Assessing soil fertility index based on remote sensing and gis techniques with field validation in a semiarid agricultural ecosystem

Among the greatest challenges of the arid and semiarid regions is the need for more crop production to meet the increasing demand of the growing population. This study aimed to compare SFI classes with both yield values and vegetation index values derived from satellite images. A total of 281 soil samples were taken at a 1-km resolution in order to quantify the spatial dynamics of soil physical, chemical and fertility indicators. Of the study area, 40.0% had very high fertile and high fertile soils, while 26.7% of the area had moderately fertile soils. Only about one-third of the total area had low and very low fertility. These results were validated using a 3-year yield values belong to parcels, and vegetation index derived from Sentinel 2A images. A strong relationship of SFI with yield (r2 = 0.88) and RE-OSAVI (r2 = 0.83) was found. Therefore, we suggested that SFI can be used to determine the sufficiency potential of soils for plant growing and management according to sustainable principles in similar ecologies provided that similar sample size should used.

Automated summary

This study demonstrated that SFI classes derived from satellite images can effectively assess soil fertility in arid and semiarid regions, showing a strong correlation with crop yield and vegetation index values. Therefore, SFI can be a valuable tool for evaluating the potential of soils for plant growing and management in similar ecologies.