Visual Climatology: Animated Presentations of
Seasonal and Interannual Climate Variations
Patrick J. Bartlein, Department of Geography, University of Oregon
Earth's climate system, and its individual components, vary across a range of temporal and spatial scales. Explaining this variation (as well as the covariation among the components) is the principal focus of courses in climatology in specific, and physical geography in general. Some of the components of the climate system are quite abstract and difficult to envision, and variability is a property of physical systems that is naturally hard to portray using static displays. Consequently, we have begun to develop ways of showing how the climate system operates using animated maps created with data from the NCEP/NCAR Reanalysis Project, a new and rich source of data for a large number of climate variables.
An example of an animated map. For more examples, see Global Climate Animations
 [ click on image to enlarge ] |
The Seasonal
Cycle of Air Temperature This image shows the month-to-month variations of air temperature around the globe, constructed using the "reanalysis data" sets described below. The broadscale patterns of air temperature that appear on the image result from the the latitudinal and seasonal variations in the receipt of solar radiation, the differential heating and cooling of the continents, and the continental- and regional-scale influences of atmospheric circulation. |
Much of the research in climatology involves the subjective analysis of maps and graphs, supported by statistical analysis, physical reasoning, and both numerical and conceptual models. The project proposed here is aimed at refining and expanding this approach for making climatology a less abstract and more visual subject, and will introduce to undergraduates the data sources and tasks actually used in climatological analysis.
The objectives of the proposed project are:
Background
How the climate system works, and how and why it varies are concepts that are important components of the material that every environmental and earth scientist should know. With the prospect that human actions are now changing climate (Intergovernmental Panel on Climate Change, Working Group I, Summary for Policymakers), a basic understanding of Earth's climate is probably something that every well educated citizen should also now possess. Unfortunately, except for a handful of variables (temperature, precipitation, cloudiness/sunshine, wind speed and direction, air pressure and humidity), climate variables are abstract in nature, and mostly unobservable across the globe. This makes climatology, despite its importance, a field of study that is difficult for students at all levels to "get a handle on." Fortunately, a new source of climate data -- "renanalysis data" -- has recently come online, and has the potential to remedy this situation. These data sets include hundreds of variables, are global in extent, and are available for the past 40 years (and soon the past 50).
Our NSF-sponsored research (TEMPO--Testing Earth-system Models using Paleoenvironmental Observations, recently renewed for the next four years) focuses on understanding the variations of the climate system from the last glacial maximum (about 20,000 years ago) to the present, an interval where nature has performed an experiment with the climate system (by changing its external controls) that is similar in magnitude (but not in kind) to the experiment now being performed by humans. In that research, we use a combination of climate-model simulations and syntheses of paleoclimatic data (see Paleoclimate simulations for North America over the past 21,000 years) to "diagnose" the past climatic variations (i.e. to understand the mechanisms responsible for creating the climatic changes recorded by the paleo data, and to examine the interactions among components of the climate system). The climate models we use are the same ones that are being used to project the response of the climate system to human modification of the composition of the atmosphere. The potential of these models to correctly simulate climate under conditions different from those at present serves as a strong test of their abilities to predict the future, and this test can be performed using the paleoclimatic data.
We use the NCEP/NCAR Reanalysis Data set in order to understand how the present climate operates in different regions. In particular, we look at the way in which variations in regional surface water- and energy-balance components are governed by global and hemispheric controls, in order to provide a perspective for evaluating the "paleo" simulations. We also use that data set, along with other similar ones, to understand how large-scale atmospheric circulation patterns control surface climatic variations at spatial scales not currently resolvable in global climate models (see Atmospheric circulation patterns and spatial climatic variations in Beringia).
Animated Climate Maps
The prospect of using the reanalysis data to "see" the variations of a number of interesting climate variables encouraged us to put together some simple animations of the seasonal cycle of some basic climate variables for GEOG. 321, Climatology, (offered each year during the fall quarter). These animations can be considered as a pilot project for the work proposed here.
The animations were put together for GEOG. 321 by Jacqueline J. Shinker, a Master's student in our department who is beginning work on her Ph.D., and who was using the NCEP/NCAR Reanalysis Project data set in her thesis on the controls of floods and droughts in the mid-continent of North America. She was assisted by Jason Rahn, an undergraduate working on the TEMPO project.
These are relatively unsophisticated animations. The maps originate as PostScript files intended for paper-copy production, and so the colors and linework are not really optimized for screen display. The animations were produced as animated .gif files, which have the advantage of being viewable on most any browser, but have the disadvantage of not being startable and stopable. From both my own and the student's perspective (judging from their written comments), the animations were successful in getting some rather complicated concepts across (in particular the relationship between pressure patterns and wind fields).
Validation of the animated-map approach
We also have strong external validation for this approach of using animated maps constructed with reanalysis data for instructional purposes. JJ Shinker created a web page that collected the animations in one place (Global Climate Animations), submitted this to the National Geographic Society's 18th Annual National Geographic Award in Cartography competition, and won. The NGS competition is the premiere student-oriented competition in geography, and is judged by a committee of cartographers from the NGS and members of the Cartography Specialty Group of the Association of American Geographers (AAG), the principal association of academic geographers in the U.S. I think that this award provides independent evidence of the novelty of our approach, and of the quality of what was essentially a first cut
Objectives and Tasks of the Proposed Project
Our objectives in the proposed work are:
Note that we are not proposing to develop ways for the students to produce the animations themselves, or to analyze the data directly. Although this is feasible, the current technology is immature (but improving), and we will explore the prospects of extending the approach to allow more direct student interaction with the data. In the meantime, however, there will more than enough information in the pre-packaged animations we plan to create to saturate the ability of students to absorb that information, and I fell that placing the additional burden of creating the animations on the students, or conducting "toy" analyses, will result in less overall information transfer.
Some of the specific tasks that must be completed to achieve the objectives are:
The expansion of the project to include interannual variations of climate naturally expands the time required for production of the animations. While the seasonal cycle of one or more climate variables can be illustrated by an animation with only 12 frames, illustrating the month-to-month variations over a 40-year interval requires the generation of 480 frames, and conquently production time will increase.
In addition to these technical tasks, an important task will be the development of the online exercises. I currently have experience with assigning exercises supported by online materials using two different approaches:
The exercises that will be developed here will, depending on the course level, represent some kind of blend of these two approaches. For the introductory courses (see below), a paper-copy exercise form will probably work best. (It's been my experience in upper-division courses that only the more technically sophisticated students can easily manage using a browser to both view material and respond to questions on an HTML form.) For these courses, the exercises will focus on having the student find and describe some of the basic features that appear on animations of the seasonal cycles of the climate variables.
For 300-level courses and above, the exercises will be more open ended, and some will be structured to require (and facilitate) group analysis--there are a number of ways to partition the variables and data (e.g. by continent, by season, by mechanism (e.g. surface water balance, surface energy balance)) that would naturally allow parts of a general analysis to be assigned to members of a group.
I plan to evaluate how well the animations and exercises work, using an exam-time questionnaire. I have found that this works well, and I have based modifications of the structure of my courses on these results. For example, the results of a questionnaire given to GEOG 102, Global Environmental Change, students led me switch from a transparency-and-overhead projector approach in GEOG 102 and 321, to the current lecture-from-the-web approach.
The courses in the geography department that the resulting animations and accompanying exercises will contribute to in a fundamental way include:
In addition, there are several other courses, including those in both physical and human geography, as well as some in other departments, that could use selected animations. For example, animations that illustrate the global hydrological cycle could find application the following courses:
The general notion of animating sequences of images to illustrate aspects of environmental change probably has wide applicability. For example, in GEOG 102, Global Environmental Change, several quick-and-dirty animations were put together to illustrate land use/land cover changes and long-term vegetation change, and these could likely be used in courses that focus on human-induced environmental change. Some examples of those kinds of animations include:
Many disciplines have data sets or materials that are sequential in nature, and the general concepts and procedures that will be refined in the proposed work could potentially have wide applicability in those disciplines.
Personnel and Their Contribution to the Project
The current set of animations were produced with equipment purchased through NSF funding, and the graduate and undergraduate student salaries were also paid from those sources. Production of these maps gives a good idea of the support required for the proposed work.
The major portion of the requested funds will apply for supporting a graduate student who will do most of the data retrieval, processing, and mapping. The graduate student will also supervise the undergraduate student, who will tasked with some of the more straightforward production work. The production of the animations is somewhat time consuming--it takes about 2 hours on a 266 MHz Pentium II machine to create and "distill" a set of 12 maps for a seasonal cycle animation, and about 30 minutes to open the resulting files and produce the animated .gif file. Animation of the interannual variations of a single variable will require the creation of 480 frames, and so the production of longer animations will require additional computing resources.
One of the limitations to using the animations in class is a mechanical one. The projector I currently use is an older one, and is rated at 230 lumens. Consequently, the room lights must be dimmed or extinguished, and window shades lowered for the projected images to be viewable. There is a tradeoff between the resolution and brightness of the projected image (at constant lamp luminosity). My experience suggests that SVGA resolution sufficiently high, and that image brightness should be optimized for an effective presentation, and the specifications of the projector were selected with that in mind.
| Graduate Student (GTF III, .49, 3 terms + summer; salary $12,982, OPE $577) | $13,559 |
| Undergraduate Student (Student Asst. 6, 3 terms + summer, 10 hours/week; salary $4,100, OPE $205) | $4,305 |
| Computer Projector (InFocus LP425Z, SVGA, 1000 lumens) | 3,999 |
| Computer (Dell Precision 410, 450 MHz Pentium III) | 4,123 |
| Total Request | $25,986 |
Letter of support from unit head
June 3, 1999
To Whom It May Concern:
I am writing to express my strongest possible support for Professor Bartlein’s proposal on "Visual Climatology," which he is submitting in response to the RFP to Transform Research into a Teaching Product. For some time Bart has been at the forefront of efforts to integrate research into undergraduate teaching, developing a set of web-based materials that have become an integral part of a sequence of courses in physical geography and climatology. Funding Bart’s proposal would allow him to take this initiative to the next stage, allowing students to use and analyze climate data and related map animations in the relevant physical geography and climatology courses.
The research that Bart himself is doing, and the work he has already put into web-based teaching materials, means that he can do much more than simply present students with idealized data. Instead, he can incorporate real "research-grade" data into the courses he is teaching, giving students the opportunity to work on the types of information and maps that are being used by research climatologists. Evidence of the value of this approach comes from the success of one of our graduate students—J. J. Shinker—who has worked on developing the kinds of map animations discussed in Bart’s proposal. She won a major National Geographic Society award for her work.
If Bart’s proposal is funded, I think it could open the door to other ways of bringing research into the undergraduate geography curriculum. Hence, I fully endorse his proposal and as Department Head I will do everything I can to facilitate his efforts if it is funded.
Sincerely,
Alexander B. Murphy
Professor, Department Head, and
Rippey Chair in Liberal Arts and Sciences
