Scalable grapevine viral disease detection with AVIRIS imaging spectroscopy
Abstract
Plant disease causes losses greater than $220 billion annually to global agriculture. Grapevine (Vitis spp.) Leafroll Virus Complex 3 (GLRV) causes >$3 billion in damages and losses to the United States grape and wine industry, the countrys most economically important fruit crop. These losses are difficult for growers to manage, as the disease has a latent period of approximately 12-18 months before foliar symptoms become apparent despite negatively impacting fruit quality during this time. Making matters worse, once infected, there is no cure, and the only effective treatment is eradication, or vine removal, to avoid further spread. Tools for early detection of plant diseases such as GLRV are urgently needed. Here, we use airborne spectroscopic data collected in September, 2020 by NASAs Airborne Visible/Infrared Imaging Spectrometer Next Generation (AVIRIS-NG) for early detection of GLRV in California vineyards. In collaboration with industry and academic colleagues, 280 acres of Aglianico, Cabernet Sauvignon, and Petite Sirah grapes were scouted for disease symptoms at peak time of year and a subset were collected for confirmation testing. Multiple Endmember Spectral Mixture Analysis (MESMA) and Colorspace2 were used to prepare data for analysis. Our preliminary analysis with a subset of this data and a Random Forest classifier identified spectral regions that allowed for differentiation between healthy and GLRV infected vines at 10m resolution with 72% accuracy. We hypothesize the spectral differences that facilitate differentiation are linked to inherent disease physiology and will spatially scale as has been seen with other physiology-linked processes. Our long term goal is to develop a generalizable, rapid, scalable, and non-destructive early disease detection method for eventual use with NASAs forthcoming Surface Biology and Geology satellite mission.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFM.B55K1325R