Decision-support system for water stress assessment and deficit irrigation management in wine grapes

Chenchen Kang

doi:10.7273/000005007

A timely and appropriate level of water deficit is desirable in wine grape production to optimize fruit quality for winemaking. Regulated deficit irrigation (RDI) is an irrigation management strategy which applies less water than the full water requirement in some growing phases (e.g., from fruit set to veraison) to achieve a mild to moderate water stress. The implementation of RDI in wine grape production requires a combination of technical knowledge, accurate assessment and monitoring, and effective decision-making to achieve the desired balance between water stress and adequate water availability to the plants. This research aimed to develop and validate a comprehensive decision-support system for precision RDI management in vineyards. The proposed system was aimed to accurately assess the soil and plant water status through hyperspectral imaging (HSI). By doing so, the system sought to provide appropriate irrigation plans to achieve the desired soil water content threshold optimally.This research comprised three studies aimed at developing and validating several data-driven models as decision-support tools for precision deficit irrigation management in wine grapes. The first two studies focused on developing ground-based approaches to detect grapevine water status using HSI. The first study aimed to develop ground-based approaches for detecting soil and grapevine water status using HSI obtained in diffused lighting conditions. It was found that using spectral data obtained under diffused light resulted in improved model performance compared to using spectral data obtained under direct sunlight. This allowed for high-resolution sensing of grapevine water status by estimating leaf water potential and stomatal conductance. The second study fused HSI with 3D point clouds to address the effect of varied leaf orientations and enabled a Multiblock Partial Least Squares-based model to estimate leaf water potential with high accuracy. The third study aimed to develop a decision-support system for managing precision RDI in vineyards. The system consists of a soil moisture prediction model and an RDI scheduling model developed based on artificial neural networks. Validation tests showed that the soil moisture prediction model could predict the soil moisture in the following week with an R2 of 0.93 and RMSE of 0.86 %, and the RDI scheduling model could estimate the weekly irrigation water amount for maintaining a target soil moisture with an R2 of 0.94 and RMSE of 8.85 L per drip irrigation emitter. These studies contribute to developing efficient data-driven approaches to assess grapevine water status and optimize deficit irrigation plans for dynamic soil water threshold. The outcomes of this research could aid in achieving a balance between yield and fruit quality in wine grape production.

Decision-support system for water stress assessment and deficit irrigation management in wine grapes

Files and links (1)

Abstract

Metrics

Details