SMC-format Data Files


Table of Contents

  1.0   Introduction
  2.0   Sample File
  3.0   Headers
    3.1  Text Headers
    3.2  Integer Headers
    3.3  Real Headers
    3.4  Comments
    3.5  SMC-format Headers Compared to BBF-format Headers
  4.0   Time-Series Files
  5.0   Response Spectra Files
    5.1  Description
    5.2  Sample File
  6.0   Differences between SMC format version 2 and version 1
  7.0   File naming conventions
    7.1  BAP output file names
    7.2  DDS#7 file names
  8.0   BAP Software
  9.0   Update History
10.0   Technical Support


1.0 Introduction

The strong-motion time series data served out by the NSMP are in one of two formats, SMC or BBF.  This documentation describes the SMC-format and compares it to the BBF-format headers.

The SMC format uses ASCII character codes and provides text headers, integer headers, real headers, and comments followed by either digitized time-series coordinates or, in version 2 of the SMC format only, sets of response spectra values.  The header information is designed to provide the user with information about the earthquake and the recording instrument.  Each SMC-format file contains either a single time series or a set of response spectra or Fourier amplitude spectra of corrected acceleration calculated from a single time series.  Typical analog strong-motion records contain traces corresponding to three orthogonal components of motion and consequently require three separate SMC-format files to represent the record.

The BBF-format uses fixed-sized blocks of binary data.  The first several blocks in each file are "header" blocks that contain auxiliary information about the time-series samples contained in the remaining "data" blocks. Each file has a minimum of two header blocks, the first of which contains 256 integer values and the second of which contains 128 real values. The time-series samples stored in the data blocks may be in either integer or real format.  A text version of the BBF-format documentation can be obtained here (bbffmt.txt).

A text version of this documentation can be obtained here (smcfmt.txt).  

This documentation was last updated on August 31, 2001.


2.0 Sample File

Here follows an example unevenly-sampled, uncorrected SMC-format time-series file from which most of the time series values have been removed. Note that the data show below may not be displayed with the exact same column spacing used in actual output files (view the ascii text version of smcfmt.txt for a correctly spaced example). 

1 UNCORRECTED ACCELEROGRAM
*
DVD0A
     1989  10  18    0004 Loma Prieta, CA Earthquake
Moment Mag=      6.9 Ms=      7.1 Ml=      7.0
station = Del Valle Dam Toe              component= 65
epicentral dist= 66.0  pk acc = 54.8
inst type=SMA-1      data source = USGS/CDWR
*
*
*
-32768 1989 291 0 4 15 -32768 451
-32768 -32768 -32768 -32768 90 65 101 9
41320 -32768 -32768 -32768 -32768 -32768 -32768 -32768
-32768 -32768 -32768 1 -32768 1265 -32768 -32768
-32768 -32768 -32768 -32768 -32768 -32768 -32768 -32768
-32768 -32768 -32768 -32768 -32768 -32768 -32768 -32768
0.1700000E+39 0.1700000E+39 0.3703700E+02 -0.1218830E+03 0.1800000E+02
0.6940000E+01 0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39
0.3761500E+02 -0.1217450E+03 0.1700000E+39 0.1700000E+39 0.1700000E+39
0.1700000E+39 0.6600000E+02 0.1700000E+39 0.1000000E+05 0.1700000E+39
0.1700000E+39 0.2540000E+02 0.6000000E+00 0.2000000E+01 0.1700000E+39
0.1700000E+39 0.1700000E+39 0.1700000E+39 0.9798830E+01 0.5482899E+02
0.9609058E+01 -0.4761129E+02 0.1700000E+39 0.1700000E+39 0.1700000E+39
0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39
0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39
0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39
|
| Put comments here.
| Explain about any peculiarities in the record, any difficulties in
| digitizing, etc.
| If no comments are necessary, place a single line with "|" in the
| first column this section.
| Set integer header cell # 16 = number of comment lines = 9 here.
| <pltlbl=> Sample plot label here <end>
|
0.0000E+0 7.3789E-2 2.4847E-3 9.0424E-1 5.9633E-3 1.3411E+0 7.3548E-3 1.4865E+0
8.8456E-3 1.6317E+0 1.0336E-2 1.8260E+0 1.1728E-2 2.1184E+0 1.3219E-2 2.5088E+0
1.4610E-2 2.5071E+0 1.6101E-2 2.8484E+0 1.7492E-2 2.6996E+0 1.8983E-2 3.0409E+0
2.0375E-2 3.2843E+0 2.1866E-2 3.2334E+0 2.3257E-2 3.3788E+0 2.4748E-2 3.2298E+0
2.6139E-2 3.5223E+0 2.7630E-2 3.4714E+0 2.9121E-2 3.8127E+0 3.0512E-2 3.6148E+0
3.2003E-2 3.7110E+0 3.3395E-2 3.6112E+0 3.4885E-2 3.6093E+0 3.6277E-2 3.3399E+0

... most of the time-series lines were removed here ...

2.9598E+1 1.9508E+0 2.9599E+1 1.6998E+0 2.9600E+1 2.1248E+0 2.9602E+1 1.9473E+0
2.9603E+1 2.1673E+0 2.9605E+1 2.2120E+0 2.9606E+1 2.3576E+0 2.9608E+1 2.5521E+0
2.9609E+1 2.5016E+0 2.9610E+1 2.4020E+0 2.9612E+1 2.3024E+0 2.9613E+1 2.5459E+0
2.9615E+1 2.4463E+0 2.9616E+1 2.4938E+0 2.9618E+1 2.3452E+0 2.9619E+1 2.6869E+0

Back to Table of Contents


3.0 Headers

There are four types of headers in SMC format files: text headers, integer headers, real headers, and comments. The 11 text header lines provide a little prose designed to allow a user to identify quickly the earthquake and recording station. Then follow 6 lines of integer header values and 10 lines of real header values. Following the text, integer, and real headers are a variable number of comment lines that can contain additional information about the earthquake or the record. Then follow either digitized time-series values or several sets of response spectra values. The number of comment lines included in the file is indicated in the 16th integer header value and the number of time series values is indicated in the 17th integer header value. Both the integer and real headers have many cells that contain a value that indicates "undefined"; these cells are either filled in subsequent stages of data processing or are merely included in the file format so that more information can be added to future files without changing the format. The value used to indicate "undefined" in integer header cells is usually  "-32768"; the value used in real header cells is usually "+1.7e+38". The Fortran format for each integer header line is "(8i10)", for each real header line is "(5e15.7)", and for each time-series line is "8(1pe10.4e1)". There are 6 integer header lines with 8 values per line, for a total of 48 integer values. There are 10 real header lines with 5 values per line for a total of 50 real header values.  There are 11 text header lines and there are as many comment lines as are indicated in integer header cell #16.

Back to Table of Contents


3.1 Text Headers (fixed format)

There are 11 text-header lines in a SMC-format time-series data file.

line 1: Contains a number in column 1 followed by text in columns 3-53, both of which indicate which type of data is contained in the file. The contents of the first line can be one of the following:

"0 UNKNOWN" (BAP will treat this like UNCORRECTED ACC)
"1 UNCORRECTED ACCELEROGRAM" (Also referred to as Volume 1 data)
"2 CORRECTED ACCELEROGRAM" (
Volume 2)
"3 VELOCITY"
(Volume 3)
"4 DISPLACEMENT" (
Volume 4)
"5 RESPONSE SPECTRA"
(Volume 5)
"6 FOURIER AMPLITUDE SPECTRA OF CORRECTED ACCELERATION" (Volume 6)

line 2: "*" in column one

line 3: Either a "*" in column one, or columns 1-4 = Station code
    column 5 = "A", "B", "C", ... to indicate which of the several accelerometer traces on the record is digitized in this file. "A" is the topmost accelerometer trace, "B" is the second from the top, etc. Note: in files that come from the DDS#7 CD-ROM, columns 1-5 of line 3 should correspond with the 5-character station-component code used in the file name. See Section 7.2

line 4: columns 1-3 = time zone if the time zone is other than GMT. This field will be blank in all the files on the CDROM(s). But it may indicate a time zone other than GMT in some files created by the V2S reformatting program (which doesn't know how to convert from all the various time zones to GMT).
    columns 6-9 = earthquake year (four digits)
    columns 12-13 = earthquake month
    columns 16-17 = earthquake day
    columns 22-25 = earthquake hour, minute in GMT
    columns 27-80 = earthquake name (left justified)

line 5: columns 1-11 = "
Moment Mag="
    columns 12-20 = moment-magnitude value
    columns 22-24 = "
Ms="
    columns 25-33 = Surface-wave magnitude value
    columns 35-37 = "
Ml="
    columns 38-46 = Local magnitude value

line 6: columns 1-10 = "
station = "
    columns 11-40 = station name, left justified (Note: Some files incorrectly have the station number instead of name in this field).
    columns 42-51 = "
component=" (Note: Some files have the "c" was in column 41 instead of column 42!
    columns 53-57 =  orientation of the recording transducer (Note: There seems to be no convention for where the value is placed in these columns.

line 7
: columns 1-17 = "
epicentral dist ="
    columns 18-26 = epicentral distance in kilometers
    columns 34-41 = "
pk acc ="
    columns 42-50 = peak value of the time series given in this file

line 8: columns 1-10 = "
inst type="
    columns 11-15 = "
SMA-1" or "CRA-1" or whatever to indicate the type of recording instrument.
    columns 22-34 = "
data source ="
    columns 36-80 = an abbreviation indicating the agency that provided the data. "
USGS", "CSMIP", etc.

lines 9, 10, and 11: a "*" in column one.

Back to Table of Contents


3.2 Integer Headers (48 values, 6 lines in "(8i10)" format)

The values given in integer header cells have the following meanings:

cell #

description
1 Value representing "undefined" or "null" in the integer header.
This is usually (always?) set to "-32768".
2 year (4 digits) GMT of first recorded sample (Note: In BAP processing the time of the first sample specified in Integer Header values 2-7 will always correspond to the time of the first sample in the original time series, even if leading pads are added. See descriptors for Integer Header values 31 and 32 below.)
3 day of year (julian day)
4 hour
5 minute
6 second
7 millisecond
8 Recorder serial number
9 Channel number of the digitized trace given in this file.
Channel numbers are counted sequentially, top-to-bottom,
with the top-most accelerometer trace counted as #1.
10 Total number of channels (= accelerometer traces) on this record.
11 Total number of channels recorded at this station
12 Sensor serial number
13 Vertical orientation (degrees from up)
0 < integer cell #13 < 180, usually 0 (UP = 0, DOWN = 180).
Null value should be used when sensor is a horizontal component.
14 Horizontal orientation (degrees east of north)
0 < integer cell #14 < 360. Null value should be used when sensor a is vertical component.
15 Sensor type code:
2 Sprengnether SA-3000 3-component fba
30 Kinemetrics FBA-13 3-component fba
31 Kinemetrics FBA-11 1-component fba
101 SMA-1 125 Kinemetrics FBA-23
102 C&GS Standard 126 Kinemetrics Episensor
103 AR-240 127 Kinemetrics FBA-4g
104 RFT-250 128 Kinemetrics FBA-2g
105 RFT-350 129 Kinemetrics FBA-1g
106 MO-2 130 Kinemetrics FBA-0.5g
107 RMT-280 131 Kinemetrics FBA-0.25g
108 SMA-2/3 132 Kinemetrics FBA-0.1g
109 DSA-1/DSA-3 133 WR1
110 DCA-300 134 S6000
111 DCA-333 135 Mark Products L22
112 A-700 136 Products L4C
113 SSA-1 137 CMG3
114 CRA-1 138 CMG3T
115 MO-2 139 CMG40T
116 FBA-3 140 CMG5
117 SMA-2 141 KS-2000
118 DCA-310 900 custom instrument
119 FBA-13 1302 Reftek Model 130-ANSS/02
120 SSA-2
121 SSR-1
122 BIDRA
123 CR-1
124 PDR-1
16 Number of comment lines following the header information.
17 Number of time-series values. Note that the number of "values"  specified
here should be the number of time-series samples if the time series is
evenly sampled, twice the number of samples if the time series is unevenly
sampled (because two values, abscissa and ordinate, are given for each
sample when the time series is unevenly sampled).
18 Problem flag - This is usually set to "undefined", but if there is a problem with
the record, such as spikes or unusual noise, it should be set to 1, in which
case additional information should be available in the comments section.
19 Type of structure: 1 = building, 2 = bridge, 3 = dam,
4 = other, null = not a structure.
20 Structure number corresponding to printed information.
21 Transducer number of the recording system as shown on the drawing.
22 Total number of transducer channels for this structure.

The following integer parameters are for buildings (when cell 19 = 1):

23 Number of floors above grade, including the roof.
24 Number of stories below grade.
25 Floor on which this sensor is located:
-1 = 1st subbasement, etc.
0 = basement
1 = grade level
2+ = floor levels above grade

The following are for bridges (when cell 19 = 2):

26 Number of spans
27 Location of transducer:
0 = free field
1 = at the base of a pier or abutment
2 = on an abutment, 3 = on the deck at the top of a pier
4 = on the deck between piers or between an abutment and a pier.

The following are for dams (when cell 19 = 3):

28 Location of this transducer:
0 = upstream or downstream free field
1 = at the base of the dam,
2 = on the crest of the dam
3 = on the abutment of the dam
29 Type of construction
1 = Reinforced concrete gravity
2 = Reinforced concrete arch
3 = earth fill
4 = other
30 Station number (This will be "undefined" in DDS#7 files if the station "number"
includes alphabetic characters, in which case it will be indicated in the station
name field in text header line 6.)
31 First recorded sample. This will be "undefined" or 1 if the first recorded sample
is the first sample given in the time series. (This info is required because
the BAP software adds leading and trailing zero pads to the time series
before filtering. BAP output files may include these leading and trailing pad
areas,  which after filtering will contain small filter transients. Integer header cells
#31 and 32 indicate where the actual recorded time series begins and ends.)
32 Last recorded sample.
33 Refer to another file flag. If this not "undefined" it indicates that users should refer to
the file indicated in the second line of the comments for more information
about the data in the current file.   Plotting programs like BAP and TSPLOT
should display that second comment line along with plots of this data.
34-39 Undefined
40-41 Reserved for user
42-47 Undefined
48 Clock source (Kinemetric Altus)
0 = RTC (internal clock)
1 = Keyboard
2 = Sync with external reference pulse
3 = GPS (= -3 if a correction for GPS rollover beginning Aug. 23, 1999, was made to original file time.

Back to Table of Contents


3.3 Real Headers (50 values, 10 lines in "(5e15.7)" format)

The values given in real header cells have the following meanings:

cell # number description
1 Value representing "undefined" or "null" in the real header. This is
usually set to "+1.7e+38". The use of the "-0.3e-38" value found
in some older files on the USGS VAXes is not recommended because
that number is out of range and not acceptable to the floating-
point representation used on PCs and most UNIX computers.
2 Sampling rate (samples per second) for evenly-sampled time series;
"undefined" for unevenly-sampled time series.
3 Earthquake latitude (decimal degrees; "+" for Northern Hemisphere)
4 Earthquake longitude (decimal degrees; "-" for Western Hemisphere)
5 Earthquake depth (km relative to sea level)
6 Source magnitude M (moment magnitude)
7 " " MS (surface-wave magnitude)
8 " " ML (local magnitude)
9 " " M (other)
10 Seismic moment (dyne-cm; 1 newton-meter = 10**7 dyne-cm)
11 Station latitude (decimal degrees; "+" for Northern Hemisphere)
12 Station longitude (decimal degrees; "-" for Western Hemisphere) Some files incorrectly have positive value here.
13 Station elevation (m relative to sea level)
14 Station offset N (m)
15 Station offset E (m)
16 Station offset up (m)
17 Epicentral distance (km)
18 Epicenter to station azimuth (degrees east of north)
0.0 < real header cell #18 < 360.0
19 Digitization units (units/cm) for analog-recorded data; Digitizing constant
(counts/volt) for digitally-recorded data.
20 Undefined for analog-recorded data; anti-alias filter corner frequency (Hz)
for digitally-recorded data.
21 Undefined for analog-recorded data; anti-alias filter poles (6dB/octave
per pole) for digitally-recorded data.
22 Natural frequency of the sensor (Hz)
23 Sensor damping coefficient (fraction of critical)
24 Recorder sensitivity (cm/g) for analog-recorded data; Coil constant
(volts/cm/sec/sec) for digitally-recorded data.
25 Undefined for analog-recorded data; amplifier gain (dB) for digitally-recorded data.
26 Undefined for analog-recorded data; pre-amp gain (dB) for digitally-recorded data.
27 Undefined in version 2. (Had been a suggested corner frequency for high-pass
filter in version 1 SMC files.)
28 Undefined in version 2. (Had been a suggested corner frequency for low-pass
filter in version 1 SMC files.)
29 Time of the time-series maximum (largest positive) value (seconds)
30 Value of the maximum value (cm/sec/sec, cm/sec, or cm)
31 Time of the time-series minimum (largest negative) value (seconds)
32 Value of the minimum value (cm/sec/sec, cm/sec, or cm)
33-36 indicate transitions for low-cut and high-cut filters.
For uncorrected (volume 1) time series provided by CSMIP, these four numbers
indicate *suggested* transition bands to be used in the correction process.
For corrected (volume 2) time series, these four numbers indicate the *actual*
low-cut and high-cut filter transitions that were used in the correction process.
The contents of cells 33 and 34 indicate the beginning and end of a transition band,
in Hz, for a low-cut filter when the contents of 34 is positive. When the contents
of 34 is negative, however, its negated value indicates the roll-off parameter that
was used in the BAP or AGRAM bi-directional Butterworth filter and the contents
of 33 is the corner frequency, in Hz, that was used in the Butterworth filter. (Note
that the roll-off "parameter" indicated for the BAP & AGRAM Butterworth filter is
equal to half of what is often called the "order" of a Butterworth filter.
The contents of 35 and 36 indicate the beginning and the end of a transition
band for a high-cut filter.
>> >> IMPORTANT NOTE: If the contents of 33-36 are "undefined" in a Vol. 2 or 3 file, the
filtering scheme will be described in the comments.
37 Undefined
38 Initial value of the filtered velocity (cm/s) after a lo-cut filter has been applied to the integrated, baseline-corrected acceleration.
39 Initial value of the filtered displacement (cm) after a lo-cut filter has been applied to the doubly-integrated, baseline-corrected acceleration.
40 Vs30 (m/s)
41-50 are undefined as yet

Back to Table of Contents


3.4 Comments

Comment lines begin, by convention, with a "|" character in colume one and may be up to 80 characters long. The comments often provide references to published documents that describe the earthquake, station, or some other aspect of the record. Comments are also used to explain any peculiarities in the record, any difficulties in digitizing, and so forth, especially when integer header cell #18 (the problem flag) is set to one. In volume 2 files, the comments may provide information about the processing the time series has received.

The Comments may include text that can be used as top-of-plot labels when the time series in the file are plotted via the BAP or TSPLOT programs. These labels are indicated in the comments by a leading "
<pltlbl=>" or "<loclbl=>"  and a trailing "<end>". (Note that the <pltlbl=>, <loclbl=> and <end> must all be in lower case.) Both types of top-of-plot label can be more than one line  long. Some sample comment lines that include top-of-plot labels:

| This time series was digitized at the USGS on 12dec97 by Sam Spade.
| <pltlbl=> blah blah blah
|
any number of lines here
|
ending with: <end>
| <loclbl=> roof, north-west corner <end>
|
More general comments here.

The
<pltlbl=> top-of-plot label is usually just one line that indicates the address or name of the recording station, if sufficient station identification would not fit on line 6 (colums 11-40). The <loclbl=> label gives information about the location of the recording transducer. It is usually quite short, and in the case of tri-axial recordings it is usually absent because all three transducers are in the same location. The <loclbl=> information is provided separately from <pltlbl=> info so the info unique to each transducer is separate from the info several transducers share. (Makes construction the plot labels more convenient for the TSPLOT program.)

When integer header cell #33 is not "undefined", the second line of the comments (the first line is usually just "|") will indicate a text file that will provide information about the data in the current file. This feature was implemented for files from the California Strong-Motion Instrumentation Program (CSMIP) and will probably not be used in files from other sources.

Integer header cell #16 indicates the number of comment lines in the file.

Back to Table of Contents


3.5 SMC-format Headers Compared to BBF-format Headers

Here follows a list of where a given value is located in SMC-format headers and BBF-format headers:

A) Integer headers:
SMC BBF Description
1 3 integer null value
2 10 year(4 digits) first sample time
3 11 julian day " " "
4 12 hour " " "
5 13 minute " " "
6 14 second " " "
7 15 millisecond " " "
8 20 recorder serial number
9 28 recorder channel number:this file
10 29 recorder channel number:total
11 30 number of channels recorded for this station
12 40 sensor serial number
13 41 vertical orientation (degrees from up)
14 42 horizontal orientation (degrees east of north)
15 43 sensor type code
B) Real headers:
SMC BBF Description
1 2 real null value
2 5 sample rate
3 10 earthquake latitude
4 12 earthquake longitude
5 14 earthquake depth
6 15 source magnitude M
10 16 seismic moment
11 40 station latitude
12 42 station longitude
13 44 station elevation
17 19 epicentral distance (km)
18 20 epicenter to station azimuth (degrees)
19 46 digitizing constant (counts/volt)
20 47 anti-alias filter corner frequency (Hz)
21 48 anti-alias filter poles (6dB/octave per pole)
22 49 natural frequency of the sensor (Hz)
23 50 sensor damping coefficient (fraction of critical)
24 51 coil constant (volts/cm/sec/sec)
25 52 amplifier gain (dB) for a DR100
26 53 pre-amp gain (dB) for a DR100

Back to Table of Contents


4.0 Time Series Files  8(1pe10.4e1)

Time-series values representing a single data trace from a strong-motion record follow the header lines in a time-series file. The Fortan format for each line is "(8(1pe10.4e1))". Eight values are given on each line, and there are as many lines as are required to provide the number of time-series values indicated in the value given in integer header cell #17. For unevenly-sampled time series (real header cell #2 is "undefined"), the values are given as a series of time-acceleration pairs; for evenly-sampled time series the values are given as a series of evenly-sampled acceleration, velocity, or displacement values, with real header cell #2 indicating the sampling rate. The number of time-series "values" given in the file and indicated in integer header cell #17 is the number of samples if the time series is evenly sampled; twice the number of samples if the time-series is unevenly sampled (because two values, time and acceleration are given for each sample when the time seriesis unevenly sampled). Time is given in seconds, acceleration in cm/sec/sec, velocity in cm/sec, and displacement in cm.

Back to Table of Contents


5.0 Response Spectra Files

5.1 Description

Lines 1-30: The first 30 lines of a response spectra file contain the header info that corresponds to the acceleration time series that was used as input to the response-spectra calculations. Header cells containing peak value, number of samples, and the like contain values appropriate for the acceleration time series.
Line 31: indicates the units used for response values. The Fortran format for this line is "3(i2,a15)".
    The numbers and units are as follows:
    1 cm     2 inches
    1 cm/sec     2 inches/sec
    1 cm/(sec squared)     2 inches/(sec squared)
Line 32: indicates the number of damping values, the number of period values, and a times-are-included-or-not flag. The Fortran format for this line is "(3i5)". The number of damping values is referred to as NDAMP below and is usually 5. The number of period values (= the number of samples in each spectrum) is referred to as NPER below and is usually 91. The whether or not (1 or 0) times of maximum response are included in the file flag is referred to as TFLAG below and is usually 0, indicating that times are not included. (The times of maximum response, lines 43-48, are not provided in BAP output files, but they are given in some of the files that will be published in DDS#7 sequel CD-ROMs.)

-------------------------------------------------
The usage "Lines" below refers to *sets* of lines
-------------------------------------------------

Lines 33: NDAMP/5 (rounded up) lines listing the NDAMP damping values in "(5e10.5)" format.
Lines 34: NPER/7 lines listing the NPER period values in "(7e11.4)" format.
--------------------
Lines 35 through 48 are repeated for each NDAMP damping value:

Line 35: A header card containing "Relative Displacement Response Spectrum at X% Damping" (where X is replaced by one of the NDAMP damping values).
Lines 36: NPER/7 lines containing relative displacement response spectral   values in "(7e11.4)" format.
Line 37: A header card containing "Relative Velocity Response Spectrum at X% Damping"
Lines 38: NPER/7 lines containing relative velocity response spectral spectral values in "(7e11.4)" format.
Line 39: A header card containing "Pseudo-Velocity Response Spectrum at X% Damping"
Lines 40: NPER/7 lines containing pseudo-velocity response spectral values in "(7e11.4)" format.
Line 41: A header card containing "Absolute Acceleration Response Spectrum at X% Damping"
Lines 42: NPER/7 lines containing absolute acceleration response spectral values in "(6e13.6)" format.
>> >> Programmers please note the different format.

Lines 43 through 48 are given only when TFLAG=1:
-------------------
Line 43: A header card containing "
Times of Maximum Response Relative Displacement at X% damping"
Lines 44: NPER/7 lines containing times of maximum response relative displacement in "(7e11.4)" format.
Line 45: A header card containing "
Times of Maximum Response Relative Velocity at X% Damping"
Lines 46: NPER/7 lines containing times of maximum response relative velocity in "(7e11.4)" format.
Line 47: A header card containing "
Times of Maximum Response Absolute Acceleration at X% Damping"
Lines 48: NPER/7 lines containing times of maximum response absolute acceleration in "(7e11.4)" format.

Back to Table of Contents


5.2 Sample Response-Spectra File

Here follows a sample BAP (version 2) response-spectra output file from which all but the first set of response spectra (the set corresponding to the first damping value) have been removed.  Note that the data show below may not be displayed with the exact same column spacing used in actual output files (view the ascii text version of smcfmt.txt for a correctly spaced example).

5 RESPONSE SPECTRA
*
DVD0A
     1989  10  18    0004 Loma Prieta, CA Earthquake
Moment Mag=      6.9 Ms=      7.1 Ml=      7.0
station = Del Valle Dam Toe              component= 65
epicentral dist= 66.0  pk acc = 54.8
inst type=SMA-1      data source = USGS/CDWR
*
*
*
-32768 1989 291 0 4 15 -32768 451
1 3 6 -32768 90 65 101 3
5914 -32768 3 -32768 -32768 -32768 -32768 -32768
-32768 -32768 -32768 1 -32768 1265 -32768 -32768
-32768 1 -32768 -32768 -32768 -32768 -32768 -32768
-32768 -32768 -32768 -32768 -32768 -32768 -32768 -32768
0.1700000E+39 0.2000000E+03 0.3703700E+02 -0.1218830E+03 0.1800000E+02
0.6940000E+01 0.7100000E+01 0.7000000E+01 0.1700000E+39 0.1700000E+39
0.3761500E+02 -0.1217450E+03 0.1700000E+39 0.1700000E+39 0.1700000E+39
0.1700000E+39 0.6600000E+02 0.1700000E+39 0.1000000E+05 0.5000000E+02
0.5000000E+02 0.2564103E+02 0.6000000E+00 0.2000000E+01 0.1700000E+39
0.1700000E+39 0.1700000E+39 0.1700000E+39 0.9790000E+01 0.5366566E+02
0.9595000E+01 -0.4893999E+02 0.1000000E+00 -0.4000000E+01 0.1700000E+39
0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39
0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39
0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39 0.1700000E+39
|
| test 1
|

1 cm           1 cm/sec           1 cm/sec/sec

5    91     0

  0.00000E+00

0.20000E-01

0.50000E-01

0.10000E+00

0.20000E+00

0.4000E-0 0.4200E-01 0.4400E-01 0.4600E-01 0.4800E-01 0.5000E-01 0.5500E-01
0.6000E-01 0.6500E-01 0.7000E-01 0.7500E-01 0.8000E-01 0.8500E-01 0.9000E-01
0.9500E-01 0.1000E+00 0.1100E+00 0.1200E+00 0.1300E+00 0.1400E+00 0.1500E+00
0.1600E+00 0.1700E+00 0.1800E+00 0.1900E+00 0.2000E+00 0.2200E+00 0.2400E+00
0.2600E+00 0.2800E+00 0.3000E+00 0.3200E+00 0.3400E+00 0.3600E+00 0.3800E+00
0.4000E+00 0.4200E+00 0.4400E+00 0.4600E+00 0.4800E+00 0.5000E+00 0.5500E+00
0.6000E+00 0.6500E+00 0.7000E+00 0.7500E+00 0.8000E+00 0.8500E+00 0.9000E+00
0.9500E+00 0.1000E+01 0.1100E+01 0.1200E+01 0.1300E+01 0.1400E+01 0.1500E+01
0.1600E+01 0.1700E+01 0.1800E+01 0.1900E+01 0.2000E+01 0.2200E+01 0.2400E+01
0.2600E+01 0.2800E+01 0.3000E+01 0.3200E+01 0.3400E+01 0.3600E+01 0.3800E+01
0.4000E+01 0.4200E+01 0.4400E+01 0.4600E+01 0.4800E+01 0.5000E+01 0.5500E+01
0.6000E+01 0.6500E+01 0.7000E+01 0.7500E+01 0.8000E+01 0.8500E+01 0.9000E+01
0.9500E+01 0.1000E+02 0.1100E+02 0.1200E+02 0.1300E+02 0.1400E+02 0.1500E+02
Relative Displacement Response Spectrum at 0.00 Damping
0.2377E-02 0.2543E-02 0.3038E-02 0.3575E-02 0.3382E-02 0.4777E-02 0.4314E-02
0.5028E-02 0.9993E-02 0.9882E-02 0.1041E-01 0.1224E-01 0.2071E-01 0.3067E-01
0.1830E-01 0.3226E-01 0.5011E-01 0.1243E+00 0.1357E+00 0.5743E-01 0.1673E+00
0.1294E+00 0.1139E+00 0.1969E+00 0.2241E+00 0.3555E+00 0.2864E+00 0.5695E+00
0.1161E+01 0.6391E+00 0.1225E+01 0.1018E+01 0.1362E+01 0.1175E+01 0.1243E+01
0.1429E+01 0.2087E+01 0.2259E+01 0.3183E+01 0.2130E+01 0.3379E+01 0.1563E+01
0.1672E+01 0.1189E+01 0.2061E+01 0.1423E+01 0.1658E+01 0.1634E+01 0.5186E+01
0.1919E+01 0.2479E+01 0.2162E+01 0.4273E+01 0.5399E+01 0.6623E+01 0.3094E+01
0.4603E+01 0.4276E+01 0.4322E+01 0.4010E+01 0.6128E+01 0.4934E+01 0.3541E+01
0.4295E+01 0.7949E+01 0.1273E+02 0.7865E+01 0.5489E+01 0.6737E+01 0.9248E+01
0.1068E+02 0.1119E+02 0.1234E+02 0.1322E+02 0.1390E+02 0.1515E+02 0.2090E+02
0.2475E+02 0.2396E+02 0.1871E+02 0.1583E+02 0.1267E+02 0.9982E+01 0.9011E+01
0.8630E+01 0.8520E+01 0.8452E+01 0.8081E+01 0.7356E+01 0.6472E+01 0.6243E+01
Relative Velocity Response Spectrum at 0.00 Damping
0.1212E+00 0.1499E+00 0.1395E+00 0.1891E+00 0.1830E+00 0.5628E+00 0.1983E+00
0.2080E+00 0.4809E+00 0.4814E+00 0.5238E+00 0.6912E+00 0.1158E+01 0.1610E+01
0.7739E+00 0.1680E+01 0.2676E+01 0.6422E+01 0.6380E+01 0.2389E+01 0.6634E+01
0.4553E+01 0.3653E+01 0.6406E+01 0.6822E+01 0.1103E+02 0.6910E+01 0.1434E+02
0.2757E+02 0.1260E+02 0.2560E+02 0.1937E+02 0.2499E+02 0.1990E+02 0.2028E+02
0.2198E+02 0.3094E+02 0.3165E+02 0.4298E+02 0.2788E+02 0.4234E+02 0.1787E+02
0.1733E+02 0.1036E+02 0.1848E+02 0.1190E+02 0.1379E+02 0.1232E+02 0.3594E+02
0.1215E+02 0.1558E+02 0.1220E+02 0.2265E+02 0.2522E+02 0.2873E+02 0.1261E+02
0.1742E+02 0.1580E+02 0.1509E+02 0.1335E+02 0.1968E+02 0.1423E+02 0.9526E+01
0.9296E+01 0.1855E+02 0.2689E+02 0.1576E+02 0.1024E+02 0.1215E+02 0.1587E+02
0.1868E+02 0.1932E+02 0.1876E+02 0.1903E+02 0.1843E+02 0.1802E+02 0.2387E+02
0.2704E+02 0.2470E+02 0.2181E+02 0.1470E+02 0.1202E+02 0.1026E+02 0.9182E+01
0.8531E+01 0.8050E+01 0.7771E+01 0.7553E+01 0.7239E+01 0.7026E+01 0.6982E+01
Pseudo-Velocity Response Spectrum at 0.00 Damping
0.3734E+00 0.3805E+00 0.4338E+00 0.4883E+00 0.4428E+00 0.6003E+00 0.4928E+00
0.5266E+00 0.9659E+00 0.8870E+00 0.8721E+00 0.9615E+00 0.1531E+01 0.2141E+01
0.1210E+01 0.2027E+01 0.2862E+01 0.6506E+01 0.6561E+01 0.2578E+01 0.7008E+01
0.5081E+01 0.4209E+01 0.6874E+01 0.7411E+01 0.1117E+02 0.8181E+01 0.1491E+02
0.2806E+02 0.1434E+02 0.2566E+02 0.1999E+02 0.2517E+02 0.2052E+02 0.2054E+02
0.2245E+02 0.3123E+02 0.3226E+02 0.4347E+02 0.2788E+02 0.4246E+02 0.1786E+02
0.1751E+02 0.1150E+02 0.1850E+02 0.1192E+02 0.1302E+02 0.1208E+02 0.3621E+02
0.1269E+02 0.1558E+02 0.1235E+02 0.2237E+02 0.2609E+02 0.2973E+02 0.1296E+02
0.1808E+02 0.1580E+02 0.1509E+02 0.1326E+02 0.1925E+02 0.1409E+02 0.9270E+01
0.1038E+02 0.1784E+02 0.2666E+02 0.1544E+02 0.1014E+02 0.1176E+02 0.1529E+02
0.1677E+02 0.1674E+02 0.1762E+02 0.1806E+02 0.1820E+02 0.1903E+02 0.2387E+02
0.2591E+02 0.2316E+02 0.1680E+02 0.1326E+02 0.9948E+01 0.7378E+01 0.6291E+01
0.5708E+01 0.5353E+01 0.4828E+01 0.4231E+01 0.3555E+01 0.2905E+01 0.2615E+01
Absolute Acceleration Response Spectrum at 0.00 Damping
0.586592E+02 0.569191E+02 0.619419E+02 0.667019E+02 0.579564E+02 0.754375E+02
0.562953E+02 0.551417E+02 0.933705E+02 0.796204E+02 0.730636E+02 0.755151E+02
0.113180E+03 0.149468E+03 0.800518E+02 0.127345E+03 0.163497E+03 0.340677E+03
0.317104E+03 0.115684E+03 0.293558E+03 0.199548E+03 0.155571E+03 0.239955E+03
0.245069E+03 0.350862E+03 0.233638E+03 0.390300E+03 0.678083E+03 0.321824E+03
0.537479E+03 0.392569E+03 0.465161E+03 0.358056E+03 0.339700E+03 0.352583E+03
0.467172E+03 0.460627E+03 0.593806E+03 0.364929E+03 0.533611E+03 0.204018E+03
0.183377E+03 0.111121E+03 0.166061E+03 0.998630E+02 0.102268E+03 0.892970E+02
0.252773E+03 0.839396E+02 0.978758E+02 0.705356E+02 0.117139E+03 0.126110E+03
0.133409E+03 0.542793E+02 0.709874E+02 0.584107E+02 0.526669E+02 0.438574E+02
0.604844E+02 0.402444E+02 0.242684E+02 0.250856E+02 0.400264E+02 0.558439E+02
0.303220E+02 0.187444E+02 0.205212E+02 0.252825E+02 0.263497E+02 0.250426E+02
0.251648E+02 0.246734E+02 0.238196E+02 0.239180E+02 0.272721E+02 0.271371E+02
0.223843E+02 0.150756E+02 0.111084E+02 0.781279E+01 0.545407E+01 0.439177E+01
0.377497E+01 0.336362E+01 0.275777E+01 0.221554E+01 0.171830E+01 0.130361E+01

... The times of maximum respose, if wanted, would be given here. The format would be similar to the last 4 sets of 91 values, only the header line for each set would have "Times of Maximum Response" inserted at the beginning.

... Then the last 4 sets of 91 values (or the last 8 sets if times are included) would be repeated for damping = 0.02, 0.05, etc.

Back to Table of Contents


6.0 Differences between SMC Format, Version 1 and Version 2

- Version 1 time series were all evenly sampled.  Version 2 uncorrected (volume 1) time series can be given either as unevenly-sampled (x,y) pairs of values or as evenly sampled y-values. The sampling rate in the real header (header cell #2) will be "undefined" when the time series is unevenly sampled.

- Version 2 allows for corrected (volume 2) time series. As a consequence, the meaning of the contents of real header cells 33-36 has been expanded. These four numbers indicate *suggested* filter parameters in files containing uncorrected (volume 1) time series, just as they did in version 1 files. But they indicate the filter parameters that were *actually* used in files containing corrected time series (volume 2) or response spectra (volume 3 data).

- Version 2 allows for files that contain response spectra (volume 3 data) in addition to time-series files. See section 5.

- The first line of version 1 files contained nothing but an asterisk (*) in the first column. The first line of version 2 files contains one of the following:

1 UNCORRECTED ACCELEROGRAM
2 CORRECTED ACCELEROGRAM
3 VELOCITY
4 DISPLACEMENT
5 RESPONSE SPECTRA
6 FOURIER AMPLITUDE SPECTRA OF CORRECTED ACCELERATION

- The third line of an SMC-format file contains the recording station code. This was true in version 1 files too, but BAP version 1 ignored it.

- Version 2 SMC files contain the station number in integer header cell #30.

- The SMC (version 2) files output from BAP (version greater than 1) can now contain the leading and trailing pads which contain filter transients. These are indicated in integer header cells #31 and #32, which contain the number of the first and last recorded sample, respectively, if padding is included in the time series and which contain "undefined" if pads are not included.

- Real header cells 27 and 28 are not used in version 2 (*** even in uncorrected time series files? ***). These cells contained suggested filter corner frequencies in version 1 files.

Back to Table of Contents


7.0 File naming conventions

7.1 BAP output file names

SMC format files produced by the BAP program have names that are assigned by the program, AAVEL.SMC and AARESPON.SMC for example. The suffix of a BAP-assigned file name is either ".SMC" or ".BBF" allow users to determine which format the file is in. The user can, of course, rename the files to anything they like.

7.2 DDS#7 file names

SMC format files on the USGS Digital Data Series DDS-7 CD-ROM, Digitized Strong-Motion Accelerograms of North and Central American Earthquakes 1933-1986,  by Linda C. Seekins, A. Gerald Brady, Catherine Carpenter, and Nicholas Brown, 1992, have names that are intended to help the user determine the contents of the files. Because file names are limited to 12 characters on PCs, we came up with the following somewhat-unwieldy convention. The files from each year are stored in a separate directory (named by year). Characters 1-3 are the day of the year (001-366); character 4 is a letter corresponding to the hour (A=0,...X=23); characters 5-6 are the minutes; characters 7-11, are the four-character station name, unfortunately split by a period in character 9; and character 12 is the component code. For example, a recording of the third component of station STNA from an earthquake occurring on July 20 (Julian Day 201), 1979 at 10:22 UTC would be in the directory 1979 in a file called 201K22ST.NAC.

The four-character station codes embedded in DDS#7 file names were taken, wherever possible, from Switzer and others (1981) or from the agency which provided the records. Generally they are only three characters long, in whichcase the fourth character is a zero. When two earthquakes occurred within the same minute, we used 1 as the fourth character. For some records, no station code was listed in the literature; in this case, we made up a four character code with X as the first character, although not all station codes beginning with X are our invention.

Back to Table of Contents


8.0 BAP Software

BAP version 1.1 software for IBM-style PCs that will correct USGS time-series data is available from the NSMP web site: /processing.html and the NSMP FTP site: ftp://ftpext.usgs.gov/pub/wr/ca/menlo.park/nsmp in the directory at "/software/bap/4pc."

The same files are also available from the USGS at the following address:

USGS Information Services
Box 25286
Denver Federal Center
Denver, Colorado, 80225-0046

Telephone: (303) 202-4200 or 1-888-ASK-USGS (1-888-275-8747)
Fax: (303) 202-4695
email: infoservices@usgs.gov
http://mapping.usgs.gov/esic/to_order.html

To order the software, request USGS Open-File Report #92-296A, titled "BAP: Basic Strong-Motion Accelerogram Processing Software; Version 1.0" for the printed user's manual and USGS Open-File Report #92-296B, titled "BAP version 1.1 software distribution diskettes" for two 1.44 MByte PC diskettes. (Note that the printed user's manual is for version 1.0, the diskettes for
version 1.1. Differences between the two versions are explained in the WHATS.NEW file given on the diskettes. The price for the user's manual is $25.25 for paper copy or $4.00 for microfiche; the price for the diskettes is $20 (prices as of April 1998). A $5.00 shipping and handling fee is applied to all orders (fee as of November 1999).

Note that BAP version 1.0 can *not* read the SMC version 2 time-series files, but BAP version 1.1 can.


9.0 Update History

Recent updates or modifications to the SMC-format are listed here.  All changes are backwards compatible with older versions of the BAP processing software.


10.0 Technical Support

Questions regarding the SMC- or BBF-formats should be directed to:

Kent Fogleman:   (650) 329-4745, fogleman@usgs.gov
         or
Chris Stephens:  (650) 329-4752, cdstephens@usgs.gov

U.S. Geological Survey, MS 977
345 Middlefield Road
Menlo Park, CA 94025


Back to Table of Contents

Back to NSMP Data Sets Page.
Back to NSMP Home Page.