Assessment of land degration using remote sensing and geographic information system in Psiloritis region, Crete, Greece

The present study focuses on the use of Remote Sensing and Geographical Information Systems for land cover/use classification, assessment of probable and potential areas of soil erosion, as well as priority areas for soil erosion protection in a typical mountainous area of the Mediterranean region. Black and white aerial photographs (1:30,000), and Arc/Info and GRASS GIS were used to build up the database. Ancillary data such as topographic maps (1:50,000) geological maps (1:50,000 and 1:200,000) and other data in the form of tables and reports were incorporated into the GIS helping to give new insights on geomorphometric variables, rock, soil and rainfall distribution. Arc/Info GIS relational database offered very efficient ways to obtain information on environmental variables in forms of maps and statistics. Probable soil erosion areas were delineated based on the universal soil loss equation (USLE). Valuable information rangeland planning and management managed to be obtained by modelling soil erosion based on rate and weight assignments into categories of soil, rainfall, vegetation cover, slope gradient and length and geology. The most severe soil erosion areas were recorded on bare soils and areas covered by sparse vegetation. Furthermore, potential soil erosion areas were delineated after zero vegetation cover was assumed. This simulated quite a different pattern of soil erosion across the landscape portrayed by a potential soil erosion map. This map presented very large areas, in terms of soil, being severely and very severely eroded. For example, severe and very severe soil erosion was probable in about 5,900 ha, but when zero vegetation cover was assumed, this area increased up to 9,600 ha or by 61. Chromic cambisols and calcaric lithosols, which occupy that part of the rangeland under the highest grazing pressure, coincided with the areas classified as severely and very severely eroded in the potential soil erosion map. Priority areas for soil erosion protection were identified based on vulnerability and the severity of soil erosion. Vulnerable areas to soil erosion were highlighted and quantified by identifying probabilistic soil erosion classes using the potential soil erosion classes as identifiers. The analysis showed that the areas which should be given the maximum priority accounted for more than 1,550 ha or 6.7 of the area of erosion study, while minimum priority for soil erosion protection was required in only 520 ha or 2.2 of the study area. This study demonstrated that the integration of Remote Sensing and other ancillary data into a GIS can provide unique information by identifying and mapping vegetation cover, and assessing probabilistic and potential soil erosion areas as well as priority areas for soil erosion protection at a scale of 1:50,000.