Submicron aerosol pollution in Greater Cairo (Egypt): A new type of urban haze?

Greater Cairo, the largest megacity of the Middle East North Africa (MENA) region, is currently suffering from major aerosol pollution, posing a significant threat to public health. However, the main sources of pollution remain insufficiently characterized due to limited atmospheric observations. To bridge this knowledge gap, we conducted a continuous 2-month field study during the winter of 2019-2020 at an urban background site, documenting for the first time the chemical and physical properties of submicron (PM1) aerosols. Crustal material from both desert dust and road traffic dust resuspension contributed as much as 24 % of the total PM1 mass (rising to 66 % during desert dust events), a figure not commonly observed in urban environments. Our observations showed significant decreases in black carbon concentrations and ammonium sulfate compared to data from 15 years ago, indicating an important reduction in both local and regional emissions as a result of effective mitigation measures. The diurnal variability of carbonaceous aerosols was attributed to emissions emanating from local traffic at rush hours and nighttime open biomass burning. Surprisingly, semi-volatile ammonium chloride (NH4Cl) originating from local open biomass and waste burning was found to be the main chemical species in PM1 over Cairo. Its nighttime formation contributed to aerosol water uptake during morning hours, thereby playing a major role in the build-up of urban haze. While our results confirm the persistence of a significant dust reservoir over Cairo, they also unveil an additional source of highly hygroscopic (semi-volatile) inorganic salts, leading to a unique type of urban haze. This haze, with dominant contributors present in both submicron (primarily as NH4Cl) and supermicron (largely as dust) modes, underscores the potential implications of heterogeneous chemical transformation of air pollutants in urban environments.

Prenatal air pollution exposure and childhood obesity: Effect modification by maternal fruits and vegetables intake.

BACKGROUND & AIMS: Prenatal exposure to air pollution is robustly associated with fetal growth restriction but the extent to which it is associated with postnatal growth and the risk of childhood obesity remains unknown. We examined the association of prenatal exposure to air pollution with offspring obesity related measures and evaluated the possible protective effect of maternal fruits and vegetables intake (FV). METHODS: We included 633 mother-child pairs from the Rhea pregnancy cohort in Crete, Greece. Fine particles (PM2.5 and PM10) exposure levels during pregnancy were estimated using land-use regression models. We measured weight, height and waist circumference at 4 and 6 years of age, and body composition analysis was performed at 6 years using bioimpedance. Maternal diet was evaluated by means of a semi-quantitative food frequency questionnaire in mid-pregnancy. Adjusted associations were obtained via multivariable regression analyses and multiplicative interaction was used to evaluate the potential modifying role of FV intake. RESULTS: Exposure to PMs in utero was not associated with measures of adiposity at 4 or 6 years of age. Associations at 4 years did not differ according to maternal consumption of FV. However, at 6 years, among children whose mothers reported consuming less than 5 servings of FV per day, one SD increase in PM10 during pregnancy was associated with increased BMI (beta 0.41 kg/m2, 95% CI: -0.06, 0.88, p-interaction = 0.037) and increased waist circumference (beta 0.83 cm, 95% CI: -0.38, 2.05, p-interaction = 0.043) and one SD increase in PM2.5 was associated with increased fat mass (beta 0.5 kg, 95% CI: 0.0, 0.99, p-interaction = 0.039) and increased percentage of body fat (beta 1.06%, 95% CI: -0.06, 2.17, p-interaction = 0.035). Similarly, higher prenatal PM2.5 and PM10 exposure was associated with increased risk for obesity and abdominal obesity at 6 years only in the low FV group. CONCLUSIONS: Exposure to fine particulate matter during pregnancy was not associated with obesity-related measures at 4 and 6 years. However, only among offspring of mothers who consumed inadequate FV, we observed higher obesity-related measures at 6 years. Our results indicate that mothers' diet during pregnancy may play a role in the relationship between air-pollution and childhood obesity.

Comprehensive chemical characterization of PM2.5 in the large East Mediterranean-Middle East city of Beirut, Lebanon.

The chemical composition of PM2.5 at two sites in Lebanon, a country in the East Mediterranean - Middle East region, is investigated in the spring and summer seasons. The average PM2.5 concentrations were of (29 ± 16) µg/m3 for Beirut urban site and (32 ± 14) µg/m3 for Beirut suburban site. This study showed that the geographic location of the East Mediterranean region, such as its proximity to the Mediterranean Sea and the dust storm intrusion are a significant contributor to the high PM levels from natural sources, which cannot be mitigated, rendering the PM2.5 WHO annual Air Quality guideline unattainable due to high natural background, which also applies to the entire Middle East region. Turkey and Eastern Europe are the dominant origin of air masses throughout our sampling days, suggesting the long-range transport as an important potential contributor to the high observed concentrations of V, Ni, and sulfate in this region most probably in other East Mediterranean countries than Lebanon too. Main local sources determined through the chemical speciation including organics are road transport, resuspension of dust and diesel private generators. A health risk assessment of airborne metals was performed and the carcinogenic risk for all the metals exceeded by 42 (adults) and 14 (children) times the acceptable risk level (10-6) at both sites. Vanadium was the predominant carcinogenic metal, emphasizing the need to replace energy production with cleaner energy on a regional level and highlighting the severe impact of air pollution on the health of inhabitants in this region's main cities.

Fine aerosol sources at an urban background site in the Eastern Mediterranean (Nicosia; Cyprus): Insights from offline versus online source apportionment comparison for carbonaceous aerosols.

A total of 348 daily PM2.5 samples were collected at an urban background site of Nicosia, capital of Cyprus, for one-year period (October 2018-October 2019) to assess the origin and sources of fine PM at the Eastern Mediterranean, a poorly characterized area of the world. The samples were analysed for water soluble ionic species, elemental and organic carbon, carbohydrates and trace metals, the combination of which were utilized to identify pollution sources by applying Positive Matrix Factorization (PMF). Six PM2.5 sources, namely long-range transport (LRT; 38 %), traffic (20 %), biomass burning (16 %), dust (10 %), sea salt (9 %) and heavy oil combustion (7 %), were identified. Despite sampling in an urban agglomeration, the chemical fingerprint of the aerosol is largely dictated by air mass origin rather than local sources. Springtime is characterized by the most elevated particulate levels due to the southerly air masses carrying particles from the Sahara Desert. Northerlies are observed throughout the year but are predominant during summer allowing the LRT source to peak (54 % during summer). Only during winter, due to extensive use of biomass combustion for domestic heating (36.6 % during winter), local sources dominate. A co-located online PMF source apportionment of submicron carbonaceous aerosols (Organic Aerosols, OA; Black Carbon) was performed by the means of an Aerosol Chemical Speciation Monitor (for OA) and an Aethalometer (for BC) for a four-month period. The comparison between the two methodologies allowed to better assess the robustness and limitations of the two methodologies. More specifically, LRT OA and biomass burning BC apportioned by the offline PMF showed a strong consistency with the online apportioned more oxidized oxygenated OA and BCwb, respectively; cross validating these sources. On the other hand, our traffic factor may contain additional hydrocarbon-like OA and BC from fossil fuel sources other than just vehicular emissions. Finally, the offline biomass burning OA source is likely to contain both primary and secondary OA.

Simple multi-residue analysis of persistent organic pollutants and molecular tracers in atmospheric samples.

We present a simple, selective and sensitive analytical method to quantitatively determine a wide range of halogenated persistent organic pollutants and molecular tracers in atmospheric samples. Identification and quantification was carried out by high-resolution gas chromatography, hyphenated with low-resolution mass spectrometry operating in electron impact (EI) and electron capture negative ionization (ECNI) mode. Optimization on a number of instrumental parameters was conducted to obtain ultra-trace detection limits, in the range of few fg/m3 for organohalogen compounds. Repeatability and reproducibility of the method was thoroughly evaluated. The analysis was validated with standard reference materials and successfully applied to actual atmospheric samples. The proposed multi-residue method provides a precise, affordable and practical procedure of sample analysis for environmental research laboratories with conventional instrumentation on a routine basis.•A simple combination of alumina, florisil and silica gel adsorbents was applied to sufficiently isolate polychlorinated biphenyls, organochlorine pesticides, polycyclic aromatic hydrocarbons, long chain n-alkanes, hopanes and steranes.•Full elution was achieved in two successive fractions, using small volumes of n-hexane and n-hexane/dichloromethane to recover all target substances.•To maximize analytical response, optimization was applied for three operating parameters in ECNI mode: i) ion source temperature; ii) emission current; and iii) electron energy.

Evidence of stockpile contamination for legacy polychlorinated biphenyls and organochlorine pesticides in the urban environment of Cyprus (Eastern Mediterranean): Influence of meteorology on air level variability and gas/particle partitioning based on equilibrium and steady-state models.

The present study investigated comprehensively the atmospheric occurrence and fate of an extensive range of polychlorinated biphenyls (PCBs; forty-two congeners), organochlorine pesticides (OCPs; twenty-seven emerging and legacy agrochemicals) and polycyclic aromatic hydrocarbons (PAHs; fifty parent and alkylated members, including the non USEPA-16 listed toxic ones), in both gas and particulate phase of the scarcely monitored atmosphere over Cyprus for the first time. Parent-metabolite concentration ratios suggested fresh application for dichlorodiphenyl-trichloroethanes (DDTs), dicofol, hexachlorocyclohexanes, endosulfan and chlorothalonil, particularly during spring (April-May). Regressions of logarithms of partial pressure against ambient temperature revealed that secondary recycling from contaminated terrestrial surfaces regulates the atmospheric level variability of PCBs, DDTs, aldrin, chlordane, dicofol, heptachlor and endosulfan. Enthalpies of surface-air exchange (∆HSA) calculated from Clausius-Clapeyron equations were significantly correlated to vaporization enthalpies (∆HV) determined by chromatographic techniques, corroborating presence of potential stockpile-contaminated sites around the study area. The Harner-Bidleman equilibrium model simulating urban areas, and the Li-Jia empirical model, predicted better the partitioning behavior of PAHs (

Urban environment and health behaviours in children from six European countries.

BACKGROUND: Urban environmental design is increasingly considered influential for health and wellbeing, but evidence is mostly based on adults and single exposure studies. We evaluated the association between a wide range of urban environment characteristics and health behaviours in childhood. METHODS: We estimated exposure to 32 urban environment characteristics (related to the built environment, traffic, and natural spaces) for home and school addresses of 1,581 children aged 6-11 years from six European cohorts. We collected information on health behaviours including total amount of overall moderate-to-vigorous physical activity, physical activity outside school hours, active transport, sedentary behaviours and sleep duration, and developed patterns of behaviours with principal component analysis. We used an exposure-wide association study to screen all exposure-outcome associations, and the deletion-substitution-addition algorithm to build a final multi-exposure model. RESULTS: In multi-exposure models, green spaces (Normalized Difference Vegetation Index, NDVI) were positively associated with active transport, and inversely associated with sedentary time (22.71 min/day less (95 %CI -39.90, -5.51) per interquartile range increase in NDVI). Residence in densely built areas was associated with more physical activity and less sedentary time, and densely populated areas with less physical activity outside school hours and more sedentary time. Presence of a major road was associated with lower sleep duration (-4.80 min/day (95 %CI -9.11, -0.48); compared with no major road). Results for the behavioural patterns were similar. CONCLUSIONS: This multicohort study suggests that areas with more vegetation, more building density, less population density and without major roads are associated with improved health behaviours in childhood.

Two-year systematic investigation reveals alterations induced on chemical and bacteriome profile of PM2.5 by African dust incursions to the Mediterranean atmosphere.

PM2.5 atmospheric samples were regularly collected between January 2013 and March 2015 at a central location of Eastern Mediterranean (Island of Crete) during African dust events (DES) and periods of absence of such episodes as controls (CS). The elemental composition and microbiome DES and CS were thoroughly investigated. Fifty-six major and trace elements were determined by inductively coupled plasma-mass spectrometry. Relative mass abundances (RMA) of major crustal elements and lanthanoids were higher in DES than in CS. Conversely in CS, RMAs were higher for most anthropogenic transition metals. Lanthanum-to-other lanthanoids concentration ratios for DES approached the corresponding reference values for continental crust and several African dust source regions, while in CS they exceeded these values. USEPA's UNMIX receptor model, applied in all PM2.5 samples, established that African dust is the dominant contributing source (by 80%) followed by road dust/fuel oil emissions (17%) in the receptor area. Potential source contribution function (PSCF) identified dust hotspots in Tunisia, Libya and Egypt. The application of 16S rRNA gene amplicon sequencing revealed high variation of bacterial composition and diversity between DES and CS samples. Proteobacteria, Actinobacteria and Bacteroides were the most dominant in both DES and CS samples, representing ~88% of the total bacterial diversity. Cutibacterium, Tumebacillus and Sphingomonas dominated the CS samples, while Rhizobium and Brevundimonas were the most prevalent genera in DES. Mutual exclusion/co-occurrence network analysis indicated that Sphingomonas and Chryseobacterium exhibited the highest degrees of mutual exclusion in CS, while in DES the corresponding species were Brevundimonas, Delftia, Rubellimicrobium, Flavobacterium, Blastococcus, and Pseudarthrobacter. Some of these microorganisms are emerging global opportunistic pathogens and an increase in human exposure to them as a result of environmental changes, is inevitable.

Atmospheric particle-bound polycyclic aromatic hydrocarbons, n-alkanes, hopanes, steranes and trace metals: PM2.5 source identification, individual and cumulative multi-pathway lifetime cancer risk assessment in the urban environment.

The occurrence of atmospheric fine particles (PM2.5)-associated polycyclic aromatic hydrocarbons (PAHs), trace metals and organic molecular markers was investigated by conducting an intensive sampling campaign at the Eastern Mediterranean urban area of Nicosia (Cyprus). Sixty-two 24-hr PM2.5 samples were collected and analyzed for fifty parent and alkylated PAHs, twenty-five long chain n-alkanes, seventeen hopanes and twelve steranes used for source apportionment. The same number and kind of samples were analyzed to determine twenty-eight trace metals. Emphasis was given to investigate the air levels of the scarcely monitored although highly carcinogenic PAHs such as dibenzopyrenes, dibenzoanthracenes, 7H-benzo[c]fluorene and 5-methyl-chrysene, not included in the USEPA's sixteen PAH priority list (USEPA-16). UNMIX receptor model was applied to apportion the sources of atmospheric emissions of the determined organic compounds and trace metals and evaluate their daily contributions to the corresponding PM2.5 associated concentrations. For comparison purposes, principal component analysis with multiple linear regression (PCA/MLR) was also applied and its results are reported. The UNMIX receptor model, compared to PCA/MLR, offered a more precise source profile and more reliable daily mass source distributions by eliminating negative contributions. The individual and cumulative multi-pathway lifetime cancer risk (posed via inhalation, ingestion and dermal contact) by exposure to PM2.5-associated USEPA-16 listed and non-listed PAHs and selected airborne trace metals (As, Cd, Co, Ni, and Pb) were assessed. To estimate the contribution of each emission source to the total cancer risk, multiple linear regression analysis was performed, using as independent variables the daily source mass contributions and as dependent variables the respective cancer risk units. The estimated total cumulative cancer risk comprising all toxic PAHs, besides those included in the priority list, and metals was higher than the USEPA's threshold by a factor of eight, denoting a potential risk for long-term exposure of a population in the urban environment.

Study of the occurrence and multi-pathway health risk assessment of regulated and unregulated disinfection by-products in drinking and swimming pool waters of Mediterranean cities.

The occurrence of a wide variety of regulated (four trihalomethanes (THM(4)), five haloacetic acids (HAA(5))) and unregulated (haloacetonitriles (HANs), halogenated ketones, chloropicrin, carbon tetrachloride, and other haloacetic acids) disinfection by-products (DBPs) was studied, in two hundred twenty-six finished drinking water samples collected in Barcelona (Spain, between 2008 and 2009), Athens (Greece, 2009-2010), Heraklion (Greece, 2009-2010), Nicosia (Cyprus, 2012-2013), and Limassol (Cyprus, 2011). The samples were analyzed by using liquid-liquid extraction, gas chromatography coupled with an electron capture detector or negative chemical ionization mass spectrometry. In addition, fourteen swimming pool water samples (from Heraklion and Athens) were also investigated regarding their DBPs content. The studied DBPs were determined concurrently with pH, total organic carbon (TOC), and bromide. Spearman's statistical analysis has shown statistically significant (p < 0.001) weak correlations between TOC, THM(4), HANs and HAA(5) but a strong correlation between THM(4) and HANs. Principal component analysis (PCA) on THM(4), HANs and HAA(5) provided a clear differentiation between the examined drinking waters, on the basis of their DBPs content. In the drinking water of coastal cities, the brominated DBPs dominated over the chlorinated ones, due to the higher bromide concentrations occurring in the corresponding raw waters. Lifetime cancer risk and hazard quotient by exposure to four THMs, dichlorocetic acid and trichloroacetic acid in drinking water and indoor swimming pools through multiple pathways were estimated. Total cancer unit risks in drinking water for Nicosia, Barcelona, Limassol and Athens exceeded in most cases the US EPA's regulatory threshold (1.0E-06). The total lifetime cancer risk evaluated for the studied indoor swimming pools was above the US EPA's negligible level for male, female, and junior swimmers.

Source apportionment of organic pollutants in fine and coarse atmospheric particles in Doha, Qatar.

In this study, we investigated the sources of organic pollutants associated with fine (PM2.5) and coarse (PM2.5-10) atmospheric particulate matter in Doha, Qatar based on an eight-month sampling campaign conducted from May to December 2015. Multiple organic compound tracers including 36 PAH members, 25 n-alkane homologs, 17 hopanes, and 12 steranes were used for organic aerosols source apportionment. Source apportionment based on specific molecular markers, molecular diagnostic ratios/indices, and positive matrix factorization (PMF) modeling showed that similar sources are responsible for both fine- and coarse-particle organic pollutants. PMF analysis showed that biogenic aerosols, fugitive dust emissions, gasoline engine emissions, diesel engine emissions, and heavy oil combustion were the five main pollution sources of organic aerosols, which agreed well with the results from the diagnostic ratios analysis. The conditional bivariate probability functions (CPF) and potential source contribution function (PSCF) indicated that both regional (i.e., mixed biogenic/secondary particles and oil refinery/shipping emissions) and local sources contributed to airborne organic aerosol concentrations observed at the site, depending on the wind speed and direction. It appears that the relatively high levels of organic pollutants were contributed by local anthropogenic sources, such as fossil fuel combustion, vehicular emissions, and fugitive dust emissions. The high levels of local contributions indicated that there might be great opportunities for Qatar to considerably reduce emissions so that population exposures to carbonaceous aerosols and the public health risks associated with air pollution can be minimized. Implications: Multiple organic tracers and various source apportionment techniques have been used for convincing source apportionment. It was found that both long-range and local sources have a significant impact on atmospheric carbonaceous particles in the area, depending on the wind conditions. Relatively high levels of organic pollutants attributed to local anthropogenic sources indicate that there are great opportunities for Qatar to establish and implement more efficient pollution control measures and policies. Regional sources such as petroleum refineries and shipping-vessels emissions in the Gulf region should also be regulated and managed through regional cooperation to improve the air quality in the region.

Concentrations of aliphatic and polycyclic aromatic hydrocarbons in ambient PM2.5 and PM10 particulates in Doha, Qatar.

Organic carbon (OC), elemental carbon (EC), and 90 organic compounds (36 polycyclic aromatic hydrocarbons [PAHs], 25 n-alkane homologues, 17 hopanes, and 12 steranes) were concurrently quantified in atmospheric particulate matter of PM2.5 and PM10. The 24-hr PM samples were collected using Harvard Impactors at a suburban site in Doha, Qatar, from May to December 2015. The mass concentrations (mean ± standard deviation) of PM2.5 and PM10 were 40 ± 15 and 145 ± 70 µg m-3, respectively, exceeding the World Health Organization (WHO) air quality guidelines. Coarse particles comprised 70% of PM10. Total carbonaceous contents accounted for 14% of PM2.5 and 10% of PM10 particulate mass. The major fraction (90%) of EC was associated with the PM2.5. In contrast, 70% of OC content was found in the PM2.5-10 fraction. The secondary OC accounted for 60-68% of the total OC in both PM fractions, indicating photochemical conversions of organics are much active in the area due to higher air temperatures and solar radiations. Among the studied compounds, n-alkanes were the most abundant group, followed by PAHs, hopanes, and steranes. n-Alkanes from C25 to C35 prevailed with a predominance of odd carbon numbered congeners (C27-C31). High-molecular-weight PAHs (5-6 rings) also prevailed, within their class, with benzo[b + j]fluoranthene (Bb + jF) being the dominant member. PAHs were mainly (80%) associated with the PM2.5 fraction. Local vehicular and fugitive emissions were predominant during low-speed southeasterly winds from urban areas, while remote petrogenic/biogenic emissions were particularly significant under prevailing northwesterly wind conditions. Implications: An unprecedented study in Qatar established concentration profiles of EC, OC, and 90 organic compounds in PM2.5 and PM10. Multiple tracer organic compounds for each source can be used for convincing source apportionment. Particle concentrations exceeded WHO air quality guidelines for 82-96% of the time, revealing a severe problem of atmospheric PM in Doha. Dominance of EC and PAHs in fine particles signifies contributions from combustion sources. Dependence of pollutants concentrations on wind speed and direction suggests their significant temporal and spatial variability, indicating opportunities for improving the air quality by identifying sources of airborne contaminants.

Overcoming the plasticity of plant specialized metabolism for selective diterpene production in yeast.

Plants synthesize numerous specialized metabolites (also termed natural products) to mediate dynamic interactions with their surroundings. The complexity of plant specialized metabolism is the result of an inherent biosynthetic plasticity rooted in the substrate and product promiscuity of the enzymes involved. The pathway of carnosic acid-related diterpenes in rosemary and sage involves promiscuous cytochrome P450s whose combined activity results in a multitude of structurally related compounds. Some of these minor products, such as pisiferic acid and salviol, have established bioactivity, but their limited availability prevents further evaluation. Reconstructing carnosic acid biosynthesis in yeast achieved significant titers of the main compound but could not specifically yield the minor products. Specific production of pisiferic acid and salviol was achieved by restricting the promiscuity of a key enzyme, CYP76AH24, through a single-residue substitution (F112L). Coupled with additional metabolic engineering interventions, overall improvements of 24 and 14-fold for pisiferic acid and salviol, respectively, were obtained. These results provide an example of how synthetic biology can help navigating the complex landscape of plant natural product biosynthesis to achieve heterologous production of useful minor metabolites. In the context of plant adaptation, these findings also suggest a molecular basis for the rapid evolution of terpene biosynthetic pathways.

Spatial variation of PM elemental composition between and within 20 European study areas--Results of the ESCAPE project.

An increasing number of epidemiological studies suggest that adverse health effects of air pollution may be related to particulate matter (PM) composition, particularly trace metals. However, we lack comprehensive data on the spatial distribution of these elements. We measured PM2.5 and PM10 in twenty study areas across Europe in three seasonal two-week periods over a year using Harvard impactors and standardized protocols. In each area, we selected street (ST), urban (UB) and regional background (RB) sites (totaling 20) to characterize local spatial variability. Elemental composition was determined by energy-dispersive X-ray fluorescence analysis of all PM2.5 and PM10 filters. We selected a priori eight (Cu, Fe, K, Ni, S, Si, V, Zn) well-detected elements of health interest, which also roughly represented different sources including traffic, industry, ports, and wood burning. PM elemental composition varied greatly across Europe, indicating different regional influences. Average street to urban background ratios ranged from 0.90 (V) to 1.60 (Cu) for PM2.5 and from 0.93 (V) to 2.28 (Cu) for PM10. Our selected PM elements were variably correlated with the main pollutants (PM2.5, PM10, PM2.5 absorbance, NO2 and NOx) across Europe: in general, Cu and Fe in all size fractions were highly correlated (Pearson correlations above 0.75); Si and Zn in the coarse fractions were modestly correlated (between 0.5 and 0.75); and the remaining elements in the various size fractions had lower correlations (around 0.5 or below). This variability in correlation demonstrated the distinctly different spatial distributions of most of the elements. Variability of PM10_Cu and Fe was mostly due to within-study area differences (67% and 64% of overall variance, respectively) versus between-study area and exceeded that of most other traffic-related pollutants, including NO2 and soot, signaling the importance of non-tailpipe (e.g., brake wear) emissions in PM.

Development of land use regression models for particle composition in twenty study areas in Europe.

Land Use Regression (LUR) models have been used to describe and model spatial variability of annual mean concentrations of traffic related pollutants such as nitrogen dioxide (NO2), nitrogen oxides (NOx) and particulate matter (PM). No models have yet been published of elemental composition. As part of the ESCAPE project, we measured the elemental composition in both the PM10 and PM2.5 fraction sizes at 20 sites in each of 20 study areas across Europe. LUR models for eight a priori selected elements (copper (Cu), iron (Fe), potassium (K), nickel (Ni), sulfur (S), silicon (Si), vanadium (V), and zinc (Zn)) were developed. Good models were developed for Cu, Fe, and Zn in both fractions (PM10 and PM2.5) explaining on average between 67 and 79% of the concentration variance (R(2)) with a large variability between areas. Traffic variables were the dominant predictors, reflecting nontailpipe emissions. Models for V and S in the PM10 and PM2.5 fractions and Si, Ni, and K in the PM10 fraction performed moderately with R(2) ranging from 50 to 61%. Si, NI, and K models for PM2.5 performed poorest with R(2) under 50%. The LUR models are used to estimate exposures to elemental composition in the health studies involved in ESCAPE.

Evaluation of land use regression models for NO2 and particulate matter in 20 European study areas: the ESCAPE project.

Land use regression models (LUR) frequently use leave-one-out-cross-validation (LOOCV) to assess model fit, but recent studies suggested that this may overestimate predictive ability in independent data sets. Our aim was to evaluate LUR models for nitrogen dioxide (NO2) and particulate matter (PM) components exploiting the high correlation between concentrations of PM metrics and NO2. LUR models have been developed for NO2, PM2.5 absorbance, and copper (Cu) in PM10 based on 20 sites in each of the 20 study areas of the ESCAPE project. Models were evaluated with LOOCV and "hold-out evaluation (HEV)" using the correlation of predicted NO2 or PM concentrations with measured NO2 concentrations at the 20 additional NO2 sites in each area. For NO2, PM2.5 absorbance and PM10 Cu, the median LOOCV R(2)s were 0.83, 0.81, and 0.76 whereas the median HEV R(2) were 0.52, 0.44, and 0.40. There was a positive association between the LOOCV R(2) and HEV R(2) for PM2.5 absorbance and PM10 Cu. Our results confirm that the predictive ability of LUR models based on relatively small training sets is overestimated by the LOOCV R(2)s. Nevertheless, in most areas LUR models still explained a substantial fraction of the variation of concentrations measured at independent sites.