Table 1.
Physical, chemical, and biological parameters measured at 3 Lake Mendota sites.
Figure 1.
N2 fixation rates, microcystin, N and P concentrations, and community dynamics in Lake Mendota from 2010 and 2011. A)
2010 average nitrogen fixation rates (µg N L−1 hr−1) and microcystin-LR concentrations (µg L−1); B) 2010 dissolved inorganic nitrogen to dissolved reactive phosphorus ratios (DIN∶DRP) and total nitrogen to total phosphorus (TN∶TP) ratios by weight. The dashed line represents the Redfield reference ratio of 7.2 N∶P by weight [58]; C) 2010 DIN and DRP concentrations (mg L−1); D) 2010 sum of Aphanizomenon and Microcystis genotypes in relative fluorescence units; E–H) Same as A–D, but for 2011, respectively. Note the log scale in panels B and F. The arrows indicate the first day N2 fixation rates were significantly above the limit of detection. For plots A, C, and D, the error bars represent the standard error of the mean between the three sites sampled. For plot E, the error bars represent the standard deviation of triplicate samples, and in plot B, the error bars represent the propagation of error for N and P between the three sites sampled. The grey box indicates the toxic-phase, which is bound by the pre- and post-toxic phases, where MCLR concentrations were greater than 1 µg L−1 (see text for details).
Figure 2.
Multivariate plots for 2010 and 2011. A)
2010 canonical correspondence analysis (CCA) of potential environmental drivers on the cyanobacterial community composition assessed using APISA. B) 2010 CCA of the cyanobacterial community composition with the top ten most abundant genotypes as drivers of the cyanobacterial community. C–D) Same as A and B but for 2011, respectively. Numbers in parentheses represent the amount of variability explained by those axes.
Figure 3.
Major genotype time-series from 2010 and 2011 in Lake Mendota.
A) Two major Aphanizomenon genotypes and a third, yet to be distinguished genotype that represents a PC-IGS fragment similar to Aphanizomenon, Anabaena, and Chroococcus. B) Three major Microcystis genotypes identified. C–D) Same as A and B, but for 2011, respectively. In 2010, Mic215 was significantly correlated to MCLR. However, no genotypes were significantly correlated to MCLR in 2011. The grey box indicates the toxic-phase, which is bound by the pre- and post-toxic phases, where MCLR was greater than 1 µg L−1 (see text for details). The error bars represent the standard error of the mean between genotypes of the three sites sampled.
Figure 4.
Conceptual diagram of the potential role of nitrogen stress in Lake Mendota, WI.
A decline in dissolved inorganic nitrogen (DIN) could have two simultaneous effects: 1) stimulate N2 fixation by diazotrophs (e.g. Aphanizomenon) that will lead to new N inputs into the photic zone and 2) lead to N starvation in non-N2 fixers (e.g. Microcystis), which may potentially stimulate toxin production. The new N inputs could then support growth, prolonging toxic blooms of non-N2 fixers throughout the growing season.
Figure 5.
Monthly values from the North Temperate Lakes-Long Term Ecological Research program (1995–2010).
A) Temperature (°C) profile from 0–24 m depth (averaged across years), with the gray boxes indicating the average ice-on dates, B) DIN∶DRP and TN∶TP ratios by weight, for each year. The dashed line represents the Redfield reference ratio of 7.2 N∶P by weight [58], C) Total Aphanizomenon and Microcystis abundances averaged across years (cells mL−1), D) 2010 DIN from the North Temperature Lakes-Long Term Ecological Research (NTL-LTER) program (mg L−1), 2010 DIN from this study (mg L−1), and 2010 MCLR concentrations (µg L−1) from this study.