The study covered Huanglongbing, citrus red mite, aphid and sooty mould, with occurrence severity recorded on a five-level scale from zero to four. Rather than predicting only whether a category was present, ST-CitrusFormer was trained to forecast both the severity class and a continuous occurrence-intensity value for every plot and day in a seven-day horizon. This dual-task design reflects the information needed for plot-level early warning over one to seven days.

The evidence came from an approximately 18-hectare demonstration orchard near the boundary between southern Jiangxi and northern Guangxi, where navel orange and Wokan mandarin were the main subtropical citrus varieties. Twelve sampling plots were differentiated by topography, variety and cultivation management. Each plot had an automatic weather station, a soil-moisture sensor and yellow sticky-trap monitoring devices, while professional plant-protection technicians conducted weekly field surveys.

Integrating these measurements at daily granularity produced approximately 18,720 samples with 14 variables: six meteorological features, three phenological features, soil moisture, elevation, slope aspect, pesticide-application timing and citrus variety. Data preparation included missing-value imputation, outlier detection, normalisation and label encoding. Continuous variables with gaps were filled with a five-nearest-neighbour procedure, while categorical variables used the modal value.

Outliers identified by the three-sigma rule were manually checked, and continuous features were standardised using training-set means and standard deviations. The records were divided in chronological rather than random order: 2019–2022 supplied about 12,540 training samples, 2023 supplied about 3,090 validation samples and 2024 supplied about 3,090 test samples. This split was intended to reproduce forecasting in which future conditions are predicted from historical observations and to reduce leakage from future information.

ST-CitrusFormer uses an encoder–decoder architecture with separate temporal and spatial attention streams. Each plot’s 14 raw variables are first projected into a 128-dimensional latent representation and combined with sinusoidal positional encoding so that temporal order remains available to the network. Three parallel one-dimensional convolutions, with kernel widths of three, five and seven days, extract short-, medium- and longer-window local patterns.

Their outputs are merged with a point-wise convolution, and a residual connection retains the shallow-layer signal before the representations enter the attention branches. The temporal branch applies eight-head self-attention independently to each plot’s history. Its sliding window allows a time step to attend only to nearby positions, reducing the stated computational complexity from quadratic in the full sequence length to linear in sequence length multiplied by window size.

The spatial branch represents the 12 sampling plots as nodes in a directed, weighted graph. Edge weights combine distance between plot centroids with management similarity derived from cultivar, terrain aspect and overlap in pesticide-application records. The distance bandwidth was fixed at the median pairwise plot distance, about 150 metres for this orchard, while the three management components were weighted equally because the authors had no prior evidence favouring one of them.

Multi-head graph attention aggregated information from the five nearest graph neighbours at each of four layers. A gated fusion module then combined the two streams. Its learned gate determined, element by element, how much information to retain from temporal and spatial attention.

Fused features passed through a two-layer Transformer decoder and a dual prediction head: a five-class softmax output represented severity level, while a linear output estimated occurrence intensity. To address the strongly imbalanced severity distribution, the classification component used Focal Loss, which reduces the contribution of abundant easy examples and increases attention to rare high-severity cases. Intensity regression used Huber Loss, with residuals up to 1.0 in normalised severity units penalised quadratically and larger residuals penalised linearly.

The combined objective weighted Focal and Huber losses at 0.6 and 0.4, respectively, and was optimised with AdamW, cosine learning-rate annealing and early stopping based on the 2023 validation set. On the 2024 test set, ST-CitrusFormer produced the best average classification F1-score across the four targets. Its score of 0.812 exceeded those of LSTM, GRU, TCN, Transformer, Informer, PatchTST and TimeXer by 7.9, 6.8, 5.7, 4.1, 3.5, 2.0 and 1.1 percentage points, respectively.

The model also recorded the lowest RMSE and MAE among evaluated baselines at one-, three- and seven-day horizons. At seven days, its RMSE was 11.6% lower than PatchTST’s and 5.2% lower than TimeXer’s. In a representative high-occurrence summer period, its seven-day trajectory followed the timing and magnitude of two observed peaks more closely than the comparison models, while LSTM and Informer showed visible lag and oscillation.

Across five random seeds and four tasks, Wilcoxon signed-rank tests found the improvements over every evaluated baseline statistically significant at p<0.05, with p<0.001 against the classical baselines. Ablation experiments indicated that several components made distinct contributions. Removing the spatial graph-attention branch caused the largest fall in average F1, a decline of 3.1 percentage points.

Removing gated fusion and Focal Loss reduced the score by 2.0 and 2.7 points, respectively, while removing the multi-scale convolutions had a smaller effect. The largest category-specific decline was for Huanglongbing, whose F1-score fell by 4.9 points without spatial attention. Attention visualisations showed temporal heads concentrating on intervals containing concentrated rainfall, abrupt humidity increases and sharp temperature changes, while within-block connections received greater spatial attention than cross-block connections.

The researchers interpreted these patterns cautiously: they are consistent with the meteorological sensitivity of red mite and aphid dynamics and with the vector-mediated spread of Huanglongbing, but attention weights alone do not establish a biological mechanism. Robustness analyses were encouraging within the orchard but limited in scope. Four management blocks—Slope–Navel, Flat–Navel, Slope–Wokan and Flat–Wokan—were each held out in turn, and 2022, 2023 and 2024 were each excluded in separate leave-one-year analyses.

Average F1 remained at or above 0.77 in all these settings, with maximum ranges of 2.1 percentage points across blocks and 1.2 points across years. These folds differed from the main chronological split, so their results are not directly comparable with the 0.812 headline score; the leave-one-year analysis also relaxed the restriction against using future observations and should not be interpreted as a leakage-free operational evaluation. The findings support stability among plots and years within the same site, not transfer to an independent orchard.

Computational cost was measurable but not negligible. ST-CitrusFormer contained 2.05 million parameters and required 2.1 milliseconds for single-sample inference on an Nvidia RTX 4090 under PyTorch 2.2, with peak GPU memory consumption of 2.7 gigabytes. It was heavier than recurrent and convolutional baselines but broadly comparable to Informer and TimeXer and used fewer floating-point operations than Informer.

No CPU-only or embedded-accelerator benchmark was conducted, so the study does not demonstrate operation on hardware such as a Raspberry Pi. Quantisation, structured pruning, knowledge distillation and device-specific profiling were identified as future requirements. The authors instead described lightweight edge sensors transmitting observations to a local gateway or cloud server as a more realistic near-term configuration.

Several limitations constrain how the results should be understood. The dataset came from one orchard with relatively homogeneous varieties, management and climate, leaving generalisation to pomelo, lemon, other citrus types, different agro-climatic zones and extreme-weather seasons unestablished. Neighbouring training and test plots may also retain spatial correlations that could make within-orchard estimates mildly optimistic.

The pipeline’s sensitivity to sensor failure, prolonged missingness and imperfect inputs was not systematically quantified, and the model did not incorporate imagery from unmanned aerial vehicles, thermal cameras or hyperspectral remote sensing. Training also requires labelled field surveys, while forecasts beyond seven days remain untested. MC Dropout and deep ensembles were evaluated for uncertainty, but calibrated uncertainty estimates were not yet integrated into operational decisions.

The dataset and code are not publicly available but may be obtained from the corresponding author on reasonable request. Within those constraints, the study shows that combining temporal attention with a geographically and managerially informed graph can improve one- to seven-day forecasts of citrus pest and disease occurrence within a finely monitored orchard. The model is best understood as a candidate plot-level forecasting component for an existing IoT-based management platform, complementing rather than replacing field inspection and specialist judgement.

Multi-site, multi-variety and multi-climate validation, spatially disjoint external testing, missing-data stress tests and explicit integration of uncertainty into decision workflows will be needed before large-scale deployment can be justified.

Subject of Research: Spatiotemporal Transformer modelling for seven-day citrus pest and disease forecasting within a single orchard

Article Title: A hybrid transformer with spatiotemporal attention for citrus pest and disease forecasting

Article References: Ji, C., & Xie, H. (2026). A hybrid transformer with spatiotemporal attention for citrus pest and disease forecasting. Discover Artificial Intelligence, 6(1), Article 1202. https://doi.org/10.1007/s44163-026-02274-0

Image Credits: AI Generated

DOI: 10.1007/s44163-026-02274-0

Keywords: citrus forecasting, spatiotemporal modelling, hybrid Transformer, graph attention, temporal attention, precision agriculture, plant pathology, orchard sensors, multi-step forecasting, pest management, hybrid, transformer

Researchers from the University of Florida surveyed nematode communities in four strawberry fields in Hillsborough County, three managed conventionally and one managed organically. Sampling took place in June 2021, roughly two months after the strawberry crop was terminated, when soil temperatures were among the hottest of the year. Using a gas-powered auger, the team collected soil cores from four separate depths, from the surface down to one metre, at known hot spots of the northern root-knot nematode, Meloidogyne hapla, a pest likely introduced to Florida on infected transplants from out-of-state nurseries.

Each sample was split and processed two ways. The traditional approach involved extracting live nematodes from the soil using centrifugal flotation, counting them under an inverted microscope and identifying plant parasites to genus or species by morphology, while free-living worms were assigned only to broad trophic groups. The second approach was a targeted form of mitochondrial metagenomics, or mtMG. Whole DNA was extracted from the entire nematode suspension, sequencing libraries were prepared and then enriched for mitochondrial DNA using hybridisation capture probes designed against nematode mitochondrial genomes. The samples were sequenced on an Illumina NovaSeq 6000, and the resulting reads were assembled into contigs and matched against a curated nematode mitochondrial reference database.

The molecular workflow required careful quality control. Of 301 contigs that passed initial classification, 35 were discarded after phylogenetic tree analysis revealed poor alignments suggestive of misidentification, a problem concentrated in the suborder Tylenchina. The team also screened for haplotypes, closely related sequence variants within species, using maximum likelihood trees with bootstrap support, distinguishing genuine population diversity from assembly errors. This conservative filtering proved essential for trustworthy species-level detections.

The comparison revealed significant differences in both richness and composition between the two methods. The annotated COX1 mtMG approach, which identified mitochondrial contigs by open reading frame annotation rather than read mapping, recovered far more taxa than the read-mapping variants and came closest to mirroring the morphological data. It tentatively identified 14 bacterial-feeding species across nine genera that morphology could only label as ‘bacterial feeders’, and it detected animal-parasitic nematodes, likely present as eggs or unidentifiable juvenile stages, that microscopy missed entirely in every sample.

Yet the molecular method had its own blind spots. Morphology detected plant parasites such as the sting nematode Belonolaimus longicaudatus, Nanidorus minor and Tylenchorhynchus species that mtMG failed to recover, almost certainly because their mitochondrial reference sequences are missing from the database. Conversely, mtMG picked up Mesocriconema species and the dagger nematode Xiphinema brevicollum that microscopy had not found. The two approaches also disagreed on omnivores and predators, which appeared only in the morphological counts, probably because their low abundance meant too few sequencing reads to assemble.

Where the methods agreed, the agreement was powerful. Both consistently detected Meloidogyne across all fields, and mtMG went further, precisely identifying M. hapla in three fields while also revealing the co-occurrence of M. javanica, M. incognita and M. luci in the organic field, along with multiple haplotypes of M. hapla that morphology cannot distinguish. This capacity to resolve species and population-level variation matters enormously, because species identity can dictate which management strategy a grower should adopt, and haplotype diversity has been linked to emerging nematicide resistance.

Despite their differences in detected taxa, both methods converged on the same ecological conclusions. Field identity was the dominant factor structuring nematode community composition, reflecting each field’s distinct management history, and sampling depth played a significant role as well. The mtMG data in particular showed that Meloidogyne species were recovered throughout the entire one-metre soil column, supporting earlier evidence that plant-parasitic nematodes migrate vertically through Florida’s deep sandy soils, likely in response to heat, tillage or fumigation. This vertical movement has practical implications, since deep soil layers may act as reservoirs from which root-knot populations re-establish after fumigation.

The authors conclude that mtMG and morphology are highly complementary tools, and that combining them yields a far more accurate picture of nematode diversity than either alone. They identify four priorities for improving the molecular approach: expanding the phylogenetic coverage of the mitochondrial reference database, refining contig filtering and classification pipelines, improving detection of under-represented trophic groups such as omnivores and predators, and adding reference sequences for rare crop pests. As the database grows, mtMG should become increasingly capable of flagging subtle genetic variants, including nematicide-resistant haplotypes, giving growers an earlier warning than morphology could ever provide.

For Florida’s strawberry industry, which produced roughly 511 million dollars’ worth of fruit in 2022 across some 12,000 acres, the stakes are considerable. Sting and root-knot nematodes remain among the most damaging pests growers face, and management increasingly relies on targeted, environmentally sustainable approaches rather than broad-spectrum fumigants. Accurate, species-level monitoring of entire nematode communities, the kind of monitoring this study shows is achievable when molecular and morphological methods are deployed together, could become a cornerstone of that transition.

Subject of Research: Comparison of mitochondrial metagenomics and morphological identification for surveying nematode communities in strawberry fields

Article Title: Improving understanding of nematode communities in agricultural settings: a comparison of mitometagenomics and morphology

Article References: Gendron, E. M. S., Oliveira, C. J., Desaeger, J., & Porazinska, D. L. (2024). Improving understanding of nematode communities in agricultural settings: a comparison of mitometagenomics and morphology. Metabarcoding and Metagenomics, 8, Article e123387. https://doi.org/10.3897/mbmg.8.123387

Image Credits: AI Generated

DOI: 10.3897/mbmg.8.123387

Keywords: nematodes, mitochondrial metagenomics, morphology, strawberry, Florida, plant-parasitic nematodes, Meloidogyne, soil biodiversity, metabarcoding, agriculture, soil depth, pest management

The cotton leafworm was no ordinary insect in late nineteenth-century São Paulo. A Neotropical moth historically distributed from southern Canada to northern Argentina, it was the primary leaf-feeding pest of upland cotton, Gossypium hirsutum. Its larvae feed voraciously after flowering, stripping foliage, impairing photosynthesis and reducing both fiber quality and yields. Between the 1950s and the 1980s, infestations in Brazil cut lint yields by 21 to 35 percent, with particularly severe losses in São Paulo. By the time the Tatuí specimen was pinned, the species was already recognized as a widespread threat across the country’s cotton-producing regions, and its economic importance was such that in 1882 and 1883 the German-American entomologist Albert Koebele, a pioneer of biological control, was sent to Brazil specifically to investigate cotton pests, including A. argillacea.

The specimen itself was identified by external morphological traits, including wing shape, coloration and a distinctive white spot on the forewing, features consistent with the historical description of the species and with Hübner’s early nineteenth-century plates of a male from the state of Bahia in northeastern Brazil. Attached to the pin is a label bearing the name of the collector, Francisco de Salles Gomes, who sent the insect from Tatuí to the Museu Paulista in 1899. That label proved to be the thread the researchers pulled to reconstruct the specimen’s entire documentary trail, from the letters that preceded its incorporation into the museum to its entry as specimen number 356 in the institution’s first entomological logbook.

Those letters are remarkable in their own right. In May and June 1899, Salles Gomes, then the Intendente, or mayor, of Tatuí, wrote to Dr. Hermann von Ihering, the German-Brazilian zoologist who directed the Museu Paulista. The first letter describes widespread larval infestations in local cotton fields and stresses the urgency of entomological investigation; the second reports the shipment of collected specimens to the museum. Salles Gomes was not a disinterested public servant. He was a founder of the Companhia de Fiação e Tecidos Santa Maria, a textile company in Tatuí, and his engagement likely reflected both political leadership and direct economic stakes in protecting the fiber that fed his mills. The correspondence captures a moment when municipal governance, private industry and scientific institutions converged around a shared agricultural problem.

The man receiving those letters was a towering figure in Brazilian natural history. Born in Kiel, Germany, in 1850, Hermann Friedrich Albert von Ihering studied medicine and zoology at Giessen, Leipzig, Berlin and Göttingen, earning his doctorate in 1873. After emigrating to Brazil in 1880, he worked as a physician and naturalist in Rio Grande do Sul before becoming the first director of the Museu Paulista in 1893. Under his leadership the museum became a leading center for zoological and anthropological research, and Ihering himself investigated agricultural pests, publishing on sap-sucking insects including Phylloxera in vineyards and on scale insects that damaged coffee, citrus and jaboticaba trees. He designated specific museum laboratories for systematic and economic entomology, and his contributions were recognized with honorary membership in the National Agricultural Society, a signal of how deeply curatorial policy at the time was woven into the economic concerns of the day.

This model was not uniquely Brazilian. In the late nineteenth and early twentieth centuries, institutions such as the Smithsonian Institution in the United States and the Natural History Museum in London similarly cataloged agricultural pests and collaborated with applied entomology programs, curating insect collections that documented species distributions and their ecological and economic impacts. The Museu Paulista’s acquisition and preservation of pest specimens fit squarely within this global trend, in which natural history museums functioned as early infrastructure for pest management science. The insect collections assembled in that era eventually formed the core of what is now the MZUSP, which was transferred in 1939 to the newly created Department of Zoology under the Secretariat of Agriculture, Industry, and Commerce, gained a dedicated facility between 1940 and 1941, and was formally incorporated into the University of São Paulo in 1969. Today the museum houses more than 13 million specimens and remains a vital center for biodiversity research in Latin America.

What makes the Tatuí moth more than a historical curiosity is its potential as a scientific resource. The specimen has not yet undergone molecular analysis, but recent advances in museomics, the application of genomic techniques to preserved museum specimens, offer promising avenues for future research. Museomic studies on other insects have revealed genetic changes linked to environmental pressures, pesticide resistance and host-plant adaptation. Sequencing the 1899 specimen could help trace long-term genetic variation and evolutionary responses to more than a century of agricultural intensification, pest control and environmental change. Comparisons with modern populations of A. argillacea would illuminate how the species’ genetic diversity and adaptive capacity have been shaped since the era of manual removal and rudimentary chemical treatments that laid the groundwork for today’s integrated pest management.

Morphological comparisons could be equally revealing. Differences in size and structural traits between historical and contemporary specimens may reflect shifts in climate conditions or host-plant interactions, and examining preserved specimens alongside modern ones could document phenotypic change over time as agricultural landscapes evolved and selective pressures favored resilience against environmental stressors. Historical climate data can also be examined alongside infestation records to evaluate the role of climatic fluctuations in pest outbreaks. Together, these integrated approaches combine entomological records with agronomic and environmental data to refine understanding of both past and present pest management challenges, insights relevant not only to entomologists but also to agronomists designing more effective control strategies.

The species’ own history underscores why such baselines matter. Native populations of A. argillacea probably fed on wild plants in the Malvaceae, the family that includes cotton, but as commercial cotton farming intensified, the moth became increasingly specialized on cultivated fields. By the early twentieth century it was reported almost exclusively from agricultural areas, with few or no records of wild hosts, a striking example of how anthropogenic pressures can drive ecological specialization in pest species. In the United States and Canada the species has been extirpated since 1998, following the decline of cotton cultivation there and the widespread use of insecticides, while in Brazil it remains a major pest whose impact varies by region and season, arriving late in the season in Southern and Central Brazil but attacking early crop stages in the Northeast. Broad-spectrum insecticides initially provided control but raised environmental concerns and led to resistance, prompting the adoption of biological control and integrated pest management approaches.

As the study’s authors emphasize, natural history collections are critical scientific resources that support conservation both directly, through species and habitat management, and indirectly, by informing research, education and policy. They provide unique records of species interactions over time and establish the historical baselines essential for assessing ecological change and guiding modern strategies in agricultural science and pest management. That the researchers could reconstruct the full history of a single specimen, from the labeled object in the collection to the correspondence that preceded its accession and its entry in the institutional logbook, is itself noteworthy, since gaps in documentation are common and often frustrate work with historical material. The 1899 moth from Tatuí thus stands as tangible evidence of the challenges posed by agricultural pests in nineteenth-century Brazil, of the local officials and scientists who confronted them, and of the enduring power of museum drawers to connect past and present in the service of sustainable agriculture.

Subject of Research: Historical documentation and museomic potential of an 1899 cotton leafworm specimen from São Paulo, Brazil

Article Title: The oldest moth in the Museu de Zoologia da Universidade de São Paulo and its connection to 19th-century cotton agriculture in São Paulo, southeastern Brazil

Article References: Duarte, M., de Oliveira e Silva, R., & Landim, M. I. (2025). The oldest moth in the Museu de Zoologia da Universidade de São Paulo and its connection to 19th-century cotton agriculture in São Paulo, southeastern Brazil. Natural History Collections and Museomics, 2, 1-15. https://doi.org/10.3897/nhcm.2.154707

Image Credits: AI Generated

DOI: 10.3897/nhcm.2.154707

Keywords: Alabama argillacea, cotton leafworm, Museu de Zoologia da Universidade de São Paulo, museomics, natural history collections, cotton agriculture, São Paulo, agricultural entomology, pest management, Hermann von Ihering, Tatuí, nineteenth century

The review, led by Mira Word Ries of Arizona State University together with an international team that includes some of the most experienced locust researchers and managers in the world, takes an unusually broad view of the problem. Rather than treating locusts purely as an entomological puzzle, the authors frame management as a social-ecological-technological system, often abbreviated as SETS. In this framing, the insects themselves, the landscapes and climates that drive their population booms, the pesticides and monitoring technologies deployed against them, and the human institutions, funding streams, and communities affected by outbreaks are all interlocking components. Failure in any one of them can undermine the entire response, as history has repeatedly shown.

The economic logic of locust control is central to the argument. Preventive management, in which early-warning systems detect gregarizing populations and small-scale treatments stop them before they coalesce into swarms, is widely regarded as far cheaper than reacting to full-blown plagues. Yet the paper highlights a persistent weakness: funding for locust management tends to surge during crises and collapse in the quiet years that follow. This boom-and-bust cycle leaves monitoring networks under-resourced precisely when they need to maintain the baseline surveys that make early intervention possible. The authors point to recurring fluctuations in funding, focus, and institutional attention as chronic vulnerabilities that no single nation or agency can solve alone.

To understand how the major locust-affected regions are actually governed, the coauthors examine the organizations responsible for management across the world’s key hotspots. These include long-standing bodies that coordinate desert locust control in Africa, the Middle East, and western Asia, along with national programs in the Americas, Australia, and Asia that deal with species such as the South American locust, the Australian plague locust, and the migratory locust. The paper assesses the relationships, strengths, and weaknesses of these organizations, revealing a patchwork in which capacity varies enormously. Some agencies operate sophisticated satellite-backed forecasting and dedicated survey teams; others struggle with basic logistics, political instability, or jurisdictions that end abruptly at borders the insects simply fly across.

Communication and trust emerge as themes as important as any insecticide. The authors emphasize that isolated agendas, limited transparency, and fragmented data sharing hamper both research and response. During the devastating 2019 to 2021 desert locust upsurge in East Africa and Southwest Asia, the world was reminded how quickly gregarious populations can multiply, with a single generation of swarms capable of increasing twentyfold and successive generations compounding the damage. Effective response in such events depends on timely data flowing from field surveys to forecasters to decision-makers, and ultimately to the farmers whose fields and pastures are at stake. Where those channels are weak, both crops and confidence erode.

The review also delves into the history of locust biology, a field with a remarkable scientific legacy. One of the pivotal insights of twentieth-century entomology was phase polyphenism: the discovery that certain grasshoppers can transform between a solitary phase, in which they live dispersed and relatively inconspicuous, and a gregarious phase, in which they band together, change color and behavior, and form the marching bands and flying swarms that give locusts their fearsome reputation. This shift is triggered by crowding and mediated by chemical and neurological signaling, and understanding its mechanisms remains central to predicting when and where outbreaks will occur. The paper traces how this biological knowledge, from phase theory to modern genomics, remote sensing, andpheromone research, has shaped management strategies across decades.

Modern technology is transforming the toolkit. Satellite imagery of vegetation and rainfall can flag the green flushes that signal favorable breeding conditions for desert locusts, while machine-learning models increasingly help forecasters anticipate where solitarious populations are likely to concentrate. GPS-tagged survey routes, mobile data collection, and improved GIS platforms allow field teams to map outbreaks in near real time. At the same time, the authors stress that technology alone is not a solution. Pesticide use remains the primary weapon against active swarms, but it carries environmental and human health costs, particularly when broad-scale treatments affect non-target organisms or when safety standards for applicators and rural communities lag behind. The review calls for stronger environmental and human health standards alongside continued exploration of biopesticides, such as fungal-based control agents, and other sustainable alternatives.

The social dimension of locust management receives rare and sustained attention. The coauthors include researchers from the humanities and social sciences who examine how communities perceive locust risk, how land-use decisions by herders and farmers shape habitat for outbreaks, and how the costs and benefits of control are distributed. Questions of food security sit at the heart of the matter: locust outbreaks can devastate the pastures and croplands of some of the world’s most vulnerable populations, and management decisions made at regional or international levels ripple down to household livelihoods. The authors argue that embedding local knowledge and community engagement into monitoring and response is not a courtesy but a necessity for effective, sustainable control.

From this wide-ranging synthesis, the paper draws a roadmap of transdisciplinary opportunities. It calls for sustained, crisis-proof funding mechanisms that keep early-warning infrastructure alive between plagues; for deeper integration of ecological forecasting with social and economic analysis; for open data standards that allow agencies on different continents to learn from one another; and for research that pairs laboratory breakthroughs on locust phase biology with field-tested, environmentally sound control methods. It also champions cross-sector collaboration among governments, international bodies, universities, and private actors, arguing that the complexity of the SETS demands exactly the kind of collective authorship the paper itself embodies.

Institutional response is already taking shape. Arizona State University launched the Global Locust Initiative in 2018 as a direct answer to the fragmentation the review describes, creating a hub that connects researchers, managers, and stakeholders across sectors and continents. The initiative welcomes participants interested in locusts and grasshoppers, transboundary pests, integrated pest management, landscape-level processes, food security, and cross-sectoral collaboration. As climate variability alters rainfall patterns and land use across the world’s arid and semi-arid zones, the conditions favorable to locust outbreaks may shift in unpredictable ways, raising the stakes for preparedness. The message of this unprecedented global collaboration is clear: locusts will always be a moving target, but with connected science, stable investment, and shared responsibility across disciplines and borders, the world need not always be caught reacting to the next plague.

Subject of Research: Transboundary locust and grasshopper pest management as a social-ecological-technological system

Article Title: Global perspectives and transdisciplinary opportunities for locust and grasshopper pest management and research

Article References: Word Ries, M., Adriaansen, C., Aldobai, S., Berry, K., Bal, A. B., Catenaccio, M. C., Cigliano, M. M., Cullen, D. A., Deveson, T., Diongue, A., Foquet, B., Hadrich, J., Hunter, D., Johnson, D. L., Pablo Karnatz, J., Lange, C. E., Lawton, D., Lazar, M., Latchininsky, A. V., … Cease, A. (2024). Global perspectives and transdisciplinary opportunities for locust and grasshopper pest management and research. Journal of Orthoptera Research, 33(2), 169-216. https://doi.org/10.3897/jor.33.112803

Image Credits: AI Generated

DOI: 10.3897/jor.33.112803

Keywords: locusts, grasshoppers, pest management, transboundary pests, desert locust, swarms, integrated pest management, food security, early warning systems, pesticide use, Global Locust Initiative, social-ecological-technological system

The research, conducted by a team of Argentine scientists at the National Institute of Agricultural Technology (INTA) and the National University of Tucumán, compared six host fruit species under controlled laboratory conditions: guava and feijoa, both native to South America, and grapefruit, peach, plum, and loquat, all introduced. The team reared fly larvae in each fruit and then measured an unusually comprehensive suite of traits spanning the entire life cycle: how long larvae took to reach pupation, how many pupae survived to adulthood, how large the resulting adults were, how well female ovaries developed, how quickly males reached sexual maturity, and how long adults survived without food.

The results, published in Frontiers in Insect Science, revealed that host species significantly affected every single trait measured. Guava, a native host and one of the fruits most heavily infested by this species in nature, supported the fastest larval development of all six fruits, with larvae pupating in just about 13 days. Grapefruit, by contrast, stretched development to nearly 30 days. Peach and feijoa fell in between at roughly 16 days, with loquat and plum taking around 19 to 21 days. On the surface, guava looks like the ideal nursery—fast development is generally considered a hallmark of host quality, because larvae that linger face greater exposure to predators, parasitoids, pathogens, and the vagaries of weather.

But the adult stage told a very different story. Flies reared in guava emerged as smaller adults, with shorter wings, reduced ovary development, and lower resistance to starvation. Females from guava also showed a higher proportion of atrophied, non-functional ovaries. Meanwhile, the exotic stone fruits—peach and plum—consistently produced the largest adults with the best-developed ovaries and the highest overall performance across multiple traits. Plum also conferred the greatest starvation tolerance, meaning that flies raised in this introduced host could survive longest without food. Loquat generally produced the poorest outcomes, yielding the smallest females, the smallest ovaries, and delayed male sexual maturation, although it did confer relatively good starvation resistance.

Some of the patterns were genuinely counterintuitive. Grapefruit, which forced the slowest larval development, nonetheless produced males that began their sexual calling behavior—the pheromone-releasing courtship display—earlier than males from any other host. Grapefruit females also showed no ovarian atrophy at all, and adult emergence from grapefruit was among the highest recorded. Feijoa, the other native Myrtaceae tested, behaved similarly to guava in supporting rapid development, but it had the lowest pupal survival of any fruit, with only about 17 percent of pupae reaching adulthood, even as it accelerated male sexual maturation. Each fruit, in other words, was a bundle of benefits and costs rather than a simple ranking of good and bad hosts.

To integrate these multidimensional results, the researchers built a heatmap of standardized trait values and performed a principal component analysis combining the biological data with published nutritional profiles of each fruit from the USDA nutrient database. The first two principal components explained roughly 71 percent of the total variance, and the analysis confirmed visually what the individual statistics implied: no fruit species sat at the optimum for all traits. Peach and plum overlapped almost completely in the PCA space, indicating similar nutritional and life-history profiles, while guava, grapefruit, and loquat each occupied distinct positions associated with different combinations of nutrients and performance outcomes.

The nutritional analysis added an intriguing layer of complexity. Guava had the highest protein, lipid, fiber, carbohydrate, and mineral content of all six fruits—on paper, it should have been the best host. Yet apart from fast development, guava performed poorly on nearly everything else. The authors suggest several non-exclusive explanations. Guava pulp contains phenolic compounds and tannins, potent anti-nutritional defenses that force developing larvae to divert metabolic energy into detoxification rather than growth. The ratio of sucrose to total phenolics in guava is dramatically lower than in peach, imposing a much heavier chemical burden. Guava’s high ash content may also disrupt osmotic balance and reduce nutrient absorption efficiency. Whatever the precise mechanism, the lesson is clear: macronutrient content alone cannot predict host quality, and the evolutionary history between insect and plant matters in ways that simple chemistry tables miss.

Perhaps the most provocative implication concerns the so-called mother-knows-best hypothesis, which holds that female insects preferentially oviposit on hosts that maximize their offspring’s fitness. Field surveys have long suggested that A. fraterculus predominantly infests native fruits, and guava and feijoa are among the most heavily attacked species in nature. Yet in the laboratory, these native hosts did not deliver superior offspring performance. The study’s authors are careful to note the caveats: laboratory conditions exclude many ecological factors—temperature fluctuations, desiccation risk, natural enemies, seasonal constraints—that shape host suitability in the real world, and field infestation levels are not always a direct proxy for female preference. Still, the results point toward a decoupling between where these flies lay their eggs and where their offspring actually thrive, a mismatch documented in other phytophagous insects as well.

One possible resolution, the researchers propose, is that different hosts are advantageous at different times. In temperate Argentina, native and exotic fruits ripen in different seasons, and a slow-developing larva inside a grapefruit may actually benefit when temperatures drop—citrus pulp stays moist for weeks and could even allow larvae to overwinter in place, a strategy documented in the Mediterranean fruit fly. Early-season hosts might favor rapid reproduction, while late-season hosts might enhance survival and stress tolerance. Under this view, no host is universally optimal because the demographic needs of the population shift across the year, and the very notion of a single best fruit dissolves into a seasonal mosaic of partial advantages.

Beyond its evolutionary interest, the work carries practical weight. A. fraterculus is actually a cryptic species complex, infests more than 177 reported host plant species, and triggers stringent quarantine restrictions that complicate fruit exports from South America. Understanding which hosts produce larger, more fecund, more stress-tolerant adults—and which merely produce fast ones—can improve predictions of pest population dynamics and sharpen integrated management strategies, from monitoring programs to sterile insect technique campaigns that depend on producing high-quality mass-reared flies. The study also flags clear priorities for future research: behavioral assays of oviposition preference paired with biochemical analyses of fruit chemistry, including secondary metabolites that current nutritional databases do not capture. For now, the message for anyone modeling this pest is sobering and liberating at once: the fruit fly’s world is not a ladder of better and worse fruits, but a landscape of trade-offs in which every host wins something and loses something else.

Subject of Research: Effects of host fruit species on larval development and adult life-history traits of the polyphagous pest Anastrepha fraterculus

Article Title: No single host is optimal: effects of fruit species on development and adult performance in Anastrepha fraterculus (Diptera: Tephritidae)

Article References: Pacelli, F., Ruiz, M. J., Devescovi, F., Nussenbaum, A. L., Garbalena, M. S., Bachmann, G. E., Milla, F. H., Vera, M. T., & Segura, D. F. (2026). No single host is optimal: effects of fruit species on development and adult performance in Anastrepha fraterculus (Diptera: Tephritidae). Frontiers in Insect Science, 6, Article 1832786. https://doi.org/10.3389/finsc.2026.1832786

Image Credits: AI Generated

DOI: 10.3389/finsc.2026.1832786

Keywords: Anastrepha fraterculus, Tephritidae, fruit flies, host plant use, life-history trade-offs, larval development, ovary development, sexual maturation, starvation resistance, polyphagy, pest management, entomology

The challenge that INApest() addresses stems from how biosecurity threats actually unfold on the ground. When a new pest arrives in a region, its fate is determined not only by its intrinsic biology—how fast it reproduces, how far its seeds or propagules travel, which climates it tolerates—but also by a web of human decisions. Landowners choose whether to report suspicious sightings, whether to adopt control measures on their properties, and whether to cooperate with neighbours and authorities. Agencies choose where to allocate surveillance effort, how to communicate risk, and whether to impose movement restrictions. Each of these decisions feeds back into the others: a detection can raise awareness among nearby landholders, which in turn can increase reporting and adoption of management, which in turn alters the apparent and actual distribution of the pest. Traditional models that treat these elements separately struggle to capture such feedbacks, and the authors argue that this fragmentation has left biosecurity planners with an incomplete picture of how responses succeed or fail.

The architecture of the new framework reflects this integrated view. Rather than modelling a landscape as a continuous surface, INApest() represents farms, properties or management units as nodes in a network, connected by pathways of natural dispersal and by human-mediated movement such as the transport of stock, machinery, fodder or seed. This network structure allows the simulation to track both the biological spread of an organism and the social spread of information—knowledge about the pest, about detection events, and about the effectiveness of different control technologies—through the same landscape. As the simulation runs, new detections occur, information propagates between neighbours, and management responses evolve organically rather than being fixed in advance. This dynamic coupling of ecology and sociology is what distinguishes the approach from earlier scenario tools that required managers to specify human behaviour as static external assumptions.

The framework decomposes any biosecurity response into four parameters that can be varied independently: the probability that infestations are detected, the proportion of landholders who adopt recommended management, the efficacy of local eradication efforts against established populations, and the degree to which spread is reduced through measures such as movement restrictions, hygiene protocols or buffer zones. By varying these parameters across thousands of combinations, managers can explore which mixes of intervention are most likely to achieve outcomes such as eradication, containment or slowed spread, and how sensitive those outcomes are to uncertainty in any single component. This parameterisation also makes the framework transparent: rather than embedding assumptions deep inside a black-box model, it exposes the levers that decision-makers can actually pull, at least conceptually, through policy, funding and engagement.

The case study used to demonstrate the method concerns Chilean needle grass, a slow-spreading pasture weed of South American origin that has become a serious problem in south-eastern Australia and New Zealand. The weed damages livestock production partly because its sharp, penetrating seeds lodge in the eyes, hides and fleece of grazing animals, and partly because it displaces palatable pasture species. Its spread is comparatively slow and well documented, which makes it an ideal test case: the authors were able to derive dispersal kernels from known natural mechanisms such as wind, water and animal movement, together with human-mediated pathways, and show that under a scenario with no management the model reproduces realistic historical spread rates. This calibration step is important because it demonstrates that the network representation is grounded in empirical reality rather than being a purely theoretical construct, giving managers reason to trust the relative comparisons the model produces.

Among the most striking findings from the scenario analysis of more than fifteen thousand parameter combinations is the outsized role of communication. When farms neighbouring known infestations are informed about the invasive threat, the level of farm-scale eradication efficacy and spread reduction needed for management success drops substantially. In other words, simply telling the right people about the risk functions as a multiplier on every other intervention: surveillance, control and containment all become more effective when the landholders closest to the front line understand what is at stake and what they can do about it. This result resonates with long-standing observations in invasion ecology that early detection and rapid response depend heavily on community engagement, but the framework makes the relationship quantitative, allowing managers to estimate how much additional management effort would be needed to compensate for poor communication, or conversely, how much investment in outreach might save in on-ground control.

The targeted simulation experiments extend this insight in two directions that are particularly relevant to contemporary biosecurity policy. First, the framework can assess cross-border consequences of local management decisions. Because the network structure explicitly represents movement pathways between jurisdictions, a simulation can show how a control programme in one region affects invasion risk in a neighbouring region, and vice versa. This matters because biosecurity responsibilities are often divided among local, regional and national authorities, and decisions optimised for one jurisdiction may impose costs or benefits on others. A tool that makes these spillover effects visible supports more cooperative and coherent policy design. Second, the experiments show how communication between landowners interacts with climate change and surveillance effort. As climates shift, the areas suitable for a pest move, and the value of surveillance in particular locations changes accordingly; meanwhile, the flow of information among landholders determines how quickly the community adapts its behaviour to the emerging threat. The framework allows these interacting pressures to be explored together rather than in isolation, which is essential for planning under global change.

The broader significance of this work lies in what it suggests about the future of biosecurity modelling. Invasive species management has long drawn on epidemiological models, population viability analyses and decision theory, but the human dimension has often been relegated to qualitative discussion. By embedding social dynamics—adoption, communication, information spread—as first-class components of a quantitative simulation, INApest() points toward a style of modelling in which ecological and social processes are given equal analytical weight. This is increasingly recognised as necessary, because most invasion pathways are human-mediated and most management actions depend on human cooperation. The authors note that the framework has the potential to be applied at multiple scales, from individual properties to regions and nations, and to explore a wide range of management, global change and social scenarios, which suggests considerable flexibility for adaptation to other pests, weeds and pathogens beyond the case study presented.

For practitioners, the practical implications are concrete. The finding that neighbour communication lowers the management threshold for success implies that investment in extension services, community liaison and rapid notification systems should be evaluated not as soft overhead but as core biosecurity infrastructure with measurable returns. The four-parameter decomposition offers a checklist for programme design: managers can ask whether their planned response has adequate detection capability, realistic expectations of adoption, sufficient eradication efficacy at known infestations, and credible spread-reduction measures, and can use simulation to identify which of these is the binding constraint in their particular situation. Where resources are limited, the model can help identify the most cost-effective allocation—for example, whether additional surveillance effort or additional communication effort yields a greater improvement in the probability of successful containment.

There are also implications for how biosecurity agencies plan under uncertainty. Scenario analysis of the kind the framework enables does not predict a single future; instead, it maps the space of plausible outcomes under different assumptions, revealing which strategies are robust across many of them. This robustness framing is well suited to global change conditions, where climate envelopes, trade patterns and land-use pressures are all shifting in ways that are difficult to forecast precisely. A strategy that performs acceptably across a wide range of simulated futures is generally preferable to one that is optimal under a single best guess but fragile elsewhere. The capacity to simulate responses that evolve organically as detections accumulate further supports this adaptive view, because real biosecurity campaigns are iterative: each detection updates knowledge, and each update should inform the next round of decisions.

Finally, the study contributes to a growing recognition that biosecurity is as much a social challenge as a biological one. The weeds, pests and pathogens that cause the greatest harm are typically those that exploit human networks of trade, travel and land management, and the responses that succeed are typically those that mobilise those same networks for detection, reporting and control. By providing a quantitative bridge between these two sides of the problem, the framework offers researchers a platform for testing hypotheses about coupled human–natural invasion dynamics, and offers managers a way to rehearse interventions virtually before committing scarce resources in the field. As global change accelerates the movement of species across borders and the pressure on existing management systems intensifies, tools that integrate these dimensions within a single analytical frame are likely to become increasingly central to protecting agricultural production, biodiversity and the communities that depend on both.

Subject of Research: A network simulation framework integrating global change, social dynamics and management interventions for invasive species biosecurity scenario analysis

Article Title: Simple network models integrate global change, social dynamics and management interventions in biosecurity scenario analysis

Article References: Buddenhagen, C. E., McGrannachan, C., Bourdôt, G., Lamoureaux, S., Garrett, K. A., Kaine, G., & Mason, N. W. H. (2026). Simple network models integrate global change, social dynamics and management interventions in biosecurity scenario analysis. NeoBiota, 106, 107–139. https://doi.org/10.3897/neobiota.106.161880

Image Credits: AI Generated

DOI: 10.3897/neobiota.106.161880

Keywords: biosecurity, invasive species, network models, scenario analysis, Nassella neesiana, Chilean needle grass, global change, social dynamics, pest management, simulation modelling, climate change, surveillance

The research began with an observation that has puzzled entomologists working with this species. In laboratory settings, when eggs are manually removed from the substrate and washed before being offered to parasitoid wasps, they suffer heavy rates of parasitism. In the field, however, naturally laid eggs are attacked far less often. Something about the intact, naturally laid egg was deterring the wasps, and the research team, led by Professor Lian-Sheng Zang of the State Key Laboratory of Green Pesticides at Guizhou University, with doctoral candidate Hai-Xia Zhan as first author, set out to identify what it was.

Field experiments provided the first clue. Naturally laid eggs were attacked markedly less frequently by three species of egg parasitoid wasps. The specialist Mesocomys trabalae almost completely avoided intact eggs, while Anastatus japonicus made fewer contacts and fewer attempts to deposit its own eggs. Egg parasitoids such as these locate hosts through a combination of chemical and visual cues, searching for the volatile signatures that eggs and the surfaces around them release. Any compound that disrupts that search, or that signals a bad host, can dramatically reduce parasitism success.

Using behavioural assays combined with chemical analysis of the egg surfaces and their emissions, the team identified 2-ethyl-1-hexanol, abbreviated 2EH, as a compound distinctly associated with naturally laid eggs. Crucially, both wasp species could detect the chemical, and when they were exposed to it directly they showed clear avoidance behaviour. This established that 2EH is not merely a by-product of egg biology but an active signal in the chemical conversation between the silkworm eggs and their enemies.

The decisive experiment came next. When the researchers applied 2EH to eggs that had been washed clean of their natural chemical coating, parasitism dropped significantly. That single result confirmed the compound’s protective role: restoring the chemical restored the defence. It also explained the laboratory anomaly in a satisfying way. Washing removes 2EH, leaving the eggs chemically naked and highly attractive to the wasps, which is precisely why washed eggs have historically served as efficient hosts for rearing parasitoids in the lab.

“Our study reveals that exposed eggs are not inherently vulnerable but instead deploy an active volatile-based chemical defence,” explains Professor Lian-Sheng Zang, corresponding author of the study. “This challenges the conventional view that eggs without physical coverings or toxins are simply passive targets.” The statement captures a conceptual shift that ripples well beyond a single species. Insect eggs have traditionally been categorized along a spectrum of protection: some species bury their eggs inside plant tissue, others encase them in hardened shells, oothecae, or layers of hairs and wax, and still others arm them with alkaloids and other sequestered toxins. A fourth strategy, an airborne repellent that keeps enemies at a distance before they ever touch the egg, had received far less attention.

The technical implications of a volatile defence are considerable. Unlike a physical barrier, a volatile compound works at a distance, interfering with the host-location process itself. Egg parasitoids depend on olfactory cues to find cryptic, immobile targets, often following plant volatiles released in response to egg deposition or detecting cues on the egg surface during close-range contact. A compound such as 2EH, which the wasps can detect and avoid, effectively jam or repels that search. For the eggs, the strategy carries its own trade-offs in principle, since any volatile cue can potentially be exploited by other organisms, but in this system the net effect measured in the field was strongly protective against the wasp species tested.

The discovery also has immediate practical value. Oak silkworm eggs are widely used as alternative hosts for the mass rearing of parasitoid wasps deployed in biological control programmes against agricultural pests. Mass-rearing operations depend on getting wasps to parasitize the eggs efficiently, and the new results suggest that the repellent effect of 2EH could complicate or limit that process under some conditions. Understanding exactly how 2EH deters the wasps gives producers a chemical handle for optimizing these systems, whether by managing the compound’s presence on production eggs or by accounting for its effects when designing rearing protocols and release strategies.

Beyond the rearing facility, the work points toward greener approaches to pest management. 2-ethyl-1-hexanol is a naturally occurring compound, and the study raises the possibility of using it, or analogues inspired by it, as an environmentally friendly insect repellent. Push-pull and disrupt-the-cue strategies, which manipulate the odour landscape to steer pest insects away from crops or their hosts, are a growing branch of sustainable agriculture. A repellent that specifically protects a vulnerable life stage, the egg, against parasitoids or other egg-feeding enemies would be a welcome addition to that toolkit, particularly because it targets innate odour-guided behaviour rather than relying on broad-spectrum toxicity.

The authors are careful to note that much remains to be learned. Further research is needed to determine whether similar volatile-mediated defences occur in other insect species, how the eggs acquire or produce 2EH, and whether natural enemies might evolve countermeasures over evolutionary time. Researchers from the Guangdong Academy of Sciences, the Beijing Academy of Agriculture and Forestry Sciences, and international collaborators including Professor Guy Smagghe contributed to the study. For now, the message is clear and surprisingly elegant: even an egg lying in full view on an oak leaf can defend itself, not with armour, but with a scent its enemies have learned to avoid.

Subject of Research: Volatile chemical defence of oak silkworm eggs against parasitoid wasps

Article Title: Oak silkworm eggs use a natural chemical defence to keep parasitic wasps away

Article References: Oak silkworm eggs use a natural chemical defence to keep parasitic wasps away. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: oak silkworm, Antheraea pernyi, 2-ethyl-1-hexanol, chemical ecology, egg parasitoids, volatile repellent, Insect Science, biological control, insect defence, pest management, Mesocomys trabalae, Anastatus japonicus

The review synthesizes scientific literature with interviews of urban farmers in Malaysia, and it arrives at a nuanced picture. Urban agriculture can reduce water consumption, cut pesticide use, shorten supply chains and lower the greenhouse gas emissions associated with transporting food over long distances. It also supports two United Nations Sustainable Development Goals directly: Zero Hunger and Sustainable Cities and Communities. Yet the authors are careful to note that sustainability and yield remain major obstacles. Without deliberate strategies, many urban farms struggle to survive beyond their initial enthusiasm.

The market data cited in the review underline how quickly interest has grown. The global urban farming market was valued at approximately 139 billion US dollars in 2022 and is projected to reach about 219 billion dollars by 2028, a compound annual growth rate of 7.88 percent. A separate market report estimates revenue rose from 146.21 billion dollars in 2023 to 159.92 billion dollars in 2024. The COVID-19 pandemic accelerated this trend, as movement restrictions disrupted food supply chains and exposed the fragility of relying on distant producers for perishable goods.

Technically, urban farming is not a single method but a family of models built around four main techniques. Aeroponics suspends plant roots in air and mists them with nutrient solution. Hydroponics grows plants directly in nutrient-rich water without soil. Aquaponics couples hydroponic plant beds to fish tanks, using fish waste as fertilizer while plants clean the water. Vertical farming stacks these systems in multiple layers, often inside warehouses or purpose-built towers, to multiply yield per unit of land. In commercial settings, these techniques are frequently combined to maximize output from tight urban footprints.

The review organizes urban farms into four broad models, each with distinct strengths. Community farms, typically outdoor and collaborative, provide fresh produce while strengthening social bonds, repurposing vacant land and improving residents’ psychological well-being. Common crops include long beans, okra, eggplant, pumpkins, cucumbers and spinach. Their weaknesses are equally clear: affordable land is scarce, land-use regulations can be restrictive, and participation often depends on transportation access and socioeconomic status. Many community farms also rely on volunteer labor and donations, making long-term resource allocation precarious.

Indoor farms sit at the opposite end of the technological spectrum. By controlling temperature, humidity, light and nutrients, they enable year-round production regardless of season and dramatically reduce pest pressure. LED lighting, climate control systems, sensors and automation allow precise optimization of growth conditions, which translates into higher yields and consistent quality. Leafy greens such as lettuce, cabbage, mustard and broccoli, along with high-value herbs, dominate indoor production. The trade-offs are substantial capital costs for infrastructure and lighting, heavy reliance on automation, and a need for specialized technical expertise that many aspiring growers lack.

Rooftop farms occupy a middle ground, exploiting underused building surfaces in dense cities. The review notes that rooftops constitute roughly one quarter of all urban surfaces worldwide, an enormous untapped area for food production. Rooftop operations range from small residential gardens to large commercial installations and can use soil beds, hydroponics or aquaponics. They offer co-benefits beyond food, including improved air quality, better building energy performance and mitigation of the urban heat island effect. But because they remain exposed to the elements, growers must select crops that tolerate fluctuating sunlight, wind, temperature and precipitation, and they often need irrigation systems and wind barriers to protect yields.

Vertical farms push intensification furthest, growing crops in stacked layers inside controlled environments ranging from shipping containers to multi-storey facilities. This approach suits cities where land is scarce and expensive, and it shares the year-round reliability of indoor farming while achieving higher output per square foot. The review highlights reduced water consumption, shorter cultivation cycles and lower pesticide dependence as key advantages. However, the integration of aeroponics, hydroponics and aquaponics within complex vertical systems creates demanding operational requirements. Standard operating procedures and skilled technicians are essential, and the authors identify technical knowledge as one of the most pressing bottlenecks for this model.

The review also surveys the global policy landscape driving adoption. Around 800 million urban residents were already engaged in farming by the mid-1990s according to the United Nations Development Program, and roughly 5,290 companies worldwide now participate in urban farming systems, materials or produce, concentrated in the United States, India and Western Europe. In the United States, the Urban Agriculture Act of 2016 provides grants, technical assistance and research funding, while many cities lease land and support community gardens. India’s National Urban Livelihoods Mission and Smart Cities Mission integrate urban agriculture into development planning. European Union policies promoting local food systems, combined with city-level incentives for rooftop and vertical farms, have fostered a diverse regulatory environment across Western Europe.

Despite this momentum, the review identifies five persistent challenges. Access to affordable land remains the first barrier, which the authors suggest addressing through government partnerships, community land trusts and agriculture-friendly zoning. Water management follows, particularly in water-stressed regions; drip irrigation, rainwater harvesting, greywater recycling and smart monitoring systems can reduce waste. Pests and diseases pose a third problem, intensified by the close proximity of crops and human habitation, and chemical pesticides are often unsuitable in soilless systems where they could contaminate water or harm fish and beneficial microbes. The authors recommend integrated pest management, biological controls, membrane filtration, ozone or UV treatment, quarantine protocols and AI-driven pest detection as safer alternatives.

Financial constraints form a fourth obstacle. The review cites Food and Fertilizer Technology Center data indicating that a warehouse plant factory of at least 2,400 square feet, capable of holding 22,000 seedlings, requires an initial investment of roughly 166,667 US dollars, covering construction, lighting, control systems and equipment. Operational costs for land, utilities and labor add further pressure, and uncertain returns can deter investors. The authors propose microloans, grants, crowdfunding, public-private partnerships, tax incentives and subsidies as mechanisms to lower entry barriers, alongside automation to reduce labor costs.

The fifth challenge is human and institutional. Knowledge gaps in system maintenance, water quality management, nutrient balancing and troubleshooting equipment failures can derail projects, while many growers also lack training in market dynamics, business management and regulatory compliance. The review calls for targeted vocational training, extension services and capacity-building programs. Equally important, it highlights weak networking among universities, governments and industries, which leads to duplicated effort, isolated knowledge and slower innovation. Stronger interdisciplinary partnerships, the authors argue, would create an enabling environment for entrepreneurship and community engagement.

Taken together, the review presents urban farming as a technology with genuine potential but not a plug-and-play solution. Its success depends on matching the right model to local space, resources and skills, and on coordinated action across financing, policy, education and research. As cities continue to grow and supply chains face repeated shocks, the authors conclude that a multifaceted approach, combining technological integration, supportive regulation and stronger stakeholder networks, will determine whether urban farming fulfills its promise as a pillar of sustainable urban food security.

Subject of Research: Urban farming models, their benefits, challenges and strategies for sustainable city food production

Article Title: A review on urban farming: Potential, challenges and opportunities

Article References: Teoh, S. H., Wong, G. R., & Mazumdar, P. (2024). A review on urban farming: Potential, challenges and opportunities. Innovations in Agriculture, 7, 1-11. https://doi.org/10.3897/ia.2024.127816

Image Credits: AI Generated

DOI: 10.3897/ia.2024.127816

Keywords: urban farming, food security, vertical farming, hydroponics, aquaponics, rooftop farms, community gardens, indoor farming, sustainable agriculture, urban agriculture policy, pest management, controlled environment agriculture

Spatial repellents work differently from the topical repellents people apply to their skin. Rather than masking the chemical cues that attract mosquitoes to human hosts or directly irritating their chemoreceptors on contact, spatial repellents disperse volatile compounds into the surrounding air that disrupt the mosquito’s host-seeking behavior from a distance. The World Health Organization currently recognizes only pyrethroid-based products as spatial repellents, including devices like MosquitoShield (containing transfluthrin), OFF! Mosquito lamps and Clip-Ons (containing metofluthrin), and ThermaCELL units (containing allethrin). While these synthetic compounds have demonstrated field efficacy, concerns about adverse effects from prolonged exposure in rodent studies have fueled growing consumer demand for plant-based alternatives.

The research team, led by April D. Lopez and senior author Immo A. Hansen, acquired seven batches of commercially available candles marketed as mosquito repellents, along with two control candles containing no essential oils at all. The products ranged from a simple unscented soy wax emergency candle to formulations containing up to 5 percent total essential oil content, with ingredient lists featuring combinations of citronella, lemongrass, rosemary, peppermint, cedarwood, lavender, eucalyptus, geranium, wintergreen, and corn mint. Not all manufacturers disclosed their exact essential oil concentrations, which the researchers noted as a significant limitation in evaluating product claims.

To test repellency, the team built a custom three-chamber taxis cage assay measuring 90 by 31 by 31 centimeters. Fifty female Aedes aegypti mosquitoes, aged 10 to 14 days and sugar-starved for at least four hours, were placed in the middle chamber. A human volunteer sat one meter upwind from the cage, and a pre-lit candle was positioned 31 centimeters from the cage between the volunteer and the mosquitoes. A small computer fan maintained a constant wind speed of 0.2 meters per second through the apparatus. After the chamber doors were opened remotely, mosquitoes were allowed to roam freely for ten minutes before being counted in each section. The researchers calculated percent attraction as the proportion of mosquitoes that moved toward the volunteer, and percent reduction in attraction by comparing candle trials to paired control trials without a candle.

The results revealed a stark divide among the products. In the absence of any candle, an average of 60 percent of mosquitoes were attracted to the human bait across all trials. Both control candles, one made of soy wax and one of bee’s wax, produced less than 15 percent reduction in attraction. However, three essential oil candles stood out dramatically: EO4 (Murphy’s Naturals Mosquito Repellent Candle), EO5 (Fabulous Frannie Essential Oil Candle Bug Away), and EO7 (Bug Bane Mosquito Repellent Candle) achieved mean reductions in attraction of 56 percent, 40 percent, and 45 percent respectively, significantly higher than both controls. The remaining three essential oil candles, including the well-known OFF! Citronella Candle and the Cutter Outdoor Candle, failed to produce statistically significant repellency despite their label claims.

To understand why some candles worked while others did not, the researchers turned to headspace solid-phase microextraction gas chromatography-mass spectrometry (HS SPME GC-MS), a technique that identifies the volatile organic compounds emitted by heated candle wax. They shaved 100 milligrams of unburned wax from each candle, equilibrated the samples at 40 degrees Celsius, and used an automated SPME Arrow fiber to extract airborne compounds for analysis on a triple quadrupole mass spectrometer. In total, 87 compounds were putatively identified across the seven essential oil candles, with compound identifications assigned by spectral matching to the Adams Essential Oil GC-MS library.

Principal component analysis of the volatile profiles showed clear separation between the high-repellency candles and the rest, indicating that the most effective products were compositionally distinct. The high-repellency candles also had greater mean total peak areas, suggesting that higher concentrations of active ingredients contribute to stronger repellent effects. Notably, EO4 and EO7 listed identical active ingredients on their labels, yet their chemical profiles did not completely overlap, pointing to the influence of essential oil sourcing, concentration, and manufacturing processes on the final volatile composition. The researchers also observed that monoterpenes made up a relatively high proportion of compounds across all candles, with EO4 and EO7 showing particularly high proportions of monoterpene hydrocarbons.

To pinpoint which specific compounds were most responsible for repellency, the team employed a LASSO regression model, a statistical approach well-suited to handling the high-dimensional data produced by GC-MS analysis. Using ten-fold cross-validation to determine the optimal shrinkage penalty, the model identified five compounds as the strongest predictors of spatial repellency. Citronellal, a monoterpene aldehyde and major constituent of citronella and lemongrass oils, received the highest positive coefficient, followed closely by trans-alpha-bergamotene, a sesquiterpene found in cedarwood and lavender oils. Ar-curcumene, another sesquiterpene present in cedarwood, citronella, and peppermint oils, also showed a positive association with repellency. Intriguingly, two compounds, dihydro carveol acetate and piperitol acetate, were negatively associated with repellency, suggesting that certain volatile components may actually undermine the overall protective effect of a candle’s formulation.

The findings carry significant practical implications for consumers and public health practitioners alike. Citronellal, the top predictor identified by the model, has been independently demonstrated as an effective mosquito repellent in multiple prior studies, lending confidence to the LASSO results. Trans-alpha-bergamotene and ar-curcumene have received less individual attention in the repellent literature, though both are constituents of essential oils from plants with known repellent properties. The identification of negatively associated compounds opens an entirely new avenue of investigation, as it suggests that candle formulation is not simply about maximizing concentration of known repellents but also about avoiding or minimizing compounds that may interfere with repellent activity.

The researchers caution that their laboratory assay, while highly sensitive, excludes environmental variables that matter under real-world conditions, including temperature, humidity, mosquito abundance, biting pressure, wind speed and direction, optical cues, and human movement. Earlier field studies of commercial citronella candles have produced contradictory results, with one reporting a 42 percent decrease in bites and another finding no repellency at all against Aedes, Culex, or Anopheles mosquitoes. The team recommends that field testing under diverse environmental conditions be conducted to inform label claims, and that further investigation of individual compounds and defined mixtures is needed to identify all active ingredients responsible for repellency. For now, the message to consumers is clear: not all mosquito repellent candles are created equal, and the efficacy of these products ultimately depends on their specific chemical formulation rather than on the mere presence of essential oils.

Subject of Research: Efficacy of commercially available essential oil-based candles as spatial repellents against Aedes aegypti mosquitoes

Article Title: The efficacy of commercially available essential oil-based candles as spatial repellents against Aedes aegypti mosquitoes – do they actually repel?

Article References: Lopez, A. D., Galvan, C. J., Holguin, F. O., Sroka, C. J., & Hansen, I. A. (2026). The efficacy of commercially available essential oil-based candles as spatial repellents against Aedes aegypti mosquitoes – do they actually repel?. Frontiers in Insect Science, 6, Article 1917728. https://doi.org/10.3389/finsc.2026.1917728

Image Credits: AI Generated

DOI: 10.3389/finsc.2026.1917728

Keywords: Aedes aegypti, essential oils, mosquito repellent, candles, spatial repellent, citronellal, volatile organic compounds, GC-MS, LASSO regression, vector-borne disease, pest management, citronella

The study, led by researchers investigating the green peach aphid Myzus persicae, integrates population theory with experimental greenhouse data to simulate a tri-trophic interaction between wingless (apterous) aphids, winged (alate) aphids, and their predators. The model, built on a system of three ordinary differential equations, extends the classic Lotka–Volterra predator–prey framework by partitioning the aphid population into two distinct morphs, each with its own dynamics. The wingless form focuses on reproduction, while the winged form serves as an escape mechanism, emigrating from danger zones to colonize new territory.

Polyphenism—the ability of a single genotype to produce different phenotypes in response to environmental cues—is a hallmark of aphid biology. Temperature, diet, and predator threats can all trigger developmental switches that yield vastly different body forms from the same genetic blueprint. When wingless aphids detect cues such as alarm pheromones or physical disturbance from predators, they begin allocating their reproductive effort toward winged offspring rather than more wingless young. This transgenerational shift effectively sacrifices immediate reproductive output to safeguard the lineage through dispersal.

While this plasticity is a marvel of evolutionary adaptation, it creates a genuine paradox for biological control. Predators such as ladybugs are essential allies in managing aphid populations, yet their very presence can inadvertently trigger the dispersal of winged aphids to new, unaffected areas of a plantation. A farmer who releases ladybugs to control an aphid outbreak may, in effect, be accelerating the spread of the pest to neighboring fields. Understanding this dynamic is therefore critical for designing effective and sustainable pest management strategies that harness predation without provoking mass escape events.

The new model draws heavily on experimental work by Sentis and colleagues, who demonstrated in the pea aphid Acyrthosiphon pisum that the presence of the harlequin ladybird Harmonia axyridis increases the production of winged offspring, with the strength of this response depending on the host plant. Alfalfa-specialized lineages showed a stronger plastic response than clover-specialized ones, suggesting that host plant specialization shapes herbivore reactions to predators within a tri-trophic context. The researchers behind the new framework adapted this insight to their own system, using Myzus persicae on broccoli as a favorable host—akin to alfalfa—and basil as a repellent host—akin to clover—acknowledging that direct equivalence is not implied due to species and plant differences.

To ground the model in real biological parameters, the team conducted greenhouse experiments at the ESALQ/USP Department of Entomology and Acarology in Piracicaba, Brazil. Using a randomized block design with two treatments—broccoli grown alone and broccoli intercropped with basil—they placed three adult wingless aphids on each broccoli plant and recorded total aphid numbers after 72 hours. The intrinsic growth rate on broccoli was estimated at 0.31, while on basil-intercropped broccoli it dropped to –0.16, indicating that basil’s repellent properties significantly lowered aphid population growth (p = 0.0375). These experimentally derived growth rates were then incorporated into the model’s equations.

The model’s structure captures the essence of aphid–predator interactions with impressive technical detail. Wingless aphids grow logistically but divert reproduction to winged forms under threat, suffer background mortality, and are consumed by predators through a type II functional response that accounts for predator saturation at high prey densities. Winged aphids are produced via polyphenism from wingless mothers, grow logistically, suffer the same mortality and predation pressures, and emigrate at a constant per-capita rate—reflecting their adaptive role in escaping threats. Predators grow by consuming both aphid morphs, with their population dynamics governed by conversion efficiency and natural mortality.

The simulation results are striking. Without predators, wingless aphids predominate and winged forms quickly go extinct due to the lack of an induction cue. With predation, however, the model shows that winged aphid populations become sustained and reach approximately twice the size of the wingless population, highlighting the dramatic polyphenic response triggered by predator presence. The repellent effect of basil in an intercrop scenario drives both aphid populations—and consequently the predator population—to extinction, demonstrating how host plant choice can fundamentally alter the outcome of predator–prey interactions.

Bifurcation analysis, used here primarily as a sensitivity tool, revealed that the predator’s attack rate and conversion efficiency are pivotal parameters in determining the strength of alate induction. Aphid populations decline across a range of conversion efficiencies from 0 to 0.1, followed by a peak where winged forms exceed wingless ones. A similar decline occurs across attack rates from 0 to 0.35, with winged aphids peaking just before the attack rate reaches 0.3. These findings underscore how the intensity of predation pressure directly shapes whether aphid colonies respond with modest or massive dispersal events.

The researchers acknowledge limitations in this preliminary framework, particularly the lack of experimental estimates for several model parameters. The initial values were chosen to enable simulations across a parameter range that demonstrates the model’s potential to describe the mechanisms modulating wing polyphenism. Nevertheless, the growth rate estimates were obtained experimentally under scenarios designed to assess plant-specific effects on aphid repellency, and the attack rate and conversion efficiency parameters were explored through parametric sensitivity analysis. As a first step toward a predictive tool for sustainable pest management, this framework opens a compelling new frontier in understanding how the invisible arms race between predators and prey plays out across agricultural landscapes—one winged aphid at a time.

Subject of Research: Predator-induced wing polyphenism in aphids modeled through a tri-trophic differential equation framework

Article Title: A preliminary framework to model wing polyphenism in aphids threatened by predators

Article References: Carvalho, R. M. R., & Godoy, W. A. C. (2026). A preliminary framework to model wing polyphenism in aphids threatened by predators. BioRisk, 24, 17-26. https://doi.org/10.3897/biorisk.24.171467

Image Credits: AI Generated

DOI: 10.3897/biorisk.24.171467

Keywords: aphids, wing polyphenism, biological control, predator-prey dynamics, Myzus persicae, polyphenism, pest management, host plant effects, ladybugs, mathematical modeling, tri-trophic interactions, greenhouse experiments