View source for PEER Data Format

Observed data from PEER will be received in a PEER NGA data format. Here are some details on the format.

PEER is a large provider of fully-processed data (e.g. for engineering applications), and this is the format they chose. In that sense, it is a very "common" format, although it is limited to PEER, and it is well defined. In the context of this project, all the currently available data are in that format (70% of the total number of records is already in that format).

== Format Details: ==

The PEER headers for acceleration files (*.AT2) are always 4 lines. The 4th line format has changed over the years
#PEER STRONG MOTION DATABASE RECORD. PROCESSING BY (PACIFIC ENGINEERING or other data provider)
# Earthquake name, date, and time; station location; record component; data source and station number (if available)
# Acceleration, velocity, or displacement and units. Filter points (optional).
# NPTS (number of points) and DT (time interval between them)

Each file contains one time series (single component as you noticed). The data is to be read horizontally, line by line, and is organized in 5 columns. The data is fully processed for engineering applications so the user has nothing to do but to read it. All the components for a given event and station are aligned in time (no need for a start time) and have the same number of points and duration.

== File names/structure ==
PEER will provide you with the files organized in folders. We can discuss if a file-naming convention is needed or not or if a list of path/filename may be read instead (more flexible).