Ref. No. [UMCES] CBL 2013-020
ACT VS12-03
month of collection. A dual-beam spectrophotometer was blanked with MilliQ water in cuvettes
in both the sample and reference positions. Matched 10 cm quartz or optical glass cells were
used for a dual-beam spectrophotometer. MilliQ samples were run intermittently during each
analytical batch to assess instrument baseline drift. Scans were run between 200 and 800 nm and
electronic files were saved for each sample. MilliQ blank and turbidity (750 nm) corrected
spectra were used to estimate CDOM abundance by non-linear regression of the absorption
spectra over 400 – 575 nm.
a[
λ
] = a[400]e
(-Sλ)
(1)
Where
a[λ]
is absorption (m
-1
) at wavelength
λ
,
a[400]
is absorption (m
-1
) at the anchor
wavelength of 400 nm, and
S
is the spectral slope (nm
-1
). Note that wavelength must be
expressed as
–
400
before fitting for the anchor value to be at 400 nm.
A[400]
is used as a
proxy for CDOM abundance in reference samples.
Chlorophyll a
Chlorophyll grab samples were analyzed on a Turner Designs 10AU fluorometer from
samples filtered on 2.5 cm GF/F filters and frozen at -20
o
C until analyzed according to Parsons,
et al. 1984. Optimum filtration volumes were determined on site. All chlorophyll analyses were
performed by the Chesapeake Biological Laboratory according to their existing SOPs. The
laboratory is a State of Maryland certified lab and has undergone previous audits by ACT during
prior evaluations. Samples were shipped to CBL in liquid nitrogen dry shippers to ensure they
remained frozen at the required temperature.
Turbidity
Turbidity concentrations of reference grab samples were determined by a Hach 1100AN
benchtop turbidity sensor in NTU. The lab analyzer was calibrated with certified standards prior
to use and a QA check of the standards were run during each analytical batch. Samples were run
immediately upon collection. The same instrument was used at each test site.
Laboratory Tests
Performance against surrogate standards and challenge environmental variables
Laboratory tests of response factor, precision, range, and reliability were conducted at
Moss Landing Marine Lab. Challenge compounds utilized in laboratory characterizations of
instrument performance are listed in Table 1 and cover the range of optical detection windows
utilized by participating hydrocarbon sensors. Laboratory challenges were performed in
insulated 500 L, black acrylic tanks in a dark room using filtered deionized water (DI) as the
background medium. Test tanks have been preconditioned by several years of use with
deionized and seawater exposures and cleanings. Temperature was maintained at 15 ± 1
o
C with
Nestlab recirculating chillers and copper heat exchange tubing. Water was continuously
circulated with submersible pumps (
ca
10 L/min) placed at opposite ends of the tank.
Temperature was monitored at opposite ends of the tank at sensor detector level by two
calibrated RBR 1060 recording thermometers. Each test level began with a 30 minute
equilibration, and reference water samples were collected at 10 minute intervals over the
7
