Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: October 25, 2011
Publication Date: January 5, 2012
Citation: Colaizzi, P.D., Schwartz, R.C., Evett, S.R., Howell, T.A., Gowda, P., Tolk, J.A. 2012. Radiation model for row crops: II. Model evaluation. Agronomy Journal. 104(2):241-255. Interpretive Summary: Water and fertilizer management of agricultural crops is important for maintaining farm profitability. Mathematical models for predicting crop growth and water use are becoming increasingly important tools for crop management; therefore, continued model improvement is essential in order for United States farmers to compete in a global market. Since solar radiation is required for crop growth and crop water use, it is important to accurately model the amount of solar radiation that is available to the crop. Most crops are planted in rows, but this complicates the computation of solar radiation available to the crop. We refined an existing solar radiation sub-model that is commonly used in crop growth and water use models. In Part 2, we showed that this refinement improved the computation of solar radiation for row crops, and we determined how sensitive the model was to certain inputs. This improvement will impact crop growth and water use models, which have the potential to improve the management of water, fertilizer, and other farm inputs.
Technical Abstract: Relatively few radiation transfer studies have considered the impact of varying vegetation cover that typifies row crops, and meth¬ods to account for partial row crop cover have not been well investigated. Our objective was to evaluate a widely used radiation model that was modified for row crops having sparse to full vegetation cover. The radiation model was combined with geometric view factors based on elliptical hedgerows that account for the spatial distribution of row crop vegetation, and this approach was compared with the more commonly used clumping index approach. Irradiance measurements included transmitted and reflected visible and shortwave, outgoing longwave, and total net radiation. The model used optimized parameters for corn (Zea mays L.), grain sorghum [Sorghum bicolor (L.) Moench], and cotton (Gossypium hirsutum L.). The elliptical hedgerow and clumping index approaches resulted in similar model agreement; however, the former resulted in up to 7.3 W m**-2 smaller RMSE and up to 7.5 W m**-2 smaller mean bias error compared with the latter. Both approaches resulted in similar model sensitivities to inputs, which varied +/-25%. Calculated shortwave irradiance fluxes were most sensitive to leaf area index (LAI; –3.25), canopy width (–1.94), ellipsoid leaf angle parameter (–0.77), and visible leaf absorption (–5.54) when LAI = 2.95 m**2 m**-2, and visible soil reflectance (0.89) when LAI = 0.21 m**2 m**-2. Calculated outgoing longwave irradiance and net radiation were most sensitive to the soil directional brightness temperature (0.55 and –0.61, respectively) when LAI = 0.21 m**2 m**-2.