Search

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., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.)

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 NO 2 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 (R 2 : 0.388-0.484) and had the worst direct (R 2 : 0.0001-0.349) and indirect transferability (R 2 : 0.07-0.352). Those models using the traditional method had the highest model fitting R 2 (0.629-0.966) with lower cross-validation R 2 (0.495-0.941) and slightly better direct transferability (R 2 : 0.0003-0.386) but suffered poor model interpretability when indirectly transferred to new locations. Our proposed models had comparable model fitting R 2 (0.601-0.966) and the best cross-validation R 2 (0.514-0.941). They also had the strongest direct transferability (R 2 : 0.006-0.590) and moderate-to-good indirect transferability (R 2 : 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., (© 2022. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.)

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 (NO 2 ) 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.

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., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier B.V. All rights reserved.)

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 (NO 2 ) concentrations in Auckland at three scales (central business district (CBD), urban, and suburban). The simulated NO 2 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 R 2 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 R 2 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 NO 2 pollutants at multi-scales in large study areas with distinct urban design and configurations., (Copyright © 2019 Elsevier B.V. All rights reserved.)

Spatial saturation studies using source-specific chemical tracers are commonly used to examine intra-urban variation in exposures and source impacts, for epidemiology and policy purposes. Most such studies, however, has been performed in North America and Europe, with substantial regional combustion-source contributions. In contrast, Auckland, New Zealand, a large western city, is relatively isolated in the south Pacific, with minimal impact from long-range combustion sources. However, fluctuating wind patterns, complex terrain, and an adjacent major port complicate pollution patterns within the central business district (CBD). We monitored multiple pollutants (fine particulate matter (PM 2.5 ), black carbon (BC), elemental composition, organic diesel tracers (polycyclic aromatic hydrocarbons (PAHs), hopanes, steranes), and nitrogen dioxide (NO 2 )) at 12 sites across the ~5 km 2 CBD during autumn 2014, to capture spatial variation in traffic, diesel, and proximity to the port. PM 2.5 concentrations varied 2.5-fold and NO 2 concentrations 2.9-fold across the CBD, though constituents varied more dramatically. The highest-concentration constituent was sodium (Na), a distinct non-combustion-related tracer for sea salt (µ = 197.8 ng/m 3 (SD = 163.1 ng/m 3 )). BC, often used as a diesel-emissions tracer, varied more than five-fold across sites. Vanadium (V), higher near the ports, varied more than 40-fold across sites. Concentrations of most combustion-related constituents were higher near heavy traffic, truck, or bus activity, and near the port. Wind speed modified absolute concentrations, and wind direction modified spatial patterns in concentrations (i.e., ports impacts were more notable with winds from the northeast).

During winter nights, woodsmoke may be a predominant source of air pollution, even in cities with many sources. Since two major earthquakes resulted in major structural damage in 2010 and 2011, reliance on woodburning for home heating has increased substantially in Christchurch, New Zealand (NZ), along with intensive construction/demolition activities. Further, because NZ is a relatively isolated western country, it offers the unique opportunity to disentangle multiple source impacts in the absence of long-range transport pollution. Finally, although many spatial saturation studies have been published, and levoglucosan is an established tracer for woodburning emissions, few studies have monitored multiple sites simultaneously for this or other organic constituents, with the ability to distinguish spatial patterns for daytime vs. nighttime hours, in complex urban settings. We captured seven-day integrated samples of PM 2.5 , and elemental and organic tracers of woodsmoke and diesel emissions, during "daytime" (7 a.m. - 5:30 p.m.) and "nighttime" (7 p.m. - 5:30 a.m.) hours, at nine sites across commercial and residential areas, over three weeks in early winter (May 2014). At a subset of six sites, we also sampled during hypothesized "peak" woodburning hours (7 p.m. - 12 a.m.), to differentiate emissions during "active" residential woodburning, vs. overnight smouldering. Concentrations of PM 2.5 were, on average, were twice as high during nighttime than daytime [µ = 18.4 (SD = 6.13) vs. 9.21 (SD = 6.13) µg/m 3 ], with much greater differences in woodsmoke tracers (i.e., levoglucosan [µ = 1.83 (SD = 0.82) vs. 0.34 (SD = 0.17) µg/m 3 ], potassium) and indicators of treated- or painted-wood burning (e.g., arsenic, lead). Only nitrogen dioxide, calcium, iron, and manganese (tracers of vehicular emissions) were higher during daytime. Levoglucosan and most PAHs were higher during "active" woodburning, vs. overnight smouldering. Our time-stratified spatial saturation detected strong spatial variability throughout the study area, which distinctly differed during daytime vs. night time hours, and quantified the substantial contribution of woodsmoke to overnight spatial variation in PM 2.5 across Christchurch. Daytime vs. nighttime differences were greater than those observed across sites. Traffic, especially diesel, contributed substantially to daytime NO 2 and spatial gradients in non-woodsmoke constituents., (Copyright © 2019 Elsevier Inc. All rights reserved.)

We assessed the association between the natural environment and asthma in 49,956 New Zealand children born in 1998 and followed up until 2016 using routinely collected data. Children who lived in greener areas, as measured by the normalized difference vegetation index, were less likely to be asthmatic: a 1 s.d. increase in normalized difference vegetation index was associated with a 6.0% (95% CI 1.9-9.9%) lower risk of asthma. Vegetation diversity was also protective: a 1 s.d. increase in the number of natural land-cover types in a child's residential meshblock was associated with a 6.7% (95% CI 1.5-11.5%) lower risk. However, not all land-cover types were protective. A 1 s.d. increase in the area covered by gorse (Ulex europaeus) or exotic conifers, both non-native, low-biodiversity land-cover types, was associated with a 3.2% (95% CI 0.0-6.0%) and 4.2% (95% CI 0.9-7.5%) increased risk of asthma, respectively. The results suggest that exposure to greenness and vegetation diversity may be protective of asthma.

Differences in urban design features may affect emission and dispersion patterns of air pollution at local-scales within cities. However, the complexity of urban forms, interdependence of variables, and temporal and spatial variability of processes make it difficult to quantify determinants of local-scale air pollution. This paper uses a combination of dense measurements and a novel approach to land-use regression (LUR) modeling to identify key controls on concentrations of ambient nitrogen dioxide (NO2) at a local-scale within a central business district (CBD). Sixty-two locations were measured over 44 days in Auckland, New Zealand at high density (study area 0.15 km(2)). A local-scale LUR model was developed, with seven variables identified as determinants based on standard model criteria. A novel method for improving standard LUR design was developed using two independent data sets (at local and "city" scales) to generate improved accuracy in predictions and greater confidence in results. This revised multiscale LUR model identified three urban design variables (intersection, proximity to a bus stop, and street width) as having the more significant determination on local-scale air quality, and had improved adaptability between data sets.