Baird (2212) (ds)

The fractionation of stable isotopes in aquatic environments has been used extensively to track the movement of elements through aquatic food webs [e.g. (Stapp et al., 1999; Zanden et al., 1999)] and to understand ecological processes (Lajtha and Michener, 1994). In particular, the changing rate of stable isotope fractionation of biological and physical processes due to changing environmental conditions has received much attention (Popp et al., 1989, 1999). Much of this research has been motivated by a realization that stable isotopes can be used to reconstruct paleo-oceanographic environments, and therefore for model assessment of coupled global ocean–atmosphere models (von Blackenburg, 1999). Many laboratory studies have been conducted to determine the rate of stable carbon isotope fractionation during phytoplankton growth (Rau et al., 1996; Laws et al., 1997; Popp et al., 1998, 1999; Burkhardt et al., 1999a). As a result of these studies, fractionation rates are understood to be a function of growth rate and ambient CO2 concentration (Rau et al., 1996; Laws et al., 1997) and cell size (Popp et al., 1998; Burkhardt et al., 1999a). Furthermore, a comparison of samples from five open-ocean stations with laboratory cultures suggests that stable isotope fractionation can be used to infer in situ growth rates (Bidigare et al., 1997). However, a recent set of experiments (Burkhardt et al., 1999b) have demonstrated that the isotope fractionation rate is dependent on the growth-rate-limiting resource. Burkhardt et al. (Burkhardt et al., 1999b) found that, under nitrate limitation, fractionation was a function of growth rate [in agreement with (Rau et al., 1996; Laws et al., 1997; Popp et al., 1998, 1999; Burkhardt et al., 1999a)], but under light limitation, fractionation was virtually independent of growth rate. This led Burkhardt et al. to conclude ‘a general relationship between p and [CO2(aq)] may not exist. These results suggest that in situ growth rates of phytoplankton cannot be estimated from a p versus [CO2(aq)] relationship’ (Burkhardt et al., 1999b). Phytoplankton growth models may offer a solution to the problem of the fractionation rate varying dependent on the growth-rate-limiting resource. Many phytoplankton growth models have been developed which determine growth rate as a function of more than one possible growthrate-limiting resource (Baird et al., 2001). In this paper, however, we show that phytoplankton growth models that are either always proportional to the maximum growth rate, or that use the Droop model relating internal