New technologies for dynamic canopy and disease management

Abstract

The grapevine canopy is the site of photosynthesis, the process whereby the energy for growth, fruit production and winter storage, is produced. Further, the canopy size and structure play a major role in determining the environment around the grape bunch, which greatly influences grape composition and, therefore, the value for winemaking. At present, determining the appropriate canopy management is a decision the viticulturist has to make with limited access to tools for objective assessment of canopy size or structure.

The canopy is also the recipient of most of the agricultural sprays used in viticulture, with protection against powdery mildew the most common reason for spraying. A growing season spray program is typically designed with little knowledge of the upcoming disease pressure and is, therefore, used to ensure that disease does not occur, potentially resulting in more sprays than necessary in a given season.

This project was established to test and develop technologies to provide non-contact measurements of canopy size and structure, canopy nutrition, canopy function and pre-symptomatic detection of disease. In particular, the project objectives included: assessing methods to determine the 3D structure of the canopy, data analysis methods that use this to provide growers with summary measures that aid canopy management decisions, hyperspectral imaging of the canopy to determine and map nutritional status and canopy functional status, hyperspectral tools to detect powdery mildew prior to it being visible to the human eye, and an assessment of the readiness of these technologies for commercialisation in viticulture.

Summary

This project was established to test and develop technologies to provide non-contact measurements of canopy size and structure, canopy nutrition, canopy function and pre-symptomatic detection of disease. In particular, the project objectives included: assessing methods to determine the 3D structure of the canopy, data analysis methods that use this to provide growers with summary measures that aid canopy management decisions, hyperspectral imaging of the canopy to determine and map nutritional status and canopy functional status, hyperspectral tools to detect powdery mildew prior to it being visible to the human eye, and an assessment of the readiness of these technologies for commercialisation in viticulture.

LiDAR, stereo-imaging, photogrammetry (structure-from-motion) from consumer drone and photogrammetry from consumer action-cams were all used to collect 3D data on vine canopies and structure. In addition, consumer action-cams were also used to estimate canopy size from continuous video. All these devices (except the drone) were mounted on commercial vineyard vehicles and used to image commercial vineyards in collaboration with a number of wine companies (Accolade Wine, Katnook Estate, Rymill Wine). Bespoke data-processing pipelines were developed in all cases to produce grower relevant data from the sensor outputs.

Hyperspectral imaging of the same vineyards from fixed wing aircraft was undertaken by Fraunhofer IFF with Airbourne Research Australia, using a bespoke imaging system built by Fraunhofer IFF. Additionally, a lab-based hyperspectral imaging system was used to develop detailed calibrations of vine nutrition against leaf images.

The lab hyperspectral system, mounted in a glasshouse environment, was used to image grapevines with various stages of powdery mildew infection and calibrations against disease status developed. Finally, the lab-based system was used to image grapevine leaves, for which key physiological traits had been measured and preliminary calibrations produced.

The most viable sensor and data analytic packages were developed using LiDAR to determine canopy structure and consumer drone imagery with photogrammetry or consumer action-cam imagery to determine canopy size. Hyperspectral data analytics were able to reliably identify vine tissue types from spectra and then analyse those spectra for relevant traits, such as nutritional status, using calibrations developed through the project. Proof of concept calibrations against vine function and disease status were also established, although pre-symptomatic detection of disease was not clearly demonstrated.

It is recommended that each of these technology areas be assessed, with further industry input, for commercialisation potential, particularly considering the dollar value of the data to viticulturist decision making. For example, if the industry values a key canopy parameter as being cost effective at $50 per hectare, then the potential to produce that data at that cost needs to be demonstrated.

For those systems at a lower technology readiness where this assessment it not yet practical, it is recommended to continue research where there is ongoing industry support.

This project was supported by Wine Australia, through funding from the Australian Government Department of Agriculture, Water and the Environment as part of its Rural R&D for Profit program, CSIRO and the Fraunhofer Institute for Factory Operation and Automation.