Use of Aerial Imaging to Evaluate Vineyard Canopy Variability

A good deal of attention has been paid to the light environment of grapevine canopies. Research on the physiological responses to light microclimate has been extensive. Studies have shown that vine yield, berry composition, and wine quality all improve with increased fruit and leaf exposure to sunlight (Dokoozlian 1990; Kliewer 1982; Smart 1985). From these findings have come the impetus for many of the canopy management practices used today. These include shoot positioning, hedging, leaf thinning, and canopy division. These activities are performed in order to optimize the canopy light microclimate environments. Grapevine light microclimate is effected by the structure and density of the canopy components including leaves, shoots, and woody material. Measurements of the grapevine canopy are difficult because of the high spatial and temporal variation within the canopy. Indices pertinent to isolated plants and row crops have been applied to grapevines. These measurements include characterizing canopies as geometrical shapes with given height, length, width, area, and volume measurements. A second approach measures leaf area and leaf density on a per vine, unit length, or unit area basis. Leaf area index (LAI; leaf area per unit of soil surface area) is one of the commonly used indices of this type. Measurements of leaf area and leaf area density (LAD; leaf area per unit volume) are used as indicators of vine vigor and as inputs into models of evapotranspiration, whole-vine photosynthesis, or sunlight penetration. Measurements of leaf area are time consuming and labor intensive because of the inherent variability found within a vine and on a larger scale within a vineyard. Some methods are destructive and can not be applied to different treatments. There are a number of techniques that are based on an interactive relationship between canopy structure and radiation interception. These techniques measure the gap fraction, i.e. the proportion of light which is not blocked by foliage in a range of azimuthal directions. Leaf area is estimated using mathematical models with the gap fraction as an input parameter. These mathematical models often assume a random foliage distribution which rarely occurs (Ollat, 1998). In addition, light sensors need to be placed within and above the grapevine canopy in order to measure the gap fraction. This limits the scale of sampling and its utilization within an commercial vineyard setting. There is a need for alternate methods of estimating leaf area that are both cost effective and accurate. 21 Project Title: Use of Aerial Imaging to Evaluate Vineyard Canopy Variability (Continued) Advances in the field remote sensing present possible methods for evaluating vine canopy characteristics including leaf area density and canopy architecture. One method involves the use of aerial image collection and analysis. The most commonly available method of aerial image acquisition is the use of four-band, digital, multi-spectral image data that are collected by an airplane (or satellite) and processed to produce a measure (vegetation index) of canopy density. Calculating this index is based on the principle that photosynthetically active vegetation shows high absorption of incident sunlight in the visible red (R) wavelengths and strong reflectance in the near-infrared wavelengths (NIR). These spectral properties are distinctive from that of soil and water, the other two predominant landscape features. There are many useful algorithms for computing vegetation indices with the most common falling into a class of ratio indices [RVI = NIR/R; NDVI = (NTR-R)/(NIR+R)]. Vegetation indices derived from these two spectral bands have been shown to correlate highly with green leaf area index, chlorophyll content, photosynthetically active biomass, vegetation density, photo synthetic rate, percent ground cover by vegetation, and grain or forage yield (Weigand et al. 1991). Studies utilizing this technology have been performed mostly in closed canopy systems such as grasslands and contiguous forest. No published study has been found where the system was a vertically shoot positioned vineyard. Vineyards do not exhibit a closed canopy system. A significant percentage of a vineyard is comprised of vineyard middles and has the spectral properties of bare soil and/ or cover crops. The influence of varying spectral backgrounds has not been studied in the context of vineyard systems. In addition, grapevine foliage distribution is not entirely understood. Canopy leaf area, density, and structure are managed within the confines of given trellising systems using practices such as shoot positioning, leaf pulling, and hedging. The spectral response, if any, of changing the canopy density using any of these practices is not understood. These factors must be taken into account when applying remote sensing technologies to vineyard management.