Air Quality Measurements Research Articles

Future Low-Cost Urban Air Quality Monitoring Networks: Insights from the EU’s AirHeritage Project

The last decade has seen a significant growth in the adoption of low-cost air quality monitoring systems (LCAQMSs), mostly driven by the need to overcome the spatial density limitations of traditional regulatory grade networks. However, urban air quality monitoring scenarios have proved extremely challenging for their operative deployment. In fact, these scenarios need pervasive, accurate, personalized monitoring solutions along with powerful data management technologies and targeted communications tools; otherwise, these scenarios can lead to a lack of stakeholder trust, awareness, and, consequently, environmental inequalities. The AirHeritage project, funded by the EU’s Urban Innovative Action (UIA) program, addressed these issues by integrating intelligent LCAQMSs with conventional monitoring systems and engaging the local community in multi-year measurement strategies. Its implementation allowed us to explore the benefits and limitations of citizen science approaches, the logistic and functional impacts of IoT infrastructures and calibration methodologies, and the integration of AI and geostatistical sensor fusion algorithms for mobile and opportunistic air quality measurements and reporting. Similar research or operative projects have been implemented in the recent past, often focusing on a limited set of the involved challenges. Unfortunately, detailed reports as well as recorded and/or cured data are often not publicly available, thus limiting the development of the field. This work openly reports on the lessons learned and experiences from the AirHeritage project, including device accuracy variance, field recording assessments, and high-resolution mapping outcomes, aiming to guide future implementations in similar contexts and support repeatability as well as further research by delivering an open datalake. By sharing these insights along with the gathered datalake, we aim to inform stakeholders, including researchers, citizens, public authorities, and agencies, about effective strategies for deploying and utilizing LCAQMSs to enhance air quality monitoring and public awareness on this challenging urban environment issue.

Measured air quality impacts after teaching parents about cooking ventilation with a video: a pilot study

Abstract Background Cooking-related emissions contribute to air pollutants in the home and may influence children’s health outcomes. Objective In this pilot study, we investigate the effects of a cooking ventilation intervention in homes with gas stoves, including a video-based educational intervention and range hood replacement (when needed) in children’s homes. Methods This was a pilot (n = 14), before-after trial (clinicaltrials.gov #NCT04464720) in homes in the San Francisco Bay Area that had a school-aged child, a gas stove, and either a venting range hood or over-the-range microwave/hood. Cooking events, ventilation use, and indoor air pollution were measured in homes for 2–4 weeks, and children completed respiratory assessments. Midway, families received this intervention: (1) education about the hazards of cooking-related pollutants and benefits of both switching to back burners and using the range hood whenever cooking and (2) ensuring the range hood met airflow and sound performance standards. The educational intervention was delivered via a video developed in conjunction with local youth. Results We found substantially increased use of back burners and slight increases in range hood use during cooking after intervening. Even though there was no change in cooking frequency or duration, these behavior changes resulted in decreases in nitrogen dioxide (NO2), including significant decreases in the total integrated concentration of NO2 over all cooking events from 1230 ppb*min (IQR 336, 7861) to 756 (IQR 84.0, 4210; p < 0.05) and NO2 collected on samplers over the entire pre- and post-intervention intervals from 10.4 ppb (IQR 3.5, 47.5) to 9.4 (IQR 3.0, 36.1; p < 0.005). There were smaller changes in PM2.5, and no changes were seen in respiratory outcomes. Impact This pilot before-after trial evaluated the use of a four-minute educational video to improve cooking ventilation in homes with gas stoves and one or more school-aged children. Participant behavior changed after watching the video, and there were decreases in indoor air pollutant concentrations in the home, some of which were significant. This brief video is now publicly available in English and Spanish (wspehsu.ucsf.edu/projects/indoor-air-quality), and this provides suggestive evidence of the utility of this simple intervention, which could be particularly beneficial for households that have children with asthma.

Application of a new multi-elevation method for determining the elemental composition of atmospheric precipitation in coastal marine zones.

Assessing local air quality using traditional methods, such as analyzing precipitation composition, is often difficult due to the complex data, which is influenced by a variety of chemical elements from distant atmospheric sources and cyclone formation areas. This study presents a new approach to overcome these challenges: a multi-elevation sampling method that improves the accuracy of local air quality measurements. By collecting precipitation samples at different ground elevations, the technique takes advantage of the natural process where raindrops and snowflakes collect more elements as they fall through the air. This method helps to distinguish local air quality from background levels found at higher, non-industrial elevations. The primary goal was to isolate and identify the elemental fingerprints of marine influences, while excluding contributions from cyclones. In February 2023, 51 meltwater samples were collected from 17 observation points in the city of Sochi under southeast wind conditions. These samples revealed a spectrum of chemical elements, predominantly of marine origin (Mg > Na > Sr > Ca > K > Ce > Mn). The results showed significant differences in elemental concentrations between upland and coastal areas. The Geochemical Indicator of Marine Origin (GIM) ranged from 22 to 3235, confirming the strong influence of marine aerosols in the area. These findings demonstrate the effectiveness of the multi-elevation sampling method in providing robust environmental assessments. This approach, combined with the GIM index, offers valuable insights for improving environmental policies and public health, particularly in coastal regions affected by marine aerosols.

Review of IoT Systems for Air Quality Measurements Based on LTE/4G and LoRa Communications

The issue of air pollution has recently come to light due to rapid urbanization and population growth globally. Due to its impact on human health, such as causing lung and heart diseases, air quality monitoring is one of the main concerns. Improved air pollution forecasting techniques and systems are needed to minimize the human health impact. Systems that fall under the Internet of Things (IoT) topology have been developed to assess and track numerous air quality metrics. This paper presents a review of IoT systems for air quality measurements, where the emphasis is placed on systems with LTE/4G and LoRa communication capabilities. Firstly, an overview of the IoT monitoring system is provided with recent technologies in the market. A critical review is provided of IoT systems regarding air quality using LTE/4G and LoRa communications systems. Lastly, this paper presents a market analysis of commercial IoT devices in terms of the costs, availability of the device, particulate matter each device can measure, etc. A comparative study of these devices is also presented on LTE/4G and possibly LoRa communications systems.

Particulate Matter Production from Prescribed Burns in the Chicagoland Area

Prescribed burns are a land management tool currently used to aid in fire mitigation and to promote desired plant species and reduce undesired species, which are often invasive species. Currently, there is a public stigma surrounding the negative effects of prescribed burns, which impacts their use near residential areas. There are also particular challenges of conducting prescribed burns in urban and suburban areas. A better understanding of these relatively small-scale prescribed burns and their impact on atmospheric chemistry and air quality can allow for better communication with the public about their positive impacts, as well as for the acknowledgement and quantification of their drawbacks. This study considers the particulate matter concentrations proximate to prescribed fires being conducted in the greater Chicago metropolitan area. We deployed low-cost sensors in the spring and fall of 2022 at four sites. Concentrations of particulate matter exceeded accepted human health exposure limits at locations that burn crews would experience, but levels quickly returned to baseline after the high-intensity phase of the burn. With the ever-expanding wildland–urban interface and increased efforts in ecological restoration, understanding the particulate matter emissions of prescribed fires conducted near populated areas can provide more data regarding air quality impacts. These impacts should be weighed against the ecological benefits of prescribed burns and potential air quality impacts from uncontrolled burns from ecosystems that are not treated with prescribed burns. Measuring air quality impacts can also inform fire management practices with the goal of reducing future emissions. With the very present effects of climate change, understanding baselines regarding prescribed fires can better equip future research and fire professionals in fighting fire with fire.

Energy retrofits: Factors affecting a just transition to better indoor air quality

In comparison with other European countries, Ireland has a disproportionately high number of poorly performing energy-inefficient buildings. Consequently, Ireland has one of the most ambitious energy retrofit programmes in Europe. The aim of this study was to evaluate the impact of deep energy renovation measures i.e. (replacement of heating system and upgrade to building envelope) on indoor air quality, thermal comfort and ventilation in a sample of primarily social housing in Ireland. A mixed methods approach including measurement of indoor air quality and a thermal comfort questionnaire survey was employed. Indoor concentration measurements of PM2.5, carbon dioxide, formaldehyde, and radon along with measurements of air temperature and relative humidity were made (N=14). Occupants (n=56) completed a thermal comfort questionnaire survey pre- and post-retrofit. Data collection ran from autumn 2020 until autumn 2023.Thermal comfort improved post-retrofit along with occupant satisfaction with the indoor thermal environment and heating-systems. Post-retrofit, higher bedroom CO2, and higher PM2.5 concentrations were recorded in both living areas and bedrooms. Occupant behaviours e.g. blocking wall vents were significant predictors of poorer ventilation while smoking indoors was related to higher concentrations of PM2.5. In general, homes that had mechanical ventilation systems installed were better ventilated. Underventilation, as indicated by visible condensation or mould was observed in half of the social homes surveyed and reported by questionnaire respondents (26 %). Tailored communications to occupants regarding the important role of ventilation and indoor air quality is required as part of retrofit to optimise impacts on IEQ.

Why do we misperceive air pollution? A scoping review of key judgmental biases

AbstractAir pollution stands as a primary environmental factor influencing mortality globally. Despite its significance, individuals frequently struggle to assess the risks associated with polluted environments. This paper delves into the psychological mechanisms that contribute to the discrepancy between public perception of air quality and objective measurements. Through an extensive literature review spanning several decades, six fundamental heuristics that underpin misperceptions of air quality are identified and examined. The article further explores the theoretical and practical implications of acknowledging these biases in the development of urban environmental policies. By addressing judgmental biases, urban planners can improve the communication of risk information among lay people and ensure that environmental improvements are acknowledged and valued by all members of the community.

Investigation of the potential of in-line particle concentration measurements in gas particle separation processes by using low-cost particulate matter sensors

In addition to the costly laboratory-based particle measurement devices, low-cost particulate matter sensors for ambient air quality measurements can also be employed for the detection of PM in gas cleaning devices. The ability to spatially resolve the actual level of particle number concentration with high temporal resolution within a gas-particle separation process is of significant value for process monitoring and control. The quantitative reliability of the measurement data from low-cost particulate matter sensors remains uncertain when the particle concentrations in the apparatus exceed the upper measurement limit of the sensors, whether temporarily or permanently. This study examines the potential of low-cost particulate matter sensors for in-line particle concentration measurements in gas-particle separation processes. The measurement performance of the low-cost particulate matter sensors is initially examined under idealised conditions in a dust chamber. This is done in order to quantify the deviations of the low-cost particulate matter sensors in comparison to a state-of-the-art precision laboratory high-cost sensor. Subsequently, both sensor types are installed at a wet separator test facility, following the quantification of their respective measurement performances. The low-cost particulate matter sensors are capable of measuring particle number concentration above their specified limit (up to ≈6.1×104cm−3), depending on the measurement conditions. By applying a 3rd polynomial correction function, obtained from the initial quantification measurement, the low-cost particulate matter sensors are able to reproduce the particle number concentration measured by the high-cost sensor. Furthermore, the low-cost particulate matter sensors are capable of reproducing short-term fluctuations in the particle number concentration.

The Cancer Equity Compass: Interactive maps showing the association between social determinants of health and cancer disparities.

149 Background: Cancer incidence and mortality in Black, Hispanic and Native American populations is over twice that in the White population for some prevalent cancers. A new, publicly available mapping tool, the Cancer Equity Compass (CEC), was developed to inform data-driven initiatives to improve equity by identifying community factors associated with cancer disparities. The tool is comprised of interactive maps that visually reveal correlations between two variables. Development of the tool began with assembling a group of nine advisors with expertise in cancer prevention, treatment and health equity. These experts inform the strategy, data sources, variables of interest, and stakeholder value propositions for the CEC. Methods: The data source for the CEC is a national, multi-level database developed using individual, health system, and community data, aggregated by county. For the three most common cancers in the U.S., breast, prostate and lung, screening and prevalence rates were collected from 2021 Medicare 5% research identifiable files and mortality rates from 2020 SEER Medicare rates. These metrics were each mapped with a variety of social determinants of health (SDOH) variables. Population rates for minorities, education, poverty, income and unemployment, and an air quality measure were from the Area Health Resources File. Environmental risks were derived from the Environmental Justice Index and included environmental burden (ranked 0-17 for pollution, toxic sites, built environment, transportation infrastructure and water pollution), air pollution (ranked 0-4), and social vulnerability (ranked 0-14 across socioeconomic variables). Results: Each map includes one cancer metric and one SDOH variable. Each variable is stratified to show the top quartile, middle 50%, and bottom quartile. The choropleth heat maps, created on Tableau, display 9 colors representing each combination of the three levels (>75%, 25-75%, <25%) for both mapped variables. The CEC currently has 140 maps presenting visualizations of county SDOH associations with cancer disparities. Conclusions: The expert advisors on the project have outlined four avenues for the CEC to aid efforts to improve cancer equity: 1) Enhance diversity in clinical trial recruitment and data registries through targeted site recruitment; 2) Identify high opportunity targets for improving equity and informing solutions; 3) Identify common correlates of disparities within and across U.S. counties to direct priorities for national policy and programs for large-scale impact; and 4) Measure change to support improvement efforts and report impact of equity initiatives. The CEC will be expanded with additional measures including healthcare access. The goal is use by stakeholders and changemakers who can apply the reported data to fuel advocacy efforts and enact positive change.

Characterization and impact of airborne particulate matter over Varanasi: A year-long study on concentration, morphology, and elemental composition

Air pollution is an important worldwide issue, especially pronounced in metropolitan and suburban regions, significantly affecting both public health and surroundings. This study investigates the particles' morphology and elemental analysis in Varanasi, a highly inhabited metropolis in the Indo-Gangetic Plain. The research was conducted over a year, from April 2019 to March 2020, utilizing Scanning Electron Microscopy with Energy Dispersive X-ray Spectroscopy, Ion Chromatography, and Atomic Absorption Spectroscopy to analyse particulate matter. Results indicated that mean values of PM2.5 and PM10 were 106.5 ± 67.2μg/m³ and 180.8 ± 71.4 μg/m³, respectively. Often, these amounts exceeded the National Ambient Air Quality Standards. SEM-EDX analysis revealed diverse particle morphologies, with significant contributions from both manmade sources including industrial activities and vehicle emissions, and natural sources, like soil dust. Elemental analysis identified major components, including Carbon, Oxygen, Fluorine, Aluminium, and Silicon. IC analysis highlighted dominant ionic species, such as Ca++, SO4−-, NO3−, and Cl−, with monthly variations reflecting different emission sources. Heavy metals concentrations such as Ni, Cd, Cr, Mn, Cu, Pb, Zn, and Fe were quantified, with concentrations varying significantly across months. The findings underscore the complex nature of aerosols in Varanasi and highlight the immediate need for targeted control over air quality measures to minimize the particulate matter's detrimental effects on the local population and ecosystem.

Spatial Analysis of Sulfur Dioxide (SO2) and Nitrogen Dioxide (NO2) Distribution Using Getis-Ord Gi* in DKI Jakarta Region, Indonesia

Abstract The increasing mobility of population and industry is the main problem in this study, so it will affect the spatial pattern in the distribution of pollutants. The parameters to measure air quality are Sulfur Dioxide and Nitrogen Dioxide. This study aims to (1) analyze air quality distribution patterns with SO2 and NO2 concentration parameters in the DKI Jakarta area in 2021, (2) analyze ambient air quality distribution patterns based on air quality threshold values, and (3) analyze the spatial distribution of SO2 and NO2 in the DKI Jakarta area in 2021. The data used for this research is the average concentration data of SO2 and NO2 pollutants in DKI Jakarta in 2021. This research uses the interpolation method using Inverse Distance Weighting and Getis-Ord Gi*. The results of this research show that the concentration of pollutants that have been interpolated using Inverse Distance Weighting produces higher NO2 than SO2. From the interpolated image, a Hot Spot analysis (Getis-Ord Gi*) was conducted. This analysis showed significant clustering of nearest neighbors, with high concentrations of SO2 in East Jakarta and NO2 in West Jakarta. Therefore, air quality needs to be monitored and managed to maintain environmental stability.

Multi-Sensor Platform in Precision Livestock Farming for Air Quality Measurement Based on Open-Source Tools

Monitoring air quality in livestock farming facilities is crucial for ensuring the health and well-being of both animals and workers. As livestock farming can contribute to the emission of various gaseous and particulate pollutants, there is a pressing need for advanced air quality monitoring systems to manage and mitigate these emissions effectively. This study introduces a multi-sensor air quality monitoring system designed specifically for livestock farming environments. Utilizing open-source tools and low-cost sensors, the system can measure multiple air quality parameters simultaneously. The system architecture is based on SOLID principles to ensure robustness, scalability, and ease of maintenance. Understanding a trend of evolution of air quality monitoring from single-parameter measurements to a more holistic approach through the integration of multiple sensors, a multi-sensor platform is proposed in this work. This shift towards multi-sensor systems is driven by the recognition that a comprehensive understanding of air quality requires consideration of diverse pollutants and environmental factors. The aim of this study is to construct a multi-sensor air quality monitoring system with the use of open-source tools and low-cost sensors as a tool for Precision Livestock Farming (PLF). Analysis of the data collected by the multi-sensor device reveals some insights into the environmental conditions in the monitored barn. Time-series and correlation analyses revealed significant interactions between key environmental parameters, such as strong positive correlations between ammonia and hydrogen sulfide, and between total volatile organic compounds and carbon dioxide. These relationships highlight the critical impact of these odorants on air quality, emphasizing the need for effective barn environmental controls to manage these factors.

Symptoms Reported by Young Adults With Asthma During Wildfire Smoke Season

PurposeThe purpose of this article is to evaluate the relationship between exposure to poor air quality (AQ) and self-reported symptoms among young adults with asthma during wildfire smoke season. MethodsSixty seven young adults (aged 18–26 years) completed the Asthma Control Test and reported asthma symptoms at three time points (baseline, 4 weeks, and 8 weeks) during wildfire season as part of a clinical trial. Bivariate correlations between Asthma Control Test and AQ measures were examined followed by predictive linear regression. Multiple symptoms were compared between participants who experienced poor AQ and those who did not. ResultsAsthma control was inversely related to AQ with increased exposure to poor AQ tied to poor asthma control. A significantly greater proportion of participants reported critical respiratory symptoms when exposed to poor AQ than those who were not. DiscussionRespiratory symptoms are key indicators that young adults can monitor to optimize their asthma management during wildfire smoke season.

Groundwater and indoor air quality of urban village housing: study in Janturan, Warungboto, Yogyakarta

Water and air are important elements for life. The presence of pathogenic bacteria in the surrounding environment can cause health problems resulting from the low quality of water and air. The purpose of this study is to measure the quality of groundwater and air based on microbiological parameters in urban village housing. This research was conducted in a descriptive survey, and a purposive sampling technique was carried out. Groundwater and air samples were taken from six residents' houses located in the Janturan area. The water quality test used the most probable number (MPN) method and the air quality test used the agar filter method using a microbiological air sampler (MAS) to trap microbes. The results of the MPN test for total coliform bacteria from six samples of groundwater showed that three samples were positive for coliform bacteria which exceeded the threshold according to Indonesian standard<50 MPN/100 ml. Assay of <em>E. coli</em> bacteria showed that all samples of groundwater positively contained <em>E. coli</em> bacteria which exceeded the threshold of 0 MPN/100 ml. The results of indoor air quality measurements from 6 residents' houses showed that five houses have air quality did not meet the standard for indoor air germ numbers based on Indonesian standard, less than 700 CFU/m3 (colony forming units per metres cubic) and 1 house that got the number of airborne germs, in the normal range, namely 356 CFU/m3. It was concluded that the water quality and air quality in the urban village did not meet health requirements.

How do occupants experience the onset and triggering of building-related symptoms in damp Australian housing? An inductive-deductive content analysis

In recent years, occupants’ time spent in dwellings has increased concurrently with an increase in the duration and intensity of extreme weather events and the worsening of housing conditions. Indoor dampness and or mould (D/M) are risk factors for building-related symptoms (BRS), yet little is known about how symptoms start, thus hindering where to focus indoor air quality measurements and improvements in occupant safety. This study describes perceived symptom initiators, triggers and coinciding factors that explain what and how symptoms occurred from the perspectives of occupants living with indoor D/M. This study uses an inductive/deductive qualitative analysis, with purposive sampling (n = 154) from data collected via an online survey. Deductive content analysis utilises a framework driven by the Determinants of Health (DoH). Themes that emerged from the inductive analysis include: 1) when occupants or others interact with D/M; 2) when indoor D/M conditions change; and 3) a home full of D/M problems. From the deductive analysis, perceived exposure relating to D/M conditions differed between the initial onset of symptoms and the later triggering of symptoms. Additional coinciding factors relating to climate, socioeconomic, policy and individual factors along with a lack of knowledge, attitudes and beliefs, emerged. This study highlights the need to raise occupant awareness, enhance housing and occupancy regulations, and introduce safe work practices to address the lack of health awareness and health significance regarding indoor D/M housing conditions. It also outlines the need for targeted research measuring bioaerosol exposure related to actions, proximity-based exposure, and wet and repeatedly wet conditions.

Quality Management System in Air Quality Measurements for Sustainable Development

Air pollution is a global health issue and a cause of premature mortality. There is an urgent need to develop air quality monitoring networks and to implement standards enabling dependable testing and delivering reliable results. European standards provide reference methods for testing ambient air quality, which are used in accredited laboratories. In this paper, we present an example of research conducted with the use of a mobile, automated station Airpointer® in an industrial area under pressure from the lime and cement industry located in southeastern Poland. During the measurement campaign, the concentrations of the studied pollutants did not exceed the permissible thresholds, yet they strongly depended on meteorological conditions. The air filter was analysed with an energy dispersive spectroscopy (EDS) microanalyzer in the scanning electron microscope (SEM). The results confirmed that dust particles present in ambient air are connected with local emission sources—industry based on the extraction and processing of minerals. The equipment and measurement techniques used in this study are effective in identifying the potential threat of air pollution. Automated, short-term measurements of air pollution can be a significant source of information, indispensable for drawing up action plans aimed at air quality protection in order to achieve sustainable development goals.

Calibration of particle sensor with neural network algorithm

AbstractBased on air quality index data for the period 2018–2022, Hungary ranks as the 80th most polluted country in the world. Given the air pollution data measured in Hungary and the health impact of air pollution, it is of utmost importance to measure air quality in Hungary focusing on PM10 and PM2.5 pollutants. One possible solution for high-density measurement is to utilize low-cost sensors at the population level. The calibration procedure has to be carried out in a way that does not incur extra costs and maintenance at the physical level. A potential solution is the development of an algorithm to perform the calibration with remote access. This publication presents a fragment of this development, where we attempted to implement the procedure using a neural network and performed a comparative analysis with official data.

Effect of bedroom environment on sleep and physiological parameters for individuals with good sleep quality: A pilot study

Bedroom conditions affect sleep, overall health, and well-being. Field studies reporting measurements of bedroom environment and sleep parameters are almost lacking in the scientific community since the majority of research has been conducted in laboratories. To fill this knowledge gap, we have investigated the associations among bedroom environmental factors (temperature, humidity, CO2, pressure), measured sleep and physiological parameters (heart rate, heart rate variability, and breathing rate) in participants with good sleep quality. Six subjects were enrolled in the study, possessing a minimum of two years of experience in using the wrist actigraph chosen for the experimentation, and exhibiting good sleep quality. The air quality measurements were conducted using the air quality monitor, whereas advanced actigraphs were employed for the assessment of various sleep stage durations. The statistical analysis, carried out applying Linear Mixed Models, revealed that an increase of 100 ppm of CO2 concentration levels is associated with an approximate 0.29 % decline in sleep quality whereas, concerning the effect of the relative humidity, we observed a 0.1 % reduction in sleep quality associated with a 1 % rise in RH. On the other hand, a 1 °C rise in air temperature is associated with an approximate 0.16 % reduction in sleep efficiency, and, at a CO2 concentration of about 900 ppm, the autonomic nervous system (ANS) indicator remains positive for temperature values < 20 °C. These results contribute to mounting evidence underscoring the crucial role of the bedroom atmosphere in facilitating high-quality sleep.

Enhancing air pollution prediction: A neural transfer learning approach across different air pollutants

Air pollution stands out as one of the most alarming environmental challenges. It poses significant risks to human health and the environment. Accurate forecasting of air pollutant concentration levels is crucial for effective air quality management and timely implementation of mitigation strategies. In this paper, the transfer learning technique is investigated using the artificial neural network (ANN), also called multi-layer perception (MLP), to transfer knowledge across different air pollutants forecasting, and therefore, to generalize over a large set of air pollutants in the same air monitoring station. By leveraging the knowledge learned from a source forecasting task, transfer learning allows us to reduce the data requirements, speed up the training of the models, and enhance the predictive performance for different air pollutants for the target forecasting task. We present a comprehensive analysis of the transfer learning across different air pollutants in the same air monitoring station on a large dataset of air quality measurements. Our results demonstrate that transfer learning significantly improves forecasting accuracy with fewer fine-tuning data, particularly when limited labeled data is available for the target task. The findings of this study contribute to the advancement of air pollution forecasting methodologies, facilitating better decision-making processes and proactive air quality management.

Indoor/Outdoor Air Quality Measurements near a Highway in Somerville, MA

Indoor/Outdoor Air Quality Measurements near a Highway in Somerville, MA