HR: 09:15h
AN: H11A-06    [PDF]
TI: Evaluation of the Noah Land-surface Model for Semi-arid Sites in the Southwestern United States
AU: * Hogue, T S
EM: thogue@seas.ucla.edu
AF: University of California-Los Angeles, Department of Civil and Environmental Engineering, 5732C Boelter Hall, Los Angeles, CA 90095-1593 United States
AU: Bastidas, L
EM: luis.bastidas@usu.edu
AF: Utah State University, Department of Civil and Environmental Engineering, Logan, UT 84322 United States
AU: Gupta, H
EM: hoshin_stc@sahra.arizona.edu
AF: University of Arizona, Department of Hydrology and Water Resources, Tucson, AZ 85721 United States
AU: Sorooshian, S
EM: soroosh@uci.edu
AF: University of California-Irvine, Department of Civil and Environmental Engineering, Irvine, CA 92697 United States
AB: Numerous experiments have been carried out to facilitate the development and evaluation of land-surface models. Most of these comparison studies have been undertaken by the Project for the Intercomparison of Land-surface Processes (PILPS) under the sponsorship of the GEWEX Global Land-Atmosphere System Study (GLASS). However, few of these model evaluations have used long-term data sets (greater than one or two years) or have been carried out in semi-arid regions. Understanding the interaction of land surface processes with climate and its impact on the water cycle is crucial for predicting the availability of water resources in semi-arid regions. A recently proposed PILPS SanPedro-Sevilleta experiment is being undertaken in the southwestern U.S. using five semi-arid vegetation sites (including two from this study). A systematic analysis is undertaken based on the framework proposed for the PILPS experiment, testing some of the hypotheses for the calibration and cross-validation of land-surface schemes. This specific study analyzes the National Center for Environmental Prediction (NCEP) Noah land-surface model, one of several "community" or "multi-group" models that are evolving in land-surface studies. The MOCOM algorithm (a general purpose multi-criteria optimization algorithm that provides an estimate of the Pareto solution space) is linked with the Noah model to estimate parameters for two semi-arid biomes, a desert shrub and grassland, which have continuous meteorological and flux measurements over a four-year period. The model shows improved performance with calibrations (over default) and the model also does fairly well for sensible heat and ground temperature over the longer term. However, the increase in latent heat during short-term climatic events (the year-to-year monsoon and winter-time El Ni¤o events) is not captured well in model simulations. The application of site-specific parameters at the two field sites does improve performance, but in the absence of calibration data, a proxy-basin set of parameters can be applied with only a slight decline in performance. Results from this analysis also show that several of the vegetation and soil parameters vary dependent on the length and period of calibration. Inadequate representation of these processes may result in large uncertainty in parameter estimates for semi-arid regions. The performance of models for long-term climate change scenarios is difficult to assess; however, the goal is that calibration and evaluation studies such as this and the PILPS SanPedro-Sevilleta experiment will lead to improved parameter estimates for use in regional and global long-term climate studies.
DE: 1833 Hydroclimatology
DE: 1878 Water/energy interactions
SC: Hydrology [H]
MN: 2003 Fall Meeting