Bringing new functionality to data visualization using Google Earth




		This image shows preliminary comparisons of WRF-CHEM model outputs to station measurements of O₃ near Mexico City, Mexico. This image allows a researcher or student to visually compare concentrations of a pollutant to the values produced by a model, and to track this comparison over time. Metadata that describes the station measurement and other station information can be accessed by clicking on the station marker. This visualization technique is significant because it easily allows visual comparisons of scientific datasets to be made in a rich geographic context. (See larger image.)

We are developing a set of tools that will exploit the capabilities of commercial "virtual globes" software, in particular the Google Earth product (which is freely available to anyone). Using a variety of information types (gridded three-dimensional model results, surface observations, and aircraft-based observations), we are writing software to provide scientists with additional information about the overall structure of the chemical conditions at the time and location of the observations. Our goal is to explore the feasibility and usefulness of this approach for helping scientists refine their models. The tool will be extended to provide visualizations in a 3D regime, such as comparing the predictions of model outputs to an aircraft track or a dropsonde.

Because these visualizations use readily available commercial tools, they can be easily distributed to scientific colleagues, educational institutions, and the public. We anticipate that schools will be interested in downloading these files with an easy interface and useful information about scientifically stimulating data.

This project supports NCAR's strategic plan in several respects. It is "Developing and providing advanced services and tools" and "Creating an Earth system knowledge environment" while "Engaging a broader and more diverse community in the atmospheric and geosciences" and "Supporting and enhancing formal science education at all levels."

In FY2007, we began leveraging an existing NCAR visualization tool, the NCAR Command Language (NCL), to create the visualizations. Existing scientific codes that perform complex operations on data are being leveraged to create visualizations that are then placed in the Google Earth context. NCL also can be used to perform data extraction and color mapping so that the tools we are building will also be able to access and visualize a wide range of information available from the NCAR Community Data Portal in a more generalized fashion.

In FY2008, this project will complete the study of use cases by finalizing software that supports automated extraction of data, visualizations, and XML descriptions that will interface to Google Earth software. We anticipate iterations with scientists to refine the products. These products will be distributed via the Community Data Portal to interested parties. We will also provide an assessment of the scientific usefulness for the evaluation of models. If time permits, this functionality will be extended into a more generalized web interface.

This work, performed in collaboration with scientists from NCAR's Atmospheric Chemistry Division, is being funded under the FY2007 NCAR Director's Opportunity fund, which is derived from NSF Core funds.