University of Toronto professor Steve Easterbrook has issued a research paper in which he outlines how software engineers can help climate scientists build computer models for projecting climate change. He notes that climate scientists with little or no training in software engineering often build the data handling and analysis tools for processing raw observational data and the results of simulation runs, as well as the tools for sharing their data with other scientists.
Easterbrook says the best way for software engineers to help climate scientists is to develop tools for handling the massive datasets and earth-system models needed to simulate a changing climate.
View a video of showing an experimental run of a climate model whose resolution approaches that of a supercomputer-based weather model.
Easterbrook notes that climate models are run on customized and constantly evolving supercomputers, which makes it almost impossible to reproduce individual experiments. Moreover, the data is processed in so many different ways that adequate metadata is needed to allow scientists not involved in experiments to meaningfully engage with the data. Easterbrook also says visualization could be used to educate other scientists, policymakers, activists, journalists, and ordinary citizens about climate change.
From Technology Review
View Full Article
Abstracts Copyright © 2010 Information Inc., Bethesda, Maryland, USA
No entries found