Geoscience Reference
In-Depth Information
2.1
1.9
y = -1.2x + 2.0
R2 = 0.84
1.7
1.5
1.3
1.1
0.9
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
SQ1/(SQ1 + SQ2)
Figure 9.7. Two components of the Cory and McKnight (
2005
) PARAFAC model that have spec-
tral properties similar to semiquinones (SQ1 and SQ2) explain the variation in the McKnight et al.
(
2001
) fluorescence index (FI). The SQ2 component is resolved in a model using only samples from
Antarctic lakes, indicating it is likely microbial in origin. As the portion of SQ2 to the total of SQ1 +
SQ2 falls, so does the FI, and a lower FI is associated with less microbially derived material. (From
Cory and McKnight,
2005
.)
(
2007
) further investigated the effect of photodegradation on Arctic samples and the FI,
demonstrating that for these samples the FI was insensitive to photodegradation. These
results provide evidence that FI is tied to underlying chemistry based on precursor mate-
rial. Finally, Cory et al. (
2010
) compared fluorometers and instrument correction factors,
and showed that if an instrument has sufficient sensitivity, application of correction fac-
tors can allow for direct comparison of FI values in different studies. For example, Jaffé
et al. (2008) analyzed, at two different laboratories, a set of diverse samples ranging from
marine surface waters to soil interstitial waters in the Arctic tundra. The FI values from the
two laboratories were highly correlated (
r
2
= 0.87), with a near zero intercept once proper
instrument correction factors were applied.
9.2.5 The “Peak T/Peak C Ratio” to Identify Sewage Impact on Rivers
Baker (
2001
) presented fluorescence EEMs from a selection of small UK rivers impacted
by sewage effluents. Samples from the effluents themselves and from the rivers upstream
and downstream of the sewage discharge demonstrated a change in fluorescence signa-
ture due to sewage input, indicating the potential use of fluorescence to analyze the water
quality of sewage-impacted rivers. Tryptophan-like peak T (called peak A in the paper)
was measured at 275 nm excitation, 350 nm emission. Fulvic-like peak C (called peak B
in the paper) was measured at the wavelengths of maximum intensity within 320 -340 nm
excitation, 410-430 nm emission. The ratio of peak T to peak C was used to differen-
tiate upstream and downstream water samples. Upstream of the discharge sites, the un-
impacted rivers had a protein-like/fulvic-like fluorescence intensity ratio of around 0.6 due
to the peak C intensity being greater than the peak T intensity. The sewage effluents had a
