DMAC Modelers Caucus Status Report

1
DMAC Modelers CaucusStatus ReportRecommendat
ions

• Steve Hankin, Chair, Modelers Caucus

2
11 members

• Steve Hankin,  convener/DMAC ST
• John Wilkin -- coastal physical modeler --
  Rutgers University/RA
• Rich Signell, coastal physical modeler -- USGS
  -- rsignell_at_usgs.gov
• Frank Bub -- operational ocean modeling --
  NAVO -- frank.bub_at_navy.mil
• Frédréique Blanc -- International GODAE data
  manager -- CLS,Space oceanography Division /
  Direction Océanographie Spatiale -- fblanc_at_cls.fr
• Ashwanth Srinivasan -- data assimilation
  modeler/data manager -- ltasrinivasan_at_rsmas.miami.e
  dugt
• V. Balaji -- modeling frameworks/metadata/grids
  -- NOAA/Geophysical Fluid Dynamics Laboratory --
  ltV.Balaji_at_noaa.govgt
• Eoin Howlett -- software developer/GIS --
  Applied Sciences Associates -- ltehowlett_at_appsci.co
  mgt
• CJ Beegle-Krause -- disaster
  management/trajectory modeling -- NOAA/HAZMAT --
  ltCj.Beegle-Krause_at_noaa.govgt
• Ben Domenico -- data tools and formats --
  Unidata -- ltben_at_unidata.ucar.edugt
• Tim Pugh -- Ocean/Marine Forecasting
  (international) -- Australian Bureau of
  Meteorology Research Centre -- tpugh_at_ves.com

3
IOOS, DMAC Modeling

• The 3rd IOOS subsystem
• Observing subsystem
• DMAC
• Modeling
• A focus on modeling to support IOOS goals
• coastal/regional processes and management
• biological and ecological models
• navigation, health and safety, disaster
  response, security
• climate

4
Why a Modelers Caucus?

• Address the data management needs of modelers
• Provide guidance to ensure priorities are set
  that reflect modeling considerations, too (often
  neglected)

5
You may recall

• 11 members Federal, academic, international,
  global, coastal, physical, (biological)

• Last DMAC-ST meeting we developed a laundry list
  of modeling data management issues. A starting
  point

6
Our immediate task Develop a modelers data
management issues list (maybe a white paper?)
7
Approach

• Established email list serve
• Established Wiki (labor intensive)
• Have held 3 telcons (7am to 9pm)
• Editing via Wiki and email

8
1. CF (Climate and Forecast) Conventions

• The IOOS modeling community has identified the CF
  conventions as vital to model interoperability.
  CF should be supported and advanced as rapidly as
  feasible. Critical areas in need of extensions to
  CF are
• 1. unstructured (finite element) grids and
  nested grids (critical to coastal modeling)
• 2. Geospatial datums (critical to GIS
  interoperability with applications)
• 3. Nested grids

9
730AM, 11 May Statement of Work Integrating
Unstructured Grids into netCDF COARDS CF
Standards - Workshop

• The goal of this work is to create a community
  process for creating metadata standards for
  unstructured grids added to the netCDF CF
  (Climate and Forecast) metadata standards

10
2. Downscaling techniques

• Vital societal functions such as tracking of
  hazardous waste spills, search and rescue, and
  inundation prediction (storm surge, flood,
  tsunami) require the ability to downscale basin
  scale models as input into coastal and regional
  models and these in turn as input into
  operational decision support and decision support
  tools. To support these capabilities requires
  development of tools and techniques for
• 1. Conditioning the outputs of global or other
  large scale models
• 2. Conditioning the outputs from multiple models
  over neighboring domains as input for a single
  problem domain

11
3. Models accessible to GIS

• DMAC needs to recommend standards, tools, and
  best practices to achieve interoperability with
  GIS systems and their users through the use of
  the OGC (Open Geospatial Consortium) WxS
  standards (WMS- Web Map Service, WFS-Web Feature
  Service, and WCS-Web Coverage Service).

12
4. Real-time notification

• DMAC real-time systems should provide
  notification in real-time of both of data
  availability (new data) and unavailability
  (interruptions to data flow or changes in content
  or format) situations.

13
5. Model metadata

• Richer and better supported metadata standards
  are needed to describe model outputs. In
  particular
• 1. Model configuration.
• 2. Key model run parameters, including
  initialization and boundary forcing.
• 3. Certification (if certified, e.g. operational
  quality for water level forecast).
• 4. Validation level, studies and data sources.
• 5. Output QA/QC level
• 6. Forecast accuracy estimates.

14
6. Assimilation-friendly data

• The DMAC value-added product generation
  capability should include assimilation-friendly
  products of in-situ and remote sensed real-time
  observations

15
7. Model intercomparison

• In order to foster the rapid improvement of model
  state estimation and forecast capabilities DMAC
  should support collaboration/intercomparison
  environments for model outputs

16
8. Uncertainty estimates

• There is a need for uncertainty estimates to be
  provided with model products. This includes
  developing a community consensus on how to
  calculate uncertainty in individual 3D time
  dependent models (beyond ensemble modeling).

17
9. Modeler Education/Outreach

• DMAC can assist the modeling community through
  education and outreach to modelers that will
  increase awareness of community standards, best
  practices and tools available to modelers

18
?? (10. Access Control) ??

• User authentication and authorization for data
  access to observational and model data using
  pull services like OPeNDAP