WP6: Quantification of biodiversity feedbacks in climate models

This is a task of the Swiss Federal Institute of Technology in Zurich, ETH.

To assess the feedback of changes in BD facets on the climate system we use the integrated COSMO-CLM2 regional climate model8. The COSMO climate model is coupled to the Community Land Model (CLM), which in turn represents the land surface feedbacks to the climate model. The COSMO-CLM2 modeling system, which captures plant photosynthesis and carbon processes constitutes a regional Earth System Model. CLM also allows to investigate potential feedbacks of modified land water or energy fluxes for the regional climate11, and in particular extreme events, and it represents several aspects of specific vegetation processes. However, there are two important constraints that prevent the model from enabling the assessment of feedbacks from BD change on climate. First, COSMO-CLM2 does not track BD change explicitly as it does not track subtle changes in vegetation structure except for very broad functional types. Therefore, data provided at finer thematic and spatial resolution from WP2 & WP3 are used as forcing to COSMO-CLM2. Second, the model is operated at a comparably coarse spatial resolution of ca. 12.5 km, which is coarser than model simulations in WP2 and WP3. We will therefore replace parts of the CLM dynamics with spatially distributed input from WP2 (dynamic simulations over Europe), WP3 (dynamic simulations over the Alps) and WP4 (more detailed vegetation classification not provided in WP2). These data will be aggregated and prepared by WP1 (spatially distributed time series of LSM properties for input in WP6) from finer spatial resolution to the COSMO-CLM2 12.5 km raster. This will allow us to compare standard COSMO-CLM2 model runs with a) those provided at higher thematic and spatial resolution so that the effect size of BD feedbacks to the climate system can be quantified, or b) with targeted scenario simulations of vegetation dynamics as requested by stakeholders. We will do so in two steps, first by replacing structural aspects of the few broad functional types with more subtle vegetation details as input to the COSMO-CLM2, and then to also replace more physiological parameters as input to CLM. The experiments with the modified COSMO-CLM2 model will be run to investigate biodiversity feedbacks to climate.

Precipitation rate simulated with the COSMO model visualized over a few hours at July 12 2006. source: ETHZ


  1. Adjust the CLM model structure and run first set of experiments: Adjust the CLM model structure such that it reads spatially distributed time series of land surface input for albedo, canopy height & roughness, LAI, and SLA provided by WP1 (M1.3). Run a first set of simulations to provide the coarse climate input data for the downscaling in WP1, and simulations in WP2 & WP3.
  2. Adjust the CLM model structure further and run second set of experiments, such that it also reads spatially distributed time series of evapotranspiration, Vcmax and stomatal conductance provided by WP1 (M1.3). Run experiments on BD feedback on climate change using the time series provided by WP2 & WP3 as forcing input (with aspects of BD either fully modeled, or held constant). Run experiments in response to stakeholder input.


  1. Baseline COSMO-CLM2 climate simulations run for input in WP1 for downscaling completed;
  2. CLM model adjusted to read input from spatially distributed fields for canopy height and roughness, SLA, Vcmax, and spatially distributed time series of albedo and LAI;
  3. CLM model adjusted to read spatially distributed patterns of stomatal conductance and time series of vegetation evapotranspiration;
  4. Scenario simulations from M6.2 completed.


  1. Paper on the sensitivity and effect size of biodiversity change as expressed in changing canopy structures, albedo, LAI etc.;
  2. Paper on the importance of improving vegetation evapotranspiration modelling in LSMs;
  3. Results from model experiments summarized for report to stakeholders and users.

Cover image: The National Oceanic and Atmospheric Administration (NOAA) on Unsplash.