The relationship between the vineyard environment and within-block variability in winegrape yield and composition

The costs associated with the spatial and temporal variation in yield and composition are a major issue for the sustainability of the Australian wine industry. The aim of this study was to identify the elements of the vineyard environment that drive within-block variation in grape yield and composition. The spatial structure, magnitude and temporal stability of grape yield and composition were measured in three case study vineyard blocks over a period of three vintages. The spatial patterns in the air and soil temperature, net radiation, topography and soil physical characteristics were also defined and the relationship between vine performance and the vineyard environment analysed. The variability was characterised in three variable terrain blocks in Victoria Australia, a 3.4 ha block of Chardonnay in the Yarra Valley Wine Region, a 6.7 ha block of Chardonnay in the Strathbogie Wine Region and a 6.9 ha block of Pinot Noir in the Mornington Peninsula Wine Region. Data on the grape yield, composition and vineyard environment were collected during the 2003, 2004 and 2005 vintages. A harvester mounted yield monitor was coupled with a GPS to collect high spatial resolution yield data. Between 90 and 111 grape samples were taken on a regular grid in each block to assess variation in the yield components and grape composition. The soil physical characteristics were examined through surface soil moisture surveys and soil pits located according to topography and a survey of apparent electrical conductivity (ECa). Air and soil temperature were recorded every 20 minutes at seven sites within each block at a height of 1.2 m within the vine canopy and a depth of 10 cm. The grape yield and sample point data were mapped using local and global kriging and a Spline function was used to interpolate the temperature data. Multiple regression models were produced to characterise the relationship between the soil physical characteristics, topography and the soil ECa. These models were then used to produce high resolution maps of soil variability. Spatial variation in yield observed in this study was in the order of three- to tenfold with a within-block range of between 5.3 and 16.5 t/ha. The magnitude of within vineyard variability in yield increased with increasing mean block yield suggesting that management decisions to increase yield may come at a cost of increasing variation. The most significant finding to come from an investigation into the relationship between spatial and temporal variability in yield is that temporal variability is not i spatially homogeneous within vineyard blocks. The definition of patterns in temporal variability in yield revealed that some areas of the blocks remained stable from year-to-year despite significant changes in the mean yield of the block. In some cases it was only small sections of the block that was driving variation in yield from one year to the next. This finding has important implications for the management and forecasting of vintage variability in yield revealing the potential to focus on smaller sections of blocks.