The impact of profiling float data on the estimation of mean climatological temperature and salinity fields.

Tim Boyer, Sydney Levitus, John Antonov, Ricardo Locarnini

Ocean Climate Laboratory/National Oceanographic Data Center
1315 East-West Highway
Silver Spring, MD 201910
USA

Profiling floats have greatly expanded the spatial and temporal reach of subsurface oceanographic temperature and salinity data collection. We now have a source of data from geographic areas and seasons which were previously data sparse or never even sampled. This creates the possibility to enhance estimation of climatological fields for both temperature and salinity, leading to better initial and boundary conditions for models, among other benefits. But this also creates new problems in the estimation of climatological fields.

All float data has been collected within the last 11 years, the majority in the last 5 years. Climatologies could potentially be skewed toward values prevalent in the last 5 years, especially for salinity.

Further, profiling floats come with their own set of problems, including pressure hysteresis and salinity drift. The sheer amount of data coming in also presents challenges in quality assurance. We use the profiling float data from the last 11 years to perform some experiments to quantify the impact of profiling floats on climatological mean fields of temperature and salinity. Further experiments use the delayed mode salinity data in climatologies to assess the impact of these data in comparison to the real time salinity. Finally, the additional quality control needed to prepare the profiling float data for use in climatological calculations is detailed. The data used are the float data from the GODAE server, along with float data available from the WOCE data set and from the GTSPP data set. Other data types are from the World Ocean Database.