CSEP 2 CATALOG FORMAT
The general philosophy is the file format should be human readable and easy to use for both researchers and developers. (The actual storage of these catalogs in CSEP testing centers might be different than this proposed format).
The catalog will be defined as a (ASCII/utf-8) text-file in csv format. Each row corresponds to a single event. This file will have the following headers:
longitude, latitude, M, time_string format="%Y-%m-%dT%H:%M:%S.%f", depth, catalog_id, [event_id]
longitude: longitude in decimal degrees
latitude: latitude in decimal degrees
time string (UTC): year-month-day hour-minute-second.fraction_second. strptime format="%Y-%m-%dT%H:%M:%S.%f": example: 1985-01-01T00:00:00.0
depth: hypocenter depth in (km)
catalog_id: indicates type of catalog
event_id [optional] column to indicate specific flags for a given event
- observed: -1
- simulated: [0, n_cat-1]
This format can be easily extended to additional file types, including more advanced storage including binary, HDF5, or database representations. Modelers can chose an arbitrary number of catalogs to store within a file. Events will be mapped to catalogs through the catalog_id field, which would prevent the build-up of many, but potentially empty, catalog files.
The catalog at the above link contains 829 events. This catalog requires 52kb of storage. The table below shows expected sizes of these catalogs for individual catalogs (middle) and forecast (right) assuming 100000 catalogs per forecast. Important: If event_id is not provided by the modeler there needs to be a ',' in the final column. See example below.
|event count||catalog size (Mb)||forecast size (Gb)|