Errata and Updates

The station scientists are responsible for the quality of their submitted data. Within the archive all incoming data are format-checked and visualized. Files with obvious errors are not imported into the archive.

Nevertheless, errors within the archive cannot be fully excluded. If errors have been detected the erroneous data are replaced by corrected datasets. If corrected data cannot be submitted within a reasonable period of time, the questionable data will be deleted in PANGAEA, and a warning will be given in the corresponding ftp files.

The version of the data is given in the first line of the station-to-archive files. New versions within PANGAEA obtain new doi-numbers. The old doi-numbers will not be deleted but linked to the new version. A comment in the PANGAEA meta-data displays the version number and the date of the change if more than one version has been submitted.

This page is written to inform the customers about detected errors and updated versions.

Ny-Ålesund (NYA) LR1100

For station NYA the LR1100 record ends in 2017. However, high resolution radiosonde measurements from 2017 on can be found here.

2020-01-01 (ongoing problem): PTR,RLM,BRB,SMS

Currently the files from PTR,RLM,BRB,SMS until 2008-08 still contain wrong long-wave fluxes. The same error exists in the long-wave data from FLO after August 1999. The wrong long-wave fluxes have been deleted in PANGAEA. It might be, that the data declared LWD are actually LWU fluxes.
The corresponding station-to-archive files are no longer readable for BSRN-users. In case you need a copy nevertheless, please contact: Amelie Driemel. The files will be replaced as soon as updates are available.

In BRB there seems to have been a tracker problem in 2014: the data from April until August as well as some from October show too low SumSW values compared with SWD at around noon and too high values in the afternoon, see also

2019-08-21: BSRN Toolbox errors in "Expanded measurements.." LR1300

In August 2019 it was noted, that the Toolbox code was giving wrong values in the extraction of LR1300 ( If the station-to-archive file contains a value of 10000 in the parameter cloud base height then the Toolbox gives out a value of 99999 which means "no cloud". This issue has to be looked into, please be careful when using the data.

2017-10-01: Barrow files fixed

Long ago, Ellsworth G. Dutton discovered a calibration glitch in the SWD ("Short-wave downward (GLOBAL) radiation") data for Barrow, Alaska. In a huge effort, the calibration values were finally corrected and the time series for Barrow (1995-2009) was resubmitted, so that Barrow data can now again be accessed on the ftp folder and in PANGAEA. On top, the data of 2010-2017 was also submitted and Barrow is again up-to-date! Thanks a lot to Chuck Long and Dave Longenecker!

2017-05-17: PAY LR3030

Between 1996 and 2004 the SWD data in logical record LR3030 seems to be wrong (values too low). The station scientist is informed, he already confirmed that the data seems to have been mixed up. The files will hopefully be replaced soon.

2016-10-25: Tiksi

The station scientist informed us that at station Tiksi, wrong calibration coefficients have probably been used to obtain the data submitted to BSRN. This will be looked into soon, and data resulting from wrong calibration coefficients will be replaced.

2015-05-06: Cabauw

On 3 November 2011 there has been a change of instruments used for the measurement of the downward longwave irradiance (DL in LR0100). This change of instruments was accompanied by a calibration change, removing a bias of +3.85 ± 1.48 W/m2 that exists in DL before 3 November 2011. After 3 November 2011, the values of DL are traceable to the WISG. We plan to correct the data (remove the bias) before 3 November 2011 and resubmit the station-to-archive files (February 2005 – November 2011). 

For further information please contact Opens window for sending emailWouter Knap.

2013-06-03: Ilorin

Some additional quality control is needed for the LW data collected at Ilorin. There are issues related to the way the measurements were taken (shaded, not shaded) and it has to be accounted for.