Geoscience Reference
In-Depth Information
and elemental analysis to study the chemistry of purely microbially derived fulvic acids
isolated from lakes and streams in the McMurdo Dry Valleys, a plant-free desert oasis near
Ross Island in Antarctica. Results showed that these microbially derived fulvic acids had
a lower carbon to nitrogen ratio and lower aromaticity than plant/soil OM, reflecting the
absence of lignin in the precursor pool of organic molecules (McKnight et al.
1994
; Aiken
et al.
1996
). An opportunity to compare Dry Valley lake samples with lake samples from
the Bunger Hills, another desert oasis in Antarctica, motivated a search for a fluorescence
index that could be used on small-volume water samples.
Based on the work of Ewald and Belin (
1987
), who examined differences in fluores-
cence between marine samples at an excitation wavelength of 370 nm in the peak C region,
a trial-and-error approach was used to identify changes in the fluorescence signal of various
fulvic acids. This study confirmed that at an excitation wavelength of 370 nm, there was
a significant difference in the emission line scan between an extracted fulvic acid from a
dry valley lake and the International Humic Substances Society (IHSS) reference fulvic
acid from the Suwannee River, which drains the Okefenokee Swamp, a large wetland in
the southeastern United States. Whole waters from the Antarctic lakes and whole waters
and isolated fulvic acids from North American streams and rivers were also analyzed. The
spectra were inner-filter corrected and a fluorescence index (FI) was developed based on
the ratio of the emission at 450 nm to 500 nm (ex = 370 nm) to quantify the sharper and
narrower peaks of the Antarctic lake fulvic acids compared to the broader peak of the
Suwannee River fulvic acid and other plant/soil-derived samples. The 450 nm wavelength
was chosen because it was close to the mean of the maximum intensity for the dataset, and
500 nm because it was close to 50% of the peak emission for the microbially derived fulvic
acids. The samples from the Antarctic lakes had the highest FI (1.7-2.0), corresponding to
DOM derived from a microbial source associated predominantly with phytoplankton prod-
uctivity. In contrast, the North American river samples had a lower FI (1.3-1.4), resulting
from their origins in plant litter and soils. EEMs obtained for a few samples confirmed that
increases in this index corresponded to a shift of the main peak C fluorophore to a lower
emission wavelength. As expected, for the isolated fulvic acid samples, a good correl-
ation was observed between the FI and the percent aromaticity as determined by
13
C-NMR
(
Figure 9.6
). However, in a set of stream fulvic acid samples, removal of more aromatic
fulvic acid by sorption onto streambed iron oxides caused a large decrease in aromaticity
but only a small change in the index. This can be seen in
Figure 9.6
where the Snake River
(SR), which is fed by Deer Creek (DC), has a lower aromaticity but comparable fluores-
cence index. This observation indicates that the index reflects primarily organic precursor
material rather than aromaticity. McKnight et al. (
2001
) commented that the FI would
potentially be useful in screening samples to identify changes in DOM source and quality
that could be explored using other more detailed measurements. McKnight et al. (
2001
)
also noted that the range of values of the index would be instrument dependent.
Subsequent work by Cory and McKnight (
2005
) refined the FI based on instrument
corrections applicable to all fluorometers. The refined FI is calculated as the ratio of emis-
sion intensity at 470 nm to that of 520 nm at an excitation of 370 nm. This study analyzed
