Simulation of the evaporation of soil water beneath a wheat crop canopy

• Ritchie's 2 stage model of soil evaporation (Es) did not describe Es well.
• This model is used widely in crop models such as DSSAT and APSIM.
• Two stage model considers that stage 2 Es depends only on soil hydraulic properties.
• We found that stage 2 Es was also a function of potential evaporation.
• An exponential decay function was developed which explained stage 2 Es well.

The evaporation of soil water (Es) is an important component of the water balance in cropping systems. Although Es is considered to be non-beneficial (not directly related to crop production), it may indirectly influence water availability for transpiration which is directly related to crop growth and thus yield. Due to the difficulty in measuring Es, empirical models are usually used, especially Ritchie's two stage model. Ritchie's model assumes that stage 1 evaporation is dependent on radiant energy at the soil surface, and that during stage 2, Es is independent of radiant energy. During stage 2, the model assumes that Es is only dependent on soil hydraulic properties, and that cumulative stage 2 Es is proportional to the square root of time. To evaluate the model, we measured Es under the canopy of wheat grown on a clay loam soil in Punjab, India, using mini-lysimeters. Soil evaporation during three different potential evaporation (Eo) periods (1.5, 3.0, 6.0 mm d−1) was compared during the 2006–2007 and 2007–2008 wheat seasons. The results suggested that stage 2 Es was affected by Eo, and cumulative Es was not described well by a square root of time function. An exponential decay function which describes the decrease in Es/Eo as a function of a soil dependent constant (b), whose value is directly related to Eo, was developed. Using this function, calculated cumulative Es for two independent drying periods was close to the observed Es values with coefficients of determination of 0.82 and 0.95.