HR: 1525h
AN: B23G-08
TI: An International Land-Biosphere Model Benchmarking Activity for the IPCC Fifth Assessment Report (AR5)
AU: Hoffman, F M
EM: forrest@climatemodeling.org
AF: Oak Ridge National Laboratory, Oak Ridge, TN, United States
AU: Randerson, J T
EM: jranders@uci.edu
AF: University of California-Irvine, Irvine, CA, United States
AU: Thornton, P E
EM: thorntonpe@ornl.gov
AF: Oak Ridge National Laboratory, Oak Ridge, TN, United States
AU: Bonan, G B
EM: bonan@ucar.edu
AF: National Center for Atmospheric Research, Boulder, CO, United States
AU: Brooks, B J
EM: bjorn@climatemodeling.org
AF: Oak Ridge National Laboratory, Oak Ridge, TN, United States
AU: Erickson, D J
EM: ericksondj@ornl.gov
AF: Oak Ridge National Laboratory, Oak Ridge, TN, United States
AU: Fung, I
EM: ifung@berkeley.edu
AF: University of California-Berkeley, Berkeley, CA, United States
AB: The need to capture important climate feedbacks in general circulation models (GCMs) has resulted in efforts to include atmospheric chemistry and land and ocean biogeochemistry into the next generation of production climate models, called Earth System Models (ESMs). While many terrestrial and ocean carbon models have been coupled to GCMs, recent work has shown that such models can yield a wide range of results (Friedlingstein et al., 2006). This work suggests that a more rigorous set of global offline and partially coupled experiments, along with detailed analyses of processes and comparisons with measurements, are needed. The Carbon-Land Model Intercomparison Project (C-LAMP) was designed to meet this need by providing a simulation protocol and model performance metrics based upon comparisons against best-available satellite- and ground-based measurements (Hoffman et al., 2007). Recently, a similar effort in Europe, called the International Land Model Benchmark (ILAMB) Project, was begun to assess the performance of European land surface models. These two projects will now serve as prototypes for a proposed international land-biosphere model benchmarking activity for those models participating in the IPCC Fifth Assessment Report (AR5). Initially used for model validation for terrestrial biogeochemistry models in the NCAR Community Land Model (CLM), C-LAMP incorporates a simulation protocol for both offline and partially coupled simulations using a prescribed historical trajectory of atmospheric CO₂ concentrations. Models are confronted with data through comparisons against AmeriFlux site measurements, MODIS satellite observations, NOAA Globalview flask records, TRANSCOM inversions, and Free Air CO₂ Enrichment (FACE) site measurements. Both sets of experiments have been performed using two different terrestrial biogeochemistry modules coupled to the CLM version 3 in the Community Climate System Model version 3 (CCSM3): the CASA model of Fung, et al., and the carbon-nitrogen (CN) model of Thornton. Comparisons of the CLM3 offline results against observational datasets have been performed and are described in Randerson et al. (2009). CLM version 4 has been evaluated using C-LAMP, showing improvement in many of the metrics. Efforts are now underway to initiate a Nitrogen-Land Model Intercomparison Project (N-LAMP) to better constrain the effects of the nitrogen cycle in biosphere models. Presented will be new results from C-LAMP for CLM4, initial N-LAMP developments, and the proposed land-biosphere model benchmarking activity.
UR: http://www.climatemodeling.org/c-lamp
DE: [0414] BIOGEOSCIENCES / Biogeochemical cycles, processes, and modeling
DE: [0428] BIOGEOSCIENCES / Carbon cycling
DE: [0429] BIOGEOSCIENCES / Climate dynamics
DE: [1622] GLOBAL CHANGE / Earth system modeling
SC: Biogeosciences (B)
MN: 2009 Fall Meeting