|Irmak, S - UNIV. OF NEBRASKA|
|Irmak, A - UNIV. OF NEBRASKA|
|Martin, D - UNIV. OF NEBRASKA|
|Payero, J - UNIV. OF NEBRAKSA|
Submitted to: Journal of Irrigation and Drainage Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: July 14, 2006
Publication Date: N/A
Interpretive Summary: Crop water use is regularly computed as a ratio to a reference crop's water use. This paper compared the computed reference crop water use from a short, smooth crop like short cool-season grass with water use from a taller and "rougher" crop like alfalfa using standardized weather variables measured in differing climates. The two reference crop water use data were computed with various equations designed for that reference crop and then using the estimation procedure proposed in the FAO-56 (Food and Agriculture Organization of the United Nations) handbook. The rato of the taller to shorter reference crop was defined as a parameter, Kr. Kr varied substantially suggesting the need to develop Kr values for a local region. In general, the variation in growing season Kr values was less than the calendar year for most of the methods. The variability between the locations was largest for the FAO-56 method, especially for the low relative humidity and wind speed conditions, suggesting the necessity for modifications in this Kr method. In general, year-to-year variability for the same location was low. Thus, the "true" performance of the reference crop water use estimates by any combination equation, including the ASCE-PM (Penman-Monteith equation) estimates needs further research. Results suggest that the ASCE-PM taller to shorter reference crop ratio, Kr, can provide reasonable Kr values for the locations where simultaneous measured taller reference crop water use and shorter reference crop water use data are not available.
Technical Abstract: For several reasons, the alfalfa-reference evapotranspiration (ETr) values need to be converted to grass-reference ET (ETo), or vice versa, to enable the crop coefficients developed for one reference surface be used with the other. However, studies are lacking in evaluating or developing procedures to provide guidance on how to make the conversions. The objectives of this study were to: (1) develop ETr-to-ETo ratios (Kr values) for different climatic regions on a monthly basis for the growing season and non-growing (dormant) season, (2) determine the seasonal behavior of Kr values within a location and between the locations to assess whether the Kr values developed for a region can be used in other locations. (3) quantify the root mean square difference (RMSD) associated when making the conversions. Monthly average Kr values from daily values were developed for Bushland (TX), Clay Center (NE), Davis (CA), Gainesville (FL), Phoenix (AZ), and Rockport (MO) for calendar year and growing season (May-September) using the ETr-to-ETo ratios from the standardized American Society of Civil Engineers Penman-Monteith (ASCE-PM) ETr, Food and Agriculture Organization Paper 56 (FAO56), 1972 and 1982 Kimberly-Penman, 1963 Jensen-Haise, and High Plains Regional Climate Center (HPRCC) Penman methods. The standardized ASCE-PM ETo values were used as the basis for the ETo estimates. The Kr values exhibited substantial variation between the locations. For example, the Kr values developed from the 1972 Kimberly-Penman method in June were 1.49, 1.39, 1.35, 1.19, 1.27, and 1.31, for Bushland, Clay Center, Davis, Gainesville, Phoenix, and Rockport, respectively. There was enough variability in the magnitude of the Kr values between the locations to justify the need for developing the Kr values for a local region. Using one Kr value developed for a local climate in other climates that have different characteristics will result in considerable errors in estimating reference ET and converting crop coefficients. In general, the variation in growing season values was less than the calendar year for most of the methods. The magnitude of variation between the locations was less for the ASCE-PM Kr values than other methods at all locations. The variability between the locations was largest for the FAO56 method, especially for the low relative humidity and wind speed conditions, suggesting the necessity for modifications in this Kr method. In most cases, the average standard deviation between the years were less than 0.13 for the calendar year and less than 0.10 for the growing season indicating that the average Kr value for a specific month should not be expected to vary by more than 0.13 from year to year for most of the methods for the calendar year and not more than 0.10 for the growing season. Some of the methods, including the ASCE-PM, produced potentially unrealistically high Kr values (e.g., 1.78, 1.80) during the non-growing season. This was attributed to the climatic and surface conditions during this period. The standardized reference surface conditions now used in the standardized ASCE-PM equation do not exist during dormant periods resulting in potentially unrealistic estimates of Kr. However, the effect of the potentially unrealistically high estimates of Kr values on ETo estimates was somewhat lower than one would have expected. This might be due to low ETo values during this period. Nevertheless, information on the “true” performance of the reference ET estimates by combination equations, including the ASCE-PM estimates, is lacking. The analyses and comparisons of the dormant periods ET by combination methods against measured data and developing new methodologies for dormant season ET and Kr need further research. The RMSD values between the ETo (method) and the ASCE-PM ETo values showed substantial variation between the methods. Considerable discrepancies were obtained in difference associated with making the conversions for the methods and for the same methods between the locations. The difference between the ETo values before and after making the conversions was least for the ASCE-PM method. These findings suggest that the ASCE-PM method seems to provide more robust and realistic Kr values than other methods. The Kr values need to be developed for a local region to obtain more accurate conversion coefficients in the absence of simultaneous lysimeter ETr and ETo measurements.