Geoscience Reference
In-Depth Information
studies (i.e., do not change filter pore size within a study as it likely means results from
different filters will not be comparable). It also raises significant concerns about the com-
parability of data across studies, especially if comparing studies with samples collected
with extremely different filter cutoffs (e.g., 0.2 µm vs. 1.2 µm) and comparison of the
tryptophan-like fluorescence or any ratios involving this fluorophore.
The influence of suspended particulates on
in situ
fluorometers is of great interest as one
of the great attractions of these instruments is the ability to provide high-frequency data
with the need for little if any sample processing, thus avoiding any potential contamination
from filtration or sample storage (Del Castillo et al.,
2001
; Spencer et al.,
2007a
; Conmy
et al.,
2009
). Utilizing a WETStar fluorometer (WET Labs), Belzile et al. (2006) showed
a strong correlation between WETStar fluorometer measurements on unfiltered samples
from a range of aquatic environments and filtered samples measured using a spectroflu-
orometer with suspended sediment concentrations of up to 35 mg L
-1
. However, a recent
study by Saraceno et al. (
2009
) shows evidence that at high concentrations of suspended
sediments that may occur in turbid rivers or during flushing events (e.g., freshets) a reduc-
tion in DOM fluorescence as measured by a WETStar fluorometer was reported. Saraceno
et al. (
2009
) state that the combined effects of scattering and absorption, due to high sus-
pended particle concentrations, resulted in an underestimate in their DOM fluorescence on
unfiltered water. Correction for particle interference, though, is possible, but as with most
optical measurements, relationships will need to be “ground-truthed” to individual sites
and water/sediment types as appropriate.
4.4 Storage
4.4.1 General Comments
The primary causes of sample instability are due to microbial and photochemical degra-
dation, and the effects of these processes on DOM absorbance and fluorescence in aquatic
samples are well documented (Moran et al.,
2000
; Osburn et al.,
2001
,
2009
; Del Vecchio
and Blough,
2002
; Stedmon and Markager,
2005
; Cory et al.,
2007
; Tzortziou et al.,
2007
;
Wickland et al.,
2007
). Therefore, samples should be filtered (see
Section 4.3.1
) and stored
in the cold and dark (e.g., refrigeration in the dark at approximately 4°C) for short-term
storage and analyzed as soon as possible after collection. It is not always realistic, however,
to analyze samples immediately after collection for a range of reasons (e.g., remote field
sites, large number of samples collected in a short time period) and so researchers routinely
have to store samples for DOM analyses. Aside from analyzing samples as soon as possible
there is little general consensus on the length of time for which it is appropriate to store
samples. The application of different storage methods and appropriate storage timeframes
should be tested by the researcher on the range of DOM samples they are examining, as
results are likely to vary with the source and history of different samples (e.g., highly
colored allochthonous dominated DOM samples vs. optically clear autochthonous dom-
inated DOM samples) and also the filtration technique and cutoff utilized. It is strongly
