06 February, 2007

Stalagmite records individual storms and intensities

In last months’ post about climate records from stalagmites and other speleothems, I concentrated on their potential for giving us detailed regional climate information over long time periods: 10s, and even 100s, of thousands of years. However, this article at Mongabay.com highlights some research concerned with records preserving climatic variations over much shorter timescales, and one of the more contentious issues in modern climate research: how the frequency and intensity of hurricanes is affected by our warming climate.

It seems that the towering storm clouds and humid atmosphere associated with hurricanes and other tropical cyclones produce rainwater which is extremely light isotopically (it has &delta18O 6 per mil more negative than normal precipitation), and a lot of it falls in a short time when a storm hits land. Amy Frappier and her colleagues decided to test if this pulse of storm water could affect the &delta18O of the groundwater enough to leave a signal in speleothems growing in underlying caves.

We used a computer-controlled dental drill to carefully mill off layers of powder from a fast-growing stalagmite [from Actun Tunichil Muknal caves, Belize], where each sample reflects cave drip water over periods of a week to a month. Analyzing these rock powders using standard techniques, we were able to detect brief spikes from recent hurricanes and tropical storms that produced rain over the cave - even when those storms struck only weeks apart!"

The effect is clearly illustrated by this figure from a paper Frappier et al. have just published in Geology [1]. Tropical cyclones which have hit Belize in the last 30 years (A) correlate well with short negative excursions in &delta18O (B). These excursions are superimposed on a longer-term pattern related to the El Nino Southern Oscillation (the 15 month offset with the El Nino record in C represents the time it takes rainwater to percolate from the surface to the cave). The accompanying carbon isotope record also shows the El Nino variation, probably due to changes in soil respiration rates, but not the storm events, showing that they are due to a separate forcing (heavy storm precipation).


The final subfigure D is interesting, as it indicates that stronger storms (plotted in A) seem to produce larger excursions. Going back to the Mongabay piece:
"We also found that the relative size of spikes that we measured was related to the intensity of the storm, which is encouraging for the prospect of reconstructing the intensities of pre-historic landfalling storms.."

There’s currently a lot of debate over possible changes in storm numbers and/or intensity as a result of anthropogenic climate change (RealClimate has a good summary of the science) – some people claim there is already a clear signal, but others say that the historical data just isn’t good enough. Speleothem records can’t help resolve one of the major issues, the detection of non-landfalling storms, but it seems that they can potentially give us a much longer term record of trends in tropical cyclone intensity than is currently available; and a better idea of how much natural variability there was in the past can only help.

[1] Geology 35, p111-114 (doi).

This post was published on Day 2 of the Just Science challenge – a full week of science and only science. You can subscribe to the RSS feed at http://www.justscience.net/?feed=rss2.
All my posts for this week

No comments: