Assessing Spatial Variability of Grape Skin Flavonoids at the Vineyard Scale Based on Plant Water Status Mapping
Luca Brillante, Johann Martinez-Luscher, Runze
Yu, and Kaan Kurtural*
*University of California Davis, 1 Shields Avenue, Davis, CA
95616 (skkurtural@ucdavis.edu)
Plant water stress affects grape (Vitis vinifera L. cv. Cabernet Sauvignon) berry composition and is variable in space due to variations in the physical environment at the growing site. We monitored the natural variability of grapevine water stress using stem water potential (Ψstem) and leaf gas exchange in an equi-distant grid in a commercial vineyard. Spatial differences were measured and related to topographical variation by modeling. Geospatial analysis and clustering allowed us to differentiate the vineyard block into two distinct zones with severe or moderate water stress that varied by 0.2 MPa. Differences in stem water potential affected stomatal conductance, net carbon assimilation, and intrinsic water use efficiency, all of which were different on all measurement dates. The two zones were sampled selectively at harvest for measurements of berry chemistry. Water status zone did not affect berry mass or yield per vine. There was significant difference in total soluble solids (3.56 Brix) and titratable acidity, indicating a direct effect of water stress on ripening acceleration. Berry skin flavonol and anthocyanin composition and concentration were measured by C18 reversed-phased high-performance liquid chromatography (HPLC). The anthocyanins were most affected by the two water stress zones. Dihydroxylated anthocyanins were more affected than trihydroxylated; therefore, the ratio of the two forms increased. Flavonols were different in total amounts, but hydroxylation patterns were not affected. Proanthocyanidin isolates were characterized by acid catalysis in the presence of excess phloroglucinol, followed by reversed-phase HPLC. Proanthocyanidins showed the least significant difference, although (+)-catechin terminal subunits were important predictors in a partial least square model used to summarize the multivariate relationships, predicting Ψstem or the management zone. The results provide fundamental information on vineyard water status that could discriminate harvest or direct vineyard operators to modify irrigation management to equilibrate berry composition at harvest.
Funding Support: USDA-NIFA Specialty Crops Research Initiative