Geoscience Reference
In-Depth Information
25
120
Scanning fluorometer
in situ
sensor calibrated
in situ
sensor raw values
100
20
80
15
60
10
40
5
20
0
0
200
300
400
500
Wavelength (nm)
600
700
800
Figure 6.10. Emission spectra from a multispectral
in situ
sensor and a scanning fluorometer demon-
strating pre- and post-calibration.
is then similar to the QSE approach. Most commonly, the NOM method utilizes discrete
filtered samples that are collected coincident with fluorescence measurements from a field
sensor (Del Castillo et al.,
2001
; Conmy et al.,
2004
). Samples are analyzed with a bench-
top fluorometer, where data are fully corrected and calibrated to QSE. A scaling factor is
applied to the
in situ
data through the equation,
FL
FL
lab
scalingfactor =
(6.2)
in situ
where FL
lab
and FL
in situ
are fluorescence intensities for discrete samples and
in situ
data,
respectively. This secondary standard method is commonly used for multichannel fluorom-
eters, where calibration factors are calculated for each wavelength channel (
Figure 6.10
)
(Del Castillo et al.,
2001
; Conmy et al.,
2004
). This method can also be used to convert
field sensor measurements to Raman Equivalents (REs), which is an intensity normalization
using the water Raman peak instead of QS, if the discrete data are calibrated in this manner.
However, analysts should be aware that at the time of this chapter's writing, direct calibra-
tion to the water Raman peak cannot be conducted with commercial field sensors due to
their wide filter bandpasses and filter wavelengths not being centered on Raman bands.
6.4.4 Correction for Inner Filter Effects
Field fluorometers are potentially subject to inner filter effects (IFEs) at high OM con-
centrations, just as are benchtop instruments. This effect is observed through an apparent
