Long-term stability of salinity measurements by autonomous observing systems with RBR inductive conductivity sensors

Quantifying the accuracy and stability of measurements collected by autonomous observing systems in the absence of regular calibration is a problem widely recognized by the scientific community. Specifically, the electrode-type conductivity cells installed in many instruments are susceptible to fouling by a biofilm formation inside the cell, which alters the conductivity and thereby the salinity measurements. Inductive conductivity sensors have an advantage over electrode-type cells in that they lack metallic elements in contact with seawater, making them more resistant to surface contamination. Another benefit of inductive sensors is that they can be designed for natural flushing, and therefore do not require a pump. As a result, they have very low power requirements, which enables longer deployments and higher sampling rates.

In this study, we assess the long-term salinity stability from multiple RBR CTDs. The data were collected by CTDs on Argo floats and long-term moorings, and by CTDs returned to RBR for factory calibration. The quantitative assessment of salinity drift is performed by comparing data with other sensors and climatological data products. All methods of assessment are subject to errors associated with delineating sensor problems from the changes in ocean water measured by different instruments. Nevertheless, the results of this study demonstrate high level of stability of the salinity measurements collected by RBR inductive conductivity sensors. For example, salinity measured by the Argo floats with RBR CTDs working in different regions of the Pacific Ocean have drifted at a rate of <0.005 psu/year over two- to four-years of deployment.