This dissertation examined a number of general vegetation-modeling issues, and the sensitivity of terrestrial net primary productivity (NPP), soil moisture and actual evapotranspiration (ET) to climatic variations in China. The specific issues addressed included: (1) the sensitivity of the performance of an equilibrium vegetation model to the choice of monthly-mean climatologies, observed validation data sets, and three map-comparison approaches; and (2) the limitations of existing map-comparison approaches in vegetation modeling; and the variation and climate sensitivity of (3) terrestrial NPP and (4) soil moisture and actual ET in China.

To address the first issue, BIOME4 (Kaplan et al., 2002), a typical example of an equilibrium vegetation model, was used along with a set of 19 different monthly-mean climatologies, three validation data sets, and several map-comparison methods. To address the second issue, the "opposite and identity" (01) index (Tang, 2008) was developed for evaluating the correspondence of two simulation results. To examine the third issue, a set of historical NPP dynamics were derived from normalized-difference vegetation index data by modifying the CASA (Potter et al., 1999) approach and then were linked to the variation of temperature and precipitation to analyze the climatic effects on terrestrial NPP in China. To examine the fourth issue, a stand-alone water balance model, LH (LPJ-hydrology), was developed by modifying the LPJ dynamic global vegetation model (Sitch et al., 2003), and applying it to a China case study.

The results of these analyses indicate that (1) the 30-year mean-climatology preceding the observed data produces the most accurate vegetation simulations; (2) the OI index is a useful tool to compare two simulation results or to evaluate simulation results against observed spatiotemporal data; (3) climate and land-use change jointly controlled NPP dynamics in the eastern monsoon zone of China. In contrast, NPP dynamics in the north-west and zone and in the Tibet Plateau frigid zone depended more on climatic variation; and (4) the spatial patterns of soil moisture and ET in China were correlated with the variation of temperature and precipitation. However, the strength of such relationship varies spatially.

This dissertation includes my published and coauthored materials.