| 英文摘要: | El Nino/Southern Oscillation (ENSO) events are known to have significant impacts on weather and climate worldwide, including reductions in rainfall over much of tropical South America with associated disruptions to water resources, agriculture, and other human and natural systems. The rainfall anomalies are ultimately due to changes in the large-scale atmospheric circulation induced by ENSO conditions in the neighboring equatorial Pacific, but they may also be modulated by land-atmosphere coupling occurring over South America. Here land-atmosphere coupling refers to several mechanisms through which the condition of the land surface influences precipitation, one of which is that soil moisture serves as a source of water vapor through evaporation and transpiration, thereby promoting precipitation. This sort of "precipitation recycling" can prolong and enhance dry spells, as lack of rain dries the soil and reduces evapotranspiration, leading to further reductions in rainfall. On the other hand, a drier land surface can mean greater heating of the land surface during the day as there is less evaporative cooling, and a hotter land surface can lead to instability in the atmospheric boundary layer, which increases the chances of convective precipitation. Land-atmosphere coupling can be quite variable depending on land cover and other factors, and can thus cause the rainfall response to ENSO events to be more spatially variable that would be expected from the large-scale atmospheric circulation anomalies. It can also cause changes in the frequency, intensity, and duration of daily and sub-daily rainfall episodes within the period of a season or more during which an ENSO event takes place.
The goal of this project is to determine the extent to which land-atmosphere coupling accounts for the spatial heterogeneity in the rainfall response to ENSO events over tropical South America. The research consists in large part of statistical analysis of precipitation and atmospheric and land surface data for tropical South America, taken from satellite and surface observations and reanalysis products. Parallel analysis is applied to model simulations from the Coupled Model Intercomparison Project version 5 (CMIP5), including simulations from the subset of models which contributed to the CMIP5 Global Land-Atmosphere Coupling Experiment (GLACE-CMIP5), in which models were integrated using climatological soil moisture so that land-atmosphere coupling could be assessed by comparison between simulations with interactive and fixed soil moisture. The statistical assessment is accompanied by model experiments using the quasi-equilibrium tropical circulation model version 2 (QTCM2), a simplified model which can simulate key aspects of the precipitation response over tropical South America, and in which key factors such as soil moisture, surface sensible heat flux, and the exchange of heat and water vapor between the boundary layer and the overlying free troposphere can be controlled and examined.
Work under this project has important broader impacts in addition to its scientific merit, given the substantial consequences of ENSO-related precipitation disruptions in the region. The results of this study are also expected to shed light on the role of land-atmosphere coupling in other regions of the tropics where similar surface conditions prevail. The work also promotes international collaboration, as it involves unfunded collaborators in two Columbian universities. Aside from the broader impacts of the research, the project also supports undergraduate research assistants through the Research in Science and Engineering (RiSE) program, a 10-week summer program which focuses on students from traditionally underrepresented populations. In addition, the project provides support and training to a graduate student, thereby providing for the next generation of the scientific workforce in this research area. |