The use of real and fictitious surfaces for territorial distribution assessment of given geographic phenomenon

Abstract

Considering the usability of real or fictitious surfaces, we analyze their formation both in terms of purpose and input data. Relief is understood as a major distribution factor of processes in the landscape. This concept is related to both continuous (e.g. air temperature, rainfall) and a discrete (various forms of land use) phenomena. While at continuous phenomena simulated surfaces of terrain distributed phenomena respect the configuration of the relief directly, simulated surfaces reflect rather discrete phenomena. Although largely shaped by terrain they differ from the relief itself, not regarding their different nature. A paper discusses within physical and human geography the issue of raster surfaces making. In case of real surfaces, authors use for calculation known regression dependences of air temperatureand precipitation on various relief parameters regarding deeper intention of drought risk assessment. This serves as an evidence of dependences of given parameters on digital elevation model which results in higher accuracy opposite to fictitious surfaces generated using spatial analyst or statistic. On the example of Focal Statistic tool (Neighborhood Statistic) and IDW interpolation method differences towards the real state (e.g. concentric shapes in IDW generated surfaces) and too smoothed and generalized areas (using different size of moving window) are described and discussed. When defining the areas of post-industrial landscape authors came to the conclusion that not very often recommended IDW interpolation does not make so big inaccuracies as compared with Focal Statistic tool.