Individual Species Predicted Habitat Suitability Models
This page provides access to predicted suitable habitat model outputs and associated reports and metadata for more than 2,000 species native or introduced to Montana. As additional species are modeled, reports will be available on this site if they are deemed worthy of distribution.
If you need a predicted habitat suitability model for a species that is not available on this website, please contact us at mtnhp@mt.gov
Modeling Approaches
Two approaches are used for modeling predicted suitable habitat for most species; inductive and deductive.
Inductive models are constructed using Maximum Entropy software (Phillips et al. 2006, Ecological Modeling 190:231-259) in conjunction with a variety of statewide biotic and abiotic layers standardized to 90 x 90-meter raster pixels and presence-only data for individual species contributed to Montana Natural Heritage Program databases and filtered to ensure spatial and temporal accuracy and remove spatial auto-correlation. The goal of inductive model outputs is to predict the distribution and relative suitability of habitat during the primary season of interest (usually breeding habitat, but overwintering habitat for winter migrant, and migratory habitat for select species that only migrate through the state) at large spatial scales.
Deductive models are simple, rule-based, associations with streams or ecological systems. The goal of deductive models is to spatially represent the stream reaches and ecological systems commonly and occasionally associated with individual species during their primary/breeding season of occupancy (e.g., breeding for resident and summer migrants, winter for winter migrants, and migratory for solely migratory species) across each species' known breeding range in Montana. The assignments of common and occasionally associated ecological systems, and how those assignments were made, can be seen in a tabular form under the "Ecological Systems Associated with this Species" section of species accounts in the Montana Field Guide.
Model Outputs
Model outputs are in the form of a logistic value that ranges from 0-1 with lower values representing areas predicted to be less suitable habitat and higher values representing areas predicted to be more suitable habitat. If enough observations were available to train and evaluate a model, the continuous output is reclassified into suitability classes (unsuitable, low suitability, moderate suitability, and optimal suitability) for each 90 x 90-meter pixel. We then aggregate the classified model output into 1-square mile (259 hectares) hexagons. Model reports include a summary of model performance, links to model ouput in GIS format, a summary of the modeling process, evaluation metrics, thresholds for suitability categories and associated areas of predicted suitable habitat, a variety of tabular and map outputs, and a detailed description of environmental input layers and observations used for modeling.
Model Limitations and Suggested Uses
Models are based on statewide biotic and abiotic environmental layers originally mapped at a variety of spatial scales and standardized to 90×90-meter raster pixels. The spatial accuracy of the training and testing data are varied (typically 20-400 meters) and may result in additional statistical noise in the model. As a result, model outputs may not be appropriate for use on smaller areas or at fine spatial scales. Model outputs should not typically be used for planning efforts on land areas smaller than one quarter of a public land survey system (PLSS) section (<64 hectares) and model outputs for some species may only be appropriate for broader regional level planning efforts. Models should be interpreted as landscape-level habitat suitability (fundamental niche) and not as estimated distributions of the species (realized niche) since suitable habitat may be unoccupied. Consequently, model outputs should not be used in place of on-the-ground surveys for species, and wildlife and land management agency biologists should be consulted about the value of using model output to guide habitat management decisions for regional planning efforts or local projects. See Suggested Contacts for Natural Resource Agencies listed at the end of each report. In general, we have found across a large number of species representing a wide variety of taxa that experts believe optimal and moderate suitability classes represent landscapes where suitable habitat is often more continuous while the low suitability class represents landscapes where suitable habitat is often less continuous, scattered, or patchy. We encourage use of these classes for management, planning, permitting, survey, and other decisions accordingly.