A Dry and Warm Future Projected by Regional Climate Models in the Subtropical Andes Mountains, Southwestern South America
摘要
The population and the ecosystems along the foothills of the Andes Mountains depend heavily on the snowmelt, glacier melt, and precipitation in the high elevations. However, this region has experienced a warming and dying trend, making these sectors particularly vulnerable to climate change. Climate projections based on Regional Climate Models (RCMs) can support decision-making processes related to adaptive measures. This study quantifies the changes in future climate over the portion of the Andes located between 21°S and 47°S in South America. First, we compared temperature and precipitation historical simulations of 25 RCMs using several statistical metrics against the CRU TS 4.07 dataset over three regions, defined according to their geographic and climatic characteristics. Then, we selected the best RCMs to quantify future climate changes relative to a baseline period (1981–2010), forced under the RCP4.5 (5 models) and RCP8.5 (7 models) scenarios. Projections indicate a decline in regional precipitation, ranging from 5% to 30%, depending on the selected scenario and time horizon. Additionally, temperature is projected to increase under all scenarios and time horizons, with the largest magnitude of warming -up to 5 °C- observed under RCP8.5 during the period 2081–2100, particularly in the northern portion of the study area. There is a clear signal of an elevation-dependent warming, specifically in the Central Andes, with a warming rate up to 0.6 °C/km. The combination of drier and warmer conditions is expected to generate profound changes in the cryosphere, posing significant challenges for regional water management and conservation of biodiversity.
Graphical AbstractThe graphical abstract provides a visual summary of the study. First, the graphic represents the reliance of the population and ecosystems surrounding the study area, the Andes Mountains in Argentina, on water resources. Then, it shows the data and methodology for the evaluation and selection of the precipitation and temperature historical simulations. We evaluated 25 historical simulations belonging to Regional Climate Models against CRU TS 4.07 data. We quantified several statistical metrics such as MAE, IVS, PCC, and the magnitude and sign of the trends. Additionally, we calculated Taylor diagrams and obtained a model ranking based on the selected metrics to identify the best-performing simulations for the study area. Then, we developed a multi-model ensemble based on the best models to represent the climatology of the region and quantify the projected changes under two scenarios in the near-future, mid-future, and far-future. Results of the future projections indicate that mean precipitation is expected to decline between 5% and 30%, depending on the region, scenario, and time horizon. The temperature is projected to increase across all the regions, with values above 5°C particularly in the far-future. The warming signal shows a dependence on elevation, being higher under RCP8.5 by the end of the century, up to 0.6°C/km. Additionally, it is expected an increase in aridity and higher evaporation across all regions, particularly under RCP8.5 by the end of the century.