Page 137 - GIS for Science, Volume 3 Preview

Using the connectivity modeling software Omniscape, the team compared the current flow across forested pixels before and after a modeled reduction of forest cover from these fires. In this simple illustration, they developed an input resistance grid with five land cover classes. Higher resistance values indicate reduced movement potential.
In the first model run (map A), researchers divided the forest cover values into four categories (low: <20%, medium: 20–40%, high: 40–60%, and very high: >60%) and scored those as 9, 6, 3, and 1 (no resistance). Areas in a non-forested but natural land cover type received a resistance weight of 10; agriculture pixels were scored 15; and urban areas, roads, and water were assigned the highest resistance value of 20. In the second connectivity model run (map B), the team integrated the VIIRS hot spot data and weighted burned areas in the low forest cover (resistance of 9) category. For both input maps, the Omniscape tool used algorithms based on electrical circuit theory (as implemented in Circuitscape software) to calculate current flow
A Pre-fire
forest connectivity
from all forested pixels in a 50-km radius moving window to a central pixel, with the amount of current emerging from each forested pixel assigned by reversing the resistance weight values (current source strength is 9 for pixels with very high forest cover and 1 for low forest cover), so the most current comes from the most highly forested pixels. The current flow outputs are summed for each moving window, and the final maps represent the cumulative movement potential for generalized forest-dependent species as a function of the structural pattern of forest cover on the landscape. The maps show areas in purple and red indicating high current flow, or movement potential, across large extents of forest. Orange and yellow areas indicate narrow bands of forest where current accumulates due to higher resistance in other neighboring pixels. Post-fire (map B), the modeled loss of forest cover, suggests further concentration of current (bright yellow) as the burned areas are avoided due to their high resistance and highlights the importance of narrow bands of forest along the edges of fires and between fire scars as potential corridors for forest-dependent species while these forests recover.
B Post-fire
forest connectivity
Circuit theory-based models of movement potential (current flow) for forest-dependent species that might avoid recent burns depicted in the VIIRS thermal hot spot imagery.
Mapping Extreme Events from Space 125