ENVIRONMENTAL MONITORING AND ASSESSMENT, cilt.194, sa.6, 2022 (SCI-Expanded)
In modeling studies, the use of spatial data derived from geographic information systems and remote sensing applications to simulate the impact of phenological and seasonal changes on soil loss has a promising effect on the accuracy of predictions. The objective of this work was to estimate the C-factor (cover management) as a dynamic-factor RUSLE (revised universal soil loss equation) model based on an NDVI (Normalized Difference Vegetation Index) approach derived from high-resolution Landsat 8 and Landsat ETM7 satellite images for 140 different rain-fed wheat parcels in terms of seasonal and phenological-based by the integrated use of remote sensing and GIS. Overall, it was found that the highest C values, an average of 0.70, were estimated for the emergence period of the wheat, while the lowest value of 0.06 was found in the booting period. Seasonally, the estimated average C values in these parcels were 0.69, 0.63, 0.13, and 0.44 for the autumn, winter, spring, and summer, respectively. Corresponding soil losses for those seasons were 1.70, 1.55, 0.28, and 1.13 t ha(-1) year(-1 )respectively. Comparatively, without considering the phenological growing periods of wheat, the annual predicted soil loss rate was 11.5% higher than the conditions considered. The present study concluded that an assessment of seasonal and phenological changes in the C-factor for fragile ecosystems with weak crop-cover development could significantly improve the accuracy of the RUSLE model predictions and effectively manage limited soil and water resources.