Relative importance of iron and molybdenum in restricting phytoplankton biomass in high phosphorus saline lakes

Eleven prairie saline (conductivity 1.8-58.8 mS cm-‘) lakes were examined over the 1994 growing season to determine what salinity-related factor or factors were responsible for controlling phytoplankton standing crops. The study lakes were characterized by high total P (0.15-24.2 mg liter-‘), total N (3.75-12.35 mg liter-‘), total Fe (552,800 p,g liter-‘), dissolved organic C (40-195 mg liter-l), pH and alkalinity, but comparatively low (usually cl00 kg liter-‘) dissolved inorganic N. Chlorophyll a (Chl a) concentrations in all but the two least saline lakes were relatively low (20 pg liter-‘), up to three orders of magnitude below those predicted by freshwater P-based models. High alkaline phosphatase activities (APA) and rapid ‘2P0, (orthophosphate) uptake indicated that the two least saline lakes were P limited; these lakes had seston deficient in P, N, and protein. APA and “2P0, uptake were below detection in the more saline lakes (conductivity >3 mS cm-‘), indicating P sufficiency; seston from these lakes was deficient in N but not protein. Nitrogen-fixing cyanophytes were important only in one of the lakes examined. Nutrient addition bioassays indicated that phytoplankton biomass was not limited exclusively by inorganic N availability, nor by a combination of MO and N. For water from all but one of the P-sufficient lakes, addition of Fe to bioassays resulted in a remarkable increase in Chl a concentrations. Addition of Fe and MO had the same effect as that of Fe alone, while the most saline lake appeared to be limited by one or more additional trace elements (but not MO). Reducing the alkalinity of the bioassay water stimulated growth in the same manner as the Fe additions, suggesting that the bioavailability of the (largely particulate) Fe already present was severely restricted by lakewater alkalinity. Some component of lake-water alkalinity (which increased with conductivity in these lakes) appears to be the key factor limiting Fe bioavailability and restricting phytoplankton standing crops in the higher salinity lakes. Phosphorus has traditionally been regarded as the nutrientlimiting phytoplankton biomass and productivity in aquatic systems (Schindler 1977). The relationship between phosphorus (P) and phytoplankton biomass and productivity has been documented in many studies of freshwater systems (Dillon and Rigler 1974; Smith 1979). In these lakes, nitrogen fixation (N, fixation) by planktonic cyanophytes is largely able to compensate for existing deficiencies in nitrogen (N), leaving P as the limiting nutrient. In saline systems the relationship between P and phytoplankton standing crop is not as robust. Phytoplankton chlorophyll a (Chl a) concentration, an indicator of primary productivity, in salt water is often far lower than predicted by freshwater phosphorus models (Campbell and Prepas 1986). Studies of marine (Howarth 1988), coastal pond (Caraco et al. 1987), and prairie saline lake systems (Bierhuizen and Prepas 1985; Campbell and Prepas 1986) indicate that N availability becomes increasingly important in limiting phytoplankton biomass as I Corresponding author. Acknowledgments We thank G. Hutchinson, F? Burgess, M. Serediak, and numerous others for their contributions to field sampling and chemical analysis. We also thank M. Agbeti for providing phytoplankton counts for Camp Lake. Discussions with R. W. Howarth, R. Marino, H. W. Paerl, and A. J. Horne were invaluable in the preparation of this manuscript. Two anonymous reviewers, W. I? Dinsmore, and R. D. Robarts also aided us in improving manuscript content and style. Financial support was provided by NSERC in the form of an operating grant to E.E.P. salinity increases. N limitation in these systems may be due to low rates of planktonic N, fixation (Howarth et al. 1988b). Many estuaries and saline lakes with high available P concentrations lack the large populations of N,-fixing cyanophytes typical of high P freshwater lakes (Howarth et al. 1988b; Marino et al. 1990). Previous explanations of low N,-fixing cyanophyte densities in estuarine phytoplankton communities included turbulence (lack of anoxic microzones; Paerl et al. 1987) and low bioavailabilities of the key trace nutrients molybdenum (MO; Howarth and Cole 1985) and iron (Fe; Howarth et al. 1988a). Because many ions are conservative in seawater, marine water chemistries tend to be tightly linked with salinity and it is difficult to isolate the biological effects of individual components. Saline lakes can provide the chemical variability needed to test some of the above hypotheses. Marino et al. (1990) found that the ratio of sulfate (SOd2-) to MO was inversely correlated with the relative abundance of N,-fixing cyanophytes in 13 SO, 2--dominated prairie saline lakes. Fe (measured as total Fe) was not correlated with N,-fixer abundance and was >25 pg liter-’ in all of the study lakes. The Marino et al. (1990) study was entirely based on surveys of relative N,-fixer biomass and chemical parameters. Our study now goes on to examine whether the bioavailability of nutrients other than MO (such as Fe and P) affect relative N,-fixer abundance and phytoplankton biomass in individual lakes. We examined water chemistry and phytoplankton communities in 11 saline lakes with high total phosphorus concentrations (up to 26,000 kg liter-‘; Evans and Prepas 1996) to examine whether lake phytoplankton biomass is limited (restricted by low bioavailability of a given nutrient or nu-