Author
JOHNSON, LEE - CALIFORNIA STATE UNIVERIS | |
Trout, Thomas | |
Gartung, Jimmie | |
HORNBUCKLE, JOHN - CSIRO AUSTRALIA |
Submitted to: Association of American Geographers
Publication Type: Abstract Only Publication Acceptance Date: 1/24/2007 Publication Date: 4/19/2007 Citation: Johnson, L.F., Trout, T.J., Gartung, J.L., Hornbuckle, J. 2007. Satellite Mapping of Horticultural Crop Cover in California's San Joaquin Valley - Potenial for Irrigation Water Resource Management. Association of American Geographers. Interpretive Summary: Technical Abstract: Estimation of crop water use, and associated irrigation demand, is commonly addressed by application of so-called crop coefficients, which express water loss as a proportion of evapotranspiration from a well-characterized reference crop such as grass or alfalfa. For horticultural crops, however, planting date, planting density, variety, and cultural practices can vary widely. It is thus impractical to specifiy, a priori, a crop coefficient profile that accommodates these potential sources of variability. It is generally recognized that percent canopy cover, as an indicator of intercepted sunlight, is positively related to crop evapotranspiration. Fieldwork was performed in California's San Joaquin Valley to measure canopy cover in support of satellite-based mapping. Radiance-calibrated Landsat Thematic Mapper image data were then converted to top-of-atmosphere (apparent ) reflectance and subsequently to normalized difference vegetation index. Over two consecutive seasons, a strong linear relationship (r2>0.9) was observed between vegetation index and field measurements up to the point of "effective full cover" (~75%). This relationship was subsequently used to map canopy cover throughout a 20 x 20 km study region. Crop water loss was then mapped by incorporating routine ground measurements of reference evapotranspiration collected at point locations by the California Department of Water Resources. The study results demonstrate the feasibility of using vegetation index imagery for improved, spatially explicit parameterization of crop coefficient-based models in regions of high-value horticultural production. |