A Datacenter Backstage: The Knowledge that Supports the Brazilian Seismic Network
Abstract:Historically, Brazilian seismology never had a clear strategic vision about how its data should be acquired, evaluated, stored and shared. Without a data management plan, data (for any practical purpose) could be lost, resulting in a non-uniform coverage that will reduce any chance of local and international collaboration, i.e., data will never become scientific knowledge.
Since 2009, huge efforts from four different institutions are establishing the new permanent Brazilian Seismographic Network (RSBR), mainly with resources from PETROBRAS, the Brazilian Government oil company.
Four FDSN sub-networks currently compose RSBR, with a total of 80 permanent stations. BL and BR codes (from BRASIS subnet) with 47 stations maintained by University of Sao Paulo (USP) and University of Brasilia (UnB) respectively; NB code (RSISNE subnet), with 16 stations deployed by University of Rio Grande do Norte (UFRN); and ON code (RSIS subnet), with 18 stations operated by the National Observatory (ON) in Rio de Janeiro.
Most stations transmit data in real-time via satellite or cell-phone links. Each node acquires its own stations locally, and data is real-time shared using SeedLink. Archived data is distributed via ArcLink and/or FDSNWS services. All nodes use the SeisComP3 system for real-time processing and as a levering back-end.
Open-source solutions like Seiscomp3 require some homemade tools to be developed, to help solve the most common daily problems of a data management center: local magnitude into the real-time earthquake processor, website plugins, regional earthquake catalog, contribution with ISC catalog, quality-control tools, data request tools, etc.
The main data products and community activities include: kml files, data availability plots, request charts, summer school courses, an Open Lab Day and news interviews.
Finally, a good effort was made to establish BRASIS sub-network and the whole RSBR as a unified project, that serves as a communication channel between individuals operating local networks.