The proposed approach responds to an increasingly diverse research landscape. Cambridge Dictionary defines botany as “the scientific study of plants,” while Treccani describes it as the branch of biology that studies plant organisms. Contemporary plant research, however, includes basic and applied studies of plant uses and extracts, including work in agronomy and medicine that may be described as “applied botany.”

Botany’s disciplinary history is also closely connected with its use and development. The word “botany” was coined by Homer in The Iliad in the eighth century BC, according to the article. Botany became an independent science from medicine during the Renaissance, while the cultivation of plants for food and aesthetic purposes dates back at least 11,700 years. These overlapping histories complicate any attempt to assign plant-related work to one field.

The article cites signs of institutional change in academic botany. Pure botany is reportedly taught less frequently, basic research has become rarer because of funding limitations, and applied research has attracted greater financial support. Some scholars have described these developments as a claim for “the end of botany,” while others report that foundational subjects such as nomenclature are receiving less attention. At the same time, “botany” and “botanist” are increasingly being replaced in academia by “plant biology” and “plant biologist.”

Outside universities, the terminology is used more broadly, sometimes for anyone working with plants for any purpose. Agronomists, arboriculturists, gardeners, herbalists, phytotherapists and plant pharmacologists may all fall under the informal label. The article argues that this broad usage can obscure professional distinctions and lead members of the public to seek advice from people whose training may not include a strong foundation in botany. Botanical gardens face a related perception problem, as their scientific functions may be mistaken for those of ornamental gardens or parks.

To represent the overlap among botany, agronomy and medicine, the article uses a ternary plot, a barycentric diagram in which three variables sum to a constant and their relative proportions are represented as positions within an equilateral triangle. Specialized research areas occupy overlapping positions rather than fitting neatly into one discipline. The author identifies a “Triangle of Botany,” defined as the area in which basic plant science remains the predominant component.

Under the proposed definitions, a botanist is a professional or scientist whose primary research focus is basic research on plant organisms or research applied to nature conservation. A botanical garden is a space dedicated to cultivating, conserving, studying and displaying plants, based primarily on scientific principles grounded in basic research, conservation research or both. This emphasis on documented living collections, scientific research, conservation, display and education broadly accords with the definition used by Botanic Gardens Conservation International.

The framework has several stated implications. Researchers focused mainly on the uses of plants rather than plants themselves would be classified as agronomists or medical researchers rather than botanists. A facility displaying and documenting only crops would be a crop collection field rather than a botanic garden, while one devoted only to ornamental plants would be a garden or park. In wider society, the proposed test is whether expertise concerns plants themselves or their uses, with the latter associated with agricultural, horticultural, herbal and pharmacological professions.

The article also distinguishes interdisciplinary, multidisciplinary and transdisciplinary approaches. Interdisciplinarity integrates multiple disciplines into one research activity, whereas multidisciplinarity studies a topic through several disciplinary perspectives at once; both remain within disciplinary research. Transdisciplinarity extends beyond and integrates disciplines, represented in the framework by a central region in which no single discipline dominates. Under this classification, phytochemistry, plant biotechnology and the biology of agronomic and forest plants retain a predominant botanical component, whereas transdisciplinary work includes topics such as alimurgical plants and aerobiology. Research dominated by agronomic or medical components is characterized as interdisciplinary or multidisciplinary agronomic or medical research rather than applied botany. The contribution is a proposed conceptual classification, supported by illustrative examples rather than an empirical evaluation of how consistently practitioners or institutions meet its definitions.

Subject of Research: A conceptual framework for defining botany, botanists and botanical gardens

Article Title: Some claim for the end of Botany… but what is Botany today?

Article References: Peruzzi, L. (2025). Some claim for the end of Botany… but what is Botany today?. Italian Botanist, 19, 15-20. https://doi.org/10.3897/italianbotanist.19.145382

Image Credits: AI Generated

DOI: 10.3897/italianbotanist.19.145382

Keywords: botany, plant science, botanical gardens, agronomy, basic research, conservation, interdisciplinarity, discipline boundaries, plant nomenclature, Some, claim, today

The study, published in the Journal of Agriculture and Food Research, was led by Curtis B. Adams of the USDA Agricultural Research Service together with Christopher W. Rogers, Juliet M. Marshall, and Kurtis L. Schroeder of the University of Idaho. Rather than comparing the two species as monoliths, the team dissected them into their market classes: three classes of spring barley (feed, food, and malt) and three classes of spring wheat (hard red, hard white, and soft white). These classes exist because selective breeding has pushed each species toward distinct end uses—malt barley must stay below a 13.0 percent grain protein ceiling for premium pricing, while hard red spring wheat is bred to exceed a 14.0 percent protein floor. Those breeding targets, the researchers suspected, might shape not only grain chemistry but also how each class responds to environment.

The experimental scale was formidable. The six crop classes were grown side by side at seven locations across northern and southern Idaho—Aberdeen, Ashton, Craigmont, Genesee, Idaho Falls, Rupert, and Soda Springs—spanning elevations from 823 to 1,760 meters, in the 2018 and 2019 growing seasons. The sites ranged from fully irrigated fields receiving complete evapotranspiration replacement to dryland plots relying solely on precipitation, and soil organic matter ranged from 9 to 40 grams per kilogram. Each class was managed with its own best practices under randomized complete block designs with four replicates, and three varieties per class were selected to create a balanced dataset of 14 distinct environments. At physiological maturity, aboveground biomass was sampled from non-border plots, hand-threshed into grain and residue, and analyzed for carbon, nitrogen, phosphorus, potassium, calcium, magnesium, sulfur, zinc, and manganese using Dumas combustion and ICP-OES at a proficiency-certified laboratory.

The headline yield finding vindicates part of the folk wisdom. When grain yields were regressed against a quantitative modeled estimate of each environment’s productivity, the slopes of the species-level relationships diverged significantly. In the lowest-yielding conditions, barley systems were modeled to produce up to 0.62 megagrams per hectare more grain than wheat. But at the high-yielding end of the spectrum, wheat was modeled to exceed barley by up to 0.36 megagrams per hectare. The pattern echoes a 2024 analysis of Argentine variety trials by Gimenez and colleagues, which found barley out-yielded early-maturing wheat in stressful environments while late-flowering wheat slightly out-yielded barley in productive ones. It contrasts, however, with a 2023 synthesis by Slafer and Savin covering worldwide comparisons, which detected no consistent species advantage at any yield level.

Adams and colleagues argue that these discrepancies resolve once breeding investment is taken into account. In their Idaho data, the species-level yield interaction was actually driven by class-level nuance: food barley trended low while malt barley trended consistently high, and the white wheat classes exhibited greater environmental responsiveness, yielding relatively poorly in lean fields but strongly in productive ones. Malt barley has received extensive breeding attention in Idaho; food barley comparatively little. The researchers suggest that the comparative performance of wheat and barley in any given region depends on the relative breeding investment in the germplasm being compared, the broader non-yield goals of that breeding, and the biochemical traits—like grain protein—that classes are selected to differ in. Regional species differences, they note, may simply offset one another when data are pooled worldwide.

The mineral nutrition results may prove even more consequential than the yield findings. Grain nitrogen concentration differed dramatically and predictably among classes, with hard red wheat highest, feed and malt barley lowest, and soft white wheat the only wheat class overlapping with barley. Beyond nitrogen, wheat grain generally accumulated higher concentrations of phosphorus, magnesium, sulfur, zinc, and manganese than barley, while barley grain carried more potassium. The manganese contrast was the most striking: barley grain Mn was low and nearly steady across environments, increasing only 5.38 milligrams per kilogram from low to high environmental Mn index, while wheat grain Mn was far greater overall and highly responsive, climbing 44.1 milligrams per kilogram across the same range. The species gap widened from roughly 13 to 50 milligrams per kilogram as environmental Mn availability increased. The researchers speculate that differences in rhizosphere acidification—a mechanism underlying luxury Mn uptake in known hyperaccumulator species—might explain the divergence.

These grain chemistry differences carry direct health implications. Mineral bioavailability depends on more than density alone, but the results suggest whole-grain wheat offers a greater source of nutritional minerals than barley, a finding consistent with earlier regional surveys from Finland and Manitoba showing wheat among the highest and barley the lowest in Mn among temperate cereals. The consistently higher grain magnesium in wheat—averaging 0.27 grams per kilogram more—also matches established whole-grain nutritional data. For populations facing mineral deficiencies, the species choice in a cropping system could thus matter for reasons entirely separate from yield or market price.

The residue story, by contrast, is one of remarkable similarity with important exceptions. Wheat and barley produced statistically identical amounts of residue, and their residue nitrogen concentrations and carbon-to-nitrogen ratios were nearly equal across environments. Since residue C/N governs microbial decomposition rates, nitrogen mineralization, and soil carbon sequestration, the findings suggest residue management can be broadly similar between the species. But other residue nutrients diverged sharply. Wheat residue carried about 7.59 grams per kilogram more potassium than barley—potassium being by far the most concentrated nutrient in residue of either crop—while barley residue accumulated more phosphorus, calcium, and zinc, with barley concentrations rising disproportionately as environmental nutrient indices increased. Earlier work by the same team estimated that wheat and barley residues extract averages of 28.4 and 23.9 kilograms of potassium per hectare, respectively, for every megagram of grain yield, underscoring the severe economic cost if residue is harvested and the nutrients within it must be replaced by fertilizer.

Perhaps the most unexpected discovery was a positive covariance between residue productivity and residue nutrient concentrations. This is counter to the well-documented negative tradeoff between grain yield and grain nutrient concentration in small grains, which the study also confirmed in the grain fraction. In the residue, biomass production and nutrient density rose together in the same environments, meaning nutrient extraction through residue is even higher in productive systems than simple averaging would predict. The variance deconstruction analysis showed that residue nutrient concentrations and residue production each contributed substantially—often more than 100 percent combined before a large negative covariance term offset them—while for grain, yield was typically the dominant driver of nutrient uptake, contributing 64 to 84 percent of variance for most nutrients. The exception was grain manganese, for which concentration, not yield, was the largest contributor to uptake variance.

Principal component analysis of whole-crop data revealed further structure in how nutrients associate. In both species, phosphorus, magnesium, and zinc stoichiometries clustered together, as did nitrogen, potassium, and sulfur—patterns the researchers attribute to favorable nutrient balance, nitrogen sufficiency from fertilizer inputs, and the well-known coordination between N and K uptake. Manganese stood negatively related to nearly all other nutrients, especially in barley, hinting at a possible Mn-P antagonism. All carbon-to-nutrient ratios declined as environmental nutrient availability rose, though at different rates. Taken together, the study delivers a dual message: the barley-beats-wheat-in-bad-years perception holds only at the environmental margins and depends heavily on which classes are being compared, while the mineral density advantage of wheat grain—tempered by barley’s potassium edge and its residue phosphorus and zinc accumulation—gives growers, breeders, and nutritionists a far richer basis for deciding which cereal belongs in which field, and for whom.

Subject of Research: Agricultural Science

Article Title: Comparative class-wise wheat and barley system yields, and mineral nutrient concentrations, stoichiometry, and uptake across wide environmental ranges

Article References: Adams, C. B., Rogers, C. W., Marshall, J. M., & Schroeder, K. L. (2026). Comparative class-wise wheat and barley system yields, and mineral nutrient concentrations, stoichiometry, and uptake across wide environmental ranges. Journal of Agriculture and Food Research, Article 103376. https://doi.org/10.1016/j.jafr.2026.103376

Image Credits: AI Generated

DOI: 10.1016/j.jafr.2026.103376

Keywords: wheat, barley, crop yield, mineral nutrients, grain nutrition, stoichiometry, crop residue, agronomy, Comparative, class-wise, system, yields

The experiment was set up on a soil that was slightly alkaline, with a pH of 7.5, and notably poor in organic carbon, nitrogen, and available phosphorus. That starting point mattered. Nutrient-limited soils are precisely the kind of environment where a fertilizer input is most likely to reveal differences between plant genotypes, because each ecotype’s internal physiology must decide how to spend the newly available resources. The researchers hypothesized that organic manure and NPK fertilizer would not act identically across ecotypes, and instead would differentially regulate vegetative and reproductive traits depending on the plant’s origin.

To test this, the team obtained three ecotypes of summer savory, from Mashhad, Shiraz, and Karaj, and subjected them to five fertilization treatments: an unfertilized control, organic manure at 25 and 50 tonnes per hectare, and NPK fertilizer at 50 and 100 kilograms per hectare. The trial was laid out as a factorial randomized complete block design with three replicates, a standard but rigorous arrangement that allows the effects of ecotype, fertilizer, and their interaction to be separated statistically. The researchers then measured a broad panel of morphological, physiological, and phytochemical traits, from plant height and leaf biomass to chlorophyll and anthocyanin content, flower number, and the percentage of essential oil in the harvested material.

The analysis, which combined analysis of variance (ANOVA), principal component analysis (PCA), and structural equation modeling (SEM), delivered a clear verdict: both ecotype and fertilizer mattered, and crucially, so did their interaction. For reproductive traits and pigment-related traits in particular, the ecotype-by-fertilizer interaction was statistically significant, meaning that the best fertilizer for one ecotype was not necessarily the best for another. This kind of genotype-by-environment interplay is exactly what breeders and agronomists watch for when deciding whether a crop recommendation can be generalized or must be tailored to a specific seed lot or region.

Some of the standout combinations were striking. The Shiraz ecotype fertilized with the higher rate of organic manure, 50 tonnes per hectare, produced the greatest vegetative growth, the highest leaf biomass, and the highest chlorophyll a content of any treatment in the trial. In other words, when the goal is green, leafy, photosynthetically active biomass, the Shiraz population responded most enthusiastically to a heavy dose of organic matter. By contrast, the Karaj ecotype under the highest rate of NPK fertilizer, 100 kilograms per hectare, produced more flowers and accumulated more anthocyanins, the red-purple pigments that often signal a shift in a plant’s metabolic priorities. And across the board, the high-rate NPK treatment was associated with a higher percentage of essential oil in the plant tissue.

That split is the conceptual heart of the paper. The authors interpret these patterns as evidence of distinct resource allocation strategies between biomass-related traits and essential-oil-related traits, rather than a uniform response to fertilization. In plain terms, a plant cannot spend its nitrogen budget twice: resources channeled into building stems, leaves, and chlorophyll are not simultaneously available for synthesizing the terpenoid compounds that make up the essential oil, and vice versa. The fact that organic manure mainly favored biomass and pigment accumulation while high-rate NPK favored oil percentage and reproductive output suggests that the two fertilizer types are not interchangeable levers but rather push the plant along different metabolic trajectories.

The multivariate statistics sharpened this picture considerably. Principal component analysis grouped the measured traits into two coordinated modules, one linking growth and biomass characteristics and the other gathering metabolite-related traits, which is consistent with the idea that these two suites of characteristics are regulated at least partly independently. Structural equation modeling then went a step further, testing hypothesized causal pathways among the trait groups. The SEM results indicated positive associations between physiological traits, phytochemical traits, and essential-oil-related traits, suggesting that a plant that is photosynthesizing efficiently and accumulating pigments also tends to be one that produces more oil. But there was a twist: phenology, the timing of developmental stages such as flowering, was negatively associated with essential oil percentage, hinting that plants that rush into reproduction may do so at the expense of oil accumulation in the harvested tissue.

For growers of medicinal and aromatic plants, the practical implications are potentially significant. Summer savory is cultivated commercially for its essential oil, which is used in food flavoring, cosmetics, and traditional medicine, and the oil content of the harvest is often the single most economically important quality metric. If the choice between organic manure and mineral NPK fertilizer, and the rate at which either is applied, can shift the balance between raw biomass and oil concentration, then fertilizer recommendations for this crop may need to be written per ecotype rather than per species. A farmer growing the Shiraz ecotype for leafy green biomass would be advised to reach for well-rotted manure, while one targeting oil yield from the Karaj ecotype might do better with a higher rate of mineral NPK, at least under the soil and climate conditions of this trial.

The authors are careful to frame their conclusions within the limits of the study. The experiment was conducted at a single location over a single growing season, which means that the observed ecotype-by-fertilizer interactions, while statistically robust within the trial, have not yet been tested across the range of soils, climates, and years in which summer savory is actually grown. Essential oil composition in particular is notoriously sensitive to environmental conditions, harvest timing, and post-harvest handling, so the authors call for multi-location and multi-season studies to clarify the mechanisms underlying fertilizer-induced changes in oil chemistry. Until such studies are done, the findings should be read as a strong signal rather than a final prescription.

Even so, the study adds to a growing body of evidence that in medicinal and aromatic crops, the old agronomic habit of treating fertilization as a purely yield-oriented decision is due for an update. When the product of the crop is not just bulk but a specific chemical fingerprint, the fertilizer decision becomes a chemistry decision as well. For summer savory, at least, the new data suggest that the ecotype a grower plants and the nutrient regime applied to it together determine whether the harvest will be a pile of lush green leaves or a distillation flask full of aromatic oil, and that getting the pairing right may be worth as much as any single input itself.

Subject of Research: Effects of organic manure and NPK fertilizer on growth, biomass, pigments, and essential oil of Satureja hortensis ecotypes

Article Title: Multi-dimensional analysis of fertilizer-treated Satureja hortensis L. ecotypes: growth dynamics, biomass production, and phytochemical fingerprints

Article References: Ghanbari Moheb Seraj, R., Esmaeili, H., Tariverdizadeh, N., Hasanzadeh, A., & Mohebodini, M. (2026). Multi-dimensional analysis of fertilizer-treated Satureja hortensis L. ecotypes: growth dynamics, biomass production, and phytochemical fingerprints. BMC Plant Biology. https://doi.org/10.1186/s12870-026-10217-1

Image Credits: AI Generated

DOI: 10.1186/s12870-026-10217-1

Keywords: summer savory, Satureja hortensis, organic manure, NPK fertilizer, essential oil, chlorophyll, anthocyanins, phenology, ecotypes, medicinal plants, plant physiology, agronomy

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 disease in question, powdery mildew, is caused by the fungus Erysiphe cichoracearum. It announces itself as minute discolored specks on sunflower leaves, from which a powdery white mass radiates outward until entire leaves are coated. Under the warm conditions typical of tropical regions, the infection can escalate to complete defoliation, and the resulting loss of photosynthetic surface translates directly into fewer and lighter seeds. While sunflower is among the world’s most cultivated oil crops, with global production around 40.21 million tons annually across roughly 26.05 million hectares, Ethiopia’s national average yield sits at only about 1.033 tons per hectare, dragged down in part by this and other fungal diseases.

The research team set out to test whether pairing host resistance with chemical control could suppress powdery mildew epidemics more effectively than either strategy alone. They evaluated two sunflower varieties, the improved Russian black and a local variety, in combination with three fungicides: propiconazole 25% EC, a systemic triazole; wettable sulphur 80%, a contact fungicide; and mancozeb 80%, another contact product. Each fungicide was applied either once or twice per season, producing fourteen treatment combinations that were arranged in a randomized complete block design with factorial structure and three replications under field conditions.

The choice of fungicides reflects three distinct modes of action. Propiconazole, as a triazole, moves systemically through the xylem of the plant and blocks the biosynthesis of ergosterol, an essential component of fungal cell membranes; without this sterol, the fungal cell wall is irreversibly damaged and the pathogen dies. Wettable sulphur interferes with cellular respiration in fungal pathogens, while mancozeb reacts with sulfhydryl groups in fungal proteins. Because these chemicals attack the fungus in different ways, the researchers reasoned that their performance under field conditions might differ substantially, and the data confirmed that prediction.

Disease assessments were rigorous. Ten plants per plot were randomly selected from central rows and marked for repeated observation, with incidence calculated as the percentage of plants showing symptoms and severity rated on a 1-to-9 scale that was converted to a percent severity index. From these repeated measurements, the team computed the area under the disease progress curve, or AUDPC, a single value that integrates initial disease intensity, the rate of epidemic development, and the duration of the epidemic, providing a compact signature of how aggressively the pathogen spread in each plot.

The results were striking. At Debub Ari, untreated local variety plots suffered a mean disease severity of 63.7 percent, while plots planted with Russian black and sprayed twice with propiconazole showed only 3.7 percent, a 94.2 percent relative reduction. At Salamago, the corresponding values were 39.25 percent for untreated local plots and 5.92 percent for the best treatment. Plots of the improved variety treated twice with wettable sulphur and plots receiving a single propiconazole spray also performed strongly, reducing severity by 86.3 percent and 81.6 percent respectively compared with the untreated local check. The AUDPC analysis mirrored these patterns: untreated local plots accumulated 778.74 percent-days of disease at Debub Ari and 489.93 percent-days at Salamago, whereas Russian black with two propiconazole applications kept the epidemic to just 102.84 percent-days at the first site, with wettable sulphur treatments close behind at the second.

Correlation analysis revealed how tightly the disease was linked to productivity. At Salamago, percent severity and AUDPC were almost perfectly correlated with each other, and grain yield showed strongly negative associations with both disease measures. Head diameter and thousand seed weight also declined as disease pressure rose. Similar negative relationships between disease parameters and yield emerged at Debub Ari, confirming that every increment of mildew severity translated into measurable losses in the components that determine final harvest weight.

Yield data told the same story from the opposite direction. The highest yields came from Russian black plots treated with propiconazole twice, reaching 2869.8 kilograms per hectare at Salamago, followed by 2841.3 kilograms per hectare with a single sulphur spray and 2834.9 kilograms per hectare with two sulphur applications. Closely similar trends appeared at Debub Ari. Untreated local variety plots, by contrast, yielded just 1289.5 kilograms per hectare at Debub Ari and 1574.6 kilograms per hectare at Salamago. Plant height and head diameter also responded to treatment, with the tallest plants and largest heads generally found in the improved variety under effective fungicide programs.

The economics proved equally persuasive. In a partial budget analysis that accounted for fungicide costs, local market prices, and a 10 percent downward adjustment of experimental yields to reflect realistic farm conditions, the highest net benefits came from Russian black sprayed once with propiconazole, followed by the two-spray propiconazole program and the sulphur treatments on the same variety. Untreated local plots delivered the lowest returns by a wide margin. Gap analysis summarized the practical advantage: treated plots reduced mean disease severity by roughly 91 percent compared with usual agro-pastoral practice and offered a 45.58 percent yield advantage over unsprayed controls.

The authors conclude that integrating the improved Russian black variety with propiconazole 25% EC at 0.1 percent concentration, followed by wettable sulphur 80% where appropriate, offers farming communities in South Omo a reliable and affordable strategy against powdery mildew. Because propiconazole is already available in local markets and provides both curative and protective action, adoption barriers are low. For a crop that anchors edible oil production across the region’s drought-prone landscapes, the study suggests that a modest investment in variety choice and spray timing can transform powdery mildew from a chronic yield thief into a manageable problem.

Subject of Research: Integrated management of sunflower powdery mildew using resistant varieties and fungicides in southern Ethiopia

Article Title: Management of sunflower powdery mildew through varieties and fungicides application in Debub Ari and Salamago districts of South Omo, Southern Ethiopia

Article References: Adila, W., Jerjero, T., & Bekele, F. (2025). Management of sunflower powdery mildew through varieties and fungicides application in Debub Ari and Salamago districts of South Omo, Southern Ethiopia. Innovations in Agriculture, 8, 1-7. https://doi.org/10.3897/ia.2025.143321

Image Credits: AI Generated

DOI: 10.3897/ia.2025.143321

Keywords: sunflower, powdery mildew, Erysiphe cichoracearum, fungicides, propiconazole, wettable sulphur, Ethiopia, South Omo, plant pathology, crop yield, integrated disease management, agronomy

The study, published in Innovations in Agriculture, addresses a persistent problem for Bangladeshi growers. Lettuce is rapidly gaining popularity in the country as a nutritious salad vegetable, yet average yields hover around just 4.0 tonnes per hectare, a strikingly low figure compared with Italy at 73.5 tonnes, Japan at 58.6 tonnes, and India at 108.9 tonnes per hectare. The authors attribute this gap largely to inadequate knowledge of modern production practices, including the selection of appropriate cultivars and the optimization of nutrient management, rather than to any fundamental climatic limitation. The crop is well suited to Bangladesh’s Rabi season, the cool winter window when many vegetables thrive.

To untangle how genetics and fertilization interact, the team grew three cultivars, Green Wave, Grand Rapids, and Iceberg, under two systems: an organic regime using only well-decomposed cow dung at 50 tonnes per hectare, and a conventional regime combining that same manure at 15 tonnes per hectare with urea, triple superphosphate, and muriate of potash at 250, 150, and 200 kilograms per hectare respectively. The urea was applied as a top dressing in two equal installments at 12 and 24 days after transplanting. The six treatment combinations were arranged in a randomized complete block design with three replications across eighteen unit plots, each 2 meters by 1.5 meters, with plants spaced 40 centimeters apart within rows set 25 centimeters apart.

Growth measurements were taken at 15, 25, and 35 days after transplanting, capturing plant height, leaf length and breadth, leaf number, and total leaf area per plant. The results revealed significant interaction effects between cultivar and fertilization method for most parameters, meaning that no single fertilization strategy was universally superior. Green Wave grown organically produced the tallest plants throughout the season, reaching 22.52 centimeters at final measurement, and also posted the largest leaf area early on, with 635.68 square centimeters per plant at 25 days after transplanting. This suggests that the slow mineralization of nitrogen in cow dung provided a steady nutrient supply that favored sustained vegetative expansion in that cultivar.

Leaf development told a similarly nuanced story. Green Wave under organic management produced the longest leaves at 25 days after transplanting, measuring 16.71 centimeters, significantly longer than every other treatment. By contrast, Grand Rapids was generally the weakest performer, recording the shortest plants and smallest leaves in several comparisons, and it ultimately yielded among the lowest of the six combinations. Iceberg, meanwhile, started relatively small, with the lowest leaf area at 15 days under conventional fertilization at just 148.00 square centimeters per plant, but accelerated late in the season to reach 722.25 square centimeters, nearly matching the 742.25 square centimeters of organically fertilized Green Wave at final measurement.

When it came to harvest, the yield hierarchy became clear. Iceberg under conventional fertilization produced the highest marketable yield at 31.0 tonnes per hectare, with 0.31 kilograms of leaf per plant and 9.30 kilograms per plot. Iceberg grown organically followed at 27.0 tonnes per hectare, a figure matched by conventional Green Wave and conventional Grand Rapids. The lowest yields came from organic Green Wave at 22.0 tonnes per hectare and organic Grand Rapids at 24.0 tonnes per hectare. The authors note that crisphead types like Iceberg respond strongly to readily available nitrogen, which helps explain their exceptional performance when mineral fertilizers deliver nutrients in immediately plant-accessible forms.

The economic analysis, however, flipped the script. Because organically produced lettuce commands a premium price of 5,583 Bangladeshi taka per tonne compared with 4,624 taka for conventionally grown produce, and because the organic system avoided the cost of synthetic fertilizers, organic Iceberg achieved the highest benefit-cost ratio of 1.84. That combination generated a gross return of 150,741 taka per hectare against production costs of 82,000 taka, leaving a net return of 68,741 taka. Conventional Iceberg came second with a benefit-cost ratio of 1.69, despite its higher yield, because greater input costs and lower prices eroded its advantage. The worst performers economically were the conventional treatments of Green Wave and Grand Rapids, each at 1.47.

These findings align with a broader body of research showing that organic systems can remain profitable even with somewhat lower yields. Earlier studies cited by the authors found that conventional mineral fertilizers rapidly boost soil nitrogen and phosphorus availability, driving higher chlorophyll content, canopy development, and fresh biomass, while compost-based amendments act more slowly but improve soil organic matter and can, when well managed, match or even surpass conventional yields over time. The slow, steady release of nutrients from decomposed manure may also support beneficial microbial activity and improved soil structure, advantages that accrue beyond a single growing season and are not captured by yield figures alone.

The practical message for farmers is that cultivar choice matters as much as fertilization strategy. Iceberg proved the most productive cultivar under both systems, while Green Wave excelled in vegetative traits under organic management. For growers weighing the trade-off between maximum output and maximum return, the study indicates that organic cultivation of Iceberg lettuce offers a sustainable and economically viable pathway, combining lower external input costs, premium consumer prices, and environmentally friendlier practices. As demand for organic vegetables grows in Bangladesh and beyond, matching the right genotype to the right nutrient system could transform lettuce from a marginal crop into a dependable source of farm income.

The research also underscores a caution about overreliance on chemical fertilizers. Excessive nitrogen use can disrupt soil nutrient balance, pollute the soil, and reduce crop quality, threatening long-term productivity. By demonstrating that a fully organic regime can deliver the best financial outcome for a high-value leafy vegetable, the Gazipur team adds to the growing evidence that sustainable intensification is not merely an ideal but a practical, profitable option for smallholder farmers in South Asia.

Subject of Research: Effects of organic versus conventional fertilization on the growth, yield, and profitability of lettuce cultivars in Bangladesh

Article Title: Organic versus conventional fertilization: impacts on growth, yield and profitability of lettuce cultivars

Article References: Akter, N., Nabi, M. N., Chowhan, S., Kayesh, E., Alam, M. S., & Roni, M. S. (2026). Organic versus conventional fertilization: impacts on growth, yield and profitability of lettuce cultivars. Innovations in Agriculture, 9, Article e183987. https://doi.org/10.3897/ia.2026.183987

Image Credits: AI Generated

DOI: 10.3897/ia.2026.183987

Keywords: lettuce, organic farming, conventional fertilization, Bangladesh, crop yield, benefit-cost ratio, cultivar selection, soil health, sustainable agriculture, Lactuca sativa, Iceberg lettuce, agronomy

The research team set up a factorial experiment that crossed three teff varieties with three seed rates and three sowing methods. The varieties included Quncho and Kora, both improved releases, alongside Abolse, a local landrace that farmers in the area have long relied upon. Seed rates of 10, 15, and 20 kilograms per hectare reflected the spectrum of practice commonly seen in the highlands, where farmers often oversow in the belief that more seed means more plants and, ultimately, more grain. The sowing methods compared traditional broadcasting, in which seed is scattered by hand across the surface, with row planting at 15-centimeter and 20-centimeter inter-row spacings, a technique increasingly promoted by extension services because it makes weeding and management easier.

The statistical outcome was unambiguous: the three-way interaction of variety, seed rate, and sowing method significantly affected crop growth and yield parameters at a probability level below 0.01. In practical terms, no single recommendation could be extracted from the data and applied universally. A variety that performed well under one combination of spacing and seed rate did not necessarily hold that advantage under another. This kind of interactive response is exactly what agronomists expect when a crop’s architecture, its competition for light and nutrients, and its susceptibility to lodging are all shaped simultaneously by how densely it stands and how its seeds are distributed in the field.

Some of the most striking results concerned plant height and biomass. The tallest plants, measuring 128.33 centimeters, came from the variety Quncho when it was sown in rows 20 centimeters apart at the lowest seed rate of 10 kilograms per hectare. Wider spacing at a lower density evidently gave individual plants room to elongate, reducing aboveground competition during the vegetative phase. Aboveground biomass, the total weight of stems, leaves, and panicles produced per unit area, peaked at 10.76 tonnes per hectare with a different combination: Quncho sown at 10 kilograms per hectare in rows spaced 15 centimeters apart. The contrast between the two combinations shows how different objectives, such as maximizing forage versus maximizing eventual grain, may call for different field layouts even with the same variety and seed rate.

Lodging, the tendency of cereal stems to bend or collapse under their own weight, emerged as a critical limiting factor and one of the study’s most consequential findings. The highest lodging index, 44 percent, was recorded when Quncho was broadcast at the highest seed rate of 20 kilograms per hectare. Dense, uniformly scattered stands produce crowded plants that stretch for light, grow taller and thinner-stemmed, and then fall over, often in response to wind or rain. Lodged teff is notoriously difficult to harvest, suffers grain losses, and can spoil in contact with wet soil. The data suggest that the very practices many farmers use to chase higher yields, thick broadcasting of an improved, high-potential variety, can instead push the crop toward structural failure.

Beyond the interaction effects, each factor on its own left a measurable imprint on the crop’s development. Seed rate, sowing method, and variety all significantly influenced the number of days to 50 percent heading, the point at which half the plants in a plot have produced their flowering heads, with a significance threshold below 0.05. Variety Kora was the slowest, taking 59.18 days to reach 50 percent heading, a meaningfully longer duration than either Quncho or the local Abolse. In highland environments where the growing season is bounded by unreliable rains, maturity time is not a trivial trait; later heading can expose the crop to end-of-season moisture stress and shorten the window for grain filling.

Sowing method and seed rate also shifted the crop’s timetable. Widening the row spacing to 20 centimeters prolonged the time to 50 percent heading across the varieties tested, while sowing at 15 kilograms per hectare likewise delayed heading compared with other rates. These phenological shifts matter for farmers synchronizing planting with rainfall patterns, and they illustrate that agronomic choices ripple through the entire life cycle of the crop, not just its final yield. A decision made at sowing about how far apart to place rows can determine whether the crop flowers before the rains taper off or weeks after.

When it came to the harvest itself, the variety effect dominated. Quncho produced the longest panicle, at 45.14 centimeters, and the highest grain yield, at 2.47 tonnes per hectare. That figure represents a 15.96 percent advantage over Kora and a 26.67 percent advantage over the local Abolse variety, margins that at field scale translate into substantial additional food and income for farming households. Sowing method made its own contribution: teff planted in rows 20 centimeters apart achieved the maximum panicle length of 44.46 centimeters and a grain yield of 2.32 tonnes per hectare, outperforming both the narrower 15-centimeter spacing and broadcasting. Among seed rates, the lowest tested rate, 10 kilograms per hectare, produced the highest grain yield and the largest panicles, a result that challenges the intuition that spending more on seed pays for itself.

The economic and practical implications of that last finding are considerable. If 10 kilograms of seed per hectare outyields 15 and 20 kilograms, farmers can simultaneously cut their seed costs and raise their harvests, provided they also adopt row planting and an appropriate improved variety. Lower seed rates also mean thinner stands, which in this study correlated with reduced lodging risk, though the lodging index peaked specifically under the broadcast, high-density treatment. For a crop that anchors the diet and the farming identity of millions of Ethiopians, a package combining variety selection, reduced seed rate, and row planting offers a rare triple win: more grain, less seed expense, and more resilient stands that stay upright through the storms of the highland rainy season.

The study’s authors frame the results as evidence that teff productivity in the central highlands depends on managing the cropping system as an integrated whole rather than adjusting inputs one at a time. The three-way interactions they documented, significant at the one percent level, mean that extension recommendations calibrated for one variety or one district may not transfer cleanly to another. As Ethiopia works to lift teff yields toward the crop’s genetic potential across its three million hectares, findings like these point toward a more precise agronomy: matching the right variety to the right seed rate and sowing method for local conditions, and in Mojana Wodera at least, that match appears to favor Quncho, low seed rates, and wide rows over the thick broadcast sowings of tradition.

Subject of Research: Interactive effects of seed rate, sowing method, and variety on teff yield and lodging in the central highlands of Ethiopia

Article Title: Interactive effects of seed rate, sowing method, and variety on Teff ( Eragrostis tef ) yield, yield components, and lodging trait performance in the central highlands of Ethiopia

Article References: Tesfa, Y., Belete, F., & Tena, W. (2026). Interactive effects of seed rate, sowing method, and variety on Teff (Eragrostis tef) yield, yield components, and lodging trait performance in the central highlands of Ethiopia. PLOS One, 21(10), e0360297. https://doi.org/10.1371/journal.pone.0360297

Image Credits: AI Generated

DOI: 10.1371/journal.pone.0360297

Keywords: teff, Eragrostis tef, Ethiopia, seed rate, sowing method, row planting, broadcasting, lodging, grain yield, varietal performance, highlands, agronomy

The research team conducted field experiments during 2020 and 2021, testing leaching quotas ranging from zero to 300 millimeters combined with schedules that varied from a single seedling-stage leaching to multiple leaching events distributed across the growing season. They tracked soil water content and salt-ion concentrations, measured root traits in different soil layers, followed nitrogen accumulation and uptake dynamics through the season, and recorded lint yield, yield components, and fiber quality. To untangle the web of interacting factors, the researchers deployed random forest modeling and exploratory partial least squares structural equation modeling, statistical tools that allowed them to identify which variables actually drove yield and quality rather than merely correlating with them.

The standout treatment, labeled W3S3, combined a moderate leaching quota with a multi-stage schedule, and its results were dramatic. Compared with the non-leached control, W3S3 increased lint yield by 112.3 percent in 2020 and 91.7 percent in 2021. Boll number, the single most important yield component in cotton, rose by 120.1 percent in the first year and 60.3 percent in the second. In an era when agricultural gains of even a few percent are celebrated, a near-doubling of lint yield through irrigation scheduling alone is the kind of result that demands attention from agronomists and water managers worldwide.

What makes the finding scientifically compelling is that the mechanism was not simply salt removal. The root responses were concentrated in the top 42 centimeters of the soil profile, where W3S3 increased surface root length density and root volume density by more than 40 percent and promoted lateral root expansion. This shallow but dense root architecture is precisely what a cotton plant needs under drip irrigation, because drip emitters wet a limited soil sphere near the surface, and roots proliferating in that moist, low-salt zone gain first access to water and dissolved nutrients. The leaching events effectively reshaped the underground environment, transforming a hostile, saline root zone into a hospitable one where fine roots could branch and forage.

Nitrogen dynamics revealed a second layer of the story. W3S3 improved soil moisture, reduced salt-ion pressure on the plants, enhanced both root and shoot nitrogen accumulation, and prolonged the effective period of nitrogen uptake by about 20 days. In cotton, late-season nitrogen uptake is critical for boll filling and fiber development, and salt stress typically shortens this window by accelerating senescence. By keeping the root zone moist and dilute in salts, the multi-stage leaching schedule allowed plants to keep absorbing nitrogen deep into the boll-setting period, effectively extending the crop’s productive lifespan. The extra three weeks of active uptake translated directly into more and heavier bolls.

The study also illuminated the tradeoffs between yield and fiber quality, a tension that has long frustrated cotton breeders and agronomists. Yield was driven primarily by boll number, nitrogen uptake, soil moisture, and sodium concentration in the root zone, while fiber quality was associated with a different set of variables: sodium levels, the root nitrogen-uptake rate, root length, soil moisture, and root surface area. Notably, the W3 leaching quota increased the spinning consistency index and fiber length, meaning that the moderate quota improved quality even where the multi-stage schedule drove the yield gains. This separation of drivers suggests that farmers and researchers can, to a meaningful degree, tune leaching schedules for yield and leaching quotas for quality rather than accepting an unavoidable compromise between the two.

The statistical modeling added mechanistic rigor to these observations. Random forest analysis ranked the importance of candidate drivers, while the structural equation modeling traced pathways linking water and ion redistribution in the root zone to root architecture, nitrogen uptake, and ultimately yield and quality outcomes. Together, the two approaches supported a coherent causal chain: leaching events redistribute water and push salts away from active rooting zones, the improved hydrochemical environment stimulates root growth and lateral branching, the expanded root system sustains nitrogen uptake longer into the season, and the combination of more bolls and better-fed fibers produces the yield and quality gains observed in the field.

The broader context makes these results especially timely. Salt-affected soils constrain agriculture across arid regions on every continent, and water scarcity is intensifying across northwestern China, making it impossible to simply flush soils with generous irrigation as past generations of farmers did. Researchers have documented the tension between water savings and salinization prevention in dryland irrigation, and long-term mulched drip irrigation has been flagged as a potential risk to agricultural sustainability precisely because restricted water application promotes salt accumulation. The Xinjiang study demonstrates that the solution lies not in choosing between water conservation and salt control, but in engineering the temporal distribution of limited leaching water so that each event does maximum work.

There are, of course, caveats and open questions. The experiments spanned two years at sites in northern Xinjiang, and leaching requirements will vary with soil texture, initial salinity, water table depth, and the salinity of the irrigation water itself. Applying 300 millimeters of extra water, even in moderate staged doses, carries a real cost in water-scarce basins, and excessive leaching can mobilize nitrogen below the root zone or raise shallow water tables. The authors’ finding that moderate quotas outperformed extremes suggests a sweet spot, but translating W3S3 into practical recommendations will require local calibration and, ideally, longer-term trials to confirm that the approach prevents salt accumulation over multiple seasons rather than merely redistributing it within a single year.

Even with those qualifications, the study reframes how scientists think about managing saline-alkali agroecosystems. The conclusion that properly timed moderate drip leaching improved cotton performance by sustaining active roots and late-season nitrogen uptake, rather than through salt removal alone, elevates root ecology from a passive consequence of irrigation to an active management target. For the millions of hectares of salt-affected farmland under drip irrigation worldwide, the message is that the invisible architecture of roots, shaped season by season by the rhythm of water moving through soil, may be the most powerful lever available for reconciling high yields, good fiber quality, and the harsh arithmetic of water scarcity.

Subject of Research: Effects of multi-stage drip leaching on root architecture, nitrogen uptake, and yield-quality tradeoffs in saline-alkali cotton fields

Article Title: Root-zone water-ion redistribution links multi-stage drip leaching to root architecture, nitrogen uptake, and yield-quality tradeoffs in saline-alkali cotton agroecosystems

Article References: Xiao, C., Chen, G., Zhang, F., Fan, J., & Zhang, Q. (2026). Root-zone water-ion redistribution links multi-stage drip leaching to root architecture, nitrogen uptake, and yield-quality tradeoffs in saline-alkali cotton agroecosystems. Plant and Soil. https://doi.org/10.1007/s11104-026-09013-5

Image Credits: AI Generated

DOI: 10.1007/s11104-026-09013-5

Keywords: cotton, drip irrigation, soil salinity, leaching, root architecture, nitrogen uptake, lint yield, fiber quality, saline-alkali soil, Xinjiang, water management, agronomy

The work, carried out by Rohini Kokare and Sunil Mane, focuses on a niche but important use case: soybean cultivation advisory for Indian agriculture. The researchers scraped unstructured documents from the ICAR-National Soybean Research Institute in Indore, Tamil Nadu Agricultural University and the TNAU Agritech portal, covering crop protection, crop production, farm mechanization, good agricultural practices, variety information and region-wise crop recommendations. From these documents they built a curated instruction dataset of 1,200 samples, each containing exactly one relational triplet, split 90:10 between training and testing with no sentence overlap.

At the heart of the system is a fine-tuned Llama-3.2-1B-Instruct model, a one-billion-parameter language model with multilingual support for eight languages. Rather than retraining the entire network, the team used Low-Rank Adaptation, or LoRA, a parameter-efficient fine-tuning technique that freezes the pre-trained weights and trains only small adapter matrices. For a frozen weight matrix of size 512 by 512, LoRA decomposes the update into two matrices of 512 by 8 and 8 by 512, yielding just 8,192 trainable parameters per adapted matrix. This dramatic reduction in trainable parameters makes the approach computationally affordable while still allowing the model to learn domain-specific behavior.

The researchers chose their target modules carefully. In Llama-style architectures, the feed-forward sub-block contains most of the parameters, so adapting only the attention mechanism would leave much of the model’s representational capacity untouched. The team therefore applied LoRA across all linear modules: the attention projections for query, key, value and output, plus the gated MLP SwiGLU blocks known as gate_proj, up_proj and down_proj. They set the LoRA rank to 8, with an alpha of 16 giving a scaling factor of two, and applied a light dropout of 0.05 on the adapter path to avoid overfitting on the specialized vocabulary of agronomy.

Instruction fine-tuning was central to the approach. Instead of simply exposing the model to domain text, the researchers trained it with explicit instructions such as “Extract entities and relationships as triples,” paired with input sentences and structured outputs. An example from the dataset maps the sentence “Macs 330 controls Bradyrhizobium japonicum” to the triplet consisting of the subject MACS 330, the relation “controls” and the object Bradyrhizobium japonicum. This format teaches the model to follow task-specific guidelines and produce consistent, machine-parseable output rather than free-form text.

Once fine-tuned, the model extracts triplets from unstructured text at inference time, and these subject-predicate-object structures are pushed into Neo4j, a graph database where entities become nodes and relations become edges. The resulting knowledge graph captures multi-relational information about soybean varieties, diseases, pests, treatments and practices in a form that supports structured queries written in the Cypher query language. This graph then serves as the retrieval backbone for a GraphRAG system, a hybrid of graph retrieval and large language model generation.

The GraphRAG retrieval layer combines vector and keyword search over the text properties of document nodes, using embeddings from HuggingFace models matched against embedding properties stored on each node. A user’s natural language query flows directly into the hybrid retriever through a LangChain pipeline without rewriting or entity extraction. Both the retrieved graph context and the original question are then passed to a fixed prompt template, and the fine-tuned Llama model generates the final answer. This design grounds responses in retrieved facts, reducing the hallucination risk that plagues out-of-the-box language models when they answer specialized questions.

The experimental results showed clear gains from the approach. When the team compared fine-tuning configurations, the model trained with all projection modules and instruction fine-tuning outperformed variants trained only on attention layers or only on feed-forward layers, measured by macro-averaged precision, recall and F1-score across the test set. On end-to-end response generation, the fine-tuned GraphRAG pipeline beat conventional vector database retrieval, standard RAG and knowledge-graph-only retrieval on both ROUGE and BLEU precision metrics, which measure how closely generated answers overlap with reference responses at the unigram, bigram and longest-common-subsequence levels.

The authors argue that the framework addresses a genuine gap. Previous research has explored agricultural chatbots and question-answering systems, agricultural knowledge graphs and general-purpose GraphRAG frameworks separately, but no existing system, to their knowledge, had combined LoRA-based generative triple extraction, automated knowledge graph construction and GraphRAG-style retrieval for a low-resource domain like Indian soybean advisory. Conventional pipelines that treat named entity recognition and relation extraction as separate steps are labor-intensive to train and prone to cascading errors, while generalized prompting or full-parameter fine-tuning is computationally inefficient and time-consuming.

The work is not finished. The researchers note that systematic verification of triplet semantics and quality, including comparison of gold versus predicted triples and elimination of noisy extractions, is scheduled as upcoming work. They also plan explicit graph-traversal-based retrieval using relation-path expansion from matched entities, and controlled comparisons against size-matched open-weight models such as Qwen2.5-1.5B-Instruct and DeepSeek small variants, as well as larger frontier models like GPT-4o and Mistral, under identical test questions and decoding settings. Even so, the study demonstrates that a small, efficiently fine-tuned language model can transform scattered agricultural documents into a queryable knowledge network, offering a template for bringing structured, trustworthy AI question-answering to domains where data is abundant but structured knowledge is scarce.

Subject of Research: LoRA-based instruction fine-tuning of large language models for agronomic triple extraction and GraphRAG knowledge retrieval

Article Title: LoRA based instruction fine tuning of large language models for agronomic triple extraction and GraphRAG knowledge retrieval

Article References: Kokare, R., & Mane, S. (2026). LoRA based instruction fine tuning of large language models for agronomic triple extraction and GraphRAG knowledge retrieval. Discover Artificial Intelligence, 6(1), Article 1381. https://doi.org/10.1007/s44163-026-02410-w

Image Credits: AI Generated

DOI: 10.1007/s44163-026-02410-w

Keywords: LoRA, instruction fine-tuning, large language models, Llama, triple extraction, GraphRAG, knowledge graph, Neo4j, agronomy, soybean, retrieval-augmented generation, parameter-efficient fine-tuning

The team’s central idea was to change the chemistry of the fertilizer microsite, the tiny zone of soil immediately surrounding each pellet where urea hydrolyzes within hours, pushing local pH above 8.5 and triggering ammonia losses. By coating or blending urea with biochars carrying a wide range of surface charges, the researchers aimed to slow nitrogen release and improve retention right where it matters. Biochar, a carbon-rich material produced by heating biomass in low-oxygen conditions, has long been studied as a soil amendment, but its properties vary enormously with feedstock, pyrolysis conditions, and post-production treatment. Rather than treating biochar as a single material, the team engineered four distinct versions and tested them head to head.

The biochars were produced from sawdust in a semi-airtight brick kiln, with temperatures climbing from 300 degrees Celsius at two hours to a peak of 807 degrees at 28 hours before declining. One batch was left pristine, with a nearly neutral surface charge of about minus 2.6 centimoles of charge per kilogram at pH 7.0. A second was chemically oxidized with hydrogen peroxide, which flooded the surface with carboxylic and phenolic functional groups and pushed the negative charge to minus 34.3 centimoles per kilogram. A third was biologically activated by inoculating it with microorganisms extracted from forest, agricultural, and pasture soils and incubating it for 15 days at alternating temperatures of 40 and 25 degrees. The fourth was pre-doped with iron, aluminum, and magnesium chlorides before pyrolysis, which impregnated the carbon matrix with metals and flipped the surface charge to slightly positive, around 1 centimole per kilogram, with a correspondingly tiny cation exchange capacity of about 2.4.

These four biochars were applied in two different architectures. In the coated formulations, urea granules were tumbled at 30 revolutions per minute with biochar, starch, and soybean oil at a urea-to-biochar ratio of 10:1, creating an external diffusion barrier around each pellet. In the blended formulations, biochar, fertilizer, and clay were physically mixed and pressed into new composite granules, so that nitrogen release depended on internal pore connectivity rather than an outer shell. Potassium nitrate was also blended with pristine and metal-doped biochars, and urea was blended with coal as an additional comparison. Because blending dilutes the nutrient content—the coated fertilizers contained roughly 40 percent nitrogen while the blends held only about 2.5 percent—application rates were adjusted so that every treatment delivered the same nutrient supply.

The field trial took place at the Agronomy Field of Patuakhali Science and Technology University in Dumki, Bangladesh, from January to May 2023, on a slightly alkaline silty clay loam with very low organic carbon of 0.98 percent and total nitrogen of just 0.05 percent. Twelve treatments were arranged in a randomized complete block design with three replications, spanning the biochar formulations, neem-coated urea, guti-urea (urea super granules), conventional prilled urea, and an unfertilized control. The hybrid maize variety Bahubali-881 was sown at 25 centimeters within rows spaced 60 centimeters apart, and nitrogen was split three ways: at land preparation, 40 days after sowing, and 75 days after sowing, just before flowering.

The standout result came from urea coated with biologically activated biochar. This treatment produced the highest grain yield of the entire experiment, 9.84 tonnes per hectare, a 174 percent increase over the unfertilized control’s 3.58 tonnes and a 29 percent increase over conventional urea’s 7.63 tonnes. The same plots produced the tallest plants at 202.73 centimeters, the longest cobs at 19.12 centimeters, the most grains per cob at roughly 590, the heaviest 1000-grain weight at 332.7 grams, and the highest SPAD chlorophyll reading at 54.7. Root biomass reached 3.81 tonnes per hectare, and straw yield peaked at 7.78 tonnes per hectare. Most strikingly, nitrogen use efficiency hit 74.4 percent, compared with just 47.6 percent for conventional urea—meaning nearly three-quarters of the applied nitrogen ended up in the crop rather than the environment.

The efficiency metrics told a consistent story across the board. The researchers calculated not only overall nitrogen use efficiency but also agronomic efficiency, the yield gain per unit of nitrogen applied; physiological efficiency, how well plant nitrogen converts into grain; and apparent recovery efficiency, the fraction of applied nitrogen recovered in plant tissue. The biologically activated biochar-coated urea led every index. Chemically oxidized biochar coatings and iron-aluminum-magnesium pre-doped coatings also matched or beat conventional urea and the commercial enhanced-efficiency products, neem-coated urea and guti-urea. When results were pooled by formulation type, coated fertilizers averaged 9.43 tonnes per hectare of grain and 65.32 percent nitrogen use efficiency, outperforming blended formulations, and oxidized biochar types averaged 9.78 tonnes per hectare and 69.3 percent efficiency.

Intriguingly, the soil itself offered few clues to the yield differences. Measurements of pH, ammonium nitrogen, and nitrate nitrogen during the growing season showed no statistically significant differences among treatments, and post-harvest analysis found most nutrients unchanged, with the notable exception of potassium, which was highest where potassium nitrate had been blended with biochar. Soil organic matter did vary, rising in several biochar-blended treatments. The authors attribute the modest soil effects to the small quantities of biochar applied, suggesting that the fertilizers’ benefits occurred locally around each pellet rather than across the bulk soil. Grain nutrient composition likewise remained stable, with no significant differences in micronutrients such as iron, zinc, and manganese, indicating that the yield gains came without sacrificing nutritional quality.

Multivariate statistics sharpened this interpretation. Pearson correlation analysis showed grain yield strongly and positively linked to plant traits—straw yield, grains per cob, root biomass, cob length, plant height, and 1000-grain weight—with correlation coefficients between 0.78 and 0.96, while residual soil nitrate after harvest correlated only weakly at 0.38. Redundancy analysis separated yield-related plant traits from soil nutrient variables along the first ordination axis, confirming that variation in harvest-time soil chemistry did not explain treatment performance. The authors are careful to note that mechanisms such as slower nitrogen release and improved synchronization between nitrogen supply and crop demand remain plausible explanations drawn from prior literature rather than processes directly measured in this study; urease activity and nitrogen transformation pathways were not evaluated.

Even with those caveats, the practical implications are substantial. Biologically activated biochar-coated urea is built from sawdust waste, soil microbes, starch, and soybean oil—no synthetic polymer coatings that linger in soil—making it an attractive candidate for smallholder maize farmers in Bangladesh and beyond, where yields still fall short of the regional irrigated potential of roughly 12.3 to 13.8 tonnes per hectare. The authors call for direct quantification of nitrogen release dynamics and loss pathways, along with validation across multiple seasons, locations, and economic conditions before large-scale adoption. If those follow-up studies hold up, a coating of microbe-conditioned charcoal could become one of the cheapest levers available for feeding more people while wasting less of the nitrogen that modern farming depends on.

Subject of Research: Biochar-coated and blended nitrogen fertilizers for improving maize yield and nitrogen use efficiency

Article Title: Role of biochar coated and blended nitrogen fertilizers on maize yield and nitrogen use efficiency

Article References: Alam, S. M., Mia, S., Saha, G., Masud, M. M., Al-Amin, M., Mim, M. J., Islam, M. M., & Jindo, K. (2026). Role of biochar coated and blended nitrogen fertilizers on maize yield and nitrogen use efficiency. Discover Plants, 3(1), Article 372. https://doi.org/10.1007/s44372-026-00835-2

Image Credits: AI Generated

DOI: 10.1007/s44372-026-00835-2

Keywords: biochar, urea, nitrogen use efficiency, maize, slow-release fertilizer, soil chemistry, Bangladesh, agronomy, sustainable agriculture, pyrolysis, enhanced-efficiency fertilizer, field trial