The loss of megafauna at the terminal Pleistocene has been linked to a wide range of Earth-system-level changes, such as altered greenhouse gas budgets, fire regimes and biome-level vegetation changes. Given these influences and feedbacks, might part of the solution for mitigating anthropogenic climate change lie in the restoration of extant megafauna to ecosystems? Here, we explore the potential role of trophic rewilding on Earth's climate system. We first provide a novel synthesis of the various ways that megafauna interact with the major drivers of anthropogenic climate change, including greenhouse gas storage and emission, aerosols and albedo. We then explore the role of rewilding as a mitigation tool at two scales: (i) current and near-future opportunities for national or regional climate change mitigation portfolios, and (ii) more radical opportunities at the global scale. Finally, we identify major knowledge gaps that complicate the complete characterization of rewilding as a climate change mitigation strategy. Our perspective is urgent since we are losing the Earth's last remaining megafauna, and with it a potential option to address climate change. This article is part of the theme issue ‘Trophic rewilding: consequences for ecosystems under global change’.

ISBN 978 1 86814 479 2All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the express permission, in writing, of both the author and the publisher.Cover photograph by Donald Cook at stock.xchng Cover design, layout and design by Acumen Publishing Solutions, JohannesburgPrinted and bound by Creda Communications, Cape Town Foreword SOUTH AFRICA and its people are blessed with diverse and thriving wildlife. We are also a developing economy with a growing population. From these facts emerges the particular situation of having most of our protected areas surrounded by land that has been transformed, to a greater or lesser extent, by human development. Large mammals, such as elephants, no longer roam the entire landscape, and their populations are no longer completely governed by the laws of nature. Protecting elephants and the ecological systems in which they exist in a practical and sustainable way that balances the needs of humans, elephants and the environment is a challenge to which I am committed.This Assessment was undertaken to reduce the degree of scientific uncertainty associated with decisions that must be made very soon and in the medium-to-long term. It helps to evaluate the costs and benefits associated with each choice, both in economic and ecological terms, and clarifies the legal framework within which they must be made. Collectively the chapters in this report reveal the many successes our country's experts, in collaboration with their peers in neighbouring countries and abroad, have achieved in understanding elephants and their needs, in fields as diverse as veterinary science, ecology, animal behaviour, population and resource modelling.Importantly, the Assessment exposes important gaps in our understanding and thus outlines necessary future avenues of research. This Assessment represents a key milestone in an ongoing Elephant Research Programme.Science does not provide all the information required to resolve the difficult issues raised by the management of elephant in a changing and humandominated world. Many of the required decisions have a strong element of human values implicit in them. How do South Africans wish to treat the other species with which they share our land? Extensive consultation and careful consideration of the values expressed by a wide range of stakeholders is also an essential part of the process of managing elephant in a democratic country.I am grateful to the many experts and interested persons who invested their time, experience and intellect to deliver this Assessment. I look forward to their continued engagement on the issue of elephant management, which is of great interest to many. Marthinus van Schalkwyk Minister of Environmental Affairs and Tourism, 2008 ContentS LiSt oF FigureS LiSt oF boxeS About the AuthorS And ContributorS Brandon Anthony is an assistant professor at the Department of EnvironmentalScienc...

A population of 805 horses (70 per cent dressage, 20 per cent show jumpers and 10 per cent trotters) with orthopaedic problems was examined for signs of lameness and back problems, irrespective of their original complaints. In the horses with a back problem the prevalence of lameness was 74 per cent, and back problems were diagnosed in 32 per cent of the lame horses. These percentages were significantly higher than those recorded in a control population of 399 horses, of which 20 per cent were lame and 12 per cent had back problems. In the group of horses with orthopaedic problems there was a strong association between lameness and back problems and, in particular, there was a high prevalence of lameness among the horses with back problems.

Surface water availability is a key driver of elephant impacts on biological diversity. Thus, understanding the spatio-temporal variations of these impacts in relation to water is critical to their management. However, elephant piosphere effects (i.e. the radial pattern of attenuating impact) are poorly described, with few long-term quantitative studies. Our understanding is further confounded by the complexity of systems with elephant (i.e. fenced, multiple water points, seasonal water availability, varying population densities) that likely limit the use of conceptual models to predict these impacts. Using 31 years of data on shrub structure in the succulent thickets of the Addo Elephant National Park, South Africa, we tested elephant effects at a single water point. Shrub structure showed a clear sigmoid response with distance from water, declining at both the upper and lower limits of sampling. Adjacent to water, this decline caused a roughly 300-m radial expansion of the grass-dominated habitats that replace shrub communities. Despite the clear relationship between shrub structure and ecological functioning in thicket, the extent of elephant effects varied between these features with distance from water. Moreover, these patterns co-varied with other confounding variables (e.g. the location of neighboring water points), which limits our ability to predict such effects in the absence of long-term data. We predict that elephant have the ability to cause severe transformation in succulent thicket habitats with abundant water supply and elevated elephant numbers. However, these piosphere effects are complex, suggesting that a more integrated understanding of elephant impacts on ecological heterogeneity may be required before water availability is used as a tool to manage impacts. We caution against the establishment of water points in novel succulent thicket habitats, and advocate a significant reduction in water provisioning at our study site, albeit with greater impacts at each water point.

Over recent decades, ungulate populations across Europe have undergone a rapid recovery. While this constitutes a conservation success, there is increasing concern about their impacts on shared resources with humans. Understanding ungulate food choices is crucial for predicting such impacts. Numerous studies have focused on single species or communities at narrow spatial scales. Here, we used 265 published diets from 87 European studies to investigate patterns of resource use by four common deer species (moose Alces alces, red deer Cervus elaphus, roe deer Capreolus capreolus and fallow deer Dama dama), and wild boar Sus scrofa at the continental scale. On average, deer diets separated mostly along a gradient from grass to browse. Fallow deer diets contained the most and moose diets the least amount of grass, but we also found large intraspecific variation among all deer species. Diets of roe deer, a presumed browser, frequently contained ≥ 25% grass. Wild boar diet contained grass in amounts similar to red deer but otherwise differed strongly from deer diets. All five ungulate species shifted to eating higher proportions of woody browse during winter. Habitat influenced variation in intraspecific diets, but the proportions of key forage types related to feeding type (i.e. grass for intermediate feeders red and fallow deer, and shrubs for the browsers moose and roe deer) remained fairly consisted across habitat types. In northern and central Europe, diet similarity between roe deer and red deer was highest during winter and spring and lowest during summer and autumn but remained constant across the seasons in southern Europe. We foresee that, as interspecific interactions driven by land‐use and climatic changes increase across Europe, further monitoring and testing will be needed to understand the dynamics of dietary niche partitioning among ungulates.