An insight to the performance of crop water stress index for olive trees

Optimization of olive oil quantity and quality requires finely tuned water management, as increased irrigation, up to a certain level, results in increasing yield, but a certain degree of stress improves oil quality. Monitoring tools that provide accurate information regarding orchard water status would therefore be beneficial. Amongst the various existing methods, those having high resolution, either temporally (i.e., continuous) or spatially, have the maximum adoption potential. One of the commonly used spatial methods is the Crop Water Stress Index (CWSI). The objective of this research was to test the ability of the CWSI to characterize water status dynamics of olive trees as they enter into and recover from stress, and on a diurnal scale. CWSI was tested in an empirical form and in two analytical configurations. In an experiment conducted in a lysimeter facility in the northwestern Negev, Israel, irrigation was withheld for 6 days for 5 of 15 trees, while daily irrigation continued for the rest of the trees. After resuming irrigation, the trees were monitored for 5 additional days. Water status measurements and thermal imaging were conducted daily between 12:00 and 14:00. Diurnal monitoring (predawn to after dusk) of the same indicators was conducted on the day of maximum stress. Continuous meteorological data were acquired throughout the experimental period. Empirical and analytical CWSI were calculated based on canopy temperature extracted from thermal images. The empirical CWSI differentiated between well watered and stressed trees, and depicted the water status dynamics during the drought and recovery periods as well as on a diurnal scale. Analytical approaches did not perform as well at either time scale. In conclusion, the empirical CWSI seems to be promising even given its limitations, while analytical forms of CWSI still require improvement before they can be used as a water status monitoring tool for olive orchards. Practically, it is proposed to compute the wet temperature analytically and set the dry temperature to 5 °C higher than air temperature.