Hydrocarbon Monitoring in Response to Personal Watercraft Regulation at Glen Canyon National Recreation Area

Glen Canyon National Recreation Area (GLCA) is located in northcentral Arizona and southcentral Utah within the Colorado Plateau physiographic province (fig. 1 ). The National Recreation Area was established in 1972 to, among other things, provide for public enjoyment through water-based recreational opportunities, and to protect resources on Lake Powell (National Park Service, undated). Lake Powell was created after the completion of construction of the Glen Canyon Dam on the Colorado River in 1963, and the lake filled to full-conservation-pool level (3,700 feet above mean sea level) in 1980. Glen Canyon Dam is the only major dam within the upper Colorado Basin located on the main stem of the Colorado River, and the dam controls almost all the flow leaving the upper Colorado River Basin. At full capacity, Lake Powell stores 27 million acre-feet of water that is used for conservation storage, electrical production, and recreation. Recreational boating on Lake Powell is by far the most popular activity in GLCA, and there are five marinas located within the Park.

Semi-volatile organic compounds (SVOCs, which include PAHs) are common contaminants in freshwater systems (Eisler, 1987; Johnson and others, 1985). Sources such as uncombusted fuel emissions, oil spills, industrial effluent, urban runoff, and atmospheric deposition are the primary origin of these compounds in ecosystems. High concentrations of PAHs are common around marinas and other areas with extensive motorboat activity because of motor exhaust and fuel/oil spills (Smith and others, 1985; Marcus and others, 1988). Owing to their high degree of aromaticity, PAH compounds are extremely stable and thus highly persistent in the environment. The fate of these compounds is of significant environmental concern because of their acute carcinogenic toxicity (Anderson and others, 1974) and their substantial persistence once they enter the ecosystem. However, many PAHs are susceptible to degradation (photolysis) upon exposure to ultraviolet radiation from sunlight. In clear waters, PAHs can degrade in minutes to hours (Huckins and others, 2006). Photolysis of PAHs has its own risks as many of the degradation products can have greater levels of toxicity than the parent PAHs. PAH toxicity can be increased by the duration of exposure to ultraviolet solar radiation; the depth and the intensity of ultraviolet solar radiation penetrating the water column also are factors effecting PAH toxicity.

Background

In an effort to reduce PAH concentrations at GLCA, the National Park Service (NPS) adopted special regulations in 2003 to manage the use of personal watercraft on Lake Powell (as described in the Record of Decision signed on June 27, 2003, for the final Environmental Impact Statement (EIS) concerning the use of personal watercraft on waters of GLCA). The special regulations (alternative B described in the EIS) defined personal watercraft emission standards, but there was a 10 year delay of implementing the standards to allow personal watercraft owners time to transition. Since December 31, 2012, the new regulations have been included in the Code of Federal Regulations Title 36, 7.70 (e)(3), which states that no one may operate a personal watercraft that does not meet 2006 emission standards set by the U.S. Environmental Protection Agency for the manufacturing of two-stroke engines. Also included as part of the Record of Decision, GLCA is required to sample and monitor for PAH concentrations in Lake Powell (http://www.nps.gov/glca/parknews/upload/pwc_rod-final.pdf, accessed April 22, 2015).

In 2004, at the request of the NPS, the U.S. Geological Survey (USGS) prepared a plan for monitoring PAH concentrations in Lake Powell (National Park Service, 2004). As part of the monitoring plan, the USGS Arizona Water Science Center (AzWSC) and the GLCA established 20 sentinel monitoring sites based on previous work at the AzWSC by Hart and others (2004). During 2004 to 2006, the AzWSC collected discrete water and sediment samples at these sites in cooperation with the NPS. Oil and grease products were detected in water at 9 of the 20 sites, and petroleum hydrocarbons were detected in water at all 20 of the sites (Hart and others, 2012).

In 2010–11, the AzWSC and GLCA again partnered to sample water and sediment at the Lake Powell monitoring sites (Schonauer and others, 2014). Water was sampled using semipermeable membrane devices (SPMDs) for PAH analysis. PAHs were detected in water most frequently at sites in the southern part of Lake Powell, where visitation and watercraft use is high (figs. 1 and 2). The most commonly detected compounds in water were fluoranthene, chrysene, 4,5-methylenephenanathrene, and 1-methylphenanthrene, which were detected at 94%, 83%, 78%, and 72% of the sites, respectively. In general, fewer PAHs were detected in sediment than in water (fig. 3). One site, Wahweap Marina, had considerably more PAH detections in sediment relative to the other 19 sites. The most commonly detected compounds in sediment were 2,6-dimethylnaphthaline and 1,6-dimethylnaphthaline, which were detected at 65% and 55% of the sites, respectively.

These studies have provided baseline information that the NPS is using to begin making managerial decisions about the use of personal watercraft on Lake Powell. These studies have also identified sites at Lake Powell where additional monitoring is necessary in order to determine if PAH concentrations in water are increasing or decreasing over time. Additional PAH sampling on Lake Powell is critical to GLCA in order to both determine the effectiveness of the newly enacted management policies, and to fully understand the relationship between watercraft and PAH concentrations in water.

Objectives

The purpose of this proposed study is to describe the current (2016–17) PAH conditions in Lake Powell in terms of PAH detections and concentrations. The current PAH detections and concentrations will be compared to previously determined 2010–11 PAH detections and concentrations in order to help NPS understand what effect current management actions of personal watercraft are having on PAH detections and concentrations.

Approach

PAH compounds have previously been sampled by the USGS in Lake Powell using passive SPMD samplers (Schonauer and others, 2014; fig. 4A). SPMDs consist of a neutral lipid, ultra-high-purity triolein, encased in a thin-walled layflat polyethylene membrane tube (Huckins and others, 2006). The membrane allows nonpolar compounds to pass through to the lipid where the compounds are concentrated (Alvarez, 2010). SPMDs continually sample the surrounding water over the period of deployment, which is typically from weeks to a month. The volume of water sampled during a SPMD deployment is a function of the sampling rate for a particular compound and the sampling duration. The estimation of the time-weighted average concentration of chemicals in the water is determined from the concentration of the chemicals accumulated in the SPMD, which is applied to uptake models which take into account the site-specific environmental variables (Alvarez, 2010). Samples collected using SPMDs typically have a large volume, and the integrative nature of SPMDs accumulates compounds and increases the probability that concentrations will be above method detection limits.

In general, SPMD samplers worked well for detecting PAHs in the Schonauer and others (2014) study because the time-integrated nature of the sample and the large sample volume resulted in low laboratory reporting levels. The SPMD sampling method, however, did present one important limitation in the accurate determination of PAH concentrations in Lake Powell: field and laboratory contamination was an issue for many compounds (fig. 5). Schonauer and others (2014) set their reporting level for PAHs at two times the maximum concentration detected in the blanks for each PAH, which resulted in censoring of at least some detected environmental concentrations for 20 compounds.

To address the limitations of the previous PAH sampling with SPMDs, two sampling methods will be utilized in the first year of this proposed study. The environmental and quality-control data collected for both methods will be evaluated and compared relative to one another. The method found to produce the best quality data will be utilized in the following years. In the first year, the SPMD sampler will once again be deployed to collect PAH compounds. SPMD construction and dialysis, however, will be completed by a different laboratory than in the Schonauer and others (2014) study. The continuous low-level aquatic monitoring (CLAM) sampler (C.I.Agent® Storm-Water Solutions; http://www.ciagent-stormwater.com; commercial introduction in 2007; fig. 4B) will be used in addition to the SPMD sampler in the first year of this proposed study to collect PAH compounds for analysis. The CLAM is a submersible, low-flow rate sampler, which continuously and actively draws water through solid-phase extraction (SPE) media contained within a disk. The extraction disks can contain a variety of media to collect a range of less- to more-polar organic compounds, including PAHs. Extraction events can be up to 36 hours long; event duration is currently limited by the extraction disk capacity of 100 liters for well-retained compounds, and by battery life. After CLAM deployment and retrieval, the extraction disk(s) is sent to a laboratory for analysis. The CLAM sampler provides an integrated sample of compounds present in the water that accumulate over the deployment period, thereby increasing the probability that compounds will be above laboratory method detection limits.

The CLAM has been previously used by the AzWSC to sample for organic compounds (Coes and others, 2014). In this previous study, concentrations were compared for a suite of wastewater compounds, including several PAHs, sampled using CLAMs, the passive polar organic chemical integrative samplers (POCIS), and discrete sampling methods. The CLAM sampling method was found to detect significantly more compounds than the other methods, primarily because of very low reporting levels relating to the large volume of water sampled (24 to 68 L).

In this proposed study, the SPMD samplers will be manufactured by the USGS Columbia Environmental Research Center (CERC). The CLAM samplers will be manufactured by C.I.Agent® Storm-Water Solutions. The AzWSC currently owns 5 CLAM samplers, which will be used for this project. SPMD dialysis will be completed at CERC, and both the SPMD extracts and the CLAM SPE disks will be analyzed for PAHs by CERC. Thirty-three PAHs will be analyzed in the SPMD and CLAM samples (table 1).

Seven sites with the greatest number of PAH detections in water in the Schonauer and others (2014) study will be resampled for PAHs (Dangling Rope Marina, Rainbow Bridge National Monument, Lone Rock Beach, State Line Marina, Antelope Marina, Wahweap Marina, and Warm Creek Bay; figs. 1 and 2). One additional site with a relatively lower number of PAH detections, Padre Bay near Dominiquez Butte, will also be monitored as a control site. During the summer months, when boat traffic and recreational use at Lake Powell is highest, many of the PAHs associated with the increased use of that season will most likely be located in the upper water column, as the lower-molecular weight and alkylated PAHs associated with fuels will float on the water surface. SPMD and CLAM samplers, therefore, will be deployed in the water column within the epilimnion. The location of the thermocline during deployment and upon retrieval of the samplers will be determined using a Sea-Bird Electronics, Inc., conductivity, temperature, and depth profiler.

CLAM samplers and SPMD samplers will be deployed at each of the 8 sites during the first year of the study, fiscal year 2016. One SPMD sampler will be deployed at each site for a month. Two CLAM samplers will be deployed at each site for up to 24 hours, one at the beginning of the SPMD deployment, and one at the end of the SPMD deployment (table 2). Data will be collected during the summer months of June to August when boat traffic and recreational use are highest. Field and laboratory quality-control (QC) data (described in detail below) collected during the first year using the CLAM sampling method will be compared to QC data collected using SPMD sampling method during the first year. If it is decided that the CLAM sampling method produces better quality data than the SPMD sampling method, then only CLAM samplers will be deployed for years two and three of the study. If the CLAM sampling method does not produce better quality data than the SPMD sampling method, then only SPMD samplers will be deployed for years two and three of the study.

The second to third year data collection will feature two deployments of CLAM (or SPMD) samplers (table 2). One deployment will be in the early summer of fiscal year 2017 when visitor use is highest, and the second deployment will occur in the fall of fiscal year 2018, when there is very limited visitor use.

The collection of field and laboratory QC data associated with both SPMD and CLAM sampling and analysis will be a vital component of the proposed work. Field QC samples will be collected to asses bias and variability associated with field and laboratory methods, and laboratory QC will be collected to assess bias and variability associated with laboratory methods. During the first year of data collection, 4 SPMD field blanks, 2 SPMD laboratory blanks, 4 CLAM field blanks, and 4 CLAM laboratory blanks will be collected to estimate potential bias caused by introduced contamination (table 2). Also during the first year, 2 SPMD replicates and 2 CLAM replicates will be collected to estimate potential variability associated introduced by field or laboratory procedures. Two SPMD laboratory spikes and 4 CLAM laboratory spikes will be completed to estimate bias associated with method performance, sample matric interference, or compound degradation. Finally, during the first year of the study, 4 CLAM samplers will have 2 SPE disks installed in series in order to investigate the possibility of incomplete sorption of PAHs by the front disk, resulting in breakthrough to, and possibly through, the back disk (fig 4C).

SPMD and CLAM QC data collected during the first year of the proposed study will be compared relative to one another to determine which sampling method has the least associated bias and variability; the sampling method with the least bias and variability will be used in the second year of the proposed study. During both the second and third years of the proposed study, 2 field blanks, 2 laboratory blanks, 2 field replicates, 2 laboratory spikes will be collected for the CLAM sampler (or for the SPMD sampler if appropriate; table 2). Also, during both the second and third years of the study, 2 CLAM samplers will have 2 SPE disks installed in series. The internal quality-control data and quality-assurance procedures of CERC will be compiled and analyzed prior to the start of the field work to assure that the laboratory is capable of meeting the project’s quality assurance needs.

This proposed study will report 2016–17 PAH detections and PAH concentrations at the previously sampled sentinel monitoring sites. In addition, 2016–17 PAH detections and concentrations will be compared to previously reported 2010–11 PAH detections and concentrations at the same monitoring sites (Schonauer and others, 2014). A secondary result of this proposed study will be a quality-assessment of PAH data collected by two different continuous samplers, the passive SPMD sampler and the active CLAM sampler.

Relevance and Benefits

This proposed study will report 2016–17 PAH detections and concentrations at the previously sampled sentinel monitoring sites. In addition, 2016–17 PAH detections and concentrations will be compared to previously reported 2010–11 PAH detections and concentrations at the same monitoring sites (Schonauer and others, 2014). The results from this proposed study will provide GLCA with the necessary information to determine if the current policies concerning personal watercraft on Lake Powell are effective management techniques for lowering PAH concentrations. Similar regulations were considered and implemented at 21 other NPS units across the United States, and data from this study can provide an example for the rest of the NPS units. 

The results from this study will also provide Reclamation with additional data to support their water-quality monitoring program on Lake Powell. Reclamation currently cooperates with the USGS Grand Canyon Monitoring and Research Center to document and understand changes that occur while water is in Lake Powell and how those changes may affect the quality of water released from Glen Canyon Dam. Water samples are collected lake-wide quarterly and near Glen Canyon Dam monthly. The samples are analyzed for field parameters, major ions, nutrients, trace elements, and plankton.

This study will support the NPS Centennial Call to Action theme of preserving American’s special places (http://www.nps.gov/calltoaction/PDF/C2A_2014.pdf). This study meets the Crystal Clear effort of protecting the health of watersheds by improving water quality for public enjoyment.

A secondary result of this proposed study will be a quality-assessment of PAH data collected by two different continuous samplers, the passive SPMD sampler and the active CLAM sampler. Continuous active samplers, such as the CLAM, are potentially an important viable method for scientists to measure organic compounds at very low concentrations. A previous study by Coes and others (2014) compared a suite of wastewater compound data collected by the passive POCIS sampler to the active CLAM sampler. The additional data for organic compounds collected by the concurrent deployment of the SPMD samplers and the CLAM samplers in this proposed study will add to the knowledge base required for USGS scientists to consider the use of continuous active samplers in their studies.