Ref. No. [UMCES] CBL 2015-011
ACT VS15-04
12
conducted in the CRM buffers as defined above. For instruments that pumped samples through
an inlet, the inlet was connected to the manufacturer’s supplied storage bag to ensure that no air
was introduced prior to exposure in the buffers.
Reference Water Sampling Schedule
– The sampling frequency was structured to examine
changes in pH over daily and weekly time scales. Specifically, an intensive sampling event was
conducted once a week that consisted of 5 sample collections within a day. During four
additional days of each week, there was a minimum of twice per day sampling, scheduled in a
manner to capture as much diurnal variation as possible. The initial intensive sampling event
occurred within the first two days of the deployment after all instruments had been deployed, and
the final intensive sampling event occurred during the last two days of the deployment. The
schedule provided a higher density of comparative data at the beginning when instruments
should have been functioning at optimum performance and again after the challenge of a four or
twelve week deployment. The sampling schedule resulted in at least 60 reference samples paired
with instrument measurements. For the 12 week deployment test at HIMB, the sampling scheme
was modified to spread out a similar number of samples over the extended time period. All
sampling times were recorded on logsheets and entered into a database for final data
comparisons.
Reference Water Sample Collection
– A standard 2.2 or 4.2L Van Dorn water sampler was used
at each field test site to collect water samples for reference pH measurements. Water sample
collections were timed to correspond directly with the instrument readings or sample intake. The
water sampler was lowered to the same depth of the instrument sampling inlets, and as close as
physically possible to the inlets (distant by no more than 0.5 m horizontal distance). The water
sampler was soaked at sampling depth for 1 minute prior to sampling. If water was not flowing,
the sampler was moved to ensure that it was flushed with the ambient water. The water sampler
was triggered to match the programmed sampling times of each instrument. Three replicate pH
samples were collected in clean 10 cm glass spectrometer cuvettes from each individual field
sample. The cuvettes were gravity filled from PTFE tubing connected to the sampling bottle
spigot such that the volume of the cuvette was exchanged three times (about 20-30 secs) before a
final sample is collected. Care was taken to ensure that no bubbles were left within the cuvette
before sealing the ports with their PTFE stoppers. Cuvettes were stored at ambient or slightly
cooled conditions during transport to the laboratory (travel times at various field sites ranged
from 5 to 20 minutes) until they were placed into the temperature equilibration chamber at the
specified temperature for analysis. In addition, at the freshwater site an additional 300 ml BOD
bottle was filled to allow for a lab-based electrode pH measurement on each field reference
sample. All samples were equilibrated to 25°C and analyzed on a lab electrode calibrated daily
before use.
Twice a week (day 1 and day 5) duplicate water samples were collected to characterize
all of the CO
2
parameters. (This sample collection occurred in conjunction with our field
duplicate sampling and one set of external partner samples came from each Van Dorn. This
protocol was mostly to facilitate limitations of volume but also helped to evaluate heterogeneity
at the mooring.) Water samples were collected and preserved for pH, pCO
2
, TCO
2
and TA
measurements following standard oceanic protocols (Dickson et al., 2007). In brief, a 500 mL
glass sample bottle (supplied by Andrew Dickson, SIO) was filled from the Van Dorn sampling
bottle using PTFE tubing and preserved with 110 µL of saturated HgCl
2
solution. Reference
samples shipped to Oregon State University (Burke Hales, OSU) were analyzed for pCO
2
and