Modeling plant species distributions under future climates: how fine scale do climate projections need to be?

Recent studies suggest that species distribution models (SDMs) based on fine-scale climate data may provide markedly different estimates of climate-change impacts than coarse-scale models. However, these studies disagree in their conclusions of how scale influences projected species distributions. In rugged terrain, coarse-scale climate grids may not capture topographically controlled climate variation at the scale that constitutes microhabitat or refugia for some species. Although finer scale data are therefore considered to better reflect climatic conditions experienced by species, there have been few formal analyses of how modeled distributions differ with scale. We modeled distributions for 52 plant species endemic to the California Floristic Province of different life forms and range sizes under recent and future climate across a 2000-fold range of spatial scales (0.008–16 km2). We produced unique current and future climate datasets by separately downscaling 4 km climate models to three finer resolutions based on 800, 270, and 90 m digital elevation models and deriving bioclimatic predictors from them. As climate-data resolution became coarser, SDMs predicted larger habitat area with diminishing spatial congruence between fine- and coarse-scale predictions. These trends were most pronounced at the coarsest resolutions and depended on climate scenario and species’ range size. On average, SDMs projected onto 4 km climate data predicted 42% more stable habitat (the amount of spatial overlap between predicted current and future climatically suitable habitat) compared with 800 m data. We found only modest agreement between areas predicted to be stable by 90 m models generalized to 4 km grids compared with areas classified as stable based on 4 km models, suggesting that some climate refugia captured at finer scales may be missed using coarser scale data. These differences in projected locations of habitat change may have more serious implications than net habitat area when predictive maps form the basis of conservation decision making.

Chapter Abstract:

Increasing temperatures have been recorded around the world, leading to changes in precipitation, sea-level rise and extreme events. Climate models are currently in use to simulate the effects of these changes on vegetation cover, which is a strong indicator of ecosystem changes in response to various drivers. Climate change, as well as anthropogenic stressors, is affecting forest dieback and tree-species migration. This chapter addresses the connections between changes in various forest types and the global soil carbon, nitrogen and hydrologic cycles, and related feedbacks between these factors and both natural and anthropogenic environmental changes. We discuss the ways these feedbacks between land use, vegetation changes and global nutrient and water cycles can lead to further climate change and soil degradation, which have profound effects on food security, and we conclude by proposing the use of soil characteristics as tools to inform land managers of challenges they may face in preserving valuable services from forested lands and cropping systems.

Ecosystem services play a crucial role in sustaining human well-being and economic viability. People benefit substantially from the delivery of ecosystem services, for which substitutes usually are costly or unavailable. Climate change will substantially alter or eliminate certain ecosystem services in the future. To better understand the consequences of climate change and to develop effective means of adapting to them, it is critical that we improve our understanding of the links between climate, ecosystem service production, and the economy. This study examines the impact of climate change on the terrestrial distribution and the subsequent production and value of two key ecosystem services in California: (1) carbon sequestration and (2) natural (i.e. nonirrigated) forage production for livestock. Under various scenarios of future climate change we predict that the provision and value of ecosystem services decline under most, but not all, future greenhouse gas trajectories. The predicted changes would result in decreases in the economic output for the state and global economy and illustrate some of the hidden costs of climate change. Since existing information is insufficient to conduct impact analysis across most ecosystem services, a comprehensive research program focused on estimating the impacts of climate change on ecosystem services will be important for understanding, mitigating and adapting to future losses in ecosystem service production and the economic value they provide.

Concern over rapid global changes and the potential for interactions among multiple threats are prompting scientists to combine multiple modelling approaches to understand impacts on biodiversity. A relatively recent development is the combination of species distribution models, land-use change predictions, and dynamic population models to predict the relative and combined impacts of climate change, land-use change, and altered disturbance regimes on species’ extinction risk. Each modelling component introduces its own source of uncertainty through different parameters and assumptions, which, when combined, can result in compounded uncertainty that can have major implications for management. Although some uncertainty analyses have been conducted separately on various model components – such as climate predictions, species distribution models, land-use change predictions, and population models – a unified sensitivity analysis comparing various sources of uncertainty in combined modelling approaches is needed to identify the most influential and problematic assumptions. We estimated the sensitivities of long-run population predictions to different ecological assumptions and parameter settings for a rare and endangered annual plant species (Acanthomintha ilicifolia, or San Diego thornmint). Uncertainty about habitat suitability predictions, due to the choice of species distribution model, contributed most to variation in predictions about long-run populations.

Prediction maps produced by species distribution models (SDMs) influence decision-making in resource management or designation of land in conservation planning. Many studies have compared the prediction accuracy of different SDM modeling methods, but few have quantified the similarity among prediction maps. There has also been little systematic exploration of how the relative importance of different predictor variables varies among model types and affects map similarity. Our objective was to expand the evaluation of SDM performance for 45 plant species in southern California to better understand how map predictions vary among model types, and to explain what factors may affect spatial correspondence, including the selection and relative importance of different environmental variables. Four types of models were tested. Correlation among maps was highest between generalized linear models (GLMs) and generalized additive models (GAMs) and lowest between classification trees and GAMs or GLMs. Correlation between Random Forests (RFs) and GAMs was the same as between RFs and classification trees. Spatial correspondence among maps was influenced the most by model prediction accuracy (AUC) and species prevalence; map correspondence was highest when accuracy was high and prevalence was intermediate (average prevalence for all species was 0.124). Species functional type and the selection of climate variables also influenced map correspondence. For most (but not all) species, climate variables were more important than terrain or soil in predicting their distributions. Environmental variable selection varied according to modeling method, but the largest differences were between RFs and GLMs or GAMs. Although prediction accuracy was equal for GLMs, GAMs, and RFs, the differences in spatial predictions suggest that it may be important to evaluate the results of more than one model to estimate the range of spatial uncertainty before making planning decisions based on map outputs. This may be particularly important if models have low accuracy or if species prevalence is not intermediate.

To anticipate the rapidly changing world resulting from global climate change, the projections of climate models must be incorporated into conservation. This requires that the scales of conservation be aligned with the scales of climate-change projections. We considered how conservation has incorporated spatial scale into protecting biodiversity, how the projections of climate-change models vary with scale, and how the two do or do not align. Conservation  planners use information about past and current ecological conditions at multiple scales to identify conservation targets and threats and guide conservation actions. Projections of climate change are also made at multiple scales, from global and regional circulation models to projections downscaled to local scales. These downscaled projections carry with them the uncertainties associated with the broad-scale models from which they are derived; thus, their high resolution may be more apparent than real.  Conservation at regional or global scales is about establishing priorities and influencing policy. At these scales, the coarseness and uncertainties of global and regional climate models may be less important than what they reveal about possible futures. At the eco-regional scale, the uncertainties associated with downscaling climate models become more critical because the distributions of conservation targets on which plans are founded may shift under future climates. At a local scale, variations in topography and land cover influence local climate, often overriding the projections of broad-scale climate models and increasing uncertainty. Despite the uncertainties, ecologists and conservationists must work with climate-change modelers to focus on the most likely projections. The future will be different from the past and full of surprises; judicious use of model projections at appropriate scales may help us prepare.

To conserve ecological connectivity (the ability to support animal movement, gene flow, range shifts, and other ecological and evolutionary processes that require large areas), conservation professionals need coarse-grained maps to serve as decision-support tools or vision statements and fine-grained maps to prescribe site-specific interventions. To date, research has focused primarily on fine-grained maps (linkage designs) covering small areas. In contrast, we devised 7 steps to coarsely map dozens to hundreds of linkages over a large area, such as a nation, province, or ecoregion. We provide recommendations on how to perform each step on the basis of our experiences with 6 projects: California Missing Linkages (2001), Arizona Wildlife Linkage Assessment (2006), California Essential Habitat Connectivity (2010), Two Countries, One Forest (northeastern United States and southeastern Canada) (2010),Washington State Connected Landscapes (2010), and the Bhutan Biological Corridor Complex (2010). The 2 most difficult steps are mapping natural landscape blocks (areas whose conservation value derives from the species and ecological processes within them) and determining which pairs of blocks can feasibly be connected in a way that promotes conservation.  Decision rules for mapping natural landscape blocks and determining which pairs of blocks to connect must reflect not only technical criteria, but also the values and priorities of stakeholders. We recommend blocks be mapped on the basis of a combination of naturalness, protection status, linear barriers, and habitat quality for selected species. We describe manual and automated procedures to identify currently functioning or restorable linkages. Once pairs of blocks have been identified, linkage polygons can be mapped by least-cost modeling, other approaches from graph theory, or individual-based movement models. The approaches we outline make assumptions explicit, have outputs that can be improved as underlying data are improved, and help implementers focus strictly on ecological connectivity.

An isolated population of the fisher (Martes pennanti) in the southern Sierra Nevada, California, is threatened by small size and habitat alteration from wildfires, fuels management, and other factors. We assessed the population’s status and conservation options for its habitat using a spatially explicit population model coupled with a fisher probability of occurrence model. The fisher occurrence model was selected from a family of generalized additive models (GAM) generated using numerous environmental variables and fisher detection–nondetection data collected at 228 survey arrays sampled repeatedly during 2002–2006. The selected GAM accounted for 69% of the Akaike weight using total above-ground biomass of trees, latitude-adjusted elevation, and annual precipitation averaged over a 5 km2 moving window. We estimated equilibrium population sizes (or carrying capacities) within currently occupied areas, and identified likely population source, sink, and expansion areas, by simulating population processes for 20 years using different demographic rates, dispersal distances, and territory sizes. The population model assumed that demographic parameters of fishers scale in proportion to habitat quality as indexed by the calculated probability of fisher occurrence. Based on the most defensible range of parameter values, we estimate fisher carrying capacity at 125–250 adults in currently occupied areas. Population
expansion into potential habitat in and north of Yosemite National Park has potential to increase population size, but this potential for expansion is predicted to be highly sensitive to mortality rates, which may be elevated in the northern portion of the occupied range by human influences, including roadkill and diseases carried by domestic cats and dogs.

Over the past few centuries, widespread disturbance of native forests of the conterminous United States has dramatically altered the composition, structure, extent, and spatial pattern of forestlands (Curtis 1956, Whitney 1994). These forests have been either permanently replaced by other land uses or degraded to varying degrees by unsustainable forestry practices, forest fragmentation, exotic species introduction, or alteration of natural disturbance regimes.

Habitat fragmentation is generally defined as the process of subdividing a continuous habitat type into smaller patches, which results in the loss of original habitat, reduction in patch size, and increasing isolation of patches (Andrén 1994). Habitat fragmentation is considered to be one of the single most important factors leading to loss of native species (especially in forested landscapes) and one of the primary causes of the present extinction crisis (Wilcox and Murphy 1985). Although it is true that natural disturbances such as fire and disease fragment native forests, human activities are by far the most extensive agents of forest fragmentation (Burgess and Sharpe 1981). For example, during a 20-year period in the Klamath–Siskiyou ecoregion, fire was responsible for 6% of forest loss, while clear-cut logging was responsible for 94% (Staus et al. 2001). Depending on the severity of the fragmentation process and sensitivity of the ecosystems affected, native plants, animals, and many natural ecosystem processes (e.g., nutrient cycling, pollination, predator–prey interactions, and natural disturbance regimes) are compromised or fundamentally altered. For many species, migration between suitable habitat patches becomes more difficult, leading to smaller population sizes, decreased gene flow, and possible local extinctions (Wilcove 1987, Vermeulen 1993).