Where do we think the global vegetation is going next?

Background/Question/Methods

The last 50 years has seen a tremendous advance in measuring and modeling global vegetation dynamics. Most prominently, the advent of satellite datasets for measurement and computer simulation capacity for global modeling has been key. Satellites easily detect vegetation structural change and can quantify landscape level disturbance rates, but cannot discriminate species assemblages that most ecologists desire. Vegetation growth is not directly detectable, but is inferred from ecosystem biophysical principles. Successional dynamics are not so well detected by current satellite technology except early regrowth after major disturbance. Gradual biome transitions are also much more difficult for satellite observation. Modeling of global vegetation is the domain of the DGVM, the Dynamic Global Vegetation Model. The biogeochemistry, or integrated energy/water/carbon/nitrogen algorithms are rather well worked out. However the logic for triggering disturbances and biome transitions is experimental, but young in development. Drought stress and fire simulation are the best developed and are even being tested for seasonal fire forecasting.  The timing and magnitude of mortality from processes, other than fire, however, are topics of current debate and development.  Biome transitions are more easily simulated via changes in water-balance, for example savanna to forest or vice versa, but are more difficult with respect to dispersal and invasion lags of neighboring Biomes induced by thermal changes.  Global modeling particularly challenges the scientist to define general principles that can represent biome dynamics worldwide, however it provides a tremendous opportunity to interface with satellite observations. Yet, even a model with robust logic requires accurate initial conditions of key system components to gauge near-term responses, which may be nearly impossible to measure globally.  Still, an enduring question is the importance of initial conditions in determining the long-term trajectory of a site.  And lastly, the incorporation of evolving land-use practices and effects within the constraints of the evolving Biome dynamics remains perhaps the most difficult task yet to be incorporated in the science of DGVMs.

Results/Conclusions

This paper will sample from many of the developments in global satellite observations, and DGVM science, and speculate where the next breakthroughs may be. If we feel reckless we may even suggest some policy implications for this thinking.