IOOS DMAC Standards Process and Lessons Learned Anne Ball DMAC Steering Team Ocean.US
IOOS DMAC Standards Process and Lessons Learned
Anne Ball
DMAC Steering Team
Ocean.US
Topics
• Brief overview– IOOS– DMAC
• DMAC standards process
• Lessons learned
• Links to more information
• Discussion
IOOS DMAC Overview
Integrated Ocean Observing System
IOOS
Tracking, predicting, managing, and adapting to changes in our coastal and ocean environment
Integrated Ocean Observing System
SEE
Safety Economy Environment
Oceans Coasts Great Lakes
Integrated Ocean Observing System
• Safety– Earlier and better predictions of severe
weather– Search and rescue– Protect water quality– Homeland security
Integrated Ocean Observing System
• Economy– Optimized shipping routes based on improved
forecasts– More accurate, longer-term forecasts for agriculture– Better planning for coastal construction and zoning– Potential sources for medicines and new technologies– Alternative energy sources– Improvements in offshore drilling practices
Integrated Ocean Observing System
• Environment– More timely and accurate predictions for oil
spills and other pollutants– Reduced vessel groundings– Better ecological protection– Fisheries– Protection and restoration of marine
ecosystems
Integrated Ocean Observing System
• Three subsystems– Observations and Data Telemetry– Data Management and Communications
(DMAC)– Modeling and Analysis
Integrated Ocean Observing System
• 11 Regional Associations
• Federal agencies
• National programs
Data Management and Communications
• DMAC Goal– design the system that efficiently and
effectively links observations to applications by enabling rapid access to diverse data from multiple sources as needed
Data Management and Communications
• Infrastructure– Data discovery, access, transfer, metadata,
archive– identify the technologies, formats, and
protocols that support an infrastructure needed to meet its goal.
– must interact with regional, national, and international systems.
Data Management and Communications
• Usability– Quality assurance, quality control, information
requirements– Data must be accompanied by the information that
supports its proper use and sound results when combined with additional data.
– This information may vary between different types of observations and collection methods.
– DMAC will adopt quality assurance, quality control, and content requirements identified by expert communities and ensure this information is carried within the DMAC infrastructure.
DMAC: One of three IOOS subsystems
ObservingSubsystem
Satellites
Aircraft
Fixed Platforms
Ships
Drifters & Floats
AUVs
Metadata standards
Data discovery
Data transport
Online browse
Data archival
Safe & EfficientMaritime Operations
Homeland Security
Natural Hazards
Climate Change
Public Health
Ecosystem Health
Living MarineResources
DMACSubsystem
Modeling & AnalysisSubsystem
DMAC Implementation
Team approach
• Steering Team
• Expert Teams
• Caucuses• Working groups• Interagency Oversight Working Group
DMAC Steering Team
• Representatives from government, industry, academia, public, and non-profits
• Coordinate and oversee DMAC standards evolution
• Identify and provide recommendations on gaps
• Conduct process in an open, objective, and balanced manner
DMAC Teams
• Expert teams– Archive– Metadata and data discovery– Transport and Access
• Working group– Systems engineering
• Address key issues as defined in DMAC plan• Review and make recommendations on
standards
DMAC Caucuses
• Outreach and community engagement– International– Private sector– Education
• K-12• Professional development
– Modeling– Regional
Interagency Coordination
Interagency Oversight Working Group (IOWG)
• Representatives of federal agencies
• Oversight of DMAC implementation within agencies
Roles
• Teams:– Review standards– Make recommendations– Identify gaps
• Organizations– Implement standards and recommendations– Fill-in gaps
Standards and standard process
Standards
Standards are key to the success of DMAC
• Identify standards needed• Coordinate with other standards processes• DMAC standards mantra:
– Adopt– Adapt– Develop only as a last resort
DMAC Standards Process
• Identifying existing standards is a first step
• Often need to “standardize” the standard– Ex. Use FGDC metadata – not good enough
to describe waves, so must develop a waves profile
• Data providers may be required to use different standards– Crosswalks, interoperability is critical
DMAC Standards Process
Three status levels for standards:• Submitted
– Has no standing within IOOS.
• Proposed– Has status within IOOS and can be used provisionally
by IOOS information system managers for evaluation purposes.
• Recommended– All IOOS data systems should consider supporting
these guidelines wherever applicable
DMAC Standards Process
Originator Submits a Standard
Internal Review
DMAC Chair –Applicable?
Gatekeeper –Complete?
Start
Expert ReviewDMAC ST
Polls
Public Comments
Community Conducts
Tests/Pilots
DMAC ST Polls
End
“SUBMITTED STATUS”
“RECOMMENDED STATUS”
“PROPOSEDSTATUS”
Website: ioosdmac.fedworx.org
Submission Entry
DMAC Standards Process
• Entry fields:– Name– Type (new, existing, adaption)– Data mgmt function (metadata, transport, …)– Abstract– Purpose– Technical description– Statutory requirement or international agreement?
DMAC Standards Process
• Entry fields (continued)– Relationships, dependencies, conflicts– Current usage– Justification– References– Acronyms– Contact information– Supporting parties/members
DMAC Standards Process
• Semi annual review schedule:– Expert review
• 1 month• September and March
– Steering team review• 1 month• October and April• Steering team initial poll (on-line)• Recommendations written
– Steering team final poll• May and November meetings
DMAC Standards Process
• Chair drafts recommendations based on polling and distribute to Steering Team
• Steering Team discusses recommendations at semi-annual meeting– Try to reach consensus– If consensus is not possible (per Terms of Reference):
• Vote is carried by 75% of those members voting “approve” or “disapprove.”
• Quorum consists of 60% of members.
• Support of multiple strategies may be the most prudent approach. As long as the consideration of multiple options does not violate the spirit of IOOS interoperability, the IOOS-ST may move forward without consensus.
DMAC Standards Process
• Steering team may decide:– Move forward (“submitted” to “proposed”)– Request additions or changes– Table and revisit in next round– Reject
Lessons Learned
First round - November 2007
Eight standards submitted:
• IOOS Vocabulary Version 1
• Required and Recommended Vocabularies for IOOS Metadata
• Metadata Catalog Services
• Data Access Protocol – DAP 2.0 (transport)
• Real Time Quality Control Tests for In Situ Ocean Surface Waves
• Quality Control Standards for Real-Time, In Situ Currents Measured by Teledyne RD Instruments
• High Frequency Radar Surface Currents
• Standards Package for the Representation and Transport of Gridded Data: netCDF+CF+OPeNDAP+aggregation
First round
Four moved to “proposed”:
• IOOS Vocabulary Version 1 • Metadata Catalog Services • Data Access Protocol – DAP 2.0 (transport) • Standards Package for the Representation and Transport of
Gridded Data: netCDF+CF+OPeNDAP+aggregation
First round
Four tabled:
• Required and Recommended Vocabularies for IOOS Metadata
• Real Time Quality Control Tests for In Situ Ocean Surface Waves
• Quality Control Standards for Real-Time, In Situ Currents Measured by Teledyne RD Instruments
• High Frequency Radar Surface Currents
First round
• Tabled submissions:– Real Time Quality Control Tests for In Situ Ocean
Surface Waves – Quality Control Standards for Real-Time, In Situ
Currents Measured by Teledyne RD Instruments – High Frequency Radar Surface Currents
• Should DMAC be involved in QA/QC?• If so, how?
First round
Tabled submission:• Required and Recommended Vocabularies for IOOS
Metadata
• Adopt multiple vocabularies?• Use technology/ontologies to map multiple vocabularies?
First round
• Findings– Wide variety of “standards” submitted– No clear definition of “standards” vs “best practices”– How do we handle QA/QC standards?– Need better instructions and/or help for submitters
Additional thoughts
• May need to better define what we’re looking for
• May need to focus more on data/metadata content
May 2008
• Addressed tabled submissions– Additional fields– Tried to refine definitions of fields– Tried to define “standards” vs “best practices”
QA/QC
• “Lessons Learned” discussion:– ST lacks technical expertise to confidently evaluate
QA/QC– Better criteria needed
• To guide ST in evaluation• To assist Originators in preparing “Work Packages”
– Draw on expertise of QA/QC-experienced organizations
• As Originators• During Expert review
– Devote more effort (time) to submission ‘Work Packages’ at front-end
QA/QC
• Is QA/QC a DMAC issue?– Regardless, DMAC must deal with QA/QC to move
forward– Develop supplemental criteria for reviewing QA/QC
submissions
QA/QC Applicability
• Does the submission:– Address an IOOS data type or parameter?– Significant to multiple applications or users?– If specific to a particular instrument, can approach be
extended to similar instruments?– Would it benefit IOOS if implemented more widely?
QA/QC
• Originating organization– Does Originator have recognized expertise in the
topic area?– Will the standard (including, if applicable, all
components of the submission) be maintained over time by a recognized organization?
• Supporting organizations– Are recognized organizations or programs supporting
the adoption of this as a standard?
More information
DMAC Documents
DMAC Plan
•Detailed plans for implementing DMAC •Created with input from federal and state governments, academia, non-profits, and private industry
http://dmac.ocean.us
DMAC Documents
Guide for IOOS Data Providers
• Draft of the IOOS Data Policy
• Guidelines for data and metadata interoperability standards and best practices
http://dmac.ocean.us
DMAC Standards Website
• Submit standards
• Provide feedback on standards (public comments)
Discussion