High phosphorus (P) loads to the Laurentian Great Lakes contribute to eutrophication, harmful algal blooms, and hypoxia. Groundwater impacted by wastewater effluent from household septic systems, common in coastal communities, is a recognized P source to nearby lakes. However, the long-term impact of neighbourhood-scale septic system decommissioning (i.e., conversion to sewer connections) on this P loading is not well understood or quantified. The objective of this study was to investigate long-term P loading to Nottawasaga Bay from groundwater plumes of decommissioned septic systems in the coastal community of Wasaga Beach. Detailed groundwater sampling characterized a legacy P plume from a septic system decommissioned 35 years ago, revealing elevated soluble reactive phosphorus concentrations extending over 40 m and reaching the shoreline. Sorption and dispersion parameters required for neighbourhood-scale modeling were derived by simulating this persistent, long but thin P plume using a numerical model. Numerical simulations of P plumes from > 800 septic systems, 0.01-1.6 km from the shoreline and active < 65 years before decommissioning, revealed P mass discharge to the lake started after similar to 30 years and will continue for > 4000 years. Relatedly, the extended P transport meant the annual mass discharge rate to the lake was consistently < 1.3 % of the annual mass input rate from septic systems to the aquifer, though it varied over time according to the septic systems' distance from shore. These findings highlight the long-term view required in accounting for decommissioned septic systems in P management strategies to protect lake water quality.

ABSTRACT Domestic wastewater is a source of persistent organic pollutants and pathogens to the aquatic environment, including groundwater aquifers. Wastewater contaminants include a variety of personal care products, pharmaceuticals, endocrine disrupters, bacteria, and viruses. Groundwater from 22 wells completed in a semi-confined to confined, fractured Silurian dolostone aquifer in southern Wellington County, Ontario, Canada, was analyzed for 14 organic wastewater contaminants (4 artificial sweeteners, 10 pharmaceuticals) as well as E. coli , total coliforms, and 6 human enteric viruses. Enteric viruses were detected in 8.6% of 116 samples, and at least one organic wastewater contaminant was detected in 82% of the wells (in order of decreasing detection frequency: acesulfame, ibuprofen, sulfamethoxazole, triclosan, carbamazepine, and saccharin). Virus indicator metrics [positive and negative predictive values (PPV, NPV), sensitivity, specificity] were calculated at the sample and well level for the organic wastewater compounds, E. coli , and total coliforms. Fecal bacteria were not good predictors of virus presence (PPV = 0%–8%). Of the potential chemical indicators, triclosan performed the best at the sample level (PPV = 50%, NPV = 100%), and ibuprofen performed the best at the well level (PPV = 60%, NPV = 67%); however, no samples had triclosan or ibuprofen concentrations above their practical quantification limits. Therefore, none of the compounds performed sufficiently well to be considered reliable for assessing the potential threat of enteric viruses in wastewater-impacted groundwater in this bedrock aquifer. Future studies need to evaluate the indicator potential of persistent organic wastewater contaminants in different types of aquifers, especially in fractured rock where heterogeneity is strong. IMPORTANCE Assessing the potential risk that human enteric viruses pose in groundwater aquifers used for potable water supply is complicated by several factors, including: (i) labor-intensive methods for the isolation and quantification of viruses in groundwater, (ii) the temporal variability of these viruses in domestic wastewater, and (iii) their potentially rapid transport in the subsurface, especially in fractured rock aquifers. Therefore, aquifer risk assessment would benefit from the identification of suitable proxy indicators of enteric viruses that are easier to analyze and less variable in wastewater sources. Traditional fecal indicators (e.g., E. coli and coliforms) are generally poor indicators of enteric viruses in groundwater. While many studies have examined the use of pharmaceutical and personal care products as tracers of domestic wastewater and fecal pollution in the environment, there is a paucity of data on the potential use of these chemical tracers as enteric virus indicators, especially in groundwater.

The Minesing Wetlands contain Canada’s only known population of Hine’s emerald dragonfly (HED), a species listed as endangered since 2011. The HED relies on groundwater discharge areas to complete its life cycle and therefore the population is potentially sensitive to changes in groundwater. The goals of this study were to, (1) characterize groundwater discharge to the eastern portion of the Minesing Wetlands, which corresponds to the HED habitat, (2) document any significant changes in groundwater chemistry since a 1998 study, and (3) establish a comprehensive baseline water quality data set to facilitate future comparisons. Consistent with HED habitats in the USA, the chemistry of Minesing Wetlands groundwater was dominated by calcium and bicarbonate. Low groundwater chloride and nitrate concentrations indicated minimal impact from activities such as road salt application and fertilizer use. Nitrate was generally only found at the wetland margin where groundwater springs emerged from the base of the uplands. The dominant form of inorganic nitrogen in wetland groundwater was ammonium (max = 1.5 mg N/L), and soluble reactive phosphorus concentrations were also relatively high (8 to 122 µg/L). The most consistent change in wetland groundwater chemistry between the 1998 study and samples collected in 2015–2017 was that chloride concentrations in 10 of 23 wells increased over time, however the maximum mean well chloride (11.6 mg/L) remained well below the guideline for the protection of aquatic life (120 mg/L). To preserve HED habitat, potential impacts to groundwater need to be considered during land use planning for the recharge area.

Analyses of the sulfur (S) and oxygen (O) stable isotope ratios (δ34SSO4 and δ18OSO4) of dissolved sulfate in streams and groundwater in Yukon Territory (YT, Canada) indicates that the dominant source of this major ion is oxidation of sulfide minerals. In these streams, sulfate has a large range in δ34SSO4 values (−19 to +10‰), which is consistent with the lower portion of the documented range for δ34S of sulfides in bedrock in YT (∼ −35 to +55‰). Furthermore, the large majority of the δ18OSO4 data plot within the expected field for sulfide oxidation in cross-plots of δ18OSO4 versus the O isotope ratios of ambient water (δ18OH2O). δ18OSO4 values plotting above that field were likely affected by microbial sulfate reduction, a process that enriches residual sulfate in both 34S and 18O. The stable isotope data indicate that dissolution of marine-evaporite gypsum, which has higher δ34SSO4 and δ18OSO4 values compared to all stream and groundwater samples, is a minor to negligible source of sulfate in the YT streams. Association of sulfate with other solutes indicates release of various metals from sulfide minerals, and suggests dominance of silicate weathering in response to oxidation of the sulfides. Variations in sulfate concentrations in YT streams are largely related to geology, while terrain and climate, including extent of permafrost, are also important factors. Long-term monitoring of sulfate concentrations in YT streams indicates increasing concentrations over time, in both streams impacted by mining, and in streams that have not been affected much by historical in-catchment industrial activities. Increases were largest in streams in northern and central-south YT in the presence of thawing continuous to discontinuous permafrost.

The stable isotope ratios of groundwater sulfate (34 S/32 S, 18 O/16 O) are often used as tracers to help determine the origin of groundwater or groundwater contaminants. In agricultural watersheds, little is known about how the increased use of sulfur as a soil amendment to optimize crop production is affecting the isotopic composition of groundwater sulfate, especially in shallow aquifers. We investigated the isotopic composition of synthetic agricultural fertilizers and groundwater sulfate in an area of intensive agricultural activity, in Ontario, Canada. Groundwater samples from an unconfined surficial sand aquifer (Lake Algonquin Sand Aquifer) were analyzed from multi-level monitoring wells, riverbank seeps, and private domestic wells. Fertilizers used in the area were analyzed for sulfur/sulfate content and stable isotopic composition (δ18 O and/or δ34 S). Fertilizers were isotopically distinct from geological sources of groundwater sulfate in the watershed and groundwater sulfate exhibited a wide range of δ34 S (-6.9 to +20.0‰) and δ18 O (-5.0 to +13.7‰) values. Quantitative apportionment of sulfate sources based on stable isotope data alone was not possible, largely because two of the potential fertilizer sulfate sources had an isotopic composition on the mixing line between two natural geological sources of sulfate in the aquifer. This study demonstrates that, when sulfate isotope analysis is being used as a tracer or co-tracer of the origin of groundwater or of contaminants in groundwater, sulfate derived from synthetic fertilizer needs to be considered as a potential source, especially when other parameters such as nitrate independently indicate fertilizer impacts to groundwater quality.

Abstract Dissolved organic matter (DOM) represents a mixture of organic molecules that vary due to different source materials and degree of processing. Characterizing how DOM composition evolves along the aquatic continuum can be difficult. Using a size‐exclusion chromatography technique (liquid chromatography‐organic carbon detection [LC‐OCD]), we assessed the variability in DOM composition from both surface and groundwaters across a number of Canadian ecozones (mean annual temperature spanning −10°C to +6°C). A wide range in DOM concentration was found from 0.2 to 120 mg C L−1. Proportions of different size‐based groupings across ecozones were variable, yet similarities between specific waterbody types, regardless of location, suggest commonality in the processes dictating DOM composition. A principal component analysis identified 70% of the variation in LC‐OCD derived DOM compositions could be explained by the waterbody type. We find that DOM composition within a specific waterbody type is similar regardless of the differences in climate or surrounding vegetation where the sample originated from.

Intensive agriculture and growing human populations are important nitrogen (N) sources thought to be associated with eutrophication. However, the contribution and seasonality of N delivery to streams from human activities is poorly understood and knowledge of the role of stream communities in the assimilation of N from human activities is limited. We used N and oxygen stable isotope ratios of dissolved inorganic N (DIN) and concentrations of artificial sweeteners to identify the relative contribution of key sources of anthropogenic N (i.e., fertilizers, human, and livestock waste) to tributaries of the Red River Valley (RRV), Manitoba, Canada. Water and algae were sampled in 14 RRV tributaries during snowmelt, spring, summer, and autumn; and water was sampled at three locations in the Red River in spring, summer, and autumn. δ15N values of DIN in tributary water differed seasonally and were greatest during snowmelt. Incorporation of ammonium δ15N provided evidence for the importance of manure N to tributaries during snowmelt. Fertilizer and municipal lagoons served as principal sources of N to streams in spring and summer. Human and livestock waste sources of N were the dominant contributor to algae at greater than 90% of sites and algae δ15N was greatest at sites downstream of municipal lagoons. We also showed that the tributaries contribute human and livestock waste N to the Red River, though much of the nitrate in the river originates outside of Manitoba. Overall, our study determined that the anthropogenic sources of N to RRV streams vary seasonally, likely due to regional hydrologic conditions. Our study also showed the potential of artificial sweeteners and ammonium δ15N as tools for identifying N sources to rivers. Moreover, we demonstrate the need for the management of N sources and the protection of stream function to control downstream transfer of N from landscapes to waterbodies.

Artificial sweeteners are ubiquitous constituents of sanitary wastewater and are not completely attenuated by wastewater treatment processes. Consequently, artificial sweeteners are increasingly employed as a tool to detect wastewater and help evaluate its impact on aquatic environments. In rural areas, septic systems are known point sources of artificial sweeteners to groundwater, however the potential contribution of artificial sweeteners to streams via this pathway is unknown. We analyzed 294 samples from 173 stream sites in Southern Ontario, Canada, for acesulfame, saccharin, cyclamate, and sucralose, and found that 91% had one or more of these compounds present. The stream sites sampled did not have municipal wastewater treatment plants upstream and therefore septic system effluent was the most likely source of the artificial sweeteners. Acesulfame, which is the most recalcitrant of the four artificial sweeteners, was by far the most ubiquitous in streams, with a 91% detection frequency, compared to 27, 8, and 3% for saccharin, cyclamate, and sucralose, respectively. Stream concentrations ranged from non-detectable to maximum values of 225, 380, 204, and 291 ng/L for acesulfame, saccharin, cyclamate, and sucralose, respectively. We calculated that water from septic effluent contributed from < 0.005% to 0.5% of streamflow, with a median value of 0.052%. This is the first study to use artificial sweetener concentrations to quantify the fraction of septic effluent reaching streams via groundwater flow. Using acesulfame concentrations, we calculated that approximately 13% of the septic effluent water generated in rural Southern Ontario eventually ends up in local streams, however this is likely a conservative estimate. Although much of the septic effluent water that reaches streams may be free of all but the most persistent anthropogenic compounds, our study demonstrates that domestic wastewater must be considered as a possible source of stream contaminants, even if municipal wastewater is not a factor in the watershed.

For more than a decade the artificial sweeteners acesulfame (ACE) and sucralose (SUC) have been applied as tracers of the input of wastewater to environmental waters. Recently concerns have been raised that degradation of ACE during treatment may hinder or restrict its use as a wastewater tracer. In this study the value of ACE and SUC as tracers was reassessed based on samples of wastewater at 12 municipal wastewater treatment (MWWT) plants and from 7 septic systems and associated septic plumes in groundwater. The results indicated stability of SUC during MWWT at most plants, and variable removal of both sweeteners during some MWWT and in the septic wastewater systems. However, the residual concentrations of ACE and SUC in municipal effluent and in septic plumes indicate that both sweeteners remain valuable wastewater tracers. The mass ratio SUC/ACE was found to be a useful parameter for examining the relative persistence of these sweeteners.

Municipal wastewater lagoons are common across North America and, unlike larger mechanical wastewater treatment plants, typically release nutrient-rich effluent directly to rivers in intermittent pulses. However, little is known about the fate of nutrients from these episodic events, which may happen under varying hydrologic or thermal conditions. We assessed fate of nitrogen (N) and phosphorus (P) from lagoon effluent during three releases to Deadhorse Creek, Manitoba, Canada. Using net nutrient uptake lengths and natural abundance stable isotope ratios of dissolved inorganic nitrogen (DIN) and primary producers, we found that DIN was processed during the summer releases though the dominant mechanism was unclear. However, nitrate was largely exported in autumn. Primary producers assimilated lagoon N but did not appear to reduce DIN concentrations. The longitudinal pattern of soluble reactive phosphorus (SRP) varied between releases and in summer 2019 the stream became a net source of SRP despite concomitant processing of DIN. We hypothesize that low demand for P in Deadhorse Creek, as suggested by upstream SRP > 0.05 mg P L-1, and nutrient ratios indicative of N limitation, reduced instream processing of P. Furthermore, our results indicated that cool or high flow conditions may result in the export of much of the lagoon nutrient load downstream. Our findings suggest the processes that transform wastewater nutrients are overwhelmed during effluent releases. Managers should consider increasing effluent dilution via continuous release of effluent rather than pulsed delivery. However, management of upstream nutrient supply may also be needed when relying upon the self-purifying capacity of rivers.

Dissolved Organic Matter (DOM) represents a mixture of organic molecules that vary due to different source materials and degree of processing. Characterizing how DOM composition evolves along the aquatic continuum can be difficult. Using a size-exclusion chromatography technique (LC-OCD), we assessed the variability in DOM composition from both surface and groundwaters across a number of Canadian ecozones (mean annual temperature spanning -10 to +6 C). A range in DOM concentration was found from 0.2 to 120 mg C/L. Proportions of different size-based groupings across ecozones were variable, yet similarities between specific hydrologic compartments, regardless of location, suggest commonality in the processes dictating the evolution of DOM composition. A principal-component analysis identified 70% of the variation in LC-OCD derived DOM compositions could be explained by the hydrological compartment. We find that hydrologic compartment has a greater influence on DOM composition than differences in climate or surrounding vegetation.

Groundwater nutrient loading to L Huron was assessed along a 1.7 km section of beach at Grand Bend, ON, Canada, where septic systems are used for wastewater disposal. The artificial sweetener acesulfame (ACE) was detected in all groundwater samples (7-842 ng/L, n = 78), revealing that the entire surficial aquifer was impacted by septic system wastewater. Nitrate concentrations (3.5 +/- 1.4 mg/L, n = 78) were correlated with ACE (r(2) = 0.54), indicating that septic systems contribute to nitrate loading in the aquifer. Chloride was also elevated (37 +/- 11 mg/L, n = 78), but was not correlated with ACE (r(2) = 0.008), indicating a non wastewater source was dominant, likely road salt. Soluble reactive phosphorus (SRP) values were low (53 +/- 9.3 mu g/L, n = 77) and were not correlated with ACE (r(2) = 0.006). Sediment profiling below two of the septic system drain-fields, showed that the sand grains had distinct secondary coatings containing P, indicating that mineral precipitation reactions played a role in limiting P concentrations present in the aquifer. Groundwater nutrient loading to the lake was estimated at 13,000 kg N/year from NO and 1.9 kg P/year from SRP. These amounts are insignificant compared to nutrient loading from a stream that drains an agricultural catchment and discharges to the lake at the north end of the study site (Parkhill Creek). This calls into question, in some cases, the rationale of decommissioning properly functioning septic systems as a mitigation measure for reducing nutrient loading to nearby water courses. (C) 2019 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved.

Little is known about the regional extent and variability of nitrate from atmospheric deposition that is transported to streams without biological processing in forests. We measured water chemistry and isotopic tracers (δ18O and δ15N) of nitrate sources across the Northern Forest Region of the U.S. and Canada and reanalyzed data from other studies to determine when, where, and how unprocessed atmospheric nitrate was transported in catchments. These inputs were more widespread and numerous than commonly recognized, but with high spatial and temporal variability. Only 6 of 32 streams had high fractions (>20%) of unprocessed atmospheric nitrate during baseflow. Seventeen had high fractions during stormflow or snowmelt, which corresponded to large fractions in near-surface soil waters or groundwaters, but not deep groundwater. The remaining 10 streams occasionally had some (<20%) unprocessed atmospheric nitrate during stormflow or baseflow. Large, sporadic events may continue to be cryptic due to atmospheric deposition variation among storms and a near complete lack of monitoring for these events. A general lack of observance may bias perceptions of occurrence; sustained monitoring of chronic nitrogen pollution effects on forests with nitrate source apportionments may offer insights needed to advance the science as well as assess regulatory and management schemes.

It has been widely documented that municipal wastewater treatment plant effluents are a major source of artificial sweeteners to surface waters. However, in rural areas, the extent to which septic systems contribute these same compounds to groundwater aquifers is largely unknown. We examined the occurrence of four commonly used artificial sweeteners in an unconfined sand aquifer that serves as a water supply for rural residents, as a receptor of domestic wastewater from septic systems, and as a source of baseflow to the Nottawasaga River, ON, Canada. Groundwater from the Lake Algonquin Sand Aquifer in the southern Nottawasaga River Watershed was collected from private domestic wells and as groundwater seeps discharging along the banks of the Nottawasaga River. Approximately 30% of samples had detectable levels of one or more artificial sweeteners, indicating the presence of water derived from septic system effluent. Using acesulfame concentrations to estimate the fraction of septic effluent in groundwater samples, ∼3.4 to 13.6% of the domestic wells had 1% or more of their well water being derived from septic system effluent. Similarly, 2.0 to 4.7% of the groundwater seeps had a septic effluent contribution of 1% or more. No relationship was found between the concentration of acesulfame and the concentration of nitrate, ammonium, or soluble reactive phosphorus in the groundwater, indicating that septic effluent is not the dominant source of nutrients in the aquifer. It is expected that the occurrence of artificial sweeteners in shallow groundwater is widespread throughout rural areas in Canada.

This study reports the first comprehensive data set of characteristic concentrations of four artificial sweeteners: acesulfame (ACE), sucralose (SUC), saccharin (SAC), and cyclamate (CYC), and their ratios with nutrients, for untreated septic system wastewater. Samples were collected from the tanks of 19 different septic systems from across Ontario, Canada; these had a variety of usages, from single‐family cottages to multiple‐dwelling (campground or resort) facilities and had no additional treatment systems. The artificial sweetener concentrations and their relative proportions were highly variable in some cases, both temporally for several individual tanks and from site‐to‐site. Variability tended to be lower for multiple‐dwelling compared to single‐dwelling systems. This variability likely reflects differing use of artificial sweetener‐containing products. The median concentrations for the complete data set of all four artificial sweeteners (in a range of 10 to 60 μg/L) were of a similar order of magnitude, but slightly higher, than has generally been reported for wastewater treatment plant influent (though these vary substantially globally). Both SUC and ACE provided adequate positive linear relationships for dissolved nitrogen and phosphorus in the septic tanks, while a summation of ACE and SUC concentrations also gave a strong correlation. In contrast, CYC and SAC showed poor linear correlation with these nutrients. These reported ranges for artificial sweetener concentrations and ratios with nutrients may be used in future studies to estimate the contributions of nutrients or other wastewater constituents (e.g., pharmaceuticals, bacteria, and viruses) from domestic septic systems to groundwater, including water supply or irrigation wells, and nearby surface water bodies.

Groundwater inputs of phosphorus (P) to the Laurentian Great Lakes are poorly known, but may contribute to eutrophication and algal bloom issues. This study's objective was to assess the contribution of P to Nottawasaga Bay from the surficial sand aquifer at Wasaga Beach, representing a coastal cottage area with decommissioned septic systems, and how this might change with time. The first part of the study involved site-scale groundwater sampling beside 4 provincial park public washrooms. Legacy P plumes were detected at two of these sites, with one being >30years since decommissioning. P transport calculations including sorption onto aquifer sediments indicate the majority of P plumes from the town's decommissioned septic systems have likely not yet reached the shoreline, >50years since installation, and will likely contribute P to the bay for many decades. The second part of the study consisted of broader-scale (town-wide) surveys of shallow beach groundwater. Dissolved P concentrations were ~50μg/L for background groundwater (in town and reference area), which is similar to literature values. This P may have been sourced from degrading organic matter, bird droppings, or soil-aquifer minerals. Sporadic elevated concentrations up to 420μg/L may be from legacy septic systems and/or natural sources. A rough calculation suggests groundwater P loading along Nottawasaga Bay's eastern shore (Wasaga Beach, 10-km; adjacent similar beaches, 40-km) is a few percent at most of that from the Nottawasaga River. Thus, it more likely affects localized periphyton and macrophyte growth rather than significantly affecting the Nottawasaga Bay P budget.

Understanding the fate of applied nitrogen (N) in agricultural soils is important for agronomic, environmental, and human health reasons, but it is methodologically difficult to study at the field scale. Natural abundance stable isotope measurements (δ15N) were used in this field study with micrometeorological measurements of nitrous oxide (N2O) emissions to identify the biogeochemical processes responsible for rapid N transformations immediately after application of liquid dairy manure. Fifteen samplings occurred between 16 Mar. 2012 and 5 Apr. 2013, with a focus on spring manure application (before and after) and a winter snowmelt period. Concentrations and δ15N values of ammonium (NH4+), nitrate (NO3−), N2O, and total N were measured throughout the year. Approximately 56 (±7)% of the NH4+–N applied in the spring could not be accounted for 3 d after manure application and was presumably lost by ammonia volatilization before it was tilled into the soil and/or removed from the inorganic N pool by microbial assimilation. Almost all of the remaining manure‐NH4+ (95 ± 1.1%) was converted within 3 wk to NO3− and N2O by nitrification and nitrifier‐denitrification, respectively. The in situ 15N isotope effect for nitrification (εNitrate–Ammonium) was calculated to be −32.0 (±5.3)‰. Overall, field‐scale measurements of δ15N at natural abundance provided valuable information that was used to distinguish sources of NH4+ (manure vs. soil organic N) and to follow the production and consumption of NO3− and the pathways of N2O production in soil.Core Ideas 49 to 63% of the manure‐NH4+ was volatilized, assimilated, or fixed 3 d after application. Only 4 to 6% of the manure‐NH4+ remained in the soil 3 wk after application. In situ 15N isotope effect for nitrification (ammonium–nitrate) was −32.0 (±5.3)‰. N2O emissions after manure application were produced by nitrifier‐denitrification. The δ15N values showed active production and consumption of NO3− during winter thaw.

BACKGROUND The objective of this study was to investigate the occurrence of glyphosate residues (glyphosate and its metabolite AMPA) in shallow groundwater in a catchment dominated by agriculture, and to examine the potential for this groundwater to store and transmit these compounds to surface waters. RESULTS Glyphosate residues were found in some of the groundwater samples collected in riparian (surface seeps), upland (mostly <20 m below ground) and wetland settings (<3 m below ground). Overall, glyphosate and AMPA were detected in 10.5 and 5.0%, respectively, of the groundwater samples analyzed as part of this study. All concentrations of glyphosate were well below Canadian guidelines for drinking water quality and for protection of aquatic life. Seasonal differences in concentrations in riparian seeps were possibly related to cycles of weather, herbicide application and degradation of glyphosate. Highest concentrations were at upland sites (663 ng L(-1) of glyphosate, 698 ng L(-1) of AMPA), apparently related to localized applications. Most glyphosate detections in wetlands were >0.5 km distant from possible areas of application, and, combined with other factors, suggest an atmospheric transport and deposition delivery mechanism. In both upland and wetland settings, highest glyphosate concentrations were sometimes not at the shallowest depths, indicating influence of hydrological factors. CONCLUSION The glyphosate/AMPA detections in riparian seeps demonstrated that these compounds are persistent enough to allow groundwater to store and transmit glyphosate residues to surface waters. Detections in the wetlands support earlier evidence that atmospheric transport and deposition may lead to glyphosate contamination of environments not intended as targets of applications. This interpretation is further supported by detections of both glyphosate and AMPA in precipitation samples collected in the same watershed. © 2016 Her Majesty the Queen in Right of Canada Pest Management Science © 2016 Society of Chemical Industry.