Using Species Distribution Models for Conservation Planning and Ecological Forecasting
C. Ashton Drew, Yolanda F. Wiersma, Falk Huettmann
Conservation practitioners and resource managers must often work with limited data to answer critical, time-sensitive questions. In many regions of the world, even the most basic information about the distribution of species is lacking. Knowing the geographic extent of a given species or ecological system is the first step in planning for its management or conservation. The sustainable management of fish stocks, timber, waterfowl populations, and biodiversity in general requires high quality spatial data on species distributions. Selecting preserves or easements to protect plants and wildlife, for instance, requires detailed knowledge of where different species are on the landscape. Such information is the foundation of science-based management and is necessary for assessing the risks of land-use actions, management scenarios, or other human activities to plant and wildlife populations (Huettmann et al. 2005). Species distribution models provide one tool for addressing the lack of species distribution data (Boyce and McDonald 1999). These models can be used to fill gaps in our knowledge by projecting habitat suitability in areas with few or no occurrence records. These models can also be used to forecast the effects of changes in environmental conditions on species distributions. Given the immense threat that climate change, land-use change, and invasive species pose to ecosystem services and to rare species in particular (Wilcove et al. 1998; Sala et al. 2000), having the ability to forecast potential future impacts allows practitioners to assess alternative policies and actions and to plan for change (Huettmann et al. 2005; Nielsen et al. 2008). Thus, forecasting will likely play an even more important role in conservation as forecasting tools become more accurate and more accessible. Many different types of models have been applied to these problems. These have included highly mechanistic models that simulate population dynamics, species interactions, and dispersal (Bugmann 1996; Schumaker et al. 2004; Battin et al.2007), empirical models that rely on correlative relationships between known distributions and environmental conditions (Guisan and Zimmermann 2000), and models that combine mechanistic and empirical approaches (Iverson et al. 2004). Here, we explore how species distribution models can be applied to two types of conservation applications: conservation planning and ecological forecasting. We begin with a brief description of these conservation applications, concentrating on some of the specific questions that species distribution models have been used to help answer. We then discuss the types of models that are available for these applications and whether, and how, those models have been validated. We conclude with two case studies in which we describe specific applications of species-distribution modeling to reserve selection and to forecasting climate impacts.