Federal University of Pelotas, Water Resources Graduate Program, Campus Porto, Rua Gomes Carneiro, 1, 96010-610 Pelotas, RS, Brazil.
Federal University of Lavras, Water Resources Department, CP 3037, 37200-900 Lavras, MG, Brazil; Federal University of Pelotas, Water Resources Graduate Program, Campus Porto, Rua Gomes Carneiro, 1, 96010-610 Pelotas, RS, Brazil.
Sci Total Environ. 2020 Jul 1;724:138315. doi: 10.1016/j.scitotenv.2020.138315. Epub 2020 Mar 31.
Rainfall erosivity is the driving factor for soil erosion and can be potentially affected by climate change, impacting agriculture and the environment. In this study, we sought to project the impact of climate change on the long-term average annual rainfall erosivity (R-factor) and mean annual precipitation in South America. The CanESM2, HadGEM2-ES, and MIROC5 global circulation models (GCMs) and the average of the GCMs (GCM-Ensemble) downscaled by the Eta/CPTEC model at a spatial resolution of 20 km in the representative concentration pathway (RCP) 8.5 were applied in this study. A geographical model to estimate the R-factor across South America was fitted. This model was based on latitude, longitude, altitude, and mean annual precipitation as inputs obtained from the WorldClim database. Using this model, the first R-factor map for South America was developed (for the baseline period: 1961-2005). The GCMs projected mean annual precipitation for three 30-year time periods (time slices: 2010-2040; 2041-2070; 2071-2099). These projections were used to run the R-factor model to assess the impact of climate change. It was observed that the changes were more pronounced in the Amazon Forest region (namely, the North Region, NR, and the Andes North Region, ANR) with a strong reduction in the mean annual precipitation and R-factor throughout the century. The highest increase in the R-factor was projected on the Central and South Andes regions (CAR and SAR) because of the increase in the mean annual precipitation projected by the GCMs. The GCMs pointed contradictory projections for the Central-South Region (CSR), indicating greater uncertainty. An increase in the R-factor was projected for this region, eastern Argentina, and southern Brazil, whereas a decrease in the R-factor was expected for southeastern Brazil. In general, the GCMs projected reductions in the R-factor and annual precipitation for South America, with the highest changes projected from the baseline to the 2010-2040 time slice.
降雨侵蚀力是土壤侵蚀的驱动因素，可能受到气候变化的影响，从而对农业和环境造成影响。本研究旨在预测气候变化对南美洲长期平均年降雨侵蚀力（R 因子）和年平均降水量的影响。本研究使用了 CanESM2、HadGEM2-ES 和 MIROC5 全球环流模型（GCM）以及 Eta/CPTEC 模型对 GCM 进行降尺度处理（空间分辨率为 20km），得到在代表性浓度路径（RCP）8.5 下的 GCM 平均集合（GCM-Ensemble）。本研究还应用了一种地理模型来估算整个南美洲的 R 因子。该模型的输入包括纬度、经度、海拔和年平均降水量，这些数据均来自世界气候数据库。使用该模型，我们首次绘制了南美洲的 R 因子地图（基准期：1961-2005 年）。GCM 预测了三个 30 年时间段的年平均降水量（时间切片：2010-2040 年、2041-2070 年、2071-2099 年）。这些预测结果被用于运行 R 因子模型，以评估气候变化的影响。结果表明，在亚马逊森林地区（即北区域 NR 和安第斯山脉北部区域 ANR），气候变化的影响更为明显，整个世纪的年平均降水量和 R 因子都明显减少。由于 GCM 预测的年平均降水量增加，中央和南部安第斯山脉地区（CAR 和 SAR）的 R 因子预计会增加。GCM 对中南部地区（CSR）的预测结果存在矛盾，表明该地区存在更大的不确定性。预计该地区、阿根廷东部和巴西南部的 R 因子将会增加，而巴西南部东南部的 R 因子将会减少。总体而言，GCM 预测南美洲的 R 因子和年平均降水量将会减少，其中从基准期到 2010-2040 年时间段的变化最大。
