Members of the XBT (Expendable BathyThermograph) community met in November in Beijing, China, at the 4th XBT Workshop: XBT Science and the Way Forward. The meeting was an international collaboration, with 30+ attendees from regions including USA, Australia, Europe, China, India, Brazil and Japan. XBT data make up around 15% of the global historical upper-ocean temperature profile database and is used for many applications, including to study ocean currents, ocean heat content, climate change and meridional heat transport. XBTs provide a unique data set that cannot be obtained from any other existing observational network. It has long been known that XBT data has biases which impact on heat content estimates, and many correction schemes have been developed over the last 30 years. The 4th XBT workshop met with the goal of publicising the current ‘best practice’ for correcting historical XBT data. The final recommendations from the meeting are:

• Bias corrections are only required for certain climate-related science applications including ocean heat content estimates and meridional ocean heat transport estimates. Most other applications of XBT data do not require correction;
• XBT correction schemes should include time, probe type and temperature variable corrections, as well as a method of accounting for a depth offset in the depth bias correction;
• Essential XBT metadata should be collected to assist in assessing future bias errors, including: probe type and manufacturer, fall rate coefficients used, recording system, software version and launch height.

Strategies to develop future corrections were discussed, and the group identified the areas for future work as: investigations into the cause of the depth offset, computational fluid dynamics modelling of different probe types to assess fall rates, comparing analog and digital recorders identify the source of temperature biases, and the influence of water temperature on the temperature bias. The full recommendations and findings will soon be published in a short paper.