Comparisons with shipboard CTD casts were done for D0001, D0002, D0003, D0004 and R0004. For each comparison, salinity from the profiler was lower than salinity from the CTD cast (D0002 difference more pronounced at depth. R0004 difference more pronounced closer to the surface). For D0004, conductivity was also lower for the profiler compared to the CTD cast. Comparison to the shipboard bottle data is outside the scope of this project.

By Lori Garzio, on 3/8/19

Missing data at the end of deployment 4 should be annotated.

By Lori Garzio, on 8/5/19

I submitted a helpdesk question on 11/12/2018 regarding an offset between recovered and telemetered data for all global HYPMs: "There appears to be a substantial offset between data with the same timestamps from recovered_wfp and telemetered methods for all deployments of all global WFP CTDs, and I have included information from two representative deployments of two different profilers as examples. It looks like the values recorded for the first common timestamp are exactly the same, and as the profiler moves along in time the offset between the values recorded for the same timestamp become increasingly more offset. Then, the profiler reaches the end of the profile (either at its shallowest or deepest point), resets, and the pattern begins again. This is most easily seen in the pressure data, but there is an offset for all science parameters.

Examples:
GI02HYPM-WFP02-04-CTDPFL000 deployment 1
GP02HYPM-WFP02-04-CTDPFL000 deployment 3

The *_intersect.csv files attached contain data where recovered and telemetered timestamps intersect. For each timestamp and science parameter there are recorded values (from .nc files downloaded from the system) and the difference. For each parameter, the difference between the values at the bottom and top of the profiler's range are exactly zero, and (particularly for pressure) the differences steadily increase until they are reset at zero at the bottom/top of the profile. This pattern continues throughout the deployment. The plots attached are zoomed in to one upcast and one downcast, and highlight the offset in the data between the two methods."

Redmine 13743

By Lori Garzio, on 8/5/19

Too much variation in salinity at the beginning of the deployment - not seen in previous deployments

By Lori Garzio, on 8/5/19