Extracting trends and correlations from noisy environmental data

Add to your list(s) Download to your calendar using vCal

• John Wettlaufer, Oxford
• Thursday 01 May 2014, 11:30-12:30
• Open Plan Area, BP Institute, Madingley Rise CB3 0EZ.

If you have a question about this talk, please contact Catherine Pearson.

Typically when we observe time series of what we believe is a representative parameter in a system a key question concerns how well correlated such data are in time. For example, how likely is the rainfall this month to be the same as that next month or next year. Here, I discuss the long-term correlations and properties of daily satellite retrievals of Arctic sea ice albedo and extent taken over the last three decades. The interpretation harnesses a recent development called multi fractal temporally weighted detrended fluctuation analysis, which exploits the intuition that points closer in time are more likely to be related than distant points. The method goes beyond treatments that assume a single decay scale process, such as a first-order autoregression, which cannot be justifiably fitted to these observations despite the commonality of doing so. It is found that long-term persistence is re-entrant beyond the seasonal scale and that on the seasonal scale the system is governed by white noise.

This talk is part of the Institute for Energy and Environmental Flows (IEEF) series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge Energy Seminars
• Department of Earth Sciences seminars
• Institute for Energy and Environmental Flows (IEEF)
• NanoDTC Energy Materials Talks
• Open Plan Area, BP Institute, Madingley Rise CB3 0EZ
• history
• ps635

Note that ex-directory lists are not shown.