You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 3 Next »

Storage of data is a basic task in any Water Resources Management activity. This data often includes observations/measurements, meteorological forecasts, historical model simulations and model forecasts. Often, this data is stored using commercially available databases including WISKI, Aquarius and CUAHSI HIS. These databases allow for storage of both timeseries data and rating curves. However, it may not be possible to store forecast data (which is characterised by multiple time vectors) or 'messy' data such as text reports, communications logs, model states, database snaps, et cetera. For that purpose, Deltares developed the Open Archive.

The Deltares OpenArchive

The OpenArchive is a file based storage solution where data can be accessed through a catalogue. The OpenArchive intimately integrates with Delft-FEWS applications but can be used independently from it. The data can be easily accessed by third party applications using web protocols.

Delft-FEWS and the OpenArchive: seamless integration

The OpenArchive intimately integrates with Delft-FEWS in a number of ways:

  • Delft-FEWS can export data to an archive.
  • Archived data can be visualized within Delft-FEWS.
  • Archived data can be used in workflows in a Delft-FEWS application (for example in a hindcast run).

Sample applications

The archive can contribute to a number of tasks often encountered in water resources management. These include hindcasting (re-forecasting), forecast verification and (web) dissemination.

Hindcasting, sometimes referred to as re-forecasting, is the process of producing forecasts for times in the past, using only data that was available at the time. Reasons for hindcasting include testing of new models or other forecasting routine such as data assimilation or statistical post-processing. Hindcasting requires the availability of forcing data. In the case of hydrological forecasts, this includes weather forecasts that were available at the time. Every hindcast should then be forced by another weather forecast. Delft-FEWS is eminently suited to carry out these hindcast exercises, especially if connected to an OpenArchive. Within a hindcasting workflow, past weather forecasts are retrieved from the archive and used to force hydrological and/or hydrodynamic models.

Hier beschrijven we hoe het archief gebruikt kan worden bij drie denkbare toepassingen:

  • Hindcasting
  • Verification
  • Web viewers

Available documentation

  1. The OpenArchive: technical background
  2. Installation of required software
  3. Setting up / configurating an OpenArchive (NB misschien is dit helemaal niet aan de orde? heeft OpenArchive een eigen configuratie?)
  4. OpenArchive maintenance
  5. Linking a Delft-FEWS application to an OpenArchive
  6. Linking third-party applications to the OpenArchive
  • No labels