This DVD contains data recorded by the USGS Parkfield Dense Array (UPSAR) for the M6.5 San Simeon, California earthquake of 22 December 2003 and the M6.0 Parkfield, California earthquake of 28 September 2004. Also included are 1 month of data following the San Simeon earthquake and 2 months of data following the Parkfield earthquake. These data are preliminary, and are subject to change. Please contact the principal investigators, Dr. Jon Fletcher (jfletcher@usgs.gov), Dr. Paul Spudich (spudich@usgs.gov), or Mr. Larry Baker (baker@usgs.gov), for the most up-to-date data. The UPSAR array is described in the paper: The USGS Parkfield, California, Dense Seismograph Array: UPSAR, Jon B. Fletcher, et al., Bulletin of the Seismological Society of America, Vol. 82, No. 2, pp. 1041-1070, April 1992. Copies are available from any of the principal investigators upon request. The UPSAR data on this DVD are organized in an inverted tree structure: upsar | +----- Year (YYYY) | +----- Julian day (DDD) | | | +----- Time of event (HHMMSS) | | | | | +----- Quick-and-dirty (fixed scale) PDF plot | | +----- Playback detail (.lis) and summary (.sum) files | | +----- VFBB-format data files : : For each triggered recording, a separate directory exists named upsar/YYYY/DDD/ HHMMSS. In addition to the VFBB-format data files, the playabck program creates 3 files: 1. A quick-and-dirty PDF plot of the first 10 seconds after the trigger of the vertical velocity channels using a fixed scale. Clipping of this plot does not mean the velocity data is clipped. 2. A .lis file which details the Parkfield Dense Array status at the time of the trigger, the station and instrument parameters used for playback, some statistics of the raw (counts) data, a report of any timing errors or missing data, and the names of the VFBB-format files created. 3. A .sum file which is a very brief summary the information in the .lis file. BE SURE TO READ THE .lis OR .SUM FILES TO VERIFY THE DATA IS OK. Each VFBB-format data file is named as DDDHHMMSC.NNN, where DDD is the Julian day, HH is the hour, MM is the minute, the single character S is an abbreviation for the second, using the alphabet from A through T to represent the 3-second intervals from 00..02, 03..05, etc. C is the component (channel) number (1..6): 1..3 are acceleration, 4..6 are velocity. NNN is the 3-character station name, which are simply P01, P02, etc. VFBB-format files have a 512 byte 2-byte- integer header, followed by a 512 byte 4-byte-floating-point header, followed by 512-byte data blocks (2-byte or 4-byte integer data, or 4-byte floating point data). The headers also have some ASCII data mixed in, as Fortran-style Hollerith data. The data are in Digital VAX-format. Warning: the name of the VFBB data files is a convention only. Do not rely on the name, for example, for the time of the first data sample; you must use the values in the headers. Also, because the instruments in the Parkfield Dense Array independently assemble their data into 200ms blocks, there is no guarantee the data from one station is time-aligned with the data from another station, except that they will be in the same 200ms interval. Again, you must use the values in the headers. There are utility routines and MatLab functions for reading the data and headers in the matlab directory. The 2 MatLab M-files there, load_bbdata.m and merge_bbdata.m, should make it easy to cut and paste time slices from the VFBB data files for the events of interest. Put them into your matlab/toolbox/local directory. These 2 M-files use the MatLab MEX-file functions bbihdr(), bbrhdr(), and bbdata(), which are in the matlab/vfbb directory. Copy the matlab/vfbb/src/*.m files (for the MatLab "help" command) to your matlab/ toolbox/local directory, along with the binary MEX files for your platform (e.g., the *.dll files under win32 for Windows PCs, or the *.mexmac files under macosx for Macintosh OS X systems). There is also a command-line tool called dhead in the system-specific directory which displays the VFBB file header values. Larry Baker US Geological Survey baker@usgs.gov 18 February 2005