Wednesday, November 7, 2007
285-6

Evaluation of Radiation Partitioning Models at Bushland, Texas.

Paul D. Colaizzi, Steven R. Evett, Terry A. Howell, Robert C. Schwartz, Judy A. Tolk, Prasanna H. Gowda, and Jose L. Chavez. Conservation and Production Research Laboratory, USDA-ARS, P.O. Drawer 10, Bushland, TX 79012

Crop growth and soil-vegetation-atmosphere continuum energy transfer models often require estimates of net radiation components, such as photosynthetic, solar, and longwave radiation to both the canopy and soil. We evaluated the 1998 radiation partitioning model of Campbell and Norman, herein referred to as CN98. The CN98 model accounts for different transmittance and albedo characteristics of the soil and canopy in the visible, near-infrared, direct, and diffuse radiation components. We also evaluated a simpler extinction-type model that assumes single transmittance and albedo values for the entire shortwave spectrum. Model output was compared with measurements of photosynthetic irradiance (assuming 1 W m-2 = 4.57 μmoles m-2 s-1) and solar irradiance transmitted to the soil and reflected from the canopy, net radiation transmitted to the soil, and total net radiation measured over the canopy. Measurements were made in large, weighing lysimeters at Bushland, TX with grain sorghum and corn crops, and with bare soil (prior to crop emergence). Both models obtained similar calculation of all parameters except for reflected photosynthetic irradiance; in that case, the CN98 model resulted in better agreement with measurements. For both models, the RMSE of observations were within 20% for transmitted solar irradiance (RMSE = 78 W m-2), reflected solar irradiance (RMSE = 16 W m-2), and total net radiation (RMSE = 63 W m-2). However, model agreement with observations were poor (RMSE/mean observations > 90%) for transmitted photosynthetic irradiance (RMSE = 50 W m-2), reflected photosynthetic irradiance (RMSE = 9 and 41 W m-2 for CN98 and extinction, respectively), and transmitted net radiation to the soil (151 W m-2).