Proposed Quanterra Datalog File Format By: Douglas Neuhauser UC Berkeley Seismographic Station doug@seismo.berkeley.edu 510-642-0931 Rev: 0 Date: 1995/02/15 Comserv output: The Comserv process provided data and related information from Quanterras dataloggers to clients in 512-byte SEED packets. Comserv currently provide 5 distinct types of packets: 1. Data Packets: 512-byte MiniSEED data packets consisting of: 48-byte Fixed Data Header Blockette 1000 (data format and blocksize description) Possible additional blockettes related to the data, such as Blockette 1001 (enhanced timing blockette). Data values, in Steim1, Steim2, ... frames. 2. Detection Packets: 512-byte MiniSEED data packets consisting of: 48-byte Fixed Data Header Blockette 1000 (data format and blocksize description) Event detection blockette 2xx. 3. Calibration Packets: 512-byte MiniSEED data packets consistin of: 48-byte Fixed Data Header Blockette 1000 (data format and blocksize description) Calibration blockette 3xx. 4. Timing Packets: 512-byte MiniSEED data packets consisting of: 48-byte Fixed Data Header Blockette 1000 (data format and blocksize description) Timing packet 500. 5. LOG Packets: 512-byte MiniSEED data packets consisting of: 48-byte Fixed Data Header (data rate = 0, channel = LOG) Blockette 1000 (data format and blocksize description) ASCII console log messages. The datalog client is designed to log information from an individual Quanterra station to disk files. Datalog treats each of the type of packets as separate streams of information, and logs them into separate files and directories. Each stream of information is logged in a SEED Telemetry File with 512-byte blocksize, which has the following format: SEED Telemetry Volume Header 8 (512-bytes) 0 ... N 512-byte packets of a consistent type (as shown above). This imples that each Quanterra data channel (eg BHZ) may have the following distinct streams of information, and thus distinct directories and files: BHZ.D BHZ data records. BHZ.E BHZ event detections. BHZ.C BHZ calibration info. The station as a whole will have the following channels: ACE.T Station timing info. LOG.L Log information Datalog requires a unique data directory for each station, and within that station data directory it creates a unique directory for each stream of information. The directories are named: channel.type where channel SEED channel name type type of info: D = data, E = event, C = calib, T = timing, L = log The type string is configurable by the [dlog] config section for the station. The current file for each channel.type information stream is named "active".