Mapping vineyard leaf area with multispectral satellite imagery

Vineyard leaf area is a key determinant of grape characteristics and wine quality. As is frequently the case in agriculture, available ground-based leaf area measurements employed by growers are not well suited to larger area mapping. In this study, IKONOS high spatial resolution, multispectral satellite imagery was used to map leaf area throughout two commercial wine grape vineyards (approximately 800 ha) in California’s North Coast growing region. The imagery was collected near harvest during the 2000 growing season, converted to at-sensor radiance, geo-referenced and transformed to normalized difference vegetation index (NDVI) on a per pixel basis. Measurements at 24 ground calibration sites were used to convert NDVI maps to leaf area index (LAI; m leaf area m 2 ground area); planting density was then used to express leaf area on a per vine basis (LAv). Image-based LAv was significantly correlated with ground-based LAv estimates developed at 23 validation sites (r 2 /0.72; P B/ 0.001). Despite challenges posed by the discontinuous nature of vineyard canopies and architectural differences imposed by shoot positioning trellis systems, remote sensing appears to offer a basis for mapping vineyard leaf area in low LAI vineyards. Such maps can potentially be used to parameterize plant growth models or provide decision support for irrigation and canopy management. # 2002 Elsevier Science B.V. All rights reserved. Corresponding author. Present address: Earth Science Division, NASA Ames Research Center, M.S. 242-4, Moffett Field, CA 94035, USA. Tel.: /1-650-604-3331; fax: /1-650-604-4680 E-mail address: [email protected] (L.F. Johnson). Computers and Electronics in Agriculture 38 (2003) 33 /44 www.elsevier.com/locate/compag 0168-1699/02/$ see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 1 6 9 9 ( 0 2 ) 0 0 1 0 6 0