Issue link: https://read.dmtmag.com/i/41362
Assessing Wildfire Response: Oneth by Land, Twoeth by Air BEYONDMAPPING T BY JOSEPH BERRY he initial attack against a wildfire generally takes three forms: ground attack, helicopter landing or helicopter rappelling. Terrain and land-cover conditions are used to determine accessible areas and relative initial-attack travel times for the three response modes. This and next month's "Beyond Mapping" column describes GIS modeling considerations and procedures for assessing and comparing alternative-response travel times. The discussion is based on a recent U.S. Forest Service (USFS) project undertaken by Fire Program Solutions (see "Author's Notes," page 11). I was privileged to serve as a consultant for the project that modeled the relative response times for all Forest Service lands from the Rocky Mountains to the Pacific Ocean (at a 30-meter grid resolution, that's a lot of little squares). Fortunately for me, I only had to work on the proto- type model, leaving the "heavy lifting" and "practical adjustments" to the team's extremely competent GIS specialist, wildfire professionals and USFS helitack experts. The project's objectives were to model the response times for different initial attack modes and provide summary maps, tables and recommendations for strategic planning and management of wildfire-response assets. Joseph Berry is a principal in Berry & Associates, consultants in GIS technology. He can be reached via e-mail at jkberry@du.edu. 10 Bring in the Birds The most challenging sub-model involved identifying helicopter land- ing zones (see Figure 1). A simple binary-suitability model is used to identify Potential Landing Zones (pLZs) by assigning a map value of 1 to all accessible terrain (gentle slopes and sub-alpine elevations) and land-cover conditions (no open water, forest or tall brush); with 0 assigned to inaccessible areas. Multiplying the binary set of maps derives a binary map of pLZs, with 1 identifying locations meeting all GEO W ORLD / AUGUST 2O11 the conditions (1 * 1 * 1 * 1 * 1 = 1); 0 indicates locations with at least one constraint. Interior locations of large, contiguous pLZ group- ings make ideal landing zones. However, edge loca- tions or small, isolated pLZ clusters must be further evaluated for clear helicopter approach/departure flight paths. At least three contiguous cells surround- ing a pLZ must have forest canopy of less than 57 feet to ensure adequate canopy clearance. In addition, it's desirable to have a negative slope differential of at least 10 feet to aid landing and takeoff. Weights and Measures Two steps are required for evaluating canopy clear- ance (see Figure 2). A reclassify operation is used to calculate a binary map, with canopy heights of 57 feet or less assigned a value of 1; 0 for taller cano- pies. A neighborhood operation (FocalSum in ArcGIS) is used to calculate the number of clear canopy cells in the immediate vicinity of each pLZ cell (i.e., a three-by-three roving window). If all cells are clear, a value of 9 will be assigned, indicating an interior location in a grouping of pLZ cells. For derived values less than 9, an edge location or isolated pLZ is indicated. If there are more than four surrounding cells with adequate clearance, at least three have to be contiguous, and the pLZ is assigned a map value of 1 to indicate that there's a clear approach/departure; 0 for locations with a sum of less than 4. Derived values indicating 3 or 4 clear surrounding cells must be further evaluated to determine if the Figure 1. A generalized outline shows a grid-based model for identifying Potential Landing Zones (pLZs) that are further evaluated for helicopter approach/departure considerations of canopy clearance and negative slope.