بررسی و ارزیابی میزان هدررفت خاک در کاربری‌های اراضی حوضه آبخیز آتشگاه با استفاده از مدل RUSLE و تصاویر ماهواره‌ای لندست (سنجنده OLI)

Soil erosion is a natural process (Lee et al., 2021). which causes the level of soil loss by various environmental factors such as weather, soil, topography and vegetation (Chen et al., 2019). However, human interventions through land use change and agricultural and construction activities can accelerate this flow (Wenker et al., 2019; Barley et al., 2017). For this reason, nowadays, soil erosion caused by land use change has become the most important issue of land degradation all over the world, and the transformation of the land form and the disruption of the main functions of the natural environment are the consequences of these geomorphic reactions (Paul et al., 2019). T (2017) aimed to study and estimate the spatial and temporal soil erosion in the periods of 1994-1999-2008-2015 in the Manderjan sub-basin located in the west of Isfahan province. Using remote sensing and GIS technologies, they concluded that the amount of soil erosion in 1994-1999-2008-2015 was 0.001 to 233, 0.001 to 297, 0.001 to 231, 0.001 respectively. It is up to 215 tons per hectare per year. Also, the height and height factor in the region with a correlation coefficient of 80% has the greatest effect in B The annual soil erosion rate was estimated by the RUSLE model. Nejad Afzali et al. (2018) used the Revised Global Model of Soil Erosion (RUSLE) to estimate soil erosion in Dehkhan watershed south of Kerman. Their results showed that the annual soil erosion in the study area is estimated at 50 tons per hectare per year. Khosravi-Aghadam et al. (2018), in order to estimate the soil erodibility factor and its relationship with some land characteristics, using the USLE model in a part of the Nazlu Chai watershed of Urmia. Their results showed that the value of K factor varies in the range of 0.079 to 0.029 tons per hour per megajoule mm. Also, in terms of erodibility, the soils of the region are in low and very low erodibility classes.

Methodology
Ateshgah watershed is located in the southwest of Ardabil city at the position of 47°50' to 48°2' east longitude and from 38°12' to 38°16' north latitude. The main branches of this basin originate from Sablan heights in the west of the basin. The area of this basin is 40.5 square kilometers and the maximum height of this basin is about 3596 meters at the extreme end of the western part of the basin and its minimum height is 1798 meters at the outlet of the basin in the eastern part. The location of Atashgah watershed is shown in Figure 1.

Research data and tools
The current research is of an applied type and its research method is an analysis based on the integration of data analysis, geographic information system, remote sensing and the use of the revised global model of soil erosion (RUSLE). The data and tools used in the research include 1:25000 digital layers of the National Mapping Organization, digital elevation model (DEM), with a spatial resolution of 30 meters, rainfall data from the National Meteorological Organization, Landsat OLI 8 satellite image for 2020 with a spatial resolution of 30 meters, the studied area from the website www.usgs.gov, the collection of educational samples was also done through field visits and the creation of false color combinations, and the soil laboratory data of the studied basin is from the watershed deputy of the country's organization of forests, pastures and watershed. In this research, ArcGIS 10.3 software was used to draw maps and analyzes related to it, as well as ENVI 5.3 software to prepare vegetation and land use layers of the study area, and statistical software such as Excel 2016 and SPSS 17 for statistical calculations and The regression relations of the equations have been used.
Landsat satellite images include the longest archive of global images with moderate resolution, multispectral data from unique sources for functional planning at various scales, including land use and land cover, change detection and monitoring of natural environment dynamics (Taherparour et al. , 2015). Therefore, Landsat OLI 8 satellite images were used in this research. The specifications of the satellite image used in this research are presented in Table
After classifying the satellite image using the Support Vector Machine (SVM) method, the obtained land use map was separated into seven land use classes, including good, poor pastures, irrigated agriculture, rainfed, residential and irrigated areas (Figure 2). The accuracy of the obtained map for 2020 was checked using the Google Earth image and ground control points, as well as the false color image of the same year. In this research, the overall accuracy for the land use map of Atashgah basin was 0.90 and the Kappa coefficient was 0.87.
3-2- Soil erosion (RUSLE)
According to (Figure A), the results of the rain erosive factor (R) vary from 74 to 34.98 MJ/mm/hectare/hour per year, the highest value of which is related to the north and southeast parts and the lowest value is related to the central parts. and southwest.
The average amount of soil erodibility factor (K), according to figure (b), varies between 0.12 and 0.37 tons/hectare per year in the study area. According to figure (c), the value of LS factor in the studied area varies between 0 and 3.98, which is higher in steep slopes, especially around waterways. Using the Normalized Vegetation Index (NDVI), the vegetation factor (C) of the Atashgah basin was prepared based on equation 4 and 5, which is presented in (Figure d and e). Based on this figure, the value of the C factor varies between -0.16 and 0.74. In general, it can be said that the eastern and central half of the basin has lower amounts of C due to the presence of dry and unused lands, and the southern and western parts of the basin have the highest amounts due to the presence of pasture lands. The soil protection operation factor (P) was also considered to be 1 due to the lack of available information from the region for the entire region.
Annual soil erosion (RUSLE), to prepare the average annual soil erosion map of Atashgah watershed through the product of rain erosion factors (R), soil erodibility (K), topography (LS), vegetation cover (C) and soil protection operations (P) It was calculated in the Raster Calculator plugin in the ArcGIS 10.3 environment using (Relation 1). The annual soil erosion values in the studied basin vary between 0.09 and 11.02 tons per hectare per year. Also, the average amount of soil erosion in the studied area is 0.16 tons per hectare per year and its standard deviation is 0.55 tons per hectare per year. In (Figure 3), the average annual soil erosion (RUSLE) map of Atashgah basin is presented.
Conclusion
Land use is one of the important factors in causing soil erosion, and in recent years, the mutual impact of land use change and soil erosion has become a major environmental concern. Considering the importance of the topic, in the current research, the amount of soil erosion on land use in the Atashgah watershed has been investigated using the RUSLE model. For this purpose, first, the land use map was classified using the Landsat OLI 8 satellite image and using the support vector machine algorithm into seven land use classes, including barren land, good pastures, poor pastures, rainfed agriculture, water, residential areas, and water. The overall accuracy and Kappa coefficient for the prepared land use map were obtained as 0.90 and 0.87%, respectively. Then the maps of R, K, LS, C and P factors of the RUSLE model were prepared in the GIS environment and after combining these layers through the Raster Calculator in the Arcmap environment, the average annual soil erosion map for the entire Atashgah watershed between 0.09 and 11.02 tons It was calculated per hectare per year. The results of the evaluation of the soil loss map on the land uses of the studied area showed that dry land use with an average soil erosion of 0.48 tons per hectare per year has the highest soil loss and good pastures with an average erosion of 0.21 tons per hectare per year has the lowest. They have soil waste among other uses in the region. In this research, it was tried to use GIS capabilities to create the required data of RUSLE model. Finally, it is suggested to control the process of land use changes in the Atashgah watershed by determining grazing capacity, vegetation management, and take steps to restore, improve and develop pastures. Therefore, it is expected that this study and the results of this research will pave the way for the implementation of better and more scientific management by competent managers and planners in this field.

Keywords: "soil erodibility", "land use", "revised global soil erosion equation", "geographical information system", "Atashgah".