Joint assimilation of leaf area index and aboveground biomass into CLM significantly reduces carbon uptake and storage in the Arctic and Boreal Region.
Xueli Huo, University of Arizona, huoxl90@email.arizona.edu
Andrew M Fox, Joint Center for Satellite Data Assimilation, afox@ucar.edu
Hamid Dashti, University of Arizona, ahangar.hamid@gmail.com
William Gallery, University of Arizona, wogallery@comcast.net
Jeffrey Anderson, NCAR, jla@ucar.edu
Charlie Devine, University of Arizona, cjdevine@email.arizona.edu
David JP Moore, University of Arizona, davidjpmoore@email.arizona.edu (Presenter)
Carbon storage in the Arctic and Boreal zone is a key control of the global carbon cycle. Future projections of carbon uptake and storage are strongly controlled by model estimates of current leaf area index (LAI) and vegetation biomass. The Community Land Model (CLM) over estimates both LAI and biomass in the Arctic and Boreal (ABoVE) region. To correct the positive bias in the two variables in CLM, we assimilated the 8-day Moderate Resolution Imaging Spectroradiometer (MODIS) LAI observation and a machine learning product of annual aboveground biomass into CLM using an Ensemble Adjustment Kalman Filter (EAKF) implemented by the Data Assimilation Research Testbed (DART). We ran ensemble CLM runs with, and without, data assimilation (free run and DA run) driven by 40 forcing ensemble members from the Community Atmosphere Model version 6 (CAM6) reanalysis. Assimilating LAI and aboveground biomass in the ABoVE region reduced these model estimates relative to the free run by 59% and 72%, respectively. The change of aboveground biomass was consistent with independent regional estimates of canopy top height (ICESat), as well as airborne lidar-derived canopy height at two NEON (National Ecological Observatory Network) sites located in the ABoVE region. Comparisons to the International Land Model Benchmarking (ILAMB) systems, DA significantly improves CLM’s performance in simulating the carbon and hydrological cycles, as well as in representing the functional relationships between LAI and other variables (aboveground biomass, total biomass, GPP, and evapotranspiration). Using corrected model states, we investigate the persistent bias in GPP by modifying the photosynthesis process within CLM to include altered photosynthetic quantum yield. Combining data assimilation with model parameterization, we were able to identify and correct biases in gross primary productivity.
Poster: Poster_Huo_2-15_188_35.pdf
Associated Project(s):
Poster Location ID: 2-15
Presentation Type: Poster
Session: Poster Session 2
Session Date: Wed (May 10) 5:15-7:15 PM
CCE Program: TE