The research team conducted a structured literature search across the Scopus and Web of Science databases, initially retrieving 1,253 records. After removing duplicates and screening titles, abstracts, and full texts using the Rayyan tool, the team narrowed the field to 39 studies that specifically addressed vineyard land suitability and produced suitability maps for vineyard establishment. This rigorous filtering process highlights how fragmented the field remains despite growing academic and practical interest in climate-resilient viticulture.

What emerged from the analysis is a field dominated by Multi-Criteria Analysis (MCA), which appeared in 17 of the 39 reviewed publications. MCA allows researchers to structure and rank different environmental variables—temperature, rainfall, soil properties, topography—based on their relative importance, making it particularly effective at regional and local scales where multiple factors must be combined, often with incomplete or non-uniform datasets. The method’s flexibility in integrating heterogeneous data has made it the go-to framework for vineyard suitability assessment worldwide.

Yet the dominance of MCA masks a deeper problem: the selection and weighting of suitability criteria vary substantially across studies, reducing methodological consistency and limiting the comparability of results. The number of criteria used ranged from just 1 to 33, with an average of 13, reflecting differences in data availability, spatial scale, and research objectives. Some studies relied on a single indicator—such as minimum temperature to assess frost risk or the Winkler Index to classify thermal suitability—while others attempted comprehensive multi-factor assessments spanning climate, soil, and topography.

Weighting methods proved equally inconsistent. While the Analytical Hierarchy Process (AHP) emerged as the most widely used formal approach, appearing in five studies, a substantial portion of the reviewed research either did not adopt a formal weighting method at all or simply assumed that all criteria contribute equally to suitability classes. This assumption can limit a model’s ability to accurately reflect real-world conditions, potentially reducing the reliability and applicability of the results. Expert-based methods, though widely used for their interpretability, introduce subjectivity into the modelling process, while data-driven approaches demand large and consistent datasets that are not always available.

The geographic distribution of research tells its own story. Europe dominated the reviewed literature with twelve studies across seven countries—Croatia, Italy, Portugal, Spain, Romania, Switzerland, and the United Kingdom—followed by significant representation from Australia, Türkiye, and the United States. This pattern aligns closely with global viticulture, reflecting regions with both established wine industries and active interest in land suitability assessment. The concentration of recent publications also reflects the rapid advancement of enabling technologies, including Geographic Information Systems (GIS), unmanned aerial vehicles (UAVs), and satellite data, which have transformed how researchers collect and process environmental information.

Beneath the methodological diversity lies a shared biological reality: grapevines are extraordinarily sensitive to their climate. Rising mean temperatures directly affect grapevine phenology by advancing key developmental stages such as budburst, flowering, and harvest. This compression of the growing cycle can lead to earlier ripening under warmer conditions, often resulting in higher sugar accumulation, lower acidity, and altered aromatic profiles—ultimately affecting wine quality. Excessive heat can induce thermal stress, impair photosynthesis, and reduce yield, while changes in precipitation patterns influence both water availability and disease pressure, with reduced rainfall intensifying drought stress and excessive rainfall promoting fungal diseases.

The increasing frequency and intensity of extreme events—heatwaves, late spring frosts, drought episodes, and extreme precipitation—represent an additional layer of complexity that most current suitability models fail to capture. These events are typically not reflected in average-based indices, yet they play a crucial role in determining both interannual variability and long-term suitability. The review notes that several studies indirectly account for extreme events through proxy indicators such as frost occurrence, heat stress, and water-related constraints, but these variables are rarely framed within an explicit climate change perspective, limiting their contribution to forward-looking and risk-oriented assessments.

The review also identified significant gaps in model outputs and analytical sophistication. None of the reviewed studies provides quantitative predictions of grape yield or production volume, relying instead on categorical suitability classifications that restrict applicability for operational decision-making. The limited use of advanced techniques such as machine learning constrains the ability to capture complex and non-linear relationships between environmental variables and suitability outcomes. Similarly, the integration of key datasets such as soil moisture remains limited, reducing the accuracy of water-related assessments.

Looking forward, the researchers argue that a paradigm shift is needed—from static suitability assessments to dynamic, climate-responsive modelling frameworks capable of supporting adaptive decision-making in viticulture. Future suitability assessments should integrate climate projections, variability, and extreme events while also incorporating agronomic and quality-related indicators, including sensory analysis of wines and chemical analyses of grapes. The authors also highlight the importance of adaptive management strategies such as irrigation, the adoption of new varieties, or the relocation of vineyards to higher elevations, noting that while initial warming trends may temporarily enhance productivity in certain regions, these benefits are likely to be short-lived. As traditional wine-growing areas face increasing constraints and new regions emerge as suitable for high-quality wine production, the governance dimension becomes central: identifying which stakeholders should be involved and assessing their willingness and capacity to engage will be crucial for equitable and sustainable adaptation pathways in a rapidly changing climate.

Subject of Research: Systematic review of vineyard land suitability assessment methods and their integration of climate change projections

Article Title: Land suitability for vineyards under climate change: a systematic review of methods, tools, and gaps

Article References: Marchese, G., Debolini, M., Bordoni, M., Meisina, C., & Vaglia, V. (2026). Land suitability for vineyards under climate change: a systematic review of methods, tools, and gaps. Frontiers in Agronomy, 8, Article 1839486. https://doi.org/10.3389/fagro.2026.1839486

Image Credits: AI Generated

DOI: 10.3389/fagro.2026.1839486

Keywords: viticulture, land suitability, climate change, Multi-Criteria Analysis, GIS, vineyard mapping, AHP, agronomy, wine quality, climate projections, remote sensing, sustainability

The experiment was designed to isolate the direct effect of temperature from the confounding influence of sunlight and whole-vine physiology. Using a fan-assisted air delivery system adapted from whole-tree gas exchange equipment, the researchers directed warm air through flexible ducts positioned beneath individual bunches, achieving a target temperature differential of roughly 10 degrees Celsius. Thermocouples inserted just under the berry skin recorded temperatures every five minutes, revealing that heated bunches never dropped below 17 degrees and peaked near 40 degrees during the hottest part of the day. The accumulated heat, measured as growing degree days, roughly doubled for heated bunches compared with controls during the treatment window.

Heating was applied continuously over seven days when the bunches had reached mid-véraison, the stage at which half the berries show the first signs of color change. In 2014 the treatment ran from midday on February 11 until midday on February 18, and in 2015 from February 16 to 23. Berry sampling was intensive: composite samples of 24 berries per experimental bay were collected every four hours during the first 48 hours of heating in 2014, then daily during heating in 2015 and every four to seven days afterward until harvest. Berries were frozen at minus 80 degrees, and the skins were carefully scored and lifted from the pulp so that pigments, acids, sugars, and hormones could be quantified in separate tissues.

The chemical analysis delivered a clear verdict on the pulp. Sugar accumulation was essentially untouched: glucose, fructose, and sucrose concentrations showed little response to heating, consistent with earlier bunch-heating studies. Tartaric acid, the other major grape organic acid, was not consistently altered. Malic acid, however, told a different story. By day 6 of the experiment, malic acid concentrations were 25 percent lower in heated berries in 2014 and 16 percent lower in 2015 compared with unheated bunches. Titratable acidity fell in parallel, and juice pH rose, although these differences had largely disappeared by harvest time. The study established that for a 10-degree increase in berry temperature above ambient, it takes three to four days for the hastened decline in malic acid to become apparent.

The skin pigments responded on a different schedule and with seasonal nuance. Five anthocyanin species were detected in the skins, with malvidin-3-O-glucoside the most abundant, accounting for roughly half of the anthocyanin glucosides throughout ripening. In 2014, heating significantly reduced malvidin, peonidin, and petunidin glucosides as well as total anthocyanins from day 4 of heating onward, and the suppression persisted until harvest. In 2015, the effect was far milder: only petunidin and delphinidin glucosides were significantly reduced, while the other anthocyanins and the total pool were largely unaffected. Notably, no effect of heating on anthocyanin concentrations was observed during the first 48 hours of treatment in either year, indicating a lag of up to four days between heat perception and measurable pigment change.

The key to that seasonal disparity appears to be abscisic acid, the hormone that triggers grape ripening and the onset of anthocyanin accumulation. Peak concentrations of free, active ABA at the onset of véraison were 30 percent higher in 2015 than in 2014, and the 2015 season was notably drier, with significantly less rainfall before and during véraison. The researchers speculate that mild-to-moderate water stress in the drier year elevated berry ABA concentrations enough to augment anthocyanin accumulation and thereby mask the inhibitory effect of heat seen in 2014. Supporting this interpretation, the conjugated storage form of ABA, the ABA-glucose ester, increased by as much as 40 percent in heated berries by the end of the heating period and remained elevated afterward in both years.

Gene expression analysis revealed how quickly the berries registered the heat. The heat shock transcription factor HsfA2 showed strong elevation in transcript abundance within four hours of heating commencing, peaking around midday during the first three to four days of treatment. Yet the downstream responses unfolded much more slowly. Changes in the expression of genes and transcription factors regulating malic acid and anthocyanins took two or more days to respond, and the magnitude of those transcriptional shifts was mostly proportional to the metabolite concentrations ultimately observed. This temporal decoupling between heat perception and compositional change is a central technical insight of the study.

On the anthocyanin side, qPCR showed that the structural genes chalcone synthase, dihydroflavonol reductase, and UDP-glucose flavonoid glucosyltransferase were downregulated by heating in 2014, with the reduction in UFGT, the gene controlling the final step of anthocyanin synthesis, taking at least two days to manifest. The transcription factor MYB5a, a general activator of phenylpropanoid genes, was reduced one day after heating, while the repressor MYB4a was upregulated within eight to sixteen hours of heat perception, apparently triggered by the HsfA2 spike. RNA sequencing on day 4 of heating in 2015 confirmed the downregulation of chalcone synthase and revealed broader suppression of phenylpropanoid variants and flavonoid hydroxylase genes, even as some late-pathway genes appeared upregulated in that single snapshot.

For malic acid, the regulatory picture centered on malic enzyme. Cytosolic NADP-dependent malic enzyme expression was upregulated by heating in both seasons, consistent with accelerated breakdown of malate into pyruvate. Fumarate hydratase, which can contribute to malate synthesis, was also upregulated, but the researchers conclude that any potential increase in synthesis was counteracted by the concurrent rise in malic enzyme activity. Genes associated with glycolysis and gluconeogenesis, including phosphoenolpyruvate carboxylase and carboxykinase, were downregulated, effectively closing off alternative routes for malate accumulation or dissimilation. The net effect of these interacting processes was a temporary but significant reduction in malic acid in heated berries.

The broader implication is sobering for cool-climate wine regions. Pinot noir carries a nonsense mutation in the gene encoding anthocyanin 3-O-glucoside-6-O-acyltransferase, rendering it unable to synthesize the acylated anthocyanins that lend other red varieties greater stability against heat, light, and the rigors of fermentation. Without that biochemical buffer, Pinot noir berries appear more vulnerable to heat-driven pigment loss. The study demonstrates that hot periods with daily maxima approaching 40 degrees Celsius sustained over a week can directly alter the ripening trajectory of Pinot noir, reducing organic acids and anthocyanins and desynchronizing the accumulation of primary and secondary metabolites that together define berry maturity balance at harvest.

Subject of Research: Effects of bunch heating on Pinot noir grape ripening, anthocyanin and malic acid concentrations, and their gene regulation

Article Title: Effects of bunch heating on Pinot noir grape ripening: changes in berry anthocyanin and malic acid concentrations and their regulation

Article References: Bennett, J. S., Albright, A. L., Friend, A. P., Boldingh, H., Cooney, J., Elborough, C., Kirk, R. A., Nguyen, H. M., Deng, C., Plunkett, B. J., Martin, D., & Espley, R. V. (2026). Effects of bunch heating on Pinot noir grape ripening: changes in berry anthocyanin and malic acid concentrations and their regulation. Frontiers in Plant Physiology, 3, Article 1701827. https://doi.org/10.3389/fphgy.2025.1701827

Image Credits: AI Generated

DOI: 10.3389/fphgy.2025.1701827

Keywords: Pinot noir, grape ripening, heat stress, anthocyanins, malic acid, abscisic acid, véraison, climate change, wine quality, gene expression, New Zealand, viticulture

The story of Tribidrag is itself a tale of scientific detective work. For decades, the identity of Zinfandel was one of the great mysteries of viticulture, with competing claims about whether the variety originated in Italy, where it thrives as Primitivo, or elsewhere in the Mediterranean. Genetic fingerprinting eventually traced the cultivar to Croatia, where it survives under its oldest recorded name, Tribidrag, and where old vines still cling to terraced slopes along the Adriatic coast. Because grapevines are propagated vegetatively, by cuttings rather than seeds, every plant of a given variety is in principle a genetic copy of its parent. Yet over centuries of cloning, somatic mutations accumulate, and growers have selected vines that differ in yield, berry size, sugar accumulation, and acidity. These selected lines, called clones, are the raw material of clonal selection programs that supply nurseries and vineyards worldwide.

The research team, led by Goran Zdunić of the Institute for Adriatic Crops and Karst Reclamation in Split and including collaborators at the University of Zagreb and Yunnan Agricultural University in Kunming, set out to bridge a persistent gap in clonal research. Although previous studies had documented variation among Tribidrag clones both in the vineyard and at the molecular level, no one had systematically connected the two. The question was deceptively simple: if two clones of the same variety look and behave differently, can we find the genetic differences responsible? To answer it, the researchers evaluated productive traits in eight clones over three consecutive growing seasons, measuring yield, yield components, and basic must chemistry, the sugar and acid parameters that determine the character of the resulting wine.

The field data told a clear story of phenotypic divergence. The eight clones varied significantly in productivity and showed significant phenotypic plasticity, meaning their performance shifted from year to year depending on growing conditions. Statistical analysis revealed a significant clone-by-year interaction, confirming that no single clone dominates in every season. Two clones maintained under the Foundation Plant Services collection in California, designated FPS03 and FPS06 and known as Primitivo selections, consistently showed lower productivity, smaller berries, higher sugar content, and lower total acidity in the 2021 and 2022 seasons compared with the Zinfandel 1–24 clone. That latter clone displayed greater productivity in all three years of the study, making it the most reliable performer in the trial. For winemakers, these differences matter enormously: smaller berries with concentrated sugars tend to produce more structured, intense wines, while higher-yielding clones favor volume and freshness.

With the phenotypic landscape mapped, the team turned to the genome. Using short-read whole-genome resequencing, they generated high-coverage sequence data for each of the eight clones and called variants across the entire reference genome of grapevine. The scale of the variation was remarkable. Each sample carried approximately 3.8 million single nucleotide polymorphisms, or SNPs, which are positions where a single DNA letter differs from the reference, along with roughly 300,000 insertions and deletions, known as InDels, where short stretches of sequence are present or absent. That level of variation within a single vegetatively propagated cultivar underscores how much somatic mutation and historical mixing of plant material can accumulate over centuries of clonal transmission. Most of the structural variants were located in intergenic regions, the stretches of DNA between genes, which is consistent with the expectation that purifying selection keeps functional gene sequences relatively stable while neutral variation piles up elsewhere in the genome.

To hunt for signatures of clonal selection, the researchers took a windowed approach, dividing the genome into 100-kilobase segments and counting pairwise SNP differences among the eight clones within each window. This method highlights genomic neighborhoods where particular clones stand out from the rest, suggesting that a mutation or set of mutations arose on a specific vine lineage and was then propagated along with it. The analysis revealed candidate regions that distinguish individual clones, providing a catalog of genomic intervals that future studies can mine for genes influencing berry size, sugar transport, acid metabolism, and other traits of viticultural interest. Because the windows are small enough to pinpoint candidate loci but large enough to be robust to sequencing noise, the approach offers a practical middle ground between whole-genome association studies and targeted candidate-gene sequencing.

Then came the twist. When the researchers performed a principal component analysis on the genome-wide SNP data, the eight clones resolved into three principal genetic clusters. If clonal selection had driven the phenotypic differences, one might expect the low-yielding, sugar-rich Primitivo clones to group together genetically, apart from the productive Zinfandel selections. Instead, the genetic groupings showed no clear geographic association among the samples and, more strikingly, no direct association with the phenotypic or productive traits of individual clones. The clones that look most different in the vineyard are not necessarily the ones that are most different in their DNA, at least not at the resolution of genome-wide SNP patterns. Measurable genetic differentiation is real, but it is not a simple mirror of measurable phenotypic divergence.

This disconnect is not a failure of the study; it is arguably its most valuable finding, and it echoes a pattern seen in other clonally propagated crops. Phenotypic differences among clones can arise from a small number of mutations with large effects, from epigenetic changes such as DNA methylation that whole-genome resequencing of the kind performed here does not capture, from structural rearrangements too large or too complex for short-read sequencing to resolve, or from interactions between genotype and environment that three field seasons, however carefully measured, only begin to sample. The significant clone-by-year interaction observed in the trial reinforces this last point: the same clone can rank differently for yield and must composition depending on the season, so any attempt to link genes to traits must account for environmental noise. The candidate SNP signatures identified in 100-kilobase windows now provide a starting list of genomic regions where deeper investigation, including long-read sequencing, epigenomic profiling, and functional validation, is most likely to pay off.

For the wine industry, the implications are practical as well as scientific. Clonal selection programs rely on the assumption that the clones they distribute are genetically distinct and predictably different in performance. The new study confirms that Tribidrag clones carry abundant genomic variation and that some of it marks individual clonal lineages, which strengthens the case for DNA-based tools to authenticate and track clonal material. At the same time, the weak link between genome-wide differentiation and field performance is a caution against assuming that a genetic profile alone can predict how a clone will behave in a given vineyard. Growers choosing between the consistently productive Zinfandel 1–24 and the smaller-berry, higher-sugar Primitivo selections FPS03 and FPS06 will still need multi-season field data, but they now have a far richer genomic map to guide the search for the mutations that ultimately matter in the glass.

The work also carries a conservation message. Tribidrag survives in Croatia as a heritage cultivar of limited distribution, and the study was supported by the Croatian Science Foundation through the RemainGrape project, which is dedicated to the prospection and genomic analysis of endangered Croatian grapevine cultivars and wild grapevine. By documenting both the phenotypic richness and the genomic depth of variation within a single ancient variety, the research makes the case that preserving old vineyards and clonal collections preserves an irreplaceable archive of somatic evolution. Every century-old vine carries mutations accumulated over generations of vegetative transmission, and each clone is a living record of both natural mutation and human selection. Sequencing eight clones has revealed how much of that record remains to be read, and how much of the connection between the grapevine genome and the wine in the bottle is still waiting to be decoded.

Subject of Research: Whole-genome sequencing and phenotypic analysis of clonal variation in the grapevine cultivar Tribidrag (Zinfandel)

Article Title: Whole-genome sequencing of eight grapevine clones of cv. Tribidrag (Vitis vinifera L.) reveals measurable differentiation not linked directly to phenotypic divergence

Article References: Zdunić, G., Duan, S., Wang, C., Lukšić, K., Marinov, L., Mucalo, A., Ozretić Zoković, M., Maletić, E., Pejić, I., & Chen, W. (2026). Whole-genome sequencing of eight grapevine clones of cv. Tribidrag (Vitis vinifera L.) reveals measurable differentiation not linked directly to phenotypic divergence. BMC Plant Biology. https://doi.org/10.1186/s12870-026-10110-x

Image Credits: AI Generated

DOI: 10.1186/s12870-026-10110-x

Keywords: Tribidrag, Zinfandel, grapevine, clonal variation, whole-genome sequencing, somatic mutation, Vitis vinifera, viticulture, SNP, phenotypic plasticity, plant genetics, clonal selection

The problem the team set out to solve is deceptively simple to state and notoriously hard to execute. Accurate, objective, and scalable disease assessment underpins everything from fungicide timing to breeding decisions, but manual scouting cannot deliver it. Human evaluators differ from one another and from themselves over time; rater fatigue and variable lighting degrade consistency further; and coarse ordinal scales lack the resolution to capture subtle treatment effects. These inconsistencies ripple directly into genetics, where phenotype-to-genotype analyses depend on reliable measurements. Even when quantitative trait loci exist, noisy phenotypes can obscure them, reducing reproducibility across breeding programs. In regions like California, scouts must also peer into shaded inner canopy tissues where symptoms hide or pathogen viability is suppressed by ultraviolet light and heat.

Automated imaging systems have long promised a way out, capturing high-throughput visual data across entire vineyard blocks. But a bottleneck has persisted on the annotation side. Powdery mildew lesions are irregular, diffuse, and visually variable, making pixel-level labeling labor-intensive and dependent on scarce expert knowledge. Building large, high-quality, domain-specific datasets is prohibitively expensive, and conventional deep learning models trained from scratch on limited labeled data tend to falter in messy outdoor environments. The challenge, as the researchers frame it, is not acquiring imagery but teaching models to learn from scarce and noisy labels.

Their answer draws on the current paradigm shift in artificial intelligence from task-specific networks to generalist foundation models. The team fused two of the most influential vision foundation models: the Segment Anything Model, or SAM, trained on more than a billion masks and capable of promptable segmentation of arbitrary objects, and CLIP, a model trained on 400 million image-text pairs that aligns visual and linguistic representations in a shared embedding space. SAM contributes high-resolution spatial precision, preserving object boundaries and fine structure, while CLIP contributes semantic reasoning grounded in natural language. In the integrated architecture, each image is processed in parallel by both encoders. A learnable multilayer perceptron projects SAM’s embedding into CLIP’s latent space, the two representations are fused by element-wise addition, and the result is conditioned on text embeddings for prompts such as “Powdery mildew” or “Canopy.” A dot-product interaction between the fused visual representation and the text embedding produces a multi-modal representation that a second MLP transforms back into the spatial layout SAM’s mask decoder needs. The output is segmentation guided jointly by pixel-level detail and language-based disease concepts.

Because labeled agricultural data is scarce, the researchers systematically compared parameter-efficient fine-tuning strategies: full fine-tuning, adapter modules inserted into key layers, and low-rank adaptation, or LoRA, which updates attention and feedforward layers through small low-rank matrices. Each strategy was tested with two SAM backbones, ViT-B and ViT-H, and with either the encoder alone or both encoder and decoder trainable. The verdict was clear: adapter-based tuning with both modules trainable performed best, and the fine-tuning strategy mattered more than raw model size. LoRA’s tiny updates proved insufficient to transfer CLIP’s rich semantic knowledge into the agricultural domain, a finding that underscores how much representational capacity is needed to absorb new visual vocabulary.

The resulting models, PM-SAM-CLIP for lesion segmentation and Canopy-SAM-CLIP for canopy extraction, were trained on modest datasets collected in New York with custom imaging rigs using active strobe illumination to maximize contrast and consistency. PM-SAM-CLIP, built on the ViT-H backbone, was fine-tuned on just 175 manually annotated images; Canopy-SAM-CLIP used 415. Despite the small training sets, the models decisively outperformed baselines including PM-SAM, DeepLab v3+, and SegFormer. On the powdery mildew test set, PM-SAM-CLIP achieved 59.63 percent image-level mean intersection over union and 64.12 percent dataset-level, beating the strongest baseline by roughly 13 to 14 percentage points and weaker baselines by more than 30. Crucially, on the worst-performing subsets of images, where competing models nearly collapsed to zero, the multi-modal model maintained scores between 21.51 and 30.25 percent, demonstrating that semantic context stabilizes performance on the hardest cases, such as sparse lesions and cluttered canopies. Canopy segmentation reached 95.59 percent image-level accuracy with strong worst-case robustness.

The real test came in California, at a USDA vineyard in Parlier planted with a genetically diverse mapping population spanning multiple Vitis species and managed under a different trellis system than the New York training data. Deployed zero-shot, without any additional fine-tuning, PM-SAM-CLIP segmented infections across two growing seasons and more than 30,000 images, contending with confounders like dust and sunburn that visually mimic mildew. A post-processing step intersecting predicted infection masks with canopy masks suppressed false positives outside biologically relevant tissue. At the vine level, severity was computed as the maximum normalized overlap between infection and canopy masks across all images of each vine, mirroring how expert scouts assess the worst infection instance. The agreement with human experts was striking: 79 percent of vines in 2023 and 81 percent in 2024 received identical scores, and more than 98 percent fell within one score level of the human rating. The pipeline even captured subtleties humans cannot, such as the difference between 75 and 76 percent infected area.

The most consequential validation came from genetics. The team fed image-derived severity scores into QTL mapping using a high-density rhAmpSeq genetic map built from 465 markers across 291 progeny of a cross between Vitis amurensis and V. vinifera. Across both years, every phenotyping method, human and automated alike, converged on the same significant QTL on chromosome 13, spanning 22.2 to 27.6 megabases. That interval overlaps the previously fine-mapped REN12 locus, a known major resistance gene, providing biological confirmation that the AI-generated phenotypes capture real genetic signal. The image-derived QTL explained 22.9 to 26.2 percent of phenotypic variance, slightly below the human-scored range but well within expected field variability. Spatiotemporal heatmaps of infection across the vineyard additionally revealed persistent high-severity zones and stable low-scoring vines, information that can guide both genotype selection and site-specific fungicide decisions.

The implications extend well beyond one disease or one crop. By fine-tuning pre-trained foundation models on tiny domain datasets, the approach sidesteps the annotation bottleneck that has long constrained AI in agriculture, and text prompts give plant pathologists an interpretable way to steer model behavior using their own vocabulary. The researchers acknowledge limitations, including occasional false positives from dust and sunburn, QR-code detection failures, and the expressiveness limits of single-keyword prompts, and they point toward edge deployment on field robots, richer textual descriptions, and integration with hyperspectral or thermal sensing. They also call for open, FAIR-compliant multi-modal agricultural datasets to accelerate foundation-model training across the plant sciences. With code, models, and datasets publicly released, the study sketches a closed loop from field imagery to genotype discovery to precision disease management, a blueprint for next-generation phenotyping in viticulture and beyond.

Subject of Research: Automated phenotyping of grapevine powdery mildew using integrated SAM-CLIP multi-modal vision foundation models

Article Title: Integrating large multi-modal models for automated powdery mildew phenotyping in grapevines

Article References: Lin, Y., Dashner, Z., Jimenez, A., Wilkerson, D., Cadle-Davidson, L., Riaz, S., & Jiang, Y. (2026). Integrating large multi-modal models for automated powdery mildew phenotyping in grapevines. Artificial Intelligence in Agriculture. https://doi.org/10.1016/j.aiia.2026.09.006

Image Credits: AI Generated

DOI: 10.1016/j.aiia.2026.09.006

Keywords: powdery mildew, grapevine, phenotyping, SAM, CLIP, foundation models, segmentation, QTL mapping, viticulture, plant disease detection, multi-modal AI, breeding

The research team set out to answer a question that has long frustrated plant pathologists and viticulturists alike: why do some grapevine cultivars tolerate infection by trunk disease fungi while others succumb, even though no grapevine genotype is fully resistant? Tolerance in this context is measured by the expression of foliar symptoms, which varies considerably among cultivars, yet the biochemical mechanisms underlying this variation have remained poorly understood. To probe the question, the researchers chose two cultivars with well-documented contrasting responses to Botryosphaeria dieback: Chardonnay, which shows greater tolerance, and Gewurztraminer, which is more susceptible. By comparing how these two cultivars reconfigure their chemistry after infection, the team hoped to identify the metabolic signatures that separate a resilient vine from a vulnerable one.

The experimental design was ambitious in scope. Rather than examining only the infection site in the woody stem, as most earlier studies have done, the researchers conducted whole-plant metabolomic analyses, sampling wood, leaves and roots at different times after inoculation. They used untargeted liquid chromatography coupled with mass spectrometry, a technique that allows scientists to detect and measure thousands of small molecules simultaneously without knowing in advance which ones will prove important. This unbiased approach is critical when studying complex plant-pathogen interactions, because defense chemistry often involves unexpected compounds and pathways that a targeted analysis would miss entirely.

A key feature of the study was the use of two different N. parvum isolates that differ in aggressiveness. Pathogen colonization assays confirmed that the two isolates possessed contrasting wood invasion capacities, with one spreading more extensively through the host tissue than the other. This distinction mattered: it allowed the researchers to separate the effects of the plant cultivar from the effects of the fungal strain, revealing that the outcome of infection depends on the specific pairing of host and pathogen rather than on a generic interaction between grapevine and wood-decay fungus. The two isolates used in the work were designated Np.Bt67 and Np.Bourgogne, and their differing behavior in the wood provided a natural experiment in fungal virulence.

When the team analyzed the resulting metabolomic data using principal component analysis, a statistical method that reveals the dominant sources of variation in large datasets, a clear pattern emerged. The cultivar of the vine and the identity of the fungal isolate were the main factors structuring metabolic variation across the experiment, with additional contributions from the organ being sampled and the time elapsed since inoculation. In other words, the chemical state of a vine after infection is not determined by the fungus alone or the plant alone, but by the interplay between a particular cultivar and a particular fungal isolate, modulated by where in the plant one looks and how long the infection has had to develop.

The most striking differences appeared in the wood, the primary battleground between vine and pathogen. In response to the most aggressive isolate, the two cultivars mounted clearly distinct metabolic responses, particularly in their profiles of stilbenes and oxylipins. Stilbenes are a family of phenolic compounds, including the famous resveratrol, that grapevines synthesize as antifungal weapons, and their production is governed by enzymes such as stilbene synthase and the upstream phenylalanine ammonia-lyase, both central players in plant defense. Oxylipins, by contrast, are oxygenated derivatives of fatty acids produced through the lipoxygenase pathway, and they serve as both antimicrobial compounds and signaling molecules that coordinate defense responses across the plant.

Here the tolerant cultivar revealed its hand. Compared with Gewurztraminer, Chardonnay showed a faster and stronger induction of stilbene and fatty acid synthesis in its wood, especially of oxylipins, compounds that could be involved in both direct defense and long-distance signaling. This rapid chemical mobilization suggests that the tolerant vine does not simply produce more of the right compounds at the end of the battle; it activates its defensive chemistry earlier and more vigorously, potentially slowing the pathogen before it can establish a firm foothold. Notably, the study found that pathogen growth was similar between the two cultivars, meaning that the difference in tolerance cannot be explained by the fungus simply growing more slowly in Chardonnay. Instead, the two cultivars are characterized by specific metabolite signatures, especially in the wood, pointing to fundamentally different interaction mechanisms rather than a common response to wood decay.

Perhaps the most surprising finding came from beyond the infection site. The researchers observed that leaves and roots also exhibited early metabolic adjustments after the fungus was introduced into the wood, and these systemic changes were more pronounced in Chardonnay than in Gewurztraminer. This indicates that the tolerant cultivar mounts a rapid whole-plant response, reprogramming its metabolism far from where the pathogen is actually attacking. Such systemic metabolic reprogramming may reduce the overall impact of infection on host tissues, priming distant organs to cope with the physiological disruption that trunk disease causes as it progresses. The finding challenges the conventional focus on the wood alone and suggests that the fate of an infected vine is decided, in part, by how quickly the entire plant can reorganize itself.

The broader implications of this work extend into viticulture and breeding. Grapevine trunk diseases are increasingly concerning in the context of climate change, and with no fully resistant genotypes available, growers currently rely on costly and labor-intensive management practices, including the surgical removal of infected wood and the replanting of dead vines. Understanding the metabolic basis of tolerance could eventually help breeders select for cultivars with the rapid stilbene and oxylipin responses seen in Chardonnay, or guide the development of treatments that stimulate these pathways in susceptible varieties. The study also underscores that fungal isolate matters as much as host cultivar, which means that breeding and protection strategies may need to account for the local diversity of Botryosphaeriaceae populations in vineyards.

By integrating pathogen colonization data with untargeted metabolomics across three organs, two cultivars, two fungal isolates and multiple time points, the research provides a more integrated view of grapevine responses to Botryosphaeriaceae infection than has previously been available. The work, conducted at the Université de Haute Alsace with plants supplied by INRAE UMR SVQV in Colmar, France, and supported by the French Ministry of Higher Education and Research, demonstrates that tolerance to trunk disease is not a single trait but a dynamic, system-wide chemical performance. As vineyards face mounting pressure from a warming climate and evolving pathogen communities, studies like this one bring the field closer to understanding why some vines endure what destroys others, and how that resilience might one day be engineered or bred into the varieties the wine world depends on.

Subject of Research: Metabolomic responses of tolerant and susceptible grapevine cultivars to infection by the trunk disease fungus Neofusicoccum parvum

Article Title: Whole plant metabolomic analysis reveals contrasted responses in tolerant and susceptible grapevine cultivars to infection with different isolates of Neofusicoccum parvum, a major trunk disease fungus

Article References: Grau, A. L., Lacrampe, N., Goddard, M.-L., & Chong, J. (2026). Whole plant metabolomic analysis reveals contrasted responses in tolerant and susceptible grapevine cultivars to infection with different isolates of Neofusicoccum parvum, a major trunk disease fungus. BMC Plant Biology. https://doi.org/10.1186/s12870-026-09951-3

Image Credits: AI Generated

DOI: 10.1186/s12870-026-09951-3

Keywords: grapevine, Neofusicoccum parvum, Botryosphaeria dieback, grapevine trunk diseases, metabolomics, stilbenes, oxylipins, plant defense, Chardonnay, Gewurztraminer, fungal pathogen, viticulture

The study was conducted in 2024 at the Tekirdağ Viticulture Research Institute, in a rainfed vineyard whose elevation ranges from roughly 47.45 to 55.46 meters above sea level. The researchers divided the block into three slope positions—lower, middle, and upper—and examined vines grafted onto 5BB rootstock and trained in a bilateral cordon system. Because the vineyard receives no irrigation, every vine depends entirely on rainfall and soil water reserves, making the site a natural laboratory for observing how water availability, which typically declines upslope, translates into physiological and biochemical change.

Across the growing season, the team measured net photosynthesis, stomatal conductance, transpiration, and two measures of water use efficiency on eleven separate occasions, under climatic conditions spanning air temperatures of 30 to 38 degrees Celsius, relative humidity of 26 to 50 percent, and photosynthetically active radiation from 600 to 1800 micromoles per square meter per second. These repeated field measurements captured how the vines responded to genuinely fluctuating Mediterranean summer conditions rather than a single snapshot, giving the dataset unusual ecological realism for a physiological study.

The results revealed a clear stress gradient running up the hill. Vines at the upper slope position showed lower rates of photosynthesis and stomatal conductance than their lower-slope counterparts, but they achieved higher water use efficiency—both in terms of carbon gained per unit of water lost through stomata and per unit transpired. In other words, the upslope vines were running a conservative water-use strategy, keeping their stomata tighter to conserve moisture at the cost of slower carbon assimilation. This is a classic drought-avoidance signature, and its emergence along a purely topographic gradient underscores how strongly slope position governs vine water status even within a single, seemingly uniform vineyard.

Biochemical analyses reinforced the physiological picture. Chlorophyll a, chlorophyll b, and total chlorophyll contents all decreased as one moved upslope, consistent with stress-impaired pigment metabolism, while the ratio of chlorophyll a to b increased—a shift often associated with acclimation to harsher light and water conditions. Meanwhile, the stress-signaling hormone abscisic acid, total phenolics, and tannins all rose with slope position. Abscisic acid is the plant hormone that orchestrates stomatal closure under drought, so its upslope accumulation dovetails neatly with the observed reduction in stomatal conductance and the shift toward water conservation.

Perhaps the most striking results came from high-performance liquid chromatography profiling of individual phenolic compounds. Gallic acid, vanillic acid, syringic acid, caftaric acid, chlorogenic acid, caffeic acid, catechin, epicatechin, rutin, and kaempferol all accumulated more abundantly in vines at the middle and upper slope positions. These compounds are not merely metabolic byproducts; they form a chemically diverse defensive shield, scavenging reactive oxygen species generated when photosynthesis is stressed, reinforcing cell walls, and deterring pathogens. Their upslope enrichment suggests that vines facing tighter water budgets invest more heavily in secondary metabolism, effectively trading growth and carbon gain for chemical resilience.

To test whether these patterns were statistically robust rather than incidental, the researchers applied principal component analysis and hierarchical clustering to the full dataset. Both multivariate approaches cleanly separated the three slope positions, with the upper slope clustering alongside the key stress markers—elevated abscisic acid, phenolics, tannins, and water use efficiency, and depressed chlorophyll and photosynthesis. Bivariate correlation analysis added a further layer: photosynthesis was negatively linked with abscisic acid, intrinsic water use efficiency, and phenolic concentrations, exposing an explicit trade-off between carbon assimilation and defense investment. A vine cannot maximize both at once, and the slope dictates where along that continuum each plant sits.

Proline, a compatible osmolyte that helps cells maintain turgor under water deficit, was also measured, along with antioxidant capacity assessed by the TEAC assay. Together with the hormone and pigment data, these measurements sketch a coherent portrait of ‘Adakarası’ as a cultivar with genuine drought resilience. Rather than collapsing under upslope stress, the vines reprogrammed their physiology and secondary metabolism in an apparently coordinated fashion—closing stomata, signaling with abscisic acid, accumulating osmoprotectants, and ramping up phenolic defenses. Such integrated plasticity is precisely the trait profile that breeders and growers seek as climate zones for quality viticulture shift poleward and upward.

The work also speaks to the concept of terroir, the oft-invoked idea that place imprints itself on wine. While terroir is usually discussed in terms of soil, mesoclimate, and tradition, this study demonstrates that within-vineyard topography alone can generate measurable gradients in leaf chemistry and gas exchange. Slope-driven differences in soil water content, erosion, and solar exposure appear sufficient to push vines into distinct physiological and metabolic states—states that plausibly influence berry composition and, ultimately, wine character. For precision viticulture, the message is that slope position deserves attention not just at the scale of hillsides and valleys, but within individual blocks.

For the Turkish wine industry, the stakes are concrete. ‘Adakarası’ is a heritage variety native to Thrace, and preserving its cultivation under a warming, drying climate depends on understanding where it thrives and why. By identifying the upper slope as a zone of conservative water use and heightened phenolic defense, and the lower slope as one of freer gas exchange, the study offers growers a physiological rationale for vineyard design, rootstock and scion matching, and future irrigation decisions. More broadly, it shows that a humble hillside, read carefully through the language of pigments, hormones, and polyphenols, can teach us how one of humanity’s oldest crops negotiates the stress of a changing world.

Subject of Research: Slope-induced abiotic stress effects on leaf phenolics and physiological activity in the grapevine cultivar Vitis vinifera 'Adakarası'

Article Title: Leaf phenolics, physiological activity and abiotic stress relationship induced by slope in grapevine: Vitis vinifera ‘Adakarası’

Article References: Candar, S., Alço, T., Seçkin, G. U., Şahin, E., Demirağ, O., Güngör, T. A., Bahar, E., & Korkutal, İ. (2026). Leaf phenolics, physiological activity and abiotic stress relationship induced by slope in grapevine: Vitis vinifera ‘Adakarası’. Plant Biosystems, 160(4), Article 217. https://doi.org/10.1007/s44473-026-00228-z

Image Credits: AI Generated

DOI: 10.1007/s44473-026-00228-z

Keywords: grapevine, Vitis vinifera, Adakarası, drought stress, phenolic compounds, abscisic acid, photosynthesis, water use efficiency, chlorophyll, slope gradient, viticulture, climate change

The research team, led by Yuanqian Ni and supervised by Zhaosen Xie and Youmei Li, took advantage of a natural contrast between two grapevine cultivars. ‘Einset Seedless’, abbreviated ENT in the study, tends to produce inflorescences at the lower tendril nodes of its lateral shoots, while ‘Pinot Noir’, or PN, overwhelmingly produces true tendrils at the equivalent positions. By comparing these two varieties node by node, the researchers could ask a precise question: when the same meristem faces the same choice, what is different inside the cells of a plant that flowers versus a plant that climbs?

The homology of tendrils and inflorescences in grapevine is well established. Classical anatomical work showed that both organs arise from the same uncommitted primordia on lateral shoots, and earlier molecular studies found that floral meristem identity genes are expressed even in developing tendrils, as if the tendril were an inflorescence that never quite committed to flowering. Decades ago, pioneering experiments by Srinivasan and Mullins demonstrated that isolated tendrils could even be coaxed into forming inflorescences in vitro, hinting that the switch between the two fates is chemically malleable. What the new study adds is a genome-wide, node-resolved map of the transcriptional landscape in which that switch is thrown.

The team first quantified the anatomical difference, confirming that ENT achieved a markedly higher flowering rate at tendril nodes one through four than PN did. They then harvested tissue from specific tendril nodes in both cultivars and performed RNA sequencing, using an intersection and exclusion strategy to isolate 549 differentially expressed genes that distinguish inflorescence-prone from tendril-prone tissue. When those genes were subjected to Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis, two signaling systems rose to the top: plant hormone signal transduction and mitogen-activated protein kinase cascades. In other words, the decision between flower and tendril appears to be broadcast through the cell by hormonal and phosphorylation-based communication channels rather than by a single master switch.

To move beyond a static snapshot, the researchers applied Mfuzz clustering, a computational method that groups genes according to the shape of their expression trajectories across a developmental series. This revealed that multiple expression programs rise and fall in patterns that track the flowering gradient running along the shoot, from the most inflorescence-prone basal nodes to the tendril-dominated upper nodes. Across every dynamic cluster, plant hormone signal transduction emerged as the predominantly enriched pathway, underscoring that phytohormones are not merely participants in this process but sit at its very center.

Guided by that transcriptional evidence, the team measured the endogenous levels of two key hormones directly in the nodal tissues: zeatin, the classic cytokinin that generally promotes reproductive meristem activity and flowering, and gibberellin GA3, which is more often associated with stem elongation and vegetative growth. The striking result was that the ratio between the two, rather than the absolute level of either hormone alone, paralleled the flowering gradient across nodes and correlated with the gene expression trajectories identified by clustering. A high zeatin-to-GA3 ratio accompanied the inflorescence-prone state, while a lower ratio accompanied tendril formation. This provides concrete physiological support for a cytokinin-gibberellin interaction model in which the balance between these two antagonistic hormones helps determine whether a lateral meristem matures into a flower-bearing branch or an elongating, climbing tendril.

The hormone story dovetails with a long history of viticultural observation. Growth retardants that suppress gibberellin synthesis have been known since the 1960s to influence fruitfulness in grapes, and cytokinin treatments have been shown to convert tendril primordia into inflorescences, with the success of that conversion depending on the genotype. The famous ‘green revolution’ dwarfing mutation in grapevine, which links reduced gibberellin response with enhanced floral induction, fits the same pattern. What the new transcriptomic work contributes is the mechanistic scaffolding: it shows that the endogenous hormone ratio in the organ itself, not just externally applied chemicals, tracks the developmental outcome, and that the downstream transcriptional response is organized into coherent, flowering-correlated trajectories.

Within the pool of differentially expressed transcription factors, one gene stood out. A MADS-box gene named FRUITFULL-LIKE, or VvFUL-L, was markedly upregulated in PN tendrils compared with inflorescence-prone ENT tissue. MADS-box transcription factors are the celebrated architects of floral development, and the FRUITFULL lineage in particular is known from arabidopsis and tomato to regulate meristem identity, flowering time, and inflorescence architecture. To probe what VvFUL-L might be doing, the team overexpressed it heterologously in arabidopsis. The transgenic plants flowered earlier than normal and produced inflorescences with reduced branching, phenotypes consistent with a role in steering lateral meristem development. The authors suggest that VvFUL-L may help tip the uncommitted grapevine lateral meristem toward the tendril program, although confirming that function in grapevine itself will require further work in the crop.

The agronomic implications are immediate and tangible. Excessive tendril growth increases vineyard management costs because tendrils must be removed or redirected by hand or machine, and they can girdle shoots and interfere with canopy machinery. If the cytokinin-to-gibberellin ratio and regulators like VvFUL-L can be manipulated, whether through breeding, targeted gene editing, or growth regulator programs, growers could eventually favor inflorescence formation at the basal nodes where fruit is wanted and suppress the runaway tendril production that wastes the vine’s resources. The study’s authors frame their findings as a foundation for future molecular and breeding studies aimed at managing tendril growth, and the identification of a small set of hormone-responsive genes and a candidate transcriptional regulator is exactly the kind of entry point such efforts require.

There is also a broader evolutionary resonance in the result. Across the grape family, the Vitaceae, tendrils and inflorescences are thought to be variations on a single ancestral organ, and expression studies of APETALA1, FRUITFULL, FLOWERING LOCUS T, and LEAFY orthologs throughout the family support that homology. The Yangzhou study now supplies a quantitative, node-by-node account of how that shared developmental potential is partitioned within a single plant, showing that the fate of each primordium is written in the language of hormone ratios and read out by MADS-box transcription factors. A curling tendril on a Pinot Noir vine, in this light, is not a failed flower but an alternative interpretation of the same instructions, and researchers are learning to read, and perhaps one day to rewrite, the grammar behind that choice.

Subject of Research: Molecular regulation of tendril versus inflorescence development in grapevine lateral shoots

Article Title: Comparative transcriptomics reveals hormone signaling and MADS-box genes in divergent development of inflorescences and tendrils in grapevine lateral shoots

Article References: Ni, Y., Bian, Y., Xie, Z., & Li, Y. (2026). Comparative transcriptomics reveals hormone signaling and MADS-box genes in divergent development of inflorescences and tendrils in grapevine lateral shoots. Plant Cell Reports, 45(9), Article 270. https://doi.org/10.1007/s00299-026-03957-9

Image Credits: AI Generated

DOI: 10.1007/s00299-026-03957-9

Keywords: grapevine, Vitis vinifera, tendrils, inflorescences, cytokinin, gibberellin, MADS-box, VvFUL-L, transcriptomics, hormone signaling, flowering, viticulture

The research, led by Maajid Bashir of Chandigarh University together with colleagues at Taibah University, Imam Mohammad Ibn Saud Islamic University, and the Islamic University of Madinah, tackles a well-known weakness of conventional image classifiers. Standard convolutional neural networks treat every region of an input image with roughly equal importance, which means that when a leaf photograph contains background clutter, uneven lighting, shadows, or healthy tissue surrounding a small lesion, the network can be distracted by irrelevant features. The team’s solution was to graft a Convolutional Block Attention Module, known as CBAM, onto a pretrained ResNet50 backbone, allowing the network to explicitly learn where to look within each image before committing to a classification decision.

CBAM works through two complementary attention mechanisms applied in sequence. The first is channel attention, which asks which feature channels, and therefore which kinds of visual patterns such as color gradients or texture edges, are most informative for the task at hand. The second is spatial attention, which highlights the specific pixel regions where those informative patterns occur. By combining both, the module effectively teaches the network to suppress background noise and concentrate its representational capacity on the pathological signatures of disease, such as the characteristic necrotic spots or chlorotic halos that distinguish one grape pathogen from another. The authors report that attention map visualizations confirm the model localizes genuine disease symptoms rather than latching onto photographic artifacts, an important sanity check for any model intended for real-world deployment.

The backbone of the system is ResNet50, a fifty-layer residual network pretrained on the ImageNet dataset, a collection of more than a million labeled natural images. Transfer learning of this kind is a cornerstone of modern applied deep learning: rather than teaching a network to see from scratch, researchers reuse the rich general-purpose visual features learned on ImageNet, such as edge detectors, texture analyzers, and shape encoders, and then adapt them to a specialized domain like plant pathology. This strategy dramatically reduces the amount of labeled agricultural data required and shortens training time, both critical considerations for research groups without access to massive computing farms.

Equally important to the reported performance is the training regimen the authors adopted. Instead of fine-tuning the entire network at once, they employed a controlled two-phase procedure. In the first phase, the pretrained ResNet50 layers were frozen, and only the newly added CBAM module and a custom classification head were trained. This allows the task-specific components to learn meaningful disease-discriminating features without the risk of large gradient updates corrupting the carefully calibrated pretrained weights. In the second phase, the entire network was unfrozen and fine-tuned end-to-end using a reduced learning rate, a technique that stabilizes convergence and improves the model’s ability to generalize to images it has never seen. This staged approach is widely regarded as best practice in transfer learning, and the results here reinforce its value.

The evaluation was deliberately conducted on two very different benchmarks. The first, PlantVillage, is a widely used public dataset of plant leaf photographs taken under controlled laboratory conditions, with uniform backgrounds and consistent lighting. On this dataset, the CBAM-augmented ResNet50 achieved an accuracy of 99.33 percent. The second, the Niphad Grape Leaf Disease dataset, or NGLD, is a far more demanding collection of 2,726 high-quality grape leaf images gathered between 2023 and 2025 and annotated by agricultural experts, capturing the messy variability of in-the-wild photography. Remarkably, the model maintained an accuracy of 99.32 percent on this challenging dataset, indicating that its performance is not an artifact of easy, curated imagery.

To contextualize these numbers, the researchers benchmarked their model against a lineup of established architectures, including the standard ResNet50 without attention, VGG16, DenseNet121, Xception, InceptionV3, and MobileNetV2. The attention-augmented version outperformed all of these reference models by a substantial margin on both datasets. This comparison isolates the contribution of the attention mechanism: the underlying feature extractor is the same in the plain ResNet50 baseline, so the performance gap can be attributed to the network’s enhanced ability to focus on salient pathological regions. For practitioners, the message is that architectural refinements, not just bigger models or more data, can deliver decisive gains in agricultural image analysis.

The implications for precision agriculture are considerable. Grape leaf diseases, including downy mildew, powdery mildew, black rot, and anthracnose, can devastate yields if not caught early, and the global viticulture sector, which the International Organisation of Vine and Wine values in the hundreds of billions of euros annually, depends on timely intervention. A smartphone-based diagnostic tool powered by a model like this one could allow farmers, agronomists, and extension workers to photograph a suspect leaf and receive an accurate diagnosis within seconds, enabling targeted fungicide application rather than blanket spraying. That shift would reduce chemical inputs, lower costs, limit environmental damage, and slow the development of pesticide resistance in pathogen populations.

The study also highlights a broader trend in machine learning research: the migration of attention mechanisms, first popularized in computer vision through modules like CBAM and later through transformer architectures, into domain-specific applications. Attention gives models a form of interpretability that plain classifiers lack, because the learned attention maps can be inspected to verify that the network is responding to biologically meaningful features. In plant pathology, where a misdiagnosis could trigger unnecessary treatments or missed outbreaks, this transparency is not a luxury but a requirement for trust and regulatory acceptance.

Challenges remain before such systems become routine in the field. Real-world deployment will require handling variable camera quality, occlusion, disease co-occurrence on a single leaf, and disease stages not represented in training data, and the authors note that no new datasets were generated in the current study, meaning future work will need to test generalization across additional regions, cultivars, and pathogens. Nevertheless, the near-identical accuracies achieved on controlled and in-the-wild data represent a striking demonstration that attention-augmented transfer learning can deliver robust, cost-effective, and deployable plant disease diagnostics. As the authors conclude, such models stand ready to serve as effective substitutes for manual inspection, bringing precision agriculture a significant step closer to everyday practice in vineyards worldwide.

Subject of Research: Deep learning-based detection of grape leaf diseases using an attention-augmented ResNet50 model

Article Title: Cbam-augmented ResNet for high-accuracy grape leaf disease detection

Article References: Bashir, M., Reshi, A. A., Shafi, S., Aljubayri, I., & Khan, M. Z. (2026). Cbam-augmented ResNet for high-accuracy grape leaf disease detection. International Journal of Machine Learning and Cybernetics, 17(9), Article 451. https://doi.org/10.1007/s13042-026-03235-w

Image Credits: AI Generated

DOI: 10.1007/s13042-026-03235-w

Keywords: deep learning, transfer learning, attention mechanism, CBAM, ResNet50, grape leaf disease, precision agriculture, convolutional neural networks, computer vision, plant pathology, image classification, viticulture

The study focused on Superior Seedless, one of Egypt’s most economically important table grape varieties, prized for its early maturity, high market price, and strong demand in European and Arabian export markets. Egypt is the world’s fourth-largest producer of table grapes, and its grape exports grew eighteenfold between 2001 and 2016, a boom that has intensified pressure on growers to raise both yields and fruit quality. Yet vineyards in recently reclaimed desert areas face mounting challenges, including declining soil organic matter, reduced soil permeability after years of continuous cultivation, and the environmental toll of heavy mineral fertilizer use. Nitrogen, phosphorus, and potassium are the workhorses of grapevine nutrition: nitrogen drives the synthesis of proteins, enzymes, chlorophyll, and phenolic compounds; phosphorus underpins cell division, DNA and RNA synthesis, and energy metabolism; and potassium governs photosynthetic activity, water uptake, long-distance transport in the xylem and phloem, and stress tolerance.

The research team, led by Mohamed Farouk Ahmed of the Egyptian Atomic Energy Authority together with Noha Eid Eliwa and Shimaa M.M. El-Mogy of the Agricultural Research Center, set out to test whether radiation treatment could unlock more of the nutritional value locked inside municipal cattle manure. The experiment took place in a private vineyard in El-Khatatba, Menoufiya governorate, on ten-year-old Superior Seedless vines grown in sandy loam soil under drip irrigation and cane-pruned to a double Y trellis with a bud load of 72 buds per vine. Over three consecutive seasons beginning in 2020, with the first season serving as an introductory period to clear residual effects of previously applied fertilizers, the researchers applied manure at 5 kilograms per vine as a soil drench around the roots in late January.

The key innovation was the radiation step. Batches of cattle municipal manure were exposed to gamma rays from a cobalt-60 source at the National Centre for Radiation Research and Technology, receiving doses of 0, 5, 10, or 15 kilograys. Previous work has shown that gamma irradiation accelerates the decomposition of organic material in compost, increasing its degree of maturity. As the radiation dose rises, the compost’s organic carbon content falls while total nitrogen increases, lowering the carbon-to-nitrogen ratio, a key indicator of compost readiness. Irradiated compost also shows elevated phosphorus and potassium levels. In essence, the radiation pre-digests the organic matter, making nutrients more readily available for mineralization in soil and, ultimately, for uptake by plant roots.

To this irradiated manure the researchers added, in some treatments, microbein, a biofertilizer produced by Egypt’s Ministry of Agriculture containing three beneficial bacterial species: Azotobacter and Azospirillum, which fix atmospheric nitrogen in a non-symbiotic form that plants can use, and Bacillus megaterium, a potassium-solubilizing bacterium that excretes organic acids capable of dissolving rock potassium minerals such as micas, illite, and orthoclases. These microbes also release growth-promoting substances including indole acetic acid, gibberellins, thiamine, and sugars that improve soil quality. The biofertilizer was applied at 30 grams per vine two weeks after the manure, in a circle around each vine, then covered with soil and irrigated. The trial followed a randomized complete block design with eight treatments, three replicates, and three vines per replicate, and the data were analyzed by one-way analysis of variance with means separated using Duncan’s multiple range test at the 0.05 probability level.

The results painted a striking picture of synergy between radiation and microbes. Non-irradiated manure combined with microbein improved vegetative growth, increasing shoot length and diameter as well as leaf number and area, but left yield attributes, cluster and berry characteristics, and leaf mineral content largely unchanged. Irradiated manure, by contrast, whether applied alone or with microbein, significantly improved nearly every measured trait. Shoots grew longer and thicker, vines carried more leaves with larger blade areas, and berries grew longer, wider, and heavier. The berries also accumulated more total soluble solids, the sugars measured in degrees of refractometer reading, while titratable acidity expressed as tartaric acid declined, pushing the TSS-to-acid ratio upward, a reliable proxy for table grape eating quality. Clusters became longer, wider, and heavier, and yield per vine climbed accordingly.

The magnitude of the response scaled with radiation dose, and the highest dose combined with microbein delivered the most dramatic numbers. The winning treatment increased leaf nitrogen by 14 percent, phosphorus by a remarkable 66 percent, and potassium by 11 percent, while raising leaf number, shoot length, and shoot diameter by 11 percent each. Total soluble solids rose 9 percent, berry weight 20 percent, and cluster weight and yield 21 percent, all while acidity dropped 25 percent. Leaf chlorophyll content, measured nondestructively with a Minolta SPAD-502 meter, also rose with both irradiation dose and microbein addition, reflecting the improved nitrogen status of the foliage. Notably, the effect of irradiation at 10 and 15 kilograys was more pronounced than the effect of the biofertilizer itself: manure irradiated at 15 kilograys and applied alone outperformed manure irradiated at 10 kilograys combined with microbein, underscoring how powerful the radiation pretreatment is on its own.

Why does zapping manure with gamma rays make it such a better fertilizer? The researchers attribute the effect to the way ionizing radiation accelerates the breakdown of complex organic compounds in the compost, hastening decomposition and increasing maturity. More mature compost mineralizes nitrogen and carbon faster, releasing nutrients in forms roots can absorb during the critical early growing season. Humic substances within the compost further enrich the soil with calcium, phosphorus, potassium, and nitrogen, and improved soil physical, chemical, and biological properties, including better drainage, moderated pH, and enhanced nutrient availability, create a more hospitable rhizosphere. When microbein is added to this irradiated substrate, the benefits compound: the improved physicochemical conditions of the soil support larger and more diverse communities of beneficial microbes, while the bacteria themselves fix nitrogen, solubilize phosphorus and potassium, produce growth regulators, and suppress root pathogens. Organic fertilization also raises soil organic carbon as microbial activity increases, buffering soil pH toward neutral and expanding the pool of available macro- and micronutrients.

The findings align with earlier studies showing that irradiated compost and biofertilizers can improve growth, yield, and fruit quality in other crops, including previous work on Superior Seedless grapevines and Valencia oranges, where fertilization with irradiated compost enhanced fruit physical and chemical characteristics, vegetative growth, and leaf mineral content, with effects intensifying at higher radiation doses. Comparable research on French Black grapes found that combining organic fertilizer with biofertilizers such as Bacillus megaterium and mycorrhizal fungi significantly increased yield and cluster number, length, and weight relative to organic fertilizer alone. Studies on Thompson Seedless and Flame Seedless grapes similarly reported that biofertilizers added to integrated mineral and organic programs boosted cluster weight, soluble solids, and petiole nitrogen, phosphorus, and potassium while reducing acidity, although responses vary with the type of organic material, the irradiation dose, and the microbial strains involved.

For a world searching for ways to feed a growing population without degrading its soils, the study offers a compelling template. Organic and microbial fertilizers improve the physical, chemical, and biological fertility of the soil while reducing dependence on synthetic fertilizers and the pollution risks they carry, and long-term viticulture in particular demands careful soil management to prevent the organic matter losses that erode yields. The Egyptian team concludes that 15 kilogray-irradiated cattle municipal manure applied with microbein is a viable alternative to chemical fertilization for seedless grapevines, one that ensures better fruit quality, higher yield, and improved nutritional status of the vines. If the approach proves scalable and economical beyond the research plot, the humble manure pile, given a brief trip through a gamma irradiator and a sprinkle of nitrogen-fixing bacteria, could become a cornerstone of sustainable viticulture in Egypt’s expanding desert vineyards and, perhaps, far beyond them.

Subject of Research: Effects of gamma-irradiated cattle manure and microbein biofertilizer on the growth, yield, and fruit quality of Superior Seedless grapevines

Article Title: The synergistic effect of irradiated manure and bio fertilizer on growth and fruit quality of superior seedless grapevines

Article References: Ahmed, M. F., Eliwa, N. E., & El-Mogy, S. M. (2025). The synergistic effect of irradiated manure and bio fertilizer on growth and fruit quality of superior seedless grapevines. BMC Agriculture, 1(1), Article 15. https://doi.org/10.1186/s44399-025-00009-7

Image Credits: AI Generated

DOI: 10.1186/s44399-025-00009-7

Keywords: grapevines, gamma irradiation, cattle manure, biofertilizer, microbein, Superior Seedless, compost maturity, viticulture, organic fertilization, plant nutrition, Egypt, sustainable agriculture

The study, published in the International Journal of Data Science and Analytics, starts from a candid acknowledgment of a problem that plagues much of agricultural machine learning: the labels. True pathological observations of infection are rare and expensive, so the researchers instead derived their disease-risk annotations from fungicide-treatment records, the practical decisions growers made to protect their vines. This makes the labels management-oriented risk proxies rather than direct measurements of pathogen presence. The team is explicit about this distinction, and it matters, because a treatment record reflects a grower’s judgment as much as the biology of the disease itself. Any model trained on such data learns to predict management behavior conditioned on weather, not infection events in the strict epidemiological sense.

The proxy-label problem had a second wrinkle. The dataset contained separate annotations for downy mildew and powdery mildew, but the two agreed on 98.9 percent of observed days, and in the validation and test years they were identical. Rather than pretending the two diseases share an epidemiology, which they demonstrably do not, the researchers collapsed the labels into a single operational any-mildew target. This pragmatic choice reflects how growers actually experience the threat: either mildew is a concern today or it is not. But the authors are careful to note that the combined target cannot support claims about distinguishing the two diseases, and biological interpretation of model outputs must remain cautious.

With the target defined, the team built an event-based formulation. A new risk event is counted only after a minimum disease-free gap has elapsed, preventing a single prolonged risky period from fragmenting into dozens of trivial episodes. Models are then asked a genuinely anticipatory question: will a new event begin within the following three to seven days? This framing, borrowed from the early-prediction literature on time series, directly measures what daily classification cannot, namely whether a model can see an episode coming before it starts. The data were split chronologically, with 2020 and 2021 for training, 2022 for validation, and 2023 for testing, ensuring that models were always forecasting into the future rather than interpolating within a shuffled mixture of seasons.

The model comparison spanned the standard spectrum of tabular and sequence learners. Logistic regression provided a linear baseline, XGBoost represented gradient-boosted decision trees, and two deep sequence architectures, long short-term memory networks and temporal convolutional networks, captured temporal dependencies in the agro-meteorological streams. Crucially, the researchers compared three input representations: calendar-only features such as day-of-year, environmental-only variables from the vineyard sensors, and the two combined. This design allowed them to isolate how much predictive power genuinely comes from the weather, a question that many published disease-forecasting studies never ask.

The results were humbling for the environmental data. Calendar-only models were highly competitive, with calendar-based LSTM and TCN models detecting 7 of the 8 test events, while a simple monthly climatological baseline, essentially a lookup table of historical risk by month, still detected 6. Environmental-only models were substantially weaker across the board, and adding environmental variables to calendar information produced inconsistent gains that varied by model family rather than delivering a uniform improvement. In other words, much of what the sophisticated models learned may have been the seasonal rhythm of vineyard disease risk, a pattern that a calendar already encodes. The authors’ SHAP analysis, which attributes model predictions to input features, confirmed that the fitted models used both seasonal and meteorological information, but it did not reveal a single dominant or consistently interpretable environmental precursor, no silver-indicator variable that reliably signals an impending outbreak.

This seasonal confounding is arguably the study’s most important cautionary finding. Mildew risk in temperate vineyards follows the growing season, so any model with access to the date can achieve respectable performance without understanding humidity, leaf wetness, or temperature at all. Without explicit seasonal baselines, reported accuracies can flatter models that are merely reproducing climatology. The Skopje team’s insistence on comparing against calendar-only and climatological baselines offers a template for how disease-forecasting studies should be evaluated, and their results suggest that several celebrated successes in the literature may deserve a second look under this stricter light.

The event-based formulation also revealed trade-offs that daily classification obscures. When the researchers compared their event-based models against conventional daily-status classifiers, they found the two objectives capture different things and exhibit different profiles in point-wise accuracy, event coverage, and alert burden. A model can score well on daily labels while missing the onset of episodes entirely, or catch most onsets while flooding the grower with false alarms. Neither formulation proved uniformly superior, which means the choice between them should be driven by the operational goal: a grower deciding when to scout needs anticipation, while a system scheduling routine monitoring may prefer stable daily estimates. Sensitivity analyses reinforced this picture, showing that results depend materially on the historical window length, the prediction horizon, and the event-separation gap, parameters that previous studies often fixed arbitrarily.

For the wine industry, the practical implications are sobering but useful. The best models anticipated most risk episodes several days ahead, which is genuinely valuable lead time for scheduling sprays and scouting, and targeting interventions more precisely can reduce fungicide use, cost, and environmental impact. But the finding that environmental sensors added inconsistent value over the calendar suggests that in-vineyard IoT weather stations, increasingly common in precision-agriculture deployments, must justify their cost through demonstrable gains over seasonal baselines on each new dataset, not through generic claims. The study also underscores the value of the underlying agro-meteorological dataset introduced by Arvanitis and colleagues in 2025, whose IoT-derived environmental records made the analysis possible, and the Skopje team states that their processed data, event annotations, engineered features, and experimental scripts will be released in a public repository, with interim availability from the corresponding author.

More broadly, the work contributes to a growing movement in machine learning toward event-oriented evaluation, in which systems are judged not on per-instance accuracy but on their ability to anticipate discrete, consequential occurrences, whether disease outbreaks, forest fires, or equipment failures. By defining events explicitly, enforcing minimum gaps, and measuring event coverage against alert burden, the vineyard study offers a methodological blueprint that extends well beyond oenology. Its honest treatment of proxy labels, its insistence on seasonal baselines, and its refusal to overclaim interpretability from SHAP attributions model the kind of rigor that agricultural AI will need as it moves from papers to practice. The next mildew wave may still be hard to predict, but thanks to this work, we now know better how to ask the question, and better how to tell whether an answer is real.

Subject of Research: Event-based machine learning early warning of vineyard mildew disease risk from agro-meteorological time series

Article Title: Event-based early warning of vineyard disease risk from environmental time series

Article References: Dimitrovski, I., Kitanovski, I., Davcev, D., Kalajdziski, S., & Mitreski, K. (2026). Event-based early warning of vineyard disease risk from environmental time series. International Journal of Data Science and Analytics, 22(1), Article 317. https://doi.org/10.1007/s41060-026-01282-8

Image Credits: AI Generated

DOI: 10.1007/s41060-026-01282-8

Keywords: machine learning, viticulture, downy mildew, powdery mildew, early warning systems, agro-meteorological time series, LSTM, temporal convolutional networks, XGBoost, SHAP, precision agriculture, disease forecasting