Spatial variability of soil degradation and yield quality traits in a marginal vineyard in Southern Italy. 

Farm-scale spatial variability of grapevine yield and related berry quality traits can potentially cause decreasing profitability. Marginal land often presents degraded soils at variable extents, which might differently respond to changing climatic conditions. Hence, analyzing the relationship between relevant soil properties and berry quality on the vineyard spatial scale, as well as the grape response to different climates might help to optimize management practices and precision agriculture interventions.To test the above relationships, a two-year (2023 and 2024) experiment was conducted in a commercial vineyard (Vitis vinifera L, cv Chardonnay/1103P, 0.9 × 2.7 m) located in a 16% sloping hill in Southern Italy (80 m asl, 40°08’29”N and 16°37’06”). Meteorological variables were recorded throughout the season by means of a standard weather station located ~ 0.3 km far from the vineyard. Soil zoning at vineyard scale was obtained by performing a proximal soil survey with an electro-magnetic induction (EMI) sensor. Georeferenced measures of the apparent electrical conductivity (ECa) were processed using Ordinary Kriging implemented in the QGIS software to produce continuous maps of ECa. The EMI maps, associated with the aspect and slope maps, were used as input for a clustering algorithm to identify homogeneous zones (HZs) within the vineyard. The results of the K-means clustering enabled the identification of 3 HZs that were analysed for soil properties, plant and grape characteristics. Fifty-two soil samples, collected at , were analysed for the assessment of the main chemical and physical properties (i.e., pH, organic carbon content, total carbonates, texture, cation exchange capacity, exchangeable bases, etc). Significant correlations were found between EMI data and several soil properties (including fine sand, Na and Mg exchangeable content, cation exchange capacity, pH), supporting soil zoning based on the ECa values. Over the two surveyed years, shoot and cluster number and mass, as well as pruning mass data were collected. At harvest (16th and 8th of August in 2023 and 2024, respectively), yield was determined on a total of 45 vines (15 per HZ), and then berry subsamples (~ 40 per vine) were used for grape quality traits determination (berry fresh mass, total soluble solids-TSS, pH and titratable acidity-TA). A factorial ANOVA was performed considering the grape characteristics as dependent variables and both the HZs and vintages as categorial factors. Over the two years, results highlight significant differences in grape characteristics among the HZs, in terms of cluster mass, grape pH, TSS and TA, while only for cluster mass and pH the differences were significant also between the vintages. Calculation of the percentage of variance of the grape quality attributable to HZs, vintage and HZs x vintage interaction enabled the recognition of the HZs as explaining from the 58 to 67% of the grape quality variability.This study was carried out within the Agritech National Research Center and received funding from the European Union Next-Generation EU (PIANO NAZIONALE DI RIPRESA E RESILIENZA (PNRR) – MISSIONE 4 COMPONENTE 2, INVESTIMENTO 1.4 – D.D. 1032 17/06/2022, CN00000022) (T7.1.1, T2.3.1).