Analysis of forest cover changes and trends in the Brazilian semiarid region between 2000 and 2018

Abstract

In this study, the Google Earth Engine (GEE) platform was used to obtain and manipulate data products derived from satellite images to quantify the forest cover changes in the Brazilian semiarid region (BSAR) between 2000 and 2018 and analyze the annual loss in this period. The relationships among the cumulative gain, loss, and gain–loss between 2000 and 2012 were also analyzed. The Mann–Kendall (MK) and Sen’s nonparametric methods were applied to identify trends, and two-way analysis of variance (ANOVA) was used to identify possible similarities among the patterns of forest cover loss in two nine-year periods (2001–2009 and 2010–2018). In 2000, this semiarid region had a forest cover of 225,412.80 km2, and in 2018, a total loss of 60,769.39 km2 occurred, i.e., a total of approximately 26.96% in 18 years. The largest area loss was observed in 2016, and the smallest was observed in 2002. The MK and Sen’s tests showed an increasing trend in the annual losses in this semiarid region at the 0.004 significance level. The results can aid efforts to identify the areas most affected by forest cover loss and determine the corresponding trends, which could contribute to implementing public policies to raise the awareness level of the local population regarding the preservation of the BSAR and mitigation of deforestation.