Delayed-Mode Calibration of Autonomous CTD Profiling Float Salinity Data by θ–S Climatology

Abstract

Autonomous CTD profiling floats are free-moving floats that report vertical profiles of salinity, temperature, and pressure at regular time intervals. The Argo program plans to deploy 3000 such floats to observe the upper 2000 m of the global ocean. These floats give good measurements of temperature and pressure, but salinity measurements may experience significant sensor drifts with time. The moving nature of these floats means that it is too expensive to retrieve them regularly for physical calibrations. Thus a system has been set up to correct the drift in these profiling float salinity data by using historical hydrographic data. An objective mapping technique is used to estimate the background climatological salinity field on ? surfaces from nearby historical data. Temporal variations in water mass properties are accounted for in the objective estimate. The float salinity data are fitted to the background climatology in potential conductivity space by weighted least squares with a time-varying slope. The error associated with estimating the background climatology is carried through in the weighted least squares calculations. The result is a set of calibrated salinity data with error estimates. Because of the need to accumulate a time series for calculating a stable slope correction term, this system is a delayed-mode quality control system, with reliable calibrations available a few months after float data are obtained. However, contemporary ship-based measurements are essential in determining whether a measured trend is due to sensor drift or due to natural variability.