Search

A recently developed open source tool kit named the Trajectory based Potential Source Apportionment (TraPSA) was used to identify the sources of respirable particulates at Kochi, India. Using 24-h average particulate matter data from samples collected at five regulatory monitoring stations over the five-year period from January 2011 to October 2016, local and regional scale analyses were made. Concentration field analysis was performed using back trajectories generated by Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model with inputs from atmospheric reanalysis data. Conditional bivariate probability function analyses were made using local meteorology data to identify the influence of local sources. Most of the stations indicated the contribution from local traffic activities during low wind conditions and from a nearby industrial area especially during high speed winds. Back trajectory analysis identified potential source areas in Kerala as well as in nearby state of Tamil Nadu as contributing to the air quality at Kochi. Arabian sea on the western side was also observed to be a potential source area for Kochi. The study demonstrated the utility of TrapSA as a tool for deriving information about the potential source areas affecting particulate matter mass concentrations.

Gaseous elemental mercury (GEM), gaseous oxidized mercury (GOM), and particle-bound mercury (PBM) were measured continuously in Rochester, NY (NY43) from January 2012 to December 2014. Continuous measurements of ozone (O3), sulfur dioxide (SO2), carbon monoxide (CO), nitrogen oxide (NO), nitrogen dioxide (NO2), particulate matter (PM2.5), and meteorological data were also made at this site. A principal component analysis (PCA) of the resulting 15 variables showed that the ambient mercury in Rochester was primarily produced by non-local sources in contrast to earlier studies that showed that local sources were present. Positive matrix factorization (PMF) analysis of the atmospheric mercury and other pollutant species concentrations showed that transport and atmospheric processes have become the major source of mercury in Rochester. Conditional bivariate probability function (CBPF) and potential source contribution function (PSCF) were used to identify local and distant mercury sources. The results in this study showed that the closure of a coal-fired power plant and promulgation of several fuel quality policies reduced local mercury emissions making long-distance transport the major source of mercury in Rochester.

Hexabromocyclododecane (HBCDD) is a hazardous, persistent, bioaccumlative brominated flame retardant. To investigate how its use has affected the Great Lakes, total HBCDD (∑HBCDD) concentrations and temporal trends in homogenized whole fish samples from the Great Lakes region (1978 to 2016) were determined. ∑HBCDD concentrations (ng/g ww) for each lake are Erie (0.49-2.60), Ontario (3.12-8.90), Michigan (3.91-9.01), Superior (5.69-13.1), and Huron (5.57-13.7). Early years (1978 to 1992) showed no significant trend. However, recent trends (2004 to 2016) suggest concentrations are increasing in Lakes Erie and Ontario, decreasing in Lakes Superior and Michigan, and not changing in Lake Huron. Decreasing trends for Lakes Superior and Michigan are likely the result of decreased usage of the compound globally, regionally, and locally. For the other lakes, increasing or zero trends are consistent with food web changes due to invasive species and climate change, which has caused more intense storms and less ice cover leading to increased sediment resuspension.

Measurements of ambient speciated mercury (Hg) concentrations and Hg wet deposition were made at two urban sites (Bronx, NY and Rochester, NY) and one rural site (Potsdam, NY) in New York State in 2013 and 2014 to: 1) determine the factors influencing Hg wet deposition concentrations, 2) identify the contribution of gaseous oxidized Hg (GOM) and particulate bound Hg (PBM) scavenging to Hg wet deposition concentrations, and 3) identify potential source areas associated with high concentration events. The Bronx had the highest mean gaseous elemental Hg (GEM) and GOM concentrations, Rochester had the highest mean PBM and the lowest GOM concentrations, and Potsdam had the lowest mean GEM and PBM concentrations. The annual volume weighted mean (VWM) Hg concentrations and Hg wet deposition fluxes in the Bronx, Rochester, and Potsdam were significantly different with mean values of 10.3 ± 8.16, 10.2 ± 9.06, and 5.07 ± 1.79 ngL−1 and 8.45 ± 0.64, 6.65 ± 0.21, and 5.25 ± 0.49 μg/m2 year−1, respectively. Hg wet deposition flux and precipitation depth were positively correlated at all three sites as were Hg concentration in precipitation and weekly GOM concentrations at the Bronx and Potsdam sites. Scavenging coefficients (SC) of 680, 630, 850 for GOM and 410, 320, and 410 for PBM at Bronx, Rochester, and Potsdam, respectively, suggest GOM is responsible for most of the scavenged Hg. Measured GOM and PBM concentrations were relatively constant before precipitation events and Hg concentrations in precipitation did not vary significantly during precipitation events implying the scavenging process mainly occurred in clouds. VWM Hg concentrations, monthly accumulated Hg flux, and SCs for GOM and PBM were higher at the urban sites and significantly different for non-snow and snow events. Local sources appeared more important at the rural site while regional sources affected high urban concentrations.

Long-term atmospheric mercury measurements at Underhill, VT (VT99), and Huntington Forest, NY (NY20), from 1992 to 2014 and 2005 to 2014, respectively, were used to determine concentration trends using Mann–Kendall’s tau test with Sen’s slope estimator. These data, measured generally downwind of large Hg sources in the Midwestern United States, provide the longest record of ambient Hg concentrations available in the United States. At VT99, concentrations of gaseous element mercury (GEM), gaseous oxidized mercury (GOM), and particle-bound mercury (PBM) declined at rates of −1.8, −3.2, and −6.7%/year, respectively. At NY20, GEM and GOM concentrations declined at rates of −1.6 and −7.8%/year, respectively; however, PBM concentrations increased at a rate of 2.0%/year, which is likely related to winter wood burning. A trajectory ensemble analysis using the potential source contribution function indicates the source locations associated with high mercury concentrations changed from Toronto–Buffalo and Pennsylva...

Polybrominated diphenyl ethers (PBDEs) were widely used as fire retardants and have been detected throughout the Great Lakes (GL) ecosystem. The concentration trends (after fish age normalization) of PBDEs in top predator fish (lake trout and walleye) of the GLs were determined from 1979 to 2016, which includes most of the period when PBDEs were manufactured and used in this region. The fish samples were collected by two national (U.S. and Canada) long-term monitoring and surveillance programs. Trends in total concentrations (age-normalized) of the five major PBDE congeners (BDE-47, 99, 100, 153, and 154) found in fish across all five lakes have varied over time. Significant increases were observed from 1990 to 2000 (16.3% per year). Rapidly decreasing concentrations (-19.5% per year) were found from 2000 to 2007. Since 2007, the decreasing trend has become smaller (less than -5.5% per year) and relatively unchanged from 2011 to 2015. BDE-47, the congener with the highest concentrations in lake trout, has decreased continuously (ranging from -6.7% to -16.2% per year) in all lakes except Lake Erie. This decrease can be associated with the voluntary and regulatory phase out of production and/or usage of PBDEs since 2000. However, it has been offset by recent (since 2007) increasing trends of the other four higher brominated BDE congeners, especially BDE-100 and 154. Production and usage of commercial penta- and octa- BDE mixtures containing primarily the five major PBDE congeners was discontinued in 2004 in the U.S.A. and 2008 in Canada. These results indicate increasing fish uptake and bioaccumulation of higher brominated BDE congeners may be related to the transformation of BDE-209 to lower brominated BDE compounds in the GL environment or food web. Considering the abundance of BDE-209 in existing products and sediment in GL region, the duration of the unchanging total PBDE concentration trend in GL fish could be longer than expected.

Following the phase-out of polybrominated diphenyl ether (PBDE) and hexabromocyclododecane (HBCDD) flame retardants (FRs) from North American markets, the use of alternative FRs has increased. In this study the occurrence and spatiotemporal distributions of 18 dechlorane analogues (collectively referred to as DECs) and 20 alternative brominated FRs (referred to as ABFRs, i.e., brominated FRs other than PBDEs and HBCDD) were investigated in top predator fish megacomposites (i.e., a composite of all 50 fish) collected yearly from each of the Great Lakes from 2004 to 2016. Frequently detected substances include dechlorane 602, 603, 604 Component B, anti- and syn-dechlorane plus, and chlordene plus, as well as several brominated benzene FRs (i.e., hexabromobenzene, pentabromotoluene, and tetrabromo- o-chlorotoluene). Concentrations of ΣDECs and ΣABFRs ranged from 0.33-31.9 ng/g lipid weight (lw) (0.01-8.3 ng/g wet weight or ww) and 0.91-54.7 ng/g lw (0.09-7.1 ng/g ww), respectively. Flame retardant contamination exhibited chemical-specific spatial variations across the five lakes. Concentrations of ΣABFRs in Lake Erie fish were generally lower than those from other lakes. By contrast, fish ΣDEC concentrations were highest in Lake Ontario and the composition of dechlorane analogues differed significantly between Lake Ontario and the other lakes, indicating likely point-source influences. Temporal analyses revealed declining trends of ΣDECs and ΣABFRs in most lakes except Lake Erie, with age corrected trend slopes of -13.5% to -8.8% and -20.1% to -7.0% per year, respectively.

The water quality prediction performance of machine learning models may be not only dependent on the models, but also dependent on the parameters in data set chosen for training the learning models. Moreover, the key water parameters should also be identified by the learning models, in order to further reduce prediction costs and improve prediction efficiency. Here we endeavored for the first time to compare the water quality prediction performance of 10 learning models (7 traditional and 3 ensemble models) using big data (33,612 observations) from the major rivers and lakes in China from 2012 to 2018, based on the precision, recall, F1-score, weighted F1-score, and explore the potential key water parameters for future model prediction. Our results showed that the bigger data could improve the performance of learning models in prediction of water quality. Compared to other 7 models, decision tree (DT), random forest (RF) and deep cascade forest (DCF) trained by data sets of pH, DO, CODMn, and NH3-N had significantly better performance in prediction of all 6 Levels of water quality recommended by Chinese government. Moreover, two key water parameter sets (DO, CODMn, and NH3-N; CODMn, and NH3-N) were identified and validated by DT, RF and DCF to be high specificities for perdition water quality. Therefore, DT, RF and DCF with selected key water parameters could be prioritized for future water quality monitoring and providing timely water quality warning.

Concentration patterns and temporal trends of legacy persistent, bioaccumulative and toxic (PBT) contaminants were determined using the Great Lake Fish Monitoring and Surveillance Program (GLFMSP) top predator fish data from 1999 to 2014 and applying Kendall-Theil robust regression after cluster-based age normalization. For most Great Lakes sites, significant decreasing concentration trends ranging from -4.1% to -21.6% per year (with the only exception being mirex in Lake Erie walleye) were found for PBTs including polychlorinated biphenyls (PCBs), dichloro-diphenyl-trichlorethane (DDTs), dieldrin, endrin, chlordane, oxychlordane, nonachlor, mirex, and hexachlorobenzene (HCB) reflecting the successful historical and ongoing reduction of fugitive releases and remediation efforts in the U.S. and Canada including physical removal (dredging) coupled with sediment sequestration. Generally, lower concentrations and faster decreasing trends are observed in western/northern sampling sites compared to eastern/southern sites as the former sites are generally more remote from population centers and industrial activities. PCBs, which can be released from ongoing sources, have the highest concentration, the second slowest decreasing trend, and increasing mass fractions of the contaminants studied suggesting that they will continue to be the legacy contaminant of greatest concern into the future.

Polyhalogenated carbazoles (PHCZs) were recently discovered in Great Lakes sediment and other aquatic systems. However, knowledge about their bioaccumulation and potential risks to fish and wildlife remains very limited. The present study investigated PHCZs in Great Lakes lake trout ( Salvelinus namaycush) and walleye ( Sander vitreus; Lake Erie only) composites collected between 2004 and 2016. Median concentrations of ∑PHCZs by lake ranged from 54.7 to 154 ng/g lipid weight or lw (6.8-28.0 ng/g wet weight). Dominant congeners included 3,6-dichlorocarbazole, 1,3,6-tribromocarbazole, and 1,3,6,8-tetrachlorocarbazole. The highest ∑PHCZs concentrations were found in Lakes Michigan and Ontario fish, followed by Lake Huron, whereas Lakes Erie and Superior fish contained the lowest concentrations. Congener profiles of PHCZs also exhibited spatial variations. After age normalization to minimize fish age influence on bioaccumulation rates, fish ∑PHCZs' concentrations declined significantly over time in all lakes except Lake Erie, with slopes ranging from -10.24% to -3.85% per year. The median toxic equivalent (TEQ) of PHCZs due to their dioxin-like activity was determined to range from 8.7 to 25.7 pg/g lw in Great Lakes fish. This study provides the first insight into the bioaccumulation and spatiotemporal trends of PHCZs in Great Lakes and suggests the need for further research on this group of chemicals.

Light-absorbing, atmospheric particles have gained greater attention in recent years because of their direct and indirect impacts on regional and global climate. Atmospheric black carbon (BC) aerosol is a leading climate warming agent, yet uncertainties in the global direct aerosol radiative forcing remain large. Based on a year of aerosol absorption measurements at seven wavelengths, BC concentrations were investigated in Dhanbad, the coal capital of India. Coal is routinely burned for cooking and residential heat as well as in small industries. The mean daily concentrations of ultraviolet-absorbing black carbon measured at 370nm (UVBC) and black carbon measured at 880nm (BC) were 9.8±5.7 and 6.5±3.8μgm-3, respectively. The difference between UVBC and BC, Delta-C, is an indicator of biomass or residential coal burning and averaged 3.29±4.61μgm-3. An alternative approach uses the Ǻngstrom Exponent (AE) to estimate the biomass/coal and traffic BC concentrations. Biomass/coal burning contributed ~87% and high temperature, fossil-fuel combustion contributed ~13% to the annual average BC concentration. The post-monsoon seasonal mean UVBC values were 10.9μgm-3 and BC of 7.2μgm-3. Potential source contribution function analysis showed that in the post-monsoon season, air masses came from the central and northwestern Indo-Gangetic Plains where there is extensive agricultural burning. The mean winter UVBC and BC concentrations were 15.0 and 10.1μgm-3, respectively. These higher values were largely produced by local sources under poor dispersion conditions. The direct radiative forcing (DRF) due to UVBC and BC at the surface (SUR) and the top of the atmosphere (TOA) were calculated. The mean atmospheric heating rates due to UVBC and BC were estimated to be 1.40°Kday-1 and 1.18°Kday-1, respectively. This high heating rate may affect the monsoon circulation in this region.

Mercury (Hg) concentration trends in top predator fish (lake trout and walleye) of the Great Lakes (GL) from 2004 to 2015 were determined by Kendall-Theil robust regression with a cluster-based age normalization method to control for the effect of changes in lake trophic status. When data from the GLs (except Lake Erie) are combined, a significant decreasing trend in the lake trout Hg concentrations was found between 2004 and 2015 with an annual decrease of 4.1% per year, consistent with the decline in regional atmospheric Hg emissions and water Hg concentrations. However, a breakpoint was detected with a significant decreasing slope (-8.1% per year) before the breakpoint (2010), and no trend after the breakpoint. When the lakes are examined individually, Lakes Superior and Huron, which are dominated by atmospheric Hg inputs and are more likely than the lower lakes to respond to declining emissions from areas surrounding the GL, have significant decreasing trends with rates between 5.2 and 7.8% per year from 2004 to 2015. These declining trends appear to be driven by decreasing regional atmospheric Hg emissions although they may be partly counterbalanced by other factors, including increasing local emissions, food web changes, eutrophication, and responses to global climate change. Lakes Michigan, Erie and Ontario may have been more impacted by these other factors and their trends changed from decreasing to non-decreasing or increasing in recent years.