Evaluating the temporal patterns of land use and precipitation under desertification in the semi-arid region of Brazil

The desertification process directly influences agricultural production and the maintenance of terrestrial life. Despite being an important agricultural centre, Brazil's semi-arid region is experiencing significant desertification. However, the relationship between land management and climate in this process is still poorly understood. This research was conducted in areas predominantly covered by the Caatinga Formation (62.2%), as well as agropastoral and pasture areas (29.9%). Our primary objective was to assess the land use change trajectories and land occupation patterns in the region while also analysing how vegetation and precipitation temporal patterns have changed from 1985 to 2019. The trends in vegetation and precipitation were evaluated using the Mann-Kendall test. Land use and occupation patterns were analysed using segmented regression and Monte Carlo simulation for future forecasting in the period 2020–240 under different scenarios (pessimistic, conservative, and optimistic). The Mann-Kendall test indicated that the precipitation trends were not statistically significant. However, the normalized vegetation index (NDVI) exhibited a significant trend in areas near the São Francisco River, an important agricultural area. Land use and occupation showed breakpoints from 1990 onwards for most classes. Monte Carlo simulations demonstrated that the Caatinga Formation will likely to remain the predominant vegetation class in the region for the next 21 years in all predicted scenarios. Agropastoral and pasture classes are projected to increase by 8% and 3%, respectively, while the area covered by native vegetation is expected to decrease by 7.5%. Furthermore, the predictions showed a decline in the rivers and lakes class, which may result in significant economic and social problems. Our findings contribute to a better understanding of the impacts of land use change in the Brazilian semi-arid region and provide science-based information for environmental public policies.