A statistical model and national data set for partioning fish-tissue mercury concentration variation between spatiotemporal and sample characteristic effects

Abstract

Many Federal, Tribal, State, and local agencies monitor mercury in fish-tissue samples to identify sites with elevated fish-tissue mercury (fish-mercury) concentrations, track changes in fish-mercury concentrations over time, and produce fish-consumption advisories. Interpretation of such monitoring data commonly is impeded by difficulties in separating the effects of sample characteristics (species, tissues sampled, and sizes of fish) from the effects of spatial and temporal trends on fish-mercury concentrations. Without such a separation, variation in fish-mercury concentrations due to differences in the characteristics of samples collected over time or across space can be misattributed to temporal or spatial trends; and/or actual trends in fish-mercury concentration can be misattributed to differences in sample characteristics. This report describes a statistical model and national data set (31,813 samples) for calibrating the aforementioned statistical model that can separate spatiotemporal and sample characteristic effects in fish-mercury concentration data. This model could be useful for evaluating spatial and temporal trends in fishmercury concentrations and developing fish-consumption advisories. The observed fish-mercury concentration data and model predictions can be accessed, displayed geospatially, and downloaded via the World Wide Web (http://emmma.usgs. gov). This report and the associated web site may assist in the interpretation of large amounts of data from widespread fishmercury monitoring efforts.

Modelling mercury concentrations in prey fish: Derivation of a national-scale common indicator of dietary mercury exposure for piscivorous fish and wildlife

“…The SAS procedure LIFEREG (version 9.1; SAS, 2009) was used to fit the NDMMF regression model (detailed in Wente, 2004) to the input dataset. Input observations were weighted based on the number of fish representing each observation (Christensen et al, 2006).…”

“…The United States Geological Survey (USGS) developed the National Descriptive Model for Mercury in Fish (NDMMF, http:// emmma.usgs.gov;Wente, 2004) to partition variation in MeHg concentration due to size, species and sample type across space and time. The NDMMF has recently been used to assess the relative importance of landscape factors contributing to variation of Hg concentrations in largemouth bass (Micropterus salmoides) across Texas (Drenner et al, 2011) and offers an approach to derive a common indicator of MeHg exposure.…”

Mercury in fish, bed sediment, and water from streams across the United States, 1998-2005

“…epa.gov/waterscience/fish/study/); the National Contaminant Biomonitoring Program (NCBP) of the U.S. Fish and Wildlife Service, which later became the Biomonitoring of Environmental Status and Trends (BEST) program of the U.S. Geological Survey (USGS) (Schmitt and others, 1999(Schmitt and others, , 2002(Schmitt and others, and 2004Hinck and others, 2004aHinck and others, , 2004bHinck and others, , 2006Hinck and others, , 2007; fish-Hg data compiled from 24 research and monitoring programs in northeastern North America (Kamman and others, 2005); and a large compilation of many State, Federal, and Tribal fish-Hg datasets (Wente, 2004; see also http://emmma. usgs.gov/datasets.aspx).…”

“…These issues include (1) use of multiple analytical laboratories and analytical methods; (2) inconsistent or unknown data quality; (3) large variations in sample characteristics, including fish species, size, and tissue sampled; (4) incomplete site information (for example, locations of some sites are not adequately described, and some georeferenced sites may not be coded as to site type, such as lake, stream, or reservoir); and (5) incomplete sample information (for example, species, length, or tissue sampled are not known). Several of these issues have been described in greater detail by Wente (2004), who has developed a promising statistical modeling approach to account for variation in fish-Hg levels by species, size, and tissue sampled. It is not known, however, whether the approach performs equally well in streams as it does in lakes, or whether it performs consistently among various regions of the Nation.…”

Relating Fish Hg to Variations in Sediment Hg, Climate and Atmospheric Deposition

“…For example, physical and chemical properties of water (e.g., temperature, aeration, pH, color, organic matter, and suspended mineralcontent) all influence the fate of Hg with regard to biological uptake, re-precipitation and settling, methylation, demethylation, and volatilization [28] [29] [30]. From a simplifying perspective, total Hg concentrations in water, sediments andaquatic organism are co-variants [13] This article focuses on analyzing the extent to which standardized data for total Hg concentrations (THg) in fish-as compiled within the Fish Mercury Datalayer (FIMDAC [1])-co-vary with lake Sediment THg, atmospheric Hg deposition (atmHg dep ), mean annual precipitation (ppt), and mean annual air temperatures for January (winter, T Jan ) and July (summer, T July ). This analysis was enabled by cross-referencing the Fish THg data to the modelled and mapped Sediment THg, atmHg dep , ppt, T Jan and T July variations across Canada, with special reference to potential climate-induced changes up to 2070.…”