Comparison of Vegetation Indices from Sentinel-2 on Table Grape Plastic-Covered Vineyards: Utilisation of Spectral Correction and Correlation with Yield
Open Access
- 17 November 2025
- journal article
- research article
- Published by MDPI AG in Horticulturae
- Vol. 11 (11) , 1385
Abstract
Climate change represents a critical challenge for viticulture worldwide, primarily through increased heat stress, more frequent and severe drought periods, and unseasonal rainfall events. There is increasing evidence of its negative effects on both thermal regimes—potentially leading to accelerated phenology and unbalanced sugar-to-acid ratios—and hydric regimes—causing water stress that impacts berry development and final yield. The use of plastic covering in vineyards is a widespread technique, particularly in regions with high climatic variability such as the Mediterranean Basin (e.g., Southern Italy, Spain, Greece), aimed at protecting both vegetation and grapes from external factors such as hail, heavy rainfall, wind, and extreme solar radiation, which can cause physical damage, promote fungal diseases, and lead to berry sunburn. This study explores the impact of six distinct commercial plastic films, with varying optical properties, on the retrieval and accuracy of vegetation indices derived from Sentinel-2 imagery in a mid-season table grape vineyard (Autumn Crisp®) in Southern Italy during the 2024 growing season. Laboratory spectroradiometric analyses were conducted to measure film-specific transmittance and reflectance factors from 200 to 1500 nm, enabling the development of a first-order linear spectral correction model applied to Sentinel-2 imagery. Vegetation indices (NDVI, CVI, GNDVI, LWCI) were corrected for plastic interference and analysed through univariate statistics and Principal Component Analysis. Results showed that after applying the spectral correction model, film T2 displayed the higher NDVI value (0.73). Films T3 and T4—characterised by high visible light transmittance (>39%) and low reflectance (2 = 0.810) and sugar content (R2 = 0.779), relationships that were not analysable with the uncorrected data; may guide film selection—opting for high-transmittance films (e.g., T2, T3) for yield or water-conserving films (e.g., T6) for stress mitigation—and irrigation strategies, such as using the corrected LWCI for precision scheduling. Future efforts should include angular effects and ground-truth validation to enhance correction accuracy and operational relevance.Keywords
This publication has 63 references indexed in Scilit:
- Exploring Multisource Remote Sensing Capacities for Spatiotemporal Analysis and Quantification of Chlorophyll of Vineyards in ArmeniaAgricultural Research, 2025
- Evaluation of canopy fraction-based vegetation indices, derived from multispectral UAV imagery, to map water status variability in a commercial vineyardIrrigation Science, 2024
- Overview of Uncertainty Propagation Methods2023
- Transparent cellulose-based film with water resistance and thermal protective for agriculture applicationBiomass Conversion and Biorefinery, 2023
- Assessment of vineyard vigour and yield spatio-temporal variability based on UAV high resolution multispectral imagesBiosystems Engineering, 2023
- Remote sensing of plastic‐covered greenhouses and plastic‐mulched farmlands: Current trends and future perspectivesLand Degradation & Development, 2022
- Grapevine stem water potential estimation based on sensor fusionComputers and Electronics in Agriculture, 2022
- A broad-band leaf chlorophyll vegetation index at the canopy scalePrecision Agriculture, 2008
- On the design and validation of surface bidirectional reflectance and albedo modelsRemote Sensing of Environment, 1992
- Red and photographic infrared linear combinations for monitoring vegetationRemote Sensing of Environment, 1979