HR: 16:45h
AN: H24D-04    [Abstracts]
TI: High-resolution ensemble solar radiation estimates through assimilation of coarse-scale retrievals into a simple physical insolation model
AU: Lee, S
EM: songweon@seas.ucla.edu
AF: UCLA, Dept. of Civil and Environmental Engineering, Los Angeles, CA 90095
AU: * Margulis, S A
EM: margulis@seas.ucla.edu
AF: UCLA, Dept. of Civil and Environmental Engineering, Los Angeles, CA 90095
AB: The incident solar radiation flux at the earth's surface is the primary driver of the energy and water exchange between atmosphere and land or ocean, and therefore plays an important role in agriculture, climate research and monitoring, long-range weather prediction and the global hydrologic cycle. In this study a simple physical radiative transfer insolation model is developed for use in both prediction and data assimilation applications. The major advantage of this model is in its parsimony through only considering the most important parameters in the insolation process. One of the novel aspects of the model is the use of the Visible Infrared Solar-infrared Split window Technique (VISST) pixel-level cloud product as a primary model input. The product is used to facilitate cloud detection and to directly estimate cloud reflectance and absorption. The parsimony of the model and high-resolution VISST cloud product (available at ~ half-hourly temporal and 0.02° spatial resolution) allow for the computationally efficient prediction of high resolution solar radiation fields. The deterministic model was first tested over the Southern Great Plains (SGP) region during the summer of 1997. The insolation predictions were shown to correlate well with observations from ground measurements from 12 Atmospheric Radiation Measurement (ARM) SGP facilities during the investigated period. To further improve the solar radiation predictions, a probabilistic approach was adopted in order to merge the model predictions with well-developed retrieval products from the Solar Radiation Budget (SRB) downward shortwave radiation data from GEWEX (Global Energy and Water Cycle Experiment) Continental International Scale Project (GCISP), which is available at hourly temporal and 0.5° spatial resolutions. Probability distributions of the model input parameters were specified to generate an ensemble of open-loop high-resolution predictions. The coarse-scale SRB retrieval estimates were then assimilated into the model using an Ensemble Kalman Filtering (EnKF) approach. The result is a high-resolution estimate that weighs the uncertainty in both the simple model and the retrieval product. Comparison of the posterior estimates to the ground observations are performed and show significant improvement over the open-loop estimates, especially in cloudy sky conditions.
DE: 0321 Cloud/radiation interaction
DE: 0360 Radiation: transmission and scattering
DE: 1814 Energy budgets
DE: 1855 Remote sensing (1640)
DE: 1878 Water/energy interactions (0495)
SC: Hydrology [H]
MN: Fall Meeting 2005