Q&A: Temperature Mismatch

Can’t see the animated gif? Click Here.

Both instruments were logging accurate data that matched the other almost exactly – due to a simple mistake, the 16plusV2 SeaCAT simply started logging AFTER the other sensor. In this case, it was because the clock on the 16plusV2 was set incorrectly. A post-recovery breakdown shows the sequence of events leading up to this error:

1. The date/time on the 16plusV2 was programmed incorrectly in the lab, set to approximately 3 months later than the actual date.
2. The 16plusV2 was programmed to start logging at a delayed start time that matched the date/time of the deployment (with the commands “startdatetime=mmddyyyyhhmmss” and “startlater”).
3. Both sensors were physically deployed on the seafloor very close to one another. The SeaCAT was still waiting to begin logging (although the operator believed that it had started logging per the delayed programming), while the other sensor was manually configured to begin logging moments before the deployment.
4. After 3 months, the incorrect clock on the SeaCAT met the delayed start date/time, and began logging.
5. Plotting the data post-recovery by date offset the data between the two sensors in the resulting figure.

At first glance, it appeared that there were significant mismatches in temperature between the two sensors. Fortunately, visualizing the data showed a noteworthy peak in temperature towards the middle of the deployment. Since this unique shape was present in both datasets, we were able to match the shape in both sensors’ temperature time-series to determine that the SeaCAT began logging later than expected. Checking the clock after this analysis verified that it was indeed set incorrectly.