S. Ian Hartwell

Cluster analysis of contaminated sediment data: Nodal analysis

The objective of the present study was to explore the use of multivariate statistical methods as a means to discern relationships between contaminants and biological and/or toxicological effects in a representative data set from the National Status and Trends (NS&T) Program. Data from the National Oceanic and Atmospheric Administration, NS&T Programs Bioeffects Survey of Delaware Bay, USA, were examined using various univariate and multivariate statistical techniques, including cluster analysis. Each approach identified consistent patterns and relationships between the three types of triad data. The analyses also identified factors that bias the interpretation of the data, primarily the presence of rare and unique species and the dependence of species distributions on physical parameters. Sites and species were clustered with the unweighted pair-group method using arithmetic averages clustering with the Jaccard coefficient that clustered species and sites into mutually consistent groupings. Pearson product moment correlation coefficients, normalized for salinity, also were clustered. The most informative analysis, termed nodal analysis, was the intersection of species cluster analysis with site cluster analysis. This technique produced a visual representation of species association patterns among site clusters. Site characteristics, such as salinity and grain size, not contaminant concentrations, appeared to be the primary factors determining species distributions. This suggests the sediment-quality triad needs to use physical parameters as a distinct leg from chemical concentrations to improve sediment-quality assessments in large bodies of water. Because the Delaware Bay system has confounded gradients of contaminants and physical parameters, analyses were repeated with data from northern Chesapeake Bay, USA, with similar results.

An assessment of chemical contaminants in sediments from the St. Thomas east End Reserves, St. Thomas, USVI.

The St. Thomas East End Reserves or STEER is located on the southeastern end of the island of St. Thomas, USVI. The STEER contains extensive mangroves and seagrass beds, along with coral reefs, lagoons, and cays. Within the watershed, however, are a large active landfill, numerous marinas, resorts, various commercial activities, an EPA Superfund Site, and residential areas, all of which have the potential to contribute pollutants to the STEER. As part of a project to develop an integrated assessment for the STEER, 185 chemical contaminants were analyzed in sediments from 24 sites. Higher levels of chemical contaminants were found in Mangrove Lagoon and Benner Bay in the western portion of the study area. The concentrations of polychlorinated biphenyls (PCBs), dichlorodiphenyltrichloroethane (DDT), zinc, copper, lead, and mercury were above a NOAA Effects Range-Low (ERL) sediment quality guideline at one or more sites, indicating impacts may be present in more sensitive species or life stages. Copper at one site in Benner Bay was above a NOAA Effects Range-Median (ERM) guideline indicating effects on benthic organisms were likely. The antifoulant boat hull ingredient tributyltin (TBT) was found at the third highest concentration in the history of NOAA’s National Status and Trends (NS&T) Program, which monitors the nation’s coastal and estuarine waters for chemical contaminants and bioeffects. The results from this project will provide resource managers with key information needed to make effective decisions affecting coral reef ecosystem health and gauge the efficacy of restoration activities.

Empirical assessment of incorporating sediment quality triad data into a single index to distinguish dominant stressors between sites

Benthic infaunal community structure, sediment contamination, and sediment toxicity data (Sediment Quality Triad) were condensed into a single index based on the area of tri-axial plots, which were examined in relation to various habitat parameters. The purpose was to assess its utility for evaluating the relative impact of contaminants versus other stressors on benthic communities. The regression relationship between the areal index and the Effects Range–Median quotient (ERMq) was used to separate contaminant-impacted sites from sites impacted by hypoxia in Chesapeake Bay. Regression using the areal index and bottom oxygen confirm the utility of the approach. Data from Delaware, Galveston, and Biscayne Bays were also examined to determine if the approach may be effective in other estuaries.

Chemical contamination in southwest Puerto Rico: A survey of contaminants in the coral Porites astreoides

Abstract. Coral (Porites astreoides) from eight sites in southwest Puerto Rico were analyzed for approximately 150 chemical contaminants, to provide a preliminary characterization of environmental contamination in the corals, and assess the relationships between chemical contamination in corals and adjacent sediments. Overall, the concentration of PAHs (polycyclic aromatic hydrocarbons) and PCBs (polychlorinated biphenyls) detected in the limited number of coral samples collected were comparable to concentrations found in sediments. However, the concentration of a chemical contaminant (e.g., PAHs) in the corals at a site was often different from what was found in adjacent sediments. The level of PCBs and DDT (dichlorodiphenyltrichloroethane) in the corals appeared higher just outside of Guanica Bay, and there was some evidence of a downstream concentration gradient for these two contaminant classes. The trace elements copper and zinc were frequently detected in Porites astreoides, and the concentrations were usually comparable to those found in adjacent sediments. Chromium was an exception in that it was not detected in any of the coral samples analyzed, although it was detected in all of the sediment samples.

An analysis of chemical contaminants in sediments and fish from Cocos Lagoon, Guam

.............................................................................................................................................