Shaping the Future: Unidata Users as Leaders
Abstract
The Unidata Summer Workshop 2000 brought together over seventy leaders in education, research and technology for the common goal of advancing meteorological education and research through the use of Unidata applications and data. The workshop was hosted by the Unidata Program Center in Boulder, Colorado, 19-23 June 2000, designed by the Unidata Users Committee, in collaboration with Cooperative Program for Operational Meteorology, Education and Training (COMET) and the Digital Library for Earth Science Education (DLESE), and expertly organized by the Unidata Program Office staff.
Workshop attendees participated in a series of lectures and lab sessions that focused on four main themes: the need and opportunities for educational leadership, exemplary uses of Unidata software, adventures in earth systems beyond weather, and new technologies and their educational uses. The blend of lecture and lab together with ongoing interaction between attendees, presenters and Unidata staff created a positive learning environment.
Introduction
The Unidata Users Committee, one of several external advisory committees for the University Corporation for Atmospheric Research (UCAR) Unidata Program, and staff from the Unidata Program Center conducted an educational workshop at its home office in Boulder, Colorado, 19-23 June 2000. The Workshop, called the Unidata Summer Workshop 2000, focused on educational opportunities and challenges facing the Unidata community. Participants saw demonstrations of exemplary uses of Unidata applications and data by leading members of the Unidata community. Hands-on sessions and pedagogical guidance were offered to help attendees incorporate these practices into their own educational and research efforts. The Workshop was sponsored by the National Science Foundation (NSF) and The Cooperative Program for Meteorology, Education, and Training (COMET). COMET participation was funded by the National Weather Service (NWS).
Background
The UCAR Unidata Program provides software and services to qualified educational and scientific research institutions for the acquisition and use of atmospheric and related geophysical data. Since its inception in 1985, over 200 educational and research organizations have joined the Unidata community by becoming software users, data relay sites, and in some cases, developers of software tools that are shared within the community.
Part of the responsibility of the Unidata Users Committee is to conduct residence workshops that bring the community together to exchange information and ideas on how best to use Unidata resources in teaching and research. COMET has many of these same responsibilities within the NWS, so this summer’s workshop was a collaborative effort between Unidata and COMET. Members of the new Digital Library for Earth System Education (DLESE) project were also instrumental in designing the Workshop.
Workshop Themes and Organization
Workshop plenary and laboratory sessions were designed around four main topics; best practices using Unidata applications and data, developments in new meteorological applications (MetApps); creative use of Numerical Weather Predication model output in teaching and research; and case studies using Unidata, COMET, and DLESE resources.
The Workshop was designed to take advantage of the expertise of Unidata members, while stimulating new thinking in a variety of subjects. Keynote speakers opened each day’s sessions with talks that set the tone for continued discussions throughout the week. Invited experts followed with lectures on contemporary academic problems, including changes in student expectations, expanding discipline boundaries, new curriculum emphases, and creative patterns of faculty-student interaction. Following precedents set during prior User Committee workshops [ref. previous BAMS articles from the Unidata bibliography], considerable time was devoted to interactive lab sessions where leaders presented techniques for employing computer graphics to help teach complex cause-and-effect relationships underlying meteorological and related phenomena. An important component of the Workshop was the ongoing interaction between participants, lecture and lab leaders, and Unidata staff.
Keynote Speakers at Plenary Sessions
Cliff Jacobs (NSF) opened the Workshop by challenging the Unidata community to become a leader in science, engineering, and education in the 21st century. He predicted that the Internet, bolstered by recommendations of four major reports (President's Information Technology Advisory Committee 1999; National Science Board 1999; National Research Council 1999; National Science Foundation 2000), will lead to an increase in funding for research and education in geoscience and information technology. He argued that although the Unidata community already embraces a diverse scientific community, it has yet to discover its full potential to contribute to the NSF's strategic goals. The Unidata community must look beyond the delivery of data and provide 1) ideas that will set research and education agendas for the 21st century and discover the connections for their use in service of society, 2) a leadership role in preparing a diverse, internationally competitive 21st century workforce and well-educated citizens, and 3) the innovation needed to develop broadly accessible, state-of-the-art information bases and shared research and education tools.
Tim Killeen (National Center for Atmospheric Research), outlined the future of geoscience education. He argued that interdisciplinary earth system science courses could revolutionize undergraduate general education, citing his experience developing a Global Change curriculum at the University of Michigan. Students are naturally interested in the world around them, and are motivated to use science as a tool to think profoundly about world issues. Electronic course materials allow for inquiry-based learning and simple models of earth system processes are developed for students to learn the importance and limitations of physical models. Course evaluations showed that, compared to single-discipline courses with one instructor, students in the Global Change courses are more challenged to think, develop a deeper interest in the course material, and feel more empowered to act on their new knowledge. He also examined the need to advance Internet usability and navigation for research and education, arguing that the time is ripe to develop a digital library for earth systems science education.
John Snow (University of Oklahoma) gave a frank overview of challenges in teaching undergraduate introductory courses. Snow asserted while large-enrollment survey courses are a necessary part of a university curriculum, they are also an important source of credit hour production. He argued that with a few exceptions, students learn relatively little in these courses. As a means to improve these courses Snow hypothesized that in many cases the target audience is too broad. These courses might be more successful if they were geared to specific groups with a common interest such as pre-service teachers, business majors or related science majors.
Susan Millar (University of Wisconsin-Madison) proposed that faculty can do a better job of engaging all students in the learning process, not by spending more time on teaching, but by changing assessment strategies. If the primary function of assessment is to rank students, the process could be improved with a change to “inquiry based assessment” that allows learning to continue during the assessment process. She referred the workshop participants to the Field-tested Learning Assessment Guide (FLAG): a treasury of inquiry-based assessment resources developed by the College Level National Institute for Science Education (NISE) Level One Team. The group's goal was to produce a rich collection of educational resources for continuous professional development of college Science, Mathematics, and Engineering (SMETE) instructors.
Terry Millar (University of Wisconsin-Madison) discussed an NSF-supported project that connected graduate students and faculty researchers with K-12 education-related professional development. This project addressed the needs of K-12 science classrooms that rely on cross-disciplinary project teams, that built on existing program strengths and university latency, that emphasized partnership, continuity and formative feedback, that were multidimensional across projects, and that provided professional development for graduate students.
Roberta Johnson (UCAR) discussed the importance of contributing to pre-college education as a way to combat the declining number of students selecting career paths in SMETE. Declining numbers may be a result of unfavorable profit/work ratios, lack of role models, cultural and gender effects, and bad experiences in early SMET education. Many teacher-training programs have not done an adequate job developing math and science educators at an early level. These programs often favor pedagogy and theory over practical skills using new teaching techniques, especially in technical subjects. It is also common that standard teacher training in math/science includes only survey courses, with limited exposure to in-depth content. In practice, new teachers may end up teaching courses for which they have little or no preparation. Johnson urged Workshop participants to become involved with teachers on a local, regional, or national level.
Dave Fulker (Unidata) discussed a vision for Unidata's future that is linked to NSF strategic plans, including the GEO 2000 planning document (http://www.geo.nsf.gov/adgeo/geo2000.htm). He quoted Tim Killeen in an early draft of GEO 2000 saying, “By 2010 we will have revolutionized science education, reducing the alienation from science of a large fraction of society, by deploying new interdisciplinary courses and developing curriculum materials that are accessible to and fascinating for science and non-science majors in the post-secondary setting and K-12 students, alike. These educational materials will provide rigorous and relevant scientific materials that address the Earth System, without flinching from the needed debates on complexity and uncertainty. [Students] will gain from inquiry-based teaching and learning and from in-depth evaluation and assessment of student outcomes.”
Fulker asked, “Does the Unidata community embrace this education agenda? Is the Unidata community ready and willing to play a leadership role in meeting this challenge? If yes (to either), what role should Unidata play, perhaps in collaboration with others such as the DLESE Program?” He noted that the Unidata community has demonstrated the capability to respond to challenges because it has had a remarkable history of leading-edge activities such as: forecasting as inquiry-based learning, enhancing education with computers, embracing the Internet, distributing data cooperatively, producing Web-based educational materials, defining a digital-library agenda and model, and collaborating in software design and testing.
Fulker noted that discussions are underway concerning a new Unidata mission to replace the current one, “Enabling universities to acquire and use atmospheric and related data”. Under consideration is the revised mission to “Provide infrastructure and software for use in academia to observe, understand, and share knowledge about Earth systems” or to “Catalyze Earth systems research and education, via data, software, and community.” Unidata goals corresponding to a revised mission might then include: 1) real-time, self-managing data flows, 2) software to visualize/analyze earth-system data, 3) distributed, organized collections of digital material, 4) computer-augmented discourse and discovery, and 5) support and engagement of diverse tool users.
Exemplary Approaches to Learning Meteorology
Perry Samson (University of Michigan) mixed humor with an excellent presentation on the rationale behind the creation of the Blue Skies CD-ROM, published with the textbook Meteorology Today. His love of teaching was apparent as he guided participants through a sampling of material on the CD, a natural extension of his efforts in the classroom. These resources can be remotely accessed and provide innovative materials for homework assignments, as well as an opportunity for students to view current weather when connected to the Internet.
Mohan Ramamurthy (University of Illinois) presented a strong case that a science literacy crisis exists in our country. He recommended that a substantial change be undertaken at the undergraduate level by teaching the connection between scientific principles and daily life. He argued that all students should “become scientists” to achieve this goal and speculated that it is possible to get students to conduct scientific inquiry using a Virtual Exploratorium concept.
Steven Ackerman (University of Wisconsin-Madison) presented a creative way to use the Internet in undergraduate education. The Internet can make teaching more effective if it is used to augment traditional resources, but if improperly used, it can make teaching less effective and more time consuming. For example adding course notes to a web page can take time, but adds little if to the learning experience. Instead, he has worked to develop on-line, inquiry-based, interactive teaching modules and self-assessment tests that allow students to be active participants in the learning process. He also shared his goals for a large-lecture introductory weather and climate course, a course not only covering basic meteorology principles, but also expanding the breadth of the science to show how meteorology is connected to all earth science systems. The excitement of science is an important component in Ackerman’s success as a teacher.
Weather and Beyond
Richard Clark (Millersville University) described a research project in which undergraduate students used Unidata-supported GEMPAK/GARP applications during an air quality field program. Students made on-site forecasts, evaluated platform safety, and learned decision-making skills while project leaders generated analysis, visualizations and interpreted archived data. The project measured urban air quality northeast of Philadelphia in July-August 1999. Undergraduate students used Raman LIDAR, tethered balloons, ground based particle/chemical samplers and airborne samplers onboard the DOE-G1 and University of Maryland research aircraft. Clark discussed how GEMPAK/GARP provided key analysis tools during the field study.
Fred Carr (University of Oklahoma) summarized the Collaborative Radar Acquisition Field Test (CRAFT) project, a national program for real-time acquisition of WSR-88D level II data that are the foundation of storm-scale research. NEXRAD Level II data are important in storm-scale numerical weather prediction because high-resolution models need high-resolution observational data for both initialization and verification. A plan is now in place for an online digital library in Norman, OK for data mining and research that will interface with other archives at NCAR.
James Steenburgh (University of Utah) showed how he uses mesoscale models to support his teaching style. His courses are based on principles that students should be active participants in the construction of understanding and knowledge, and that real-time weather analysis and forecasting fosters student and faculty interactions. Steenburgh uses real-time mesoscale models in his courses because they expose students to the strengths and weaknesses of next-generation NWP systems. They also provide a dynamically consistent, four-dimensional dataset for visualization of weather systems over the complex terrain of the inter-mountain West. Steenburgh concluded that real-time mesoscale modeling enhances undergraduate and graduate meteorological instruction, and that his teaching objectives are best met through real-time weather discussions and forecasting labs.
New Technologies and Resources
William Hibbard (University of Wisconsin) described the Visualization for Algorithm Development (VisAD) component library. Since most of today’s computers are connected to each other via the Internet and each may use a different operating system, communication between computers and standardization of software applications is a growing problem. The Java programming language is designed to overcome these problems through operating system independence. Hibbard explained that VisAD uses Java for its platform independence and to support data sharing and real-time collaboration among geographically distributed users.
VisAD is designed to support a wide variety of user interfaces ranging from simple data browser applets to complex applications that allow groups of scientists to collaboratively develop data analysis algorithms. Components can be built and assembled to create applications, map elements of data to different spatial and temporal dimensions to create animations, 3-D terrain-like representations, or multiple overlays.
Charles Murphy (Kean University of New Jersey) gave an overview of MetApps, a Java-based geoscientific information framework for developing data display and analysis tools. The goal of the MetApps team is to develop a set of turnkey, platform-independent Java-based tools for interactive meteorological analysis using network resources. Efforts will be made to enlist other developers in enhancing existing or developing new components. Murphy explained that a MetApps task force, representing Unidata community needs, is charged with determining the specification of application requirements, defining Use Cases, and testing the prototype tools. The MetApps that were available during the Workshop were a Surface Data Viewer, a 3-D Interactive Skew-T, a Satellite Image Viewer, and a Gridded Data Viewer.
Mary Marlino (DLESE), described the Digital Library for Earth System Education Program. Its goals are to help educators and learners find, evaluate, use and create resources that support active learning about the Earth system. The vision for the geoscience community is for a facility that provides easy access to high-quality instructional materials about the earth at all educational levels. It is viewed as a community center that fosters interaction, collaboration and sharing among earth science educators and learners. DLESE elements would include collections of high quality, peer reviewed instructional materials, and interfaces to large data sets about the earth. It would be a discovery system for a diverse population of users that would have both computer and human-mediated services. The operating principles are that this digital library would grow from the community and respond to community input and needs, thus building a federated, distributed effort. The library would also measure impacts of learning and teaching practices and engage users as contributors.
Summary of Lab Sessions
Presentation of Internet-Based Learning Materials in a University Setting
John Nielsen-Gammon (Texas A&M) described the use of web-based instructional modules, a team learning concept and teacher guidance to create an engaging and effective environment for teaching meteorology to undergraduates. Workshop participants played the role of students with pairs of participants working together at each computer workstation. Each team worked through a computer module entitled “Factors Affecting the Diurnal Cycle of Temperature: Clouds and Moisture”
Following the computer exercise each participant was given a quiz creatively designed to reinforce the concepts presented. It not only tested individual knowledge, but also allowed for team answers and included an appeals process for improving scores. John emphasized the benefit of having students work in teams of two to promote cooperative learning, alternating lecture with computer tasks to keep students engaged, and having adequate in-class teaching assistants to answer student questions and deal with technical problems.
Making and Using Java-Based Learning Activities
Tom Whittaker (University of Wisconsin) conducted a hands-on lab session based on QuizImage and VisitView. These are Java-based applications that can operate over the Web and are designed for building interactive learning exercises.
QuizImage builds quizzes and interactive exercises requiring identification of features appearing in images. Because of its use of maps, graphics, and imagery it has wide geoscience applicability. It is network enabled and fits nicely with multi-media presentations.
Workshop participants used a QuizImage exercise based on a theme of cloud feature identification using visible and infrared satellite imagery. Participants, worked in pairs and learned to make their own short lessons, in the form of html files that could be web-deliverable. They then shared their work with other participants for discussion and critique.
Whittaker also demonstrated the use of VisitView, an application used by the NWS for distance learning. VisitView allows the NWS to conduct interactive, high-resolution residence and teletraining sessions in a simulated AWIPS environment. VisitView's potential for use in a university setting was discussed with emphasis on it potential for collaborative learning.
Creating Learning Environments
Doug Yarger (Iowa State University), assisted by Dave Dempsey (San Francisco State University), demonstrated the use of a constructivist model in which students become active participants in their own learning. Yarger's exercises challenge students to “do what scientists do:” collect and organize data, make hypotheses, test them through experimentation, and apply the conclusions to real-life applications.
Participants worked through several learning modules based on fundamental concepts of short and longwave radiation, adiabatic processes, humidity variables, and cloud formation processes. One module asked participants to explore the process of how air is warmed during the day and to test several variables that influence air temperature. Another module demonstrated changes in temperature and vapor pressure of air as it ascends and descends a mountain.
Each exercise shared a common philosophy: a student learns best by observing physical data or processes at work, formulating hypotheses for why they work as they appear to, systematically altering individual observations and observing the reaction of the entire system, and drawing conclusions from responses they have observed.
Synoptic and Mesoscale Case Studies
Lynn McMurdie (University of Washington) and Jeff Weber (UCAR/Unidata demonstrated the use of GEMPAK/GARP applications to display and analyze classic storm systems contained in COMET cases studies. Her lab sessions allowed participants to complete case study exercises of the three-dimensional structure of two mid-latitude cyclones, including winds, jet streams fronts, cloud and precipitation structure and quasi-geostrophic forcing. Hand analysis of both surface and upper-air maps was used to compliment computer-based analysis.
Jeff Weber described the COMET case studies, a collaborative project with UCAR's Joint Office for Science Support (JOSS). The COMET case study website (www.joss.ucar.edu/cometCases) provides access to all datasets, plus links sources of other studies.
Participants in this lab session engaged in lively exchanges of ideas regarding the use of case studies. Most agreed that COMET case studies are a valuable teaching resource and have adapted them to meet the needs of their own teaching methods.
Severe Weather Teaching and Visualization Techniques
William Gallus (Iowa State University) challenged participants with a web-based training module on severe weather forecasting with each participant having the opportunity to complete a forecast. Gallus has designed both warm-season and cold-season severe weather components of the exercise, and each has three levels of complexity to match each student's level of expertise. Set as a true forecasting exercise, participants were presented with a chronological sequence of observations, to simulate how forecasters receive data during an actual operational shift. At each “hour” through the exercise, new information was made available, allowing the forecast decision-making process to evolve. Questions were also asked during the forecast process. The forecasting module then stepped forward to “verification time” and presented the actual weather, allowing participants to critique their forecast.
Gallus also presented a video of a virtual tornadic supercell thunderstorm developed in the virtual reality Convective storm Automated Virtual Environment (CAVE) at Iowa State University. CAVE is a small viewing booth in which a person watches a 3-dimensional computer simulation of a supercell thunderstorm on the Iowa plains. The viewer is allowed to navigate around the storm, using pan and zoom features and to study the storm's evolution and structure.
Best of the Web
Ben Domenico (UCAR/Unidata) and Mel Shapiro (NOAA/ATL) assembled a collection of “Best of the Web” websites in the geosciences, based mostly on nominations received from workshop participants before and during the Workshop. Participants were asked to identify websites they find most useful to support their educational or research efforts. Participants then spent their lab time exploring these web sites.
Shapiro reaffirmed his commitment to traditional hand analysis tools including pencil, eraser, feather duster, and protractor and chided “PC (point and click) meteorologists” who are overly dependent on computer analyses and forecasts. He then presented the NOAA/Climate Diagnostic Center's highly versatile and interactive website. He described the breadth of historical datasets available and the flexibility of the Java-based website for data access.
MetApps: Building Tools for the Future
The MetApps project was previously described in Charles Murphy’s plenary talk. MetApps are Java-based, platform independent, data display tools, designed to run on a local computer or remotely, across the Web, and can use data that may exist locally or on remote servers. Murphy, Dempsey, and Don Murray (UCAR/Unidata) ran concurrent lab sessions demonstrating three MetApps prototypes: Satellite Image Viewer, Gridded Data Viewer, and an interactive 3-D Skew-T. Following an introduction of each application, participants worked through exercises that demonstrated the features of the application. This was as much an opportunity for the MetApps developers to get feedback from potential users, as it was to showcase the MetApps themselves.
Focus Groups
On the fifth and final day of the Workshop, participants divided into five focus groups to discuss a charge from Unidata Director, Dave Fulker. Fulker outlined strategies for Unidata's future that might help it enable the Unidata community to respond to the educational, technical, and scientific challenges it faces. He asked participants to respond to the following questions about the Unidata’s future and that of its community. (Fulker, Ben Domenico, Charles Murphy, Doug Yarger and Anthony Rockwood led the focus groups.)
Many participants endorsed both NSF's integrated earth systems research and education agenda and the expansion of the Unidata community to support collaborations with other geoscience disciplines. However, even many of the most enthusiastic proponents of Unidata expansion envision the atmospheric sciences community continuing as its core community.
Some urged that Unidata expand assertively by soliciting funds to develop tools for other geoscience communities and to foster cross-disciplinary collaborations. Others suggested a more conservative approach, in which Unidata would focus first on developing tools generic and powerful enough to serve multiple disciplines, then let other communities attracted by the tools come to Unidata for support.
Fulker expressed Unidata's high comfort level with this approach because tool development is and has been Unidata's greatest strength. Consistent with this less aggressive strategy, some suggested that Unidata should support members of the community who establish cross-disciplinary collaborations but not actively initiate or promote such collaborations. Some were concerned that by expanding into other disciplines and by dedicating resources to develop new, more generic tools, Unidata's support for the atmospheric sciences community, particularly its support for mature software packages such as GEMPAK/GARP and MCIDAS, would be threatened. Others worried that Unidata's success has derived in significant measure from its focus on the atmospheric science community, and expansion to serve a larger community would dilute that focus and endanger Unidata's institutional effectiveness.
Several participants noted the value of collaborations between Unidata and organizations such as DLESE, which can help ensure that Unidata develops tools that best serve their pedagogical purposes.
Some participants urged that the Unidata community commit to support community colleges and the primary and secondary education (K-12) community, which constitutes the source of future scientists. However, most agreed that, rather than trying to support these large communities directly, Unidata should support members of the university geoscience community who in turn support these non-university education communities. An intermediate strategy could be to work with UCAR to acquire block funding for minigrants to those engaged in K-14 outreach.
While acknowledging that face-to-face interactions are sometimes essential, some participants encouraged Unidata to develop collaborative communication tools in addition to its data distribution, analysis and display tools. However, before the potential of these communication tools could be recognized, Unidata would probably need to support demonstration projects in the community.
The bottom line for expansion of Unidata's mission is funding. Although some argued that expansion could pay for itself by creating new funding sources, others cautioned that Unidata should not attempt to expand until NSF, consistent with its own rhetoric, commits the funds for it.
By the end of the Workshop participants agreed to draft the following resolution to NSF:
Unidata's mission since its inception has been “enabling universities to acquire and use atmospheric related data”, but that the needs and opportunities of the Unidata community continue to change. A common theme among Workshop speakers was the changing educational, technical and scientific environment. Science education has begun to shift away from traditional, teacher-centered classroom methods and toward student-centered, constructivist, inquiry-based approaches. Information technologies continue to improve, making better data access, analysis, and display modes possible. Research within the geosciences is shifting toward a more integrated, interdisciplinary study of earth systems.
Summary
The Unidata Summer Workshop-2000 proved to be a highly successful and enjoyable experience. Keynote speakers challenged the participants to be creative and innovative in the educating and developing future scientists. Guest teachers led participants in thought-provoking and stimulating labs sessions, encouraging them to expand the use of new teaching methods within the increasing diversity of our earth science curricula. Ongoing interactions between colleagues, new friends, respected leaders and Unidata experts provided us all with a fresh perspective on the responsibility we have to our science.
While the Workshop clearly demonstrated the extraordinary resource that is Unidata, it was also a unique opportunity for Unidata to come face-to-face with its users and receive constructive criticism that will help it remain strong in the future. Unidata has always been highly responsive to changing technology, data types, applications of data, and the needs of the community it serves. A certainty for the future is that it will be filled with challenges, both known and unforeseen, and Unidata seems ready and willing to meet those challenges.
References
President's Information Technology Advisory Committee, 1999: Information Technology Research: Investing in Our Future. Report to the President. February 24, 1999. Available at http://www.ccic.gov/ac/report/pitac_report.pdf.
National Science Board, 2000: Environmental Science and Engineering for the 21st Century: The Role of the National Science Foundation. February 2000. Available at http://www.nsf.gov/pubs/2000/nsb0022/reports/nsb0022.pdf
NSB Task Force on the Environment-Interim Report-July 1999. NSB 99-133. 80 pp. Available at http://www.nsf.gov/nsb/tfe/nsb99133/start.htm.
National Science Foundation, 2000: NSF Geosciences Beyond 2000: Understanding and Predicting Earth's Environment and Habitability. Available at http://www.nsf.gov/pubs/2000/nsf0028/nsf0028.txt
National Research Council, 1999: Our Common Journey: A Transition Toward Sustainability. National Academy Press, 384 pp
Wetzel, M., D. Dempsey, S.K. Nilsson, M. Ramamurthy, S. Koch, J. Moody, D. Knight, C. Murphy, D. Fulker, M. Marlino, M. Morgan, D. Yarger, D. Vietor, and G. Cox, 1998. Faculty Workshop on Using Instructional Data for College Level Education in the Atmospheric and Earth Sciences. Bulletin of the American Meteorological Society, 79, 2153-2160.