Surf Science: How Big Were Waves in 1840?

Tony Butt

by on

Updated 179d ago

What are the chances that Mundaka was pumping when Christopher Columbus set sail in 1492? Or that Mullaghmore was going off during the Easter Rising of 1916?

In a previous article (see HERE) I talked about the North Atlantic Oscillation (NAO). The NAO is a climatic cycle that switches between a ‘fluid’ phase that gives us a constant stream of low pressures, and ‘blocking’ phase that gives us a large, slow-moving anticyclone.

Example of an anticyclone.

Example of an anticyclone.

A parameter called the NAO index (NAOi) tells us how stormy the North Atlantic is at any given time. And, of course, a stormy North Atlantic means big surf, so the NAOi can also tell us something about wave heights. This could be really useful if we wanted to get an idea of how wave heights change over really long time periods.

First we would need to find out how well the NAOi actually mimics wave heights in the North Atlantic. Scientists from the Southampton Oceanography Centre have done this by comparing wave-height data with NAOi values over the period 1991 to 2000, to see how well correlated they were. In general, very good correlations were found for northwest-facing coastlines, meaning that the NAO index can be used as a proxy for wave heights, at least in these areas.

Now, by looking at the long-term history of the NAOi, we can get an idea of the long-term fluctuations of North Atlantic wave heights. I found some archives of NAOi values from 1820 to 2000, so I plotted them to have a look at the ups and downs.

The NAO index from 1820 to 2000.

The NAO index from 1820 to 2000.

A few notable features can be seen, which some of us might remember. For example, you can see a peak around the early 1990s, which did indeed coincide with several stormy winters and big surf.

But 184 years is still a relatively small amount of data, and they don’t let you see the longer-term cycles within which the ups and downs might be buried. To resolve those cycles and get a proper idea of how the North Atlantic behaves, we would need NAOi values going back several hundreds of years.

Unfortunately, we don’t have direct NAOi data going that far. However, there are a few other parameters that can be used to infer NAOi values. We would have to be careful not to try to be too precise, but it nevertheless might still give us a few clues about the long-term behaviour of North Atlantic wave heights.

The two most commonly used parameters are tree rings and ice cores. Scientists from the University of Arizona have reconstructed NAOi values from 1429 to 1983 using tree-ring chronologies from Morocco and Finland combined with ice cores from Greenland. The tree rings were used to indicate changes in rainfall, and the ice cores to indicate changes in temperature, both of which are closely connected to the NAO. The time series, going back 555 years, now long enough to seriously examine those embedded cycles.

The best way to visualize this is to plot a spectrum from the time series, which you can see below. You can see that there are peaks at 4, 8, 13, 20, 30 and 55-year periods, with an extra-large peak corresponding to periods of between 20 and 30 years. In other words, the strongest embedded cycle in that 555-year time series is around 20 to 30 years long. Conclusion: particular features in the time series are more likely to repeat themselves every 20 to 30 years than at any other interval.

So, without sticking our necks out and trying to be too precise, we can say that particular features in the history of North Atlantic wave heights, such as the peak around the early 1990s, are most likely to repeat themselves at intervals of about 20 to 30 years.

Spectrum of inferred NAO index, derived from data going back 555 years.  Don’t worry about the units – the most important feature is the peaks, showing the length of the most important cycles.

Spectrum of inferred NAO index, derived from data going back 555 years. Don’t worry about the units – the most important feature is the peaks, showing the length of the most important cycles.