Articles
Enhancing sustainable water management: utilizing UAV-based NIR/SWIR hyperspectral imaging to evaluate grapevine water status in a variably irrigated vineyard
Article number
1395_9
Pages
61 – 66
Language
English
Abstract
In 2022, a study was initiated, focusing on Cabernet Sauvignon vines situated in the San Joaquin Valley.
The objective was to investigate the potential of remote sensing as a critical element in sustainable water management, particularly in capturing spectral and spatial information.
To conduct the research, an automated irrigation system was implemented, facilitating variable irrigation across 48 distinct watering zones.
These zones covered 12 different irrigation regimes, each replicated 4 times in a randomized manner.
Throughout the entire growth season, spectral information was gathered from all zones using unmanned aircraft vehicles (UAVs) equipped with hyperspectral imaging capabilities, specifically in the near infrared (NIR) and short-wave infrared (SWIR) range spanning from 900 to 1700 nm.
These wavelengths include water absorption bands, utilized in machine-learning regression models to predict plant water status.
To validate the spectral information accuracy, plant water status was concurrently measured using techniques such as stem water potentials (Ψstem) and leaf gas exchange.
Measurements were taken every two weeks, commencing in June until harvest, resulting in a total of 5 flights and approximately 250 individual readings in 2022. During data analysis, highly accurate segmentation methods (>99% accuracy) were employed to extract the pure canopy signal from the images.
This extracted information was then used to train machine learning models for predicting water status measurements.
The preliminary model for 2022 achieved a coefficient of determination (R2) of 0.54 and a root mean square error (RMSE) of 0.11 MPa through a 5-fold cross-validation routine for the prediction of Ψstem.
This project signifies a significant step toward the development of innovative methods for precise monitoring and management of irrigation in vineyards.
The objective was to investigate the potential of remote sensing as a critical element in sustainable water management, particularly in capturing spectral and spatial information.
To conduct the research, an automated irrigation system was implemented, facilitating variable irrigation across 48 distinct watering zones.
These zones covered 12 different irrigation regimes, each replicated 4 times in a randomized manner.
Throughout the entire growth season, spectral information was gathered from all zones using unmanned aircraft vehicles (UAVs) equipped with hyperspectral imaging capabilities, specifically in the near infrared (NIR) and short-wave infrared (SWIR) range spanning from 900 to 1700 nm.
These wavelengths include water absorption bands, utilized in machine-learning regression models to predict plant water status.
To validate the spectral information accuracy, plant water status was concurrently measured using techniques such as stem water potentials (Ψstem) and leaf gas exchange.
Measurements were taken every two weeks, commencing in June until harvest, resulting in a total of 5 flights and approximately 250 individual readings in 2022. During data analysis, highly accurate segmentation methods (>99% accuracy) were employed to extract the pure canopy signal from the images.
This extracted information was then used to train machine learning models for predicting water status measurements.
The preliminary model for 2022 achieved a coefficient of determination (R2) of 0.54 and a root mean square error (RMSE) of 0.11 MPa through a 5-fold cross-validation routine for the prediction of Ψstem.
This project signifies a significant step toward the development of innovative methods for precise monitoring and management of irrigation in vineyards.
Authors
E. Laroche-Pinel, K. Vasquez, G. Partida, L. Brillante
Keywords
imaging spectroscopy, machine-learning, precision viticulture, variable rate irrigation, Vitis vinifera L
Groups involved
- Division Temperate Tree Nuts
- Division Vine and Berry Fruits
- Division Temperate Tree Fruits
- Division Precision Horticulture and Engineering
- Division Plant-Environment Interactions in Field Systems
- Division Tropical and Subtropical Fruit and Nuts
- Working Group Precision Management of Orchards and Vineyards
Online Articles (59)