Statistical and Dynamical Exploration of Land-Ocean-Atmosphere Interactions in Two Contrasting Monsoon Regions: China and Northern Australia

University Of Wisconsin-Madison, Madison WI

Investigators

Abstract

Monsoon systems cover much of the tropics and subtropics, and monsoon rains provide the agricultural and domestic water resource for much of the world's population. Thus, variations of the monsoons can have dramatic human consequences, and a better understanding of the processes which cause monsoon variability is desirable. Research to date has shown that some portion of monsoon variability can be linked to anomalies in remote sea surface temperatures (SSTs) such as those associated with El Nino events, which cause large-scale changes in atmospheric circulation that modulate rainfall in the monsoon regions. A second potential cause of monsoon variability is the condition of the local land surface, which is affected by rainfall and can potentially have a further influence on rainfall through a variety of land-atmosphere feedback processes. This project seeks to identify the separate effects of remote SSTs and local land-atmosphere interactions in modulating rainfall and atmospheric conditions over monsoon regions. The research is focused on two regions chosen to represent subtropical and tropical monsoon systems, China and Northern Australia respectively. Previous work by the Principal Investigator (PI) suggests that modification of the land surface by a reduction in vegetation cover produces opposing results in the two regions, with an earlier summer monsoon onset in China (linked to greater ocean-land thermal contrast) and enhanced springtime rainfall, and a delayed and weaker monsoon in Northern Australia (linked to reduction in the source of moisture for precipitation due to transpiration from plants). The research addresses the relative roles of remote SSTs and local land-atmosphere interactions through a suite of regional climate model experiments and statistical analysis. The model simulations will examine the role of land-atmosphere feedbacks through experiments in which aspects of the land surface are held fixed (e.g. fixing the leaf area so that vegetation does not wilt under dry conditions). The statistical analysis uses a technique pioneered by the PI known as Generalized Equilibrium Feedback Analysis (GEFA). The method uses multivariate lagged correlation to identify causal relationships between monsoon precipitation and forcing from remote SSTs and local land surface conditions (including vegetation as represented by the Leaf Area Index). As noted above, monsoon variations have strong human impacts, including impacts on crops, livestock, and the duration of the Australian fire season. If successful, the research performed here may lead to improvements in models used to forecast monsoon precipitation. In addition, the PI plans to lead a 2-day workshop on land-atmosphere interactions, regional climate modeling and land surface modeling, which will serve to introduce students and early-career scientists to methods used for research on regional climate and land-atmosphere interactions.