Searchable Product Archive and Discovery Engine Now Available
After two years of development and coordination with data providers, IRIS DMC and the USArray Program are pleased to announce the release of the Searchable Product Archive and Discovery Engine (SPADE).
The goal of SPADE is to provide a coherent system to manage the submission and search for value-added data products provided by researchers and data centers participating in the USArray program and other cooperating scientific programs. By coupling raw data, images, documents, and metadata together, SPADE facilitates data discovery across a broad range of scientific disciplines and specific areas of study, many of which are possible because of EarthScope-funded geophysical data gathering programs like USArray.
A data product is, at one level, encapsulated information derived or calculated from raw data (seismic or other). At another level, a data product can be essentially anything, including scientific documents, images, station information, and other data. These ‘levels’ of data products, as they are called, refer to orders of derivation and representation from the original raw data and lower level products. Raw data occupies Level 0 and Level 1, and SPADE will begin presenting Level 2 and Level 3 products as it enters production. Higher levels of data are possible and all a part of the overall design goal of this archival system.
SPADE’s configuration-driven architecture allows new product types to be added to the system at any time, while still providing the product-specific query for each new product type. This allows researchers and data centers to the ability to create new and different product types and make those products immediately available to users.
Users can access the system in several ways, including interactively through the Web and with stand-alone programs that communicate over the Internet using SOAP-based Web services and the SPADE Application Programming Interface (API).
The archive supports efficient queries across the product space. That is, a user may construct queries that match to fields common among all product types, but may also form queries that filter on parameters specific to a certain product type, refining the search space. The resultant matches are displayed for the user as a list that can then be examined and downloaded.
Because SPADE couples data products with supporting metadata, the returned products are in the form of XML bundles: a self referential directory of index file and supporting data files. Work is proceeding on tools to provide formatted presentations of the resultant XML bundle, configured specifically to each product type and method of presentation.
Current products in the archive include scanned WWSSN film chip images for a number of significant earthquakes thanks to work done by Willy Lee, and ongoing submission of Record Section plots for the USArray Backbone. We are currently working on developing a Centroid Moment Tensor product for the global CMT catalog, and MagnetoTelluric Impedance Functions with Oregon State. There are also a number of example products currently in the archive as well, which are used for testing and demonstration, including Hypocenters, test CMTs, and USArrayStation information. These test products are unofficial and are not at all representative, up-to-date, or complete. These products will be removed as more new products come online.
Going forward, the SPADE project will benefit most from the contributions of enterprising scientists, providing meaningful value-added products, findings, and determinations as a result of geophysical data processing. It is our intention to have SPADE be a digital library to ‘grow’ knowledge from the seeds planted by earlier works, all as a result of a community of users who glean information from it and give back in the form of new products and findings. In this way, SPADE hopes to contribute to the burgeoning geoscience cyberinfrastructure through the promotion of knowledge sharing and discovery.
SPADE (now SPUD).
by Linus Kamb , Rob Casey and Tim Ahern (IRIS Data Management Center)