Calculating landscape surface area from digital elevation models

There are many reasons to want to know the true surface area of the landscape, especially in landscape analysis and studies of wildlife habitat. Surface area provides a better estimate of the land area available to an animal than planimetric area, and the ratio of this surface area to planimetric area provides a useful measure of topographic roughness of the landscape. This paper describes a straightforward method of calculating surface-area grids directly from digital elevation models (DEMs), by generating 8 3-dimensional triangles connecting each cell centerpoint with the centerpoints of the 8 surrounding cells, then calculating and summing the area of the portions of each triangle that lay within the cell boundary. This method tended to be slightly less accurate than using Triangulated Irregular Networks (TINs) to generate surface-area statistics, especially when trying to analyze areas enclosed by vector-based polygons (i.e., management units or study areas) when there were few cells within the polygon. Accuracy and precision increased rapidly with increasing cell counts, however, and the calculated surface-area value was consistently close to the TIN-based area value at cell counts above 250. Raster-based analyses offer several advantages that are difficult or impossible to achieve with TINs, including neighborhood analysis, faster processing speed, and more consistent output. Useful derivative products such as surface-ratio grids are simple to calculate from surface-area grids. Finally, raster-formatted digital elevation data are widely and often freely available, whereas TINs must generally be generated by the user.