Abstract Listing by Session

SY10 Climate Change Case Studies: Using Historical Data to Predict Future Responses

Marlborough Room 1      Tuesday, December 6th 2011

Organizer(s): Toni Lyn Morelli, U.C. Berkeley

As managers and policymakers look to enable biological systems to adapt to anthropogenic climate change, and thus conserve biodiversity, scientists struggle to understand exactly how changes are affecting populations and ecosystems. The need for baseline data is apparent, but such data are scarce. In this symposium, researchers will present the latest sources of historical data and their use to reveal patterns in biological responses to climate change. Examples that will be discussed include museum specimens, genetic samples, long-term animal census data, forest and fire history, palentological research, and citizen science. Through these presentations we will show that there are untapped opportunities to capitalize on existing datasets for conserving species and understanding response to global change. We will close the symposium with a discussion of how better to utilize and synthesize existing data in a way that optimizes their use in conservation.

SY10.1   14:00  Historical Resurveys: Challenges in Revisiting the Past to Quantify Ecological Change and Project the Future Beissinger, SR*, Museum of Vertebrate Zoology, UC Berkeley ; Thorne, JH, UC Davis; Santos, MJ, UC Berkeley; Morelli, TL
Historical resurveys - where biodiversity surveys from the past are resampled - provided important opportunities to understand the influence of past environmental change on biodiversity, establish new benchmarks to understand future change, test the accuracy of range change predicted by bioclimatic models, and quantify past phenotypic, genotypic or trophic changes by comparing historic and modern specimens. Potential sources of historic datasets include field notes, correspondences, specimens, reports and images. Three main challenges must be overcome when using historic data: (1) Data quality varies greatly due to nonstandard protocols and imprecise locations for historic surveys; (2) Historic surveys are often limited to detection and nondetection data, which necessitates the use of occupancy modeling to obtain unbiased estimates of presence-absence; and (3) Ascribing causation to observed changes is often difficult because measures of change of external drivers are often uncertain. We illustrate these points through two exemplar projects: (1) the Grinnell Resurvey Project, where birds and mammals were sampled from 1911-1929 throughout the Sierra Nevada Mountains in California, USA, and recently resurveyed; and (2) the Wieslander Dataset, where vegetation over 40,000 km2 were mapped in the 1930s along a 300 km front of the Sierra Nevada and were recently remapped.
SY10.2   14:30  Identifying climate change refugia and population extinctions using landscape genetic analysis and occupancy modeling of historical and resurvey data Morelli, TL*, U.C. Berkeley
Natural history collections provide an opportunity to understand how populations have responded to recent climate change. As part of the Grinnell Resurvey Project, my research capitalizes on high-quality historical field survey data and specimen collections to understand how species have responded to global change over the last century. Specifically, I used occupancy modeling, population genetics, and geospatial analysis to identify population shifts, landscape connectivity, and climate change refugia across the Sierra Nevada. Through surveys conducted in 2010 and 2011, I found that Belding’s ground squirrels (Urocitellus beldingi) have disappeared out of nearly half of their historically surveyed California range. The extirpated sites were lower in elevation and had higher in mean annual temperature; populations maintained at hot, low sites were found in “anthropogenic refugia”, such as irrigated county parks and agricultural fields. I then analyzed the genetic diversity of U. beldingi and identified barriers to gene flow across California. Finally, I used genetic and occupancy data to identify climate change refugia, areas that increase persistence of populations and, as a result, maintain higher genetic diversity. The results of this study will help to identify landscape features central to climate change refugia and thus aid land managers in allocating limited resources to develop climate change adaptation strategies, including increasing resilience on the landscape.
SY10.3   14:45  Investigating the response of animals to temperature shifts at a variety of temporal scales Smith, Felisa A.*, University of New Mexico ; Murray, Ian W., University of New Mexico
Arguably the most pressing environmental issue facing society today is that of anthropogenic climate change. Considerable debate centers around the magnitude, timing and nature of biotic responses to such changes. The increasing availability of fine-scale historical information has led both to an appreciation for the rapidity and frequency of past shifts in the earth climate system, as well as an interest in exploring past biotic responses. Here, we examine the response of small mammals to late Quaternary climate fluctuations at a variety of temporal scales. We focus on Neotoma (woodrats) because they construct middens, which can be preserved for thousands of years. Paleomiddens yield information on morphology, genetics and diet. We integrate this historical record with museum specimens and modern studies to characterize climatic thresholds leading to species extirpation and/or range alterations rather than in situ adaptation. Our results demonstrate remarkable congruence across the geographic range at a variety of temporal scales. Chronosequences demonstrate woodrats respond as expected on the basis of Bergmann's rule: colder climatic conditions select for larger body size and warmer conditions select for smaller body size. For this genus, morphological adaptation is the dominant mode in coping with changing climatic regimes. Our work documents the entire range of responses – phenotypic adaptation, migration and species replacements, and local extirpation.
SY10.4   15:00  Assessing species vulnerability to climate change: use of breeding bird survey data to develop distribution models and predictions for the 21st century MAGGINI, R*, Swiss Ornithological Institute, Sempach, Switzerland ; Lehmann, A, Climatic Change and Climate Impacts, Institute for Environmental Sciences, University of Geneva, Switzerland; Zimmermann, NE, Swiss Federal Research Institute WSL, Birmensdorf, Switzerland; Bolliger, J, Swiss Federal Research Institute WSL, Birmensdorf, Switzerland; Foppen, R, European Bird Census Council, Beek-Ubbergen, the Netherlands; Schmid, H, Swiss Ornithological Institute, Sempach, Switzerland; Beniston, M, Climatic Change and Climate Impacts, Institute for Environmental Sciences, University of Geneva, Switzerland; Zbinden, N, Swiss Ornithological Institute, Sempach, Switzerland
Climate change is affecting biodiversity worldwide and represents a new challenge for conservation. Project ClimBird has three main objectives: i) to assess to which extent breeding birds have already shifted their distribution along the elevational gradient in Switzerland; ii) to predict how the forecasted changes in climate but also in land use will modify their distribution across the country in the future; iii) to identify the most vulnerable species so as to efficiently inform authorities of the seriousness of the threat and to prioritize conservation actions. The assessment of the elevational range shift was performed using the data of the Swiss national monitoring program for common breeding birds and revealed a significant upward shift for 33 species. The current distribution across Switzerland was modelled using data from different in-house sources: monitoring, atlas, ornithological databases. The modelled distributions were then projected for the 21st century according to combined scenarios of climate and land use change. Results clearly show that the greatest species turnover is expected for the alpine region and more specifically for the Eastern Alps. A vulnerability index was defined on the basis of the change in distributional areas and the ranking suggests that most vulnerable species will be species with particular habitat requirements and alpine species for which Switzerland has a key responsibility in the European alpine landscape.
SY10.5   15:15  Current trends in french bats population highlghts by old heterogenous dat Kerbiriou, C*, Conservation des Espèces, Restauration et Suivi des Populations, UMR 7204 MNHN-CNRS-UPMC, 61 rue Buffon, Paris, France ; Julien, JF, Conservation des Espèces, Restauration et Suivi des Populations, UMR 7204 MNHN-CNRS-UPMC, 61 rue Buffon, Paris, France ; Marmet, J, Conservation des Espèces, Restauration et Suivi des Populations, UMR 7204 MNHN-CNRS-UPMC, 61 rue Buffon, Paris, France ; Robert, A, Conservation des Espèces, Restauration et Suivi des Populations, UMR 7204 MNHN-CNRS-UPMC, 55 rue Buffon, Paris, France ; Lemaire, M, Muséum d\'histoire naturelle de Bourges, Les Rives d’Auron, Allée René Ménard, 18000 Bourges; Arthur, L, Muséum d\'histoire naturelle de Bourges, Les Rives d’Auron, Allée René Ménard, 18000 Bourges; Lois, G, NatureParif, 84, Rue de Grenelle 75007 ; Couvet, D, Conservation des Espèces, Restauration et Suivi des Populations, UMR 7204 MNHN-CNRS-UPMC, 55 rue Buffon, Paris, France
In the context of biodiversity loss, we need information of population trend at large time and space scale, however well documented animal population dynamics are generally scarce, short time series and based on heavy protocols requiring animal manipulation, which are usually impossible to conduct in species of conservation concern. For bat species, an alternative approach could consist in doing appropriate analysis of participating networks monitoring. We firstly observed that whereas a great proportion of European bat species have a bad conservation status due to various pressures (agriculture intensification, urbanization or forest management), during the last 10 years, French bats populations seem to stabilize and even for some species, to slightly increase. In order to have a better understanding of the current trends in bat populations we used data rediscovered from old registers (data provide by tags' museum registers, count in roosts and data from care centers of wildlife, from 1939 until now) and sometime we even had the opportunity to interview their producer. Here, we compare the strong decline observed thanks to these old data with recent counts in roost cavities by evaluating differences in species distribution, roots communities’ composition and population abundance variations. Using population dynamics modeling we conclude that in order to attempt meaningful analysis of such time series and provide a source of data for implement biodiversity indicator, it is necessary to include local knowledge of people involve on field survey in these analyses (existence of disturbances, site protections) with the aim to assess the impact of climate changes and land use changes.
SY10.6   15:30  Subjective decisions and uncertainty in species distribution models Baumgartner, JB*, School of Botany, The University of Melbourne ; Regan, TJ, School of Botany, The University of Melbourne; Wintle, BA, School of Botany, The University of Melbourne; Elith, J, School of Botany, The University of Melbourne
Species distribution models (SDMs) are commonly used for making predictions about the impacts of threatening processes, such as climate change, on species’ distributions. These models are typically correlative, identifying determinants of species occurrence by finding statistical associations between occurrence localities and environmental characteristics. However, although practical and widely adopted, this approach suffers from a range of uncertainties that emerge from subjective judgements made throughout the modelling process. Two key sources of subjective uncertainty lie in the choice of relevant environmental covariates, and in the treatment of habitat components critical to the species’ persistence (e.g. geological features). The present research investigates the consequences of these subjective decisions on the outcomes of species distribution modelling through a case study of an alpine specialist, the endangered Australian mountain pygmy-possum, Burramys parvus. The Maxent algorithm was used to develop a range of SDMs for this species, each of which incorporated a unique subset of ecologically-relevant environmental correlates for habitat suitability, as well as alternative methods for representing the influence of boulderfields on the suitability of habitat for pygmy-possums. This study highlights the uncertainties arising from subjective modelling choices when predicting the impacts of climate change on biodiversity.
DISCUSSION - Discussion period commences after all presentations have finished