897 Environmental Toxicology and Chemistry, Vol. 27, No. 4, pp. 897–908, 2008  2008 SETAC Printed in the USA 0730-7268/08 $12.00 + .00 MERCURY CONCENTRATIONS IN BLOOD AND FEATHERS OF PREBREEDING FORSTER’S TERNS IN RELATION TO SPACE USE OF SAN FRANCISCO BAY, CALIFORNIA, USA, HABITATS JOSHUA T. ACKERMAN,*† COLLIN A. EAGLES-SMITH,†‡ JOHN Y. TAKEKAWA,§ JILL D. BLUSO, and TERRENCE L. ADELSBACH‡ †U.S. Geological Survey, Western Ecological Research Center, Davis Field Station, One Shields Avenue, University of California, Davis, California 95616 ‡U.S. Fish and Wildlife Service, 2800 Cottage Way, Suite W-2605, Sacramento, California 95825 §U.S. Geological Survey, Western Ecological Research Center, San Francisco Bay Estuary Field Station, 505 Azuar Drive, Vallejo, California 94592 Humboldt State University, Wildlife Department, One Harpst Street, Arcata, California 95521, USA ( Received 28 March 2007; Accepted 17 October 2007) Abstract—We examined mercury concentrations and space use of prebreeding Forster’s terns (Sterna forsteri) in San Francisco Bay, California, USA, to assess factors influencing mercury levels in piscivorous birds. In 2005 and 2006, we collected blood and feathers from 122 Forster’s terns and radio-marked and tracked 72 terns to determine locations of dietary mercury uptake. Capture site and capture date were the most important factors explaining variation in blood mercury concentrations (geometric mean  standard error: 1.09  0.89 g/g wet wt), followed by sex and year. Accordingly, radiotelemetry data revealed that Forster’s terns generally remained near their site of capture and foraged in nearby salt ponds, managed and tidal marshes, and tidal flats. In contrast, capture site and capture date were not important factors explaining variation in feather mercury concentrations, probably because feathers were grown on their wintering grounds several months prior to our sampling. Instead, sex and year were the most important factors explaining mercury concentrations in breast feathers (9.57  8.23 g/g fresh wt), and sex was the most important factor for head feathers (6.94  7.04 g/g fresh wt). Overall, 13 and 22% of prebreeding Forster’s terns were estimated to be at high risk for deleterious effects due to mercury concentrations in blood (3.0 g/g wet wt) and feathers (20.0 g/g fresh wt), respectively. Breeding terns are likely to be even more at risk because blood mercury concentrations more than tripled during the 45-d prebreeding time period. These data illustrate the importance of space use and tissue type in interpreting mercury concentrations in birds. Keywords—Blood Mercury Telemetry Terns San Francisco Bay INTRODUCTION Piscivorous birds are good indicators of mercury contam- ination and risk to wildlife in aquatic food webs because they forage at a high trophic level [1] and methylmercury bio- magnifies through aquatic food chains [2]. Despite a large body of literature assessing waterbird exposure to mercury [2–5], few studies have simultaneously examined mercury concen- trations and space use in birds. Instead, most studies have examined mercury contamination in birds at a regional or land- scape scale [6,7], possibly because their mobility is presumed to be a key disadvantage of using birds as biomonitors [8]. However, many birds often show strong fidelity to foraging, roosting, and breeding sites [9,10], and therefore they can also be used to assess site-specific contamination at smaller spatial and temporal scales. The usefulness of birds as biomonitors to assess variation in local contamination will depend on ad- equately documenting their movement and foraging locations as well as sampling appropriate tissues that represent recent mercury accumulation. Feathers are often used to measure mercury exposure be- cause they are easily obtained and noninvasively sampled. Feather mercury represents blood mercury concentrations at * To whom correspondence may be addressed (jackerman@usgs.gov). Published on the Web 12/3/2007. the time of feather growth and often is derived from mercury stored in body tissues [6,11,12]. Feathers are the major elim- ination pathway for methylmercury, and an adult bird’s plum- age can contain from 56 to 93% of the total body burden of mercury [6,13,14]. However, using feathers to monitor ex- posure can be problematic because mercury concentrations vary significantly among and within feather tracts and also depend on the timing of molt [11]. For example, feathers grown early during molt have higher mercury concentrations than feathers grown later, as the total body burden of mercury is reduced throughout the molting sequence [6,11]. Although us- ing hatchling feathers is one way to assess recent mercury exposure, as are eggs [8,15,16], this limits sampling to only a short time period each year during the breeding season. Mercury concentrations in blood, on the other hand, rep- resent exposure at the time of sampling and are a dynamic equilibrium of recent dietary mercury uptake and internal tis- sue redistribution [17]. Mercury concentrations in blood are more highly correlated with mercury concentrations in internal tissues than are feathers (Ackerman et al., unpublished data). Almost all mercury in blood is methylmercury bound to red blood cells [5,18] and has a half-life of one to three months [19,20]. Therefore, blood may be the best tissue to estimate short-term mercury exposure in wild birds [5], and it can be sampled throughout the year in a nonlethal manner. We examined blood and feather mercury concentrations in