Occurrence of Urban-Use Pesticides and Management with Enhanced

Books and Reference; News ..... Samples were stored at 4 °C until processing within 7 d. ... The watershed was broken into six subcatchments draining...
2 downloads 0 Views 2MB Size
Download PDF
Article Cite This: Environ. Sci. Technol. XXXX, XXX, XXX−XXX

pubs.acs.org/est

Occurrence of Urban-Use Pesticides and Management with Enhanced Stormwater Control Measures at the Watershed Scale Jordyn M. Wolfand,†,‡,∥ Carolin Seller,†,‡,§ Colin D. Bell,†,∥ Yeo-Myoung Cho,†,‡ Karl Oetjen,∥,⊥ Terri S. Hogue,†,∥ and Richard G. Luthy*,†,‡

Downloaded via UNIV OF EDINBURGH on March 23, 2019 at 05:04:22 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

NSF Engineering Research Center for Re-inventing the Nation’s Urban Water Infrastructure (ReNUWIt) and ‡Department of Civil and Environmental Engineering, Stanford University, Stanford, California 94305, United States § Department of Environmental Chemistry, Eawag, Swiss Federal Institute of Aquatic Science and Technology, 8600 Dübendorf, Switzerland ∥ Department of Civil and Environmental Engineering, Colorado School of Mines, Golden, Colorado 80401, United States ⊥ SCIEX, 1201 Radio Road, Redwood City, California 94065, United States S Supporting Information *

ABSTRACT: Urban-use pesticides are of increasing concern as they are widely used and have been linked to toxicity of aquatic organisms. To assess the occurrence and treatment of these pesticides in stormwater runoff, an approach combining field sampling and watershed-scale modeling was employed. Stormwater samples were collected at four locations in the lower San Diego River watershed during a storm event and analyzed for fipronil, three of its degradation products, and eight pyrethroids. All 12 compounds were detected with frequency ranging from 50 to 100%. Field results indicate pesticide pollution is ubiquitous at levels above toxicity benchmarks and that runoff may be a major pollutant source to urban surface waters. A watershed-scale stormwater model was developed, calibrated using collected data, and evaluated for pesticide storm load and concentrations under several management scenarios. Modeling results show that enhanced stormwater control measures, such as biochar-amended biofilters, reduce both pesticide storm load and toxicity benchmark exceedances, while conventional biofilters reduce the storm load but provide minimal toxicity benchmark exceedance reduction. Consequently, biochar amendment has the potential to broadly improve water quality at the watershed scale, particularly when meeting concentration-based metrics such as toxicity benchmarks. This research motivates future work to demonstrate the reliability of full-scale enhanced stormwater control measures to treat pollutants of emerging concern.

■

INTRODUCTION Urban stormwater is a major cause of water quality impairment. While pathogens, nutrients, metals, and suspended sediments are typical pollutants of concern, there is increasing awareness of trace organic contaminants (e.g., detergents, pharmaceuticals, household products) and their impact on downstream water quality.1 Urban-use pesticides are particularly of interest since they are commonly found in stormwater runoff2−4 and contribute to stream toxicity.5−7 Due to their broad toxicity, organophosphate pesticides have been phased out, replaced in popularity by the phenylpyrazole compound, fipronil, and the pyrethroid class of pesticides, such as bifenthrin. Fipronil is primarily used for structural pest control and also for flea and tick prevention in pets. In California, fipronil is not registered for agricultural uses,8 but it is found in agricultural and urban uses throughout the rest of the US as well as globally.9 Fipronil is relatively recalcitrant to degradation under © XXXX American Chemical Society

natural environmental processes (hydrolysis half-life >100 d at pH = 7) and frequently detected in urban waterways.2,10,11 The solubility of fipronil ranges from 1.9 to 2.4 mg/L, and fipronil is moderately hydrophobic with a log Kow of about 4.0.12 Its transformation products, which are also often found in the environment, include the following: fipronil sulfone (oxidation byproduct), fipronil desulfinyl (photolysis byproduct), and fipronil sulfide (reduction byproduct).2,10,11 Fipronil degradation products are detected as much or more frequently than the parent compound11 and are generally more persistent in stream sediments.13 Urban runoff is hypothesized to be a primary source of fipronil to urban water bodies.2,11 Received: October 17, 2018 Revised: February 16, 2019 Accepted: February 19, 2019

A

DOI: 10.1021/acs.est.8b05833 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

Figure 1. Location map and sampling locations within the lower San Diego River Watershed.

Because storm runoff is considered a large source of these urban pesticides to the environment, we hypothesize that total pesticide loads could be controlled by targeting treatment of stormwater. Stormwater control measures (SCMs), also known as green infrastructure (GI) or best management practices (BMPs), are structural control measures that capture, delay, and treat urban runoff. Consequently, SCMs can be used to provide flood control, reduce peak flows, increase groundwater recharge, and improve water quality. Examples include green roofs, biofilters, permeable pavement, and vegetated swales. In addition to the water quantity and quality benefits, SCMs provide various other environmental, social, and economic benefits such as green space and reducing urban heat island effect.30 Because of the benefits SCMs can provide urban centers, many cities have made or plan to make large investments in SCMs. For example, Los Angeles plans to invest over $6 billion (2015 dollars) by 2037 to address stormwater management in the Los Angeles River watershed with neighborhood, street, and regional scale SCMs.31 The City of San Diego has estimated $3.1 billion (2015 dollars) is needed to comply with flood protection and water quality requirements by 2035 and proposes nonstructural management strategies in addition to installation of SCMs.32 Given the magnitude of investment planned in stormwater management, researchers and practitioners continue to innovate SCM technologies, particularly with the aim of improving water quality. For example, iron-amended sand filters provide phosphate removal,33 woodchip biofilters enhance nitrate removal,34 and fungi-enhanced systems can reduce pathogens and pesticides.35,36 Of particular promise are biochar-amended biofilters.37 Biochar, a derivative of organic matter that has been processed under high temperature and pressure, has extensive pore space and available sites to sorb a variety of pollutants. More cost-effective than activated

Pyrethroids, similar in structure to the natural pyrethrins found in pyrethrum flowers, are in the top 10 classes of pesticides used globally for both agricultural and urban applications.14 Pyrethroids are commonly found in consumer products used in and outside the home for insect control. Common chemicals in this class include bifenthrin, permethrin, cypermethrin, cyfluthrin, and deltamethrin. Pyrethroids are hydrophobic and tend to accumulate in sediment and biota (log Kow ranging from 4.5 to 7.0).15 Pyrethroids have been documented in urban streams at acutely toxic concentrations,5,16−23 and several studies have identified urban runoff as the primary source.3,4,24,25 For example, sampling around Sacramento, CA, found in-stream sediment concentrations of bifenthrin of 6.4−17 ng/g dry, while sediment from storm drain mouths ranged in concentration from 238 to 744 ng/g dry. The same study estimated that a single 3 h storm contributed as much bifenthrin mass as six months of dry weather runoff (primarily due to irrigation).4 Regulation of pesticides varies by country and by state, but fiproles (fipronil and its transformation products) and pyrethroids are increasingly pollutants of regulatory concern. Recently, several total maximum daily loads (TMDLs) have been established for pyrethroids in mixed land use watersheds in California, including the Salinas, Sacramento, Santa Maria, and San Joaquin River watersheds.26−28 In the Ballona Creek watershed (Los Angeles County, CA), a TMDL was originally established for aquatic toxicity due to legacy pollutants such as polychlorinated biphenyls (PCBs) and dichlorodiphenyltrichloroethane (DDT), but fiproles and pyrethroids have been recently determined to be the primary cause of aquatic toxicity in that watershed.29 Similar patterns may follow in other urbanized watersheds where legacy pollutants have been phased out but replaced with fiproles and pyrethroids. B

DOI: 10.1021/acs.est.8b05833 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology carbon,37 it has been employed in laboratory- and mesocosmscale experiments to successfully reduce nutrients, metals, trace organic contaminants, and pathogens.37−42 Modeling, laboratory, and pilot studies suggest biochar-amended filters are able to remove greater than 99% of urban-use pesticides.40,43 This research seeks to measure the occurrence of urban-use pesticides in urban runoff and the subsequent reduction of pesticide loading and in-stream toxicity achievable by implementing conventional and biochar-amended biofilters at the watershed scale. We use the lower San Diego River watershed as a case study and examine the presence and distribution of urban-use pesticides during a winter storm event. We then develop a stormwater model to evaluate pesticide load and concentration reduction under the following conditions: (1) without management, (2) with conventional biofilters, and (3) with biochar-amended biofilters. Several studies have used computational modeling to predict pesticide transport via runoff and load,44−48 but to our knowledge, this is the first to consider removal of urban-use pesticides by SCMs at the watershed scale.

fecal indicator bacteria TMDL, responsible parties within the watershed have proposed a mix of nonstructural and structural control measures to attain the dry weather and wet weather load reductions by 2021 and 2031, respectively, with a total estimated cost of $483 million (2013 dollars).54 As SCMs are planned to be installed to treat fecal indicator bacteria, an opportunity exists to also treat pesticides, potentially reducing urban stream toxicity. The lower San Diego River watershed is an ideal case study for several reasons: (1) it is highly urbanized; (2) it is currently listed as a 303(d) impaired water for toxicity, and it is likely that urban-use pesticides are a toxicity contributor; and (3) relevant stakeholders plan to make considerable investments in SCMs in the near future. Field Sampling. Urban runoff was sampled during a winter storm event (February 17−18, 2017) at four sites in the lower San Diego River Watershed: three on the mainstem and one at the tributary Forester Creek (Figure 1, Table S1).55 Precipitation was measured with tipping bucket rain gages (Table S1) and flow was determined from stage and velocity measurements (Figure S3).55 Precipitation at the four sites ranged from 27−35 mm (1.06−1.37 in) over about 12 h (Table S1). Storm samples at Fashion Valley, Forester Creek, and Mission Trails were collected using peristaltic pumps and polytetrafluoroethylene (PTFE) tubing with a stainless-steel intake screen mounted to the channel bottom. At Mission Road, samples were taken manually by dunking sample bottles under the water surface due to field equipment malfunction. Every 1−2 h throughout the storm event, samples were collected in 4 or 2.5 L solvent-rinsed amber glass bottles for pesticide analysis, for a total of 29 samples, including field replicates. A paired 1 L grab sample was also taken at every collection time for analysis of dissolved organic carbon (DOC) and total suspended solids (TSS). Samples were stored at 4 °C until processing within 7 d. Sample Processing and Analytical Method. Samples were collected, processed, and analyzed with methods adapted from USGS Techniques and Methods 5-C2.56 Twelve target analytes were quantified: fipronil, fipronil sulfone, fipronil sulfide, fipronil desulfinyl, bifenthrin, cypermethrin, cyfluthrin, deltamethrin, esfenvalerate, fenpropathrin, λ-cyhalothrin, and permethrin. Briefly, stormwater grab samples were centrifuged at 2000 rpm for 15 min and filtered with 0.7 μm Whatman glass microfiber filters (Grade GF/F; MilliporeSigma, St. Louis, MO). The solids and filtrate were extracted separately to assess the distribution of pesticides in the solid versus liquid phase. Liquid-phase target analytes were extracted from the filtrate with solid-phase extraction (SPE). The filtered suspended sediments were spiked with the surrogate standard (d5bifenthrin; (±)-bifenthrin-(biphenyl-2′,3′,4′,5′,6′-d5); MilliporeSigma, St. Louis, MO ) and digested using microwaveassisted extraction (MAE; MARS Xpress from CEM, Matthews, NC). Matrix interferences were removed by acidactivated copper and stacked SPE cartridges. Samples were then concentrated to 300 μL, and internal standards (d10phenanthrene and d10-pyrene; MilliporeSigma, St. Louis, MO) were added. Sample extracts were quantified by gas chromatography−tandem mass spectrometry (GC−MS/MS). Detailed extraction and analytical methods are found in the Supporting Information. The limit of quantification (LOQ) was estimated at the lowest concentration standard (in-vial concentration of 3.9

■

METHODS Study Site. The lower San Diego River watershed, located in San Diego County, CA, is roughly 445 km2 and drains to the Pacific Ocean (Figure 1). Land use is 30% residential, 12% transportation (e.g., roadways, parking lots, and rights-of-way), 11% commercial, institutional or industrial, and 44% open space or designated for parks and recreation (Figure S1). About 26% of the watershed area is impervious, and the watershed has separate storm and sanitary sewers. Precipitation in this region is highly seasonal, with most rain falling between October and April, for an average annual rainfall (water years 1987−2016) of about 242 mm (9.54 in).49 Precipitation for the 2017 water year was 133% of this annual average with 323 mm (12.7 in).49 Toxicity in freshwater, freshwater sediments, and marine sediments is common in the San Diego region. Stormwater effluent appears to be a significant source of toxicity; 90% of 20 permitted NPDES facilities in the San Diego region showed one or more instances of effluent toxicity in monitoring data from 2001 to 2008.50 While historically organophosphate pesticides appeared to be the cause of such toxicity, pyrethroid pesticides, particularly bifenthrin and cypermethrin, are increasingly detected in the region now that organophosphate use is restricted.50 Within the San Diego River watershed, pyrethroids and fiproles have been detected in the 10s to 100s ng/g range in sediments.51 Water concentrations within the watershed, particularly during dry weather, tend to be below detection limits.51 Patterns are difficult to elucidate because of limited monitoring during wet weather and varied detection limits, but a previous study52 in water year 2012 found levels of bifenthrin in San Diego River sediments ranging from ND ( fipronil desulfinyl > fipronil sulfide) is typical of environmental samples from shallow and flowing surface waters11 where oxidation and photolysis are key degradation processes. Pyrethroids were consistently found in liquid- and solidphase stormwater samples (88% detection), but with less regularity than fiproles (100% detection). Whole-sample concentrations were within range of other reported runoff concentrations in California.3,4,25,66 For example, we observed a median whole-sample bifenthrin concentration of 7.5 ng/L,

Figure 4. Concentration of fiproles and bifenthrin in solid- and liquid-phase stormwater sampled in the lower San Diego River watershed during a February 2017 storm event. Midline of the box represents the median of observed values; lowest bound (left) and highest bound (right) of the box represent the 25th and 75th percentiles, respectively. Left and right whiskers represent the 5th and 95th percentiles, respectively. Points represent data below and above the 5th and 95th percentiles. F

DOI: 10.1021/acs.est.8b05833 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

Figure 5. Whole-sample concentration of fiproles and bifenthrin in the lower San Diego River watershed at each sampling location during a February 2017 storm event. Asterisk (*) denotes statistical significance (p < 0.05) between locations using an ANOVA and Tukey’s Honest Significant Difference method. Midline of the box represents the median of observed values; lowest bound (left) and highest bound (right) of the box represent the 25th and 75th percentiles, respectively. Left and right whiskers represent the 5th and 95th percentiles, respectively. Points represent data below and above the 5th and 95th percentiles.

To assess whether bifenthrin and fipronil were in equilibrium in the sampled runoff, we compared the soil partition coefficient (KD) with the reaction quotient (Q) for the median observed liquid- and solid-phase concentrations (see the Supporting Information for detailed calculations). For both compounds Q > KD, and therefore, concentrations in runoff are not in equilibrium, in agreement with previous studies.25,67 In fact, flux of bifenthrin and fipronil is from the solid phase to the liquid phase as observed concentrations in the solid-phase are greater than expected assuming equilibrium. Pesticides can be applied in various formulations such as granules, liquids, wettable powders, and emulsifiable concentrates. The formulation, type of surface they are applied to, and precipitation event characteristics may affect pesticide washoff68,69 and therefore their distribution in the liquid vs solid phase. Future work should therefore consider both the liquid- and solid-associated fractions of pesticides in urban runoff as environmentally relevant. Pesticide Concentrations Correlate with Each Other but Not with Location or TSS. Correlation between whole-sample pesticide concentrations was determined with Kendall’s tau (τ) to accommodate the censored data set with multiple reporting limits.70 Concentrations of bifenthrin correlated strongly to concentrations of many of the other pyrethroids including cypermethrin, permethrin, deltamethrin, and cyfluthrin (Figure S5). Concentrations of fipronil correlated strongly to two of its degradation products, fipronil sulfone and fipronil desulfinyl, as well as to cypermethrin and λ-cyhalothrin (Figure S5). The strong correlations suggest that use patterns of pyrethroids and fipronil may be similar. Observed pesticide concentrations were determined to be log-normally distributed by a Shapiro−Wilk test for normality. One-way analysis of variance (ANOVA) was used to assess statistical differences between log-transformed concentrations at multiple sampling locations (α = 0.05). Tukey’s Honest Significant Difference method (α = 0.05) was used to test differences among log-transformed mean concentrations between each sampling location if ANOVA revealed significant differences (p < 0.05). Statistical tests revealed few significant differences between sites; the only differences were greater concentrations of fipronil and fipronil desulfinyl at Forester Creek compared to Mission Road (Figure 5). It is possible the greater concentrations of fiproles at the Forester Creek site are

due to different land use patterns among catchments. Forester Creek drains Cajon watershed, which is the most urbanized of the watersheds (44% impervious) with only 13% open space (vacant or parks/recreation) and 55% residential use. The area that drains to Mission Road, on the other hand, is only 20% impervious. Also, Forester Creek is an upstream tributary and drains only 14% of the watershed area, hence there is less of a chance for dilution. However, the overall similarity in concentrations among sites suggests pyrethroid and fipronil use is widely distributed throughout the watershed. TSS concentrations ranged from 8.67 to 113 mg/L (median = 51.8 mg/L; Figure S6) and were highly correlated to flow, exhibiting a first-flush phenomenon (Figure S7). DOC concentrations ranged from 4.3 to 21.6 mg/L (median = 8.9 mg/L; Figure S8). Whole-sample bifenthrin concentration is poorly correlated with TSS (R2 = 0.011). Previous studies25 also note TSS is not the primary determinant of pyrethroid concentration, and that timing of pesticide application or rainfall intensity are better determinants of pyrethroids in runoff.25 This, and that only 61% of bifenthrin mass was found in the solid phase, indicates pyrethroid management must include SCMs that both retain suspended solids (e.g., basins) and that allow for adsorption and biodegradation of dissolved pollutants (e.g., biofilters). Simulation of Pesticide Removal by Biofilters at the Watershed Scale. Infiltrating and/or Biochar-Amended Biofilters Provide Substantial Load Reduction. Total simulated loads for the single storm event for the lower San Diego River (435 km2) were 390 g of fipronil and 68 g of bifenthrin. These single event loads (0.90 g/km2 of fipronil and 0.16 g/km2 of bifenthrin) are within range of other studies.11,44 The total load for 10 pyrethroids in the Ballona Creek watershed (254 km2, Los Angeles County, CA) is estimated to be between 16 and 39 g/km2/yr.44 Annual storm load of fipronil to Salt Creek (2.6 km2, Orange County, CA) is estimated to be about 0.70 g/km2/yr.11 Simulated storm loads for all scenarios are shown in Table S4. Simulations from a hydrologic and water quality model show infiltrating conventional biofilters provide substantial load reduction: 55% and 79% reduction of load for fipronil and bifenthrin respectively, compared to baseline conditions with no management (Figure 6). Noninfiltrating (treat-and-release) conventional biofilters are less effective at reducing pesticide G

DOI: 10.1021/acs.est.8b05833 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

reduction of bifenthrin and fipronil toxicity benchmark exceedances, respectively, but were particularly sensitive to the tested range in performance (from 90 to 99%; Figure 7). However, only the noninfiltrating biochar-amended biofilters reduced concentrations below the toxicity benchmark for the duration of the storm event (Figure 7). This suggests that treatment is the best mechanism to meet concentration-based targets, such as toxicity exceedances. Only noninfiltrating biochar-amended biofilters both treat and release clean enough water to provide dilution of the untreated runoff in the channel. Load reduction, in contrast, can be achieved by infiltration of pollutant-laden water. It should be noted that runoff concentrations were compared to chronic invertebrate toxicity benchmarks, for a conservative approach. Toxicity benchmarks for fish (acute and chronic) and acute benchmarks for invertebrates are greater (110 and 800 ng/L for fipronil and bifenthrin, respectively). Previous results for removal of fecal indicator bacteria with enhanced SCMs differ from the results herein on urban-use pesticides. Infiltrating (both enhanced and conventional) SCMs were found to reduce fecal indicator bacterial load.71 Neither enhanced nor conventional SCMs were able to reduce indicator bacteria concentrations below concentration-based TMDL limits in the Ballona Creek watershed (Los Angeles County, CA) when 90% of the watershed runoff was treated.71 This contrast may be due to the large concentration range of fecal indicator bacteria in runoff (over 5 orders of magnitude) compared to urban-use pesticides (3 orders of magnitude).71 We conclude that water quality stakeholders need to identify priorities within their watershed and design SCMs accordingly. For example, in our simulation, infiltrating biofilters provide 45% runoff volume reduction compared to noninfiltrating biofilters (from 2.06 × 106 to 1.13 × 106 m3). This may provide flood risk reduction and increase water for groundwater recharge, depending on subsurface conditions, and both benefits may be advantageous depending on the end management goals. In this case, either conventional or biochar-amended biofilters can be employed to reduce load, while management of concentrations in runoff proves more difficult. If runoff reduction is not a concern, and stakeholders would like to maintain environmental flows, noninfiltrating biochar-amended biofilters can be deployed to reduce both runoff concentrations and load. In fact, under simulation conditions, noninfiltrating biochar-amended biofilters are the only way in which concentrations can be reduced enough to be below toxicity benchmarks consistently.

Figure 6. Percent storm load reduction of fipronil and bifenthrin under various watershed-scale biofilter simulation scenarios.

load: only 26% and 65% of load is removed for fipronil and bifenthrin, respectively, compared to baseline conditions (Figure 6). Biochar-amended biofilters, whether infiltrating or noninfiltrating, are effective at reducing both fipronil and bifenthrin load (82−93%), and always perform better than their conventional counterparts (Figure 6). While enhanced SCMs are known to perform better than conventional SCMs individually, we demonstrate that these benefits may also be observed at the watershed scale. Only Biochar-Amended Biofilters Can Meet Concentration-Based Toxicity Benchmarks. During the simulated February 2017 storm event, the aquatic life toxicity benchmarks for bifenthrin (1.3 ng/L) and fipronil (11 ng/L) were exceeded by 91% of modeled concentrations at 5 min time steps. Installing infiltrating conventional biofilters had no effect on toxicity benchmark exceedances; that is, the benchmarks were still exceeded for 91% of the duration of the storm event. Noninfiltrating (treat-and-release) conventional biofilters performed slightly better and reduced toxicity benchmark exceedances by 1−20% compared to the no-management baseline (Figure 7). Infiltrating biochar-amended biofilters performed moderately better, with 2−46% and 16−97%

■

ENVIRONMENTAL IMPLICATIONS High concentrations of urban-use pesticides were observed in the lower San Diego River watershed, illuminating the need for future monitoring and management. Simulations from a hydrologic and water quality model show that watershedscale installation of conventional and biochar-amended biofilters could substantially reduce pyrethroid loads. Achieving concentration-based targets proves more difficult without the implementation of enhanced (biochar-amended) systems. Depending on pollution and water management goals, various types of SCMs (enhanced vs conventional and infiltrating vs noninfiltrating) provide different benefits. When green infrastructure is installed for other primary benefits, such as reduced runoff quantity, aesthetics, and increased green space, thoughtful design of geomedia may provide significant water quality benefit as well. In addition, as pesticides are not in

Figure 7. Percent reduction of EPA aquatic life benchmark exceedances for fipronil and bifenthrin during a winter 2017 storm event under various watershed-scale biofilter simulation scenarios. H

DOI: 10.1021/acs.est.8b05833 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

■

ACKNOWLEDGMENTS We thank Ken Schiff and the team at the Southern California Coastal Water Research Project (SCCWRP) for organizing the field sampling, providing precipitation and flow data, and providing comments on the manuscript. In addition, we thank Patricia Gonzales and Kim Quesnel who assisted with the field work in San Diego. We are grateful to Chris Higgins for the use of his GC−MS/MS. Lastly, we thank Aniela Burant, Jennifer Teerlink, and the team at the California Department of Pesticide Regulation for assisting with interpretation of results and reviewing the manuscript. Project and personal support was provided by the National Science Foundation’s Engineering Research Center for Reinventing the Nation’s Urban Water Infrastructure (ReNUWIt, NSF ERC 1028968) and the UPS Foundation.

equilibrium in runoff due to varied application formulations and patterns, treatment should target both liquid- and solidphase concentrations. Previous studies have also identified treatment of dissolved stormwater pollutants as an area of importance requiring future research.72 While biochar-amended biofilters show great promise when it comes to trace organic pollutants, particularly dissolved pollutants, it should be noted that this technology is in its infancy, particularly when it comes to field installations; our study uses results from laboratory and pilot experiments.40,43 Efficacy of biochar under true field conditions remains an area of necessary investigation. While there is considerable uncertainty in removal of pollutants in SCMs, our study examines the end pointsthat is, no removal versus complete removal, and the scale at which SCMs would need to be implemented to show water quality benefit. We conclude that managers should consider using enhanced geomedia, such as biochar, when building SCMs, to maximize water quality treatment. Future work should further quantify how pesticide application and hydrologic conditions affect transport. Our study is limited by concentration data from a single storm event. Varied storm characteristics (e.g., antecedent dry conditions, rainfall intensity, and rainfall distribution) and application formulations and patterns may affect pesticide transport throughout the watershed as well as treatment efficacy within SCMs. In addition to removal of pesticides through distributed urban infrastructure, source control is a crucial part of reducing pesticide impacts on water quality. The US EPA approved new nationwide labeling requirements for fipronil in April 2017 and California-specific labeling requirements in October 2017. Nationwide label changes prohibit applications of fipronil to enter or runoff into storm drains or surface waters. In California, restrictions are more stringent; applications are prohibited on garage doors and driveways, between November 1 and February 28, and when rain is predicted within 48 h postapplication. More field data are required to understand whether these labeling changes will play a role in reducing pesticide load. Both source control and distributed treatment will prove essential in reducing pesticide impacts to the urban environment in the long term.

■

■

REFERENCES

(1) Burant, A.; Selbig, W.; Furlong, E. T.; Higgins, C. P. Trace Organic Contaminants in Urban Runoff: Associations with Urban Land-Use. Environ. Pollut. 2018, 242, 2068−2077. (2) Gan, J.; Bondarenko, S.; Oki, L.; Haver, D.; Li, J. X. Occurrence of Fipronil and Its Biologically Active Derivatives in Urban Residential Runoff. Environ. Sci. Technol. 2012, 46 (3), 1489−1495. (3) Weston, D. P.; Chen, D.; Lydy, M. J. Stormwater-Related Transport of the Insecticides Bifenthrin, Fipronil, Imidacloprid, and Chlorpyrifos into a Tidal Wetland, San Francisco Bay, California. Sci. Total Environ. 2015, 527−528, 18−25. (4) Weston, D. P.; Holmes, R. W.; Lydy, M. J. Residential Runoff as a Source of Pyrethroid Pesticides to Urban Creeks. Environ. Pollut. 2009, 157 (1), 287−294. (5) Hintzen, E. P.; Lydy, M. J.; Belden, J. B. Occurrence and Potential Toxicity of Pyrethroids and Other Insecticides in Bed Sediments of Urban Streams in Central Texas. Environ. Pollut. 2009, 157 (1), 110−116. (6) Phillips, B. M.; Anderson, B. S.; Hunt, J. W.; Siegler, K.; Voorhees, J. P.; Tjeerdema, R. S.; Mcneill, K. Pyrethroid and Organophosphate Pesticide-Associated Toxicity in Two Coastal Watersheds (California, USA). Environ. Toxicol. Chem. 2012, 31 (7), 1595−1603. (7) Weston, D. P.; Lydy, M. J. Toxicity of the Insecticide Fipronil and Its Degradates to Benthic Macroinvertebrates of Urban Streams. Environ. Sci. Technol. 2014, 48 (2), 1290−1297. (8) California Department of Pesticide Regulation. California Pesticide Information Portal (CalPIP), https://calpip.cdpr.ca.gov/ main.cfm (Feb. 28, 2019). (9) Simon-Delso, N.; Amaral-Rogers, V.; Belzunces, L. P.; Bonmatin, J. M.; Chagnon, M.; Downs, C.; Furlan, L.; Gibbons, D. W.; Giorio, C.; Girolami, V.; Goulson, D.; Kreutzweiser, D. P.; Krupke, C. H.; Liess, M.; Long, E.; Mcfield, M.; Mineau, P.; Mitchell, E. A.; Morrissey, C. A.; Noome, D. A.; Pisa, L.; Settele, J.; Stark, J. D.; Tapparo, A.; Van Dyck, H.; Van Praagh, J.; Van Der Sluijs, J. P.; Whitehorn, P. R.; Wiemers, M. Systemic Insecticides (Neonicotinoids and Fipronil): Trends, Uses, Mode of Action and Metabolites. Environ. Sci. Pollut. Res. 2015, 22 (1), 5−34. (10) Stone, W. W.; Gilliom, R. J.; Ryberg, K. R. Pesticides in U.S. Streams and Rivers: Occurrence and Trends during 1992−2011. Environ. Sci. Technol. 2014, 48 (19), 11025−11030. (11) Budd, R.; Ensminger, M.; Wang, D.; Goh, K. S. Monitoring Fipronil and Degradates in California Surface Waters, 2008−2013. J. Environ. Qual. 2015, 44 (4), 1233−1240. (12) Gunasekara, A. S.; Truong, T.; Goh, K. S.; Spurlock, F.; Tjeerdema, R. S. Environmental Fate and Toxicology of Fipronil. J. Pestic. Sci. 2007, 32 (3), 189−199. (13) Lin, K.; Haver, D.; Oki, L.; Gan, J. Persistence and Sorption of Fipronil Degradates in Urban Stream Sediments. Environ. Toxicol. Chem. 2009, 28 (7), 1462−1468.

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.8b05833.

■

Article

Detailed extraction and analytical methods, model parameters, equilibrium calculations, TOC and TSS results, and additional figures (PDF)

AUTHOR INFORMATION

Corresponding Author

*Tel: (650) 721-2615. E-mail: [email protected]. ORCID

Jordyn M. Wolfand: 0000-0003-2650-4373 Colin D. Bell: 0000-0001-8712-9859 Richard G. Luthy: 0000-0003-0274-0240 Notes

The authors declare no competing financial interest. I

DOI: 10.1021/acs.est.8b05833 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology (14) Fenner, K.; Canonica, S.; Wackett, L. P.; Elsner, M. Evaluating Pesticide Degradation in Emerging Opportunities. Science (Washington, DC, U. S.) 2013, 341 (6147), 752−758. (15) Laskowski, D. A. Physical and Chemical Properties of Pyrethroids. Rev. Environ. Contam. Toxicol. 2002, 174, 49−170. (16) Tang, W.; Wang, D.; Wang, J.; Wu, Z.; Li, L.; Huang, M.; Xu, S.; Yan, D. Pyrethroid Pesticide Residues in the Global Environment: An Overview. Chemosphere 2018, 191 (308), 990−1007. (17) Kuivila, K. M.; Hladik, M. L.; Ingersoll, C. G.; Kemble, N. E.; Moran, P. W.; Calhoun, D. L.; Nowell, L. H.; Gilliom, R. J. Occurrence and Potential Sources of Pyrethroid Insecticides in Stream Sediments from Seven U.S. Metropolitan Areas. Environ. Sci. Technol. 2012, 46 (8), 4297−4303. (18) Holmes, R. W.; Anderson, B. S.; Phillips, B. M.; Hunt, J. W.; Crane, D. B.; Mekebri, A.; Connor, V. Statewide Investigation of the Role of Pyrethroid Pesticides in Sediment Toxicity in California’s Urban Waterways. Environ. Sci. Technol. 2008, 42 (18), 7003−7009. (19) Hladik, M. L.; Kuivila, K. M. Assessing the Occurrence and Distribution of Pyrethroids in Water and Suspended Sediments. J. Agric. Food Chem. 2009, 57 (19), 9079−9085. (20) Weston, D. P.; Asbell, A. M.; Hecht, S. A.; Scholz, N. L.; Lydy, M. J. Pyrethroid Insecticides in Urban Salmon Streams of the Pacific Northwest. Environ. Pollut. 2011, 159 (10), 3051−3056. (21) Amweg, E. L.; Weston, D. P.; You, J.; Lydy, M. J. Pyrethroid Insecticides and Sediment Toxicity in Urban Creeks from California and Tennessee. Environ. Sci. Technol. 2006, 40 (5), 1700−1706. (22) Hu, Y.; Ji, L.; Zhang, Y.; Shi, R.; Han, W.; Tse, L. A.; Pan, R.; Wang, Y.; Ding, G.; Xu, J.; Zhang, Q.; Gao, Y.; Tian, Y. Organophosphate and Pyrethroid Pesticide Exposures Measured before Conception and Associations with Time to Pregnancy in Chinese Couples Enrolled in the Shanghai Birth Cohort. Environ. Health Perspect. 2018, 126 (7), 077001. (23) Ding, Y.; Harwood, A. D.; Foslund, H. M.; Lydy, M. J. Distribution and Toxicity of Sediment-Associated Pesticides in Urban and Agricultural Waterways from Illinois, USA. Environ. Toxicol. Chem. 2010, 29 (1), 149−157. (24) Gilbreath, A. N.; McKee, L. J. Concentrations and Loads of PCBs, Dioxins, PAHs, PBDEs, OC Pesticides and Pyrethroids during Storm and Low Flow Conditions in a Small Urban Semi-Arid Watershed. Sci. Total Environ. 2015, 526, 251−261. (25) Weston, D. P.; Lydy, M. J. Stormwater Input of Pyrethroid Insecticides to an Urban River. Environ. Toxicol. Chem. 2012, 31 (7), 1579−1586. (26) TMDL Projects - Pyrethroid Pesticides TMDL and Basin Plan Amendment | Central Valley Regional Water Quality Control Board, https://www.waterboards.ca.gov/rwqcb5/water_issues/tmdl/ central_valley_projects/central_valley_pesticides/pyrethroid_tmdl_ bpa/ (accessed Aug 21, 2018). (27) Salinas River Watershed Sediment Toxicity and Pyrethroid Pesticides in Sediment TMDL | Central Coast Regional Water Quality Control Board, https://www.waterboards.ca.gov/ centralcoast/water_issues/programs/tmdl/docs/salinas/sed_tox/ index.html (accessed Aug 21, 2018). (28) Santa Maria River Watershed TMDL - Toxicity and Pesticides | Central Coast Regional Water Quality Control Board, https://www. waterboards.ca.gov/centralcoast/water_issues/programs/tmdl/docs/ santa_maria/pesticide/index.html (accessed Aug 21, 2018). (29) Bay, S. M.; Greenstein, D. J.; Maruya, K. a; Lao, W. Toxicity Identification Evaluation of Sediment (Sediment TIE) in Ballona Creek Estuary Final Report; SCCWRP, 2010. (30) Tzoulas, K.; Korpela, K.; Venn, S.; Yli-Pelkonen, V.; Kaźmierczak, A.; Niemela, J.; James, P. Promoting Ecosystem and Human Health in Urban Areas Using Green Infrastructure: A Literature Review. Landsc. Urban Plan 2007, 81 (3), 167−178. (31) CH2M; Paradigm Environmental; Black & Veatch. Enhanced Watershed Management Program (EWMP) for the Upper Los Angeles River Watershed; Ballona Creek Watershed Management Group, 2015. (32) Geosyntec Consultants. City of San Diego Storm Water Fee Study; 2016.

(33) Erickson, A. J.; Gulliver, J. S.; Weiss, P. T. Capturing Phosphates with Iron Enhanced Sand Filtration. Water Res. 2012, 46 (9), 3032−3042. (34) Halaburka, B. J.; Lefevre, G. H.; Luthy, R. G. Evaluation of Mechanistic Models for Nitrate Removal in Woodchip Bioreactors. Environ. Sci. Technol. 2017, 51 (9), 5156−5164. (35) Taylor, A.; Flatt, A.; Beutel, M.; Wolff, M.; Brownson, K.; Stamets, P. Removal of Escherichia Coli from Synthetic Stormwater Using Mycofiltration. Ecol. Eng. 2015, 78, 79−86. (36) Wolfand, J. M.; LeFevre, G. H.; Luthy, R. G. Metabolization and Degradation Kinetics of the Urban-Use Pesticide Fipronil by White Rot Fungus Trametes Versicolor. Environ. Sci. Process. Impacts 2016, 18 (10), 1256−1265. (37) Mohanty, S. K.; Valenca, R.; Berger, A. W.; Yu, I. K. M. M.; Xiong, X.; Saunders, T. M.; Tsang, D. C. W. W. Plenty of Room for Carbon on the Ground: Potential Applications of Biochar for Stormwater Treatment. Sci. Total Environ. 2018, 625, 1644−1658. (38) Ulrich, B. A.; Im, E. A.; Werner, D.; Higgins, C. P. Biochar and Activated Carbon for Enhanced Trace Organic Contaminant Retention in Stormwater Infiltration Systems. Environ. Sci. Technol. 2015, 49 (10), 6222−6230. (39) Ulrich, B. A.; Vignola, M.; Edgehouse, K.; Werner, D.; Higgins, C. P. Organic Carbon Amendments for Enhanced Biological Attenuation of Trace Organic Contaminants in Biochar-Amended Stormwater Biofilters. Environ. Sci. Technol. 2017, 51 (16), 9184− 9193. (40) Ulrich, B. A.; Loehnert, M.; Higgins, C. P. Improved Contaminant Removal in Vegetated Stormwater Biofilters Amended with Biochar. Environ. Sci. Water Res. Technol. 2017, 3 (4), 726−734. (41) Mohanty, S. K.; Boehm, A. B. Escherichia Coli Removal in Biochar-Augmented Biofilter: Effect of Infiltration Rate, Initial Bacterial Concentration, Biochar Particle Size, and Presence of Compost. Environ. Sci. Technol. 2014, 48, 11535−11542. (42) Bock, E. M.; Coleman, B.; Easton, Z. M. Effect of Biochar on Nitrate Removal in a Pilot-Scale Denitrifying Bioreactor. J. Environ. Qual. 2016, 45 (3), 762. (43) Ashoori, N.; Teixido, M.; Spahr, S.; LeFevre, G. H.; Sedlak, D. L.; Luthy, R. G. Evaluation of Pilot-Scale Biochar-Amended Woodchip Bioreactors to Remove Nitrate, Metals, and Trace Organic Contaminants from Urban Stormwater Runoff. Water Res. 2019, 154, 1−11. (44) Chinen, K.; Lau, S.; Nonezyan, M.; McElroy, E.; Wolfe, B. Predicting Runoff Induced Mass Loads in Urban Watersheds: Linking Land Use and Pyrethroid Contamination. Water Res. 2016, 102, 607− 618. (45) Luo, Y.; Spurlock, F.; Jiang, W.; Jorgenson, B. C.; Young, T. M.; Gan, J.; Gill, S.; Goh, K. S. Pesticide Washoff from Concrete Surfaces: Literature Review and a New Modeling Approach. Water Res. 2013, 47 (9), 3163−3172. (46) Luo, Y.; Jorgenson, B. C.; Thuyet, D. Q.; Young, T. M.; Spurlock, F.; Goh, K. S. Insecticide Washoff from Concrete Surfaces: Characterization and Prediction. Environ. Sci. Technol. 2014, 48 (1), 234−243. (47) Stone, W. W.; Crawford, C. G.; Gilliom, R. J. Watershed Regressions for Pesticides (WARP) Models for Predicting Stream Concentrations of Multiple Pesticides. J. Environ. Qual. 2013, 42 (6), 1838. (48) Jorgenson, B.; Fleishman, E.; Macneale, K. H.; Schlenk, D.; Scholz, N. L.; Spromberg, J. A.; Werner, I.; Weston, D. P.; Xiao, Q.; Young, T. M.; Zhang, M. Predicted Transport of Pyrethroid Insecticides from an Urban Landscape to Surface Water. Environ. Toxicol. Chem. 2013, 32 (11), 2469−2477. (49) San Diego County Water Authority. Annual Rainfall - Lindbergh Field, https://www.sdcwa.org/annual-rainfall-lindbergh-field. (50) Anderson, B.; Phillips, B.; Markiewicz, D.; Stillway, M. Toxicity in California Waters: San Diego Region; Water Boards, 2012. (51) State Water Resources Control Board. California Environmental Data Exchange Network (CEDEN), http://www.ceden.org/ index.shtml. J

DOI: 10.1021/acs.est.8b05833 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology (52) Farinas, G. R. Analysis of Pyrethroid Pesticides in Sediments From San Diego County Streams Using Liquid Chromatography-Triple Quadrupole Mass Spectrometry; San Diego State University, 2012. (53) State Water Resources Control Board. Impaired Water Bodies: 2014 and 2016 Integrated Report, https://www.waterboards.ca.gov/ water_issues/programs/tmdl/integrated2014_2016.shtml (accessed Oct 10, 2018). (54) San Diego River Watershed Comprehensive Load Reduction Plan Phase II; City of San Diego, 2013. (55) Steele, J.; Griffth, J.; Noble, R.; Schiff, K. Tracking Human Fecal Sources in an Urban Watershed During Wet Weather - SCCWRP Technical Report 1002, 2017. (56) Hladik, M. L.; Smalling, K. L.; Kuivila, K. M. Methods of Analysis - Determination of Pyrethroid Insecticides in Water and Sediment Using Gas Chromatography/Mass Spectrometry, 2009; DOI: 10.3133/ tm5C2 (57) US Environmental Protection Agency. National Functional Guidelines for Superfund Organic Methods Data Review, 2016. (58) US Environmental Protection Agency. System for Urban Stormwater Treatment and Analysis IntegratioN (SUSTAIN), 2014. (59) SanGIS Data Warehouse. San Diego Geographic Information Source - JPA, http://www.sangis.org/download/index.html (Feb. 28, 2019). (60) County of San Diego BMP Design Manual; County of San Diego, 2016. (61) National Engineering Handbook Part 630 Hydrology: Chapter 7 Hydrologic Soil Groups; USDA, 2009. (62) Heber Green, W.; Ampt, G. A. Studies on Soil Physics. J. Agric. Sci. 1911, 4 (01), 1−24. (63) Anderson, B. S.; Phillips, B. M.; Voorhees, J. P.; Siegler, K.; Tjeerdema, R. Bioswales Reduce Contaminants Associated with Toxicity in Urban Storm Water. Environ. Toxicol. Chem. 2016, 35 (12), 3124−3134. (64) US Environmental Protection Agency. Aquatic Life Benchmarks and Ecological Risk Assessments for Registered Pesticides, https://www.epa.gov/pesticide-science-and-assessing-pesticide-risks/ aquatic-life-benchmarks-and-ecological-risk (Feb. 28, 2019). (65) Ensminger, M. P.; Budd, R.; Kelley, K. C.; Goh, K. S. Pesticide Occurrence and Aquatic Benchmark Exceedances in Urban Surface Waters and Sediments in Three Urban Areas of California, USA, 2008−2011. Environ. Monit. Assess. 2013, 185 (5), 3697−3710. (66) Weston, D. P.; Lydy, M. J. Urban and Agricultural Sources of Pyrethroid Insecticides to the Sacramento-San Joaquin Delta of California. Environ. Sci. Technol. 2010, 44 (5), 1833−1840. (67) Liu, W.; Gan, J. J.; Lee, S.; Kabashima, J. N. Phase Distribution of Synthetic Pyrethroids in Runoff and Stream Water. Environ. Toxicol. Chem. 2004, 23 (1), 7−11. (68) Trask, J. R.; Harbourt, C. M.; Miller, P.; Cox, M.; Jones, R.; Hendley, P.; Lam, C. Washoff of Cypermethrin Residues from Slabs of External Building Material Surfaces Using Simulated Rainfall. Environ. Toxicol. Chem. 2014, 33 (2), 302−307. (69) Jones, R. L.; Trask, J. R.; Hendley, P.; Cox, M. J.; Chepega, J. C.; Harbourt, C. M.; Davidson, P. C. Effects of Formulation on Transport of Pyrethroids in Residential Settings. Environ. Toxicol. Chem. 2016, 35 (2), 340−347. (70) Helsel, D. R. Statistics for Censored Environmental Data Using Minitab and R; John Wiley & Sons, Inc.: Hoboken, NJ, 2012. (71) Wolfand, J. M.; Bell, C. D.; Boehm, A. B.; Hogue, T. S.; Luthy, R. G. Multiple Pathways to Bacterial Load Reduction by Stormwater Best Management Practices: Trade-Offs in Performance, Volume, and Treated Area. Environ. Sci. Technol. 2018, 52, 6370−6379. (72) LeFevre, G. H.; Paus, K. H.; Natarajan, P.; Gulliver, J. S.; Novak, P. J.; Hozalski, R. M. Review of Dissolved Pollutants in Urban Storm Water and Their Removal and Fate in Bioretention Cells. J. Environ. Eng. 2015, 141 (1), 04014050.

K

DOI: 10.1021/acs.est.8b05833 Environ. Sci. Technol. XXXX, XXX, XXX−XXX