Modelagem matemática das perdas de água, solo e nutrientes em Latossolo argiloso sob sistemas de uso e manejo

Abstract

Soil erosion due to runoff causes serious damage to agricultural productivity and to the environment. Thus, this work aims to mathematically model water, soil and nutrient losses in a loamy Oxisols with different management systems by means of the physical soil attributes vegetation cover, in order to avoid erosion by runoff. The experimental area was located in the municipality of Santa Tereza do Oeste - PR, under soybean cultivation. The treatments consisted of three systems of management: traditional notillage system, no-till system, no-tillage system with scarification and twelve treatments of no-tillage system with quality with cover species in a consortium or not, called "recuperators" of soil structure: millet, pigeon pea, Crotalaria juncea, pigeon pea, Crotalaria spectibilis, black oats, white oats, black oats, rye, black oat with forage turnip, black oats with white lupine and black oats with forage peas. A compacted range was performed, except for traditional treatment, totaling 29 treatments. The water, sediment and nutrient data were collected at each precipitation in the collector channels installed at each treatment along the soybean cycle. The design was completely randomized. The replicates consisted of each rain collection, totaling 20 successful collections. A descriptive analysis and a Tukey test at 5% of significance were performed to compare the treatments. Regarding the modeling, the Levenberg-Marquardt Method was used to adjust the curve of the losses and it was generated submodels to estimate values of straw and green cover over the time cycle of soybean. In general, the losses of water, soil and nutrients did not differ between treatments and highest losses are related to higher density and lower macroporosity, while the relation between the dry mass of the straw and the green cover was inversely proportional. The use of the function defined by parts was adequate to estimate green coverage values throughout the soybean cycle. This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001.