COLLABORATIVE RESEARCH: Nonlinear Dynamics of Soil Moisture Climate at Continental Scales: Climate Origins of Droughts

点击翻译按钮获取中文摘要

The goal of this collaborative research project will identify and analyze the interactions and feedbacks between the land and atmosphere that result in observed complexities and lack of predictability of hydroclimatology. In particular, it is intended to study the dynamics of soil moisture and its excursions into dry (drought) and wet (flood) states. The first hypothesis is that local recycling of moisture at continental scales, and time periods on the order of seasons to years leads to soil moisture dynamics that can be described by a stochastic differential equation with multiplicative noise. The solution to such nonlinear system is a probability distribution of soil moisture with at least two modes; that is, two preferred states, one dry and one wet. This contrasts with the traditional unimodal view of the soil moisture distribution. This distribution can be studied to obtain answers like the probability of the onset of drought or vice versa the end of a drought period. Another question of interest is the duration of droughts. The multimodal distribution can also be the basis of a new classification of climatic regions. The second hypothesis is that at time scales on the order of the occurrence of individual events (i.e., weeks) the soil moisture has a delayed effect on atmospheric dynamics. This phenomena is due to the moisture supply as well as to dynamic influences resulting from soil moisture effect on surface temperature and atmospheric pressure. The implication of this hypothesis is a highly nonlinear delayed differential equation for the description of soil moisture. The solution of this equation leads to a fixed equilibrium, a limit cycle with bifurcations or to chaos. This will be used to study amplification of seasonal effects over continental regions like the Amazonia as well as the predictability of soil moisture and hydroclimatology in general. Principal Investigators and staff from the University of Arizona and the Massachusetts Institute of Technology will collaborate on this project.