 0.080 Axis 2 0.034* 0.072 0.208 0.201 Fallowing*Axis 2 0.158 0.065 0.241 Not Included Fallowing*Axis 1 0.062 0.140 Not Included Not Included Axis 1*Axis 2 0.195 Not Included Not Included Not Included

Table 16. Effects of village characteristics on average annual income (ANCOVA model 1). Axes refer to axis scores from NMS analysis (see Table 4 and Appendix 3).

Effect F df p Bishnoi 0.469 1 0.509 Fallowing 2.944 1 0.117 Axis1 9.722 1 0.011* Axis2 6.049 1 0.034* Bishnoi * Axis1 1.930 1 0.195 Fallowing * Axis1 4.424 1 0.062 Fallowing * Axis2 2.323 1 0.158

Table 17. Effects of village characteristics on average annual income (ANCOVA model 2). Axes refer to axis scores from NMS analysis (see Table 4 and Appendix 3).

Effect F df p Bishnoi 1.694 1 0.220 Fallowing 1.733 1 0.215 Axis1 7.185 1 0.021* Axis2 3.977 1 0.072 Fallowing * Axis1 2.531 1 0.140 Fallowing * Axis2 4.187 1 0.065

Figure 1. A) Map of Rajasthan. B) Map of village locations in relation to the major cities Jodhpur and Pali (pictured). Push pin icons mark village sites.

Figure 2. Time series of Multivariate ENSO Index (MEI). Positive MEI values indicate El Niño events and negative values indicate La Niña events.

Figure 3. Map of Rajasthan (above) and the study villages (below). Black arrow indicates Jodhpur city.

Figure 4. A) Schematic representation of one transect run in Khejarli Kalla village. The vertical white arrow points to the village center. The black arrow with the white border represents the “buffer transect” 500m from the village center. The white arrow with the black border represents the transect where Khejeri trees were counted. Each transect heading has a starting point 500m from the village center and an end point 1km from the village center. KKL stands for Khejarli Kalla. B)The solid black barbell line is a representation of a supplemental transect run equidistant from two original transect starting points. SSW stands for south of southwest.

140

120

100

40 Number of Khejeri Trees Khejeri of Number

0 -1.5000 -1.0000 -0.5000 0.0000 0.5000 1.0000 1.5000 NMS Axis 2 Scores

Figure 6. Aerial photo of one village in the study area, Khejarli Kalla village. The various land types within the village are labeled.

Figure 7. Photo of male blackbuck antelope crossing the road in Baniawas village.

*photo taken by Jonathan Hall

Appendix A: Chapter 3 Village demographic data. Average annual household income values given in Indian Rupees (INR). 1 USD = 47 INR approximately. Livestock populations include cattle, buffalo, sheep, goats, donkeys, and camels.

Total Khejeri Average Annual Bishnoi Trees within Household Income Farmland Total Livestock Village Present? Transect (Rs.) Total Population Area (ha) Population Baniwas no 33 25,000.00 1159 791.00 2164 Bidasni yes 131 37,500.00 574 518.00 235 Birami yes 29 26,200.00 2251 2181.20 5851 Birdawas no 26 31,500.00 1253 987.00 1251 Golia yes 71 26,300.00 576 729.80 715

94 Guda-Bishnoi yes 76 83,700.00 16207 4446.00 6610 Gujarawas no 40 25,833.00 412 660.00 395 Kakelao yes 96 34,700.00 2919 4162.00 3330 Khejarli Kallan yes 80 35,821.00 2129 1990.60 3220 Khejarli Kurd yes 69 29,600.00 695 1047.60 2265 Lolawas no 28 34,714.00 1547 717.28 1770 Miyasani no 48 15,000.00 1369 831.20 4909 Modi Sutara no 49 28,375.00 1887 657.00 638 Mortuka no 78 35,222.00 1210 1292.00 2315 Palasni no 53 28,176.00 7904 2989.12 12800 Phitasani no 88 23,917.00 700 322.00 585 Pisawas no 28 32,083.00 646 916.90 1633 Sangahasni yes 96 53,125.00 1126 500.00 703 Singahasni no 60 15,750.00 647 746.00 1350

Appendix B: Chapter 3 Axis scores for the 19 villages in the study. In this case the more negative the axis score the more positive the association with the axis.

Village Axis 1 Scores Axis 2 Scores

Baniwas 0.4758 0.2226

Bidasni -0.1106 -0.6286

Birami 0.0480 0.5772

Birdawas 0.0747 -0.2004

Golia 0.4969 -0.0314

Guda-Bishnoi -2.5766 -0.6099

Gujarawas 0.5922 -0.0551

Kakelao -0.5284 -0.0233

Khejarli Kallan -0.3707 -0.2040

Khejarli Kurd 0.1300 0.0191

Lolawas -0.1565 -0.2824

Miyasani 1.1250 1.3251

Modi Sutara 0.2980 -0.0530

Mortuka -0.2435 -0.2516

Palasni -0.7435 0.8968

Phitasani 0.7570 -0.0065

Pisawas 0.0332 -0.2286

Sangahasni -0.7751 -1.1388

Singahasni 1.4742 0.6727

Appendix C: Chapter 5 Demographic data collected from semi-structured interviews, School of Desert Sciences survey, and the 2001 Rajasthan census.

Community Number of Bishnoi Total Agricultural Land Area Khejeri Total Practice Bishnoi Village Present? Population Area (ha) (ha) Trees Livestock Fallowing? Population Bidasni yes 574 518 28.8 131 235 yes 448 Birami yes 2251 2181.2 92.9 29 5851 yes 834 Guda- Bishnoi yes 16207 4446 708 76 6610 yes 8024 Kakelao yes 2919 4162 148 96 3330 yes 1240 Khejarli Kallan yes 2129 1990.6 414.6 80 3220 yes 875 96 Khejarli Kurd yes 695 1047.6 209.3 69 2265 yes 88 Sangahasni yes 1126 500 17 96 703 yes 270 Golia yes 576 729.8 12.1 71 715 no 207 Baniwas no 1159 791 83 33 2164 yes - Modi Sutara no 1887 657 257.1 49 638 yes - Mortuka no 1210 1292 311.5 78 2315 yes - Pisawas no 646 916.9 81.8 28 1633 yes - Birdawas no 1253 987 54.8 26 1251 no - Gujarawas no 412 660 38 40 395 no - Lolawas no 1547 717.28 35 28 1770 no - Miyasani no 1369 831.2 23.6 48 4909 no - Palasni no 7904 2989.12 287.14 53 12800 no - Phitasani no 700 322 59 88 585 no - Singahasni no 647 746 63 60 1350 no -

Appendix D: Chapter 5 NMS axis scores for the nineteen villages in the study.

Village Axis 1 Axis 2 BNW -0.16449 -0.01331 BDS -0.64214 1.26786 BRM 0.36513 -0.92034 BRD -0.10953 0.28124 GOL -0.49789 0.74305 GBS 1.44316 -1.5815 GJW -0.80509 1.00378 KLA 0.75623 -0.6719 KKL 0.37466 -0.49535 KKD -0.34473 -0.10123 LLW 0.00803 0.1437 MYS 0.05251 -0.67545 MDS 0.30246 0.57923 MRK -0.03721 -0.17044 PLS 0.76108 -1.81182 PTS -0.33312 1.07675 PSW -0.44788 0.17454 SAN -0.10473 0.79073 SIN -0.47143 0.38046

Appendix E: Chapter 5 NMS axis scores for villages in the study excluding Guda Bishnoi.

Village Axis 1 Axis 2 BNW 0.1583 -0.0254 BDS -0.7628 -1.2537 BRM 0.7000 1.0201 BRD -0.1532 -0.1508 GOL -0.3545 -0.8042 GJW -0.4083 -1.2570 KLA 0.1976 1.2464 KKL 0.2808 0.7839 KKD 0.3525 -0.1484 LLW -0.0950 0.0477 MYS 0.7152 0.4915 MDS -0.6987 0.0530 MRK 0.2301 0.1912 PLS 1.1178 2.0672 PTS -0.7700 -0.8450 PSW 0.1582 -0.4076 SAN -0.6420 -0.4573 SIN -0.0260 -0.5518