Investigation of Soil Salinity Changes in Torbat Heydariyeh of Khorasan Razavi Province in Relation to Topography and Vegetation using MODIS Satellite Remote Sensing Data

Abstract
Introduction: Soil salinity is a major cause of desertification. It can be caused by several factors including and not limited to improper irrigation, low irrigation water quality, saline groundwaters, saline geological formations, and improper land management. Since direct measurement of soil salinity is costly and time-intensive, the main purpose of this research is verifying whether satellite remote sensing data can be used as a proxy to estimate soil salinity.
Materials and Methods: In the present study, soil salinity changes in Torbat Heydariyeh city of Khorasan Razavi province were investigated using MODIS satellite data in the period of 2005, 2010, 2015 and 2020. We used a combination of MODIS visual images band 4 and 6 to produce the salinity index. In order to evaluate the alternative indices and the effect of salinity on other environmental variables, the correlation between salinity, altitude, slope and EVI (Enhanced Vegetation Index) was also examined. To prepare the ground truth map in 2020, a field sampling was conducted in late June of 2020 from the top 0-5 cm of soil surface in a random sampling design from 80 points. Soil salinity was determined in saturated extract. The obtained salinity values were compared with the equivalent values in the 2020 salinity map based on the kappa index to determine the accuracy of the calculations.
Results and Discussion: Although the salinity level in the region is not very high, but the trend of change indicates an increase in salinity in the region, mainly in the southern and central parts and parts of the north of the region. In terms of topography, most of the area is flat with minor elevations in the middle and northern section. Soil salinity is believed to be correlated with the topography of a region. Salts tend to accumulate more in the lowlands and highlands become saline only if their parent formation is saline. To investigate the effect of this factor on soil salinity distribution in Torbat-e Heydarieh, an accurate digital altitude map of 12.5 meters resolution from the ALOS satellite was used. The correlation between salinity, altitude and slope was evaluated at different times. However, this relationship was not significant in any of the studied periods. But the interesting point was the higher correlation between salinity and altitude, which seems reasonable considering the distribution of evaporitic and Marine formations in the plain areas. We did not obtain significant changes in vegetation in the region between 2005 and 2020. If one would divide the EVI index results into three cover classes, 60% of the area would be classified as without coverage or with very poor cover (2235 square kilometers), 20% as medium (745 square kilometers) and high (745 square kilometers), respectively. Most of the vegetation of the region is scattered in the northern plains on the alluvial sediments. Accordingly, ​​approximately 60% (2235 square kilometers) of the total areas fell in the class of no salinity (0-0.17), 35% (1304 square kilometers) in light salinity (0.17-1.2) and only 5% of the area (186 square kilometers) in the extreme class (more than 0.2). To confirm the validity of the salinity index, the salinity measured in the field was compared with the corresponding values ​​from the produced map and the kappa coefficient was equal to 0.65%, indicating the validity of the measurements performed. However, the correlation between different parameters and salinity showed no correlation with slope, altitude and EVI, which is conceivable due to the scattered vegetation of the area. Interestingly, the values ​​of height and slope correlations were higher than vegetation. It seems that the severe lack of vegetation and the accumulation of evaporative formations in a few spots have led to a lack of correlation between these factors.
Conclusion: The results of soil salinity estimation in Torbat Heydarieh showed that remote sensing data is a viable alternative to direct soil salinity measurements. However, soil salinity did not correlate with other environmental factors including vegetation cover, altitude and slope gradient. Therefore, these factors cannot help us as proxies of soil salinity in these arid areas. According to the accuracy of the salinity index, it can be used to monitor salinity in the region. These results can help land managers to deal with salinity and ultimately desertification.