The development and testing of the UCD Advanced Canopy-Atmosphere-Soil Algorithm (ACASA) for use in climate prediction and field studies.

摘要

The University of California, Davis Advanced Canopy-Atmosphere-Soil model (UCD ACASA) is presented and its output is compared with a comprehensive set of observations at six diverse sites. ACASA is a multi-layer canopy-surface-layer model that solves the steady-state Reynolds-averaged fluid flow equations to the third order. ACASA includes a fourth-order, near-exact technique to calculate leaf, stem, and soil surface temperatures and surface energy fluxes. Plant physiological response to micro-environmental conditions is also included using Ball-Berry/von Caemmerer-Farquhar formulations. Results from comparing model and observed estimates agree within 95% statistical confidence for all six sites with few exceptions. Results also indicate that ACASA generally produces flux estimates that are better correlated with observations than those from the Biosphere-Atmosphere Transfer Scheme (BATS). Sensitivity tests show that reducing the vertical resolution, linearizing surface temperature calculations, and simplifying the treatment of surface layer turbulence each altered mean sensible and latent heat flux estimates by amounts that are statistically significant in many cases.; ACASA is coupled to the PSU/NCAR Mesoscale Model Version 5 (MM5) as a surface-layer flux scheme. Preliminary runs for July 1998 over western North America show that this coupling has been successful, with little evidence of numerical instability failures in the linkage between MM5 and ACASA. Results for mean flux estimates and near surface air temperatures suggest that the model is sensitive to initial soil moisture conditions mostly through altering soil thermal conductivity. High sensitivity exists in choosing between ACASA and the surface-layer scheme chosen for the reference run, where mean sensible and latent heat fluxes differ by over 100 Wm−2 for much of the desert southwest. Changing initial soil moisture altered July 18–31 mean near-surface air temperature and specific humidity estimates by several K and g kg−1, respectively. The MM5-default simulation produced July 18–31 total accumulated precipitation amounts over the and continental interior exceeding 20 cm, while MMS-ACASA predicted 3 to 8 cm over the same period in the region.