Toward the improved use of remote sensing and process modeling in California's premium wine industry

摘要

Californian premium winegrowers are making increasing use of optical remote sensing as an additional tool for monitoring canopy density and managing vineyards. In particular, high spatial resolution (2m) vegetation index imagery has been shown to be useful for subdividing individual fields ("blocks") for harvest based upon end-of-season vigor, as inferred by canopy density. Block segmentation can result in more uniformly mature wine "lots" and, in some cases, ultimately in improved wine quality. In partnership with the wine and commercial remote sensing industries, NASA/Earth Science Enterprise investigators continue to examine relationships among vine stress, canopy development, and resulting wine quality by combining remote sensing with an agro-ecosystem process model adapted from BIOME-BGC. The model, which predicts fluxes of water and carbon, uses remotely sensed leaf area index (LAI) to modulate photosynthesis and transpiration across the landscape. The modeling framework potentially enables improved specification of irrigation and nutritional requirements for greater block uniformity. Landscape analysis represents a departure from much of the vineyard remote sensing application to-date, which has tended to emphasize relative canopy differences within a particular block. As an initial step in this direction, the authors seek to determine the robustness of remote sensing for retrieving LAI across different blocks in the face of such potential confusion factors as trellis system, sun/view angle, topography, grape variety, soil type/brightness, and atmosphere.