MARXAN site selection analysis results

This data set is the output of MARXAN site selection work performed by The Wildlands Project. We assembled three major tracks of data to perform these analyses: Ecological Variation, Focal Species, and Special Elements.

To represent the variation in ecological conditions that exists across the region we used a data layer of Ecological Land Units (ELUs) developed by The Nature Conservancy. ELUs are unique combinations of elevation, geology and landform - that are important to the distribution and abundance of ecological communities in the ecoregion.

We used data from a focal-species analysis, performed by Carlos Carroll, that identified areas of high quality (source) habitat for three species of carnivores: Canada lynx, American marten, and eastern gray wolf. All three species are native to the study area but are considered threatened or extirpated in some or all of the region. Vertebrate carnivores were used as focal species because they are vulnerable or sensitive to human activities and human-induced landscape change.

For our special elements track we used a geographic data layer of modeled occurrences of small to moderately-sized, high quality ecosystems developed by The Nature Conservancy. This database consisted of nine ecosystem types: open wet basins, pine barrens, open dry flats, beach dune, river systems/coves, floodplains, forested wetlands, steep slopes and cliffs, and summits.

MARXAN, a program used for site selection, was used to find a set of planning units that met our conservation goals while minimizing cost. In our study, we created hexagonal planning units of 10 km2. We populated these hexagons with spatial data from our 181 conservation features (162 for environmental variation, 9 for special elements, and 10 for focal species and set conservation goals. Hexagons containing more than 75% of currently protected areas (GAP 1 and 2) were locked into every solution. Hexagons containing more than 50% urban areas were locked out of every solution. We created 12 distinct scenarios with varying levels of goal setting (high, high-medium, medium, low) and 3 levels of boundary length modifier (a measure of fragmentation.

This data set represents the average MARXAN score for each hexagon averaged over all 12 scenarios. A hexagon with a score of 100 was included in every run in every scenario. A hexagon with a score of zero was never included in any scenario.