Skip to main content
The University of Massachusetts Amherst

“Critical Evaluation of Methods and Outcomes for Habitat/Ecological Systems Classification and Mapping in the Northeast and Midwest U.S.”

Wednesday, February 11, 2015 | 10:30 am
134 Morrill Science Center Conference Room (speaker will present remotely)
David D. Diamond and Don Faber-Langendoen
David D. Diamond, Missouri Resource Assessment Partnership
Don Faber-Langendoen, NatureServe

The goal of this project was to review and compare four vegetation map products for the eastern USA, with an eye toward providing a single ‘best map product’ insofar as possible. Map producers included LANDFIRE, Southeast GAP Analysis, TNC, and NatureServe.  The LANDFIRE map covered the entire region, as did the NatureServe map, which is a compilation and revision of the LANDFIRE and GAP maps.  The producers all used ecological systems (or finer or coarser sets or subsets of them) as mapping targets together with cultural and ruderal types.  TNC, Southeast GAP, and LANDFIRE used somewhat similar methods, relying on plot-based ground data and a combination of direct classification of pixels and map overlays.  TNC did no new remote sensing analyses and LANDFIRE used few map overlays versus TNC and GAP.  NatureServe combined and modified the LANDFIRE and GAP maps, primarily based on expert review.  The four resulting maps are quite different, and they could not be readily mixed and matched to create a new, ‘best map’ in areas of overlap.  At coarser thematic resolution (e.g. Macrogroups within the National Vegetation Classification rather than Ecological Systems), the match among the maps was better.  The TNC map was more cartographically appealing and ecologically logical than other maps, but variation in accuracy among the maps could not be determined.  Low-cost solutions to improve the existing maps will result in small but possibly substantive gains, even though visible seams and discontinuities will remain in any result from such an effort.  Depending on their need, end users may want to create more accurate maps at their own cost or, they may find that the existing national maps are adequate for their purposes, despite the inaccuracies at the finest scales of mapping.  This webinar will summarize lessons learned from this comparison, with an eye to improving future mapping efforts of this kind.

David D. Diamond has been Director of the Missouri Resource Assessment Partnership at the University of Missouri for the past 18 years.  Previously he worked in Texas for The Nature Conservancy (4 years) and the Texas Parks & Wildlife Department (8 years).  Working as an ecologist, he wrote the first "modern" plant community classification for Texas.  Recently, he has worked on vegetation and geophysical site type mapping and conservation opportunity area identification at multiple scales, from regions (Lower Midwest) to states (Texas and Oklahoma) to local areas and management units (St. Louis Region, National Park units).  He has worked with colleagues to compile enduring features data for the conterminous USA, perform an ecoregion-based GAP-style analysis of enduring features, and provide an analysis of enduring features diversity hotspots by ecoregion.

Don Faber-Langendoen, Senior Research Ecologist, NatureServe.  As Senior Research Ecologist with NatureServe’s Conservation Science department, Don develops methods for the conservation of ecosystems.  He works with staff and the Network of Natural Heritage Programs and Conservation Data Centers in North America on vegetation classification, conservation status (endangered ecosystems) and ecological integrity assessment methods.  Don has a doctorate from St. Louis University, in association with the Missouri Botanical Garden. He is a member of the Editorial Board for Applied Vegetation Science.  Don lives in Syracuse, New York.