Modelo de estimativa da produtividade de cafeeiros no Cerrado com base em métricas da paisagem

Abstract

In this study we studied the levels of Environmental Fragility (EF) of areas occupied by coffee agriculture, associating the crop productivity indexes. We used as a case study the Feio River basin, in Patrocínio. We mapped the EF of the Feio River basin from the year of 2016 thru 2019 using an adapted from our reserves methos based from Ross (1994). Based on EF and yield observations, we created a model to estimate the yield potential of coffee plants tested in the Feio River basin; we repeated the experiment in a region outside the study area to validate the applicability of the predictive model we created. In this outer region, it compares predicted yield levels with historical EF and historical data and average data for the years of 2016, 2018 and 2020. The model adapted for predicting yield potential showed high correlation in the regression test (mean R² = 0.9). In the nearby pre-installations, we applied the model in an area of Feio Stream, within the nearby areas, they were also included in the productivity in the available data. Thus, the Feio River basin has areas of greater fragility in its central portion, mainly occupied by savannah fields or activities with a high environmental impact, such as degraded pastures and conventional planting systems. The areas of low environmental fragility are mostly occupied by coffee growing; however, there is a tendency to replace it with a no-tillage system (soybean), which is dangerous for conservation and, consequently, for the productivity of coffee plantations. With the results of the association of these levels of fragility with the productive levels of coffee, we show that there is an inversely proportional relationship, where the greater the environmental fragility of an area, the lower its productive potential.The use of decision support metrics in coffee farming enhances the importance of nature conservation, encouraging the use of natural resources in the occupation.