Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Publications at this Location » Publication #292610
Title: Estimating potential evapotranspiration with improved radiation estimation

Author
item QIU, J - Collaborator
item LIU, S - Collaborator
item MO, X - Collaborator
item Crow, Wade

Submitted to: BARC Poster Day
Publication Type: Abstract Only
Publication Acceptance Date: 4/10/2013
Publication Date: 4/18/2013
Citation: Qiu, J., Liu, S., Mo, X., Crow, W.T. 2013. Estimating potential evapotranspiration with improved radiation estimation [abstract]. 2013 BARC Poster Day, April 18, 2013, Beltsville, MD.

Interpretive Summary:

Technical Abstract: Potential evapotranspiration (PET) is of great importance to estimation of surface energy budget and water balance calculation. The accurate estimation of PET will facilitate efficient irrigation scheduling, drainage design, and other agricultural and meteorological applications. However, accuracy of PET estimation is frequently compromised by sparse radiation measurements. Especially during monsoon season, distinctive absorption and scattering processes dominating radiation propagation in atmosphere vary significantly in space and time, and thus affect the accuracy of instantaneous solar irradiance. For instance, it is found a typical variation of water vapor of 7 g cm^-2 will lead to instance solar irradiance variation of approximately 90 W m^-2 (10%) in semi-humid monsoon season, and further affect the accuracy of instantaneous PET estimation, which was sensitive to radiation term, especially in summer time. In this study, we proposed a novel way to implement the global solar irradiance model, i.e., Paulescu and Schlett scheme to better estimate surface energy budget with the aid of remote sensing and ground measurements, and further monitor the atmospheric evaporation capacity over North China Plain (NCP). Considering that different attenuation processes of radiation propagation carry different weights on atmospheric transmittance, to achieve effective estimation of radiation, the most sensitive and highly varied water vapor content was selected by sensitivity analysis, and retrieved from MODIS atmospheric product for accuracy improvement, and other less sensitive variables (ozone concentration, Angstrom’s turbidity factor, etc) were adopted using parameterization schemes. Following this, the regional PET of study area is obtained and its spatio-temporal characteristics are depicted. Validation against ground measurements indicates that simulated daily averaged solar radiation produced a bias of 5.0 to 38.0 W m^-2, RMSE of 26.3 to 51.2 W m^–2, and R2 of 0.83 to 0.96, is comparable to or outperforms existing satellite radiation products (Geostationary Operational Environmental Satellite-West, HelioClim-1, and NOAA/NESDIS Surface Radiation Budget). It also shows our simulations do an overall better job than those from widely used Angstrom scheme, and outperform Zillman scheme simulations in arid and semi-arid areas. Consequently, PET over the NCP was then simulated by the Penman Monteith method with the proposed radiation scheme. During summertime, highest PET was found in forested areas and grassland, followed by a large region of farmland in the middle of NCP. Regional daily PET reached maximum of 5.3 mm, whereas comparatively low PET was found around Shandong peninsula. This coastal-continental longitudinal trend was reversed in other seasons. Finally, the uncertainty of PET estimation introduced by MODIS products was analyzed, for future effective selection of remotely sensed data.