Modeling the integrated ecology, biogeochemistry, and hydrology of the global terrestrial biosphere in the Community Land Model (CLM4) (Invited) (original) (raw)

Agu Fall Meeting Abstracts, 2010

Abstract

Terrestrial ecosystems influence climate through physical, chemical, and biological processes that affect planetary energetics, the hydrologic cycle, and atmospheric composition. Much of our understanding of how terrestrial ecosystems affect climate comes from numerical models of Earth’s climate and their representation of the terrestrial biosphere. These models initially simulated only hydrometeorological processes at the land surface. They have since evolved to simulate the coupled ecology, biogeochemistry, and hydrology of terrestrial ecosystems so that the biosphere and atmosphere form a coupled system. As such, the simulated hydrologic cycle is an emergent property of the ecology and biogeochemistry represented in the model. Here, we use the Community Land Model (CLM4), the land component of the Community Earth System Model, to examine interactions among hydrology, ecology, and biogeochemistry with respect to energy, water, and carbon fluxes. Specifically, carbon uptake during gross primary production (GPP) is linked to water loss during evapotranspiration (ET). Model functional errors in the parameterization of canopy radiative transfer, leaf photosynthesis and stomatal conductance, and canopy scaling lead to substantial errors in simulated GPP and ET. Improvements to these parameterizations gained from theoretical considerations and empirical studies reduce biases in GPP, with concomitant improvement in ET. Most of the bias reduction comes from the revised photosynthesis-stomatal conductance formulation; improved canopy radiation has lesser effect; and changes to canopy scaling have minor effect. A key model parameter is the maximum rate of leaf carboxylation, which is highly uncertain for large-scale climate models and for which various disparate estimates have been published. The effect of model parameter errors on GPP and ET is of comparable magnitude to that of model functional errors and offset bias reductions from improved model parameterizations. Our results imply that this key leaf-level physiological parameter cannot be defined independent of model-specific parameterizations. Our analyses suggest that we still have much to learn about the biochemistry of photosynthesis, the biophysics of evapotranspiration, their interdependencies, and how to represent these processes in models of the terrestrial biosphere for climate simulation.

Keith Oleson hasn't uploaded this paper.

Let Keith know you want this paper to be uploaded.

Ask for this paper to be uploaded.