Search

Abstract Salient object detection (SOD), one of the most important applications in the field of computer vision, aims to extract the most visually appealing regions of scenes. However, the improvement of the accuracy of existing salient object detection in optical remote sensing images (ORSI‐SOD) is usually accompanied by an increase of network complexity, which affects the application of these models. Motivated by this, a novel lightweight edge‐supervised neural network for ORSI‐SOD is proposed, named CSFFNet. Specifically, the backbone (ResNet34) is first lightened by feature encoding module (FEM), building a lightweight subnet for feature extraction. Then, in the transformer‐based feature pyramid enhancement module (FPEM), the convolutional features obtained in the FEM are enhanced by long‐distance dependence to obtain multi‐scale features containing rich saliency cues. Based on this, the feature fusion module (FFM) is designed to capture cross‐scale long‐range dependencies and effectively fuse high‐level semantic information with low‐level detail information. Thus, the increase in network complexity due to multi‐level decoding is avoided. Finally, the segmentation results are optimized by using salient edges as auxiliary information, which effectively improves the contrast and completeness of the results. Experimental results on two public datasets demonstrate that the lightweight CSFFNet achieves competitive or even better performance compared with state‐of‐the‐art methods.

To comprehensively understand the explosion risk in underground energy transportation tunnels, this study employed computational fluid dynamics technology and finite element simulation to numerically analyze the potential impact of an accidental explosion for a specific oil and gas pipeline in China and the potential damage risk to nearby buildings. Furthermore, the study investigated the effects of tunnel inner diameter (d = 4.25 m, 6.5 m), tunnel length (L = 4 km, 8 km, 16 km), and soil depth (primarily Lsoil = 20 m, 30 m, 40 m) on explosion dynamics and on structural response characteristics. The findings indicated that as the tunnel length and inner diameter increased, the maximum explosion overpressure gradually rose and the peak arrival time was delayed, especially when d = 4.25 m; with the increase in L, the maximum explosion overpressure rapidly increased from 1.03 MPa to 2.12 MPa. However, when d = 6.5 m, the maximum explosion overpressure increased significantly by 72.8% from 1.25 MPa. Evidently, compared to the change in tunnel inner diameter, tunnel length has a more significant effect on the increase in explosion risk. According to the principle of maximum explosion risk, based on the peak explosion overpressure of 2.16 MPa under various conditions and the TNT equivalent calculation formula, the TNT explosion equivalent of a single section of the tunnel was determined to be 1.52 kg. This theoretical result is further supported by the AUTODYN 15.0 software simulation result of 2.39 MPa (error < 10%). As the soil depth increased, the distance between the building and the explosion source also increased. Consequently, the vibration peak acceleration and velocity gradually decreased, and the peak arrival time was delayed. In comparison to a soil depth of 10 m, the vibration acceleration at soil depths of 20 m and 30 m decreased by 81.3% and 91.7%, respectively. When the soil depth was 10 m, the building was at critical risk of vibration damage.

Abstract Background Few studies have assessed air pollution exposure association with birthweight during both preconception and gestational periods. Methods Leveraging a preconception cohort consisting of 14220 pregnant women and newborn children in Shanghai, China during 2016–2018, we aim to assess associations of NO2 and PM2.5 exposure, derived from high-resolution spatial-temporal models, during preconception and gestational periods with outcomes including term birthweight, birthweight Z-score, small-for-gestational age (SGA) and large-for-gestational age (LGA). Linear and logistic regressions were used to estimate 3-month preconception and trimester-averaged air pollution exposure associations; and distributed lag models (DLM) were used to identify critical exposure windows at the weekly resolution from preconception to delivery. Two-pollutant models and children’s sex-specific associations were explored. Results After controlling for covariates, one standard deviation (SD) (11.5 μg/m3, equivalent to 6.1 ppb) increase in NO2 exposure during the second and the third trimester was associated with 13% (95% confidence interval: 2 – 26%) and 14% (95% CI: 1 – 29%) increase in SGA, respectively; and one SD (9.6 μg/m3) increase in PM2.5 exposure during the third trimester was associated with 15% (95% CI: 1 – 31%) increase in SGA. No association have been found for outcomes of birthweight, birthweight Z-score and LGA. DLM found that gestational weeks 22–32 were a critical window, when NO2 exposure had strongest associations with SGA. The associations of air pollution exposure tended to be stronger in female newborns than in male newborns. However, no significant associations of air pollution exposure during preconception period on birthweight outcomes were found. Conclusion Consistent with previous studies, we found that air pollution exposure during mid-to-late pregnancy was associated with adverse birthweight outcomes.

Increasing evidences have revealed a close relationship between various environmental exposures and reproductive health. The real-world complex exposure scenario along with intricate interactions effects of these exposures has high demands of exposome-wide association studies for human risk assessment. With the development of exposomic studies, applying hair exposome to link environmental factors and reproductive health outcomes has obvious potential advantages. However, the reliability of utilizing hair to characterize exposome is always of high concern for researchers. In this review, we briefly summarized studies about the effects of environmental exposures on several typical reproductive health outcomes, described state-of-the-art situation of applying hair exposome for reproductive health risk assessment, elucidated the advantages (e.g., containing abundant exposure factors, low expenses for sampling, and tracing exposure history), and proposed the unsolved issues in this area. We also conducted a comprehensive discussion about the reliability of various hair biomarkers and primarily established the biomarker databases. Finally, a standardized framework on using hair exposome for reproductive health study was proposed to provide fundamental approach for future studies. We concluded that it is promising to apply hair exposome to evaluate the effects of the environmental factors on human reproductive health.

BackgroundMedical Waste (MW), conceptualized as waste generated in the diagnosis, treatment, or immunization of human beings or animals, posing massive threat to public health. Environment-friendly public attitudes promotes the shaping of pro-environmental behavior. However, the public attitudes of MW and the potential determinants remained scarce. The present study aims to reveal globally public attitudes towards MW and captured the determinants.MethodsWe integrated the crawler technology with sentiment analysis to captured the public attitudes toward MW across 141 specific countries from 3,789,764 related tweets. Multiple cross-national databases were integrated to assess characteristics including risk, resistance, environment, and development. The spatial regression model was taken to counterbalence the potential statistical bias.ResultsOverall, the global public attitudes towards MW were positive, and varied significantly across countries. Resilience (β = 0.78, SD = 0.14, P < 0.01) and development (β = 1.66, SD = 0.13, P < 0.01) posed positive influence on public attitudes towards MW, meanwhile, risk (β = -0.1, SD = 0.12, P > 0.05) and environment (β = 0.09, SD = 0.09, P > 0.05) were irrelated to the shaping of positive MW public attitudes. Several positive moderating influences was also captured. Additionally, the cross-national disparities of the determiants were also captured, more specific, public attitudes towards MW in extremely poor areas were more likely to be negatively affected by risks, resilience and development.ConclusionsThis study focused mainly on the public attitudes as well as captured the potential determinants. Public attitudes towards MW were generally positive, but there were large cross-national disparities. Stakeholders would need to designate targeted strategies to enhance public satisfaction with MW management.

Abstract Environmental exposure to ambient polycyclic aromatic hydrocarbons (PAHs) can disturb the immune response. However, the evidence on adverse health effects caused by exposure to PAHs emitted from specific sources among different vulnerable subpopulations is limited. In this cross‐sectional study, we aimed to evaluate whether exposure to source‐specific PAHs could increase systemic inflammation in older adults. The present study included community‐dwelling older adults and collected filter samples of personal exposure to PM2.5 during the winter of 2011. Blood samples were collected after the PM2.5 sample collection. We analyzed PM2.5 bound PAHs and serum inflammatory cytokines (interleukin (IL)1β, IL6, and tumor necrosis factor alpha levels. The Positive Matrix Factorization model was used to identify PAH sources. We used a linear regression model to assess the relative effects of source‐specific PM2.5 bound PAHs on the levels of measured inflammatory cytokines. After controlling for confounders, exposure to PAHs emitted from biomass burning or diesel vehicle emission was significantly associated with increased serum inflammatory cytokines and systemic inflammation. These findings highlight the importance of considering exposure sources in epidemiological studies and controlling exposures to organic materials from specific sources.

Various geostatistical models have been used in epidemiological research to evaluate ambient air pollutant exposures at a fine spatial scale. Few studies have investigated the performance of different exposure models on population-weighted exposure estimates and the resulting potential misclassification across various modeling approaches. This study developed spatial models for NO2 and PM2.5 and conducted exposure assessment in Beijing, China. It explored three spatial modeling approaches: variable dimension reduction, machine learning, and conventional linear regression. It compared their model performance by cross-validation (CV) and population-weighted exposure estimates. Specifically, partial least square (PLS) regression, random forests (RF), and supervised linear regression (SLR) models were developed based on an ordinary kriging (OK) framework for NO2 and PM2.5 in Beijing, China. The mean squared error-based R2 (R2mse) and root mean squared error (RMSE) in leave-one site-out cross-validation (LOOCV) were used to evaluate model performance. These models were used to predict the ambient exposure levels in the urban area and to estimate the misclassification of population-weighted exposure estimates in quartiles between them. The results showed that the PLS-OK models for NO2 and PM2.5, with the LOOCV R2mse of 0.82 and 0.81, respectively, outperformed the other models. The population-weighted exposure to NO2 estimated by the PLS-OK and RF-OK models exhibited the lowest misclassification in quartiles. For PM2.5, the estimates of potential misclassification were comparable across the three models. It indicated that the exposure misclassification made by choosing different modeling approaches should be carefully considered, and the resulting bias needs to be evaluated in epidemiological studies.

Abstract Secondary organic aerosol (SOA) represents a major fraction of atmospheric fine particles. Both biogenic and anthropogenic volatile organic compounds (VOCs) can contribute to SOA through (photo-) oxidation. However, the current understanding of their combined, interactive effect on SOA formation and composition is still limited, challenging the accuracy in assessing global SOA budget, sources, and climate effect. Here we combine laboratory experiments and modelling to show that isoprene can suppress SOA formation from photo-oxidation of anthropogenic aromatics (toluene and p-xylene) with the presence of NO x , and similar SOA suppression phenomena are observed when replacing isoprene with propene. We find that the decreased SOA in such mixed-VOC conditions can be largely attributed to OH scavenging effect, resulting in reduced consumption of parent aromatics. However, various changes in SOA oxidation state (i.e., O/C) and oxidation pathways (i.e., more carbonyls formation) are observed following addition of isoprene, and the SOA chemical composition may not be similar to any single parent hydrocarbon, which implies the existence of complex interactions between the degradation chemistry for alkenes and aromatics. Under the conditions of this work, the OH scavenging effect is largely determined by gas-phase chemistry, which is expected to be widespread in binary or more complex systems in ambient air. More broadly, we infer that the global budget of anthropogenic SOA and its corresponding radiative forcing could be affected by biogenic emission of isoprene, particularly in urban environments with appreciable vegetation coverage.

Soil-derived dust emissions have been widely associated with health and environmental problems and should therefore be accurately and reliably estimated and assessed. Of these emissions, the inhalable PM10 and PM2.5 are difficult to estimate. Consequently, to calculate PM10 and PM2.5 emissions from soil erosion, an approach based on an artificial neural network (ANN) model which provides a multilayered, fully connected framework that relates input parameters and outcomes was proposed in this study. Owing to the difficulty in obtaining the actual emissions of soil-derived PM10 and PM2.5 over a broad area, the PM10 and PM2.5 simulated results of the ANN model were compared with the published results simulated by the widely used wind erosion prediction system (WEPS) model. The PM10 and PM2.5 emission results, based on the WEPS, agreed well with the field data, with R2 values of 0.93 and 0.97, respectively, indicating the potential for using the WEPS results as a reference for training the ANN model. The calculated r, RMSE and MAE for the results simulated by the WEPS and ANN were 0.78, 3.37 and 2.31 for PM10 and 0.79, 1.40 and 0.91 for PM2.5, respectively, throughout Southern Xinjiang. The uncertainty of the soil-derived PM10 and PM2.5 emissions at a 95% CI was (−66–106%) and (−75–108%), respectively, in 2016. The results indicated that by using parameters that affect soil erodibility, including the soil pH, soil cation exchange capacity, soil organic content, soil calcium carbonate, wind speed, precipitation and elevation as input factors, the ANN model could simulate soil-derived particle emissions in Southern Xinjiang. The results showed that when the study domain was reduced from the entire Southern Xinjiang region to its five administrative divisions, the performance of the ANN improved, producing average correlation coefficients of 0.88 and 0.87, respectively, for PM10 and PM2.5. The performances of the ANN differed by study period, with the best result obtained during the sand period (March to May) followed by the nonheating (June to October) and heating periods (November to February). Wind speed, precipitation and soil calcium carbonate were the predominant input factors affecting particle emissions from wind erosion sources. The results of this study can be used as a reference for the wind erosion prevention and soil conservation plans in Southern Xinjiang.

Few studies have examined the association between greenness exposure and birth outcomes. This study aims to identify critical exposure time windows during preconception and pregnancy for the association between greenness exposure and birth weight. A cohort of 13 890 pregnant women and newborns in Shanghai, China from 2016–2019 were included in the study. We assessed greenness exposure using Normalized Difference Vegetation Index (NDVI) during the preconception and gestational periods, and evaluated the association with term birthweight, birthweight z -score, small-for-gestational age, and large-for-gestational age using linear and logistic regressions adjusting for key maternal and newborn covariates. Ambient temperature, relative humidity, ambient levels of fine particles (PM _2.5 ) and nitrogen dioxide (NO _2 ) assessed during the same period were adjusted for as sensitivity analyses. Furthermore, we explored the potential different effects by urbanicity and park accessibility through stratified analysis. We found that higher greenness exposure at the second trimester of pregnancy and averaged exposure during the entire pregnancy were associated with higher birthweight and birthweight Z -score. Specifically, a 0.1 unit increase in second trimester averaged NDVI value was associated with an increase in birthweight of 10.2 g (95% CI: 1.8–18.5 g) and in birthweight Z -score of 0.024 (0.003–0.045). A 0.1 unit increase in an averaged NDVI during the entire pregnancy was associated with 10.1 g (95% CI: 1.0–19.2 g) increase in birthweight and 0.025 (0.001–0.048) increase in birthweight Z -score. Moreover, the associations were larger in effect size among urban residents than suburban residents and among residents without park accessibility within 500 m compared to those with park accessibility within 500 m. Our findings suggest that increased greenness exposure, particularly during the second trimester, may be beneficial to birth weight in a metropolitan area.

Background: Evidence of maternal exposure to ambient air pollution on congenital heart defects (CHD) has been mixed and are still relatively limited in developing countries. We aimed to investigate the association between maternal exposure to air pollution and CHD in China. Method: This longitudinal, population-based, case-control study consecutively recruited fetuses with CHD and healthy volunteers from 21 cities, Southern China, between January 2006 and December 2016. Residential address at delivery was linked to random forests models to estimate maternal exposure to particulate matter with an aerodynamic diameter of ≤ 1 µm (PM1), ≤2.5 µm, and ≤10 µm as well as nitrogen dioxides, in three trimesters. The CHD cases were evaluated by obstetrician, pediatrician, or cardiologist, and confirmed by cardia ultrasound. The CHD subtypes were coded using the International Classification Diseases. Adjusted logistic regression models were used to assess the associations between air pollutants and CHD and its subtypes. Results: A total of 7055 isolated CHD and 6423 controls were included in the current analysis. Maternal air pollution exposures were consistently higher among cases than those among controls. Logistic regression analyses showed that maternal exposure to all air pollutants during the first trimester was associated with an increased odds of CHD (e.g., an interquartile range [13.3 µg/m3] increase in PM1 was associated with 1.09-fold ([95% confidence interval, 1.01–1.18]) greater odds of CHD). No significant associations were observed for maternal air pollution exposures during the second trimester and the third trimester. The pattern of the associations between air pollutants and different CHD subtypes was mixed. Conclusions: Maternal exposure to greater levels of air pollutants during the pregnancy, especially the first trimester, is associated with higher odds of CHD in offspring. Further longitudinal well-designed studies are warranted to confirm our findings.

Background: Experimental research suggests that fine particulate matter (PM2.5) exposure might affect embryonic development. However, only few population-based studies have investigated the impact of maternal exposure to PM2.5 on the early pregnancy loss. Objectives: To estimate associations between clinically recognized early pregnancy loss (CREPL) and exposure to ambient PM2.5 at individual residences during peri-conception periods, with the aim to identify susceptible exposure time windows. Methods: CREPL cases and normal early pregnancy controls (of similar age and gravidity presenting within one week, a total of 364 pairs) were recruited between July 2017 and July 2018 among women residing in Tianjin, China. Average ambient PM2.5 concentrations of ten exposure windows (4 weeks, 2 weeks and 1 week before conception; the first, second, third and fourth single week, the first and second 2-week periods, and the entire 4-week period after conception) at the women's residential addresses were estimated using temporally-adjusted land use regression models. Associations between PM2.5 exposures at specific peri-conception time windows and CREPL were examined using conditional logistic regression models, adjusted for covariates. Results: Based on adjusted models, CREPL was significantly associated with a 10 μg/m3 increase in PM2.5 exposure during the second week after conception (OR = 1.15; 95% CI: 1.04, 1.27; p = 0.005), independent of effects at other time windows. There was also an association of CREPL with PM2.5 during the entire 4-week period after conception (OR = 1.22; 95% CI: 1.02, 1.46; p = 0.027). There was little evidence for associations with exposure during pre-conception exposure windows. Conclusions: Maternal exposures to ambient PM2.5 during a critical time window following conception are associated with CREPL, with the second week after conception possibly being the exposure window of most vulnerability. Future studies should focus on replicating these findings and on pathogenic mechanisms. Keywords: Adverse pregnancy outcomes, Air pollution, Land use regression models, Miscarriage, Particulate matter

Although carbonyl compounds are a key species with atmospheric oxidation capacity, their concentrations and sources have not been sufficiently characterized in various atmospheres, especially in desert areas. In this study, atmospheric carbonyl compounds were measured from 16 May to 15 June 2018 in Tazhong in the central Taklimakan Desert, Xinjiang Uygur Autonomous Region, China. Concentrations, chemical compositions, and sources of carbonyl compounds were investigated and compared with those of different environments worldwide. The average concentration of total carbonyls during the sampling period was 11.79 ± 4.03 ppbv. Formaldehyde, acetaldehyde, and acetone were the most abundant carbonyls, with average concentrations of 6.08 ± 2.37, 1.68 ± 0.78, and 2.52 ± 0.68 ppbv, respectively. Strong correlations between formaldehyde and other carbonyls were found, indicating same or similar sources and sinks. A hybrid single-particle Lagrangian integrated trajectory was used to analyze 72 h back trajectories. The values of C1/C2 (formaldehyde to acetaldehyde, 3.22–4.59) and C2/C3 (acetaldehyde to propionaldehyde, 15.00–17.03) from different directions and distances of the trajectories were consistent with the characteristics of a remote area. Relative to various environments, the carbonyl concentration in the Tazhong desert site was lower than that in urban areas and higher than that in suburban and remote areas, implying contributions from local primary and secondary sources. The obtained data can be used to improve the source and sink estimation of carbonyls at the regional scale.

To better understand the source and health risk of metal elements in PM2.5, a field study was conducted from May to December 2018 in the central region of the Liaoning province, China, including the cities of Shenyang, Anshan, Fushun, Benxi, Yingkou, Liaoyang, and Tieling. 24 metal elements (Na, K, V, Cr, Mn, Co, Ni, Cu, Zn, As, Mo, Cd, Sn, Sb, Pb, Bi, Al, Sr, Mg, Ti, Ca, Fe, Ba, and Si) in PM2.5 were measured by ICP-MS and ICP-OES. They presented obvious seasonal variations, with the highest levels in winter and lowest in summer for all seven cities. The sum of 24 elements were ranged from to in these cities. The element mass concentration ratio was the highest in Yingkou in the spring (26.15%), and the lowest in Tieling in winter (3.63%). The highest values of elements in PM2.5 were mostly found in Anshan and Fushun among the studied cities. Positive matrix factorization (PMF) modelling revealed that coal combustion, industry, traffic emission, soil dust, biomass burning, and road dust were the main sources of measured elements in all cities except for Yingkou. In Yingkou, the primary sources were identified as coal combustion, metal smelting, traffic emission, soil dust, and sea salt. Health risk assessment suggested that Mn had non-carcinogenic risks for both adults and children. As for Cr, As, and Cd, there was carcinogenic risks for adults and children in most cities. This study provides a clearer understanding of the regional pollution status of industrial urban agglomeration.

Air pollution is an increasing threat to human health in China. In this study, daily PM10 and PM2.5 samples were collected simultaneously at the Jinan Environmental Monitoring Station (EMS)in Jinan, China from 15 November 2016 to 15 March 2017. The aim of this work was to improve the understanding of the characteristics and sources of air particles and determine different levels of PM2.5 pollution and its constituent elements, water-soluble ions and carbonaceous species. Nitrate (NO3−), organic materials (OM) and sulfate (SO42−) were identified as the three main components of PM2.5 pollution. With increasing pollution level, the contributions of SO42−, NO3− and NH4+ increased at greater rates, unlike that of OM. The proportion of SO42− exceeded that of NO3− and became predominant in severe PM2.5 pollution (SP; 250 μg m−3 ≤ PM2.5 ≤ 500 μg m−3). This work demonstrates that SO42− has a dominant role in SP level and, consequently, requires greater research attention. It is demonstrated that relative humidity (RH) enhances the rate of sulfate formation more than that of nitrate. Therefore, under the current Chinese emergency response measures, it is necessary to further reduce emissions of SO2 and NO2. Four clusters of backward trajectories identified dominant pollution vectors originating from highly industrialized areas that exacerbate the poor air quality in Jinan. It is, therefore, necessary to undertake regional control measures to reduce pollutant emissions.

As one of the biggest cities in North China, Jinan has been suffering heavy air pollution in recent decades. To better characterize the ambient particulate matter in Jinan during heavy pollution periods, we collected daily PM2.5 (particulate matter with aerodynamic diameters equal to or less than 2.5 μm) filter samples from 15 October 2017 to 31 January 2018 and analyzed their chemical compositions (including inorganic water-soluble ions (WSIs), carbonaceous species, and inorganic elements). The daily average concentration of PM2.5 was 83.5 μg/m3 during the sampling period. A meteorological analysis revealed that both low wind speed and high relative humidity facilitated the occurrence of high PM2.5 pollution episodes. A chemical analysis indicated that high concentrations of water-soluble ions, carbonaceous species, and elements were observed during heavy pollution days. The major constituents of PM2.5 in Jinan were secondary aerosol particles and organic matter based on the results of mass closure. Chemical Mass Balance (CMB) was used to track possible sources and identified that nitrate, sulfate, vehicle exhaust and coal fly ash were the main contributors to PM2.5 during heavy pollution days in Jinan, accounting for 25.4%, 18.6%, 18.2%, and 13.3%, respectively.

A new structure bulk tobacco curing barn was presented. To study the temperature and humidity field in the new structure tobacco curing barn, a 3D transient computational fluid dynamics (CFD) model was developed using porous medium, species transport, κ-ε turbulence and discrete phase models. The CFD results demonstrated that (1) the temperature and relative humidity predictions were validated by the experimental results, and comparison of simulation results with experimental data showed a fairly close agreement; (2) the temperature of the bottom and inlet area was higher than the top and outlet area, and water vapor concentrated on the top and outlet area in the barn; (3) tobacco loading density and thickness of tobacco leaves had an explicit effect on the temperature distributions in the barn.

Ninety-six particulate matter (PM10) chemical source profiles for geological sources in typical cities of southwest China were acquired from Source Profile Shared Service in China. Twenty-six elements (Na, Mg, Al, Si, K, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, As, Se, Sr, Cd, Sn, Sb, Ba, Be, Tl and Pb), nine ions (F−, Cl−, SO42−, NO3−, Na+, NH4+, K+, Mg2+ and Ca2+), and carbon-containing species (organic carbon and elemental carbon) were determined to construct these profiles. Individual source profiles were averaged and compared to quantify similarities and differences in chemical abundances using the profile-compositing method. Overall, the major components of PM10 in geological sources were crustal minerals and undefined fraction. Different chemical species could be used as tracers for various types of geological dust in the region that resulted from different anthropogenic influence. For example, elemental carbon, V and Zn could be used as tracers for urban paved road dust; Al, Si, K+ and NH4+ for agricultural soil; Al and Si for natural soil; and SO42− for urban resuspended dust. The enrichment factor analysis showed that Cu, Se, Sr and Ba were highly enriched by human activities in geological dust samples from south-west China. Elemental ratios were taken to highlight the features of geological dust from south-west China by comparing with northern urban fugitive dust, loess and desert samples. Low Si/Al and Fe/Al ratios can be used as markers to trace geological sources from southwestern China. High Pb/Al and Zn/Al ratios observed in urban areas demonstrated that urban geological dust was influenced seriously by non-crustal sources.