Evaluation of soil erodibility index using the 137Cs fallout redistribution analysis

Author(s)

Vladia Correchel

• Document type: Doctoral Thesis
• Press: Piracicaba.
• Institution: Universidade de São Paulo (USP). Centro de Energia Nuclear na Agricultura (CENA/STB)
• Defense date: 2004-02-13

Examining board members

Osny Oliveira Santos Bacchi; Klaus Reichardt; Gerd Sparovek

• Advisor: Osny Oliveira Santos Bacchi; Isabella Clerici de Maria

Abstract

Among the Universal Soil Loss Equation (USLE) parameters, the soil erodibility factor (K), defined as the rate of soil loss per unit of area and per unit of erosivity index for a specific soil as measured on unit plot (Wischmeier & Smith, 1978), is the only soil intrinsic factor. The K values determined on standard unit plots are considered as the real K values. However, K factor can be also estimated by empirical models based on physical, chemical and mineralogical soil attributes. Eight of such empirical methods for K factor estimation were evaluated by comparing the rates of soil loss calculated by USLE and 137Cs fallout redistribution analysis. This procedure was applied both in eight standard erosion plots and six field transects on different soils. In the standard erosion plots the measured soil erosion rates were compared with the rates estimated by the 137Cs method and the results indicate a good relationship. The K values estimated by the different models for the specific soil of the standard plots were compared with the measured K value. The results obtained by the models proposed by Wischmeier et al. (1971) and Denardin (1990) for Brazilian soils were more close to the measured K value. The same K values were also evaluated by comparing the soil erosion rates estimated by USLE and 137Cs methods and the results are in agreement with the direct comparison. This procedure was also applied for the evaluation of K estimation models on the field transects. Due to the limitations associated to both models, the soil erosion rates estimated by USLE and 137Cs methods were very discrepant which did not allow the comparison of K estimation methods when applied for the soils of the transects. (AU)