2014/09/25 - On 23 September, CLS switched IOBX02 data from Toulouse back to Largo. Drifting buoy data received at Meteo-France with header IOBX02 KARS are corrupted. The problem is not at CLS America that produces reliable native BUFR data with this header. A feedback was sent to NOAA/NWP yesterday. The problem ECMWF reported about moored buoy data fitted with header IOBX11 LFVW could be the same. For this latter, two dataflew are received at ECMWF: those coming from Exeter (EGRR) have been wrong since the 10th of September whilst those coming from Offenbach (EDZW) are correct. Here also CLS certifies their native BUFR data are correct.
2014/09/22 - ECMWF reports they can no longer decode IOBX11 LFVW messages (TAO/PIRATA/RAMA moored buoy networks). An element (0 03 007) not known and not present in any table is reported. The problem is forwared to CLS Toulouse.
2014/09/19 - CLS switches the sending of buoy data from Largo (KARS) to their backup service in Toulouse (LFVW) waiting for the problem is fixed in the US. Data fitted received at Meteo-France with header IOBX02 LFVW are correctly decoded.
2014/09/18 - Drifting buoy data with header IOBX02 KARS received at Meteo-France and ECMWF have no longer been valid since the 10th of September. Data sent with this header represent 75% of the whole drifting buoy observations. Investigations are ongoing at CLS to fix the problem.