Child pages
  • Sprintsession June 4th, 2008

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Migrated to Confluence 4.0

A third sprintsession will be held on June 4th, 2008, in the meeting room Innovation of the new institute on Delta Technology Deltares including drinks in De V at the end of the day. As is now the tradition the whole day is open to everybody, so please feel free to walk in and out when you feel like it, contribute to each of the topics on the agenda or to raise other items that are not on the agenda yet. It promises to be yet another productive session. McTools is now at revision 1193 (June 3rd, 2008 10am). This means that the threshold of 1000 has been passed! The free dinner goes to Pieter van Geer.

Agenda

The day will start with from 9:30 - 11:00 a number of presentations and demonstrations.

  • 09:30 - 09:45 Introduction (Mark van Koningsveld) Brief introduction of the day
  • 09:45 - 10:15 Postprocessing 3D Delft3D results (Thalweg, Firmijn Zijl and Gerben de Boer) - Firmijn developed a number of tools for the postprocessing of 3D Delft3D resultats and added these to McTools. He will inform you about the functionalities that have now become available to you.
  • 10:15 - 11:00 Online model morphological storm impact (Maarten van Ormondt et al) Maarten will inform you about the current status of the online model for morphological storm impact prediction. The last sprintsession you all helped to make a workplan to set up such an operational model train in preparation of the EU FP7 project Micore. Versie 1.0 of this model train is now in operation. The results can be monitored here. There will be room for discussions on further development of this instrument (e.g. coupling with BeachWizzard).

From 11.00 until lunch the use of DelftAlmighty will be discussed. In previous sprintsessions and lunchmeetings, DelftAlmighty has been demonstrated regularly. This already gave rise to several follow-up discussions regarding the use of DelftAlmighty in projects (some pilot applications where made like the hindcasting of the Myanmar hurricane - zie Deltares website). Spontaneous use of DelftAlmighty sometimes revealed the existence of some application thresholds. This Sprintsession Maarten is available to recieve feedback on which bottlenecks remain. These will be solved with priority.

The rest of the day will be filled up with hands on treatment of a number of topics that were handed to us by you.

  • Starting 13:00 MAD (Model Application Database) At the request of Andries Roelfzema some time has been allocated in the afternoon for the Model Application Database (MAD). E number of people have been approached to come and enter their models into MAD. Fedor Baart is available for assistance. Some effort will be spent investigating how MAD and DelftAlmighty can be coupled (this will enhance visibility and useability).
  • Starting 14:00 Testbanks The need for testbanks, to garantuee the quality of certain often used or important routines, surfaces regularly during major development efforts. This way regular testing of important routines during development efforts, to assess whether an improvement here did not cause any other problems there, becomes possible. During the previous Sprintsession the Team City Buildserver was addressed briefly in the activities surrounding MorphAn. Since then a number of routines was set up to facilitate testing in a transparant and uniform way. One of the items addressed will be how the currently developed tests can be run on the Buildserver. The discussion is wide open to involve other initiatiatives (e.g. XBeach)
  • Additional topics
    • Voortgang afstudeerwerk Thijs Damsma (te bespreken met Dano, Dirk, Mark en Thijs),
    • voortgang Morphan project (Sander, Mark, Pieter en Kees)
  • Lopende zaken rondom toolbox onderhoud en documentatie (Robin, Gerben en Mark)

17:00 vertrek richting De V voor een hapje en een drankje (tot ca 19:00 uur)

Everyone is obviously free to join and contribute to each of these topics or to raise other items that are not on the agenda yet.

Results

  • 19 People attended
  • Firmijn Zijl
    • presentation
    • afternoon make a plan on how to include these tools in the toolbox and Muppet
  • Maarten van Ormondt
    • brief presentations of the models that where setup, coupled and run
    • show of the website
    • developments
      • other wind sources
      • include BeachWizard
      • include more XBeach models
      • discuss about the desired outputs
    • problems with precision of writing spectra data for hotstart files solved
    • nesting problems solved (to overcome resolution steps from model to model)
    • output for XBeach from SWAN solved (spectral data could not be written cumulatively)
    • plans for online train
      • investigate coupling with FEWS
      • include BeachWizard before the end of June (Bologna meeting) (Ap and Irv)
      • include flowfields on the wiki (running average of the currents display the last field? curved colored vector plots) (Mark and Maarten)
      • profiles (extract profiles Jarkus location) (Mark)
      • demo application storm 8-9 nov 2007 (experiment with what would be interesting output)
      • include Schiermonnikoog (Jaap, Jamie and Maarten) Dano!!!
      • move train to a faster computer (Theo en Mark)
    • start the modeling including uncertainties where Rijkswaterstaat work is ending (we should bring together existing lines and develop skills locally to act globally)
    • search for links with FloodControl 2015 (what can we do in realtime with duneerosion). Coupling with BeachWizard helps to get the winterprofile before the storm. Mark and Dano should make a plan ... plan to be submitted to Karel, then find other partners and try to get the project approved - make the link with the US).
    • combination of Matroos and FEWS (should be combined in the future) by 2012 we have to deliver an advanced forecasting system to Rijkswaterstaat (in the management contract)
    • make a coupling to smaller Delft3D models (include operational Delft3D models)
    • how to organise (infrastructure - create a research environment (this should be a project), glue projects to this environment) Lead Karel Heynert
  • Make mailinglist
  • Update documentation
  • Work on DelftAlmighty link