Potential field detection of Flavescence dorée and Esca diseases using a ground sensing optical system

• Published in: Biosystems engineering Vol. 215; pp. 203 - 214
• Main Authors: Daglio, G., Cesaro, P., Todeschini, V., Lingua, G., Lazzari, M., Berta, G., Massa, N.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2022
• Subjects: Ground sensing; NDVI; Precision viticulture; Vegetation indices; VIS/NIR; Yellows diseases; VIS/NIR; Precision viticulture; Ground sensing; NDVI; Vegetation indices; Yellows diseases
• ISSN: 1537-5110; 1537-5129
• DOI: 10.1016/j.biosystemseng.2022.01.009

Every year, in Italy many new cases of Flavescence dorée (FD) and Esca disease (ED) are recorded in Vitis vinifera, resulting in plant death and reduced production. FD is caused by phytoplasmas and its main symptoms comprise growth reduction, leaf reddening and downward rolling. The vines affected by ED, a fungal disease, show foliar inter-venal necrosis and chlorosis. The aim of this work was to test a commercial optical sensor for the detection of FD and ED in plants. Three vineyards of the red grapevine Dolcetto cv. were mapped and plants showing FD or ED symptoms were georeferenced. In July 2015 and 2016, monitoring by optical sensors (OptRx®) was undertaken and leaf samples from plants were collected to confirm the presence of FD phytoplasmas by molecular analyses. The OptRx® records the reflectance and calculates two indices: normalised difference vegetation index (NDVI) and normalised difference red edge index (NDRE). A visible/near infrared (VIS/NIR) index was calculated from the two reflectances. Data were processed by either statistical analysis of raw data, to detect differences in the plant health status and between the two years, or by using filters to evaluate sensitivity and specificity of the sensor in detecting the diseases and create prediction indices. Diseased plants showed lower NDVI and NDRE values than healthy plants. A negative signal from the filter system was always associated with healthy plants. Further studies are needed in order to understand the most appropriate threshold and to implement the diagnostic algorithm to assess the presence of disease. •Diseased plants showed lower values of NDVI and NDRE than healthy ones.•Diseased plants presented higher values of VIS/NIR compared to healthy ones.•Index absolute values were highly variable across vineyards and years.•A normalised moving average metric was used to detect problem areas in the vineyard.•The employed algorithm correctly classified all healthy plants.