Computing Transpiration of Sunlit and Shaded Cotton Foliage under Variable Water Stress

The accurate calculation of transpiration from plant canopies is critical in order to judiciously manage limited water resources in many agricultural areas. The objective of this study was to compute transpiration from the sunlit and shaded fractions of a cotton (Gossypium hirsutum, cv, Acala SJ2) canopy using the Penman-Monteith energy balance equation. These computations were validated by a heat pulse system measuring stem sap flow, and by measuring the crop surface energy balance by the Bowen ratio method during two field seasons in 1988 and 1989. Adjusting the relationship between absorbed radiation and leaf conductance for decreasing leaf water potential sensitized the Penman-Monteith energy balance computation to variable water stress conditions. Results showed that the computed transpiration of the shaded canopy comprised 0.21, 0.28, and 0.33 of the total during severe water stress in normal atmospheric demand, non-stress in normal atmospheric demand, and non-stress in very high atmospheric demand, respectively. The average daily computed transpiration for the two seasons was within 5% of that measured by the heat pulse system and the Bowen ratio. The hourly computation exhibited the expected sensitivity to fluctuations in atmospheric demand and to the change of available soil moisture. The linear regression of computed transpiration for measured water uptake by the heat pulse system yielded an R2 value of 0.96. Quadratic regression analyses during non-stress and stress for computed hourly transpiration vs. evapotranspiration yielded R2 values of 0.97 and 0.92, respectively. These analyses show that midday evapotranspiration is systematically higher than computed transpiration, which would presumably be due to evaporation from the soil.