A comprehensive review of the development of land use regression approaches for modeling spatiotemporal variations of ambient air pollution: A perspective from 2011 to 2023.

Land use regression (LUR) models are widely used in epidemiological and environmental studies to estimate humans' exposure to air pollution within urban areas. However, the early models, developed using linear regressions and data from fixed monitoring stations and passive sampling, were primarily designed to model traditional and criteria air pollutants and had limitations in capturing high-resolution spatiotemporal variations of air pollution. Over the past decade, there has been a notable development of multi-source observations from low-cost monitors, mobile monitoring, and satellites, in conjunction with the integration of advanced statistical methods and spatially and temporally dynamic predictors, which have facilitated significant expansion and advancement of LUR approaches. This paper reviews and synthesizes the recent advances in LUR approaches from the perspectives of the changes in air quality data acquisition, novel predictor variables, advances in model-developing approaches, improvements in validation methods, model transferability, and modeling software as reported in 155 LUR studies published between 2011 and 2023. We demonstrate that these developments have enabled LUR models to be developed for larger study areas and encompass a wider range of criteria and unregulated air pollutants. LUR models in the conventional spatial structure have been complemented by more complex spatiotemporal structures. Compared with linear models, advanced statistical methods yield better predictions when handling data with complex relationships and interactions. Finally, this study explores new developments, identifies potential pathways for further breakthroughs in LUR methodologies, and proposes future research directions. In this context, LUR approaches have the potential to make a significant contribution to future efforts to model the patterns of long- and short-term exposure of urban populations to air pollution.

The impact of low emission zones on personal exposure to ultrafine particles in the commuter environment.

Auckland is a city with limited industrial activity, road traffic being the dominant source of air pollution. Thus, the time periods when social contact and movement in Auckland were severely curtailed due to COVID-19 restrictions presented a unique opportunity to observe impacts on pedestrian exposure to air pollution under a range of different traffic flow scenarios, providing insights into the impacts of potential future traffic calming measures. Pedestrian exposure to ultrafine particles (UFPs), was measured using personal monitoring along a customised route through Central Auckland during different COVID-19-affected traffic flow conditions. Results showed that reduced traffic flows led to statistically significant reductions in average exposure to UFP under all traffic reduction scenarios (TRS). However, the size of the reduction was variable in both time and place. Under the most stringent TRS (traffic reduction of 82 %), median ultrafine particle (UFP) concentrations reduced by 73 %. Under the less stringent scenario, the extent of reduction varied in time and space; a traffic reduction of 62 % resulted in a 23 % reduction in median UFP concentrations in 2020 but in 2021 similar traffic reductions led to a decrease in median UFP concentrations of 71 %. Under all scenarios, the magnitude of the impact of traffic reductions on UFP exposure varied along the route, with areas dominated by emissions from construction and ferry/port activities showing little correlation between traffic flow and exposure. Shared traffic spaces, previously pedestrianised, also recorded consistently high concentrations with little variability observed. This study provided a unique opportunity to assess the potential benefits and risks of such zones and to help decision-makers evaluate future traffic management interventions (such as low emissions zones). The results suggest that controlled traffic flow interventions can result in a significant reduction in pedestrian exposure to UFPs, but that the magnitude of reductions is sensitive to local-scale variations in meteorology, urban land use and traffic flow patterns.

Evidence and Mass Quantification of Atmospheric Microplastics in a Coastal New Zealand City.

This study investigated the atmospheric deposition of microplastics (MPs) in Auckland, New Zealand, from two sampling sites over a 9-week period. The sizes, morphologies, number counts, and mass concentrations of specific polymers were determined for airborne MPs using a combination of a Nile Red-assisted automated fluorescence microscopy technique in series with pyrolysis-gas chromatography-mass spectrometry (Pyr-GC/MS). This enabled a larger number of MPs to be analyzed from each sample compared to traditional spectroscopic techniques. Microplastic number concentrations increased exponentially with decreasing size. The results show the importance of using consistent methodologies and size cutoffs when comparing microplastic data between studies. Eight polymers were quantified in the atmospheric deposition samples, with polyethylene (PE), polycarbonate (PC), and poly(ethylene terephthalate) (PET) being the most commonly observed. The largest MP deposition rates at an urban rooftop correlated with winds originating from the marine environment with speeds between 15 and 20 m s-1, indicating that airborne MPs in coastal regions may originate from wave-breaking mechanisms. This study represents the first report of using Pyr-GC/MS to determine the chemical compositions and mass concentrations of atmospheric microplastics, along with corresponding data on their sizes, morphologies, and number counts.

Development of transferable neighborhood land use regression models for predicting intra-urban ambient nitrogen dioxide (NO2) spatial variations.

Land use regression (LUR) models have been extensively used to predict air pollution exposure in epidemiological and environmental studies. The lack of dense routine monitoring networks in big cities places increased emphasis on the need for LUR models to be developed using purpose-designed neighborhood-scale monitoring data. However, the unsatisfactory model transferability limits these neighborhood LUR models to be then applied to other intra-urban areas in predicting air pollution exposure. In this study, we tackled this issue by proposing a method to develop transferable neighborhood NO2 LUR models with comparable predictive power based on only micro-scale predictor variables for modeling intra-urban ambient air pollution exposure. Taking Auckland metropolis, New Zealand, as a case study, the proposed method was applied to three neighborhoods (urban, central business district, and dominion road) and compared with the corresponding counterpart models developed using pools of (a) only macro-scale predictor variables and (b) a mixture of both micro- and macro-scale predictor variables (traditional method). The results showed that the models using only macro-scale variables achieved the lowest accuracy (R2: 0.388-0.484) and had the worst direct (R2: 0.0001-0.349) and indirect transferability (R2: 0.07-0.352). Those models using the traditional method had the highest model fitting R2 (0.629-0.966) with lower cross-validation R2 (0.495-0.941) and slightly better direct transferability (R2: 0.0003-0.386) but suffered poor model interpretability when indirectly transferred to new locations. Our proposed models had comparable model fitting R2 (0.601-0.966) and the best cross-validation R2 (0.514-0.941). They also had the strongest direct transferability (R2: 0.006-0.590) and moderate-to-good indirect transferability (R2: 0.072-0.850) with much better model interpretability. This study advances our knowledge of developing transferable LUR models for the very first time from the perspective of the scale of the predictor variables used in the model development and will significantly benefit the wider application of LUR approaches in epidemiological and environmental studies.

The Effect of Route Choice in Children's Exposure to Ultrafine Particles Whilst Walking to School.

Children walking to school are at a high risk of exposure to air pollution compared with other modes because of the time they spend in close proximity to traffic during their commute. The aim of this study is to investigate the effect of a walker's route choice on their exposure to ultrafine particles (UFP) on the walk to school. During morning commutes over a period of three weeks, exposure to UFP was measured along three routes: two routes were alongside both sides of a busy arterial road with significantly higher levels of traffic on one side compared to the other, and the third route passed through quiet streets (the background route). The results indicate that the mean exposure for the pedestrian walking along the background route was half the exposure experienced on the other two routes. Walkers on the trafficked side were exposed to elevated concentrations (>100,000 pt/cc) 2.5 times longer than the low-trafficked side. However, the duration of the elevated exposure for the background route was close to zero. Public health officials and urban planners may use the results of this study to promote healthier walking routes to schools, especially those planned as part of organized commutes.

Socioeconomic Status and Route Characteristics in Relation to Children's Exposure to Air Pollution from Road Traffic While Walking to School in Auckland, New Zealand.

Pedestrian exposure to traffic-related air pollution depends on many factors; including the nature of the traffic on nearby roads and the route characteristics. In this study, we investigated the effects of differences in vehicle fleet and transport infrastructure on children's exposure to traffic-related air pollution during their morning walk to school in Auckland, New Zealand. Five pairs of routes to schools-each including a low and a high socioeconomic status (SES) school-were selected and traversed over 7-8 days for each pair while measuring particle number concentration (PNC) and GPS coordinates. At the same time, a sample of 200 license plates of cars from each school's neighborhood was captured using videos, and the age of each car was extracted using an online database. Both the mean age of the cars and the percentage of old cars (>14 years) were found to be higher in the low SES areas. However, neither measure of vehicle age was associated with a significant difference in exposure to PNC for pedestrians. Route features including narrow footpaths and a higher density of traffic lights may be associated with higher levels of PNC exposure. These findings provide insight into the role of urban design in promoting healthy commutes to school.

Evaluating the Effect of Ambient Concentrations, Route Choices, and Environmental (in)Justice on Students' Dose of Ambient NO2 While Walking to School at Population Scales.

The commuting microenvironment accounts for a large part of students' diurnal exposure to air pollution, especially in cities in developed countries where air pollution is caused predominantly by vehicle traffic. Accurate quantification of students' exposure and pollution dose during their commute from home to school requires their home addresses and details of the schools they attend. Such details are usually inaccessible or difficult to obtain at population scales due to privacy issues. Therefore, estimates of students' exposure to, and dose of, air pollution at population scales have to rely on simulated origins and destinations, which may bias the results. This contribution overcomes this limitation by quantifying students' terrain-based dosage of ambient nitrogen dioxide (NO2) during their commute from home to school while walking along (a) the shortest-distance routes and (b) an alternative lowest-dose route. This is determined at population scales for students in Auckland, New Zealand using a rich dataset of observed home addresses and schools attended for 14,091 walking students. This study also determines the bias introduced when using simulated addresses (as opposed to observed data) to calculate the same result. Finally, we examine exposure inequalities among students of different socioeconomic backgrounds at school, at home, and during walking commutes. Results show that only 17.48% of students in the whole of Auckland can find alternative lowest-dose routes. The portion is higher (26%) in central Auckland because of its better road network connectivity. The trade-off analysis identifies that for only about 30% of students, a 1% increase in route length is associated with a >1% reduction in dosage if using the alternative lowest-dose route. Greater benefits were observed in suburban Auckland (a less-polluted area) than in central Auckland, which highlights the importance of taking an alternative lowest-dose route, especially for students whose shortest-distance routes overlap with or run parallel to an arterial road. The use of simulated addresses resulted in underestimates of both the length and reduced dosage of the alternative routes by up to a quarter in comparison with the results derived from the observed data. Limited evidence of exposure inequality based on commuter exposure was found, but patterns in the central city were opposite to those in the suburbs.

Assessing schoolchildren's exposure to air pollution during the daily commute - A systematic review.

Air pollution is mostly caused by emissions from human activities, and exposure to air pollution is linked with numerous adverse human health outcomes. Recent studies have identified that although people only spend a small proportion of time on their daily commutes, the commuter microenvironment is a significant contributor to their total daily air pollution exposure. Schoolchildren are a particularly vulnerable cohort of the population, and their exposure to air pollution at home or school has been documented in a number of case studies. A few studies have identified that schoolchildren's exposure during commutes is linked with adverse cognitive outcomes and severe wheeze in asthmatic children. However, the determinants of total exposure, such as route choice and commute mode, and their subsequent health impacts on schoolchildren are still not well-understood. The aim of this paper is to review and synthesize recent studies on assessing schoolchildren's exposure to various air pollutants during the daily commute. Through reviewing 31 relevant studies published between 2004 and 2020, we tried to identify consistent patterns, trends, and underlying causal factors in the results. These studies were carried out across 10 commute modes and 12 different air pollutants. Air pollution in cities is highly heterogeneous in time and space, and commuting schoolchildren move through the urban area in complex ways. Measurements from fixed monitoring stations (FMSs), personal monitoring, and air quality modeling are the three most common approaches to determining exposure to ambient air pollutant concentrations. The time-activity diary (TAD), GPS tracker, online route collection app, and GIS-based route simulation are four widely used methods to determine schoolchildren's daily commuting routes. We found that route choices exerted a determining impact on schoolchildren's exposure. It is challenging to rank commute modes in order of exposure, as each scenario has numerous uncontrollable determinants, and there are notable research gaps. We suggest that future studies should concentrate on examining exposure patterns of schoolchildren in developing countries, exposure in the subway and trains, investigating the reliability of current simulation methods, exploring the environmental justice issue, and identifying the health impacts during commuting. It is recommended that three promising tools of smartphones, data fusion, and GIS should be widely used to overcome the challenges encountered in scaling up commuter exposure studies to population scales.

Implications for air quality management of changes in air quality during lockdown in Auckland (New Zealand) in response to the 2020 SARS-CoV-2 epidemic.

The current changes in vehicle movement due to 'lockdown' conditions (imposed in cities worldwide in response to the COVID-19 epidemic) provide opportunities to quantify the local impact of 'controlled interventions' on air quality and establish baseline pollution concentrations in cities. Here, we present a case study from Auckland, New Zealand, an isolated Southern Hemisphere city, which is largely unaffected by long-range pollution transport or industrial sources of air pollution. In this city, traffic flows reduced by 60-80% as a result of a government-led initiative to contain the virus by limiting all transport to only essential services. In this paper, ambient pollutant concentrations of NO2, O3, BC, PM2.5, and PM10 are compared between the lockdown period and comparable periods in the historical air pollution record, while taking into account changes in the local meteorology. We show that this 'natural experiment' in source emission reductions had significant but non-linear impacts on air quality. While emission inventories and receptor modelling approaches confirm the dominance of traffic sources for NOx (86%), and BC (72%) across the city, observations suggest a consequent reduction in NO2 of only 34-57% and a reduction in BC of 55-75%. The observed reductions in PM2.5 (still likely to be dominated by traffic emissions), and PM10 (dominated by sea salt, traffic emissions to a lesser extent, and affected by seasonality) were found to be significantly less (8-17% for PM2.5 and 7-20% for PM10). The impact of this unplanned controlled intervention shows the importance of establishing accurate, local-scale emission inventories, and the potential of the local atmospheric chemistry and meteorology in limiting their accuracy.

Urban environmental health interventions towards the Sustainable Development Goals.

The aim of the UN Sustainable Development Goals (SDGs) is to achieve a better and more sustainable future for all by 2030. Since the majority of the global population lives in cities, it is crucial to identify, evaluate and implement urban interventions (such as zero carbon housing, active transport, better urban connectivity, air pollution control, clean household fuels, and protection from heat and flood events) that will improve health and wellbeing and make our natural and built environment more sustainable. This Virtual Special Issue (VSI) comprises of 14 diverse case studies, methods and tools that provide suggestions for interventions which directly or indirectly support the achievement of the UN SDGs.

Erionite in Auckland bedrock and malignant mesothelioma: an emerging public and occupational health hazard?

Overseas, emerging research has shown that where erionite is present in bedrock as a zeolite, and then subsequently disturbed and blown into the atmosphere, resulting exposure is associated with health effects similar to those caused by asbestos, including malignant mesothelioma (MM). Erionite-induced MM is thought to be particularly prevalent in the construction and quarrying industries, in regions where rock containing erionite is disturbed. In 2015, the then Government Chief Scientist, Sir Peter Gluckman, reported that erionite was a more potent carcinogen than asbestos, and more recent studies have established its presence in the Auckland Region. However, globally at present, there are no established occupational exposure limits for erionite, standard sampling and analytical methods or exposure mitigation guidelines. Given the many major construction projects being carried out in Auckland at the present time, which involve the removal of large quantities of bedrock containing erionite, an assessment of the health risks such activities pose to the public is needed.

Arsenic in Hair as a Marker of Exposure to Smoke from the Burning of Treated Wood in Domestic Wood Burners.

Timber treated with the anti-fungal chemical copper chrome arsenate is used extensively in the New Zealand building industry. While illegal, the burning of treated timber is commonplace in New Zealand and presents a health risk. Outdoor ambient monitoring of arsenic in airborne particulate matter in New Zealand has identified levels that exceed the maximum standards of 5.5 ng m-3 (annual average) at some urban locations. In this study, two-week-old beard hair samples were collected during the winter months to establish individual exposure to arsenic using Inductively Coupled Plasma-Mass Spectrometry. These results were then compared with questionnaire data about wood burner use for the two weeks prior to sampling, and spatial trends in arsenic from ambient monitoring. Results suggest that the burning of construction timber that may contain arsenic is associated with a higher level of arsenic in hair than those who burn logs or coal exclusively. There is no association between the area-level density of wood burners and arsenic levels but a significant correlation with individual household choice of fuel as well as the smell of wood smoke in the community, suggesting very localised influences. Strategies are needed to raise awareness of the risks of burning treated timber and to provide economically-viable alternatives.

A site-optimised multi-scale GIS based land use regression model for simulating local scale patterns in air pollution.

Standard Land Use Regression (LUR) models rely on one universal equation for the entire city or study area. Since this approach cannot represent the heterogeneous controls on pollutant dispersion in central, urban and suburban areas effectively the models are not transferable. Further, if different land use types are not adequately sampled in the measurement campaign, model estimates of local-scale pollutant concentrations may be poor. In this study, this deficiency is overcome with a site-optimised multi-scale GIS based LUR modelling approach developed. This approach is used to simulate nitrogen dioxide (NO2) concentrations in Auckland at three scales (central business district (CBD), urban, and suburban). The simulated NO2 distribution clearly shows a higher concentration of pollution along arterial roads and motorways as expected. Areas of limited dispersion (such as among high-rise buildings of the CBD) are also identified as high pollution areas. Predictor variables vary between scales; no single variable is common to all the scales. The leave-one-out cross validation (LOOCV) revealed that the multi-scale LUR model achieved an R2 of 0.62, 0.86 and 0.73, respectively, at the CBD, urban, and suburban scales. The corresponding LOOCV root-mean-square-errors (RMSE) were 5.58, 3.53 and 4.41 µg·m-3 respectively. Based on these statistical measures the multi-scale LUR model performs slightly better than the universal kriging (UK) model and the standard LUR model, and significantly better than the inverse distance weighting (IDW) and ordinary kriging (OK) models. When evaluated against external observations at eight fixed regulatory monitoring stations, the multi-scale LUR model out-performed all four of the other models considered and achieved an R2 value of 0.85 with the lowest RMSE (8.48 µg·m-3). This approach offers a robust alternative for modelling and mapping spatial concentrations of NO2 pollutants at multi-scales in large study areas with distinct urban design and configurations.

A Reflex Protocol for Creatinine Testing Reduces Costs and Maintains Patient Safety.

BACKGROUND: The Jaffe and enzymatic methods are the 2 most common methods for creatinine measurement. The Jaffe method is less expensive but subject to interferences. Some laboratory scientists have called for the Jaffe method to be retired. OBJECTIVE: To determine the most cost-effective and safe protocol for creatinine measurement. METHOD: We performed a retrospective database review of all outpatient creatinine measurements for 1 year, testing the risk-based reflex testing protocol we had implemented for creatinine measurement. Samples were first measured using the Jaffe method and were reflexed to the enzymatic method if the estimated glomerular filtration rate (eGFR) was between 55 and 65 mL per min per 1.73 m2. RESULTS: There were 104,530 creatinine measurements, of which 11,420 (10.9%) were reflexed to the enzymatic method. The Jaffe method had a positive bias of 0.08 mg per dL (-6.14 mL/min/1.73 m2 eGFR). A total of 3.4% of the paired reflexed specimens were discordant. Also, 133 (1.2%) of the Jaffe results were classified as false negatives and 3411 (29.9%) were classified as false positives. None of the false-negative results and 5 of the false-positive results were considered clinically significant. Using the reflex protocol saved approximately $40,000 per year. CONCLUSIONS: The reflex protocol for creatinine measurement can reduce costs with acceptable risk.

Air Pollution Exposure in Walking School Bus Routes: A New Zealand Case Study.

Walking School Buses (WSBs), organized groups for children to walk to school under the supervision of adults, help reduce traffic congestion and contribute towards exercise. Routes are based largely on need, traffic safety and travel time, with exposure to air pollution not generally considered. This paper explores whether reductions in exposure can be achieved based on the side of the road travelled using data collected in Auckland, New Zealand. Exposure to air pollution was measured for a 25-min commute consisting of a 10-min segment along a quiet cul-de-sac and a 15-min segment along a main arterial road with traffic congestion heavier in one direction. Two participants were each equipped with a portable P-Trak ultrafine particle monitor and a portable Langan carbon monoxide monitor, and walked the route on opposite sides of the road simultaneously, for both morning and afternoon, logging 10-s data. The results suggest that pedestrians travelling on the footpath next to the less congested side of the road in the morning avoid many short-term peaks in concentration and experience significantly lower mean exposures than those travelling on the footpath next to the more congested side. Significant reductions in air pollution exposure could be made for children by taking into account the side of the road in WSB route design.

Solution to the Problem of Calibration of Low-Cost Air Quality Measurement Sensors in Networks.

We provide a simple, remote, continuous calibration technique suitable for application in a hierarchical network featuring a few well-maintained, high-quality instruments ("proxies") and a larger number of low-cost devices. The ideas are grounded in a clear definition of the purpose of a low-cost network, defined here as providing reliable information on air quality at small spatiotemporal scales. The technique assumes linearity of the sensor signal. It derives running slope and offset estimates by matching mean and standard deviations of the sensor data to values derived from proxies over the same time. The idea is extremely simple: choose an appropriate proxy and an averaging-time that is sufficiently long to remove the influence of short-term fluctuations but sufficiently short that it preserves the regular diurnal variations. The use of running statistical measures rather than cross-correlation of sites means that the method is robust against periods of missing data. Ideas are first developed using simulated data and then demonstrated using field data, at hourly and 1 min time-scales, from a real network of low-cost semiconductor-based sensors. Despite the almost naïve simplicity of the method, it was robust for both drift detection and calibration correction applications. We discuss the use of generally available geographic and environmental data as well as microscale land-use regression as means to enhance the proxy estimates and to generalize the ideas to other pollutants with high spatial variability, such as nitrogen dioxide and particulates. These improvements can also be used to minimize the required number of proxy sites.

Use of a handheld low-cost sensor to explore the effect of urban design features on local-scale spatial and temporal air quality variability.

Portable low-cost instruments have been validated and used to measure ambient nitrogen dioxide (NO2) at multiple sites over a small urban area with 20min time resolution. We use these results combined with land use regression (LUR) and rank correlation methods to explore the effects of traffic, urban design features, and local meteorology and atmosphere chemistry on small-scale spatio-temporal variations. We measured NO2 at 45 sites around the downtown area of Vancouver, BC, in spring 2016, and constructed four different models: i) a model based on averaging concentrations observed at each site over the whole measurement period, and separate temporal models for ii) morning, iii) midday, and iv) afternoon. Redesign of the temporal models using the average model predictors as constants gave three 'hybrid' models that used both spatial and temporal variables. These accounted for approximately 50% of the total variation with mean absolute error±5ppb. Ranking sites by concentration and by change in concentration across the day showed a shift of high NO2 concentrations across the central city from morning to afternoon. Locations could be identified in which NO2 concentration was determined by the geography of the site, and others as ones in which the concentration changed markedly from morning to afternoon indicating the importance of temporal controls. Rank correlation results complemented LUR in identifying significant urban design variables that impacted NO2 concentration. High variability across a relatively small space was partially described by predictor variables related to traffic (bus stop density, speed limits, traffic counts, distance to traffic lights), atmospheric chemistry (ozone, dew point), and environment (land use, trees). A high-density network recording continuously would be needed fully to capture local variations.

Characteristics and applications of small, portable gaseous air pollution monitors.

BACKGROUND: Traditional approaches for measuring air quality based on fixed measurements are inadequate for personal exposure monitoring. To combat this issue, the use of small, portable gas-sensing air pollution monitoring technologies is increasing, with researchers and individuals employing portable and mobile methods to obtain more spatially and temporally representative air pollution data. However, many commercially available options are built for various applications and based on different technologies, assumptions, and limitations. A review of the monitor characteristics of small, gaseous monitors is missing from current scientific literature. PURPOSE: A state-of-the-art review of small, portable monitors that measure ambient gaseous outdoor pollutants was developed to address broad trends during the last 5-10 years, and to help future experimenters interested in studying gaseous air pollutants choose monitors appropriate for their application and sampling needs. METHODS: Trends in small, portable gaseous air pollution monitor uses and technologies were first identified and discussed in a review of literature. Next, searches of online databases were performed for articles containing specific information related to performance, characteristics, and use of such monitors that measure one or more of three criteria gaseous air pollutants: ozone, nitrogen dioxide, and carbon monoxide. All data were summarized into reference tables for comparison between applications, physical features, sensing capabilities, and costs of the devices. RESULTS: Recent portable monitoring trends are strongly related to associated applications and audiences. Fundamental research requires monitors with the best individual performance, and thus the highest cost technology. Monitor networking favors real-time capabilities and moderate cost for greater reproduction. Citizen science and crowdsourcing applications allow for lower-cost components; however important strengths and limitations for each application must be addressed or acknowledged for the given use.

Low-cost sensors and crowd-sourced data: Observations of siting impacts on a network of air-quality instruments.

Low-cost sensors offer the possibility of gathering high temporal and spatial resolution crowd-sourced data-sets that have the potential to revolutionize the ways in which we understand individual and population exposure to air pollution. However, one of the challenges associated with crowd-sourced data ('citizen science'), often from low-cost sensors, is that citizens may use sites strongly affected by local conditions, limiting the wider significance of the data. This paper examines results from a low-cost network measuring ground-level ozone to evaluate the impact of siting on data quality. Locations at both reference stations and at private homes or research centers were used, and thought of as a typical 'crowd-sourced' network. Two instruments were co-located at each site to determine intra-site variability and evaluated by standard performance statistics and local-scale activity logs. The wider application of the data for both regional Inter-site variability was evaluated to show-case the wider value and usefulness of crowd-sourced data. Analysis of intra-site variability showed little differences at most sites (<5ppb). Large differences in intra-site variability were detected when sensors were exposed to direct sunlight (causing thermal variations within the instrument) and proximity to large emission sources. Short-term local activities, such as lawn-mowing, were identifiable in the data, but had minimal impact on standard reporting time-scales, and so did not pose as being significant limitations or errors. Inter-site evaluation demonstrated that dense networks of low-cost sensors can add value to existing networks, with minimal impact on the overall data-set quality. Sensors located in crowd-sourced locations nearby to regulatory analyzers were able to capture similar trends and concentrations, supporting their ability to report on wider conditions. Thus crowd-sourced approaches to monitoring (with suitable calibration and data quality control checks) may be an appropriate method for increasing the temporal and spatial resolution of air quality networks.