Modeling the Brazilian Cerrado land use change highlights the need to account for private property sizes for biodiversity conservation.

Simulating future land use changes can be an important tool to support decision-making, especially in areas that are experiencing rapid anthropogenic pressure, such as the Cerrado-Brazilian savanna. Here we used a spatially-explicit model to identify the main drivers of native vegetation loss in the Cerrado and then extrapolate this loss for 2050 and 2070. We also analyzed the role of property size in complex Brazilian environmental laws in determining different outcomes of these projections. Our results show that distance to rivers, roads, and cities, agricultural potential, permanent and annual crop agriculture, and cattle led to observed/historical loss of vegetation, while protected areas prevented such loss. Assuming full adoption of the current Forest Code, the Cerrado may lose 26.5 million ha (± 11.8 95% C.I.) of native vegetation by 2050 and 30.6 million ha (± 12.8 95% C.I.) by 2070, and this loss shall occur mainly within large properties. In terms of reconciling conservation and agricultural production, we recommend that public policies focus primarily on large farms, such as protecting 30% of the area of properties larger than 2500 ha, which would avoid a loss of more than 4.1 million hectares of native vegetation, corresponding to 13% of the predicted loss by 2070.

CLIMBra - Climate Change Dataset for Brazil.

General Circulation and Earth System Models are the most advanced tools for investigating climate responses to future scenarios of greenhouse gas emissions, playing the role of projecting the climate throughout the century. Nevertheless, climate projections are model-dependent and may show systematic biases, requiring a bias correction for any further application. Here, we provide a dataset based on an ensemble of 19 bias-corrected CMIP6 climate models projections for the Brazilian territory based on the SSP2-4.5 and SSP5-8.5 scenarios. We used the Quantile Delta Mapping approach to bias-correct daily time-series of precipitation, maximum and minimum temperature, solar net radiation, near-surface wind speed, and relative humidity. The bias-corrected dataset is available for both historical (1980-2013) and future (2015-2100) simulations at a 0.25° × 0.25° spatial resolution. Besides the gridded product, we provide area-averaged projections for 735 catchments included in the Catchments Attributes for Brazil (CABra) dataset. The dataset provides important variables commonly used in environmental and hydroclimatological studies, paving the way for the development of high-quality research on climate change impacts in Brazil.

Projected climate change impacts in rainfall erosivity over Brazil.

The impacts of climate change on soil erosion may bring serious economic, social and environmental problems. However, few studies have investigated these impacts on continental scales. Here we assessed the influence of climate change on rainfall erosivity across Brazil. We used observed rainfall data and downscaled climate model output based on Hadley Center Global Environment Model version 2 (HadGEM2-ES) and Model for Interdisciplinary Research On Climate version 5 (MIROC5), forced by Representative Concentration Pathway 4.5 and 8.5, to estimate and map rainfall erosivity and its projected changes across Brazil. We estimated mean values of 10,437 mm ha-1 h-1 year-1 for observed data (1980-2013) and 10,089 MJ mm ha-1 h-1 year-1 and 10,585 MJ mm ha-1 h-1 year-1 for HadGEM2-ES and MIROC5, respectively (1961-2005). Our analysis suggests that the most affected regions, with projected rainfall erosivity increases ranging up to 109% in the period 2007-2040, are northeastern and southern Brazil. Future decreases of as much as -71% in the 2071-2099 period were estimated for the southeastern, central and northwestern parts of the country. Our results provide an overview of rainfall erosivity in Brazil that may be useful for planning soil and water conservation, and for promoting water and food security.