IRIS Knowledge Base :: How do I interpret the SAC ASCII data files?

Knowledgebase Home

Search the Knowledgebase

Browse by Category

How do I interpret the SAC ASCII data files?

Q: I have a few questions regarding the Mexicali Event files that I downloaded:

1. I attached a image showing all the files I received from the Seismogram tab -- I understand "BK.JRSC." refers to the server and the station name, but can you confirm:
"BQ2, BQ3, etc." refer to the various channels?
What do the various time stamps refer to? Event and after shocks?

2. I requested data from 4/4/10 15:00 to 4/4/10 16:00 (PDT) and opened up the following file "BK.JRSC..BHE__at__2010-04-04T22.37.32.448Z.SAC_ASCII" (attached), which contains 7920 rows (below the metadata, the first row contains "-1811 -1823 -1840 -1862 -1862"). I understand the BHE channel records 40 samples per second, which suggests that this file only contains data for 198 seconds (7920/40). Can you explain why I didn't get an hour of data? Also, what do the various values, such as -1811 -1823 -1840 -1862 -1862, represent?

A: BQ2 and BQ3 are indeed channel names. Since you requested B*, you're getting a number of different broadband channels.

   According to MDA, these are electric field sensors:
http://www.iris.edu/mda/BK/JRSC/10/BQ2?timewindow=2004/12/29%2000:00:00-2599/12/31%2023:59:59

   The time stamps on each file is the start time of the first sample in the SAC ASCII file. Sometimes you will see multiple times for the same channel because there was a discontinuity in the data. SAC data is treated as continuous, so when there is a break, a new file must be created to mark the new starting point.

   The 7920 rows of data information are expressed as 32-bit signed integer, scaled down by the SCALE factor included in the header of the SAC ASCII file.

   Each row of data contains 5 data points (five column data), so you would multiply your number of rows by 5 to get the approximate number of samples: 990 seconds or 16.5 minutes, which would put you at minute 54, approximately. The other BHE files might represent some overlaps, then, since their start times are earlier than this.