MERCURY IN PHYTOPLANKTON AND ZOOPLANKTON

 

Edward A. Nater (ed.nater@soils.umn.edu), Deborah L. Swackhamer, Jacob A. Fleck

University of Minnesota

 

ABSTRACT

 

Total mercury was determined in samples of phytoplankton and zooplankton collected from 14 sampling stations in Lake Michigan. Samples were collected on 7 voyages conducted over a period of two years. Spatial trends in the data showed relatively similar levels at all stations with the exception of site 310, located on the Michigan shore just north of the mouth of the Kalamazoo river, which had mean Hg values in phytoplankton and zooplankton more than twice the mean of all other stations. Temporal trends showed lower zooplankton Hg concentrations in spring and early summer; however, during those periods the "zooplankton" samples contained significant quantities of filamentous diatoms, thus "diluting" the zooplankton concentrations. Phytoplankton concentration means were higher in fall and winter and relatively constant from April through October.

 

INTRODUCTION

 

This research was part of the USEPA's Lake Michigan Mass Balance Study (LMMBS). Our portion of the study concentrated on mercury in phytoplankton and zooplankton, the lower trophic levels in Lake Michigan. Sampling was conducted in accordance with the overall plan for the LMMBS and in concurrence with other researchers on the same project.

 

METHODS

 

Plankton was collected from Lake Michigan waters at the mid-lake stations (18M, 23M, 27M, 40M, 47M), the biology boxes (110, 140, 180, 240, 280, 310, 340, 380), and the Chicago station (5). Collections were made jointly by Nater's and D.L. Swackhamer's research groups who were analyzing for mercury and for persistent organics, respectively. The plankton fractions were separated based on net capture sizes. Working definitions for zooplankton and phytoplankton were established by separation at the following net sizes: zooplankton > 100µm > phytoplankton > 10 µm.

Zooplankton were collected by vertical net tows using a 1 m diameter net with 100 µm mesh netting. Repeated tows were conducted until a sufficient quantity of sample was obtained for analyses. Typically, from one to three tows were required, though more tows were sometimes required in shallower waters due to the smaller volumes of water being sampled. Phytoplankton were collected using the "phytovibes", a pair of large inverted-pyramidal nets supported by a stainless steel frame and equipped with a vibrating shaker to prevent plankton from plugging the pores. Water was pumped into the phytovibes using submersible pumps and nylon tubing at a rate of 30 to 40 L min-1; slower rates were associated with pumping from greater depths. The depth of sampling was selected to match the chlorophyll-A fluorescence peak obtained from the seabird recording. Once initiated, sampling continued until sufficient quantities of phytoplankton were obtained for determination of mercury and persistent organics (PCBs and others). Sampling times ranged from 9 to more than 15 hours, depending on plankton concentrations. Shorter times were usually associated with early to mid-summer sampling times. Sampling occurred at any hour of the day.

Samples were frozen in the field in PFA Teflon bottles and freeze-dried back in the laboratory. Subsamples analyzed for total mercury were weighed into clean 30 mL Teflon vessels and then digested in a mixture of concentrated HNO3 and concentrated H2SO4 at 70˚ C overnight. Digestates were analyzed for total mercury by cold vapor atomic fluorescence spectrometry using a modification of the method of Bloom and Crecelius (1983).

 

RESULTS AND DISCUSSION

 

The net size-based working definitions used did not always accurately separate phytoplankton from zooplankton. Filamentous diatoms dominated the water column during late spring and early summer and, due to their long lengths and relative abundance, were readily captured by the 100 µm net and often dominated the zooplankton net fractions.

Temporal trends were observed in the plankton data (Figure 1). In general, the means for phytoplankton held fairly steady around 30 ng g-1 from April until or during October, then increased in October and January. Zooplankton showed a contrasting trend, with an increase from June to August and then a slow decline until spring of the next year. The trend appeared to be repeated in the following year, although different sampling periods were used during the two years. This decrease in April through June, however, may be due to the large quantities of filamentous diatoms that were captured in the 100 µm zooplankton nets during spring and early summer, thus "diluting" the zooplankton values.

It is important to note that the data for the January cruise consist of 5 phytoplankton and 3 zooplankton samples, all taken from sites 5 and 380 with the addition of one sample from MB19M; consequently, these data do not hold much statistical power. However, the trends follow those observed from August to October for both phytoplankton and zooplankton.

 

 

 


 


Figure 1. Temporal trends in Phytoplankton and Zooplankton Hg concentration means. Calendar months are indicated for the 1994 and 1995 cruises, respectively.

 

 

Spatial trends were also observed in the plankton data (Figure 2), though they were less clear than the temporal trends. The most obvious spatial trend is the relatively high values observed for both phytoplankton and zooplankton at site 310. This site is located very close to the Michigan shore just north of the mouth of the Kalamazoo river. The means for total mercury in phytoplankton (79.0 ng g-1) and zooplankton (108.1 ng g-1) at site 310 were more than twice the means for phytoplankton (30.9 ng g-1) and zooplankton (50.0 ng g-1) at all other sites. However, the values for site 310 were extremely variable, ranging from lows of 16.8 and 19.1 ng g-1 to highs of 175.8 and 376.3 ng g-1, respectively for phytoplankton and zooplankton. This site accounted for the three highest phytoplankton values and two of the top three zooplankton values.

These high but variable concentrations are suggestive of strong influence from a localized source, the most obvious potential source being the Kalamazoo river. The high variability could also be consistent with a point source in that wind direction, currents, and sampling position with respect to the mouth of the Kalamazoo river could have a major effect on what was collected. Unfortunately, however, the Kalamazoo river was not sampled during the tributary analyses, so this is currently just speculation.

 


Figure 2. Spatial trends in Phytoplankton and Zooplankton Hg concentration means. Sites indicated are sampling sites for the USEPA Lake Michigan Mass Balance Study.

 

 

 


REFERENCES

 

Bloom, NS, Crecelius, EA (1983), Mar. Chem. 14:  49-59.