Effect of Pavement Cover Variation on Erodibility Factor of USLE Model UsingGeostatistics Technique (Case Study: Yazd- Ardakan plain)

The Soil erodibility Factor is one of the important and effective factor that is used for estimating water erosion by USLE model and its different versions. The main topic of this paper is the spatial distribution of K factor and knowing its effective factors. One of the factors which is usually ignored for K stimation,is the desert pavement density as the natural factor of soil conservation. The other purpose of this paper is the estimating of the variations of K factor due to landuse change, land plowging and combination with sub-layers soil. For doing that in cross-section of Yazd-Ardakan plain (including Coarse grain plain, Medium grain plain and fine grain plain) with different pavement density, by random classification sampling method, the soil samples were collected in 37 points from 0-10 cm of surface soil in each parts of plain. The rate of infiltration was measured by using double ring method. Desert pavement cover in to conditions (natural and disturbed) was measured using the exist scales in Munsell’s notebook. Grain size distribution by hydrometry method and organic matter by standard methods were analyzed. Then the spatial pattern of variations by ArcGIS9.3 and Surfer9.0 softwares and zonation by kriging method for K factor and its effective parameters were determined. To evaluate the obtained semivariogram, some statistical indices like MAE(Mean Absolute Error), MBE(Mean Bias Error), MSE(Mean Square Error), RMSE(Root Mean Square Error) and determination coefficient were applied. The results show that in fine grain plains due to decreasing the desert pavement until less than %20, it can not be considerd as the conservation factor and the relation between pavement desert and K factor is not sufficient. On the other hand, in Coarse grain and Medium grain plains (that desert pavement is more than %50), the K factor lies in the low until very low. The conservative role of desert pavement in Coarse grain and Medium grain plains that by combining pavement, the K factor is increased 10 times more than natural conditions. The results of model fitting on spatial changes show that Gaussian model is the best. Totally, ordinary kriging is suitable method for spatial variability description of K and its effective factors.