Policy Committee Meeting Summary:

29-30 October 2007

Unidata Program Center, Boulder, Colorado

Participants

Members

UPC Staff

Steven Businger—Chair

Mohan Ramamurthy

Rich Clark

Ben Domenico

Vanda Grubisic

Ethan Davis (for John Caron)

Rudy Husar

Jo Hansen

Jim Koermer

Linda Miller

Paul Ruscher

Terry Mitchell

 

Russ Rew

Agency Representatives

Jeff Weber for Monday 9:45 agenda item

Jeff de La Beaujardiere, NASA

Tom Yoksas

Bernard Grant, (NSF/ATM)
Clifford Jacobs (NSF/ATM)

The following staff members were present for portions of the meeting: Steve Chiswell, Steve Emmerson, Anne Wilson

 

 

Committee Representative

NCAR and UOP Representatives

Gary Lackmann (Users Committee Chair)

Jack Fellows, Director, UOP

 

Hanne Mauriello, UOP Director's Office (Monday morning)

Not in Attendance

Tim Spangler, Director, COMET Program (Monday only)

Leroy Spayd, NOAA

Roger Wakimoto, Director NCAR/EOL

Mike Bevis

Steve Worley, NCAR/CISL/SCD (Monday only)

Gene Takle

 

David Tarboton

 

Melanie Wetzel

 

See: Unidata Acronym/Glossary List
See: Meeting Index

Monday, 29 October 2007

Administrative Matters (Businger)

  • The Draft Agenda was approved
  • The July 2007 meeting notes were approved.
  • Next Meeting Dates: 28-29 April 2008 at NSF Headquarters in Arlington, Virginia
  • Introductions: new members, Mike Bevis and Vanda Grubisic, were introduced. Thank you and farewell to Jim Koermer for his service on the committee.

Director's Report (Ramamurthy)

Mohan's presentation began with recognition of the new policy committee members and new Users Committee member, Bill Gallus. He also noted that committee member, Rich Clark, will receive the AMS 2008 Teaching Excellence Award at the AMS Annual Meeting, and Unidata former director, Dave Fulker, will receive the 2008 AMS Cleveland Abbe Award.

Since the last committee meeting in July:

  • Unidata's Strategic Plan has been completed and released to the community at large where it has been favorably received.
  • Data flowing through the IDD has remained steady except for TIGGE and CONDUIT. In the former, data volume is averaging12-13 GB/hour but there are bursts up to 23 GB/hour. The number of participating sites continues to grow.
  • A CONDUIT meeting is planned for the AMS annual meeting.
  • Use of visualization packages, GEMPAK and IDV continues to increase while McIDAS-X use is decreasing. Interestingly, the majority of new interest in Unidata McIDAS is from the international community.
  • A beta release of netCDF 4 was made available in April. Developers turned their attention to mentoring SOARS student Shanna-Shaye Forbes as she developed a netCDF C++ interface. They will begin work on integrating netCDF with OPeNDAP, an activity supported by funds from a recently funded proposal. In addition, the proposed work includes extending the OPeNDAP protocol to support Unidata's CDM.
  • UPC's Annual Training Workshops attracted 75 registered participants with 41 from academic and research communities, and 28 from government organizations. The largest ever netCDF developers workshop was successful. All five participants from China were denied visas and could not attend the workshop.
  • In addition to the training workshops held locally, a regional workshop took place at the University of Oklahoma. Jointly hosted by Unidata and Oklahoma, it featured training sessions on THREDDS, Integrated Data Viewer, and an overview of and vision for the Unidata program by Mohan.
  • As reported at the July meeting, Unidata's Equipment Awards were made to six institutions including one to Plymouth State, which has proposed to host a regional workshop May 18-20 2008.
  • An OGC Interoperability workshop that took place in September attracted 63 attendees with a theme of Standards-based Web Services Interfaces to
    Existing Atmospheric/Oceanographic Data Systems.
  • LEAD PIs visited NSF on 30 August to discuss the project's future. The PIs met with officials in ATM, OCI and CISE and presented an invited seminar to NSF staff and a separate meeting with NSF officials in which a possible approach to LEAD-2 was discussed.
  • An OPeNDAP proposal to the NSF SDCI solicitation in which Russ Rew is co-PI was successful; in other proposal activity the VOC proposal has been tabled for submittal at a later date, Unidata and the Shanghai Typhoon Institute of Shanghai Meteorological Bureau (SMB) are developing a proposal whose goal is to transfer the Unidata IDV to STI for the development of an IDV for tropical cyclone analysis and prediction.

In effect, the state of the Program is quite similar to its position at the July meeting when only two areas were noted as areas of concern, funding and staffing. It's recognized that it's impossible to separate the two.

Budget Report (Mitchell)

With increasing budgetary constraints, we have always kept an eye out for funding above and beyond our NSF core funding. Taking a look at the proposal statistics for the last 10 years shows that Unidata is at about a 54% success rate for funded proposals compared with UOP's overall 77% rate. Both are good compared to the national average of about 30-40%.

Turning to our analysis of FY07, we ended the year on track with our projections. A key point to make is the carryover from FY06, which consisted of $655k in FY07 funding that was received late in FY06. Following that trend, we also received some FY08 funding late in the fiscal year to add to our carryover balance for FY08. So that, in combination with funding on the CADIS and NEXRAD projects, contributed to the considerable FY08 carryover amount.

As expected, we expended more than the $3.431M in new core funding for FY07. Both direct and indirect costs have increased over the current award period, but the percentage breakout by expenditure type has held steady, with salaries, benefits, and overhead being the biggest outlays.

We're hoping that along with carryover, the anticipated 4.5% increase in core funding for FY08 will offset our deficit spending and put us in a favorable position as we head into the new award period starting in FY09. Without any new direct funds in the next two years, however, we will have depleted all of our carryover and reserves by the end of FY09. We will continue to seek out direct funding as much as possible to supplement our core funding and to leverage our resources to maintain a healthy budget both in the short and long-term.

Data Flow/Case Studies (Weber, Ramamurthy)

Jeff has created a multi-layered, comprehensive web document that will assist users in locating and understanding data available through the program center. Jeff's document leads off with data types, formats, access protocols, and clients headers. Each of those headers contains links to more images and information. For example, "Data Type", when selected, expands to a diagram that illustrates the current display capabilities and maps out the process of how to use that type of data with Unidata tools. In addition, every object provides information about the data or format, and can be followed to find examples of how the data will look when visualized with Unidata clients. Jeff emphasized that input on the document’s organization and structure are welcome, and he invited committee input on it.

Case Studies/THREDDS Data Repository

A case study project has always been popular with the Unidata community. Jeff reported on the successful evolution of the current project that originated with Mohan and COMET director, Tim Spangler. The project is internally funded by STORM funds generated via management fee to UCAR. Outside collaborators are a UNC-Charlotte team consisting of Dr. Brian Etherton, faculty advisor, and Ms. Shelly Holmberg, graduate student, who are creating the first series of case studies and the scientific write ups associated with them. The intent is for these case studies to be amendable and not static by using the THREDDS Data Repository to augment them and enrich them, and, when applicable, COMET learning modules.

In addition, they will be living or dynamic, allowing for the community to augment and add value to existing case studies by contributing related observations, analyses, educational, curricular, and multimedia materials, and other views on the case. An important element of that perspective is to build an easy to use framework in which community members can develop and build on future case studies.

Users Committee Report (Lackmann)

Gary presented a rundown of the recently held Users Committee meeting. Sean Arms, University of Oklahoma grad student, who is the student rep on the Users Committee, conducted a survey of 127 graduate students that showed that 40% of them were unaware of Unidata's data products or software. Some means for dealing with the survey results included a student section of Unidata's CommunitE-letter was mentioned along with setting up a student forum on one of the social networking sites like Facebook. Discussion will continue, but no definite action was taken at the meeting.

The committee speculated that AWIPS-II development and the transition of N-AWIPS (GEMPAK) to AWIPS-II will impact many users during the next proposal period. Unidata will need to track those developments and may need to make course corrections during the period. The question of the type of open source license by Raytheon remains unanswered. Questions about porting GEMPAK to Java need to be answered. There was a discussion of a potential letter to NWS regarding the open source license, but no decision was made.

Agency Reports

NSF (Jacobs)

NSF's budget funding remains a priority with the administration. A 7% projected annual growth through 2014 is anticipated. GEO's request for FY 2008 is up 5.9% over 2007. Congress has not passed any of the appropriation bills, and at this point, NSF, like other agencies, is operating on a continuing resolution. With the Thanksgiving break coming up, it seems likely that will continue to be the case.

There is strong support for climate change science, research on natural hazards, the near-term priorities of the Ocean Research Priorities Program, and in GEO funding fundamental research remains a priority.

In GEO news, the search for a director continues. Guided by a strong panel, the GEO-Vision 2007 document is nearing completion. The document is defined by an outline that includes four points: Planet Earth, The Earth--A Complex System, The Earth--A System of Change, and fourth, Meeting the Challenges. NSF is seeking community input for this document.

In agency-wide activity NSF seeks to define, then facilitate transformative research. It does this in several ways, among which is making small, SGER, grants for exploratory research. NSF has been tasked by the National Science Board to develop and Foundation wide Transformative Research Initiative. In response to the request, NSF developed a three-pronged approach that includes infusing support of Potentially Tranformative Research (PTR) across all programs, learning how to facilitate PTR, and leading the community through proposal funding possibilities for PTR. There was the comment that the NSB is not on board with all of these suggestions.

NASA (de La Beaujardiere)

Jeff created and implemented a NASA Unidata mailing list in July 2007. Its purpose is communicating Unidata information to NASA stakeholders. Thus far its membership is limited, but Jeff is confident it will grow. [NOTE: following the meeting Unidata staff forwarded a list of 191 NASA based users.]

Goddard Earth Sciences (GES) Data and Information Services created a case study using the OPeNDAP 4 data server (Hyrax). The case study is composed of four principle components:

  • An Atmospheric Infrared Sounder,
  • A total ozone mapping spectrometer,
  • A tropical rainfall measuring mission, and
  • A microwave sounding unit.

There's been positive feedback by its users and ultimately a THREDDS catalog will be created.

NASA funding opportunities include an Education and Public Outreach solicitation whose goal is supporting coordinating and integrating those activities. Four awards of 3.5M each, one in Earth Science, will be made.

In the NASA ROSES (Research Opportunities in Space and Earth Sciences) six solicitations remain open in the following areas: Ocean Surface Topography, Earth Surface & Interior, EarthScope InSAR & Geodetic Imaging Component, Terrestrial Hydrology, and Energy & Water Cycle Ocean Biology & Biogeochemistry.

Discussion

  • Jeff will provide a NASA budget update slide, similar to the slide usually presented in the NOAA slides in his next presentation.

Unidata and Climate Data Services

The agenda item was framed by the following questions/points: 1) What is needed, 2) what is missing, 3) how can Unidata facilitate, and 4) related items, such as Unidata's contribution to facilitating IPCC AR5 Data Services.

Some indications of Unidata's presence in the climate community are already apparent. One is extensive use of netCDF in the climate community--almost all climate model output is stored and accessed in netCDF. Next, the broadly defined goals in the Strategic Plan are meant to comprehend it, and the Users Committee added a climate scientist to its membership a year ago.

Lawrence Buja, NCAR/CGD, joined the meeting for this agenda item, and opened the discussion with a presentation entitled Climate Data Case Study. In addition to nearly-comprehensive netCDF usage for data storage, the presentation also affirmed Unidata's visibility in metadata and data standards. In the climate data arena, data integrity is critical, and thus verification of its integrity is critical to the climate enterprise.

Discussion

Some discussion items:

  • Climate data is larger in scale, and tools to analyze and visualize it must be able to deal with long time periods, global scales, and large datasets. .
  • At this time, Unidata tools don't lend themselves to tracking a single variable over a longer period of time--a requirement in the climate community.
  • The metrics assessment undertaken a year ago by Unidata indicated community interest in greater collaboration with the climate community.

The committee recognized that a move to include the climate community is a significant shift in the program's activities; but it seems clear that it is desirable.

COSMIC Data (All)

UCAR President, Rick Anthes, opened the agenda item, providing a video illustrating the origins of the GPS-Met observing system that was launched in 1995 and was named SuomiNet for legendary scientist, Vern Suomi. GPS-Met was the precursor of COSMIC. Unidata, which participated in GPS-Met from its inception, continues to distribute the data on the IDD/LDM network. These data were and are available at no cost.

Following the SuomiNet video, Chris Rocken described the COSMIC data and provided a background of the program in a presentation. Presently six satellites orbit the earth and transfer data to three ground-based stations. The data benefit members of the meteorology, climate, ionosphere, and geodesy communities. Both research and education communities derive benefit from them.

At this time, like SuomiNet, COSMIC data are freely available upon approval of an e-mail request made to the National Space Organization (NSPO). The NPSO Director, along with UCAR President Anthes, reviews the requests. Thus far, every request has been approved. Users are also asked to cite COSMIC in publication and presentation materials.

The data are already ingested by operational centers like NCEP and ECMWF in their assimilation systems, and visualizations of some of the data have been created by Unidata's GEMPAK and IDV developers.

The Unidata survey

In an action item that originated at the July Policy Committee meeting, Unidata staff members, Policy Committee chair, together with Dr. Anthes, prepared a survey for distribution to the Unidata Community. Of fifty-five respondents, 27 percent were familiar with the data, and most of those (96%) received the data on the FORMOSAT feed. Respondents are enthusiastic about data distribution on Unidata's event-driven LDM and are in favor of visualization and analysis capabilities of the data becoming part of the IDV and GEMPAK.

Discussion

  • Everyone agreed that the data use policy would need to be revised for Unidata users, but registration will still be required
  • The advantages for Unidata's COSMIC data users would be access to the tools and suite of services Unidata offers.
  • The committee wondered whether Unidata would be obliged to educate the community about the data
  • The prevailing sentiment was that each student would have to become a registered user in a classroom setting.

Some wrinkles remain in bringing COSMIC data to Unidata's user community through LDM, but at this time there's agreement that Unidata and COSMIC need to pursue solutions to effect the data delivery.

Developing metrics to measure progress on the Strategic Plan's goals

The recently-completed Strategic Plan will provide guidelines for structuring the 5-year funding proposal, and staff will develop a set of metrics to measure Unidata's effectiveness in realizing the goals. While this discussion ranged far and wide, its primary focus was software where (among others) the following points were made:

  • Most Unidata software developers track both the number of users and number of sites. Exceptions are LDM, which tracks number of downloads only, and LEAD, which tracks number of sites.
  • In this context it was pointed out that THREDDS shares text and information in addition to data, which complicates its metric. In this regard, committee members thought that data available on the internet, should be like music available on the internet, easy to find and easy to use.
  • Milestone-setting for software development, for example, "500 new users in the next quarter for the IDV," yielded this response: Unidata doesn't measure gains in users because its primary goal is not having software that's the most popular and ubiquitous. Unidata isn't in competition to achieve the greatest "market share." Instead, its goal is high-quality software: reliable, efficient, and useful even if it's not the most popular. Popularity is a side effect of good software, but if that (market share) is the only goal, developers end up adding "features" to packages instead of designing good software.
  • NSF, however, needs a measurement that probably includes a milestone component to establish that there's been a good return on its investment.
  • Software development could (should?) be a framework for collaboration, which is key to knowledge creation.

In discussion surrounding community and collaboration, the following points were made:

  • The commonality between and among disciplines has to be part of the larger community growing plan. The IPCC panel was mentioned as an example.
  • It is also possible that providing space for collaboration with non-traditional collaborators is an important consideration.
  • Committee members wondered how to go about identifying the 2013 partners. There are a couple of possibilities:
    • Rely on organic growth as Unidata has done in the past
    • Identifying possible future areas for collaboration was mentioned. Biology/Ecology was mentioned as an area where potential collaborators and projects are probable and even desirable, and could yield societally important results.

There was consensus among committee members that as part of its proposed activity in the next five years, Unidata should build bridges, or create spaces, or recognize commonalities among diverse disciplines or all of these. The concept of collective space did not exist three years ago, and it's important to recognize and be creative with that concept.

And, although the end of the proposal's period of performance is 2013, the proposal should, to some extent, look beyond that date.

NSF Facilities Users Workshop: How can Unidata Help with cyberinfrastructure for the field project community?

Recognizing the need to interact with its user community, EOL, University of Wyoming, and Colorado State University held the first-ever users workshop in Boulder in September. Roger Wakimoto, NCAR observer on the committee and director of NCAR's Earth Observing Lab, presented a report on the NSF Observational Facilities Users' Workshop. Unidata-specific points highlighted in the presentation were:

  • Infrastructure and services through the field phase and
  • data stewardship

Roger's presentation also included a couple of slides that were created and presented at the workshop by Unidata staff member Don Murray and Chris Burkhardt (EOL). IDV developers have actively pursued collaboration with EOL in field projects. A general view is that this is a good time to assess the role of Unidata tools, technologies, and services in facilitating the field project community.

Discussion

  • Remote display of real-time data is needed – virtually dropping in to a project.
  • Community repository idea was supported, but has overhead and some drawbacks. Central support (ala Unidata) might be a better model to reduce duplication of efforts
  • Need a browse capability for project data (e.g. SDSMT browser) – maybe a model for UCAR efforts
  • Many community members use IDL/MatLab/S+ for analysis now

Next Steps in Proposal Development (all)

As noted earlier in the meeting, Unidata's five-year (2008-2013) funding proposal will be submitted in 2008 near the end of February or the beginning of March. This committee has provided input to the process beginning September 2006 through the May and July 2007 meetings to the present. The committee agreed that community expansion, including geodesy and air quality members is indicated. The committee, did not, however, prioritize the recommended inclusions. There was a suggestion that the intersection of the biosphere-atmosphere (biology/ecology) communities may seem far out, but it must be recognized and "on the radar."

In the area of atmospheric chemistry, the committee felt that Unidata should attempt to identify its role in data delivery and tool development for this community. The area is intimately connected with air quality analysis and thus the air quality community.

While there was recognition that while Unidata cannot move quickly on these ideas (can't be all things to all people, and must respect the core community) Unidata's proposal will explore these possibilities. At the same time, the program must be responsive to the needs of educators while maintaining and expanding the capabilities of the excellent visualization and analysis tools already in place.

Wrap-up

The wrap-up portion focused on diversity: primarily, what steps should be initiated to achieve it. There were a couple of specific suggestions: 1) A Unidata representative at a diversity workshop to be held the last week of March in Boulder; 2) the deliberate inclusion of an MSI representative on the Users Committee.

A second question was Unidata's role in the inclusion of community colleges in its community. Because this is a multi-faceted, complex topic, the discussion was tabled until a future meeting. However, there has been considerable e-mail dialog amongst committee members subsequent to the Policy Committee meeting.

Resolutions and Action Items

Resolutions

The resolutions were approved by e-mail vote immediately following the meeting.

  1. The Policy Committee recommends that the Program Center continue to facilitate climate data services development.
  2. The Policy Committee encourages the Program Center to explore the feasibility of delivering real-time air quality data to the community.

Action Items

The action items were reviewed following the meeting. E-mail suggested revisions are noted in parenthetic remarks.

  1. The Policy Committee recommends that the Unidata Users Committee explore the possibility of adding an NCAR representative to its membership which could be ex-officio. (There was discussion that ex-officio makes the appointment less attractive; however, Policy Committee agency reps, and observers are ex-officio, non-voting. The issue remains unresolved.)
  2. The committee recommends that Unidata proceed in coordinating with COSMIC and its sponsors the distribution of COSMIC data using Unidata technology. This would include a possible modification of the registration agreement currently in place. (Progress has been made in appropriate modification of the data use agreement, and expectations are that a revised agreement will be in place soon.)
  3. The NASA and NOAA agency representatives will provide updates on new observing instruments associated with their respective agencies, e.g., EOS, NPOESS, and GOES.
  4. The Committee requests that Unidata send a representative to EPA's data summit to be held in spring 2008.
  5. The Policy Committee requests that the Users Committee determine which of the available EPA data should be distributed using Unidata technology.

 

Direct comments and questions to jhansen@unidata.ucar.edu