As most of you are aware, a problem with the IRIS/IDA Mark 7 data logger was discovered earlier this year.  A very comprehensive report can be found at

            http://ida.ucsd.edu/WhatsNew/

and I encourage you to read it for more specific details of the problem.

As a quick summary of the issue, an error with the Digital Signal Processor implementation introduced a problem that can result in increased noise for several of the data streams from the MK 7 Data Logger.  Data, originally digitized at 100 samples per second, is processed by the DSP to derive the lower sample rate channels such as the BH and VH channels.  Under specific conditions, each pass through the DSP introduces more error resulting in greater noise.  The bad news is that the original 100 sample per second stream is not recorded; the highest sample rate stream that is retained is the 20 sample per second BH stream.  Therefore the problem can not be completely corrected by reprocessing.  Presently the error in the BH stream has a standard deviation of about 25 counts.  By reprocessing the data, the error in the BH (and all other streams) can be reduced to 1 to 2 counts. The good news is that the BH, LH and VH streams can be improved by processing.

The result of this is that approximately 4 network years of IRIS/IDA data need to be re-archived at the DMC.  At the present time a significant amount of these reprocessed data are already at the DMC. These data are available using any of the normal DMC request tools. This short note is to identify some special considerations when requesting these data.

Originally IRIS/IDA data (Network Code II) did not make use of the Location Identifier, and it was set to Space - Space.  As part of the Mark 7 Data repair project, all II data will be re-archived with new location identifiers.  This affects all IDA data, not just those from the Mark 7 data loggers whose data needed repairing.

The following table indicates how the Location Identifier will be used.

In other words, data susceptible to this problem had their LI changed to XX, and those not susceptible were changed to 00 or 10 as appropriate.  All repaired data were given code 00.  II data with LI space-space will no longer be distributed by the DMC.

The DMC has activated a filter that normally prevents any XX data from being sent to end users.  For users that would like to receive the original, un-reprocessed MK7 data the DMC can de-activate the filter to pass data with location identifier XX.  If you have such interest contact Rick Benson at the IRIS DMC () and he can arrange to have the original XX data sent to you.

[See original DMS Electronic Newsletter article about how Location Identifiers are used.]

Submitted by Tim Ahern, IRIS DMC