Colorado Water Dec 2003 - Laserfiche WebLink

Home Browse Search

REFINING EVAPOTRANSPIRATION ESTIMATES FOR COLORADO by Marvin E. Jensen INTRODUCTION Dr. Robert Ward, Director of the Colorado Water Resources Institute, requested a brief article describing the current state -of- the -art in estimating evapotranspiration (ET) by crops for use in planning a program for refining estimates of ET for Colorado. Where applicable, he requested a sum- mary of lessons learned, or some of my past experiences, in deriving, developing and using modern technology for es- timating crop ET. The following brief comments are based on the points that we discussed. CURRENT TECHNOLOGY General Approach The most common methodology to estimate crop ET that is used in the U.S. and internationally is to first estimate evap- orative demand based on climate. Then, a crop coefficient curve that varies by crop growth stage is applied. These procedures were developed in the 1960s and are explained in detail in publications by Jensen et al. (1990), Allen et al. (1996), and Allen et al. (1998). Emerging technology involves remote sensing of cropped surfaces coupled with surface energy balance calculations ( SEBAL) based on local climate to calculate ET over indi- vidual fields or large areas (Bastiaanssen et al. 1998; Allen et al., 2002). The SEBAL methodology can also be used to determine crop coefficients for fields under current levels of irrigation management and soil salinity. This approach also has the advantage of delineating areas actually cropped excluding turn areas, ditches etc. and non - uniform irrigation practices. Reference Crop ET Reference crop ET is a measure of the evaporative demand under current climatic conditions. The most current and internationally accepted method for estimating reference crop ET is the Penman - Monteith (P -M) equation. Recently, the P -M procedures have been simplified and standardized to estimate reference ET for a standardized short crop (like grass) or a tall crop (like full cover alfalfa) using standard- ized vegetated surface parameters (Walter et al., 2002a; Walter et al. 2002b). The standardized equation, for hourly or daily values, has been evaluated under a wide range of conditions (Itenfisu, 2003). One of the most significant is an assessment of calculated half -hour ET rates for alfalfa compared with measured alfalfa ET from a 3 x 3 m weigh- ing lysimeter near Bushland, Texas under several high wind days (Howell, 2003). Measured alfalfa ET on June 13, 1998 was 17.4 mm /day (0.685 inch /day) when wind speeds were over 10 m/s (over 22 mi/h) and daytime vapor - pressure defi- cit exceeded 4.0 kPa. A similar weather pattern occurred on June 20, 1998 when measured alfalfa ET was 16 mm /day (0.63 inch /day). Half -hour estimates using the standardized equation closely tracked measured values. The total esti- mated daily ET was 5 -10% low on these windy days. This example indicates that the standardized equation not only is reliable under normal climatic conditions, but also under severe advective conditions that are often encountered on windy days in eastern Colorado. Crop Coefficients Crop coefficients originally were based on measuring soil -water depletion over several days by sampling the soil gravimetrically or using the neutron probe. The ratio of measured ET to reference crop ET is called the crop coef- ficient (K,) commonly expressed as a curve for each crop. Most current crop coefficients are based on measured daily ET for a well - watered crop using a large weighing lysimeter surrounded by preferably about 330 ft (100 m) of the same crop grown under the same soil water conditions. The ratio of measured daily ET to measured or estimated daily short or tall crop reference crop ET is the daily K. value. A set of KC values must be used with the appropriate reference crop ET. Because of the high initial cost of equipment, annual operating costs and time requirements, only a few large weighing lysimeter systems have been installed in the western U.S. Some were installed in the 1960s at Kimberly, ID, near Fresno, CA in the 1980s, and at Bushland, TX in the late 1980s. Crop coefficient curves developed by Wright (1982) at Kimberly, ID for use with an alfalfa base reference crop have been widely used in the Colorado area. These are currently being updated so that the coefficients can be used with the sum of growing- degree -days (GDD) in addition to a time scale or percentage of the growing period. General- ized procedures for developing crop curves by linear seg- ments for use with a short crop (grass) reference crop ET are available in a recent FAO publication (Allen et al., 1998). Transferability of Crop Coefficients Can crop coefficients developed in a climate somewhat dif- ferent from that in Colorado be used to estimate crop ET? For most applications, crop curves developed in another