Classification And Determination Of Erodibility Indises Of Hamedan Province Marls

Because of high sensitivity to erosion of marls, It is sent thousands tons of fine grains sediments and clay to reservoirs of dams and artificial feeding stations annually. This process increase the expenses of filteration of drinkable water and forward settlement of artificial feeding water strongly, Therefore, the study of different dimensions of marls is important for specialists and researchers. In this project it was tried that effective variables of erosion rate, erosion features and classification of marls were recognized and some models were produced for classification of marls and erosion tye of them. By this reasoning we prepared the lithological, slope and climatic maps of area and then with putting together them, the work units map was made. Then, 49 samples of marly soils that belonged to 41 erosional points were taken and for determinig of amount of physico-chemical variables were sent to laboratory. Also, characteristics of sampling place such as slope, Geographical situation, privilege of 7 parameter for soil surface factor and so on were noted. Also, the test of artificial rain was done by kampHorst´s rainfall simulator apparatus in the 41 mentioned points and 41 samples were taken and the amount of runoff and sediment of them were measured by simple method in the laboratory. Furthermore, the 7 samples of marls were analyzed by X-ray diffractometer and the 8 thin setion of marls and their accompaniment rocks such limestones and sandstones provided and studied by polarizing microscope.Then the abnormal data were normalized by logarithm and afterwards the all data were evaluated by correlation analysis, Bivariate and multivariate regression analysis, cluster analysis, discriminant analysis and scalogram method. The result of this research show that the marls esentialy consist of Calcite, Quartz, Albite, Illite, Mica, Clinochlore and expandible minerals, and have Oligo-Miocene age. The marls of area were assigned four groups, five groups and six groups ordely by chemical, physical and physico-chemical variables, The provided fisher´s linear discriminant functions from these classification are enable to predict the marls groups for cases with a high sensitive that is orderly equivalent 97.6%, 87.8% and 97.6%. The number of chemical variables that is effective for mentioned classification at above equal three variables and number of effective physical or physico-chemical variables is six. The chemical variables are more usefull for classification of marls, because they have high sensibility and more precision for prediction, Furthermore the number of them is fewer. Evaluation of relationship between measured variables and erosion features was shown that pH is only variable that amount of it has a significant difference and effective for recognition of erosion features and Sensibility of fisher´s linear discruminant functions for prediction of erosion features by help of all chemical, physical and physico- chemical variables are orderly equal with 75.6%, 82.9% and 78% that show the physical variables are more usefull than chemical and physico-chemical variables for prediction of erosiolal features. Evaluation of relationship between measured variables with soil lost were shown that the soil lost has a significant relation with only 9 variables that the plastic index, clay percent and clay ratio are orderly the best three of them. The result of this project showed that marl type has a significant relation with amount of erosion, but the soil lost and the marl type have not significant relation with erosion features. In this research, several functions have been produced for prediction of erosion features, marl type and the amount of soil lost.