Strides, steps and stumbles in the march of the seasons
Brian E. Mapes, Nikolaus Buenning, In-Sik Kang, George N.
Kiladis, Richard B. Neale, David M. Schultz, Klaus M. Weickmann
ABSTRACT
As the Earth orbits the sun, sunlight at the top of the atmosphere
varies in a smooth repeating pattern that is almost entirely composed
of annual and semiannual Fourier harmonics in time. The Earth system's
response is mainly smooth, with large linear responses to the forcing,
but also contains fast or high frequency (HF) changes. HF features of
climatology must spring from non-linear mechanisms, which are key
challenges for our understanding and models of climate dynamics. To
highlight HF climate features, a consistent objective signal
processing algorithm has been applied to 365-day climatological time
series of many variables at all locations in many datasets, both
observational and simulated. Because these HF features must be
understood
within the context of the total climatology, an
interactive software package, the Annual Cycle Explorer (ACE), was
written to automate processing and display the results. ACE is offered
free on CD or DVD for popular computer platforms, loaded with several
datasets and a predefined menu of settings to reproduce figures in
this article along with other interesting findings. This article
shows a sample of interesting seasonal phenomena, from simple to
subtler, including: 1. The annual harmonic of near-surface
temperature, which already poses some challenges for climate system
models; 2. The semiannual mid-winter minimum of storms in the Northern
Hemisphere, related to nonlinear jet-stream variations; 3. Rapid
(HF) onsets of summer rainy seasons; 4. Mid-summer rainfall minima in
western parts of subtropical ocean basins; and 5. Historically
intrigueing but statistically weak evidence of an oscillation in late
northern winter, related to January Thaw folklore in North America.
BAMS, proposed and encouraged 2003, submitted June 2004, revised
April 2005, refused October 2005. Watch for it in MWR most likely.