Submitted to: American Journal of Enology and Viticulture
Publication Type: Peer reviewed journal
Publication Acceptance Date: 6/6/2006
Publication Date: 11/1/2006
Citation: Stamatiadis, S., Taskos, D., Tsadilas, C., Christofides, C., Tsadila, E., Schepers, J.S. 2006. Relation of ground-sensor canopy reflectance to biomass production and grade color in two merlot vineyards. American Journal of Enology and Viticulture 57:415-422. Interpretive Summary: Understanding spatial variability in grape vine growth and grape productivity is difficult, but important because these factors influence wine quality. Aircraft imagery is costly and is frequently not very timely in terms of making important management decisions. Tractor-mounted, four-band reflectance sensors were positioned above the canopy. Reflectance data collected near veraison (time of color change in the grapes) showed a good relationship with pruning weight. This relationship was better when averaged across four plants than for single plants. A negative correlation between canopy reflectance and anthocyanin content of grapes was significant in one of the two vineyards and implied an inverse relationship between biomass production and grape color. These results demonstrate the potential value of proximal remote sensing for the application of site-specific management of vineyards in order to optimize production, improve wine quality and reduce chemical inputs.
Technical Abstract: Recent studies with optical remote sensing have demonstrated the relationship between canopy reflectance, biomass production and certain quality attributes of grapes in red winegrape vineyards. Multispectral reflectance data are currently delivered by airborne platforms at a cost and may not be available to producers in time for critical management decisions to be implemented. Ground-based sensors are an emerging technology designed to overcome many of the limitations associated with satellite- or aircraft-based sensing systems. This study provides information about the potential of ground-based canopy sensors in predicting biomass production and quality attributes of grapes in two differing Merlot vineyards. The multispectral sensors were mounted on a tractor vehicle and recorded canopy reflectance from two different viewing angles and fields of view along selected rows of vines. The normalized difference vegetation index (NDVI) was compared to pruning weight, phenol, anthocyanin and sugar content of grapes measured in 25 to 32 sampling positions within each field over two growing seasons. Sensor canopy reflectance was able to predict the spatial variation of biomass production in the two vineyards with varying degrees of precision. A nadir viewing angle of the canopy near veraison provided estimates of NDVI that were better predictors of biomass production, while masking the sensor optics provided more reliable estimates of canopy reflectance. The quadratic relationship between NDVI and pruning weight improved with decreasing sensor resolution (from one plant to four plants). The modified Wide Dynamic Range Vegetation Index (WDRVI) reduced saturation effects and improved the linearity between canopy reflectance and biomass production. A negative correlation between canopy reflectance and anthocyanin content of grapes was significant in one of the two vineyards and implied an inverse relationship between biomass production and grape color. These results demonstrate the potential value of proximal remote sensing for the application of site-specific management of vineyards in order to optimize production, improve wine quality and reduce chemical inputs.