Estimation of Evapotranspiration as a function of Photosynthetic Active Radiation
Abstract
The purpose of this research project is to more accurately measure the water balance and energy movements to properly allocate water resources at the Snapper Creek Site in Miami-Dade County, FL, by quantifying and estimating evapotranspiration (ET). ET is generally estimated using weather based equations, this project focused on estimating ET as a function of Photosynthetic Active Radiation (PAR). The project objectives were first to compose a function of PAR and calculated coefficients that can accurately estimate daily ET values with the least amount of variables used in its estimation equation, and second, to compare the newly identified ET estimation PAR function to TURC estimations, in comparison to our actual Eddy Covariance (EC) ET data and determine the differences in ET values. PAR, volumetric water content (VWC), and temperature (T) data were quality checked and used in developing singular and multiple variable regression models fit with SigmaPlot software. Fifteen different ET estimation equations were evaluated against EC ET and TURC estimated ET using R2 and slope factors. The selected equation that best estimated EC ET was cross validated using a 5 month data set; its daily and monthly ET values and sums were compared against the commonly used TURC equation. Using a multiple variable regression model, an equation with three variables (i.e., VWC, T, and PAR) was identified that best fit EC ET daily data. However, a regression was also found that used only PAR and provided ET predictions of similar accuracy. The PAR based regression model predicted daily EC ET more accurately than the traditional TURC method. Using only PAR to estimate ET reduces the input variables as compared to using the TURC model which requires T and solar radiation. Thus, not only is the PAR approach more accurate but also more cost effective. The PAR-based ET estimation equation derived in this study may be over fit considering only 5 months of data were used to produce the PAR function. However, the relationship between PAR and daily ET was established for the data available. Future work will continue to examine this relationship.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFM.H43C1348W
- Keywords:
-
- 1816 HYDROLOGY / Estimation and forecasting