Learn More

  • Learn more on how NatureServe has used PDM to map species in Latin America
  • For more information on PDM, download the white paper "Element Distribution Modeling: A Primer". (PDF, 328 KB)
  • Download fact sheet on PDM. (PDF, 675 KB)

Products & Services

Predictive Distribution Modeling

Predictive Distribution Modeling (PDM) is an innovative GIS-based method used to produce predictive maps of where elements (i.e., species, ecological community type) are likely to occur and likely not to occur. The probability of occurrence is quantified and is directly related to underlying environmental variables and the locations of known occurrencesw. There are several advantages to using PDM (also known as "element distribution modeling") for inventory and conservation planning:

  • Maps of documented occurrences ("dot maps") convey no information on likelihood of element occurrence in areas that have not been surveyed. Range maps from field guides and similar treatments are too coarse to inform on-the-ground action or study.
  • Good predictive distribution maps make field inventories more efficient and effective. They show where to commit limited inventory resources for the highest likelihood of documenting a target element.
  • Predictive distribution maps are crucial to state comprehensive wildlife conservation strategies and other agency planning efforts (e.g., USFS Regional and Forest Plans, BLM Resource Management Plans, USGS Gap Analysis).
  • Predictive distribution maps for multiple elements, all produced with consistent and defensible methods, are very well suited for identifying spatial patterns in biological diversity.
PDM results for a rare bird species in the Andes of Peru and Bolivia
PDM results for a rare mammal distribution in the Andes of Peru and Bolivia: yellow points are known occurrences, green is the predicted occurrence from PDM, and the blue line is the original range map.

Predictive maps are produced through PDM three steps:

  • compile spatial data associated with the target element and the environment in the area of interest;
  • build a statistical model based on the association of the element to environmental variables (e.g., vegetation, soils, landform, climate) at sites of known occurrence, and
  • map that model via GIS across the area of interest. Models may be deductive (based on suspected habitat affinities) or inductive (based on extrapolating from known occurrences)
PDM results for a rare bird species in the Andes of Peru and Bolivia
Schematic diagram indicating the steps involved in Predictive Distribution Modeling

Several natural heritage programs have successfully used PDM to guide inventory work, most notably the Oregon Natural Heritage Information Center, New York Natural Heritage Program, and Wyoming Natural Diversity Database (WYNDD). NatureServe has used PDM extensively in Latin American mapping efforts. In addition, NatureServe and WYNDD have provided PDM training to over 20 network programs as well as other public and private partners.

For a more detailed discussion of the strengths and limitations of PDM, download the white paper "Element Distribution Modeling: A Primer". (PDF, 328 KB)