A Quality-Control Procedure for Surface Temperature and Surface Layer Inversion in the XBT Data Archive from the Indian Ocean*

Abstract

Expendable bathythermograph (XBT) depth error associated with the probe’s fall rate has been reported in many earlier studies. Hanawa et al. (1995) made a comprehensive evaluation of the fall rate problem and suggested a global algorithm to correct the XBT depth, though the applicability of the algorithm in waters of extreme temperature is yet to be resolved. The other two potential sources of error in XBT data are the sea surface temperature (SST) and surface layer inversion (SLI). A quality-control procedure has been developed to address these sources of error in SST and SLI of XBT data from Indian Ocean, archived at Responsible National Oceanographic Data Centre for Indian Ocean region (RNODCINDO). The surface temperature error due to start-up transient in the digitizer electronics was reported by Roemmich and Cornuelle (1987). They observed that the magnitude, sign, and duration of this transient vary from instrument to instrument and even from cast to cast. One way to overcome the start-up transient problem is to flag the data up to the transient depth for poor quality (Bailey et al. 1993). The other alternative is to correct the surface temperature by bucket temperature, measured prior to the launch of an XBT probe. In the former case, SST, a prominent parameter required in general circulation models (GCMs) and other major environmental applications, will be missing from the prodigious XBT coverage over the global ocean. The procedure of correcting XBT surface temperature (XST) with bucket temperature sounds good. However, in many cases such corrections were not possible as the bucket temperatures were not reported along with the XBT data. Even when reported, the values need not be accurate enough as it