Spatial Dynamics and Ecology of Large Ungulate Populations in Tropical Forests of India PDF
Preview Spatial Dynamics and Ecology of Large Ungulate Populations in Tropical Forests of India
N. Samba Kumar K. Ullas Karanth · James D. Nichols Srinivas Vaidyanathan · Beth Gardner Jagdish Krishnaswamy Spatial Dynamics and Ecology of Large Ungulate Populations in Tropical Forests of India Spatial Dynamics and Ecology of Large Ungulate Populations in Tropical Forests of India N. Samba Kumar • K. Ullas Karanth • James D. Nichols • Srinivas Vaidyanathan • Beth Gardner • Jagdish Krishnaswamy Spatial Dynamics and Ecology of Large Ungulate Populations in Tropical Forests of India N. Samba Kumar K. Ullas Karanth Centre for Wildlife Studies Centre for Wildlife Studies Bengaluru, Karnataka, India Bengaluru, Karnataka, India James D. Nichols Srinivas Vaidyanathan University of Florida Foundation for Ecological Research, Gainesville, FL, USA Advocacy and Learning Morattandi, Tamil Nadu, India Beth Gardner School of Environmental and Forest Sciences Jagdish Krishnaswamy University of Washington Centre for Biodiversity and Conservation Seattle, WA, USA Ashoka Trust for Research in Ecology and the Environment Bengaluru, Karnataka, India ISBN 978-981-15-6933-3 ISBN 978-981-15-6934-0 (eBook) https://doi.org/10.1007/978-981-15-6934-0 © The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2021 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore It is impossible for us to convey in the dry pages of text here, the sheer scale of all the hard field work, high motivation, intense focus and passion for wildlife, brought to bear on the task of massive data collection by our army of volunteers, as well as dedicated staff and field assistants at the Centre for Wildlife Studies during 1989–2017. They mastered the art of bravely sneaking past wild elephants, dodging angry sloth bears, and even worse, literally being eaten alive by tiny jungle ticks that burrowed into their skins. These “transect surveyors” stayed in dingy field camps, ate monotonous meals and withstood the harsh work regimen and disciplinary code imposed on them. The high quality and massive quantity of the data we present here are testimonies to their commitment. We humbly dedicate this monograph to all these citizen scientists. It has been our privilege to work with them for over three decades. Foreword I met Samba Kumar in the summer of 2007 when he spent a sabbatical at USGS Wildlife Research Center at Patuxent to work on analyses related to his PhD effort, the effort that ultimately resulted in this monograph on ungulate ecology and con- servation. At the time, one of the coauthors, Beth Gardner, was a post-doc in our group. To my mind, this time was Patuxent’s heyday not only in terms of staffing and scientific activity but also its vibrant and dynamic atmosphere. In those days, some of the hierarchical modeling ideas applied in this monograph were just being developed by members of our group, our collaborators, and colleagues. At least some of the ideas that came out of Patuxent during this period were inspired by col- laboration with Ullas Karanth who approached me through Jim Nichols who has been a long- term mentor to both of us. This collaboration between USGS Patuxent Lab and Centre for Wildlife Studies in India contributed to innovations such as the application of spatial capture–recapture models to camera trap data on tigers, occu- pancy models to sign survey data on mammals as well as the spatial distance sam- pling models to line transect data on ungulates, presented in this monograph. My engagement with spatial distance sampling work done by Samba Kumar was a part of this broader collaboration, which continues to this day (Karanth and Nichols 2017). This monograph by Samba and his coauthors is really two monographs rolled-up into one. On one hand, it is a monograph on spatial ecology and conservation of large ungulates in an important ecosystem, the Nagarahole–Bandipur landscape of southern India. This critical ecosystem includes a suite of ungulate species and also Indian tigers which several of the authors have been studying for most or all of their careers. Second, it is a monograph on applied hierarchical modeling. The important ecological and conservation context of this work is established in Chap. 1. As the authors emphasize, there is a sparsity of reliable information on ungulate abundance and distribution in south Asia, and this study addresses that information gap. The authors emphasize a number of important ecological and methodological challenges which structure the development and implementation of their study, statistical models, and analysis of the data. This sets the stage for the second viewpoint that this monograph is also a unique and very practical mono- graph on applied hierarchical modeling, and specifically what I would call “hierar- chical distance sampling” (HDS). Distance sampling is one of the most important methodologies in wildlife monitoring, and the type of data analyzed by Samba and coauthors are routinely collected in many different systems throughout the world. vii viii Foreword Hierarchical distance sampling concerns the situation where transects or points are surveyed using a distance sampling protocol, but the underlying abundance or den- sity of each sample unit varies spatially according to some ecological process. In HDS models, understanding the structure and dynamics of how density varies among units is a fundamental objective. In Chap. 2, the core methodological chapter of the monograph, the authors provide a comprehensive specification of the HDS model, allowing for group size to affect detection probability, covariates that affect density and detection, spatial variation in the form of a flexible model of spatial dependence, and Bayesian variable selection. In Chap. 3, the HDS model is applied to the suite of five ungulate species using a number of covariates which represent habitat, physical environment, and management effectiveness. Together, these approaches allow geographic predictions of density for these important ungulates that occupy various niches in this community. One novel methodological aspect of the monograph is that the model is a novel integration of density surface models (DSM; Miller et al. 2013) with a flexible model of spatial dependence (a conditional autoregression or CAR model). The authors provide a “fully Bayesian” implementation of the DSM framework in which both observation and process models are analyzed jointly, in contrast to the original two-stage analysis of Hedley and Buckland (1999). This monograph is the most comprehensive application of fully Bayesian DSMs in the literature, and it is one element which will be widely appreciated. In Chap. 3, the authors give a compre- hensive analysis of an extraordinary monitoring data set representing more than 1400 km of survey transects, and in the appendix they provide a complete R script to show how the models are analyzed in NIMBLE which greatly increases the prac- tical utility of this work. The management context of this study is detailed excellently in Chaps. 4 and 5. Chapter 4 is about quantifying the spatial distribution of proximate threats to ungu- lates, the development of a “composite human disturbance index” or CDI, and eval- uating management strategies to increase ungulate abundance. The authors formulate various management strategies involving elements of habitat manage- ment and law enforcement and then project the impact of those actions on ungulate density under new habitat scenarios induced by six alternative management regimes. The authors look at existing strategies and budgets to compute a “return on manage- ment efforts.” Anti-poaching measures provide the most “bang for buck,” which suggest changes to the existing management regime. The final chapter summarizes the big picture of ungulate conservation in this system and, importantly, outlines a vision for the future derived from key conservation, monitoring, models, and man- agement concepts developed as part of this work. The authors envision a formal adaptive management framework for the management of this landscape integrating decision-making with the biological (density/habitat models) and conservation con- text (increasing ungulate populations). This is perhaps an aspirational vision but very much consistent with the direction of the conservation field as a whole includ- ing the work of some of the monograph’s authors. In summary, “Spatial dynamics and ecology of large ungulate populations in tropical forests of India” represents a complete treatise on applied hierarchical Foreword ix modeling motivated by an important conservation problem, the development of a rigorous methodological framework along with the provision of an accessible implementation, and then the use of the model to evaluate explicit management actions. As an important ecological case study, or as a methodological exposition, this monograph is sure to be widely useful and informative. I hope wildlife managers, researchers, and conservationists who are investing much energy and resources on recovering all threatened ungulate species will avidly consider recommendations provided here in their efforts. USGS Patuxent Wildlife Research Center J. Andrew Royle Laurel, MD, USA March 27, 2020 References Hedley SL, Buckland ST, Borchers DL (1999) Spatial modelling from line transect data. J Cetacean Res Manage 1(3):255–264 Karanth KU, Nichols JD (eds) (2017) Methods for monitoring tiger and prey populations. Springer Nature Singapore Pte. Ltd. Miller DL, Burt ML, Rexstad EA, Thomas L (2013) Spatial models for distance sampling data: recent developments and future directions. Methods Ecol Evol 4(11):1001–1010 Preface Wild ungulates inhabiting tropical forests of the world are bearing the brunt of increasing human impacts of the current Anthropocene epoch. Once ubiquitous and abundant throughout their range, distributions and abundances of most species are declining at unprecedented rates. In spite of the recognition of the critical role of wild ungulates in maintaining the structure, stability and functioning of ecosystems they live in, science-based efforts focused at ungulate conservation are scarce, par- ticularly in the tropical regions where ungulates attain greatest diversity. Before the second half of the twentieth century, conservation efforts directed at populations of wild ungulates were constrained by the absence of rigorous methods to evaluate ungulate-habitat relationships to prioritize effective conservation inter- ventions. Subsequent development of better field survey methods and robust statisti- cal inferential approaches have laid the foundations for improved monitoring. However, the actual implementation of these improved methods by wildlife manag- ers as well as researchers has consistently lagged in tropical forested regions, which are inherently more difficult environments for conducting monitoring compared to Savannas and other open habitats. The lack of adoption of modern monitoring, we believe, is also—at least partially—due to the apathy of wildlife professionals in Asia. Numerous reports of methods aimed at “censusing” ungulates (in other words getting a total count), which is an impossibility in reality, are examples of this weak- ness. This managerial apathy to rigorous science arises also because wildlife researchers have generally not published large-scale case studies illustrating useful application of such science. Several factors that compound challenges faced in monitoring ungulate popula- tions need urgent attention. Most tropical forest ungulate species occur at relatively lower densities and consequently require field efforts to attain sufficient sample sizes demanded by design-based inference methods to estimate population density and abundance. The problems get further compounded because field counts of ungulates simply do not address measurement errors, most important being imper- fect detection, leading to the “noise” in the data often drowning the “signal” sought by the surveyors. This situation characterizes virtually all approaches to abundance estimation, including the case of distance-sampling-based line transect survey, which traces its history to the middle of the twentieth century, and is widely used for ungulate xi xii Preface species that can be visually detected. Early exposure of some of the authors to these issues set the stage for the evolution of this monograph. In 1989, the first author (Kumar), then an amateur naturalist, participated as a volunteer in the line transect survey of ungulate species conducted by the second author (Karanth). The survey objective was to rigorously assess the prey availability for tigers and their sympatric predators in a macro-ecological study forming the lat- ter’s graduate work at University of Florida. Karanth became deeply interested in line transect methodology, through his early interactions with field biologists Mel Sunquist, John Seidensticker, R. Rudran, and John Eisenberg. However, most cru- cially, he was also guided toward the right statistical approaches to the line transects by David Anderson at Colorado State University (in 1987) and James Nichols (third author) in 1989. In 1994, Kumar decided on a career in wildlife science, and joined as lead researcher in Karanth’s team at the Centre for Wildlife Studies, after quitting his position as a project manager in India’s space agency. Soon their focus progressed from implementing rigid field survey protocols (to ensure quality of data collected) to survey design, analysis, and estimation methods rooted in modern sampling the- ory applied under a likelihood-based inferential framework. The progress toward modern line transect methodology was enriched and facilitated greatly in 2000 by interactions with Stephen Buckland, David Borchers, Len Thomas, and Samantha Strindberg at the University of St. Andrews, which had by then become a global leader in this field. The continuing mentorship by the versatile James Nichols who straddled both the universes of likelihood theory and Bayesian inference, led Kumar and Karanth to the exciting field of hierarchical modeling of wildlife populations under a Bayesian paradigm pioneered by researchers at the US Geological Survey—William Link, Andy Royle, and Robert Dorazio. What impressed us most was the ability of their hierarchical models to elegantly deal with many thorny realities in the data generated from field sampling, without distracting us from ecological parameters that really mattered most. Along the way, other coauthors joined these efforts strengthening the team of authors. The most exciting aspect of working in such a team, consisting of passionate field biologists, quantitative ecologists, landscape ecologists, and mathematical statisticians, has been the ability we gained collec- tively to bridge the yawning gap between the rapidly unfolding methodological advancements and their full-fledged implementation in macro-ecological studies conducted in real world. In this monograph, we address the application of statistically robust hierarchical modeling approaches for understanding and conservation of tropical forest ungulate populations. Our aim is to demonstrate model development that can address thorny field sampling issues typically glossed over in many monitoring schemes, and, the application of these models to mechanistically explain spatial abundance patterns of five threatened ungulate species in a conservation priority landscape. Here, we expose readers to different important components of a science-based animal popu- lation monitoring program.
The list of books you might like
