Arctic. The first few days on board are usually hectic as we reinstall
instrumentation, fire everything up, and then start trouble shooting the
inevitable problems. As first weeks of cruises go, this has not been one
of our better ones; we’ve suffered rather more technical problems than
usual. To make things more challenging, half of the problems are with
kit that we’re running on behalf of colleagues from other institutes,
and with which we’re not fully familiar.
We joined Oden in Longyearbyen on Svalbard last Monday (Aug 8). The
meteorology team – myself and John from Leeds, Piotr from Stockholm, and
Anna from SMHI in Norrkoping – immediately started sorting out all the
instrumentation that needed reinstalling. Most of the work was done back
in May, and all the sensors installed and tested, but we didn’t want to
leave them on the mast for the two months before the cruise started.
We got everything installed and powered up, and almost immediately found
our first problem. The Licor gas analyser, which makes water vapour
measurements, wasn’t starting up properly. We took the spare up, and
tested that: working! So, a complete swap of sensor and it’s interface box.
The next problem was the radio sounding kit. This is the ground station
for weather balloons. Data from weather balloons is used to initialise
weather forecast models, providing measurements of temperature,
humidity, and winds from the surface up to an altitude of about 25 km.
When we first ran through a dummy sounding – just running the radiosonde
on deck – to check the configuration of the system, we got no reception
of the radio signal. After several days of intermittently trying to get
this working – checking all the cables to antennas, double checking them
on a standalone radio receiver, etc, and many email exchanges with the
manufacturer’s engineers, we think the radio receiver card in the
sounding station has failed. A week on and we’re still testing and
trying out possible fixes but with no luck yet. This is a major blow to
the meteorology programme; there are very few direct measurements in the
Arctic, and forecast centres are always keen to get any they can.
Fortunately, it doesn’t actually impact most of the other science we’re
Meanwhile, back on the foremast another problem had cropped up. The
sonic anemometer – a sensor that uses pulses of sound to measure the
3-dimensional turbulent wind and air temperature – was outputting error
messages instead of data. The error messages indicated a mixture of
implausible measurements and a complete lack of signal on different
pairs of transducers. I initially thought this was a calibration
problem; the instrument was calibrated in a nice warm lab, worked OK
when first installed in May, but was now operating in sub-zero
temperatures. We tried to do a field-calibration on the lift platform
alongside the top of the mast, but with no luck. So, stripped it all off
the mast for tests back in the lab. So far we’ve failed to get it
working, but have determined that temperature changes affect which
transducers work and which don’t. Meanwhile we’ve installed another
sonic anemometer; this is working, but doesn’t have a heated head like
the first one, so if we get heavy icing conditions we’ll lose data.
Because most of the science on this cruise is piggy-backing on a
Canadian led sea-bed mapping expedition, we have to fit in around the
other work on board. We can only get up the mast when the ship is
stationary and conditions not too harsh. For the last couple of trips up
this has meant working between 11pm and 3am the following morning. Last
night’s trip up the mast was also pushing the limits of workable
conditions; with winds of 16 m/s we were swaying around rather more than
we were entirely comfortable with.
But for now at least, we’re finally getting the majority of the
measurements we really need. A great relief.