Use of canopy temperature for precision irrigation of wine grapes
Abstract
The objective of this research was to develop a method for calculating a water stress index (CWSI) for wine grape without direct measurement of well-watered and non-transpiring vines. ‘Syrah’ grapevines were grown under arid conditions in treatment plots that were drip-irrigated at a frequency of one or three times per week to supply either 70 or 35% of evapotranspiration-based vine water demand (ETc) in Parma, ID USA (43.78N, 116.94W; 750 m asl). Vines located in the trial perimeter were irrigated at >100% ETc. Infrared radiometers (20 cm diameter temperature sensing area), installed ~30 cm above sun-exposed, fully-expanded leaves at the top of the vine canopy, were used to record canopy surface temperature of vines in irrigation treatment plots and in the trial perimeter. Meteorological sensors recorded air temperature, relative humidity, total solar radiation and wind speed. An artificial neural network was trained to predict canopy temperature under well-watered conditions using data collected during the 2h period encompassing solar noon. The linear regression of predicted versus observed values for well-watered vines had a correlation coefficient of 0.91 with a root mean square error of 0.95 °C.
The canopy temperature of non-transpiring vines was estimated from the cumulative probability distribution of the temperature difference between the canopy of vines irrigated at 35% ETc and ambient air. A CWSI was calculated for vines in irrigation trial plots using their measured canopy temperatures and the estimated and predicted values for non-transpiring and well-watered vines. The CWSI for irrigation treatment plots was linearly correlated with midday leaf water potential and its slope and intercept varied with irrigation frequency. Results showed that meteorological data input into a neural network model combined with measured canopy temperature can be used to calculate a real-time CWSI without direct measurement of well-watered or non-transpiring vines.
Issue: GiESCO 2015
Type: Extended abstract
Authors
1 USDA Agricultural Research Service, Horticultural Crops Research Unit, 29603 U of I Lane Parma, ID 83660 U.S.A.
2 USDA Agricultural Research Service, Northwest Irrigation and Soils Research, 3793 N. 3600 E Kimberly, ID 83341 U.S.A.
Contact the author*
Keywords
crop water stress index, leaf water potential, neural network modeling, infrared thermography