FedMCC: Federated multi-center clustering algorithm to improve privacy healthcare

In recent years, healthcare data from various sources such as clinical institutions, patients, and pharmaceutical industries have become increasingly abundant. However, due to the complex healthcare system and data privacy concerns, aggregating and utilizing these data in a centralized manner can be challenging. Federated learning (FL) has emerged as a promising solution for distributed training in edge computing scenarios, utilizing on-device user data while reducing server costs. In traditional FL, a central server trains a global model sampled client data randomly, and the server combines the collected model from different clients into one global model. However, for not independent and identically distributed (non-i.i.d.) datasets, randomly selecting users to train server is not an optimal choice and can lead to poor model training performance. To address this limitation, we propose the Federated Multi-Center Clustering algorithm (FedMCC) to enhance the robustness and accuracy for all clients. FedMCC leverages the Model-Agnostic Meta-Learning (MAML) algorithm, focusing on training a robust base model during the initial training phase and better capturing features from different users. Subsequently, clustering methods are used to ensure that features among users within each cluster are similar, approximating an i.i.d. training process in each round, resulting in more effective training of the global model. We validate the effectiveness and generalizability of FedMCC through extensive experiments on public healthcare datasets. The results demonstrate that FedMCC achieves improved performance and accuracy for all clients while maintaining data privacy and security, showcasing its potential for various healthcare applications.

Microplastics in Australian indoor air: Abundance, characteristics, and implications for human exposure

Studies on airborne microplastics (AMPs) have reported higher abundance of AMPs in indoor air compared to outdoors. Most people spend more time indoors compared to outdoors, and it is therefore important to identify and quantify AMPs in indoor air to understand human exposure to AMPs. This exposure can vary among different individuals as they spend their time in different locations and different activity levels, and thus experience different breathing rates. In this study, AMPs ranging from 20 to 5000 μm were sampled across different indoor sites of Southeast Queensland using an active sampling technique. The highest indoor MP concentration was observed at the childcare site (2.25 ± 0.38 particles/m³), followed by an office (1.20 ± 0.14 particles/m³) and a school (1.03 ± 0.40 particles/m³). The lowest indoor MP concentration was observed inside the vehicle (0.20 ± 0.14 particles/m³), comparable to outdoor concentrations. The only shapes observed were fibers (98 %) and fragments. MP fibers ranged from 71 to 4950 μm in length. Polyethylene terephthalate was the prominent polymer type at most sites. Using our measured airborne concentrations as inhaled air concentrations, we calculated the annual human exposure levels to AMPs using scenario-specific activity levels. Males between the ages of 18 to 64 were calculated to have the highest AMP exposure at 3187 ± 594 particles/year, followed by males ≥65 years at 2978 ± 628 particles/year. The lowest exposure of 1928 ± 549 was calculated for females between the ages of 5 to 17. This study provides the first report on AMPs for various types of indoor locations where individuals spend most of their time. Considering acute, chronic, industrial, and individual susceptibility, more detailed human inhalation exposure levels to AMPs should be estimated for a realistic appraisal of the human health risk, including how much of the inhaled particles are exhaled. SYNOPSIS: Limited research exists on the occurrence and the associated human exposure levels to AMPs in indoor locations where people spend most of their time. This study reports on the occurrence of AMPs at indoor locations and associated exposure levels using scenario-specific activity levels.

An ensemble learning with active sampling to predict the prognosis of postoperative non-small cell lung cancer patients

Background

Lung cancer is the leading cause of cancer death worldwide. Prognostic prediction plays a vital role in the decision-making process for postoperative non-small cell lung cancer (NSCLC) patients. However, the high imbalance ratio of prognostic data limits the development of effective prognostic prediction models.

Methods

In this study, we present a novel approach, namely ensemble learning with active sampling (ELAS), to tackle the imbalanced data problem in NSCLC prognostic prediction. ELAS first applies an active sampling mechanism to query the most informative samples to update the base classifier to give it a new perspective. This training process is repeated until no enough samples are queried. Next, an internal validation set is employed to evaluate the base classifiers, and the ones with the best performances are integrated as the ensemble model. Besides, we set up multiple initial training data seeds and internal validation sets to ensure the stability and generalization of the model.

Results

We verified the effectiveness of the ELAS on a real clinical dataset containing 1848 postoperative NSCLC patients. Experimental results showed that the ELAS achieved the best averaged 0.736 AUROC value and 0.453 AUPRC value for 6 prognostic tasks and obtained significant improvements in comparison with the SVM, AdaBoost, Bagging, SMOTE and TomekLinks.

Conclusions

We conclude that the ELAS can effectively alleviate the imbalanced data problem in NSCLC prognostic prediction and demonstrates good potential for future postoperative NSCLC prognostic prediction.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-022-01960-0.

Impacts of sampling-tube loss on quantitative analysis of gaseous semi-volatile organic compounds (SVOCs) using an SPME-based active sampler

Active samplers are widely used in the quantification of gaseous semi-volatile organic compounds (SVOCs). A sampling tube is often assembled upstream of the sampler, especially in the active samplers used for separating the particle-phase and gas-phase SVOCs and in the newly-designed active sampler based on solid-phase microextraction (SPME). However, gaseous SVOCs can be easily adsorbed by the sampling tube, which may induce significant errors to the quantitative results. Taking the SPME-based active sampler as an example, a mass-transfer model was developed to characterize the sampling-tube loss of gaseous SVOCs. Experiments involving six SVOCs were conducted. The model predictions (with a best-fit surface/air partition coefficient of SVOCs) were found to be consistent with the measurements. Both model predictions and experimental data indicated that the measured concentrations were significantly lower than the actual concentration (around 60% lower) due to the sampling-tube loss. The duration of sampling-tube loss (τ_e, minutes to days) varied with the volatility of SVOCs (vapor pressure, V_p), i.e., log τ_e linearly increased as increasing log V_p. The relationship could be helpful for determining the sampling strategies to eliminate (reduce) the effects of sampling-tube loss according to the volatility of SVOCs. The above conclusions may be also applicable for other active samplers of gaseous SVOCs. However, further studies are required to quantify the effects of sampling-tube loss for other active samplers due to the difference in the size and shape of the sampling tube between them and the SPME-based active sampler. The corresponding mass-transfer model and experimental procedure may require adjustment as appropriate.

What is the spatial-temporal behavior of a low, medium and high adsorptive compound in two contrasting natural sediments in OECD 218/219 test systems?

Artificial sediment used in studies according to OECD 218/219 (Sediment Water Chironomid Toxicity Test Using Spiked Sediment/Water) does not necessarily mirror the characteristics of natural sediments. To investigate the influence of sediment characteristics on the spatial-temporal behaviors of bixafen (KfOM = 2244 mL/g), fluopyram (KfOM = 162 mL/g) and N,N-dimethylsulfamide (KfOM ≈ 0 mL/g), experiments according to OECD 218/219 with two contrasting natural sediments were conducted. The silt loam sediment provided a high content of organic matter (OM) (13.1%), while the OM (0.45%) of the sandy sediment was low. Diffusion into (OECD 219) or out (OECD 218) of the sediment was dependent on the extent of adsorption, which is linked to the model compounds ́ adsorption affinities and the sediments ́ OM. Consequently, N,N-dimethylsulfamide showed unhindered mobility in each experimental set up, while the high adsorption affinities of fluopyram and bixafen limited the diffusion in the respective sediments. Therefore, in experiments with the silt loam sediment, both compounds revealed a limited mobility and either accumulated in the top 5 mm of the sediment (OECD 219) or remained homogenously distributed over the sediment depth (OECD 218). A greater mobility was observed within the sandy sediment.The influence of OM as found in a study using artificial sediment could be confirmed. Moreover, the applicability of a TOXSWA model was reassured to predict the measured concentrations at different sediment depths. TOXSWA is used in the regulatory exposure assessment to simulate the behavior of pesticides in surface waters. Calibration of three driving input parameters by inverse modelling (diffusion-, adsorption coefficient and OM) revealed no potential for improvement. The core sampling technique used and the model may contribute to a more realistic determination of concentration to which the Chironomid larvae are exposed to. This applies to water sediment test systems where the test organisms do not evenly inhabit the sediment.

Working memory and active sampling of the environment: Medial temporal contributions

Working memory (WM) refers to the ability to maintain and actively process information-either derived from perception or long-term memory (LTM)-for intelligent thought and action. This chapter focuses on the contributions of the temporal lobe, particularly medial temporal lobe (MTL) to WM. First, neuropsychological evidence for the involvement of MTL in WM maintenance is reviewed, arguing for a crucial role in the case of retaining complex relational bindings between memorized features. Next, MTL contributions at the level of neural mechanisms are covered-with a focus on WM encoding and maintenance, including interactions with ventral temporal cortex. Among WM use processes, we focus on active sampling of environmental information, a key input source to capacity-limited WM. MTL contributions to the bidirectional relationship between active sampling and memory are highlighted-WM control of active sampling and sampling as a way of selecting input to WM. Memory-based sampling studies relying on scene and object inspection, visual-based exploration behavior (e.g., vicarious behavior), and memory-guided visual search are reviewed. The conclusion is that MTL serves an important function in the selection of information from perception and transfer from LTM to capacity-limited WM.

Field application of a novel active-passive sampling technique for the simultaneous measurement of a wide range of contaminants in water

A first test of the field capabilities of a novel in situ sampling technique combining active and passive sampling (APS) was conducted in the sea. The proof-of-concept device uses a pump to draw water into a diffusion cell where dissolved target substances are accumulated onto sorbents which are selective for different classes of contaminants (i.e., metal cations, polar and non-polar organic compounds), simultaneously. A controlled laminar flow established in the diffusion cell enables measurements of contaminant concentrations that are fully independent from the hydrodynamic conditions in the bulk solution. APS measurements were consistent with those obtained using conventional passive sampling techniques such as organic diffusive gradients in thin films (o-DGT) and silicone rubber (SR) samplers (generally < 40% difference), taking into account the prevailing hydrodynamic conditions. The use of performance reference compounds (PRC) for hydrophobic contaminants provided additional information. Field measurements of metal ions in seawater showed large variability due to issues related to the device configuration. An improved field set-up deployed in supplementary freshwater mesocosm experiments provided metal speciation data that was consistent with passive sampling measurements (DGT), taking into account the hydrodynamic conditions. Overall, the results indicate that the APS technique provides a promising approach for the determination of a wide range of contaminants simultaneously, and independently from the hydrodynamic conditions in the bulk solution.

BTEX compositions and its potential health impacts in Malaysia

This study aims to determine the composition of BTEX (benzene, toluene, ethylbenzene and xylene) and assess the risk to health at different sites in Malaysia. Continuous monitoring of BTEX in Kuala Lumpur City Centre, Kuala Terengganu, Kota Kinabalu and Fraser Hill were conducted using Online Gas Chromatograph. For comparison, BTEX at selected hotspot locations were determined by active sampling method using sorbent tubes and Thermal Desorption Gas Chromatography Mass Spectrometry. The hazard quotient (HQ) for non-carcinogenic and the life-time cancer risk (LTCR) of BTEX were calculated using the United States Environmental Protection Agency (USEPA) health risk assessment (HRA) methods. The results showed that the highest total BTEX concentrations using continuous monitoring were recorded in the Kuala Lumpur City Centre (49.56 ± 23.71 μg/m³). Toluene was the most dominant among the BTEX compounds. The average concentrations of benzene ranged from 0.69 ± 0.45 μg/m³ to 6.20 ± 3.51 μg/m³. Measurements using active sampling showed that BTEX concentrations dominated at the roadside (193.11 ± 114.57 μg/m³) in comparison to petrol station (73.08 ± 30.41 μg/m³), petrochemical industry (32.10 ± 13.13 μg/m³) and airport (25.30 ± 6.17 μg/m³). Strong correlations among BTEX compounds (p<0.01, r>0.7) at Kuala Lumpur City Centre showed that BTEX compounds originated from similar sources. The values of HQ at all stations were <1 indicating the non-carcinogenic risk are negligible and do not pose threats to human health. The LTCR value based on benzene inhalation (1.59 × 10^-5) at Kuala Lumpur City Centre were between 1 × 10^-4 and 1 × 10^-5, representing a probable carcinogenic risk.

Characterization of the accumulation of metals and organic contaminants on a novel active-passive sampling device under controlled water flow conditions

Recently, a new sampling device combining active and passive sampling (APS) was developed for the measurement of time-averaged concentrations of metal species and both polar and non-polar organic contaminants in water. By coupling a diffusion cell (loaded with a set of sorbents selective for different substances) with a small pump and a flow meter, the APS device is able to perform in situ measurements that are independent of the hydrodynamic conditions in the exposure medium. In the present study, the diffusion layer thickness (δ) at the sorbent/solution interface within the diffusion cell was characterised under controlled flow conditions. Laboratory tests indicated that, in the range of flow rates investigated, the average diffusion layer thickness (δ¯) varied from ∼60 to ∼110 μm, depending on the type of substance measured and the position of the sorbent with respect to the flow direction. Due to its ability to maintain an approximately constant δ¯, good to excellent agreement was found between measurements performed with the APS device in non-complexing media and concentrations measured in discrete water samples for all the substances investigated. These results suggest that the APS device could overcome issues affecting the quantitative interpretation of measurements by conventional passive sampling devices and serve as a useful tool for simultaneously monitoring a wide range of contaminants in water.

Outdoor air monitoring: Performance evaluation of a gas sensor to assess episodic nuisance/odorous events using active multi-sorbent bed tube sampling coupled to TD-GC/MS analysis

In order to evaluate the performance of a commercially available metal oxide semiconductor gas sensor (TGS 2602, Figaro Engineering Inc.) for activating a monitoring system when a nuisance/odorous pollution episode of volatile organic compounds (VOCs) occur, a widely used active sampling methodology based on multi-sorbent bed tubes (Carbotrap, Carbopack X and Carboxen 569) and analysis through automatic thermal desorption-gas chromatography/mass spectrometry was used. Daily 24 h samples of multi-sorbent bed tubes were taken over a period of 14 days using an air collector pump sampler specially designed in the LCMA-UPC laboratory. Simultaneously, daily episodic samples were taken according to the activation of another LCMA-UPC sampler by the metal oxide semiconductor gas sensor. Sampling was done throughout January-February 2019 at El Morell (Tarragona, Spain), near the petrochemical area. All episode samples present higher concentrations of VOCs than 24 h samples, with an average ratio of 3.5 times for Total VOCs. VOC familial distributions present very similar values in 24 h and episode samples (r² = 0.7466), correlating significatively (F-Snedecor, p < 0.05). A higher level of VOCs in the atmosphere in general, not derived from a specific compound or a VOC/s family/ies, seems to be the trigger of the activation of the sampler by the sensor. On the other hand, no significant correlations are observed between alcohols concentrations and relative humidity (F-Snedecor, p < 0.05). Additionally, Total VOCs concentrations in episode samples are in agreement with higher percentages of NE-SSE wind directions, coming from the petrochemical complex. Hence, these aspects validate the use of the evaluated sensor for its application for the activation of samplers in air quality evaluations when episodic events occur, an interesting and innovative technique. Thus, this study is an important contribution to the understanding of the performance of gas sensors and proposes an expansion of their field of use.

Monitoring gas- and particulate-phase short-chain polychlorinated paraffins in the urban air of Dalian by a self-developed passive sampler

The concentration of short-chain polychlorinated paraffins (SCCPs) in the urban air of Dalian, China was monitored from September 2016 to August 2017 with a self-developed passive sampler (PAS1) and an active high-volume sampler, simultaneously. PAS1 successfully collected the entire target SCCPs in the ambient air. Air SCCPs sampled by PAS1 were found be in the linear uptake stage during 181 days of sampling. Passive and active samples showed comparable congener profiles, and the dominant contributors of SCCPs in the two kinds of samples were similar. A significant linear correlation was observed between the total concentration of SCCPs sampled by PAS1 and active sampler in the four seasons. The passive sampling rates of the PAS1 for the gas and particulate phases of SCCPs were measured. The quantitative structure-property relationship of the sampling rate of PAS1 (R_air) for gas-phase SCCPs was studied. From the molecular point of view, R_air was mainly affected by the molecular weight and sub-cooled liquid vapor pressure of SCCPs. In general, SCCPs in the urban air of Dalian mainly existed in gas phase, lower molecular weight SCCPs primarily occurred in the gas phase, whereas higher molecular weight SCCPs were predominately adsorbed or absorbed on airborne particles. The air concentration of SCCPs in the four seasons were different, the correlation of the concentration of SCCPs in the air with the meteorology parameters was conducted. The exposure risk by intake air SCCPs of the residents around the sampling sites was evaluated according to the European risk assessment standards.

A novel active-passive sampling approach for measuring time-averaged concentrations of pollutants in water

Passive sampling with in situ devices offers several advantages over traditional sampling methods (i.e., discrete spot sampling), however, data interpretation from conventional passive samplers is hampered by difficulties in estimating the thickness of the diffusion layer at the sampler/medium interface (δ), often leading to inaccurate determinations of target analyte concentrations. In this study, the performance of a novel device combining active and passive sampling was investigated in the laboratory. The active-passive sampling (APS) device is comprised of a diffusion cell fitted with a pump and a flowmeter. Three receiving phases traditionally used in passive sampling devices (i.e., chelex resin, Oasis HLB, and silicone rubber), were incorporated in the diffusion cell and allowed the simultaneous accumulation of cationic metals, polar, and non-polar organic compounds, respectively. The flow within the diffusion cell was accurately controlled and monitored, and, combined with diffusion coefficients measurements, enabled the average δ to be estimated. Strong agreement between APS and time-averaged total concentrations measured in discrete water samples was found for most of the substances investigated. Accuracies for metals ranged between 87 and 116%, except Cu and Pb (∼50%), whilst accuracies between 64 and 101%, and 92 and 151% were achieved for polar and non-polar organic compounds, respectively. These results indicate that, via a well-defined in situ preconcentration step, the proposed APS approach shows promise for monitoring the concentration of a range of pollutants in water.

Gas/particle partitioning and particle size distribution of PCDD/Fs and PCBs in urban ambient air

Urban ambient air samples, including gas-phase (PUF), total suspended particulates (TSP), PM₁₀, PM_2.5 and PM₁ airborne particle fractions were collected to evaluate gas-particle partitioning and size particle distribution of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) and polychlorinated biphenyls (PCBs). Clausius-Clapeyron equation, regressions of logKp vs logP_L and logK_OA, and human respiratory risk assessment were used to evaluate local or long-distance transport sources, gas-particle partitioning sorption mechanisms, and implications for health. Total ambient air levels (gas phase+particulate phase) of TPCBs and TPCDD/Fs, were 437 and 0.07pgm^-3 (median), respectively. Levels of PCDD/F in the gas phase (0.004-0.14pgm^-3, range) were significantly (p<0.05) lower than those found in the particulate phase (0.02-0.34pgm^-3). The concentrations of PCDD/Fs were higher in winter. In contrast, PCBs were mainly associated to the gas phase, and displayed maximum levels in warm seasons, probably due to an increase in evaporation rates, supported by significant and strong positive dependence on temperature observed for several congeners. No significant differences in PCDD/Fs and PCBs concentrations were detected between the different particle size fractions considered (TSP, PM₁₀, PM_2.5 and PM₁), reflecting that these chemicals are mainly bounded to PM₁. The toxic content of samples was also evaluated. Total toxicity (PUF+TSP) attributable to dl-PCBs (13.4fg-TEQ₀₅ m^-3, median) was higher than those reported for PCDD/Fs (6.26fg-TEQ₀₅ m^-3). The inhalation risk assessment concluded that the inhalation of PCDD/Fs and dl-PCBs pose a low cancer risk in the studied area.

Silicone wristbands compared with traditional polycyclic aromatic hydrocarbon exposure assessment methods

Currently there is a lack of inexpensive, easy-to-use technology to evaluate human exposure to environmental chemicals, including polycyclic aromatic hydrocarbons (PAHs). This is the first study in which silicone wristbands were deployed alongside two traditional personal PAH exposure assessment methods: active air monitoring with samplers (i.e., polyurethane foam (PUF) and filter) housed in backpacks, and biological sampling with urine. We demonstrate that wristbands worn for 48 h in a non-occupational setting recover semivolatile PAHs, and we compare levels of PAHs in wristbands to PAHs in PUFs-filters and to hydroxy-PAH (OH-PAH) biomarkers in urine. We deployed all samplers simultaneously for 48 h on 22 pregnant women in an established urban birth cohort. Each woman provided one spot urine sample at the end of the 48-h period. Wristbands recovered PAHs with similar detection frequencies to PUFs-filters. Of the 62 PAHs tested for in the 22 wristbands, 51 PAHs were detected in at least one wristband. In this cohort of pregnant women, we found more significant correlations between OH-PAHs and PAHs in wristbands than between OH-PAHs and PAHs in PUFs-filters. Only two comparisons between PAHs in PUFs-filters and OH-PAHs correlated significantly (r_s = 0.53 and p = 0.01; r_s = 0.44 and p = 0.04), whereas six comparisons between PAHs in wristbands and OH-PAHs correlated significantly (r_s = 0.44 to 0.76 and p = 0.04 to <0.0001). These results support the utility of wristbands as a biologically relevant exposure assessment tool which can be easily integrated into environmental health studies.

Graphical abstract

Occurrence and seasonal distribution of polycyclic aromatic hydrocarbons and legacy and current-use pesticides in air from a Mediterranean coastal lagoon (Mar Menor, SE Spain)

The occurrence and seasonal distribution of polycyclic aromatic hydrocarbons (PAHs) and legacy and current-use pesticides (CUPs) in air were characterized around the Mar Menor lagoon using both active and passive sampling devices. The seasonal distribution of these pollutants was determined at 6 points using passive samplers. Passive sampler sampling rates were estimated for all detected analytes using an active sampler, considering preferentially winter data, due to probable losses in active sampling during summer (high temperatures and solar irradiation). The presence of 28 compounds (14 CUPs, 11 PAHs and 3 organochlorinated pesticides) were detected in air by polyurethane passive sampling. The most commonly detected contaminants (>95% of samples) in air were chlorpyrifos, chlorpyrifos-methyl and phenanthrene. The maximum concentrations corresponded to phenanthrene (6000 pg m^-3) and chlorpyrifos (4900 pg m^-3). The distribution of contaminants was spatially and seasonally heterogeneous. The highest concentrations of PAHs were found close to the airport, while the highest concentrations of pesticides were found in the influence area of agricultural fields (western stations). PAH and herbicide concentrations were higher in winter than in the other seasons, although some insecticides such as chlorpyrifos were more abundant in autumn. The presence of PAHs and legacy and current-use pesticides in air confirmed their transference potential to marine coastal areas such as the Mar Menor lagoon.

A new sorbent tube for atmospheric NOx determination by active sampling

In this paper we used hydrated mayenite as reactive substrate for NO_x active sampling in the air, which is novel. The performance of the mayenite-based sorbent for the NO_x tubes was evaluated in two different monitoring surveys (autumn 2015 and winter 2016), characterized by different environmental conditions. Sorbent tubes filled with mayenite were exposed simultaneously to triethanolamine (TEA)-based sorbent tubes and to a chemiluminescence detector, as reference. The comparison of the NO_x concentration levels measured by active sampling, using mayenite as NO_x sorbent, showed a close relationship with the chemiluminescence analyzer. The effect of the environmental conditions on the performance of both mayenite and TEA-based sorbent tubes was evaluated and limitations connected to the use of TEA were discussed.

Apparatus and method for time-integrated, active sampling of contaminants in fluids demonstrated by monitoring of hexavalent chromium in groundwater

Annual U.S. expenditures of $2B for site characterization invite the development of new technologies to improve data quality while reducing costs and minimizing uncertainty in groundwater monitoring. This work presents a new instrument for time-integrated sampling of environmental fluids using in situ solid-phase extraction (SPE). The In Situ Sampler (IS2) is an automated submersible device capable of extracting dissolved contaminants from water (100s-1000smL) over extended periods (hours to weeks), retaining the analytes, and rejecting the processed fluid. A field demonstration of the IS2 revealed 28-day average concentration of hexavalent chromium in a shallow aquifer affected by tidal stresses via sampling of groundwater as both liquid and sorbed composite samples, each obtained in triplicate. In situ SPE exhibited 75±6% recovery and an 8-fold improvement in reporting limit. Relative to use of conventional methods (100%), beneficial characteristics of the device and method included minimal hazardous material generation (2%), transportation cost (10%), and associated carbon footprint (2%). The IS2 is compatible with commercial SPE resins and standard extraction methods, and has been certified for more general use (i.e., inorganics and organics) by the Environmental Security Technology Certification Program (ESTCP) of the U.S. Department of Defense.

Presence and transport of the antimicrobials triclocarban and triclosan in a wastewater-dominated stream and freshwater environment

The presence of the antimicrobials triclocarban (TCC) and triclosan (TCS)in Fountain Creek, a wastewater-dominated stream, and the Arkansas River, Colorado, USA was measured in the surface water, suspended sediments, and bed sediments during spring runoff (high flow) and summer base flow (low flow) conditions. Fountain Creak is a tributary of the Arkansas River. Passive polar organic chemical integrative samplers (POCIS) were used along with active sampling (water grab samples) to measure and TCS concentrations in these surface waters. The concentration of TCC and TCS, based on POCIS measurements, ranged from 4.5 to 47.3 ng/L and 3.9 to 28.3 ng/L, respectively, at the five sample sites monitored in this study under both flow conditions. The range of concentrations of TCC and TCS in suspended sediments was 0.7-57.3 ng/g and 0.7-13.3 ng/g, respectively, and was closely tied to the quantity of organic carbon in the suspended sediment, which ranged from 1.6 to 14.5%. The quantity of organic carbon in suspended sediment during the summer base flow was influenced by runoff from the burn area of a large forest fire that occurred between the two sampling periods. The primary transport mechanism of TCC and TCS in these surface waters was in the dissolved phase, with 64-99% of TCC and 68-99% of TCS transported in the dissolved phase. The total amount of TCS and TCC in bed-sediments was relatively low, with the maximum amount at any one site being 0.38 ± 0.15 ng/g TCS and 4.09 ± 5.26 ng/g TCC. Fountain Creek contributed up to 76% and 69% of the TCC and TCS, respectively, that is transported directly below its confluence with Arkansas River. Fountain Creek drained approximately 3.0 g/day TCS (in spring), 2.9 g/day TCS (in summer) and 1.9 g/day TCC (in spring), 2.0 g/day TCC (in-summer) into the Arkansas River, which suggests consistent input of TCC and TCS into Fountain Creek, such as in discharge of treated wastewater that is independent of changing creek flow conditions.