Relationships Between Crop Water Stress Index and Alfalfa Yield and Evapotranspiration

Abstract

ABSTRACT THE objectives of this study were to compare two methods of determining Crop Water Stress Index (CWSI), and to relate CWSI to the ratio of measured evapotranspiration (E) to potential evaporation (Ep) and to dry forage yield of alfalfa (Medicago sativa L.). The first method requires determination of a lower baseline, the relationship between vapor pressure deficit (V) and canopy minus air temperature (Tc - Ta) when the crop is fully transpiring, and an upper baseline, the relationship between V and Tc - Ta when the crop is non-transpiring. The CWSI is the relative position between the upper and lower baselines. An infrared thermometer was used to measure Tc at 30-min intervals throughout 7 days in 1982 in well-watered alfalfa in Arizona to determine the lower baseline. The highest value of Tc - Ta obtained from alfalfa irrigated using a line-source sprinkler system near Las Cruces, New Mexico was taken to be the upper baseline (4 C). The infrared thermometer was used to measure Tc on 22 days in 1982 at five locations along the irrigation gradient in New Mexico. The CWSI was then determined for each day at each location on the gradient. Evapotranspiration was calculated for three cutting periods in 1982 using the water balance method at the five locations on the gradient, and Ep was calculated using the Penman equation. Dry forage yield was determined three times at the five locations. Significant relationships (P<0.01) were obtained between CWSI and measured E/Ep (r2 = 0.85), and between CWSI and yield (r2 = 0.79). The second method used energy balance considerations to determine CWSI, and required evaluation of the aerodynamic resistance (ra). Three methods were used to evaluate ra, which gave considerable differences in the CWSI estimates. When ra was evaluated using the wind profile method, CWSI gave a poor estimate of measured E/Ep and yield. This was thought to be caused by the lack of complete soil cover by the alfalfa canopy in this study. However, when ra was evaluated using the convective heat transfer method, the CWSI was significantly correlated (P<0.01) with measured E/Ep (r2 = 0.90). The third method, the leaf boundary layer method, did not provide as good a CWSI estimate as did the second method.