Taking off the training wheels: the properties of a dynamic vegetation model without climate envelopes, CLM4.5(ED)
- 1National Center for Atmospheric Research, Boulder, Colorado 80305, USA
- 2Los Alamos National Laboratory, Los Alamos, New Mexico 87454, USA
- 3Lawrence Berkeley National Laboratory, Berkeley, California, USA
- 4Woods Hole Research Center, Falmouth, Massachusetts, USA
- 5Department Environment, Earth and Ecosystems, Open University, Milton Keynes, UK
- 6Department Atmospheric Chemistry, Max Planck Institute for Chemistry, Mainz, Germany
Abstract. We describe an implementation of the Ecosystem Demography (ED) concept in the Community Land Model. The structure of CLM(ED) and the physiological and structural modifications applied to the CLM are presented. A major motivation of this development is to allow the prediction of biome boundaries directly from plant physiological traits via their competitive interactions. Here we investigate the performance of the model for an example biome boundary in eastern North America. We explore the sensitivity of the predicted biome boundaries and ecosystem properties to the variation of leaf properties using the parameter space defined by the GLOPNET global leaf trait database. Furthermore, we investigate the impact of four sequential alterations to the structural assumptions in the model governing the relative carbon economy of deciduous and evergreen plants. The default assumption is that the costs and benefits of deciduous vs. evergreen leaf strategies, in terms of carbon assimilation and expenditure, can reproduce the geographical structure of biome boundaries and ecosystem functioning. We find some support for this assumption, but only under particular combinations of model traits and structural assumptions. Many questions remain regarding the preferred methods for deployment of plant trait information in land surface models. In some cases, plant traits might best be closely linked to each other, but we also find support for direct linkages to environmental conditions. We advocate intensified study of the costs and benefits of plant life history strategies in different environments and the increased use of parametric and structural ensembles in the development and analysis of complex vegetation models.