Vineyard precision farming by means of satellite data: optical and polarimetric radar data safeguarding the Frascati D.O.C. area

Some decades of observations have demonstrated the usefulness of spaceborne optical sensors in agriculture. Progress in spatial, spectral, and temporal resolutions brought in by recent satellites, like QuickBird and Ikonos, Spot-5 or Kompsat-2 is resulting beneficial to precision farming practices (Arkun et al., 2005). Imagery from space spots spatial and temporal anomalies of crop vigor and biomass and allows timely remedial treatments (Johnson, 2004). Likewise, the crop response to the adopted local agronomic strategies is revealed (Johnson et al., 2004), while preserving the capability of monitoring land use at regional level. In several developed countries, customized services, such as grid soil sampling, yield mapping, variable rate application of fertilizers and pesticides, are now available (Srinivasan, 2006). Especially in the last decade, some premium-wine producers approached space remote sensing technology with the main intent of enhancing the quality of their product. Indeed, while the cellar practices are being increasingly controlled and automated, on the field the operations mainly rely on individual skills of personnel and the quality of grapes remain subject to climatic variability. Leaf Area Index (LAI) of vines is related to fruit ripening rate, disease incidence and grape quality (Winkler, 1958) and the phenological state is crucial for the timing of the correct cultivation practices and of harvest (Moran et al., 1997). Soil Moisture Content (SMC) maps provide additional information for managing pesticides and fertilizations or, where allowed, for watering. Such parameters can be monitored by space-borne sensors and effectively handled by Geographic Information Systems (GIS) (Pearson et al., 1997a). Hence, space Earth Observation technology begins to look an attractive element of the wine producing chain. Since 2000, an airborne hyperspectral system (ITRES, 1996) has been deployed in Australia as a support to the wine industry (Cochrane, 2000). Several experiments took place in California, U.S.A., to assess the serviceability of optical images from Unmanned Aerial Vehicle (UAV) (Johnson et al., 2003), or from space Johnson et al. (2004). In Europe, airborne remote sensing has been proposed for the detection of dead vine trees Chanussot et al. (2005). In Italy, an experiment was carried out in 2005 in the Franciacorta wine producing area based on the retrieval of the Normalized Vigor Index (NVI) from multispectral images acquired by IKONOS Brancadoro et al. (2006) The potential of SAR images in monitoring the development and conditions of vegetation has been investigated in a number of studies e.g., Brisco and Brown (1998), Ferrazzoli (202),Toan et al. (1997), Della Vecchia et al. (2008). Specific analyses of radar images of vineyards are scarce in literature and manly focused on the effect of the periodicity in rows and vine supporting structures (Lewis et al., 1999). Indeed, the radar sensitivity to tree biomass stems from the different volumes of woody matter which affect the wave-plant-soil interaction mechanisms. In their development cycles, crops like maize, sun ower, colza, or alfalfa, considerably change the number, dimensions and shapes of the scattering elements, thus modifying their radar return. The situation is different when the plants have a stable woody structure which is only slightly modified by developing twigs, leaves and fruits. In this case, monitoring biomass evolution by radar is made difficult by the small variations of backscattering with respect to a strong, almost saturated (at least at C-band) background. Measurements on vineyards are even more difficult, given the high number of poles and metallic wires supporting the runners. For given radar frequency, polarization and observation angle, the backscattering depends on several parameters, including slope of the imaged vineyard, soil roughness, soil moisture content, weeds, cultivation aspect, geometry of supporting structures, plant type and state, and, finally, on the fruit biomass. In an attempt to gain some insight on the potential of SAR in monitoring vineyards, and, especially, grape biomass, several experiments have been carried out by means of different sensors, both optical and Radar, at several frequencies, ground resolutions and different polarimetric mode. Moreover, ground observations have been carried out to better understand the complex interaction between SAR and vineyards. After a first presentation of the state of the art on vineyard precision farming, several experiments by means of radar instruments will be presented and results critically discussed. Finally, a precision farming system prototype is presented.