Abstract Artificial snowmaking is used by winter recreation resorts to elongate the ski season in areas that lack sufficient snow, yet the occurrence of contaminants in artificial snow remains largely uncharacterized. Contaminants in artificial snow can lead to the introduction of pollutants into surrounding ecosystems (e.g., plants, water streams, soils) or create pathways for human exposure. Using the Arizona Snowbowl Ski Resort in Flagstaff, Arizona, as a case study, we characterized the chemical (organic and inorganic) profile of snow made from reclaimed water samples as compared to natural snow samples. Pharmaceuticals and anthropogenic compounds were identified in snowmaking from reclaimed water samples using nontargeted mass spectrometry paired with molecular networking, and identities of selected compounds were confirmed and quantified using targeted mass spectrometry alongside authentic chemical standards. Nineteen pharmaceuticals were found to be increased in high artificial snowmaking samples as compared to natural snow, including carbamazepine, fexofenadine, flecainide, and lamotrigine, which were present at low ng/L concentrations in one set of high artificial snowmaking samples. A suite of inorganic contaminants was also assessed and quantified using inductively coupled plasma mass spectrometry to reveal that nine metals─sodium, phosphorus, potassium, chromium, manganese, iron, zinc, arsenic, and barium─were increased in abundance in high artificial snowmaking samples compared to levels in normal snow. Taken together, this study suggests that snow made from reclaimed water may contain anthropogenic organic and inorganic analytes, warranting further study on their environmental and human health effects.
Abstract Micro-estuaries receive elevated loads of organic contaminants, yet the influence of water-column structure on their vertical distribution remains unclear. We investigated contaminant dynamics in three structurally distinct micro-estuaries during the dry season using monthly water-column and sediment sampling. Across 63 samples, 274 organic contaminants were detected. Contaminant mixtures shifted seasonally from runoff-to baseflow-dominated conditions, transitioning from pesticide-dominated profiles to more persistent industrial and effluent-derived compounds, including PFAS and sucralose. Water-column stratification strongly influenced contaminant partitioning: strongly stratified systems showed surface accumulation, while well-mixed systems exhibited uniform vertical distributions. Intermediate systems displayed partial segregation, reflecting intermittent mixing and prolonged deep-water residence times. Sediment–deep water distribution coefficients did not correlate with the sediment organic carbon content, suggesting that the dynamic estuarine conditions prevent steady-state sorption–desorption equilibrium. These findings demonstrate that stratification shapes contaminant persistence and biological exposure, underscoring the need for monitoring and risk assessments that incorporate vertical structure.
Abstract Machine learning (ML) holds immense promise for predictive modeling but is often misapplied without a systematic understanding of the full model development pipeline. This tutorial provides a comprehensive, end-to-end workflow─from raw data to deployed models─explicitly designed for environmental chemists with limited prior experience in ML modeling while also providing practical guidance for other users seeking to strengthen their modeling workflows. Using a public high-performance liquid chromatography small molecule retention time (SMRT) data set containing over 80,000 small molecules, this tutorial demonstrates critical stages: data cleaning via chemical similarity and scaffold analysis, application-oriented data splitting, feature engineering with molecular fingerprints and graphs, model development using commonly used LightGBM and Graph Convolutional Network (GCN), and model interpretation with different methods. This work highlights how domain knowledge guides each step to avoid common pitfalls, such as overestimating model performance through inappropriate data splits. The work concludes with a robust deployment framework that ensures models are both accurate and accessible. All code is openly available, providing an example implementation for users to reference when developing ML models across diverse applications.
Abstract 6-PPDQ, an oxidation product of the antioxidant 6-PPD added to tire rubber, reaches the environment with tire abrasion on roads. Firm data on the environmental persistence of 6-PPDQ in different compartments is lacking. The microbial degradation of 6-PPDQ was studied in simulation tests under standardized conditions in the laboratory. Biodegradation half-life of 6-PPDQ (DT50) increased from surface water (12–17 d) over water-sediment systems including nonextractable residue (NER I) under aerobic conditions (46–55 d), soils including NER I under aerobic conditions (27–141 d, median 61 d), water–sediment systems under anaerobic conditions (140–202 d) to soil under anaerobic conditions (214–887 d, >10 000 d in one model). On this basis, 6-PPDQ would be considered nonpersistent according to ECHA’s definition under aerobic conditions in water, in water–sediment systems, and in soil (median value and three of the four tested soils). 34% of the applied radioactivity (AR) of [14C]-6-PPDQ was mineralized under aerobic conditions in 60 d in surface water, 27% in the water–sediment system in 100 d and 34% in soil in 120 d. Under anaerobic conditions, mineralization was 4%–5% AR, only, both in water–sediment and in soil. Hydroxylated and carboxylated 6-PPDQ were identified as transformation products, of which 11% AR remained after 60 d. In sediment and soil no single transformation product exceeded 1% AR at the end of the experiment; instead, NERs were formed. Under aerobic conditions, 55% (water–sediment) and 49% (soil) of the AR of degraded 6-PPDQ occurred as NER; it increased to 67% (water–sediment) and 84% (soil) under anaerobic conditions. These first quantitative data on the extent and the rate of biodegradation and mineralization of 6-PPDQ help to understand the occurrence of 6-PPDQ in the environment and support environmental exposure modeling and risk assessment.
Hydroxyethers (HEs) are widely used oxygenated volatile organic compounds (OVOCs). Among them, 2-propoxyethanol (CH3CH2CH2OCH2CH2OH, 2-PE) is of particular interest due to its extensive use and the limited mechanistic information available regarding its atmospheric degradation. These characteristics place 2-PE as a compound of emerging environmental interest and motivate a detailed investigation of its reactivity with key atmospheric oxidants. This compound is commonly used as a solvent and emulsifying agent for hydrophobic substances, and it has been proposed as a biodiesel additive. Once released into the atmosphere, 2-PE undergoes degradation processes mainly by reactions with oxidants (OH and NO3 radicals, Cl atoms). This study provides a kinetic and mechanistic investigation of the atmospheric degradation of 2-PE using FTIR (Fourier transform infrared spectroscopy) and GC-MSTOF (gas chromatography/mass spectrometry time of flight). Rate coefficients at ambient temperature and pressure, were (units cm3 molecule-1 s-1): (1.99 +/- 0.12) & times; 10-10, (2.54 +/- 0.15) & times; 10-11 and (7.20 +/- 0.53) & times; 10-15 for Cl, OH center dot and NO3 center dot reactions, respectively. Detected products include formaldehyde, acetaldehyde, propyl formate, 2-hydroxyethyl formate and 2-hydroxyethyl propanoate, quantified by FTIR, and nitrated compounds detected in the presence of NO x . Based on product distributions and structure-reactivity relationships, two predominant reaction pathways are proposed, both initiated by the oxidant attack on the -CH2- group adjacent to the ether. Atmospheric lifetimes indicate that OH center dot reaction dominates the 2-PE removal. Estimated GWP and ozone formation indices confirm that 2-propoxyethanol is a short-lived VOC with limited direct climate impact, while providing a basis for comparison with structurally related glycol ethers and other VOCs.
Freshwater lakes are vulnerable to land-use changes that promote the input and accumulation of organic pollutants in sediments, posing risks to aquatic ecosystems. This study assessed the relationship between watershed land use and the occurrence of selected pesticides, antimicrobial agents, and alkylphenols in surface sediments of Lake Victoria in Africa. Surface sediment samples (n = 38) were collected in September 2019 across the Uganda, Kenya, and Tanzania sides of the lake. The samples were extracted using ultrasound-assisted extraction (UAE) and analyzed by gas chromatography-mass spectrometry (GC-MS). 16 out of 23 target compounds were detected, with pesticides showing low concentrations (1 (4-tert-octylphenol, 2-phenylphenol, 4-phenylphenol, beta-HCH, and 4-n-octylphenol) that pose potential aquatic ecotoxic risks to aquatic organisms. This study demonstrates that activities related to agriculture, industrialization, and urban development are the main drivers of sediment contamination in the lake, highlighting the need for targeted catchment management and monitoring.
Understanding the physicochemical properties of fly ash from industrial and municipal solid waste incineration is essential for its safe and efficient utilization. This study investigated the particle size distribution and elemental size partitioning in fly ash aerosols from a waste-to-energy (WtE) facility with grate-fired boilers, along with the composition of boiler deposits and ash collected in the electrostatic precipitator (ESP). The physicochemical properties of fly ash were investigated using online aerosol instruments combined with size-selective collection (low-pressure impactor and cyclone-filter setup) for gravimetric and elemental analyses. The particle size distribution was multimodal, with a distinct mass peak at 0.5 μm well separated from two overlapping modes at 30 and 200 μm. Fine particles (<1 μm) represented 16% of the total mass but were dominant in number. Elemental analysis showed that fine particles mainly consisted of Cl, Na, K, Zn, and S. Fine particles were enriched in potentially toxic yet valuable metals (Zn, Cd, Cu, Sb, Pb, Sn). Coarse particles (>1 μm) were dominated by Ca, Si, and Al, while the above-mentioned metals were depleted. Boiler ash resembled the coarse fraction, and ESP ash was a mix of both fine and coarse particles. Increased knowledge of the elemental composition in different size fractions may enable large-scale size separation for improved resource recovery and safer utilization. The composition of the fine particles support targeted salt or metal recovery, while coarse particles are suitable for construction applications.
Linear solvation energy relationship (LSER) models are nowadays often used to predict physicochemical properties of chemicals, such as partition coefficients, retention factors in chromatography, and solubilities. Due to their mechanistic foundation and transferability across phases, LSER models are particularly valuable for predicting partition coefficients in data-poor systems where experimental data sets are not available. However, their broader applicability is currently constrained by the limited availability of experimentally determined solute descriptors. We developed three different models, based on three different approaches, to predict the solute descriptors S, E, A, B, and L for LSER applications: (1) a group contribution model that includes an initial screening algorithm to identify functional groups as structural patterns, (2) a k-nearest neighbors model, and (3) a graph-convolutional neural network. All models were developed on the same curated data set for each solute descriptor. An independent test set was used to evaluate the overall model performance, with rmse values ranging from 0.08-0.13 for A, 0.10-0.15 for B, 0.17-0.23 for S, 0.09-0.19 for E, and 0.25-0.45 for L across the three approaches. By enabling the prediction of these descriptors directly from molecular structure, our modeling framework addresses a key bottleneck that has so far limited the scalability and broader application of the LSER models. In addition, we investigated whether a consensus approach would enhance overall prediction quality. Additionally, the predicted solute descriptors were directly used to derive environmentally and analytically relevant partition coefficients, demonstrating that reliable LSER-based property predictions are feasible even for chemicals lacking experimental descriptor data or large property-specific training sets. By enabling descriptor generation at scale, this work improves the practical applicability of LSER modeling and strengthens its role as a transferable and data-efficient tool for environmental fate and risk assessment.
Urban wastewater systems were historically designed under assumptions of climatic stationarity; however, climate change is rapidly altering the boundary conditions under which they operate. In this Perspective, we synthesize emerging evidence to conceptualize wastewater systems as climate-responsive infrastructure embedded within dynamic environmental feedbacks. We propose an input-output boundary framework to characterize climate change-driven alterations in wastewater systems. On the input side, climate change affects wastewater systems through three primary pathways: increased flow variability, shifts in physicochemical characteristics of wastewater, and changes in the contaminant spectrum. On the output side, increased risks of untreated discharge or improperly treated effluents and amplified greenhouse gas emissions intensify the environmental footprint of wastewater systems. Collectively, these nonstationary pressures undermine the reliability and effectiveness of key wastewater components, such as wastewater collection networks, biological processes, and advanced treatment units, thereby challenging conventional design and operation. Finally, we highlight four key opportunities of climate-resilient wastewater management, including predictive modeling and adaptive operational strategies enabled by digitalization, resilient infrastructure and process design, decentralized treatment with resource recovery, and nature-based solutions. Reframing wastewater systems as adaptive, low-emission, and circular infrastructures is essential to sustain urban water security under accelerating climate change.
Humans use soil to supply 97-99% of their calories, but the unprecedented pressure being placed on soil is now causing rapid soil degradation, including through erosion, loss of soil organic matter, acidification, contamination, salinization, and biodiversity loss. Here, we show how this degradation is causing an inherent decrease in global food productionthe loss of soil organic matter, for example, is predicted to cause a 4.3% decrease in yields for staple crops (representing calories for 640 million people) while soil erosion is projected to cause a loss in global crop production of 10% to 2050. While the adverse effects of this degradation on crop yield can often be largely masked or reversed by increased use of anthropogenic inputs such as fertilizer and irrigation, we show that some forms of soil degradation lead to permanent decreases in food production. Regardless, in all cases, we examine how both the soil degradation itself and the additional inputs required thereafter to sustain productivity result in substantial planetary harm, such as through climate change and eutrophication. It is clear that urgent action is required for ensuring both food security and planetary health, including through the development of integrated frameworks for improved policy- and decision-making.
Machine learning (ML) has become a powerful paradigm for extracting structures from complex environmental data and supporting scientific inference across diverse subfields. Its potential, however, is often limited by gaps in the appropriate application of domain knowledge to machine-learning workflows, variability in data quality, and methodological choices that can distort model behavior or its interpretation. This Tutorial provides practical guidance on how domain expertise can be effectively integrated into the design of environmentally meaningful machine learning models and outlines a coherent workflow that integrates crucial stages, including data preprocessing, model development, evaluation, and interpretability. It also examines recurring pitfalls that arise along this pipeline and explains how they shape the credibility and reliability of machine-learning findings in environmental contexts. By consolidating these principles, this Tutorial aims to provide researchers with a clearer foundation for using machine learning in ways that are scientifically grounded, methodologically rigorous, and better aligned with the needs of environmental decision-making.
Per- and polyfluoroalkyl substances (PFAS) are widespread environmental contaminants that are largely observed as mixtures. PFAS-focused ecological risk assessments largely assess individual PFAS, but understanding PFAS mixture exposure and potential ecological and biological effects is critical to characterizing real-world risk. To address this, we exposed CD-1 mice to individual PFAS and mixtures of PFAS that are relevant to surveyed environmental media and aqueous film-forming foams (AFFF) used on Department of Defense (DoD) sites. Our study was performed in two parts: (1) assess the whole-body concentration of PFAS and (2) assess tissue compartment concentrations of PFAS (i.e., serum, liver, kidney, and brain). Organ weights and clinical chemistry were collected in the tissue compartment study to measure potential toxicological effects of individual and mixtures of PFAS. Mixed-effect models were used to evaluate how well relationships between dose and body compartments predicted exposure with and without the influence of mixtures. A PFOS-relative weighting factor was evaluated for the potential to predict select effects from dose and tissue concentrations. We apply the laboratory-based finding to a desktop ecological risk assessment scenario and find that while changes in PFAS dietary concentrations may alter exposure and effect estimates, differentiating reference vs impacted sites is difficult. Our results indicate that dose addition methods are successful predictors of PFAS exposure and liver effects and, importantly, provide efficient predictive screening tools for ecological risk assessors evaluating sites' PFAS risks
Persistent and mobile organic contaminants (PMOCs) evade water treatment technologies and current monitoring efforts, posing a threat to water resources. To investigate this largely unexplored chemical space, we developed an analytical workflow based on sample enrichment by evaporative concentration and chromatographic separation by zwitterion-HILIC to quantify known and discover previously unknown PMOCs in water samples. We applied this workflow to drinking and surface water samples in the Greater Boston area of Massachusetts, USA, in a pilot effort to address the geographical gap in PMOC analysis, as the vast majority of studies take place in European countries. We targeted 19 analytes for quantitation via isotope dilution. TFA was estimated at hundreds of ng/L in 6 municipal tap water samples, and 13 analytes were quantified at 10-100s ng/L in surface water samples. 191 total compounds were annotated via nontarget analysis in drinking water, including 12 level 1, 32 level 2, and 44 level 3 identifications, and we report 12 new level 2+ drinking water contaminants, including 1-methylguanidine, 1,2-dimethylguanidine, and trimethylarsine oxide. Suspect screening of surface water samples revealed 80 PMOC candidate compounds. We cross-referenced surface and drinking water data to find 50 common contaminants despite a lack of hydrological connection between sample sources, 12 of which occurred in tap, bottled, and surface water, including halogenated methanesulfonic acid disinfection byproducts and the newly reported 1,2-dimethylguanidine. Our results reinforce the need for tailored analytical methods and widespread geographic efforts to achieve a holistic understanding of PMOC contamination and its implications for aquatic and human health.
This study employed both qualitative and quantitative approaches to examine Supermarket Food Waste (SFW) and Packaging (SPK) generations, as well as current management practices implemented by supermarkets and public agencies, using Bangkok, Thailand as a case study. Two supermarket chains provided SFW data and participated in interviews, along with three agencies from the Bangkok Metropolitan Administration (BMA) and four private waste processors were interviewed to explore the current SFW management practices. Carbon Footprint (CF) and Life Cycle Cost (LCC) analyses were also conducted to support the development of policy recommendations for sustainable SFW management. Results indicated that fruits and vegetables and plastics constituted the largest portion of SFW and SPK, accounting for 92.09% and 80.20% respectively, with a packaging-to-food weight ratio of 0.09 kg SPK/kg SFW. Thematic analysis revealed that strong public-private partnerships between supermarket parent companies and public agencies significantly influenced operational efforts, leading to successful SFW management. Based on these findings, policy recommendations were developed across three dimensions: environmental, economic, and policy; tailored to supermarkets and public agencies to support sustainable SFW management.
This research assessed biosolids from sludge treatment wetlands (STW) as possible biofertilizers, comparing them with digestate, compost, and microalgae biomass in greenhouse trials involving three crops (i.e., lettuce, radish, and ryegrass). Additionally, the potential biostimulant effects were evaluated through bioassays conducted under controlled conditions using water extracts of the biosolids. The biosolids showed a notable increase in radish growth compared with the unfertilized control, with increases of 93% and 95% in fresh and dry weight, respectively. The protein content of all crops grown under biosolids treatment was similar to that of microalgae, and it exceeded the control treatments by 6%, 9%, and 17% for lettuce, ryegrass, and radish, respectively. Furthermore, using biosolids reduced nutrient loss through leaching (i.e., N mineral forms), which was observed in urea treatments. Phytohormone-like bioassays revealed that biosolids extracts showed gibberellin-like activity, aiding seed germination and growth. Biosolids extracts also delayed senescence by slowing chlorophyll breakdown in wheat leaves and stimulated secondary root formation in bean seedlings. Overall, the results demonstrated the dual role of STWs' biosolids as biofertilizers and biostimulants, with performance comparable to urea and most of the biofertilizers used for comparison. Future studies should include field trials to confirm agronomic performance under real conditions.
This study elaborates a theoretical framework to understand metal bioaccumulation beyond the limitations of the equilibrium-based Biotic Ligand Model (BLM). By integrating the dynamics of metal speciation in aquatic media with the kinetics of metal biouptake, the formalism predicts metal bioavailability under conditions where the BLM fails, such as high degrees of metal complexation, diffusion-limited metal uptake, and/or internalization of intact metal complexes. The theory accounts for the reactive transport of free metal ions and complexes, incorporating different uptake mechanisms by facilitated and passive diffusion. Extended Best expressions are derived for the flux of metal biouptake by coupling extracellular metal chemodynamics (intertwined diffusion and complexation) with Michaelis-Menten uptake kinetics. The approach provides a unified rationale for various bioaccumulation situations, including the uptake of lipophilic complexes, and the concomitant uptake of free and complexed metals through distinct or shared diffusion-facilitated pathways with competitive, noncompetitive, and/or uncompetitive inhibitions. Computational examples are detailed to illustrate the intricate interplay between metal species transport dynamics and biouptake kinetics for all bioaccumulation cases, particularly addressing how the metal internalization flux is impacted by the (bioavai)lability of metal complexes, whether or not they are internalized intact. The benefits of the formalism are further illustrated through an analysis of well-characterized experimental data on neodymium (Nd) uptake by Chlamydomonas reinhardtii in the absence/presence of well-defined organic ligands. This analysis not only derives key metal biouptake and bioaffinity parameters but also provides solid evidence of the BLM's failure to describe the data. The quantitative interpretation of Nd bioaccumulation data leads to the identification of two potential uptake mechanisms, and a methodology to distinguish between them is discussed. Overall, this work establishes a more accurate and comprehensive theoretical foundation for predicting metal bioaccumulation in aquatic systems, thereby fundamentally challenging the common equilibrium-based perception of metal bioavailability.
Bromine (Br2) plays a crucial role in atmospheric oxidative chemistry, particularly in polar, marine, and stratospheric environments, where it significantly impacts ozone depletion and mercury oxidation. In this study, high-resolution measurements of the Br2 absorption cross section were evaluated across 325-850 nm (11765 to 30,769 cm-1), with emphasis on features due to the B3Πou ← X1Σg + transition at 530 nm (18,867 cm-1). UV-vis spectra revealed well-resolved vibronic features, enabling the accurate assignment of absorption peaks. Simulated rovibronic spectra reproduced the measured features and confirmed spectral assignments. The data set produced in this work shows improved signal-to-noise and peak distinction compared to prior studies and reveals differences of up to 25% relative to literature values, and to 6.7% relative to the JPL evaluation values, in structured regions of the spectrum, which directly alters calculated Br2 photolysis rates used in atmospheric chemistry models. These refined cross sections serve as robust reference data for laboratory and atmospheric research, supporting improved photochemical modeling of reactive halogen species. A broadband cavity-enhanced absorption spectroscopy (BBCEAS) instrument was employed to test the application of the evaluated absorption cross-section in a field-ready instrument and setting. The BBCEAS method achieved a detection limit of 97 ppt and an effective optical path length of 22.6 km, facilitating sensitive detection suitable for field applications.
