Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Data-Driven Models for Canopy Temperature-Based Irrigation Scheduling

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  Transactions of the ASABE. 63(5): 1579-1592. (doi: 10.13031/trans.13901) @2020
Authors:   Bradely A. King, Krista C. Shellie, David D. Tarkalson, Alexander D. Levin, Vivek Sharma, David L. Bjorneberg
Keywords:   Canopy temperature, Crop water stress index, Irrigation scheduling, Leaf water potential, Sugarbeet, Wine grape.


Artificial neural network modeling was used to predict crop water stress index lower reference canopy temperature.

Root mean square error of predicted lower reference temperatures was <1.1°C for sugarbeet and Pinot noir wine grape.

Energy balance model was used to dynamically predict crop water stress index upper reference canopy temperature.

Crop water stress index for sugarbeet was well correlated with irrigation and soil water status.

Crop water stress idex was well correlated with midday leaf water potential of wine grape.

Abstract. Normalized crop canopy temperature, termed crop water stress index (CWSI), was proposed over 40 years ago as an irrigation management tool but has experienced limited adoption in production agriculture. Development of generalized crop-specific upper and lower reference temperatures is critical for implementation of CWSI-based irrigation scheduling. The objective of this study was to develop and evaluate data-driven models for predicting the reference canopy temperatures needed to compute CWSI for sugarbeet and wine grape. Reference canopy temperatures for sugarbeet and wine grape were predicted using machine learning and regression models developed from measured canopy temperatures of sugarbeet, grown in Idaho and Wyoming, and wine grape, grown in Idaho and Oregon, over five years under full and severe deficit irrigation. Lower reference temperatures (TLL) were estimated using neural network models with Nash-Sutcliffe model efficiencies exceeding 0.88 and root mean square error less than 1.1°C. The relationship between TLL minus ambient air temperature and vapor pressure deficit was represented with a linear model that maximized the regression coefficient rather than minimized the sum of squared error. The linear models were used to estimate upper reference temperatures that were nearly double the values reported in previous studies. A daily CWSI, calculated as the average of 15 min CWSI values between 13:00 and 16:00 MDT for sugarbeet and between 13:00 and 15:00 local time for wine grape, were well correlated with irrigation events and amounts. There was a significant (p < 0.001) linear relationship between the daily CWSI and midday leaf water potential of Malbec and Syrah wine grapes, with an R2 of 0.53. The data-driven models developed in this study to estimate reference temperatures enable automated calculation of the CWSI for effective assessment of crop water stress. However, measurements taken under conditions of wet canopy or low solar radiation should be disregarded as they can result in irrational values of the CWSI.

(Download PDF)    (Export to EndNotes)