Sim-Lin Lau

Characteristics of highway stormwater runoff in Los Angeles: metals and polycyclic aromatic hydrocarbons.

Stormwater runoff from three highway sites in Los Angeles, California, was monitored, during the 2000 to 2003 wet seasons. Correlations among heavy metals, polycyclic aromatic hydrocarbons (PAHs), and storm characteristics were performed using datasets collected for 62 storm events. Statistical correlation analyses of the event mean concentrations (EMCs) and mass first-flush ratios (MFFs) with storm characteristics were conducted to determine if the first flush is related to site or storm characteristics. This study agreed with other highway runoff characterization studies, in that strong correlations were observed among the heavy metals and between heavy metals and total PAHs, and total suspended solids were well correlated with most heavy metals. Only antecedent dry days among storm characteristics were reasonably well-correlated with the EMCs of heavy metals and total PAHs, and dissolved and total metals exhibited similar MFFs, with approximately 30 to 35% of the mass being discharged in the first 20% of the runoff volume.

First Flush Stormwater Runoff from Highways

Stormwater is now receiving attention from regulatory agencies and has become an important component in watershed planning. In many cases, pollutant mass emissions from stormwater exceed those from wastewater treatment plants. Land use has been identified as an important parameter in predicting stormwater quality. Land uses associated with vehicular activity, such as parking lots, are thought to be high contributors of stormwater pollutants. Other factors, such as greater pollutant concentrations or mass emissions at the onset of rainfall, usually called a “first flush,” or higher emissions from the first storm of the season, usually called a “seasonal first flush,” have been identified. In order to determine the magnitude of the first flush from freeway runoff, three sites in the west Los Angeles area were sampled for 14 storms during the 1999-2000 rainy season. Samples were collected very early in the storm in order to compare water quality from the first runoff to water quality from the middle of the storm. A large range of water quality parameters and metals were analyzed. The data show large first flushes in concentration profiles and moderate first flushes in mass emission rates.

First Flush of Organics in Highway Runoff

Stormwaters from three highway sites were monitored over two wet seasons for organics, Chemical oxygen demand, oil and grease and polyaromatic hydrocarbons (PAHs) were measured. The sites exhibited a first flush in most cases for most parameters. The mass first flush ratio (the ratio of the normalized transported mass of pollutant to the normalized runoff volume) generally was above 1.8 for the first 25% of the runoff volume, and in some cases as high as 2.8. Dissolved PAHs were generally at or below detection limits. Mass emission rates for particulate PAHs are reported. The results suggest that best management practices (BMPs) that address the initial runoff can have greater effectiveness than other types of BMPs.

Identification of subwatershed sources for chlorinated pesticides and polychlorinated biphenyls in the Ballona Creek watershed.

Santa Monica Bay forms part of the western border of the greater Los Angeles region. The Ballona Creek watershed is highly urbanized and past studies indicate that Ballona Creek is the largest source for most pollutants to Santa Monica Bay. This study evaluates the contribution of subwatersheds to PCB and chlorinated pesticide loading during wet weather flow. Fifteen storm drains from these subwatersheds were sampled during three storms during the 2005-2006 winter rainy season. A series of grab samples were taken over the duration of the storms. The suspended solids were analyzed for polychlorinated biphenyls (PCBs) and chlorinated pesticides. A geographic information system (GIS) was used to calculate the runoff volume from each subwatershed to estimate pollution mass loading. There was no statistical difference among subswatersheds; however, a disproportionate mass of PCB loading came from site 5, which had no obvious sources. No specific subwatersheds were identified as key sources for chlorinated pesticides. These results may serve as a model for other locations with concerns for historic PCB and chlorinated pesticides loadings.

Predicting runoff induced mass loads in urban watersheds: Linking land use and pyrethroid contamination.

Pyrethroid pesticide mass loadings in the Ballona Creek Watershed were calculated using the volume-concentration method with a Geographic Information Systems (GIS) to explore potential relationships between urban land use, impervious surfaces, and pyrethroid runoff flowing into an urban stream. A calibration of the GIS volume-concentration model was performed using 2013 and 2014 wet-weather sampling data. Permethrin and lambda-cyhalothrin were detected as the highest concentrations; deltamethrin, lambda-cyhalothrin, permethrin and cyfluthrin were the most frequently detected synthetic pyrethroids. Eight neighborhoods within the watershed were highlighted as target areas based on a Weighted Overlay Analysis (WOA) in GIS. Water phase concentration of synthetic pyrethroids (SPs) were calculated from the reported usage. The need for stricter BMP and consumer product controls was identified as a possible way of reducing the detections of pyrethroids in Ballona Creek. This model has significant implications for determining mass loadings due to land use influence, and offers a flexible method to extrapolate data for a limited amount of samplings for a larger watershed, particularly for chemicals that are not subject to environmental monitoring. Offered as a simple approach to watershed management, the GIS-volume concentration model has the potential to be applied to other target pesticides and is useful for simulating different watershed scenarios. Further research is needed to compare results against other similar urban watersheds situated in mediterranean climates.

Implication of Oil and Grease Measurement in Stormwater Management Systems

Oil and grease is not easily measured by automated samplers, which makes the collection of a flow-weighted composite difficult. A grab sample is often substituted. This paper examines more than 40 storm events to determine when a grab sample most closely approximates the value of a flow-weighted composite sample. The best time for sample collection is related to storm duration, and rainfall prediction may be useful to select the best sample time. The standard deviation between a randomly timed grab sample and composite sample was reduced by approximately 5 mg/L using the best time procedure. The reduction in standard deviation was less for other grab sample strategies, and sampling after 3 hours was nearly as effective as the best time strategy.

A field studies and modeling approach to develop organochlorine pesticide and PCB total maximum daily load calculations: case study for Echo Park Lake, Los Angeles, CA.

Echo Park Lake is a small lake in Los Angeles, CA listed on the USA Clean Water Act Section 303(d) list of impaired water bodies for elevated levels of organochlorine pesticides (OCPs) and polychlorinated biphenyls (PCBs) in fish tissue. A lake water and sediment sampling program was completed to support the development of total maximum daily loads (TMDL) to address the lake impairment. The field data indicated quantifiable levels of OCPs and PCBs in the sediments, but lake water data were all below detection levels. The field sediment data obtained may explain the contaminant levels in fish tissue using appropriate sediment-water partitioning coefficients and bioaccumulation factors. A partition-equilibrium fugacity model of the whole lake system was used to interpret the field data and indicated that half of the total mass of the pollutants in the system are in the sediments and the other half is in soil; therefore, soil erosion could be a significant pollutant transport mode into the lake. Modeling also indicated that developing and quantifying the TMDL depends significantly on the analytical detection level for the pollutants in field samples and on the choice of octanol-water partitioning coefficient and bioaccumulation factors for the model.

Discussion on “Alternatives to methanol—water elution of solid-phase extraction columns for the fractionation of high log Kow organic compounds in aqueous environmental samples” by Durhan et al., J. Chromatogr., 629 (1993) 67–74

With reference to the recently published article by Durhan et al. [l], similar problems in eluting non-polar polyaromatic hydrocarbons (PAHs) from C,, solid-phase extraction (SPE) columns were also observed in our laboratory. Our results confirm and expand Durhan et al.‘s with additional PAHs and different elution solvent mixtures. The C,, SPE method developed by the US Environmental Protection Agency (EPA) [2] was used to fractionate the non-polar PAHs from storm drain waters for short-term chronic toxicity tests. Our recovery study of the C,, SPE method used standard water solutions which contained eight commonly detected PAHs in storm drain waters (naphthalene, 2-methylnaphthalene, acenaphthene, fluorene, anthracene , pyrene , chrysene and benzo[a]pyrene). Low recoveries of PAHs from the C,, column