The heterogeneous distribution of snow in mountain regions often constitutes a great problem for an adequate modeling of snow dynamics. Remote sensing is usually used for the detection of snow distribution, but the fixed spatiotemporal resolution of these sources greatly reduces their efficiency, especially when the significant scale of the process under study is smaller than the sensor resolution. This is the case of the snow in semiarid areas, where rapid snow melting on sunny and warm days may substantially change the snow cover within just a few days, or where small snow patches (O~1 m) usually persist during many weeks even in early summer. Terrestrial photography (TP), which is taken from the ground, is a powerful alternative to remote sensing when scale issues arise, since its spatiotemporal resolution can be adapted to the scale of process, that is, to study snow evolution in such regions. This work shows a graphic user interface (GUI) developed in MATLAB to facilitate the specific steps in these images analysis required to quantify the area covered by snow: (1) georeference process, to provide spatial coordinates to the images, and (2) snow detection process, to distinguish snow-no snow pixels. Several inputs, such as the location of the camera, the coordinates of the central point of the image, a Digital Elevation Model (DEM) or some control points, are required. Different formats of the output images can also be selected. This tool is evaluated over Sierra Nevada Mountain (Southern Spain), where three different locations could successfully capture the variability of the snow behavior. The capability and versatility of TP for and easy and continuous monitoring of snow is shown from the results.
Pimentel, Rafael; Pérez-Palazón, María José; Herrero, Javier; and Polo, María José, "Monitoring Snow Cover Area In Semiarid Regions Using Terrestrial Photography" (2014). CUNY Academic Works.