Empirical Estimation of Biota Exposure Range for Calculation of Bioaccumulation Parameters Aroon R Melwani,*À Ben K Greenfield,À and Earl R Byron` ÀSan Francisco Estuary Institute, 7770 Pardee Lane, 2nd Floor, Oakland, California 94621, USA `CH2M HILL, 2485 Natomas Park Drive, Suite 600, Sacramento, California 95833, USA (Received 5 May 2008; Accepted 2 October 2008) ABSTRACT Bioaccumulation factors (BAFs) and biota–sediment accumulation factors (BSAFs) are frequently used to predict contaminant bioaccumulation in risk assessments. Development of these parameters is often hindered by uncertainty regarding the spatial scale of contaminant transfer from sediments to biota. We present a simple statistical method for optimizing bioaccumulation parameters (BAF and BSAF) in aquatic species, such as fish, whose exposure history may occur over broad spatial scales. For 6 finfish species sampled in San Francisco Bay, San Diego Bay, or the Southern California Bight, California, USA, the spatial scale of correlation was optimized using regression analysis. The ranges identified for pairing biota and sediment observations generally corresponded to the known life histories of the species and with laboratory tests comparing relationships observed for 28-d Macoma spp. This procedure may be useful for identifying appropriate species and spatial scales to predict bioaccumulation and for developing data sets of corresponding sediment and tissue residues. Keywords: Bioaccumulation factor Exposure range BSAF Sediments INTRODUCTION Legacy pollutants have severely impacted natural aquatic systems, necessitating costly risk assessments and cleanup actions. The bioavailability of organic pollutants, and thereby the potential for bioaccumulation, has been shown to vary widely among estuarine and coastal water bodies potentially at risk (Boese et al. 1995; Boese et al. 1997; Mason and Lawrence 1999; Kraaij et al. 2002; Battelle et al. 2005). Bioaccumulation, the net increase of a chemical by an organism because of uptake from all environmental sources, is frequently modeled using bioaccumulation factors (BAFs) and biota–sediment accumulation factors (BSAFs). Bioaccu- mulation factors are the ratio of biota to sediment contam- ination concentration (Eqn. 1). BAF ¼ C t C s ð1Þ Biota–sediment accumulation factors are the same ratio (Eqn. 2), corrected for lipid content of the biota and organic carbon content of the sediment (reviewed in Wong et al. 2001; Burkhard et al. 2004). BSAF ¼ C t =f L À Á C s = f OC À Á ð2Þ where C t is the tissue concentration, C s is the sediment concentration, f L is the fraction of lipid in tissue, and f OC is the fraction of organic carbon in sediment (USEPA 2000). For organic pollutants, the use of lipid and organic carbon normalization rests on the principle that pollutants are predominantly associated with these matrices, producing more reliable relationships (Clark et al. 1988). The use of BSAFs and BAFs to predict biota exposure from sediment-associated pollutants relies on several key assump- tions, which should be considered before their application. These include the assumptions that currently monitored sediments are in steady state with the organism and are the primary source of contamination to the species being modeled. For the selected fish species, contamination is assumed to be primarily due to bioaccumulation from contaminated benthic prey, such as invertebrates and smaller fish, closely associated with the sediment. The exposure to contaminants from waterborne sources other than the sedi- ments, including uptake from ambient water, respiratory surfaces (e.g., gills and external body), and prey not associated with sediments (e.g., phytoplankton, zooplankton, and pelagic forage fish), are assumed to be relatively small. These assumptions have been shown to generally apply when assessing bioaccumulation in chemicals of higher hydro- phobicity (K ow ), as indicated by log octanol–water partition- ing coefficients between 6 and 7 (Burkhard, Cook, et al. 2003). As a result, sediments as the ultimate source of organic contaminant exposure to benthic fish and invertebrates have been indicated in many recent modeling approaches (e.g., Morrison et al. 2002; Burkhard, Cook, et al. 2003). Pelagic species are less attractive for sediment risk assessment and decision-making than benthic species because there is more uncertainty regarding the indirect (i.e., food web mediated) contribution of sediments to the contaminant burden of pelagic species. Biota–sediment accumulation factors are widely applied in the scientific literature (e.g., Boese et al. 1995; Tracey and Hansen 1996; Burkhard, Cook, et al. 2003; Burkhard, Endicott, et al. 2003) and have commonly been used in sediment risk assessments (e.g., Byron et al. 2003; USEPA 2003). For PCBs, a BSAF of 4 is expected for finfish, whereas benthic invertebrates typically have values around 1 (Ankley et al. 1992; Maruya et al. 1997; Kraaij et al. 2002). However, substantial variation exists among locations, as observed in syntheses undertaken at national and global scales (Wong et al. 2001; Burkhard et al. 2005). Differences often arise as a result of multiple factors. Food web structure and resulting  To whom correspondence may be addressed: aroon@sfei.org Published on the Web 10/15/2008. Integrated Environmental Assessment and Management — Volume 5, Number 1—pp. 138–149 138 Ó 2009 SETAC Original Research