The impact of global warming on the aridity in South America (SA) is investigated. For this purpose, the methodology for generating surrogate climate-change scenarios with a RCM is employed. For the present climate (C...The impact of global warming on the aridity in South America (SA) is investigated. For this purpose, the methodology for generating surrogate climate-change scenarios with a RCM is employed. For the present climate (CTRL) the RCM is initialized with and driven by ECMWF/ERA-Interim reanalysis data. Two aridity indices are used: the Budyko and the UNEP indices. The results for the CTR are in agreement with other model studies which indicate future warming;rainfall increases in southeastern South America, Ecuador and Peru and decreases in the central and eastern Amazon. In general the model reproduces the aridity in the continent compared with the observed data for both indices. The distribution of aridity over SA in surrogate climate-change scenario shows an increase of the dryness in the continent. Over Amazonia the aridity increases 23.9% (for the UNEP index) and 3.1% (for the Budyko index), suggesting that portions of the Amazonia forest are replaced by dry land area. The semi-arid zone over northeast Brazil expands westward, attaining the interior of north Brazil. In this region the aridity increases 20% (for the UNEP index) and 0.6% (for the Budyko index) indicating that areas of humid regime may be occupied by areas with dry land regime. The RCM was also integrated driven by the AOGCM ECHAM5/MPI-OM for the reference climate (CTRL2) and under A1B SRES scenario. The results for the present-day climate are similar in CTRL2 and CTRL, and are in agreement with CRU data. The distribution of the aridity for the present climate seems to be better represented in CTRL using both Budyko and UNEP indices. The changes in aridity (future climate minus control) are higher in the run forced by the A1B SRES scenario. Although the UNEP and Budyko indices show potentialities and limitations to represent the aridity distribution over SA, the changes in aridity due to a pseudo-scenario of global warming are higher using the UNEP index.展开更多
文摘The impact of global warming on the aridity in South America (SA) is investigated. For this purpose, the methodology for generating surrogate climate-change scenarios with a RCM is employed. For the present climate (CTRL) the RCM is initialized with and driven by ECMWF/ERA-Interim reanalysis data. Two aridity indices are used: the Budyko and the UNEP indices. The results for the CTR are in agreement with other model studies which indicate future warming;rainfall increases in southeastern South America, Ecuador and Peru and decreases in the central and eastern Amazon. In general the model reproduces the aridity in the continent compared with the observed data for both indices. The distribution of aridity over SA in surrogate climate-change scenario shows an increase of the dryness in the continent. Over Amazonia the aridity increases 23.9% (for the UNEP index) and 3.1% (for the Budyko index), suggesting that portions of the Amazonia forest are replaced by dry land area. The semi-arid zone over northeast Brazil expands westward, attaining the interior of north Brazil. In this region the aridity increases 20% (for the UNEP index) and 0.6% (for the Budyko index) indicating that areas of humid regime may be occupied by areas with dry land regime. The RCM was also integrated driven by the AOGCM ECHAM5/MPI-OM for the reference climate (CTRL2) and under A1B SRES scenario. The results for the present-day climate are similar in CTRL2 and CTRL, and are in agreement with CRU data. The distribution of the aridity for the present climate seems to be better represented in CTRL using both Budyko and UNEP indices. The changes in aridity (future climate minus control) are higher in the run forced by the A1B SRES scenario. Although the UNEP and Budyko indices show potentialities and limitations to represent the aridity distribution over SA, the changes in aridity due to a pseudo-scenario of global warming are higher using the UNEP index.
