Gis-Based Niche Modeling for Mapping Species' Habitat

Abstract

Ecological "niche modeling" using presence-only locality data and large-scale environmental variables provides a powerful tool for identifying and mapping suitable habitat for species over large spatial extents. We describe a niche modeling approach that identifies a minimum (rather than an optimum) set of basic habitat requirements for a species, based on the assumption that constant environmental relationships in a species' distribution (i.e., variables that maintain a consistent value where the species occurs) are most likely to be associated with limiting factors. Environmental variables that take on a wide range of values where a species occurs are less informative because they do not limit a species' distribution, at least over the range of variation sampled. This approach is operationalized by partitioning Mahalanobis D2 (standardized difference between values of a set of environmental variables for any point and mean values for those same variables calculated from all points at which a species was detected) into independent components. The smallest of these components represents the linear combination of variables with minimum variance; increasingly larger components represent larger variances and are increasingly less limiting. We illustrate this approach using the California Gnatcatcher (Polioptila californica Brewster) and provide SAS code to implement it.

High-resolution niche models via a correlative approach: Comparing and combining correlative and process-based information

“…We modeled occupancy during May and June based on the reproductive behavior seen in the literature and at our site (R. W. Moody, Tinker Air Force Base, unpublished 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 data) using the partitioned Mahalanobis distance statistic (Browning et al, 2005;Rotenberry et al, 2006), which requires presence-only data, to model occupancy. The Mahalanobis distance ( 2 D ) is a measure of occupancy based on p variables measured at n locations:…”

“…where y is the 1  p vector of measurements taken at any point, and  is the 1  p vector of means for each environmental data layer (Clark et al, 1993;Rotenberry et al, 2006). To overcome potentially restrictive model output (Knick and Rotenberry, 1998), we partitioned the statistic via principal component analysis performed on the correlation matrix of environmental data taken at animal locations (Rotenberry et al, 2002;Browning et al, 2005;Rotenberry et al, 2006).…”

“…To overcome potentially restrictive model output (Knick and Rotenberry, 1998), we partitioned the statistic via principal component analysis performed on the correlation matrix of environmental data taken at animal locations (Rotenberry et al, 2002;Browning et al, 2005;Rotenberry et al, 2006). Partitioned Mahalanobis models emphasize a minimum set of habitat characteristics (compared to an optimum set of habitat characteristics, which a full-rank Mahalanobis model seeks to achieve; Rotenberry et al, 2006), and are calculated by:  distribution with n+1-p degrees of freedom (where n = total number of components, p = target component), which allows rescaling of the Mahalanobis distance (which can range from 0 to infinity) for ease of display (Rotenberry et al, 2006). Niche-only models included two thermally-derived and one prey GIS layer; combination niche-habitat models included three niche layers plus the top six habitat layers; and habitat-only layers included only those top six habitat layers.…”

“…For the purposes of statistical tests (see below), we selected enough pseudo-absence points to equal the sample size in each validation dataset. We conducted PCA using the correlation matrix of development dataset locations intersected with pixel values of all model layers to determine the components that were most limiting (and thus most likely to be representative of a realized niche; Rotenberry et al, 2006). Mahalanobis distances were calculated using Python scripts (Appendix C).…”

eHabitat, a multi-purpose Web Processing Service for ecological modeling

“…This approach is typically used for ecological niche modeling, in which a spatial prediction model for a given species is computed from a set of environmental parameters, or 'indicators' (see e.g. Clark et al, 1993, Knick and Dyer, 1997, Rotenberry et al, 2006. In this context, Geographic Information Systems (GIS) have proven to be very useful tools for conservation because of the ease of handling various thematic layers and using multi-criteria decision trees for extracting information.…”