عنوان مقاله [English]
Introduction: Soil erosion is the most widespread form of soil degradation jeopardizing food security worldwide. In Iran, gully erosion is important because about 90% of the country has arid and semi-arid climates and rainfall is not adequately distributed. In such conditions, the absence or lack of vegetation cover with increasing runoff causes more than 2 billion tons of soil losses annually. Therefore, conducting basic and applied researches on soil erosion via different methods for comprehensive management of natural resources is emphasized. The prerequisite for all kinds of erosion, such as gully erosion, is the prediction of the risk of gully formation in different areas susceptible to erosion.
Materials and Methods: The study area is located in Ghaleh Gorg watershed which sub-basin of Shahid Modarres that large part of these farmlands has been destroyed by gully erosion. The purpose of this study was to apply fuzzy logic operators for gully erosion zoning. In this research, six effective parameters on soil erosion including mean weight diameter of aggregate, sodium adsorption ratio, salinity, percent of fine sand, silt and clay were determined according to standard laboratory methods. After re-classification, standardization of prepared layers was carried out by the Fuzzy method. Hence, Fuzzy-based layers were integrated using operators of Fuzzy algebraic sum, Fuzzy algebraic production and Fuzzy gamma with 0.2, 0.5, 0.8, 0.9 values and the obtained results were evaluated. The validation of prepared maps was done based on two methods of map matching percentage of areas with very high and high sensitivity with gullies map of the study area and quality sum index (Qs).
Results and Discussion: The Fuzzy results of raster layers showed the relative accumulation of silt, clay and fine sand grains in the middle to western parts of the region due to leaching and soil aggregation of Aghajari formation with a high slope of >20%. The deposition of this sediments was with 0-5% slope in the middle and western part of the basin. Results of layer integration indicated the fact that the fuzzy summation and multiplication method are not suitable approaches for final mapping because of their high increasing and decreasing effects, respectively. About Fuzzy Gamma operator 0.2, the results revealed that about 17.07% of the area was in the high and very high-risk zone and 67.07% of the area was in the low risk zone. In Fuzzy Gamma 0.5, about 31.16% of the area was in high risk and 55.38% in low risk zone. And only 60.38 percent of the gullies was in the high-risk area. Thus, these both operators 0.2 and 0.5 cannot be an acceptable method for preparing the final fuzzy map. The results of gamma operator 0.8 showed that about 43.21% of the area was in high and very high risk classes and 42.45% of the area was in low and very low risk classes. In the gamma operator 0.9, about 60.92% of the area was in high and very high risk zone and 17.1% of the area was in low and very low-risk zone. Also, regarding the distribution range of gullies, 94.93% of gullies was in high and very high risk classes, which is more acceptable and better than gamma 0.8. According to the obtained results, Fuzzy gamma 0.9 with 94.93 matching percentage of areas containing very high and high sensitivity and maximum quality sum index (0.73) among different operators was selected as the best method for preparing Fuzzy map in the study area.
Conclusion: According to the results of this study and its comparison with field observations, effective factors contributing to the initiation and development of gully erosion were sensitivity of the geological formations, soil texture type, salinity and alkalinity and non-implementation of biological and biomechanical operations to the soil and vegetation cover restoration. Around 3855 hectares (60.38%) out of 6327.5 hectares of the total studied area were at high and very high erosion risk. Furthermore, 2056 hectares (94.93%) out of the 2166 hectares of the gullies area were at high and very high risk of erosion, denoting the high accuracy of the final map. Among the available methods, Fuzzy Gamma 0.9 with the highest overlap between the area of high and very high-risk classes of the gully map (94.93%), and the highest qualitative sum index (0.73), was selected as the best fuzzy method for zoning in the study area.