Exploration of Climate Data Using Interactive Visualization
Abstract
In atmospheric and climate research, the increasing amount of data available from climate models and observations provides new challenges for data analysis. We present interactive visual exploration as an innovative approach to handle large datasets. Visual exploration does not require any previous knowledge about the data as is usually the case with classical statistics. It facilitatesiterative and interactive browsing of the parameter space in order to quickly understand the data characteristics, to identify deficiencies, to easily focus on interesting features, and to come up with new hypotheses about the data. These properties extend the common statistical treatment of data, and provide a fundamentally different approach. We demonstrate the potential of this technology by exploring atmospheric climate data from different sources including reanalysis datasets, climate models, and radio occultation satellite data. Results are compared to those from classical statistics revealing the complementary advantages of visual exploration. Combining both, the analytical precision of classical statistics and the holistic power of interactive visual exploration, the usual work flow of studying climate data can be enhanced.
F. Ladstädter, A. K. Steiner, B. C. Lackner, B. Pirscher, G. Kirchengast, J. Kehrer, H. Hauser, P. Muigg, and H. Doleisch, "Exploration of Climate Data Using Interactive Visualization," Journal of Atmospheric and Oceanic Technology, vol. 27, iss. 4, p. 667–679, 2010.
[BibTeX]
In atmospheric and climate research, the increasing amount of data available from climate models and observations provides new challenges for data analysis. We present interactive visual exploration as an innovative approach to handle large datasets. Visual exploration does not require any previous knowledge about the data as is usually the case with classical statistics. It facilitatesiterative and interactive browsing of the parameter space in order to quickly understand the data characteristics, to identify deficiencies, to easily focus on interesting features, and to come up with new hypotheses about the data. These properties extend the common statistical treatment of data, and provide a fundamentally different approach. We demonstrate the potential of this technology by exploring atmospheric climate data from different sources including reanalysis datasets, climate models, and radio occultation satellite data. Results are compared to those from classical statistics revealing the complementary advantages of visual exploration. Combining both, the analytical precision of classical statistics and the holistic power of interactive visual exploration, the usual work flow of studying climate data can be enhanced.
@ARTICLE {ladstaedter10explorationClimateData,
author = "Florian Ladst{\"a}dter and Andrea K. Steiner and Bettina C. Lackner and Barbara Pirscher and Gottfried Kirchengast and Johannes Kehrer and Helwig Hauser and Philipp Muigg and Helmut Doleisch",
title = "Exploration of Climate Data Using Interactive Visualization",
journal = "Journal of Atmospheric and Oceanic Technology",
year = "2010",
volume = "27",
number = "4",
pages = "667--679",
month = "April",
abstract = "In atmospheric and climate research, the increasing amount of data available from climate models and observations provides new challenges for data analysis. We present interactive visual exploration as an innovative approach to handle large datasets. Visual exploration does not require any previous knowledge about the data as is usually the case with classical statistics. It facilitatesiterative and interactive browsing of the parameter space in order to quickly understand the data characteristics, to identify deficiencies, to easily focus on interesting features, and to come up with new hypotheses about the data. These properties extend the common statistical treatment of data, and provide a fundamentally different approach. We demonstrate the potential of this technology by exploring atmospheric climate data from different sources including reanalysis datasets, climate models, and radio occultation satellite data. Results are compared to those from classical statistics revealing the complementary advantages of visual exploration. Combining both, the analytical precision of classical statistics and the holistic power of interactive visual exploration, the usual work flow of studying climate data can be enhanced.",
images = "images/ladstaedter10exploration.jpg, images/ladstaedter10exploration1.jpg, images/ladstaedter10exploration3.jpg, images/ladstaedter10exploration2.jpg",
thumbnails = "images/ladstaedter10exploration_thumb.jpg, images/ladstaedter10exploration1_thumb.jpg, images/ladstaedter10exploration3_thumb.jpg, images/ladstaedter10exploration2_thumb.jpg",
url = "//dx.doi.org/10.1175/2009JTECHA1374.1"
}