Vegetation and climate variability in North America—a study using the SSiB4/TRIFFID biophysical/dynamic vegetation model

In the past three decades, numerous modeling sensitivity studies have established the importance of land/atmosphere interaction in climate system. However, in most of these studies, vegetation types were fixed during the period of model integration. Recently, new evidence emerged from satellite data analyses indicates that including a fully coupled process is of imminent important in explaining the relationship discovered in these analyses.

We apply off-line SSiB4/TRIFFID over North America for this study. The TRIFFID is a dynamic vegetation model, in which the relevant land-surface characteristics of vegetation cover and structure are modeled directly. The areal coverage, leaf area index (LAI), and canopy height of each type are updated based on a carbon balance approach, in which vegetation change is driven by carbon assimilation, allocation, and accumulation, as well as competition between plant functional types (PFT). SSiB4 is a biophysical model, including the consideration of surface heterogeneity, and a parameterization scheme to directly relate surface albedo, surface roughness length, zero displacement height, and surface aero dynamic resistances within and under canopy to LAI and vegetation height for different PFTs.

The offline SSiB4/TRIFFID is tested using the observed precipitation and reanalysis-based meteorological forcing from 1948 to 2000 with 1 degree horizontal resolution. The results indicate that the model with relatively realistic forcing was able to produce reasonable vegetation distributions. The spatial distribution of dominator types is generally consistent with the current vegetation map derived from satellite, with deficiency in some PFTs' distributions. The simulated LAIs are also compared with satellite derived products.

To investigate the mechanism of ecosystem, water, carbon, and radiation interactions, the analysis is conducted to find relationships between simulated LAI and environmental conditions. It is found that the vegetation characteristics simulated by SSiB4/TRIFFID responds primarily to four factors: air temperature, soil moisture, photosynthesis, and solar radiation (CO2 is fixed in simulation). Soil temperature plays a dominant role in northern N. America and southeast U.S. In central U.S., both soil temperature and soil moisture play the role. In western U.S., higher temperature reduces vegetation productivity. Further south, in Mexico, the plant growth is only controlled by soil moisture due to high temperature year around there. Soil moisture is not a control factor in Canada but dominate in central and western U.S.