Forecasting severe grape downy mildew attacks using machine learning
Problem
Grape downy mildew in Bordeaux vineyards requires repetitive fungicide spraying (average 10.1 treatments in 2013); growers need to identify heavily attacked vineyards to target treatments.
Why it matters
Reducing fungicide treatments is a major environmental and public health issue.
What the AI did
Using 9 years of disease observations, the authors developed and compared generalized linear models and machine learning algorithms predicting the probability of high GDM incidence and severity from disease onset date and monthly temperature/precipitation.
Results
not stated in source (abstract truncated before performance figures)
Impact
Approach could target fungicides to high-risk vineyards, but model performance is not stated in the abstract.
Record
| Venue | PLoS ONE |
|---|---|
| Published | 2020 |
| Work type | article |
| Authors | Mathilde Chen, François Brun, Marc Raynal, David Makowski |
| Citations | 68 (OpenAlex, at import time) |
| References | 73 |
| Model or method | generalized linear models and machine learning algorithms; specific types not stated |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1371/journal.pone.0230254 |
| Open access | gold |
| Keywords | grape downy mildew, machine learning, forecasting, fungicide reduction, viticulture |
Importance rating
real but narrower advance
Abstract calls reducing fungicide treatments a major environmental and public health issue for Bordeaux vineyards; regional agricultural scope, not global-scale quantified impact.