Abstract. In the present study an integrated neural network based model named ImpelERO was applied to estimate the impact of conventional cropping system on erosion indices, including soil vulnerability index, erosion risk class and soil loss rates by potato, alfalfa and maize plantations. Our results revealed that the soil vulnerability indices ranged from 0.21 to 0.52, 0.15 to 0.41 and 0.2 to 0.5 by potato, alfalfa and maize cultivations, respectively. The values of erosion risk classes by potato and maize cultivation ranged from V1 to V4 in which categorize the region as non-sensitive to very sensitive to erosion and by alfalfa cultivation varied between V1 to V3 which classify the study area as non-sensitive to sensitive to erosion The values of soil losses varied between 7.1 to 59 t ha -1 yr -1 with an average of 15.11 t ha -1 yr -1 by potato, 4.9 to 32.4 t ha -1 yr -1 with an average of 8.42 t ha -1 yr -1 by alfalfa and 6.8 to 52.7 t ha -1 yr -1 with an average of 13.65 t ha -1 yr -1 by maize cultivation. It was concluded that planting perennial alfalfa compared to row planted potato and maize has a great effect on controlling soil erosion and its indices at the study area.
Soil erosion is a severe problem worldwide, and controlling it will be a major challenge in the future. Neural networks, as an artificial intelligence technology, have grown rapidly over the past few years and have an ability to deal with nonlinear multivariate systems. An integrated Model to Predict European Land use named ImpelERO is a decision trees/neural network hybrid model. The overall approach of ImpelERO was applied in 49 selected sites from Mashhad-Chenaran plain to quantify the soil erosion features including soil vulnerability index, soil loss rate, erosion risk class and soil depth reduction by sugar beet cultivation under conventional and conservational management practices. Our results revealed that the soil vulnerability indexes ranged from 0.12 to 0.54 and 0.1 to 0.44 by conventional and conservational practices, respectively. The values of soil losses in the study area varied between 4 to 59.7 t ha-1 year and 3.4 to 38.7 t ha-1 year with an average of 11.72 and 7.83 t ha-1 year by conventional and conservation management practices, respectively. The mean values of erosion risk classes ranged from V3 in conventional to V2 in conservational practices which categorize the region as accepted tolerable to sensitive to erosion. The long term soil productivity reduction for time horizons 2020, 2050 and 2100 revealed that the conservational practices have greater contribution on preventing productivity reduction than conventional management practices.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.