Lake Bonney, which is characteristic of many of the lakes in the antarctic dry valleys with respect to salinity profiles, derives its primary hydraulic input from glacial streams during the austral summer. A wide variety of salts are introduced into the lake through this inflow. Because of continuous ablation and no outflow, salinities in Lake Bonney range from near 0 parts per thousand (ppt) below the ice cap to 247 ppt (nearly 7 times open ocean) at 35 meters (m) (Spigel et al. 1991). Our current research project on Lake Bonney (see Priscu, Ward, and Downes, Antarctic Journal, in this issue) utilizes the stable isotope nitrogen-15 to measure rates of nitrogen transformation directly. Many of our experiments using this isotope require ammonium (NH₄⁺) or nitrite (NO₂^-) to be extracted from solution. The wide range of salinities encountered at our experimental depths in Lake Bonney prompted us to test extraction efficiencies of nitrogen and potential isotopic discrimination as a function of salinity. Such tests are necessary if accurate measurements of nitrogen transformations are to be obtained.

We used a modification of the NO₂^- extraction protocol of Olson (1981) that is based on chemical complexation of sample NO₂^- with aniline sulfate under acidic conditions.

This reaction yields a diazonium salt that is condensed with alkaline beta-naphthol to form a base-soluble azo dye which is partitioned from the aqueous phase through repeated extractions in a nonpolar solvent. After the absorbance [at a wavelength of 500 nanometers (nm)] of the solvent-dye mixture was measured, the solvent-dye mixture was dried and analyzed for its 15-nitrogen content using emission spectrometry (Timperly and Priscu 1986). (NH₄⁺) was extracted using the molecular sieve, zeolite (IONSIV W85). Zeolite was introduced into the sample at 1 milligram per milliliter and allowed to react for 30 minutes during which time the sample was mixed vigorously several times. Extraction was followed by filtration of the zeolite-ammonium complex onto a precombusted Whatman G/FC filter. Extraction efficiency was determined by measuring (NH₄⁺) in the sample before and after extraction.

The effects of salinity on NO₂^- extraction were tested by preparing a series of 40 micromolar (20 atom-percent 15-nitrogen) NO₂^- solutions in saline (sodium chloride) solutions which were similar to salinities at our sampling depths in Lake Bonney. Sodium chloride was utilized because it is the predominant salt in the lake (Priscu and Spigel unpublished data). Results from this experiment showed that absorbance (a measure of NO₂^- extracted) decreased rapidly with salinity (figure 1A) indicating a strong influence of salinity on NO₂^- extraction efficiency. The nitrogen-15 atom percent also decreased with increasing salinity showing that the extraction technique discriminated against nitrogen-15 when higher levels of sodium chloride were present.

Because it is necessary to have precise amounts of nitrogen for emission spectrophotometry (see Timperly and Priscu 1986), we further examined the mass of nitrogen in the dye measured with an elemental analyzer (Carlo-Erba 1500) and compared it to spectrophotometric measured absorbance of the dye in a 1 centimeter cell at 500 nanometers (figure 1B.

These measurements were made on a series of deionized water samples (type I water) amended with varying levels of NO₂^-. The relationship between absorbance and nitrogen mass was essentially linear (r²=0.93, y-intercept=-0.929 absorbance units, slope=0.013 absorbance units per microgram nitrogen per milliliter). It should be noted that the nitrogen mass includes both the NO₂^- in the sample and the nitrogen contained in aniline (one atom per mole aniline).

Zeolite extraction efficiency of NH₄ also decreased with increasing salinity in lake water samples from selected depths (figure 2). Extraction efficiency can be improved significantly if NH₄⁺ in the samples is steam distilled for 1 hour in a rotary evaporator (to eliminate salt interference) before zeolite discriminates against 15-nitrogen as salinity is increased. Experiments are currently in progress to verify this latter prelimniary finding.

In conclusion, we have shown that, when working in lakes of the dry valleys near McMurdo Sound, a number of preliminary experiments must be conducted for accurate extraction of various forms of inorganic nitrogen. If results from these experiments are not considered, sample analysis amy be impossible or inaccurate results may occur.

This work was supported by National Science Foundation grant OPP 91-17907 to John Priscu.

References

Olson, R.J. 1981. 15N tracer studies of the primary nitrite maximum. Journal of Marine Research, 39, 203-226.

Priscu, J.C., B.B. Ward, and M.T. Downes. 1993. Water column transformations of nitrogen in Lake Bonney, a perennially ice-covered antarctic lake. Antarctic Journal of the U.S., 28(5).

Spigel, R.H., I. Forne, I. Sheppard, and J.C. Priscu. 1991. Differences in temperature and conductivity between the east and west lobes of Lake Bonney: Evidence for circulation within and between lobes. Antarctic Journal of the U.S., 26(5), 221-222.

Timperly, M.H., and J.C. Priscu. 1986. Nitrogen-15 analysis by optical emission spectrometer. Analyst, 111, 23-28.