The research, published in Environmental Monitoring and Assessment by Eslam Saad, Mohamed Elbawab, and Kareem Tonbol of the Arab Academy for Science, Technology and Maritime Transport, focused on what the authors call the port-urban exposure zone. Between May and October 2022, the team ran two monthly monitoring campaigns at a total of 24 sites: 11 inside the port area itself and 13 in the urban district that wraps around it. Armed with a portable multi-gas monitor and particulate sensors, they measured carbon dioxide, carbon monoxide, nitrogen dioxide, sulfur dioxide, ammonia, volatile organic compounds, and particulate matter in two size fractions, PM2.5 and PM10. From these readings they derived air quality index values for selected pollutants, turning raw concentrations into a measure of potential health relevance.

The study’s most provocative finding concerns carbon monoxide and ammonia. Spatial analysis showed that port activities may be associated with elevated concentrations of both gases, with the berths used for fuel handling and bulk fertilizer recording significantly higher levels than the urban sites outside the port fence. That pattern makes chemical sense: fuel transfer operations release combustion byproducts, while the loading and unloading of nitrogen-rich fertilizers can loft ammonia dust and vapor directly into the ambient air. Unlike diffuse urban pollution, these hot spots are tightly tied to specific cargo operations, which means they are also tightly tied to specific, actionable interventions such as vapor recovery systems, enclosed transfer equipment, and stricter handling protocols.

Sulfur dioxide told the opposite story. Concentrations of SO2 and traffic-related pollutants measured outside the port appeared to contribute to the urban background rather than emanating from the docks, suggesting that the city’s vehicle fleet, not ship operations, drives much of the ambient sulfur and traffic-linked pollution that residents experience. This distinction matters enormously for policy. If regulators assume ships are the culprit for every pollutant wafting through a port city, they may pour resources into shore power and cold ironing while the exhaust pipes that matter most are bolted to buses and trucks idling in gridlocked streets. The Alexandria data show that both sources are real, but they leave different fingerprints in different places.

Short-term exposure observations at one site inside the port revealed the single highest carbon monoxide levels recorded in the study, with air quality index values suggesting potentially unhealthy conditions during intensive cargo and ship-handling activities. These are precisely the moments when dockworkers, tugboat crews, and mooring personnel are on the front line, working in close proximity to running diesel engines, auxiliary ship generators, and heavy equipment. While the screening-level nature of the measurements precludes definitive health risk calculations, the finding flags a clear occupational dimension to port air quality that has often been overshadowed by concerns about residential exposure. It also underscores why continuous, real-time monitoring inside working ports could become a standard safety practice.

Not all the news was grim. Particulate matter levels were relatively low across the study area, both for the fine PM2.5 fraction that penetrates deep into the lungs and the coarser PM10 fraction that irritates airways. This is notable given Alexandria’s reputation for dusty coastal conditions and heavy industry, and it suggests that, at least during the monitoring window, the port’s particle emissions were not dominating the local picture. The researchers are careful to frame their results as screening-level spatial evidence rather than definitive exposure assessments, a caveat that reflects both the short campaign duration and the inherent variability of port operations, ship traffic, and weather.

Weather itself plays a starring role in the study’s interpretation. The team compiled hourly meteorological observations from the Ras El-Tin station for each field session, covering the period from 14:00 through 03:00 the following day. The campaigns were dominated by moderate northerly to northwesterly winds, with campaign mean wind speeds ranging from 3.0 to 7.2 meters per second. These steady sea breezes acted as a natural ventilation system, dispersing pollutants and serving as the atmospheric backdrop against which the measurements unfolded. In a port city, such wind regimes determine whether emissions linger over the docks or drift inland over neighborhoods, and understanding them is essential to any dispersion modeling that follows from this work.

The Alexandria study sits within a growing international literature on port emissions, from assessments of cruise ship impacts in Naples to ship emission externalities in Piraeus, and it aligns with toolkits developed by the International Maritime Organization and the International Association of Ports and Harbors for assessing port emissions. What distinguishes the Egyptian work is its dual-domain design, sampling simultaneously inside and outside the port boundary to separate maritime from urban contributions. That approach directly addresses a chronic problem in port-city air quality research, where monitoring stations placed for general urban purposes rarely capture the sharp spatial gradients that exist within a few hundred meters of a working berth.

The authors argue that their findings underline the local air quality processes controlled by both maritime and urban sources, and they point to the efficacy of implementing a green port strategy. Green port concepts, already adopted to varying degrees in leading Asian and European ports and explored in prior studies of Egyptian ports, encompass measures such as shore-side electricity for berthed vessels, low-sulfur fuel requirements, electrified cargo equipment, and emissions monitoring as a management tool. The new data give such strategies an evidence base tailored to Alexandria, identifying fertilizer and fuel berths as priority targets for source-specific controls while confirming that port-wide particulate levels are not the primary concern.

For the roughly 135 million people worldwide who live in port cities, and for the millions more whose goods pass through them, this study offers both a warning and a template. The warning is that the dockside environment can briefly reach potentially unhealthy air quality conditions precisely during peak operations, and that ammonia and carbon monoxide hot spots cluster at specific, identifiable berths. The template is a pragmatic, instrument-driven monitoring campaign that any mid-sized port authority could replicate: modest equipment, a few dozen sites, coordination with national meteorological services, and an analysis framework built around standard air quality indices. As global shipping volumes grow and climate goals tighten, studies like this one make clear that the path to cleaner port cities begins with knowing exactly where the smoke comes from.

Subject of Research: Air pollutant monitoring at the port-urban interface of Alexandria Port, Egypt

Article Title: Short-term monitoring of air pollutants at the port-urban exposure zone: a case study of Alexandria Port, Egypt

Article References: Saad, E., Elbawab, M., & Tonbol, K. (2026). Short-term monitoring of air pollutants at the port-urban exposure zone: a case study of Alexandria Port, Egypt. Environmental Monitoring and Assessment, 198(9), Article 1012. https://doi.org/10.1007/s10661-026-15828-7

Image Credits: AI Generated

DOI: 10.1007/s10661-026-15828-7

Keywords: air quality, Alexandria Port, port emissions, carbon monoxide, ammonia, particulate matter, PM2.5, air quality index, green ports, Egypt, maritime pollution, urban pollution

The team focused on four metals with well-documented reproductive and developmental toxicity: chromium, lead, cadmium, and nickel. Each pair of samples came from the same household, allowing a direct, person-by-person comparison between external contamination and internal exposure. Quantification was performed with inductively coupled plasma optical emission spectrometry, a technique sensitive enough to resolve the low microgram-per-liter concentrations typical of urinary biomarkers and the microgram-per-gram loadings found in settled household dust. The study was approved by the ethics committee of Isfahan University of Medical Sciences, and informed consent was obtained from all participants.

In the dust samples, the concentration hierarchy was unambiguous: lead dominated, followed by chromium, then nickel, with cadmium trailing far behind. Lead’s mean concentration in indoor dust reached 111.10 micrograms per gram — a figure consistent with the elevated levels reported in residential dust studies from Istanbul, Toronto, Chengdu, and Macao, where lead routinely tops the metal profile of settled indoor particles. The urinary measurements told a parallel story at much lower absolute levels. Lead again ranked first, at a mean of 10.52 micrograms per liter, followed by chromium at 8.30 micrograms per liter and nickel at 4.14 micrograms per liter. Cadmium, notably, did not register as a leading urinary marker in the reported hierarchy, mirroring its low abundance in the dust matrix.

The spatial analysis produced the study’s most consequential contrast. When the researchers grouped participants by urban zone, urinary metal concentrations differed significantly across zones, with p-values below 0.001. The Air Quality Index showed the same strong spatial variability, also at p < 0.001. Yet the dust measurements refused to follow suit: dust metal concentrations showed no significant spatial differences, and dust lead in particular yielded a p-value of 0.514 — statistically indistinguishable across neighborhoods. In other words, the outdoor environment varied dramatically by location, the women’s internal burdens varied in step with it, but the contamination inside their living rooms was essentially uniform across the city.

One household variable did explain a substantial share of the dust variance. Building age correlated strongly with dust lead levels, with a Spearman rank correlation coefficient of 0.788 and a p-value below 0.001. This is a textbook result in indoor contamination science: older buildings carry legacy lead from deteriorating paint, historical plumbing, and decades of accumulated particulate matter that modern structures simply do not contain. The finding aligns with international reviews of indoor dust, which consistently identify building age, paint condition, and proximity to traffic as the dominant predictors of lead loading in household dust, regardless of the surrounding neighborhood’s ambient air quality.

But the correlation between dust and urine — the link the study was designed to test — failed to materialize. Dust and urinary metal profiles were not significantly correlated, a null result the authors interpret as evidence that environmental contamination does not translate linearly into maternal internal exposure. This decoupling has a mechanistic basis that the toxicology literature has long recognized. Metal bioaccumulation is governed by biodynamics: uptake rates across the gut and lungs, sequestration in bone and soft tissue, and excretion kinetics all vary between individuals and between metals. Lead absorbed decades earlier can be remobilized from maternal bone during pregnancy, meaning a pregnant woman’s urinary lead may partly reflect her own lifetime exposure history rather than the dust currently under her sofa.

Diet, drinking water, occupational contact, and behavioral factors — hand-to-mouth activity, cleaning frequency, ventilation habits — all contribute additional exposure pathways that a single dust sample cannot capture. Urinary biomarkers, meanwhile, integrate exposure across all routes over recent days to weeks, and are further modulated by hydration status, a confounder that pregnancy exposure researchers have flagged as requiring explicit adjustment. The Isfahan results are a quantitative demonstration of why multi-matrix assessment — measuring both environmental reservoirs and biological indicators — is now considered essential for characterizing exposure in vulnerable populations, rather than relying on either measurement alone.

The stakes are not abstract. A growing body of systematic reviews links maternal exposure to lead, cadmium, mercury, and chromium with adverse pregnancy outcomes, including preterm birth and low birth weight, and with neurodevelopmental effects in children. Geographic clustering of elevated blood metal levels among pregnant women has been documented in the United States, and biomonitoring studies in Puerto Rico and Canada have mapped the predictors of urinary and blood metal burdens during pregnancy. Isfahan adds a distinctive context to this literature: a semi-arid metropolis with documented heavy metal contamination in its particulate matter, airborne dust episodes, and a housing stock old enough to carry substantial legacy lead indoors.

Methodologically, the study’s strength lies in its paired design and its explicit spatial framing. By collecting urine and dust from the same households and stratifying by urban zone alongside air quality data, the researchers could test — rather than assume — the coupling between environment and body burden. The answer was a clear negative for the dust-to-urine pathway in this cohort, and a clear positive for the influence of ambient air quality and neighborhood on internal exposure. The practical implication is that remediating indoor dust alone, while worthwhile given the strong lead concentrations found and the building-age effect, would not by itself equalize maternal metal burdens across the city. Exposure reduction strategies would need to address the ambient and dietary routes that the urinary data reflect.

The authors conclude that integrated multi-matrix assessments are essential to accurately characterize pregnant women’s environmental risks, and their data make the case concretely: 80 paired samples, four metals, and two measurement matrices that told two different stories about the same homes. For environmental health researchers, the Isfahan findings are a caution against using household dust measurements as a proxy for internal exposure in biomonitoring studies. For public health officials in rapidly urbanizing cities with aging housing stocks, they are a reminder that the contaminants that matter most to a pregnancy may be arriving by routes no vacuum cleaner can reach — and that the neighborhood a woman lives in may shape her body burden more than the dust in her own home does.

Subject of Research: Heavy metal exposure in pregnant women measured via paired urinary biomarkers and indoor dust samples in Isfahan, Iran

Article Title: Spatial variability of heavy metals in pregnant women: linking urinary biomarkers, indoor dust contamination, and air quality in Isfahan, Iran

Article References: Moradnia, M., Movahedian Attar, H., Hajizadeh, Y., Darvishmotevalli, M., & Chavoshani, A. (2026). Spatial variability of heavy metals in pregnant women: linking urinary biomarkers, indoor dust contamination, and air quality in Isfahan, Iran. Environmental Geochemistry and Health, 48(16), Article 646. https://doi.org/10.1007/s10653-026-03508-x

Image Credits: AI Generated

DOI: 10.1007/s10653-026-03508-x

Keywords: heavy metals, pregnant women, indoor dust, urinary biomarkers, lead exposure, air quality index, Isfahan, environmental health, ICP-OES, spatial variability, cadmium, chromium

Maiduguri presents an unusually demanding setting for this kind of research. The city has absorbed rapid urban growth while coping with the social and economic pressures of a prolonged insurgency, and its densely populated neighborhoods face mounting environmental stress as a result. Evidence linking residents’ environmental awareness to their socio-demographic characteristics has been scarce for urban northern Nigeria, where most environmental research has focused on the oil-polluted Niger Delta or on broad national assessments. By focusing on a single metropolitan council in the country’s northeast, the study fills a geographic and evidentiary gap that policymakers in the region have long had to navigate without local data.

The methodological core of the study is a quantitative cross-sectional survey of 200 respondents, combined with harmonised secondary environmental data covering the years 2015 to 2024. The researchers measured environmental awareness using a twelve-item instrument and then examined how awareness scores varied across socio-demographic categories. Categorical comparisons and regression models were used to test the associations, with a binary logistic regression retained only as a sensitivity analysis to check the robustness of the findings. On the environmental side, the reconstructed time series for deforestation, air quality index, temperature change and biodiversity were summarized descriptively, and detrended Pearson correlations were computed to explore how the different indicators moved together over the decade.

The results on the social side are strikingly uneven. Education and occupation emerged as the characteristics associated with environmental awareness, while awareness also differed across age groups. Gender and location, by contrast, were not statistically significant in the reported comparisons. This pattern suggests that in Maiduguri, environmental knowledge is not distributed evenly across the population but clusters around people whose schooling and working lives give them greater exposure to environmental information. The finding aligns with a wider literature on environmental literacy, which has repeatedly linked formal education and professional context to environmental knowledge, values and reported behavior, but it now has a concrete empirical anchor in one of Nigeria’s most environmentally pressured cities.

The environmental indicators tell their own directional story. The reconstructed series showed changes in deforestation, the air quality index, temperature change and biodiversity over the ten-year window. Two correlations stood out in the detrended analysis: a preliminary negative co-movement between deforestation and biodiversity, with a correlation coefficient of minus 0.90, and a positive co-movement between air quality and temperature, with a coefficient of 0.74. In plain terms, the data suggest that years in which forest loss accelerated tended to be years in which biodiversity declined, and that air quality and temperature tended to worsen or improve together. These are exactly the kinds of coupled dynamics that climate and land-use researchers have warned about at the global scale, now visible in a single city’s decadal record.

The authors are careful, and rightly so, about what these numbers can and cannot support. Because only ten annual observations were available for each environmental series, the correlations are descriptive and exploratory rather than causal or confirmatory. Short time series are notoriously vulnerable to spurious relationships, a problem long recognized in the statistical literature on regression with limited data, and detrending helps but does not eliminate the risk. The deforestation-biodiversity correlation of minus 0.90 is strong enough to be suggestive, but with ten data points it cannot establish that forest loss drove biodiversity decline, only that the two moved in opposite directions over the same period. Similarly, the air quality and temperature co-movement of 0.74 points to a shared trajectory that deserves longer monitoring rather than a settled conclusion.

There is also an important limitation on the survey side. Because the original item-level responses for the twelve-item awareness instrument were not retained, the researchers could not assess full-sample internal consistency, including Cronbach’s alpha, the standard measure of how well a set of survey items hangs together as a single scale. Environmental awareness is therefore treated as a researcher-developed composite, and all awareness-related findings are interpreted with this reliability limitation in mind. This kind of transparency is unusual and valuable: rather than presenting a clean but fragile result, the study flags exactly where the evidentiary floor is soft, inviting future surveys to preserve item-level data and replicate the socio-demographic patterns with a fully validated instrument.

The ethical and procedural scaffolding of the research is documented in detail. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki, received formal ethical clearance from the Institutional Ethics Committee and the Research and Development office of Integral University in December 2025 before the 2025 field survey, and obtained informed consent from all participants. All gathered information was anonymized for publication, and the authors declare no competing interests. The work received no external funding and was supported institutionally by Integral University, Lucknow, which assigned the manuscript a communication number for record-keeping after submission.

What the findings mean for policy is framed deliberately as context rather than proof. The authors present policy, governance and Sustainable Development Goal implications as contextual recommendations, not as statistically tested effects, which is an honest reading of a study that combines a 200-person survey with ten-year environmental series. Still, the practical signal is clear: if education and occupation are the strongest correlates of environmental awareness in Maiduguri, then awareness campaigns that ignore schooling and livelihood channels are likely to miss their mark, and targeted activities aimed at less-educated and differently employed residents could raise the baseline of environmental knowledge where it is lowest. The study’s support for continued environmental monitoring is equally pointed, since the decadal trends in deforestation, air quality, temperature and biodiversity can only be confirmed, or overturned, by longer and better-resourced data collection.

Maiduguri’s decade of environmental change, reconstructed here from fragmented secondary data, offers a rare window into how a rapidly growing city in an insecure region absorbs ecological pressure. The study does not claim that awareness alone will reverse deforestation or cool the air, and it explicitly avoids causal language where the statistics cannot bear it. What it does deliver is a paired diagnosis: a social map of who in the city knows about environmental problems, and a preliminary environmental record of what those problems look like on the ground. For a city rebuilding amid conflict, that combination of evidence, with its limitations stated plainly, is a starting point for the targeted awareness work and sustained monitoring that the Sustainable Development Goals demand, and a reminder that in data-scarce settings, careful and honest description is itself a scientific contribution.

Subject of Research: Socio-demographic determinants of environmental awareness and decadal environmental trends in Maiduguri, Nigeria

Article Title: Socio-demographic determinants of environmental awareness and decadal environmental trends in Maiduguri Metropolitan Council, Borno State, Nigeria

Article References: Hammajam, B. P., & Khan, A. S. (2026). Socio-demographic determinants of environmental awareness and decadal environmental trends in Maiduguri Metropolitan Council, Borno State, Nigeria. SN Social Sciences, 6(9), Article 413. https://doi.org/10.1007/s43545-026-01688-y

Image Credits: AI Generated

DOI: 10.1007/s43545-026-01688-y

Keywords: environmental awareness, Maiduguri, Nigeria, deforestation, biodiversity, air quality index, temperature change, socio-demographics, cross-sectional survey, Sustainable Development Goals, urbanization, environmental monitoring

The study focused on eight monitoring stations operated by India’s Central Pollution Control Board in Ahmedabad, one of the world’s most air-pollution-stressed cities. The team set out to predict the AQI twenty-four hours ahead at every station simultaneously, a task that naturally invites spatio-temporal models designed to exploit correlations between neighboring sensors. They compared two graph-based deep learning architectures, a Spatio-Temporal Graph Convolutional Network known as ST-GCN and a hybrid GCN-LSTM model, against four per-station approaches: Random Forest, XGBoost, Long Short-Term Memory networks, and Gated Recurrent Units. Every model faced the identical test set, evaluated across 31,831 usable station-hours, ensuring that differences in performance could be attributed to the models themselves rather than to favorable data splits or evaluation tricks.

One of the study’s most important contributions lies in how the target variable was constructed. Rather than computing a generic pollution score, the authors followed the CPCB’s official AQI methodology, which assigns pollutant-specific averaging periods, sub-index breakpoints, and strict data-completeness standards before a station’s daily index can be calculated. This matters because the AQI is the number that public health advisories are actually built on, and a model that excels at predicting raw PM2.5 concentrations may still fail when those concentrations are translated into the categorical scale that triggers warnings. By anchoring the prediction task to the regulatory definition, the researchers aligned their benchmark with the question that genuinely matters to city officials and residents: what will the official index be tomorrow?

The preprocessing pipeline also received unusual care. All data-dependent operations, including gap handling, outlier capping, and normalization, were fitted exclusively on the training segment of the data, eliminating a subtle but pervasive form of information leakage known as data snooping. In many published air quality studies, preprocessing statistics computed over the full dataset quietly leak information from the future into the training process, inflating reported accuracy. By restricting every transformation to training data and applying identical evaluation masks to all models on a shared test index, the authors built a benchmark whose results can be trusted, even when those results are unflattering to the fanciest architectures.

So which model won? The GCN-LSTM hybrid achieved the best pooled continuous performance, with a root mean square error of 35.84 and a coefficient of determination of 0.463 across the full test set. But here is the uncomfortable part: the persistence baseline, which requires no training at all and simply carries today’s value forward, came second, with an RMSE of 37.50 and an R-squared of 0.412. The ST-GCN followed closely at 36.18 and 0.452. The margin separating the most sophisticated graph neural network from a naive forecast of no change was slim, and bootstrap analysis, which resamples the test data to quantify uncertainty, indicated that the gains over persistence were rather slight. In other words, much of the apparent advantage of deep learning may fall within the range of statistical noise.

The authors did not stop at a single headline metric. They probed the robustness of their conclusions from several angles. Rolling-origin evaluation, in which models are repeatedly retrained and tested on successive time windows, assessed how performance holds up as the forecast period moves forward in time. Multi-seed experiments, in which each neural network is trained from several different random initializations, tested whether reported results depend on a lucky starting point. Matched-feature studies varied the input configuration and revealed that model rankings are sensitive to which predictors are available, a finding that complicates comparisons across studies that use different input sets. Perhaps most striking was the graph-structure ablation: replacing the distance-based connection scheme between stations with a trivial identity adjacency, in which each station is effectively isolated, produced no discernible difference in performance.

That last result deserves particular attention because it strikes at the core assumption of spatio-temporal graph models. These architectures are built on the premise that explicitly encoding spatial relationships, such as the physical distance between monitoring stations, allows information to flow between sensors and improves prediction. If an identity adjacency performs just as well on this network of eight Ahmedabad stations, it suggests that either the stations are not close enough for inter-station dynamics to matter at a twenty-four-hour horizon, or that the temporal signal at each station already contains whatever spatial information is useful. For a city-scale deployment, this raises a practical question about whether the added complexity of graph machinery is justified at all.

The picture grew darker when the researchers examined performance by AQI category. Rare high-pollution classes, precisely the episodes that matter most for public health, were predicted worse than common ones, and no model managed to correctly forecast the Severe category even once. This failure mode is well known in imbalanced classification but is often obscured by pooled metrics, which are dominated by the frequent moderate categories. A model can post a respectable RMSE while being systematically blind to the pollution episodes that send people to emergency rooms. The authors’ class-wise evaluation makes this blind spot explicit, and it is a practice that future air quality forecasting studies would do well to adopt as standard.

Why does this matter beyond Ahmedabad? Air pollution is linked to cardiovascular mortality, respiratory disease, and a wide range of economic and social costs, and roughly half the world’s population is exposed to increasing pollution levels according to global assessments. Cities across South Asia and beyond are investing in machine learning pipelines for next-day air quality advisories, often guided by literature that reports impressive accuracy figures without persistence baselines, leakage-free preprocessing, or uncertainty quantification. The Ahmedabad benchmark demonstrates that when those methodological safeguards are in place, the headline numbers shrink dramatically. An R-squared of 0.463 for the best model means that less than half the variance in next-day AQI is explained, leaving substantial room for meteorology, emissions variability, and atmospheric chemistry that the models do not capture.

The study’s deeper lesson is about scientific hygiene rather than any particular architecture. Solid baselines, matched comparisons, leakage-free pipelines, bootstrap uncertainty, rolling-origin validation, and multi-seed stability checks are not bureaucratic overhead; they are the difference between a result that generalizes and one that evaporates on contact with reality. The authors’ finding that GCN-LSTM edges out persistence, but only barely, and that graph structure itself appears dispensable, should prompt the air quality forecasting community to re-examine published claims made without such controls. For cities deciding where to spend limited monitoring and computing budgets, the message is pragmatic: before deploying a graph neural network, check whether simply assuming tomorrow will look like today, augmented with modest statistical modeling, gets you most of the way there. In Ahmedabad, at least, the answer appears to be yes.

Subject of Research: Comparative benchmarking of machine learning and spatio-temporal graph neural network models for 24-hour-ahead air quality index prediction at multi-station monitoring networks

Article Title: Comparative analysis of machine learning and spatio-temporal graph models for multi-station air quality index prediction

Article References: Yagnik, H., Gupta, R., & Shah, K. (2026). Comparative analysis of machine learning and spatio-temporal graph models for multi-station air quality index prediction. Theoretical and Applied Climatology, 157(11), Article 706. https://doi.org/10.1007/s00704-026-06623-y

Image Credits: AI Generated

DOI: 10.1007/s00704-026-06623-y

Keywords: air quality index, AQI forecasting, graph neural networks, GCN-LSTM, ST-GCN, persistence baseline, machine learning, Ahmedabad, CPCB, data leakage, bootstrap analysis, spatio-temporal modeling

The significance of the result lies less in the headline accuracy figure than in what it implies for the millions of people living in rapidly urbanizing regions where air quality monitoring infrastructure is thin. The Air Quality Index, or AQI, condenses the concentrations of several regulated pollutants into a single number that public health authorities use to issue warnings, close schools, and restrict traffic. Producing that number reliably normally requires continuous measurements of particulate matter, sulfur dioxide, carbon monoxide, and nitrogen dioxide from calibrated instruments. When a district has only sparse or intermittent monitoring, the index becomes unreliable exactly when residents need it most, and the health consequences of missed pollution episodes can be severe.

Akiner and Ghasri attacked the problem with a two-stage design that reflects a key insight about atmospheric data: pollution does not behave the same way under all weather conditions. In the first stage, they applied K-Means clustering, an unsupervised algorithm that groups observations by similarity without being told what the groups should be, to a combination of meteorological variables and pollutant concentrations. The algorithm partitioned the ten-year record into distinct atmospheric regimes, essentially discovering recurring weather-and-pollution patterns on its own. Each cluster corresponds to a characteristic set of conditions, such as stagnant, cold episodes in which pollutants accumulate near the surface, or windy, well-mixed periods in which concentrations disperse rapidly.

This clustering step matters because it gives the downstream model a structured picture of the atmosphere instead of forcing it to learn every relationship from raw, undifferentiated data. In the second stage, the researchers employed XGBoost, a gradient-boosted decision tree algorithm that has become a workhorse of tabular machine learning. XGBoost builds an ensemble of decision trees sequentially, with each new tree trained to correct the residual errors of the ensemble so far, and it includes regularization terms that discourage overfitting. The model was tasked with estimating AQI values from meteorological parameters together with the key pollutants PM10, PM2.5, SO2, CO, and NO2, conditioned on the regime structure identified by the clustering stage.

The crucial design choice is the framework’s reliance on widely available meteorological data to compensate for scarce pollutant measurements. Weather variables such as temperature, wind speed, humidity, and pressure are collected routinely by national meteorological services almost everywhere on Earth, and Türkiye’s General Directorate of Meteorology, together with the Ministry of Environment, Urbanization, and Climate Change, provided the database underpinning the study. Because these inputs are cheap and abundant, the framework reduces dependence on dense sensor infrastructure. In effect, the model learns how the local climate converts emissions into pollution episodes, and can then reconstruct or forecast air quality even where direct pollutant observations are limited.

The reported performance is remarkable for a real-world urban dataset. An overall accuracy of approximately 98 percent means the model reproduces the observed AQI with enough fidelity to support timely health advisories, giving authorities lead time to warn vulnerable populations, adjust traffic, or deploy inspection resources. Beyond raw accuracy, the authors used feature-importance analysis to identify which meteorological drivers dominate AQI variability in Kağıthane. This kind of interpretability is increasingly demanded of machine learning in environmental applications, because a black-box prediction is of limited use to a planner who needs to know whether ventilation, temperature inversions, or precipitation scavenging is controlling pollution on a given day.

Kağıthane itself is a fitting laboratory for the method. The district sits in a valley that channels both traffic and terrain-driven airflow, and previous research has documented elevated particulate levels in the Kağıthane Valley and the wider Istanbul metropolitan area. Istanbul combines intense vehicular emissions, a population exceeding fifteen million, complex topography, and a coastal climate whose winds and humidity strongly modulate pollutant dispersion. A model that performs well in such a demanding setting is more likely to transfer to other challenging environments than one tuned on a flat, well-monitored city.

The authors emphasize that the framework is modular and readily transferable across different geographic and temporal contexts. Both K-Means and XGBoost are standard, well-documented algorithms, and the input data are commonly available, which means a municipality does not need supercomputing resources or proprietary technology to adopt the approach. The pipeline can be retrained on local data, the clustering stage will rediscover the atmospheric regimes specific to a new location, and the boosted trees will re-weight the meteorological drivers accordingly. That scalability is the study’s central claim: a practical answer for regions facing environmental monitoring constraints, from fast-growing cities in the Global South to districts of wealthy cities where sensor coverage is uneven.

The work also arrives amid a broader shift in how air pollution is measured and modeled. Researchers worldwide are experimenting with low-cost sensor networks, satellite retrievals of aerosol optical depth, and deep learning architectures for spatiotemporal forecasting, and systematic reviews have catalogued both the promise and the pitfalls of machine learning in air pollution epidemiology. Against that backdrop, the Istanbul study makes a deliberately pragmatic argument: rather than waiting for perfect instrumentation, cities can extract far more value from the data they already possess by combining unsupervised structure discovery with powerful supervised regression. The approach complements, rather than replaces, physical dispersion models, which remain essential for source attribution and regulatory analysis.

For public health, the implications are immediate. Air pollution is linked to cardiovascular and respiratory disease, adverse birth outcomes, and growing evidence of effects on mental health and children’s cognitive development, and the burden falls disproportionately on neighborhoods with the least monitoring. A framework that turns routine weather data into near-accurate AQI forecasts can level that playing field, enabling evidence-based environmental policy in places that have historically flown blind. As urbanization accelerates and climate change alters the meteorological conditions that govern pollutant dispersion, tools like this hybrid model may become as standard a piece of civic infrastructure as the weather forecast itself, quietly translating the atmosphere’s behavior into actionable warnings for the people who live beneath it.

Subject of Research: Hybrid machine learning for air quality index forecasting in data-scarce urban environments

Article Title: Overcoming data shortages through hybrid machine learning in air quality forecasting: a case study of Kağıthane, Istanbul

Article References: Akiner, M. E., & Ghasri, M. (2026). Overcoming data shortages through hybrid machine learning in air quality forecasting: a case study of Kağıthane, Istanbul. Theoretical and Applied Climatology, 157(10), Article 609. https://doi.org/10.1007/s00704-026-06545-9

Image Credits: AI Generated

DOI: 10.1007/s00704-026-06545-9

Keywords: air quality, machine learning, XGBoost, K-Means clustering, Air Quality Index, Istanbul, Kağıthane, meteorology, urban pollution, data scarcity, public health, environmental forecasting

The research team, led by Priyal Singhal and Upendra Singh of Manipal University Jaipur together with colleagues at VIT Bhopal University, trained and validated their framework on 435,742 daily air quality records collected by India’s Central Pollution Control Board between 1990 and 2015. The dataset spans monitoring stations across major states including Maharashtra, Uttar Pradesh, Andhra Pradesh, Punjab, and Rajasthan, recording concentrations of sulphur dioxide, nitrogen dioxide, respirable suspended particulate matter, and suspended particulate matter. Because the historical records predate routine measurement of PM2.5, carbon monoxide, ozone, ammonia, and lead at most stations, the researchers computed a CPCB-derived AQI proxy based on the three pollutants reliably available throughout the period, with particulate matter serving as the historical Indian monitoring proxy for PM10.

The first stage of the pipeline addresses the missing data problem, which is far from random in this dataset. Gaps correlate with station age and with the monsoon months of June through September, when outages are most frequent. Rather than deploying computationally expensive generative adversarial networks such as TimeGAN, the team developed a lightweight noise-augmented linear interpolation scheme. The method fills gaps by interpolating between observed neighbours, then injects calibrated Gaussian noise with a standard deviation set to two percent of each column’s observed dispersion, preventing the artificially smooth segments that plague simple interpolation. A final clipping step enforces physical non-negativity. The approach comes with a formal variance-preservation guarantee: the ratio of imputed to observed variance stayed within 0.04 percent of unity across pollutants, and sensitivity analysis confirmed that performance is stable for noise scales between 0.01 and 0.05. Compared against mean-fill, forward-fill, KNN, MICE, Kalman smoothing, and MissForest under an identical downstream pipeline, the scheme delivered the lowest error at fifteen to twenty times lower computational cost than the heaviest alternatives.

At the heart of the framework sits a hybrid ensemble combining two fundamentally different modelling philosophies. A long short-term memory network with a fourteen-day input window learns sequential dependencies, capturing fortnight-scale dynamics such as multi-day pollution accumulation and monsoon wash-out episodes that discrete lag features cannot fully encode. In parallel, an XGBoost gradient-boosted tree model exploits nonlinear feature interactions across twenty-six engineered predictors, including lagged values, rolling means, calendar variables, and inter-pollutant ratios. The two sub-models are fused through a dynamic R-squared-weighted averaging scheme derived from a variance-minimising convex combination argument. The measured residual correlation between the two learners was just 0.31, indicating genuine complementarity, and covariance-optimal weights estimated independently fell within 0.04 of the R-squared-derived weights, confirming the fusion rule operates near the theoretical optimum.

The most distinctive contribution, however, is the closed-loop use of explainability. Rather than computing SHAP values merely for human inspection, as previous LSTM-SHAP frameworks have done, the team used SHAP attributions from the XGBoost sub-model as an active feature-selection signal. SHAP, grounded in cooperative game theory, assigns each feature an exact marginal contribution to every prediction, and for tree models the TreeExplainer algorithm computes these values in polynomial time. The researchers independently verified feature importance for the LSTM using gradient sensitivity analysis, finding strong agreement between the two methods: Spearman rank correlation of 0.87, with the top three features identical across both approaches. The top ten features by SHAP importance were then used to retrain both sub-models, cutting input dimensionality by 62 percent.

The results are striking. The final XAI-enhanced ensemble achieved a root mean squared error of 0.3817, a mean absolute error of 0.2773, and an R-squared of 0.9994 on the held-out test set, representing a 54.7 percent RMSE reduction and a 58.7 percent MAE reduction over the standalone LSTM baseline. Translated to original AQI units, the error corresponds to roughly 1.91 AQI points, well below the 50-point width of the narrowest CPCB AQI category and at or below the noise floor of the underlying measurement instruments, whose typical uncertainty propagates to roughly three to four AQI units. SHAP-guided selection alone yielded an additional 38 percent RMSE improvement over the standard ensemble, and the across-trial variance of test error fell by 56 percent, an effect the authors interpret through bias-variance theory: removing low-signal features suppresses estimator variance without materially increasing bias.

The team subjected these gains to unusually rigorous statistical scrutiny. All models were retrained across thirty independent random seeds, with paired t-tests and Wilcoxon signed-rank tests applied under Bonferroni correction. Every improvement remained significant by several orders of magnitude, with Cohen’s d effect sizes exceeding 2.8, far above the conventional threshold of 0.8 for a large effect. A five-fold rolling-origin time-series cross-validation, in which each fold advances the train-test boundary forward in time, reproduced the model ranking with RMSE of 0.3928 versus 0.6382 for the standard ensemble, and bootstrap confidence intervals for the two models did not overlap. The framework also outperformed a Temporal Fusion Transformer, which the authors attribute to the limited training scale under-utilising the transformer’s capacity while the ensemble’s explicit feature engineering exploits the available signal more efficiently.

The authors are candid about the caveats. The near-deterministic correlation of 0.99 between RSPM and the AQI target means that part of the headline R-squared reflects a definitional relationship, though control experiments demonstrate genuine learning: a deterministic baseline computing AQI directly from RSPM achieves substantially higher error, and removing RSPM from the feature set entirely still yields an R-squared of 0.9851. Winter pollution peaks are systematically under-predicted by four to eight AQI units, a bias traced to the absence of meteorological covariates such as temperature, wind speed, and boundary layer height in the historical records, with stubble-burning episodes in the Indo-Gangetic Plain contributing an episodic component that a seasonal correction only partially recovers. The test set also contains few samples in the Poor, Very Poor, and Severe AQI categories, precisely those most relevant for public health intervention, and the framework’s calibration to Indian RSPM-dominated profiles would require adaptation in PM2.5-dominated or ozone-dominated regions.

On the deployment front, the numbers are compelling. Per-sample inference takes just 7.9 milliseconds on a commodity CPU, roughly five orders of magnitude faster than the CPCB’s reporting cadence, and the reduced ten-feature input lowers memory and input-output demands. The explainability stage adds about twenty-one minutes at training time only. Because the pipeline is agnostic to pollutant count, porting it to post-2015 CPCB data with the full eight-pollutant suite requires only substituting the sub-index inputs and re-running SHAP selection. The study’s central lesson may prove its most enduring contribution: interpretability and accuracy are not competing objectives in environmental forecasting but complementary ones, and the feedback loop that turns explanations into better models offers a template that extends well beyond air quality, from rainfall prediction to any domain where high-stakes decisions demand both precision and justification.

Subject of Research: Explainable hybrid deep learning for air quality index prediction using Indian CPCB monitoring data

Article Title: Explainable hybrid deep learning framework for air quality index prediction in Indian cities

Article References: Singhal, P., Saurabh, P., Sharma, M., Badge, J., & Singh, U. (2026). Explainable hybrid deep learning framework for air quality index prediction in Indian cities. Discover Artificial Intelligence, 6(1), Article 1395. https://doi.org/10.1007/s44163-026-02131-0

Image Credits: AI Generated

DOI: 10.1007/s44163-026-02131-0

Keywords: air quality index, explainable AI, SHAP, LSTM, XGBoost, ensemble learning, deep learning, missing data imputation, CPCB India, air pollution, environmental forecasting, machine learning

The team, led by Aniket Chavan together with Dilip H. Lataye of the Department of Civil Engineering, analyzed data collected throughout 2023 at a Continuous Ambient Air Quality Monitoring Station in Nagpur, a rapidly expanding Tier-2 city in central India. Rather than relying on a single metric, the researchers examined eight major pollutants simultaneously: particulate matter of two size fractions, PM10 and PM2.5, along with sulfur dioxide, carbon monoxide, ozone, nitrogen dioxide, ammonia, and benzene. These were paired with four meteorological parameters: temperature, relative humidity, wind speed, and rainfall. The analysis was carried out in the R programming environment using the Openair package, a specialized toolkit for atmospheric data visualization and statistics, supplemented by principal component analysis, a dimension-reduction technique that reveals hidden structure in large multivariate datasets.

The headline finding is stark. In February, the city’s Air Quality Index climbed to 492, deep within the Severe category, a level at which even healthy people face serious health risks and outdoor activity becomes hazardous. By contrast, during the monsoon months the AQI fell below 100, crossing into the satisfactory range. That dramatic seasonal swing is not primarily a story about emissions rising and falling, the researchers argue, but about the atmosphere’s capacity to dilute and disperse whatever is emitted. In winter, cooler air, shallow boundary layers, and stagnant winds trap pollutants close to the ground; in the monsoon, rain scrubs the air and vigorous mixing lifts contaminants away.

To quantify that dispersal capacity, the team calculated the ventilation coefficient, a product of wind speed and the height of the atmospheric mixing layer that expresses how much clean air is available per unit time to dilute urban emissions. In Nagpur, this coefficient ranged from a low of 668 square meters per second in January to a high of 1385 square meters per second in May. The implications for policy are striking: because winter air can carry away only about half as much pollution as summer air, emission reductions in winter need to be roughly 3.7 times more aggressive than summer measures to achieve the same ambient concentration targets. In other words, a uniform, year-round abatement strategy will systematically underperform precisely when public health is most at risk.

Correlation analysis of the pollutant suite revealed which contaminants travel together, offering clues about their origins. PM10 and PM2.5 were almost inseparable, with a correlation coefficient of 0.955, indicating that the two size fractions largely share the same sources and atmospheric behavior. Carbon monoxide and benzene, meanwhile, showed a correlation of 0.889. Both are classic signatures of incomplete combustion, from vehicle exhaust, biomass burning, and other fuel-consuming processes, so their tight coupling points to common combustion sources driving a substantial share of the city’s pollution burden. This kind of statistical fingerprinting matters because it tells regulators where interventions will pay the greatest dividends: targeting combustion sources yields simultaneous reductions in several harmful pollutants at once.

Principal component analysis with varimax rotation condensed the twelve variables into two dominant components that together explained 73.1 percent of the variance in the dataset. The first component grouped the particulate pollutants with nitrogen dioxide, a cluster the authors interpret as anthropogenic combustion, the combined exhaust of traffic, industry, and burning fuels. The second component paired ozone with temperature, capturing the photochemical processes by which sunlight and heat drive the formation of secondary pollutants in the atmosphere. The separation is chemically meaningful: primary pollutants emitted directly from tailpipes and stacks behave differently from ozone, which is not emitted at all but manufactured in the air through sunlight-driven reactions involving nitrogen oxides and volatile organic compounds.

One of the study’s more surprising results is what it did not find. Many cities show a weekend effect, with pollutant concentrations dipping on Saturdays and Sundays as commuting traffic declines. In Nagpur, no statistically significant weekend effect emerged for any pollutant, with all p-values exceeding 0.1. The authors interpret this as evidence that pollution sources operate continuously, without a strong weekly rhythm dominated by commuter traffic. That pattern suggests contributions from sources that do not rest on weekends, such as industrial operations, residential fuel burning, and goods transport, and it cautions against policies that assume traffic-focused weekend restrictions will meaningfully clean the air.

The health stakes are made concrete by the study’s exceedance analysis. During winter months, PM2.5 and PM10 concentrations exceeded the standards set by India’s Central Pollution Control Board on 27 to 29 days, meaning that for most of the season the city’s most dangerous pollutants were out of compliance nearly every day. Fine particulate matter is especially concerning because particles smaller than 2.5 micrometers penetrate deep into the lungs and enter the bloodstream, contributing to cardiovascular and respiratory disease. Benzene, also tracked in the study, is a known carcinogen, adding a toxic dimension to the combustion-driven pollution mix identified by the correlation and component analyses.

From these findings the researchers distill a set of policy recommendations aimed specifically at cities like Nagpur rather than at Delhi or Mumbai. Emission reduction targets should vary by season, with the heaviest cuts demanded in winter when the ventilation coefficient collapses. Co-benefit strategies that target combustion sources should be prioritized, since a single intervention there reduces particulates, carbon monoxide, benzene, and nitrogen dioxide simultaneously. And monitoring and enforcement resources should be reallocated toward the winter months, when dispersion is poorest and every ton of emissions does the most damage. The study’s methodological approach, combining Openair’s visualization and statistical capabilities with principal component analysis on continuous monitoring data, offers a replicable template that other data-rich but understudied cities can adopt.

Beyond its immediate findings, the work highlights a growing gap in global air quality science. Most research attention and most pollution control infrastructure have concentrated on megacities, yet hundreds of Tier-2 cities across Asia and Africa are industrializing and motorizing rapidly, often with weaker monitoring networks and less tailored policy. Nagpur’s baseline shows that the dynamics in such cities, from the magnitude of seasonal swings to the absence of a weekend effect, can differ in ways that matter for regulation. As continuous monitoring stations proliferate and open-source analytical tools lower the barrier to sophisticated analysis, the authors’ demonstration that a single year of CAAQMS data can yield actionable, seasonally resolved insight may prove as influential as the specific numbers they report for Nagpur itself.

Subject of Research: Statistical analysis of air quality dynamics and meteorological influences in Nagpur using the Openair R package and principal component analysis

Article Title: Application of Openair package in R programming and PCA: unveiling air quality dynamics and meteorological influences

Article References: Chavan, A., & Lataye, D. H. (2026). Application of Openair package in R programming and PCA: unveiling air quality dynamics and meteorological influences. Environmental Science and Pollution Research. https://doi.org/10.1007/s11356-026-38275-w

Image Credits: AI Generated

DOI: 10.1007/s11356-026-38275-w

Keywords: air pollution, Openair, R programming, principal component analysis, Nagpur, PM2.5, PM10, ventilation coefficient, Air Quality Index, meteorology, combustion sources, Tier-2 cities

The research team, led by Mahesh Bankar of Dr. Babasaheb Ambedkar Technological University in Lonere, together with Vinayak Patki, Sachin Pore, and Mithun B. Patil, focused on the three pollutants that dominate the air quality picture at the Solapur Municipal Corporation monitoring station: sulphur dioxide, nitrogen oxides, and respirable suspended particulate matter. These pollutants are the primary inputs used to compute the Air Quality Index in India, a single number that condenses complex atmospheric chemistry into a scale that officials and the public can act upon. Rather than treating the index as a purely statistical artifact, the team set out to learn the underlying nonlinear relationships between pollutant concentrations and the resulting index, relationships that simpler models routinely miss.

The technical pipeline behind the model reflects the messy reality of real-world environmental data. Monitoring stations in developing cities often suffer from gaps caused by instrument downtime, calibration issues, and communication failures, and the Solapur dataset was no exception. The researchers applied interpolation to fill in missing values, a technique that estimates absent readings from the surrounding data points. They then eliminated outliers, the anomalous readings that can arise from sensor malfunctions or transient local events, and normalized the features so that all input variables operated on comparable scales. Each of these preprocessing steps matters: unhandled gaps and outliers can destabilize neural network training, while unscaled features can cause the optimizer to weight certain pollutants disproportionately simply because their units produce larger numbers.

At the heart of the study is a three-layer deep neural network trained with the Adam optimizer, a widely used adaptive gradient descent algorithm that adjusts learning rates individually for each parameter. The network was trained to minimize the mean squared error between its predictions and the observed Air Quality Index values, a standard loss function that penalizes large deviations heavily and thereby pushes the model toward accuracy on the most consequential errors. Three-layer architectures of this kind strike a practical balance: they possess enough depth to capture nonlinear interactions among pollutants, such as the way particulate levels respond differently to precursor gases depending on season and meteorology, while remaining small enough to train efficiently on the modest datasets typical of a single monitoring station.

The performance figures reported by the team are striking. The model achieved a mean absolute error of 3.88 index points and a root mean squared error of 5.92, both small relative to the range of values the Air Quality Index can span. The coefficient of determination of 0.983 indicates that the network’s predictions track the observed values almost perfectly across the evaluation period. For context, an error of under four index points means the model can generally distinguish not just between broad categories such as moderate and poor air quality, but between fine gradations within those categories, which is precisely the level of fidelity needed for meaningful public health advisories.

Equally important is the comparative analysis the researchers conducted against baseline models. Linear regression, the workhorse of classical statistics, assumes that the Air Quality Index responds linearly to pollutant concentrations, an assumption that atmospheric science has repeatedly shown to be false. Support vector regression, a more flexible machine learning method that maps inputs into higher-dimensional spaces, performs better but still struggled to match the deep network. The deep neural network’s superiority over both baselines demonstrates that the relationships embedded in air pollution data are genuinely nonlinear and that architectures capable of representing those nonlinearities deliver tangible gains in predictive skill rather than merely theoretical elegance.

The significance of this work extends well beyond a single city in Maharashtra. Cities across India and the wider developing world operate sparse monitoring networks, often with just one or two reference-grade stations per urban area, and the data those stations produce are frequently incomplete. Sophisticated forecasting systems built on numerical atmospheric models demand extensive computational resources, detailed emissions inventories, and meteorological inputs that many municipalities cannot assemble. A neural network trained on pollutant concentrations from a single station, by contrast, requires only data that municipal corporations already collect and report publicly. The Solapur study used air quality data freely available from the Maharashtra Pollution Control Board, underscoring how accessible such systems have become.

The study also situates itself within a rapidly expanding global literature on machine learning for air quality. Recent years have seen researchers apply recurrent architectures, convolutional networks, hybrid models combining gradient boosting with deep learning, and even quantum-inspired neural networks to pollutant forecasting in cities from Delhi to Kampala. Systematic reviews of the field conclude that artificial intelligence methods now consistently outperform traditional statistical approaches for both monitoring and forecasting tasks. What the Solapur study adds is a demonstration that this revolution reaches down to second-tier cities, not just megacities with dense sensor networks and abundant funding, and that a carefully preprocessed dataset from a single station can support a genuinely high-performing model.

There remain important caveats and directions for future work. A model trained on data from one station captures the pollution dynamics of that location and may not generalize to other neighborhoods or cities without retraining, particularly where emission sources differ, as they do between industrial corridors and residential districts. The study also predicts the index from concurrent pollutant concentrations rather than forecasting it days in advance, and true early-warning systems require lead time. Nonetheless, the framework the researchers describe, from interpolation and outlier removal through normalization, network design, and rigorous evaluation with multiple error metrics, provides a replicable template that other municipalities can adapt to their own monitoring records.

The broader stakes could hardly be higher. Exposure to particulate matter and pollutant gases is linked to respiratory disease, cardiovascular illness, and premature death, and accurate, timely air quality information is a prerequisite for protective measures ranging from school closures to traffic restrictions. If deep learning models of the kind developed for Solapur can be deployed across the hundreds of Indian cities that monitor air quality, they could convert routine regulatory data into actionable forecasts at almost no additional cost. The authors report no funding or conflicts of interest, and they state that the data and code supporting the findings are available from the corresponding author upon reasonable request, an openness that may accelerate exactly that kind of adoption. In a field where the gap between data collection and public benefit has often been wide, this study shows how a modest neural network, trained on freely available measurements, can begin to close it.

Subject of Research: Deep neural network prediction of the Air Quality Index from pollutant concentrations at an urban monitoring station in Solapur, India

Article Title: Deep neural network-based prediction of air quality index (AQI) for municipal corporation station, Solapur, Maharashtra State, India

Article References: Bankar, M., Patki, V., Pore, S., & Patil, M. B. (2026). Deep neural network-based prediction of air quality index (AQI) for municipal corporation station, Solapur, Maharashtra State, India. Theoretical and Applied Climatology, 157(10), Article 651. https://doi.org/10.1007/s00704-026-06588-y

Image Credits: AI Generated

DOI: 10.1007/s00704-026-06588-y

Keywords: air quality index, deep neural network, air pollution, Solapur, Maharashtra, machine learning, particulate matter, sulphur dioxide, nitrogen oxides, environmental monitoring, Adam optimizer, regression baselines

The study, led by Yuanhao Du and Xiuli Geng of the University of Shanghai for Science and Technology, together with Sheng Cheng and Hongliu Zhang, addresses two weaknesses the authors identify in the existing literature. First, most air quality index (AQI) prediction models do not adequately account for the noisy character of AQI data, which blends pollutant concentrations, meteorological influences, and measurement artifacts into a single fluctuating number. Second, previous models have largely focused on a single region, leaving the cross-regional prediction problem—arguably the more useful one for public health planning—largely unsolved. The new framework is an explicit attempt to handle both challenges within one pipeline.

The first stage of the pipeline is data preparation, and it begins with a technique called complementary ensemble empirical mode decomposition with adaptive noise, or CEEMDAN. Rather than filtering the raw AQI signal with fixed mathematical filters, CEEMDAN adaptively decomposes it into a set of simpler components at different scales, adding controlled white noise in complementary pairs so that the residual noise in the decomposition largely cancels itself out. The result, according to the researchers, is a significant reduction in the noise effects that plague raw monitoring data. Once the data has been denoised in this way, the team applies phase space reconstruction, a method drawn from nonlinear dynamics that rebuilds the hidden geometric structure of the time series by embedding past values into a higher-dimensional space. This step is intended to expose the dynamic characteristics of the AQI series—the delayed dependencies and oscillations that a predictive model must capture in order to look even one day into the future.

With the data conditioned, the framework turns to its predictive engine: a deep echo state network modified by what the authors call a dynamic activation mechanism, abbreviated DCM-DESN. Echo state networks belong to the reservoir computing family, in which a large, sparsely connected recurrent layer—the reservoir—transforms input signals into a rich set of internal states, while only the readout weights connecting the reservoir to the output are trained. This makes them fast to train and well suited to time-series problems, and stacking reservoirs into a deep architecture extends their memory and representational capacity. The catch is that deep reservoirs typically demand a large number of neurons to perform well, which inflates computational cost. The dynamic activation mechanism adjusts the nonlinear connection mode of the deep echo state network, and by doing so it reduces the model’s dependence on the number of reservoir neurons—allowing the network to reach strong performance with a leaner internal structure.

Training the readout weights of such a network is itself an optimization problem, and this is where the second custom ingredient enters. The researchers propose an improved version of atom search optimization, a nature-inspired algorithm that models the movement of atoms under interaction forces to search a solution space. Their improved atomic search optimization, or IASO, is used to tune the network weights of the DCM-DESN, replacing gradient-based tuning of the reservoir-to-output connections with a global search that is less likely to be trapped in poor local solutions. The combination—CEEMDAN and phase space reconstruction on the front end, a dynamically activated deep reservoir in the middle, and IASO on the training side—forms what the authors describe as an intelligent AQI monitoring framework built on multi-source data.

The transfer learning component is what elevates the framework from a single-city forecaster to a cross-regional one. Transfer learning, a strategy now central to modern machine learning, allows a model trained on one task or domain to be adapted to another with far less data than training from scratch would require. In this study, the researchers constructed multiple cross-city transfer-learning tasks using daily data from Xi’an, Wuhan, Changsha, and Shenzhen covering the period from January 1, 2021 to March 24, 2024. For each task, the IASO-optimized DCM-DESN is first pretrained on data from a source city. The pretrained reservoir-based model is then transferred to the target city and frozen—its internal dynamics held fixed—while only the output weights are fine-tuned using a limited target-domain adaptation set drawn from the destination city.

The design choice to freeze the reservoir and retrain only the readout is both computationally elegant and practically significant. Because reservoir computing trains only its output layer in the first place, adaptation to a new city reduces to a lightweight re-fitting of a small set of weights, using just a limited sample of local data. In operational terms, that means a city with a short monitoring history could borrow the learned atmospheric dynamics of a data-rich neighbor and still produce accurate one-day-ahead AQI forecasts. The inputs to the model are historical pollutant concentrations and related environmental variables available up to the current day, which are used to directly predict the AQI one day ahead—a horizon short enough to be actionable for public health advisories yet long enough to require genuine predictive skill.

To evaluate the framework, the team measured performance across the various transfer tasks using three standard regression metrics: mean absolute error (MAE), root mean square error (RMSE), and the coefficient of determination (R²). Together these metrics capture both the typical size of forecast errors and the proportion of variance in the observed AQI that the model explains. The authors report that the proposed model demonstrated excellent performance across the different tasks, validating the architecture’s ability to generalize across cities with distinct pollution profiles. The four cities themselves span a meaningful range of conditions: Xi’an in the northwest, Wuhan and Changsha in central China, and Shenzhen in the subtropical south, each with its own mix of emissions sources and climatic drivers.

The study also connects to a broader body of work on decomposition-based air quality forecasting. Previous research has combined CEEMDAN with recurrent architectures such as independent recurrent neural networks and long short-term memory networks, and with attention-augmented convolutional models, to squeeze better forecasts out of noisy AQI series. Spatiotemporal graph convolutional networks have been applied to exploit correlations among monitoring stations, and transfer learning has been explored for predicting different pollutants and for assessing events such as COVID-19 lockdown effects on air quality. What distinguishes the new framework is the integration of these strands—adaptive decomposition, phase space reconstruction, deep reservoir computing with a dynamic activation mechanism, metaheuristic weight optimization, and cross-city transfer—into a single end-to-end pipeline aimed squarely at the one-day-ahead AQI problem.

The implications reach beyond the four cities studied. Accurate real-time understanding of AQI evolution is described by the authors as essential for atmospheric pollution prevention and urban public health management, and a framework that can be adapted to a new city with limited local data could lower the barrier for municipalities that lack long monitoring records. The work was supported by the National Natural Science Foundation of China under grant number 72271164, and the authors state they have no competing financial or non-financial interests to disclose. As cities worldwide grapple with the health burden of air pollution—exposure that the World Health Organization continues to flag as a major environmental health risk—tools that make accurate forecasting portable from one urban environment to another may prove among the most consequential applications of machine learning in atmospheric science. The study’s data will be made available on request, and the full article is published in Air Quality, Atmosphere & Health, volume 19, article number 218.

Subject of Research: Cross-city air quality index prediction using transfer learning and an optimized deep echo state network

Article Title: An intelligent AQI prediction method using transfer learning and an IASO-optimized DCM-DESN under multi-source data: A case study of four Chinese cities

Article References: Du, Y., Geng, X., Cheng, S., & Zhang, H. (2026). An intelligent AQI prediction method using transfer learning and an IASO-optimized DCM-DESN under multi-source data: A case study of four Chinese cities. Air Quality, Atmosphere &amp; Health, 19(10), Article 218. https://doi.org/10.1007/s11869-026-02109-y

Image Credits: AI Generated

DOI: 10.1007/s11869-026-02109-y

Keywords: air quality index, transfer learning, deep echo state network, CEEMDAN, phase space reconstruction, atom search optimization, reservoir computing, cross-city prediction, machine learning, air pollution forecasting, China, public health

The research team, led by Mushtaq Ahmad and Sirima Panyametheekul, deployed low-cost laser particle sensors at four distinct locations across the Chulalongkorn University campus: the School of Agricultural Resources, the Chulalongkorn University Language Institute, the Central Library, and the Chamchuri 10 building, also known as the UltimateX Library. These sites were deliberately chosen to represent different microenvironments, from high-occupancy teaching spaces with mechanical ventilation to quiet study halls where students spend extended hours. Outdoor sensors were placed near Phayathai and Rama IV Roads, two of Bangkok’s busiest traffic arteries, to capture the pollution load generated by vehicles and campus activity. Each sensor recorded data at ten-minute intervals, producing an enormous dataset spanning both the dry and rainy seasons across three full annual cycles.

The instrumentation itself is a case study in modern environmental monitoring. The team used Plantower PMS5003 G5 sensors, compact laser-scattering devices that count particles as small as 0.3 micrometers. Before trusting the data, the researchers calibrated the sensors against reference-grade instruments: a U.S. EPA Federal Equivalent Method monitor for PM2.5 and a research-grade TSI DustTrak for PM10. The calibration results were impressive for PM2.5, with a coefficient of determination of 0.921 and a root mean square error of just 1.83 micrograms per cubic meter, though the sensors slightly underestimated concentrations. Performance for PM10 was more moderate, with an R-squared of 0.712, reflecting the well-known difficulty optical sensors face when estimating coarse particle mass, which is more sensitive to particle shape, density, and humidity-driven swelling.

The headline finding concerns seasonality. Monthly mean PM2.5 concentrations ranged from 0.8 to 20.8 micrograms per cubic meter indoors and 0.8 to 44.3 micrograms per cubic meter outdoors, with PM10 reaching up to 40.0 and 51.9 micrograms per cubic meter respectively. In every case, the dry season months from December to February produced the highest readings, a pattern the researchers attribute to a lower planetary boundary layer, regional biomass burning, transboundary pollution transport, and stagnant weather conditions during haze episodes. During those dry-season months, daily PM2.5 concentrations exceeded the World Health Organization’s 24-hour guideline of 15 micrograms per cubic meter in nearly every monitored building, both indoors and out. The only exception was the indoor air of the Central Library, whose filtration apparently kept fine particles at bay even as the city outside choked.

That library result points to the study’s most intriguing thread: the indoor-to-outdoor ratio, a simple but powerful metric that reveals whether a building’s air is dominated by outdoor infiltration or by sources within. An I/O ratio below one suggests outdoor air is the main driver, while values above one signal significant indoor generation or pollutant accumulation. The ratios varied wildly across buildings and years. The School of Agricultural Resources posted a mean ratio of 1.79 in 2023, with monthly values exceeding 2.0 during the mid-year months, hinting at occupant activities, resuspended dust, or insufficient ventilation. The Central Library, by contrast, achieved remarkably low ratios of 0.06 in 2022 and 0.29 in 2023, evidence of effective exclusion of outdoor particles, before jumping to 1.46 in 2024, possibly reflecting changes in ventilation operation or occupancy. Chamchuri 10 stayed below unity for three years, then surged to a mean of 1.55 in 2025. These swings demonstrate that building design, ventilation strategy, and human behavior can matter as much as the pollution outside.

Correlation analysis added another layer of nuance. Both indoor and outdoor particulate concentrations showed negative correlations with relative humidity and temperature, suggesting that meteorological factors did not directly drive pollution levels in this tropical setting. This finding contrasts with studies from temperate climates, where temperature differences drive the stack effect and window-opening behavior strongly modulates indoor air. In Bangkok’s perpetually warm and humid environment, where relative humidity exceeds 70 percent year-round, the seasonal signal appears dominated instead by regional pollution dynamics, particularly the agricultural burning that sweeps smoke across Southeast Asia each dry season.

Beyond measuring concentrations, the team translated their data into human terms by estimating exposure concentrations and potential inhaled doses for three age groups: children aged 6 to 11, adolescents aged 12 to 17, and adults. Using inhalation rates and exposure durations drawn from the U.S. EPA Exposure Factors Handbook, they calculated that adults accumulated the highest absolute inhaled doses, owing to their larger lung volumes and longer daily exposure times. But the researchers caution that a higher dose does not automatically mean higher risk. Children, whose lungs are still developing and who breathe more air per unit of body weight, may suffer disproportionately greater health consequences from the same concentration. Their developing respiratory and immune systems, combined with higher ventilation rates per kilogram of body mass, mean that identical exposure levels can translate into greater internal doses and more lasting harm.

The air quality index analysis offered a glimmer of hope amid the concern. Most monitoring days fell within the Very Good, Good, or Moderate categories of Thailand’s PM2.5-based AQI, and the trend from 2024 to 2025 improved compared with 2022 and 2023, when dry-season readings pushed into the Unhealthy range. Daily PM2.5 concentrations exceeded Thai national standards on only 0.24 to 5.32 percent of monitoring days, while PM10 exceeded standards on 1.70 to 14.1 percent of days. Short pollution episodes still occurred, however, and the study’s authors emphasize that these transient spikes can meaningfully raise exposure during exactly the periods when people are least prepared for them.

The practical implications extend well beyond the campus gates. Because indoor PM concentrations tracked outdoor levels at several sites, the researchers argue that building managers should factor outdoor conditions into ventilation decisions, ramping up filtration or deploying portable air cleaners during haze episodes rather than simply drawing in more outside air. At the policy level, they call for stricter vehicle emission standards, noting that Thailand’s adoption of EURO 5 and EURO 6 rules has been delayed, alongside better public transportation, stronger industrial regulation, and subsidies for HEPA filtration in schools and public buildings. The study explicitly ties these measures to United Nations Sustainable Development Goals on health, education, sustainable cities, and climate action, since the same combustion sources that produce PM2.5 also emit greenhouse gases.

The authors are careful to acknowledge the limits of their work. The measurements come from a single campus and cannot be generalized to all of Bangkok’s buildings, and the team did not analyze particle chemistry, monitor indoor activities like cooking or smoking, or measure air exchange rates directly, meaning the I/O ratios indicate relative relationships rather than true infiltration factors. The exposure estimates rely on generalized scenarios rather than individual time-activity data, and a sensitivity analysis confirmed that a 20 percent change in assumed inhalation rate or exposure duration shifts the calculated dose by exactly 20 percent. Still, as one of the longest continuous paired indoor-outdoor PM records in a tropical educational setting, the study delivers a clear message: in a megacity choking through its haze season, the air inside your building is only as clean as the walls, filters, and ventilation choices that separate it from the street.

Subject of Research: Long-term indoor and outdoor PM2.5 and PM10 exposure assessment in university buildings in Bangkok, Thailand

Article Title: Indoor and outdoor PM 2.5 and PM 10 exposure assessment in university buildings: A campus-based case study in Bangkok, Thailand

Article References: Ahmad, M., Panyametheekul, S., Thaveevong, P., Ngamsritrakul, T., Bennett, C., Khan, M. T., & Zhang, Y. (2026). Indoor and outdoor PM2.5 and PM10 exposure assessment in university buildings: A campus-based case study in Bangkok, Thailand. Case Studies in Chemical and Environmental Engineering, 14, Article 101491. https://doi.org/10.1016/j.cscee.2026.101491

Image Credits: AI Generated

DOI: 10.1016/j.cscee.2026.101491

Keywords: PM2.5, PM10, indoor air quality, Bangkok, air quality index, particulate matter, haze, exposure assessment, inhaled dose, university buildings, tropical megacity, WHO air quality guidelines