Yield prediction assessment before bloom and at veraison in a cv. Airén high yielding vineyard in Toledo (La Mancha, Spain)

Abstract

Anticipation in the possible responses of grapevines to environmental variations is key to adjust field work in view of a more effective management. This idea has been the driving force behind the current work, which seeks to understand the interaction patterns of the vine with its habitat throughout the growing cycle.

The experimental site is located at ‘Viñedos del Río Tajo’ company. It is a complex of 250 hectares of irrigated vineyards of cv Airén, Vitis vinifera, L., located in Guadamur (Toledo, Spain). Vine spacing is 2.0 x 3.3 m2, and the training system is a single curtain with minimal pruning, where the cordon is 1.80m above the ground.

The study is based on data collected throughout 2023 and 2024 from 10 plots. Within these plots, 3 to 4 replicates of 25 plants each were sampled. In each replicate, two measurements were taken in consecutive rows, each covering 2 meters of row, corresponding to one plant.

The number of clusters was collected at separated clusters phenological phase. Berry weight and the number of berries per cluster were taken at veraison. Yield partitioning was determined at harvest along 4 m of row per replication. Other parameters, such as plant water status, surface area, pruning weight, and shoot weight were also measured.

In addition, 100 berries per replication were used for basic must analysis (SST, pH, titratable acidity). Besides, YAN (yeast available nitrogen) and K+ content was also determined.

Based on the geospatial, climatic, and phenological data measured, the purpose is to identify the main factors influencing the yield. Up to now, the global yield, expressed as the product (yield/m of row · SST), has mainly been explained by surface area/kg, number of clusters per meter of row, average shoot weight, and number of berries per cluster, arranged in order of significance (R2=0.87**). Due to the high proportion of variance explained by these variables, which we can already know in spring, we perform an early prediction of the yield based on the number of clusters per meter row, the number of shoots per cluster and the average cluster weight.

This work highlights the importance of integrating spatial and temporal analysis tools into decision-making in the wine sector, and presents a replicable methodological framework for other vine crops with similar objectives.