Raquel Garcia.Pdf
Total Page:16
File Type:pdf, Size:1020Kb
UNCERTAINTY IN PROJECTED IMPACTS OF CLIMATE CHANGE ON BIODIVERSITY A FOCUS ON AFRICAN VERTEBRATES RAQUEL A. GARCIA University of Copenhagen | 2014 Uncertainty in projected impacts of climate change on biodiversity A focus on African vertebrates Raquel A. Garcia PhD thesis | January 2014 University of Copenhagen | Faculty of Science ACADEMIC ADVISORS Prof Miguel B. Araújo CMEC, University of Copenhagen, Denmark Imperial College, Silwood Park, UK National Museum of Natural History–CSIC, Spain InBio/CIBIO, Évora University, Portugal Dr Mar Cabeza Metapopulation Research Group, University of Helsinki, Finland SUBMITTED TO The PhD School of The Faculty of Science, University of Copenhagen, Denmark January 2014 ASSESSMENT COMMITTEE Prof Jane Hill University of York, Department of Biology, UK Dr Richard Pearson University College London, Research Department of Genetics, Evolution and Environment, UK Dr David Nogués-Bravo CMEC, University of Copenhagen, Denmark COPYRIGHT © 2014 Raquel A. Garcia (Synopsis, Design) © 2012 Blackwell Publishing Ltd (Chapter I) © 2014 The Authors Journal of Biogeography Published by John Wiley & Sons Ltd (Chapter II) © 2014 The Authors (Chapter III) © 2014 The Authors (Chapter IV) © 2014 The Authors (Chapter V) Contents | iii Contents Preface v Summary / Resumé vii Acknowledgements x Synopsis 1 The uncertain nature of assessments under climate change 5 Variation in data and model decisions 9 Robustness to ecological assumptions 11 Broadening the scope of assessments 13 Embracing uncertainty 15 References 18 Chapter I 27 Exploring consensus in 21st century projections of climatically suitable areas for African vertebrates Chapter II 65 Matching species traits to projected threats and opportunities from climate change Chapter III 87 Conservation implications of omitting rare and threatened species from climate change impact modelling Chapter IV 115 Multiple dimensions of climate change and implications for biodiversity Chapter V 147 Do projections from bioclimatic envelope models and climate change metrics match? Preface | v Preface This thesis is the result of a four-year PhD pro- ject based at the Center for Macroecology, Evo- lution and Climate at the University of Copen- hagen, in Denmark, and the National Museum of Natural History (CSIC), in Madrid, Spain. The project was supervised by Prof Dr Miguel Araújo and Dr Mar Cabeza. The thesis work also included various stays at the University of Hel- sinki in Finland, the 'Rui Nabeiro' Biodiversity Chair at Évora University in Portugal, and the South African National Biodiversity Institute in Cape Town, South Africa. The work was funded by a PhD grant from the Portuguese Foundation for Science and Technology. The thesis consists of two parts. The first part is a synopsis describing the background and aims of the thesis, summarising the main findings, and discussing the work presented in a broader perspective. The second part consists of five chapters that form the core of the thesis. They include four analytical chapters and one review chapter. At the time of submission, two were published as scientific articles, and the remaining three were in review. Summary / Resumé | vii Summary Evidence for shifts in the phenologies and dis- need to both integrate uncertainties in assess- tributions of species over recent decades has ments, and reduce or circumvent them where often been attributed to climate change. The possible. Integration of uncertainties is illus- prospect of greater and faster changes in cli- trated in two examples. The first uses ensem- mate during the 21st century has spurred a bles of bioclimatic envelope models for sub- stream of studies anticipating future biodiver- Saharan African vertebrates, built with alterna- sity impacts. Yet, uncertainty is inherent to tive climate data and model algorithms. En- both projected climate changes and their effects semble forecasting provides a means for ex- on biodiversity, and needs to be understood ploring the breadth and spatial variation of before projections can be used. This thesis uncertainties, and for building consensus seeks to elucidate some of the uncertainties among projections. Several consensus method- clouding assessments of biodiversity impacts ologies are compared here, including a newly from climate change, and explores ways to proposed methodology that preserves informa- address them. While the focus is mostly on sub- tion about the variability of projections in the Saharan African vertebrates, the methodologi- ensemble. The second example examines model cal advances and conclusions presented are far- outputs for sub-Saharan African amphibians in reaching and have wider relevance. the light of species' vulnerability to climate Throughout the chapters in this thesis, pro- change. An analytical framework is developed jections under changing climates for sub- for distinguishing between different climatic Saharan African vertebrates, based on biocli- threats and opportunities revealed by the bio- matic envelope models, are shown to be af- climatic envelope models, and analysing how fected by multiple uncertainties. Different they each are altered by the consideration of model algorithms produce different outputs, as specific response-mediating traits. do alternative future climate models and sce- Efforts to reduce uncertainties in biodiver- narios of future emissions of greenhouse gases. sity impact assessments are equally important. Another uncertainty arises due to omission of However, many sources of uncertainty cannot species with small sample sizes, which are easily be reduced, not least the omission of difficult to model. The effect of such bias species that are narrow-ranging, poorly known, against narrow-ranging species is often over- or even unknown to science. This uncertainty looked in assessments of biodiversity impacts, can instead be circumvented through the use of but our results for sub-Saharan African am- alternative approaches to assessing impacts. phibians show that it trickles down to conser- This thesis discusses one candidate approach vation strategies. Finally, assumptions about that is independent of species' data: the use of the climatic tolerance of species, their dispersal climate change metrics. By describing the expo- ability, and other characteristics are also shown sure of regions to multiple changes in the mag- to alter model projections for sub-Saharan nitude, timing, position, or availability of cli- African amphibians. matic conditions, metrics can provide infer- Despite numerous calls to address the un- ences about the potential threats and opportu- certainty challenge, appropriate treatment of nities for the biodiversity in those regions. The uncertainty has yet to be formalised in assess- diversity of existing metrics is reviewed here, ments of biodiversity impacts under climate and the picture that emerges is one of multifac- change. The chapters in this thesis highlight the eted changes in climate, with unequal spatial viii | Summary / Resumé patterns around the world. To help interpret The uncertainties discussed in this thesis, the diversity of climate change metrics, a con- and many others not covered here, impair the ceptual framework is proposed for using them conservation of biodiversity under changing in biodiversity impact assessments. Early test- climates in Africa and elsewhere. Explicitly ing of this framework, by comparing inferences addressing all uncertainties of projected im- from metrics and from the bioclimatic envelope pacts appears overwhelming. Yet, if model models for sub-Saharan African amphibians, projections are to be useful for conservation suggests that climate change metrics might be a planners, they must be as transparent as possi- useful addition to the biodiversity impact as- ble by including an honest description of their sessment toolbox. level of confidence given the current knowl- edge. Resumé De seneste årtiers evidens for ændringer i konklusioner mere vidt-rækkende og har bred arters fænologi og geografiske udbredelser er relevans. ofte blevet tilskrevet klimaforandringer. En gennemgående tendens i denne Udsigten til hurtigere og mere omfattende afhandling er at forudsigelser for afrikanske ændringer i klimaet i det 21. århundrede har hvirveldyr, baseret på bioclimatic envelope affødt en strøm af studier, der søger at models (klima-baserede modeller for en arts klarlægge de forventede konsekvenser for udbredelse) under klimaforandringer, er fremtidens biodiversitet. Dog ligger der en påvirkede af flere usikkerheder. Forskellige iboende usikkerhed i forudsigelser af model-algoritmer giver forskellige resultater, klimaforandringer, og deres indflydelse på og det gør alternative klimamodeller og biodiversiteten, som det er nødvendig at afklare scenarier for fremtidige drivhusgas- før disse forudsigelser kan anvendes. Denne udledninger også. En anden usikkerhed opstår afhandling sigter mod at udrede nogle af de som følge af at arter med små sample-sizes usikkerheder der forplumrer en vurdering af udelukkes, fordi de er svære at modellere. klimaforandringernes indflydelse på Effekten af denne skævvridning væk fra arter biodiversiteten, og udforsker måder at med små udbredelser overses ofte i håndtere dem på. Selv om fokus hovedsagligt vurderinger af konsekvenser for ligger på afrikanske hvirveldyr syd for Sahara, biodiversiteten, men vores resultater for er de metodologiske fremskridt og padder syd for Sahara viser at der er en effekt, Summary / Resumé | ix der afspejles helt ned i konkrete strategier for udbredelser,