Sediments tell about hurricanes

How geologists, historians and statisticians decipher centuries-old weather extremes

Clipping lake sediment core Lower Mystic Lake, Boston. Light-dark annual shifts, hurricane event 1706 recognizable by the location of coarse material. Vertical length about 2.3 cm. © Mark Besonen, University of Massachusetts Amherst, USA
Read out

Whether floods in England, hurricanes in the US or forest fires in Greece - extreme weather events occur worldwide with disastrous consequences again and again. But not only today, but also earlier there were such "natural disasters". Their traces can be meticulously reconstructed from geological and historical sources. Using mathematical calculation models, they provide the basis for climate risk analyzes, the prognosis of future damage.


"If you want to know, for example, how many hurricanes have occurred on the New England coast over the last few centuries, today's aircraft or satellite measurements are of little use, " explains Manfred Mudelsee, specialist for climate risk analyzes of the Institute of Meteorology at the University of Leipzig and owner of the company "Climate Risk Analysis ". Instead, one reaches for a drill and picks a core from the bottom of Lower Mystic Lake near Boston, Massachusetts. The sediments store information about such extreme weather events - you just have to read them. "

Analysis of sediment cores

By analyzing the lake sediment core, you can determine the time when a hurricane raged in this area. Even if it was already several centuries ago. "At that time, the lake was low in oxygen and the seasonal fluctuations in the sediment input remained undisturbed by burrowing creatures such as worms, " ​​explains Mudelsee. By counting the different deposits - light for sand and dark for biogenic origin - the scientists derive important information: time. A verification by means of an absolute dating method, such as the measurement of the radiocarbon, increases the reliability of the data.

"To find the historical hurricane events, it is best to search for layers of coarse gravel material using X-ray in the sediment core, " explains the Leipzig climate physicist. "Because if a hurricane uprooted over Boston trees, it will transport coarse material into the lake and deposit it there - the event is thus documented. This is a typical example of a proxy measurement in which one uses a proxy for what is not directly observable. "Display

The verification of such data takes place on the one hand by other sediment cores, on the other hand by historical information. These are enough for Boston at least until the year 1630, the beginnings of colonial colonization, back. In this way, geoscientists in Amherst, Massachusetts, along with M. Mudelsee, were able to generate a time series of hurricane events over the past 1, 000 years. Their publication is currently in progress.

Th ringer Flood of 1613

The Th ringer Sintlut of May 1613. From: Hellmann, G. (1913) Publications of the Royal Prussian Meteorological Institute 256: 21-57. Copyright extinguished

Pal o weather extremes can also be deciphered with the help of historical records: manuscripts from monasteries, town chronicles, requests from flood victims to the authorities for financial help or pamphlets, the killing of the disaster. So also from the Th ringer Sintflut, which brought on May 29, 1613 Weimar and other parts of the country near the destruction. First of all, in this case, it was a matter of circumnavigating the Time Kl cliff, Mudelsee explains. The date of the documents is still based on an old calendar system. The acquisition of the new Gregorian calendar, actually introduced in 1582, happened later in Thuringia. Correspondingly, the data had to be corrected by ten days

Another problem: old documents are getting worse compared to newer ones; they are lost, for example, by the turmoil of war, before they can be held in further, secondary sources. In retrospect, this effect would suggest a dramatic increase in flood risk in the second millennium, "explains Mudelsee. Therefore, as many sources as possible must be evaluated, which document the flood events along a single river.

The researchers compare meteorological information such as "It rained for two weeks continuously" with hydrological information such as "The tide was growing slowly". Also cross-references, the social environment of the reporting or the persliche überreibungsma Mudelsee includes with the considerations. "The reconstruction of flood chronologies thus equates to an exciting detective work, " smiles the climate researcher.

Approach to the truth

In the center of research regardless of hurricane or flood is always the occurrence rate of events, that is, the number of events per year. If these were known exactly, it would not be possible to prevent all damage in the future, but at least the extent of the expected damage could be minimized through better planning. "The problem is, however, that only a limited amount of data on extreme weather events is available to us in the past, " says Mudelsee. You can therefore only try to approach the true occurrence rate with appropriate confidence in the forecasts.

The determination of errors is therefore complex and requires elaborate mathematical simulations. But these calculations are absolutely necessary as a measure for determining the credibility of the forecasts. If there is no certainty about the secrets of the record or is concealed in media and scientific publications, then this is an indication that either amateurs are at work or that there is other than scientific motivation, explains Mrt Mudelsee emphatically. "But that hurts twice: both the economy and the reputation of science."


Further information:

(Climate Risk Analysis, 07.09.2007 - AHE)