Satellite Solutions for Precision Viticulture: Enhancing Sustainability and Efficiency in Vineyard Management

Abstract

The priority problem in intensive viticulture is reducing pesticides, and fertilizers, and improving water-use efficiency. This is driven by global and EU regulatory efforts. This review, systematically examines 92 papers, focusing on progress in satellite solutions over time, and (pre)processing improvements of spatio-temporal and spectral resolution. The importance of the integration of satellites with ground truth data is highlighted. The results provide precise on-field adaptation strategies through the generation of prescription maps and variable rate application. This enhances sustainability and efficiency in vineyard management and reduces the environmental footprint of vineyard techniques. The effectiveness of different vegetation indices in capturing spatial and temporal variations in vine health, water content, chlorophyll levels, and overall vigor is discussed. The challenges in the use of satellite data in viticulture are addressed. Advanced satellite technologies provide detailed vineyard monitoring, offering insights into spatio-temporal variability, soil moisture, and vine health. These are crucial for optimizing water-use efficiency and targeted management practices. By integrating satellite data with ground-based measurements, viticulturists can enhance precision viticulture, reduce reliance on chemical interventions, and improve overall vineyard sustainability and productivity.

1. Introduction

Intensive viticulture management up to now has led to an increase in the quantity and quality of grape production mainly through the external inputs of agrochemicals (i.e., pesticides and fertilizers). These are harmful to non-renewable resources such as soil, to the ecosystem and biodiversity, and to both workers and consumers [1,2,3]. Agrochemical drift affecting air, land, and water bodies, along with mobility and the degradation of pesticides, is a well-documented issue [4,5,6]. The European Union (EU) is among the highest users of pesticides globally, with 500 active substances approved and 374,000 tons sold annually [7,8]. At least 9 million deaths each year are attributed to various forms of pollution, including pesticides, plastics, heavy metals, and toxic chemicals. Over 60% of these deaths are linked to cardiovascular disease [9].

Global reports highlight the deteriorating biophysical status of 5670 million ha of land, attributing 29% of the decline to human activities, including agriculture [10]. In Europe, industrial activities and waste disposal have led to 2.8 million contaminated sites, with soil degradation costing over EUR 50 billion annually [11]. Heavy metals, mineral oil, and persistent pesticide residues are the most frequent contaminants at European-investigated sites; 75% of them had excessive nutrient inputs and 58% contained two or more types of residues. In the soils of 11 EU Member States with extensive agricultural areas and high pesticide use, the most common pollutants are glyphosate and its metabolite aminomethylphosphonic acid, along with dichlorodiphenyltrichloroethane, and its metabolites. Beside those broad-spectrum of fungicides boscalid, epoxiconazole, and tebuconazole have been identified [12].

The EU’s regulatory efforts aim to reduce pesticide use and promote a healthier environment through initiatives like the EU Action Plan for 2050: A Healthy Planet for All and the European Green Plan. These initiatives align with international sustainability goals, such as the United Nations Sustainable Development Goals, and aim to achieve climate neutrality by 2050, ensuring a toxic-free environment [10,13].

Viticulture, as permanent monoculture large-scale crops, faces significant pest, disease, and weed pressures. The genetic vulnerability of Vitis vinifera subsp. sativa to fungal pathogens downy mildew (Plasmopara viticola), powdery mildew (Erysiphe necator), botrytis (Botrytis cinerea); pests phylloxera (Daktulosphaira vitifoliae Fitch), moth (Eupoecilia ambiguella), and leafhoppers (Erythroneura vitis); and bacterium (Xylella fastidiosa) necessitates intensive pesticide use, with average treatment frequency index in French vineyards up to 20 times per year [14,15,16]. Without pesticide use, 75% of the yield can be destroyed. Predictive models indicate an increase in downy mildew from 5 to 20% through Europe by 2050. Those downy and powdery mildew predictive models are weather-designed and mark the start of fungicide use and application in intervals of 7 to 14 days [17,18]. However, rising temperatures and the presence of semi-natural areas have led to a decrease in pesticide use in conventional and organic vineyards [19].

Water management is another critical concern of our time, and water scarcity and drought are recognized as key priorities in the European Green Plan [13], addressed in several major European strategies. The water footprint in vineyards varies significantly with irrigation practices and climate. In a temperate climate with irrigation, 110 L of water is needed to produce 125 mL of wine. In drought-prone regions, this amount can exceed 240 L due to the increased need for irrigation to compensate for lower natural water availability [20]. As in a vicious cycle, long-term irrigation raises concerns about salinity stress and soil health [21]. According to the first-ever country-driven Global Map of salt-affected Soils released by the FAO, over 833 million hectares of soils worldwide are already salt-affected, with 1.5 million ha of farmland taken out of production each year [22].

Climate change effects on grape and wine quality, along with the negative environmental impact of current viticulture techniques (pesticides, irrigation, soil degradation, erosion), have led the European Union to call for precise on-field adaptation strategies. Unmanned Aerial Vehicles (UAVs) provide high-resolution imagery; however, their operational costs, logistical challenges, and limited coverage area present significant limitations. Advanced Earth observation technologies offer promising solutions for sustainable viticulture, especially in viticulture zoning, vine growth, and health monitoring [23,24]. The widespread availability of high- and very high-resolution satellite data, along with frequent time series observations, is opening new research avenues in viticulture. Microsatellites, providing daily revisits and high spatial resolution, are becoming crucial for monitoring vine vegetation, identifying vulnerable vineyard areas, early detection of pests and diseases, and managing weeds and water [25]. The vegetation indices, Normalized Difference Vegetation Index (NDVI), Green Normalized Difference Vegetation Index (GNDVI), Enhanced Vegetation Index (EVI), Normalized Difference Water Index (NDWI), Crop Water Stress Index (CWSI), Soil-Adjusted Vegetation Index (SAVI), Modified Soil-Adjusted Vegetation Index (MSAVI), and Chlorophyll Absorption Ratio Index (CARI), each offer unique, non-destructive, and rapid insights into vineyard health and productivity. These indices support the creation of satellite-derived prescription maps for variable-rate technology. Replacing UAVs with satellite systems lowers costs, simplifies data collection, and allows for more frequent monitoring, improving the potential for timely interventions and better vineyard management. In light of these evolving challenges, using satellite data offers a promising solution for sustainable viticulture by promoting targeted strategies and reducing pesticide use. This shift aligns with the broader trend of adopting advanced remote sensing technologies.

This review investigates the role of satellite technology in mitigating the environmental impacts of viticulture, enhancing on-field adaptation strategies, and promoting sustainable grape production. A key novelty is its detailed analysis of satellite technology evolution, emphasizing improvements in spatial and spectral resolution, revisiting frequencies, and data accuracy. Special focus is given to image (pre)-processing techniques to enhance spatial and spectral resolution and comparability across regions, enabling detailed monitoring of vineyard changes. The integration of satellite data with ground-based measurements and models to create accurate prescription maps for variable-rate treatments in viticulture is highlighted. The review comprehensively assesses various vegetative indices’ (VIs) effectiveness in capturing spatial and temporal variations in vine health, water content, chlorophyll levels, and overall vegetation vigor, crucial for optimizing vineyard management. Additionally, it explores challenges associated with using satellite data in vineyards, including mixed pixels, spectral limitations, and the need for rigorous data pre-processing and validation. These efforts aim to identify vulnerable areas, provide early detection of abiotic and/or biotic stress, and optimize resource utilization through targeted management strategies, with the aim of improving environmental sustainability and enhancing grape production quality.

2. Materials and Methods

The systematic review aimed to comprehensively gather and synthesize peer-reviewed articles focusing on the use of satellite data in precision viticulture. The primary goal was to explore how satellite solutions enhance sustainability and efficiency in vineyard management. The review followed an adapted version of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement from 2020 [26], ensuring a systematic approach to literature selection and analysis. A comprehensive search strategy was applied to the title, abstract, and keywords of articles in selected databases (Scopus, Web of Science, Science Direct, PubMed, MDPI, and Google Scholar). Combinations of keywords used were “satellite image” OR “satellite data” AND “precision viticulture” OR “vineyard” OR “grapevine” AND “monitoring and mapping” OR “variable rate application maps” AND “spatial variability” AND “vegetation indices (Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI))” AND “temperature/water stress”. This approach aimed to identify relevant peer-reviewed journals, conference articles, and book chapters on the topic. The inclusion of papers depended on whether they addressed the usefulness of satellite data in the vineyard, with a focus on their spatial and time resolution, and efficiency in the reduction of pesticide, fertilizer, water, and labor use. The key vegetation indices (NDVI and NDWI) were determined based on their frequency in scientific papers, with the topic “NDVI” OR “NDWI” AND “vineyard” AND “satellite” indexed in Scopus (85 for NDVI; 7 for NDWI) and Google Scholar (1.030 for both), and relevance in assessing key aspects of vineyard management since 2014. NDVI is used for evaluating vegetation health, while NDWI provides valuable insights into water stress, with both indices being well established and validated in viticulture research. As the goal is to perceive alternatives to the Unmanned Aerial Vehicle (UAV) methods of monitoring and mapping in vineyards, those papers that exclusively use UAVs without satellite data were excluded from the final dataset. Preference was given to papers published between 2019 and 2024 to ensure the most current information available on satellite applications. The papers were primarily in English. In total, 308 papers were identified across all databases since 2019. After excluding studies that focused solely on UAVs, the number of relevant papers was reduced to 120. Of these, 92 papers were reviewed in detail, offering a comprehensive overview of the use of satellite imagery in precision viticulture, with an emphasis on monitoring spatial variability, vegetation indices, and stress factors in vineyards.

Two reviewers independently screened the titles, keywords, and abstracts of the identified papers based on the predefined inclusion and exclusion criteria. Papers that met the initial screening criteria were subjected to a full-text review by the same two reviewers working independently. Relevant data were extracted from the included studies using a standardized data extraction form. The extracted data were then cross-verified by the reviewers to ensure consistency and accuracy. Any disagreements between the reviewers were resolved through discussion, with a third reviewer consulted if consensus was not reached.

A narrative synthesis of the findings from the included studies is given and the results are summarized qualitatively, highlighting key themes and trends in the data. Data Extraction and Quality Assessment of the 19 most important studies are shown in Table S1 (Supplementary Material Table S1).

The extracted data included the following:

Study characterization (authors, year of publication, title, data sources, specific indices analyzed), interventions and comparisons, main outcomes and practical importance or applications of the study findings, and limitations.

During the data extraction and quality assessment of the included studies several frequently used vegetation indices were identified, including the NDVI, NDWI, GNDVI, EVI, SAVI, MSAVI, and CARI. To determine the relevance of these indices in vineyard research, a comprehensive search was conducted using the Scopus and Google Scholar databases. The search terms included “specific vegetation index” AND “vineyard” AND “satellite”, and results were filtered to include publications from the last decade. The search gave the following number of scientific papers: GNDVI (3 and 949), EVI (11 and 2060), SAVI (8 and 1380), MSAVI (2 and 398), and CARI (2 and 170) in Scopus and Google Scholar, respectively. The NDVI and GNDVI were primarily used for assessing vegetation health and chlorophyll content, which are critical indicators of vine vigor. CARI was used to analyze chlorophyll absorption, offering a deeper understanding of vine health. EVI is important because of its effectiveness in densely vegetated vineyards and its ability to minimize atmospheric interference, making it particularly useful in regions with variable atmospheric conditions. NDWI and the CWSI were utilized for monitoring water content and stress within the vineyard. NDWI was particularly valuable for assessing plant water status. CWSI, while less commonly used, provided direct measurements of water stress, aiding in the identification of water-deficient areas. Lastly, SAVI and MSAVI were selected for their ability to accurately monitor areas with sparse vegetation, where traditional indices like NDVI might be influenced by soil brightness. These indices adjust for soil reflectance, ensuring that the vegetation signals are not overshadowed by the underlying soil.

3. Results

3.1. The Evolution of Satellite Missions to Assess Vineyard Conditions, Detect Pests and Diseases, and Optimize Resource Use

The evolution of satellite missions for viticulture has progressed significantly, from the pioneering Landsat 1 mission with a spatial resolution of 80 m to the current Landsat 9 and Copernicus Sentinel missions, offering enhanced capabilities for precision viticulture [27,28,29,30].

Landsat 1, launched on 23 July 1972, was the first Earth-observing satellite explicitly designed to study and monitor our planet’s landmasses. It was equipped with a Return Beam Vidicon (RBV) with three bands and a Multispectral Sensor (MSS), which recorded data in four spectral bands (green, red, and two infrared) and had a repeat coverage of 18 days [27]. In 2013, Landsat 8 (Landsat Data Continuity Mission) [31] was successfully launched and operates to the present day. It has the Operational Land Imager (OLI) acquiring data for nine multispectral channels in the solar reflective (VSWIR) part of the spectrum (0.43–2.29 μm) and the Thermal Infrared Sensor (TIRS) acquiring data for two channels sensitive to emitted thermal radiation. The Landsat 8’s spatial resolution for Visible–Near-Infrared–Shortwave (Vis-NIR-SWIR) is 30 m, for the panchromatic, 15 m, and for the Thermal Infrared band, 100 m. The TIRS instrument suffered from two anomalies: an optical design flaw first detected on-orbit that resulted in excessive stray light contamination, particularly affecting the longest wavelength channel (11.5–12.5 μm). As a result, the absolute radiometric error in Band 11 could only reach 9% compared to the 2% requirement, impeding the ability to retrieve accurate surface temperatures [32]. TIRS also suffered an early on-orbit failure of the scene-select mirror (SSM) position encoder, which required alternative operational procedures for the remainder of the mission. To address the issues encountered with TIRS, improvements such as stray light correction in the TIRS data processing and an alternate operations concept that limits the use of the encoder were implemented. The Landsat 9, launched in 2021, builds upon Landsat 8’s design, providing excellent performance with the Operational Land Imager 2 (OLI-2) for reflective band multispectral imaging in the VSWIR wavelengths and Thermal Imaging Sensor 2 (TIRS-2). The Landsat 9’s spatial resolution is 30 m for the Vis–NIR–SWIR bands, 15 m for the panchromatic band, and 100 m for the Thermal Infrared band, which can be adjusted to 30 m in the processed data. In addition to its superior data quality, Landsat 9 provides an 8-day global land revisit frequency when combined with Landsat 8 data [28].

The European Space Agency (ESA) counterpart for Landsat operates the series of Earth observation Sentinel satellites as part of the Copernicus program. The mission is focused on Earth observation, land monitoring, emergency response, and climate change studies.

The Sentinel-1 mission comprises a pair of radar satellites, Sentinel-1A and Sentinel-1B, utilizing synthetic aperture radar technology in the C-band for all-weather, day-and-night imaging in agricultural monitoring and ground deformation measurement. The Sentinel-1 sensor operates at 5405 MHz with a radiometric accuracy of 1 dB, offering four different modes with varying resolutions (up to 5 m), such as Strip Map (SM) and Interferometric Wide Swath (IW). The SM mode provides a high spatial resolution of 5 × 5 m in a coverage area of 80 km in single (horizontal HH, vertical VV) or double polarization (HH + HV, VV + VH) suitable for environmental monitoring and crises. The IW mode covers larger areas up to 250 km, with a slightly lower resolution, 5 × 20 min single or double polarization, the same as in SM mode. At the same time, the radar beam also moves along the azimuth, which enables improved imaging radar resolution. This technique is Terrain Observation with Progressive SAR (TOPSAR) and results in a scene composed of three-by-nine sub-areas (the so-called “swath” and “burst”). The benefits of the Sentinel-1 mission are the capability of dual polarization, very short revisit time (1–3 days for Europe), and rapid product delivery within an hour of acquisition. The improved split-spectrum method for TOPSAR interferograms using Sentinel-1 data has been developed to reduce errors in terrain observation, highlighting the utility of Sentinel-1 TOPSAR in viticulture applications [33]. Sentinel-1 can delineate low topsoil moisture areas in Mediterranean vineyards for irrigation management [34]. Complementary use of Sentinel-1A SAR and NASA ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) datasets enabled a comprehensive understanding of vine growth dynamics, stress responses, and environmental impacts in Sonoma County, enhancing vineyard management practices and decision-making processes [29].

Sentinel-2 is even more effective in capturing vineyard spatial variability and analyzing pure canopy pixels due to its radar-based technology [34]. With a 10–20 m spatial resolution, 13 spectral bands, and a revisit time of every 5 days, it is used in the monitoring of vine vigor and health through growth and in different training conditions [23,35]. Sentinel-2 data are used in mapping spatial and temporal variability in vineyards and among vineyards in specific terroirs [24,30]. The satellite-derived Sentinel-2 and thermal infrared maps of evapotranspiration can be used for optimization of irrigation management strategies through monitoring water use and stress in vineyards of the Central Valley of California [36,37]. Spatially variable-rate nitrogen fertilizer maps were made for Sicilian vineyards to reduce production costs, environmental impact, and climate footprints per kg of produced grapes, according to the European Green Deal challenges [13,38]. Sentinel-2 data are also a cost-effective tool in assessing frost damage and the impact of management recovery. The Enhanced Vegetation Index, Near-Infrared, and Red Edge 7 indices exhibited high sensitivity in detecting frost damage, while the Modified Triangular Vegetation Index 1 provided the most precise information about recovery 40 days post-frost event [39,40]. The medium-resolution imagery of Sentinel-2 data is suitable for the detection of heat stress on grapevines in irrigated vineyards [41]. It provides spatially variable practical information on vineyard conditions such as soil variability, water status, phenological phases, and environmental stress conditions. Optimization of inputs (water, fertilizers, pesticides) has a valuable role in the reduction of the climate footprint of precise viticulture.

• However, the use of Sentinel-2 data for monitoring qualitative parameters like total soluble solids and total acidity in grape berries is limited due to the low correlation of those with NDVI [35]. As those require high precision, the conventional methods for the analysis of basic grape ripeness parameters remain essential, especially in vertically positioned training systems where satellite technologies face limitations.

The primary mission of Sentinel-3 satellites, launched in 2016 (3A) and 2018 (3B), is a precise measurement of topography, surface temperature, and color of the ocean, sea, and land for continuous environmental and climate monitoring. Data from the Sentinel-3 mission with a higher temporal and spatial resolution furthermore increased the ability of vigor and health satellite monitoring in Mediterranean vineyards [42]. It is also used to estimate evapotranspiration and vine water status. Sentinel-3 data in this case are combined with Sentinel-2 data mainly through the technique Two-Source Energy Balance Model (TSEB) [43,44]. This integration provides superior insights into vine water status, improving the accuracy of estimating spatio-temporal variability of actual transpiration and water stress in vineyards. The problem is that the model is not accurate at low transpiration rates or when vines are under severe water stress, due to potential over- or underestimation. However, the TSEB model with sharpened land surface temperature from Sentinel-2 showed superior performance in estimating transpiration and assessing water stress in vineyards [43].

Commercial satellite imagery, such as WorldView 2–3 (operated by Maxar Technologies, Longmont, CO, USA), offers high-resolution satellite imagery with remarkable detail, at a panchromatic resolution of 50 cm [45]. The Planet satellite provides frequent revisits and a vast archive of imagery, allowing for daily monitoring of changes on Earth’s surface. The WorldView-3 multispectral images of 3 m resolution were used to estimate grapevine water status offering large-scale information; however, the ground truth measurements are needed for the model calibration at the beginning of the growing season [46]. The high-spatial-resolution Planet images allow real-time weekly vineyard-scale monitoring of stem water potential (Ψstem) through vegetation indices [46,47]. Moreover, those are extremely important in the development of a ‘global’ model for in-season monitoring of Ψstem, offering a cost-effective and efficient method for real-time assessment of vine water status in Mediterranean vineyards [47].

3.2. Satellite Images Pre-Processing and Processing

As the electromagnetic signal recorded from the satellite sensor passes through the atmosphere twice, a crucial step to obtain accurate and reliable satellite images is image pre-processing. This includes a geometric correction to remove distortions caused by the sensor, platform, and measuring instrument including time variation. Besides these intrinsic distortions, there are also those impacted by atmospheric conditions such as refraction, reflection, turbulence, and Earth-related factors like curvature, rotation, and topographic effects. These distortions can lead to changes in scale, irregular angular relationships among image elements, displacement of objects, and occlusion of image elements. Correction involves modeling to understand and quantify distortion sources and to derive correction formulas. In the process of geometric correction, ground control reference points are used to establish accurate location information. These control points are used with satellite image acquisition details like sensor specifications, acquisition geometry, and viewing angles to ensure precise geometric correction. Additionally, factors such as yaw, pitch, and roll values of the satellite platform during image capture are considered to enhance the correction process. Once the correspondence between specific points in the distorted satellite image and the ground truth reference map is determined, mathematical transformations are used to rectify the image and align it with the reference coordinate system [48].

To mitigate distortions caused by the Earth’s atmosphere, which can absorb, scatter, and reflect sunlight, impacting the quality of the images, atmospheric corrections are used [45]. Atmospheric correction begins with the use of radiative transfer models, mathematical models used to simulate how electromagnetic radiation interacts with the atmosphere at different wavelengths. These models take into account factors such as the composition and density of atmospheric constituents, as well as the angle and intensity of sunlight. Several atmospheric correction algorithms exist, each tailored to specific sensor characteristics, spectral bands, and atmospheric conditions. Complex physically based models such as Moderate Resolution Atmospheric Transmission (MODTRAN) [49,50], Simulation of the Satellite Signal in the Solar Spectrum (5S) [51], Vector version of the Second Simulation of the Satellite Signal in the Solar Spectrum (6S) [52], Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes (FLAASH) [53], Image Correction (iCor) [54], Framework for Operational Radiometric Correction for Environmental monitoring (FORCE) [55], Py6S [56], Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS) [57], and Landsat Surface Reflectance Code (LaSRC) [58] are commonly used for precise atmospheric corrections. Those require aerosol optical depth, aerosol type, water vapor, and ozone concentration, as well as geometric information like solar and sensor zenith and azimuth angles and the introduction of those data from diverse sources (in situ or satellite) increases the problem of inherent errors in the final imagery. To address this, simple image-based models such as the Dark Object Subtraction method [59], the Empirical Line method, and histogram matching are used. These models use the image metadata file for estimations that lead to enhancing the accuracy of surface reflectance values in images without the need for explicit atmospheric parameters [60]. The Landsat and Sentinel-2 provide the analysis-ready data with the corrected atmospheric effects that can be directly used in the calculation of NDVI [61,62].

Once atmospheric effects are estimated and corrected, satellite imagery is often converted from top-of-atmosphere radiance or digital numbers to surface reflectance values. Reflectance measures how much light is reflected by the Earth’s surface at different wavelengths, with surface reflectance values less affected by atmospheric conditions and thus directly comparable between different images and regions. Validation of the atmospheric correction accuracy involves comparing corrected imagery with ground-based measurements, evaluating feature consistency across multiple images, and conducting sensitivity analyses to assess atmospheric parameter impacts on correction results.

To enable accurate and consistent satellite imagery data, normalization adjustment of the radiometric values of different images is performed, making pixel values from multiple images comparable and suitable for analysis. Gain and offset corrections ensure the sensor’s response linearity, by scaling pixel values to account for sensor sensitivity variations and adjusting for zero-level offsets or dark current noise. Saturation correction methods, such as thresholding, rescaling, or dynamic range adjustments, address pixel saturation issues where recorded values exceed the sensor’s dynamic range, thereby preventing loss of information in saturated image regions. Radiometric correction also incorporates noise reduction techniques to improve image quality and enhance the signal-to-noise ratio. Common noise reduction methods include spatial filtering, wavelet denoising, and statistical approaches. These ensure that the radiometric values of satellite images reflect the true physical properties of the Earth’s surface, enabling reliable quantitative analysis and interpretation. The processing of satellite images for viticulture involves various steps to extract valuable information. Initially, the selection of appropriate satellite data with high temporal, spatial, and radiometric resolution is crucial [63]. The first step often involves image enhancement through contrast enhancement and noise reduction. In this step, different sharpening methods can be used as pan-sharpening for enhancing multispectral image resolution using panchromatic data [45], or data mining sharpening method to sharpen thermal imagery using shortwave multispectral data [43]. After this, image classification and segmentation are performed. Classification assigns pixels into categories based on spectral characteristics, while segmentation groups them into meaningful objects or regions using spectral, spatial, and contextual information. The selection of pure canopy vegetation and soil pixels can be undertaken with a supervised classification method that takes advantage of the semi-automatic classification plugin of QGIS [43]. Comparing images over time ensures the detection of changes in land cover, land use, and environmental conditions, often using algorithms to identify and quantify these changes. Feature extraction estimates biophysical parameters like vegetation indices and surface temperature, as well as deriving topographic information such as elevation models and slope maps from stereo satellite imagery. To enhance the spatial resolution of Sentinel-2A satellite images for detailed analysis in Italian Aglianico vineyards, a combination of techniques was employed. Brook et al. [64] utilized a modified pan-sharpening scheme by Park et al. [65] to increase the original 20 m and 60 m spatial resolution to 10 m across the Vis–NIR–SWIR. This approach involved fusing multispectral images with higher-resolution panchromatic images, enhancing both spectral and spatial resolution. Additionally, convolutional neural networks (CNNs) were implemented to improve the spatial resolution and spectral characteristics of the images. To ensure spectral accuracy, reflectance data were calibrated using the empirical line method, and atmospheric effects were corrected using the sen2cor software (v. 2.4.0) provided by ESA [66]. These advancements in image processing and deep learning techniques enable a more detailed analysis of water-related parameters and plant characteristics, facilitating the study of soil water availability, water flow, and water-use efficiency in vineyards.

The continuous wavelet transform (CWT) and CNN techniques can be employed to reconstruct time series data and identify vineyards accurately [67]. Additionally, the integration of Very High-Resolution (VHR) images from optical and microwave bands, along with neural network models, aids in extracting biophysical parameters for precision viticulture [68]. Furthermore, algorithms can be designed to process satellite images to locate representative pixels within vineyard blocks, optimizing spatially explicit sampling protocols for assessing fruit maturation and quality. NDVI3 is a novel satellite NDVI-based sampling protocol that includes three consecutive NDVI pixels (left tail, center, right tail) within the vineyard block. NDVI3 is highly accurate in the estimation of total soluble solids, pH, titratable acidity, and total anthocyanins in large uniform vineyards. An estimated p-value of NDVI3 was higher than 0.90 for 12 of 13 blocks in comparison to commercial methods with a p-value of <0.05 for 3 of 13 blocks [69].

3.3. Integration of Satellite Data with Ground-Based Measurements and Models to Achieve Accurate and Reliable Prescription Maps for Variable Rate Treatments

Ground-based measurements provide essential validation data for satellite-derived products [44,70,71]. Parameters such as vegetation indices, soil moisture, and canopy characteristics derived from satellite imagery, when used with field measurements collected with ground-based sensors or manual sampling, enhance the spatial coverage, temporal resolution, and overall accuracy and reliability of data. Studies by Palazzi et al. [70] and Kang et al. [44] emphasize the importance of combining Sentinel-2 data with ground measurements to monitor spatial and temporal variations in vineyard ground cover and accurately estimate the Leaf Area Index (LAI) for evapotranspiration modeling. Kang et al. [44] highlight the impact of LAI uncertainties on evapotranspiration estimation, especially in vineyards with dense canopies, underscoring the need for appropriate satellite-based LAI estimation methods. The authors in this study used the TSEB model that performs well under the low-to-medium LAI, and shows underestimations under high LAI, and in densely clumped vine canopies. Furthermore, Cunha et al. [71] demonstrate a strong correlation, with an average deviation of 3 days, between satellite-derived vineyard phenology and ground-based observations, particularly during critical phenological phases like flowering and veraison, under eco-physiological conditions of Douro in Portugal. The satellite-based phenological observations are efficient alternatives to traditional ground-based methods. However, the integration of satellite data with ground-based measurements allows for the creation of accurate prescription maps for variable rate treatments, ensuring that the right quantities are applied in specific locations based on real-time data analysis and monitoring. Variable rate application maps or prescription maps are created from the range of geo-referenced data, canopy vigor maps located in the field using a Global Navigation Satellite System, remote sensing, historical in situ information (soil nutrient levels, meteorological conditions, yield, other agronomic data), decision support systems, and GIS software [72]. Syncing these maps with GPS technology helps us define zones for the precise input use of fertilizers, pesticides, or irrigation based on specific conditions within a vineyard to optimize input use. Variable rate application maps minimize environmental impact, help reach synchronicity, and maximize yield in precise viticulture. Satellite-derived variable-rate application maps demonstrate significant potential for approximating aerial imagery in vineyard vigor assessment, with correlation coefficients reaching up to 0.80 following pixel selection and spatial interpolation. However, temporal and dataset variations introduce inconsistencies in the quantitative interpretation of vigor strength between the aerial and satellite datasets [73]. The generation of detailed canopy characterization prescription maps using multispectral images and information provided by a decision support system led to a reduction in pesticide use, with savings of up to 40% compared to conventional spray applications [74]. From the satellite-derived NDVI and NDWI maps and decision support system for nitrogen fertilization in vineyards, spatially variable-rate nitrogen fertilizer maps are generated for Syrah and Nero d’Avola varieties with an optimal fertilization time according to vine vigor and leaf water content [38]. Thermal satellite imagery can effectively monitor vineyard water use and stress, capturing spatial heterogeneity and transient stress events, which are not reflected in vegetation index-based estimates [37]. Recently, the shortwave infra-based indices from Sentinel-2 (hemispherical shortwave albedo, LAI, and soil and canopy water status) were integrated with surface energy data from Landsat-7 and -8 through the Shuttleworth and Wallace model. This approach enhanced the frequency and reliability of high-resolution daily evapotranspiration products suitable for operational irrigation management. Thermal-based models are used to calculate the latent heat flux through land surface temperature input from platforms like Landsat that provide medium resolution (100 m, resampled to 30 m) of thermal data at a revised time of 8–16 days. The Optical Trapezoid Model (OPTRAM) approach assesses soil and vegetation moisture using shortwave infrared bands from satellites like Landsat-8 and Sentinel-2, and establishes correlations between NDVI and Shortwave Infrared Transformed Reflectance (STR), enabling the estimation of soil water content through a water index (W) [36]. These advancements significantly enhance the accuracy and frequency of data, and support more precise irrigation strategies and the optimization of water-use efficiency in vineyards.

Variable rate irrigation in vineyards can be used based on NDVI. There is a significant correlation between the NDVI and Crop Water Stress Index (CWSI) through which homogenous management zones within vineyards with different water needs can be identified [75]. Campos et al. [76] generated plant protection product prescription maps from vigor maps obtained by manual field measurements and remote-sensing-based methods for canopy characterization in commercial vineyards. The leaf wall area and tree row volume in manual field measurements were then calculated at three canopy stages (beginning of flowering, berries pea size, and beginning of ripening). The high-resolution satellite images with 3 m of spatial resolution and daily revisit times were obtained from the PlanetScope constellation of nanosatellites. The plant protection product prescription maps enable variable rate application of pesticides and a significant reduction in their use. Advancements in variable-rate sprayers, equipped with ultrasonic and laser scanning sensors, allow targeted application of plant protection products, optimizing resource use and minimizing environmental impact. The savings in pesticide application volume were 58%, maintaining similar or even better canopy coverage [77]. The satellite-derived prescription maps in viticulture are shown to be quite accurate and useful for various applications, including the mapping of vine vigor, terroirs, and water stress, as well as for delineating soil characteristics and management zones. The integration of satellite data with other sources, such as UAV imagery and field data, enhances the precision and applicability of these maps [35,64,76]. The research indicates that these tools can provide valuable insights for precision viticulture, supporting site-specific management strategies to optimize grape production and quality.

3.4. The Most Commonly Used Vegetative Indices from Satellite Images

The seven most commonly used VIs from satellite data given in

Table 1. Vegetative Indices for Satellite Monitoring in Vineyards.

Index (Acronym)	Formula Equation	Application and Advantages	Limits	References	Spectral Region
Normalized Difference Vegetation Index (NDVI)	${\displaystyle \frac{\mathrm{N}\mathrm{I}\mathrm{R}-\mathrm{R}\mathrm{e}\mathrm{d}}{\mathrm{N}\mathrm{I}\mathrm{R}+\mathrm{R}\mathrm{e}\mathrm{d}}}$	Simple, widely used, indicator of vegetation health, vigor, and stress	Saturation in high biomass	Tassopoulos et al. [23]; Stolarski et al. [24]; Di Gennaro et al. [35]; Comparetti et al. [38]	NIR, Red
Normalized Difference Water Index (NDWI)	${\displaystyle \frac{\mathrm{N}\mathrm{I}\mathrm{R}-\mathrm{S}\mathrm{W}\mathrm{I}\mathrm{R}}{\mathrm{N}\mathrm{I}\mathrm{R}+\mathrm{S}\mathrm{W}\mathrm{I}\mathrm{R}}}$	Indicator of water content and changes in vines	Affected by soil moisture	Comparetti et al. [38]; Palazzi et al. [70]; Borgogno-Mondino et al. [73]	NIR, SWIR
Green Normalized Difference Vegetation Index (GNDVI)	${\displaystyle \frac{\mathrm{N}\mathrm{I}\mathrm{R}-\mathrm{G}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}}{\mathrm{N}\mathrm{I}\mathrm{R}+\mathrm{G}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}}}$	Chlorophyll and nitrogen content	Non-vegetation factors	Tassopoulos et al. [23]; Cogato et al. [39]; Helman et al. [47]; Brook et al. [64]	NIR, Green
Enhanced Vegetation Index (EVI)	$2.5 * {\displaystyle \frac{\mathrm{N}\mathrm{I}\mathrm{R}-\mathrm{R}\mathrm{e}\mathrm{d}}{\left(\mathrm{N}\mathrm{I}\mathrm{R}+6\mathrm{R}\mathrm{e}\mathrm{d}-7.5\mathrm{B}\mathrm{l}\mathrm{u}\mathrm{e}\right)+1}}$	Reduces background noise (atmosphere, canopy), for vegetation health and phenology	More complex, requires additional parameters (Blue band)	Tassopoulos et al. [23]; Cogato et al. [39]; Cogato et al. [41]; Helman et al. [47]	NIR, Red, Blue
Soil-Adjusted Vegetation Index (SAVI)	${\displaystyle \frac{\mathrm{N}\mathrm{I}\mathrm{R}-\mathrm{R}\mathrm{e}\mathrm{d}}{\mathrm{N}\mathrm{I}\mathrm{R}+\mathrm{R}\mathrm{e}\mathrm{d}+\mathrm{L}}}$ × (1 + L)	Reduces the soil brightness; for heat stress, (semi)-arid conditions	Requires adjustment factor L	Cogato et al. [41]; Helman et al. [47]	NIR, Red
Modified Soil-Adjusted Vegetation Index (MSAVI)	${\displaystyle \frac{2\mathrm{N}\mathrm{I}\mathrm{R}+1-\sqrt{{(2\mathrm{N}\mathrm{I}\mathrm{R}+1)}^2-8(\mathrm{N}\mathrm{I}\mathrm{R}-\mathrm{R}\mathrm{e}\mathrm{d})}}{2}}$	Reduces soil brightness; for (semi)-arid (sparse soil)	Requires adjustment factor L	Tassopoulos et al. [23]; Del-Rio et al. [81]	NIR, Red
Chlorophyll Absorption in Reflectance Index (CARI)	${\displaystyle \frac{\mathrm{R}\mathrm{E}\mathrm{D} \mathrm{E}\mathrm{D}\mathrm{G}\mathrm{E} 5}{\mathrm{R}\mathrm{E}\mathrm{D}}}$${\displaystyle \frac{{\sqrt{(\alpha RED}+RED+b)}^2}{{\left({\alpha }^2+1\right)}^{0.5}}}$ $\mathsf{\alpha }={\displaystyle \frac{RED EDGE5-GREEN}{150}}$ $\mathrm{b}=GREEN-\left({\displaystyle \frac{(RED EDGE5-GREEN)}{150}}550\right)$	Chlorophyll content	Complex calculation	Cogato et al. [39]	Green, Red, Red-Edge

* NIR—Near-Infrared Region, SWIR—Shortwave Infrared Region, L—soil brightness correction factor (common values 0–1), R700—reflectance value at 700 nm, R670—reflectance value at 670 nm, R550—reflectance value at 550 nm.

provide comprehensive insights at both vine and soil levels, contributing to a more accurate assessment of vineyard health and productivity. NDVI and the NDWI [78] are widely used spectral indices that effectively capture spatial and temporal variations in grapevine quality and production [73,79]. NDVI provides information on vegetation greenness (chlorophyll) and is effective for expressing vegetation status and quantifying vegetation attributes, quickly delineating vegetation and vegetative stress. Furthermore, it provides valuable insights into parameters such as LAI, biomass production, chlorophyll concentration in leaves, plant productivity, and fractional vegetation cover [23,24,35,38]. The NDWI measures liquid water molecules in vegetation canopies that interact with incoming solar radiation. NDWI utilizes narrow channels centered at 0.86 μm and 1.24 μm, which penetrate vegetation canopies equally, providing a distinct advantage. Also, NDWI is less affected by atmospheric scattering compared to NDVI [78] and it gives a quicker response to drought than the NDVI [80].

The estimation of both the NDVI and NDWI indexes is associated with atmospheric effect, saturation phenomenon, sensor characteristics, calibration procedures, and data processing techniques, so the values of those indexes can differ from one satellite to another. One of the future goals is to address these challenges and ensure consistency in NDVI measurements across different sensors and platforms [82]. Overall, the accuracy of estimated NDVI indices is intricately tied to the condition of the plants and/or soil. The reliability of NDVI data can be improved through their correlation with ground observation data, temporal and spatial interpolation, and the calibration of multispectral satellite-based models, as demonstrated in recent studies [73,79]. Furthermore, the correlation between satellite-derived NDVI and NDWI’ (detrended NDWI) with vine midday stem water potential and soil volumetric water content was estimated. It was found that models relating ground measures to spectral indices varied depending on the vigor class of the vineyard (whether high or low), highlighting the necessity of having prior knowledge of vineyard spatial variability [73]. The vegetation indices (NDVI, LAI, and NDWI) and integrated machine learning models can be used as accurate predictors of table grape yield across various growth stages. The study revealed that NDVI demonstrated the highest accuracy in predicting grape yield for the years 2017 and 2019, whereas LAI showed better performance in 2019. However, the artificial neural network (ANN) approach consistently achieved high accuracy across all years, with NDVI being the most accurate overall. Validation using ground reference yield datasets confirmed the reliability of the models [83].

Similarly, the GNDVI index provides insights into vine health, water stress, and yield prediction [23,39,47,64]. This index is correlated with Ψstem at the vineyard scale, and it is important for irrigation decision-making for vineyards [47]. The Enhanced Vegetation Index is used for monitoring vegetation health, growth, and phenology [23,39,41,47]. The EVI and the 705 nm bands are significantly correlated to relative humidity, R = 0.62 and R = 0.65, respectively [39]. The CARI index gives information on chlorophyll levels and is highly correlated to photosynthetic activity, vigor, and overall vine health. Together with Green and Red Edge Bands, CARI is negatively correlated with thermal environmental parameters (air and soil temperature and growing degree days) [39]. VIs are used in combination to assess LAI estimation approaches and their impact on Evapotranspiration (ET) modeling sensitivity in vineyards [44].

Among different vegetative indices (GNDVI, NDVI, EVI, and SAVI) derived from high spatial resolution (3 m) Planet images, SAVI shows slightly superior performance due to its ability to mitigate soil effects [47]. Recently the Modified Soil-Adjusted Vegetation Index MSAVI and NDVI were shown as useful tools for precise characterization of vineyard phenology. The MSAVI, which considers both vegetation and soil reflectance, is particularly effective for monitoring early growth stages, while NDVI is better suited for later ones. Specifically, a great distinction between early Chenin Blanc, Sauvignon Blanc, Syrah, and Chardonnay, middle Malbec, Marselan, Merlot, and Tempranillo, and late budbreaking varieties Cabernet Franc, Riesling, and Gewürztraminer was established through MSAVI analysis. Moreover, MSAVI can be used to monitor drought stress and environmental changes during leaf development. The maximum values of NDVI in this study were in September and coincided with harvest time; partly this was the result of different ampelotechniques employed before in the vineyard. When combined with meteorological data, those indexes provide valuable pieces of information for choosing a suitable variety in a specific region [81].

The Crop Water Stress Index (CWSI) is calculated based on the temperature difference between the vine canopy and the surrounding air temperature, using thermal infrared imagery. CWSI can be used as an alternative to direct, time-consuming, and costly measuring Ψstem for irrigation scheduling. Despite significant correlations for different CWSI methods, its application is highly dependent on methodological variations such as soil water availability, vine acclimation, and potential modeling error [43]. ΨMDstem estimates are more accurate for vigorous vineyard areas. Despite simplified methods, uncertainties in ΨMDstem and soil fraction volumetric water content (VWC) estimates are generally consistent with ground measures, with ΨMDstem uncertainty notably higher compared to VWC [73]. This underscores the fact that no universally applicable predictive model for both ΨMDstem and VWC can be developed without accounting for the specific characteristics of the vineyard.

3.5. Challenges in the Use of Satellite Data in Vineyards

The use of satellite images in vineyards faces challenges due to the discontinuous nature of grapevine canopies, their partial coverage, and the significant influence of background and shadows on the measured reflectance signal [84]. Di Gennaro et al. [35] previously reported the presence of mixed pixels due to row-based architecture during their monitoring of a Montepulciano vineyard trained to a tendone trellis system. Similarly, Devaux et al. [30] also observed mixed pixels while monitoring vine growth throughout the entire season, yet they successfully differentiated weeding practices despite these challenges. To address this, rigorous pre-processing of spectral bands is necessary to separate vegetation from the background. Once this pre-processing is completed and spectral bands are combined using index formulas, the spatial variability of the index can be investigated. This information is then interpreted to derive georeferenced prescription maps, which can be used to implement viticultural practices at varying rates or intensities. The ultimate goal is to achieve uniform crop characteristics, maximize grape yield and quality, and simultaneously minimize production costs and environmental impact. Recently, a new method using time series data from Sentinel-2, a decametric resolution satellite, to differentiate LAI of grapevines and manage inter-rows (grass-covered vs. tilled), was proposed. This approach offers a cost-effective and scalable solution for monitoring vineyard characteristics over large areas, reducing the reliance on extensive field observations [85].

In cool climate viticulture and the vertical shoot positioning of shoots, the application of NDVI faces challenges of significant background noise. To address these challenges, higher spatial resolution imagery or advanced row recognition algorithms are needed to enable effective analysis [35,43,86]. The limited spectral bands of satellite sensors represent a challenge in accurate detection and differentiation in vineyards. The multispectral sensors provide data in a few key bands such as the visible, near-infrared, and shortwave infrared wavelengths. The hyperspectral PRISMA images provide detailed spectral bands that allow for precise discrimination in vineyard management, capturing variations in topsoil properties [25]. However, their use entails higher costs for sensor hardware and data acquisition, along with the need for specialized expertise and computational resources for data processing and analysis. To mitigate these challenges, targeted data acquisition strategies can be employed, focusing on specific monitoring tasks to optimize resource utilization and reduce overall expenses [87]. By strategically acquiring hyperspectral data only when necessary, vineyard managers can enhance the efficiency of their operations while still benefiting from the detailed insights provided by advanced sensing technologies.

Satellite images are a reliable tool for canopy map generation; moreover, they align with aerial ones showing consistency in spatial vigor distribution [79]. These images effectively monitor vineyard grape composition variability across seasons and vineyards, providing predictive accuracy comparable to high-resolution airborne imagery [88]. The radiometric concerns such as image calibration, atmospheric correction, and spatial resolution can represent challenges in obtaining reliable information so rigorous pre-processing techniques are essential for ensuring the accuracy and reliability of derived data. Quantitative interpretation of mapped vigor can vary depending on the datasets used and the timing of acquisition. Bias modeling can help mitigate these variations, aiding in a more accurate interpretation of vigor maps.

The reliance on ground truth data for validation in satellite studies [35,43,44,73] presents several inherent limitations that affect their accuracy and practicality in viticulture. Although ground data play a crucial role in periodically calibrating vigor maps and validating data, collecting them is often labor-intensive, costly, and challenging to execute consistently across large or distant vineyards. This results in sparse and unevenly distributed data points that may not fully capture the spatial variability within vineyards. For the calibration of multispectral medium-resolution satellite-based models for mapping the vigor of Moscato Bianco vines and soil water content (ΨMDstem and VWC), which cover a shorter time range and do not perfectly align with the date or hour of satellite acquisitions, Leave-One-Out cross-validation processing strategies were employed [73]. This method, as described by Picard and Cook [89], involves systematically leaving out one ground observation at a time to assess its impact on model performance, thereby identifying the most informative observations for classification purposes.

Although approaches like the Leave-One-Out method have been used to mitigate some limitations in model calibration [73], the scientific rigor of ground measures may still fall short due to inadequate repetitions and spatial distribution. Logistical constraints such as access to vineyard sites and weather conditions further hinder the timely and comprehensive collection of ground truth data. Variations in sampling methods and timing can introduce biases and inconsistencies, compromising the reliability of comparisons with satellite-derived information. Additionally, the effectiveness of these frameworks depends upon the availability and quality of in situ meteorological data, which vary significantly across different regions [36,40]. The Climate Forecast System Re-analysis fails to account for wildfire-induced smoke haze, leading to errors in insolation and net radiation [90]. To address these challenges and improve the reliability of ground truth data, strategies such as integrating sensor networks for continuous monitoring, employing advanced sampling techniques to ensure representative coverage, and fostering collaborative research initiatives for standardized data collection practices are crucial. By enhancing the quality and consistency of ground truth data, satellite studies offer better support to precision viticulture.

Access to satellite imagery may be limited due to factors like subscription costs and data availability restrictions, posing challenges for vineyard managers. Processing and interpreting satellite data require specialized skills and knowledge of remote sensing techniques. Integrating satellite-derived information with ground-based observations and vineyard management practices requires expertise in data analysis and agronomy and may require specialized expertise and user-friendly decision support tools. The agronomic interpretation of satellite-derived maps in viticulture depends heavily on viticulturists’ expertise, emphasizing the need for robust collaboration among researchers, agronomists, and viticulturists to integrate these technologies into vineyard management strategies. By fostering such collaboration, the integration of satellite technologies holds the promise of improving productivity, sustainability, and decision-making in viticulture. The cost-effectiveness and timeliness of satellite imagery acquisition is another challenge that can be overcome by testing different satellite platforms that deliver timely and reliable data when needed for decision-making.

The temporal scope of studies reviewed here often spans approximately one vegetation season, ranging from as short as 11 months [34] to up to 2 [39,90] or even 3 years in the longest cases [64,88]. However, this timeframe may not sufficiently capture the full spectrum of seasonal dynamics critical for effective vineyard management, particularly concerning the nuances across various grape varieties, agroecological conditions, and training systems. Seasonal and site-specific variations significantly influence the accuracy and reliability of models [73], underscoring the need to validate findings across longer periods and diverse environments and practices within viticulture. Moving forward, it is imperative to assess whether these insights hold across different grape varieties, environmental conditions, and management approaches to ensure their broader applicability and relevance. Moreover, the relatively small sample sizes of vineyard plots—from as few as 2 [34] to 12 plots [49]—may further limit the generalizability of these findings across larger or more heterogeneous regions. Only a few studies have focused on large areas: a study of vine water status over 36 plots for 3 years [91], and a study of a heatwave on 107 non-irrigated vineyard blocks [92]. The linear regression model between stem water potential and Sentinel-2 data effectively predicted vine water status across different years, grape varieties, and grass management practices. However, the model was particularly useful during the veraison, rather than the ripening, which limits its practical use for continuous monitoring [91]. The multi-way partial least squares regression model used on Sentinel-2 data successfully predicted yield losses in vineyards across a large region, but the accuracy of the model was moderate, 0.66 in validation [92]. This highlights the ongoing challenge of extrapolating research outcomes to broader viticultural contexts, urging more extensive and diverse data collection to enhance the robustness of future studies.

4. Conclusions

Conclusions and Future Direction of Use of Satellite Data in Viticulture

Advanced satellite technologies, such as Sentinel-2 and Landsat-8, provide high-resolution images that enable territorial-scale monitoring of vineyards. These technologies offer insights into the spatio-temporal variability of vines and soil, topsoil moisture, water status, and different training systems. Heatwaves, late frost damage and recovery time, the intensity of inter-row cover, and differences between conventional and organic vineyards can all be accurately assessed using these technologies. Through evapotranspiration measurements, satellite vegetation indices such as NDVI and NDWI, and grapevine stem water potential monitoring, viticulturists gain valuable insights into grapevine health, physiological processes, and growth requirements. These data are crucial for improving water-use efficiency through precise irrigation scheduling and water management practices, especially in drought-prone regions. Based on spectral data, spatially variable-rate fertilizer maps tailored to the specific needs of individual plots aid in precise nitrogen application. Moreover, evaluating vigor, yield, and moisture variability in vineyards is important for early pest and disease-susceptible areas, as is the adoption of targeted approaches that reduce reliance on chemical interventions. The generation of detailed satellite-based prescription maps can reduce pesticide use by up to 40% while maintaining effective pest control.

To enhance the accuracy and reliability of real-time data, remove noise, and improve satellite image quality, an overview of pre-processing corrections and processing steps is given. These steps ensure that the extracted information accurately represents vineyard conditions, leading to precise and effective management. Integrating satellite data with ground-based measurements and models enhances spatial coverage and temporal resolution, further improving accuracy and reliability.

Looking ahead, satellite data are irreplaceable in environmental studies, enabling the monitoring of land-use changes over time and assessing human impacts on the environment. Satellite imagery tracks changes in vineyard cover and analyzes trends in vegetation dynamics, land surface temperature, and other environmental parameters critical for understanding the effect of climate change on ecosystems and biodiversity. In the future, global collaboration will be critical for the improvement of vineyard management and sustainability through the use of satellite data. Standardizing methodologies, sharing data, and developing comprehensive monitoring systems can provide valuable insights into environmental trends and the drivers of change. Predictive modeling and early warning systems, driven by satellite data, can help anticipate and mitigate environmental risks. Capacity-building initiatives and education programs will empower stakeholders to effectively use these data. Addressing data gaps and technological limitations is important for maximizing the potential of satellite data to manage environmental challenges and advance sustainable development. Robust policy frameworks, community engagement, and ongoing investment in satellite infrastructure can help revolutionize vineyard ecosystem management in the middle of the evolving environmental challenges.