Introduction: Soil erosion is the most prevailing form of soil degradation which is really play an important role on the mass balance index of soil particles in the watersheds. Moreover, regarding the on-site and off-site effects of erosion essentially has to measure and predict the soil loss using different methods. Specifically gully erosion is a form of water erosion with the huge amount of soil dislodgement. Due to the complexity and variability of soil erosion it is necessary to apply different techniques in order to monitor the soil erosion changes. Remote sensing technology and the use of spectrometry and reflectometry basics is a suitable solution and option for monitoring coastal areas affected by erosion and deposition events, which provide high quality temporal and spatial data. Soil color is an appearance property which is meaningfully effective on soil reflectance. Generally, soils with high amount of organic matter has low reflectance because of the darkness while the light soils has high reflectance from surface (high Brightness index, BI) which is effective on soil temperature. Therefore we try to use RS and radio spectrophotometry to find a relation between soil color and its reflectance.

Materials and Methods: The study area is located at Zahirieh watershed of Khuzestan province which is between Ahvaz and Masjedsoleyman cities with approximately 7100ha area. The average of rainfall is 218.6mm, the maximum temperature is 54 and the minimum is 7 degrees. Regarding the separation of erosional and depositional surfaces in the study area; first, using the visual inspection of Landsat satellite false color images, 8 regions were divided into several regions, then random sampling points were created using the random point generator tool in ArcGIS 10.4 software to implement the random sampling method within the block. Finally, 12 sampling points representing erosion surfaces and 14 sampling points representing depositional surfaces were selected and sampled to determine surface soil characteristics. Surface soil color was determined using Mansell's color book in natural daylight in two dry and moist conditions. After collecting the soil samples in air-dry moisture condition and also in wet condition, their spectroscopic analysis was done by FieldSpec3 device and this moisture condition was considered for all the soil samples of eroded and depositional surfaces. Statistical analyzes and mean comparisons were performed using SPSS 26 statistical software. Corrections of satellite images and transformations were made in ENVI 4.7 software, and visual outputs and maps were made in ArcGIS 10.4 software.

Results and Discussion: Results depicted that among the evaluated soil color indicators, the dry weight parameter is significant at the level of 1%. This level of significance shows well that the Value index in the dry state can be used as an effective parameter to identify and separate erosion and deposition levels in the study area. There is a difference between the values of the statistics for red, green and blue RGB in the dry state for erosion and depositional surfaces, and these differences are also evident for the moist state. In the depositional surfaces, with the drying of the soil, blue, red, and green reflections all decrease, but this decrease is double and about six times for blue. The reduction of blue reflections in the RGB system leads to an increase in the yellowness of the color. In the case of the soils of erosion surfaces, we can see the pattern of the photo and we see the enhancement of reflections and consequently the lightening of the color of the soil when the soil is dry. According to what has been seen in the Munsell system, it seems that this issue has a direct relationship with the amount of organic matter and the ratio of fulvic acid to humic acid in the organic matter of the soil. Moreover, the results of the comparison of the average bands of Landsat 8 shows that bands 2, 3 and 4 are able to separate erosion and sedimentary surfaces at the 1% level, but thermal bands cannot be used to separate surfaces. Due to the difference in the color characteristics of erosional and sedimentary surfaces, as a result, it is possible to separate them based on reflectometric characteristics, and it is possible to separate eroded and sedimentary surfaces by using color indices.

Conclusion: Due to the difference in the color characteristics of erosional and depositional surfaces, as a result, it is possible to separate them based on reflectometric characteristics, and it is possible to separate eroded and depositional surfaces by using color indices. The results showed that it was possible to model surface soil characteristics using quantified surface soil color data, and this hypothesis was confirmed by statistical investigations.