Uso do modelo WEPP (Water Erosion Prediction Project) modificado para estimar taxas de erosão em estradas florestais

• Autor(a) principal: Garcia, Alessandra Reis
• Data de Publicação: 2001
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/608
• topic: Estradas florestais; Erosão; WEPP; Forest roads; Erosion; WEPP; CNPQ::CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTAL::MANEJO FLORESTAL

The general objective of this work was to determine the total volume of the water runoff and the production of sediments from segments of forest roads under normal rainfall conditions, and, with the results, to test the WEPP model through comparisons between data observed and data predicted by the model. The research was carried out in a reforestation project with Pinus caribaea Morelet and Pinus oocarpa Schiede plantations, of the Fazenda Monte Alegre, in Agudos São Paulo State, Brazil. The area covered by the research was of 16 road segments. Two slopes (1 and 7%) and two lengths (20 and 40 m) were studied, totalling four treatments: length of the segment of 20 m and slope of 1%; length of 40 m and slope of 1%; length of 20 m and slope of 7%; and length of 40 m and slope of 7%. Of each treatment, four repetitions were made. For the determination of amount of eroded matter, collector drums were placed in the lowest part of the roads. Later, files of climate, precipitation, soil, slope and segment length were included and adapted to the erosion prediction WEPP model, with the purpose of validating the model, viewing the elaboration of a model suited to the Brazilian forest conditions. The analyses results allowed the following conclusions: a) in the observed volumes of runoff, the effect of road segment length was significant, while the slope effect did not present a significant difference; b) in the observed values of soil weights, the slope effect of the segment was more significant than the length effect; c) in the values of runoff volume and soil weight predicted by the WEPP model, the slope effect of the segment was more significant than the length effect; d) as the precipitation increased, the runoff volume increased in a quadratic way, which was observed in all the treatments; e) as the precipitation increased, a linear increase of soil weight occurred in the segment length of 20 m and slope of 1%. However, in the other treatments, the increase observed in soil weight was quadratic; f) in all the treatments, the observed soil weight increased exponentially in function of the increase in the runoff volume; g) in the validation of the WEPP model, the runoff volumes predicted were, in average, 166.58% greater than the observed data; h) the observed soil weight values in the 1% slope presented a higher variation of 1125.09% in the road segment with length of 20 m, and 724.41% in the segment with 40 m in relation to the values predicted by the WEPP model. This indicates that, in lower slope values, the behaviour of the model was not satisfactory, subestimating losses; i) the soil weight values predicted by the WEPP model in the 7% of slope presented a higher variation of 6.73% in the road segment of 20 m, and 120.25% in the segment of 40m, in relation to the observed values; and j) the Brazilian model for prediction of erosion by water (WEPP-Brasil), is not totally adjusted to these conditions.