Temporal and spatial variations in the biogeochemical cycling of cobalt in two urban estuaries: Hudson River Estuary and San Francisco Bay Antonio Tovar-Sa´nchez a,b, ) , Sergio A. San˜udo-Wilhelmy a , A. Russell Flegal c a Marine Sciences Research Center, Stony Brook University, Stony Brook, NY 11794-5000, USA b IMEDEA (CSIC-UIB), Instituto Mediterra ´neo de Estudios Avanzados, Esporles 07190 (Mallorca, Islas Baleares), Spain c Environmental Toxicology, University of California, Santa Cruz, CA 95064, USA Received 19 November 2003; accepted 12 March 2004 Abstract Despite the fact that Co is an essential trace element for the growth of marine phytoplankton, there is very limited information on the cycling of this trace metal in the marine environment. We report here the distribution of dissolved ( !0.4 mm) and particulate (O0.4 mm) Co in surface waters of the Hudson River Estuary (HRE) and San Francisco Bay (SFB). Samples were collected during several cruises ( from 1990 to 1995 in SFB and from 1995 to 1997 in the HRE) along the whole salinity gradient. Dissolved Co concentrations (mean G 1 standard deviation) were nearly identical in magnitude in both estuaries despite differences in climate, hydrography, riverine-flow conditions and land-usage (HRE=0.91 G 0.61 nM; SFB=1.12 G 0.69 nM). Dissolved Co levels in each system showed non-conservative distributions when plotted as a function of salinity, with increasing concentrations downstream from the riverine end-members. Desorption from suspended particulates and sewage inputs, therefore, seems to be the major processes responsible for the non-conservative behavior of Co observed. Mass balance estimates also indicated that most of the estuarine Co is exported out of both estuaries, indicating that they and other estuarine systems are principal sources of this essential trace element to the open ocean. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: cobalt; estuaries; Hudson River Estuary; San Francisco Bay 1. Introduction The distribution and biogeochemical behavior of dis- solved metals in estuaries are functions of the complex dynamics of those environments. For example, dissolved metals are introduced into estuaries from external (e.g. rivers, industrial and municipal effluents, groundwater, rain and atmospheric deposition) and internal (e.g. re- mobilization from benthic and resuspended sediments) sources. Particle scavenging and hydraulic flushing to the ocean are the most important removal mechanisms influencing metal levels in estuaries (Turekian, 1977; Millward and Turner, 1995; Breuer et al., 1999). The actual metal distributions observed in estuaries reflect the imbalance among those inputs and sinks. Identifying the magnitude of those different processes is relevant for understanding the mechanisms responsible for some of the high levels of metals observed in some urban estu- aries (Klinkhammer and Bender, 1981; Flegal et al., 1991; San˜udo-Wilhelmy and Gill, 1999). A few previous studies have focused on the cycling of Co in the open ocean (e.g. Knauer et al., 1982; Saito and Moffett, 2002), in coastal waters (Johnson et al., 1988; San˜udo-Wilhelmy and Flegal, 1996) and in estu- aries (e.g. Moffett and Ho, 1996; Sirinawin et al., 1998; Wen et al., 1999; Martino et al., 2002). However, those studies did not address how much the cycling of this metal is influenced by anthropogenic perturbations. The ) Corresponding author. E-mail address: atsanchez@notes.cc.sunysb.edu (A.Tovar-Sa´ nchez). Estuarine, Coastal and Shelf Science 60 (2004) 717e728 www.elsevier.com/locate/ECSS 0272-7714/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.ecss.2004.03.010