Detecting and modeling large-scale interactions between vegetation, precipitation, and temperature over temperate-semiarid and boreal climate regimes.

摘要

Terrestrial vegetation exerts an important influence on climate variability via the exchange of mass, energy, and momentum between the land surface and the atmosphere. This dissertation uses statistical techniques and stochastic models to investigate large scale vegetation/climate interactions in remotely-sensed vegetation datasets and observational climate records.;Vegetation feedbacks on climate variability over the North American Grasslands are evaluated by the methodology of Granger causality, which examines statistical causal relationships in a coupled system. Results indicate that positive vegetation anomalies earlier in the growing season significantly "Granger cause" lower rainfall (and higher temperatures) later in summer. Coupled with the positive influence of precipitation on vegetation, these interactions suggest an oscillatory variability of vegetation, which is identified in observations.;The observed vegetation-precipitation covariability is then simulated using a coupled stochastic model, which is derived from eco-hydrological principles in a semiarid environment. The model demonstrates that vegetation/precipitation interactions have distinct frequency characteristics, and are oscillatory at intraseasonal time scales, in agreement with observations. The model also indicates that this oscillatory behavior arises because enhanced vegetation depletes soil moisture faster than normal, which induces drier and warmer climate anomalies via the strong soil-moisture/precipitation coupling in this region.;Extended analyses also identify intraseasonal oscillatory variability in vegetation anomalies over the boreal forests. This characteristic variability is likely induced by interactions between vegetation and temperature, which maintain a climatological thermal balance within the soil and the lower boundary layer of the atmosphere via the removal of excess heat from the surface through enhanced evapotranspiration. The cooling effect of vegetation on temperature is detected by Granger causality analyses, and is found to be most significant over the boreal forests in lower and central Siberia.;Altogether, the findings of this dissertation highlight the role vegetation plays in regulating the water cycle and the energy balance between the surface and the atmosphere. These results present observational evidence for large-scale vegetation feedbacks, and also reveal important characteristics of the vegetation-climate system that deserve further investigation in the future.