Deforestation and land use change seem to have created serious problems in Golestan province, north of Iran . The objectives of this study were to evaluate the effect of deforestation, grazing, exclusion and rangeland degradation on soil quality attributes in loess-derived landforms . The results were shown that Loess/soil sequences in study area have been proven to be excellent paleoclimate archives. The results showed that the amount of organic matter decreased three units when it turns from forest to farmland, and increased two units from farmland to rangeland. The amount of CaCO 3 in surface layer of deforested area was more than the forest soils. The amount of soil N in forest and soil P and K in rangeland were high. Root decomposition and uptake by plants had an important role on distribution of N. Bulk density and porosity in forest and MWD in rangeland were higher than in other land uses because of the decrease in organic matter due to farming activities. Soil respiration in forest was highest as compared to other land uses. Decomposition of organic matter in farmland with no addition of plant residues have caused low respiration rate. Different enzymes activities (L-asparaginase and Dehydrogenase) compared to microbial respiration indicate that enzymes contribute to specific biological processes while soil microbial respiration basically depends on the general activity of soil microbial population. Different quality of the plant residues in different ecosystems may also create different impacts on biological indices. It could be concluded that amount of organic matter, soil N, bulk density, porosity, MWD, soil respiration and enzymes activities are suitable indicators for soil quality evaluation in this area. Calibration of the LISEM soil erosion model for Pasang catchment showed that the LISEM model can simulate runoff and soil loss from the catchment. Soil losses were estimated bu model and were differed from 85 to 1595 (Kg/ha) according to rainfall amounts. The effect of a number of land use and land management scenarios was simulated with LISEM. The simulation predicted that an increase in erosion amount about 21% was reached by changing of forest to current land uses. Decrease in soil loess amount of about 30% could be reached by implementing conservation methods and exclusion in farmland and degradated rangelands.
