Project Overview

Palaeotempestology is the study of past storms using geological and documentary evidence. The destructive nature of hurricanes means that these storms leave traces in the geological record of varying resolution and strength. Commonly used geological proxies for reconstructing past storms include storm sediment deposits from lakes, marshes and beaches, shell deposits, corals, tree rings, and speleothems.

How are hurricane signals recorded in speleothem calcite?

Stalagmites are well established natural archives of past climate. They grow as rainwater percolates through the soil and epikarst and drips onto the top of the stalagmite, depositing a thin layer of calcite with each drop. The chemical signature (isotopes and trace elements) of the drip water reflects the above-ground temperature and precipitation and this is preserved in the geochemistry of the newly deposited calcite.

Tropical cyclone rainfall events are characterised by short-lived, intense precipitation with extremely low δ18O values that can affect areas hundreds of kilometres in diameter. The isotopically distinct rainwater behaves as a natural isotopic tracer infiltrating the soil and karst system and creating a negative departure from the long term isotopic trend of drip waters and cave calcite. This occurs after a lag time, dependent upon the flow pathway and the depth of the overlying soil and bedrock (see diagram below). Because each new layer of calcite will contain an oxygen isotope signature similar to that of the original rainfall associated with the hurricane, individual hurricanes that passed over the cave during the growth period of the stalagmite can be identified when the stalagmite is sampled at a subannual resolution.

While normal rainfall in the tropics tends to have relatively high oxygen isotope values (around 0 to –5 per mil), hurricane rainfall oxygen isotope values are significantly lower (between -8 to -15 per mil). One hurricane alone can lower integrated annual groundwater δ18O values by approximately 1.5 per mil in a localised area and by 0.5 per mil over an area several hundreds of kilometres across. Hurricane strikes, therefore, appear in the stalagmite oxygen isotope record as sharp negative excursions in δ18O. The greater the sampling resolution of the stalagmite calcite, the more pronounced the hurricane δ18O signal.

cross section of a hurricane
Above, Schematic diagram of hurricane circulation and typical rainwater isotope values.

The low δ18O values of hurricane precipitation occur because hurricanes are very efficient fractionators. Heavy isotopes (18O, 2H) are preferentially removed from the surrounding vapour as they condense and fall as rain or snow, causing the isotope ratios to decrease with height. As the raindrops fall, they become further enriched in heavy isotopes due to mixing with isotopically heavy water vapour closer to sea level. Because rain removes heavy isotopes from the air, the isotope ratios in a given hurricane will decrease with time, amount of rainfall (i.e., the amount effect), and inward from the edge of the storm. The long lifetime of these immense storm systems and the large volumes of rain associated with them, contribute to the characteristically low oxygen isotope signature of hurricane rainfall.

Speleotherm Formation Diagram
Left, diagram illustrating the variability of hypothetical flow paths feeding stalagmites in a particular cave. The stalagmite hurricane signal would likely be clearest if the stalagmite were fed by drip water transported via conduit flow pathways. Diffuse flow pathways would likely yield the most attenuated hurricane signal. The exposure of infiltrating rainwater to evaporation and mixing with older, stored water has the potential to obscure the hurricane signal. This demonstrates the importance of fully characterising the drip site associated with stalagmites collected for the study (as described below). Diagram adapted from Lachniet (2009)

What would a stalagmite-based hurricane proxy record look like?

A very short pilot study conducted by Dr Amy Frappier (now on our research team) yielded a 23 year record of hurricane activity using a stalagmite from Actun Tunichil Muknal in Central Belize. The stalagmite ATM7 isotope record revealed sharp negative δ18O excursions, each of which coincided with a known hurricane that passed over Belize during the 23 year time interval. Hurricane Mitch appears as two sharp negative spikes marking the hurricane’s initial and return pass of the site.

Delta 18 Oxygen record for stalagmite ATM7
Above, a high resolution δ18O record from stalagmite ATM7 from Central Belize showing brief isotopic negative excursions that coincide with historical hurricane landfalls near the site (named). This figure is adapted from Frappier et al., Geology, 2007

In the Field: Stalagmite Selection and Cave Monitoring

Within a cave system not all speleothems grow at the same rate or record climatic signals at the same resolution. This is due to differences in how each stalagmite is fed, the hydrological pathways through the overlying karst and environmental conditions within the cave. Some stalagmites may record a long-term averaged climate signal while others may record individual rain events. Morphological characteristics, drip monitoring, and monitoring surface climate will help to identify stalagmites capable of recording the climate signal suitable for this project (i.e., with sub-annual resolution).

Cave environmental conditions are also known to affect stalagmite growth. Cave air partial pressure carbon dioxide (pCO2) has been shown to greatly affect the rate of calcite deposition. It is critical that cave air pCO2 is characterised as it can vary seasonally, thereby affecting the timing of calcite deposition and creating bias within the climate record. The temperature and humidity of the cave can also disrupt the isotopic signal captured by the stalagmite through the process of disequilibrium fractionation so this will also be monitored.

Ultimately, the geochemical signature of the stalagmite is interpreted to reflect the climate at the time the stalagmite was growing. To increase our confidence that the stalagmite is faithfully recording climate we also monitor surface climate for comparison to the monitoring data within the cave.

Taking measurements in the field

In the Lab: Constructing the Record

The bulk of the lab work for this project is being conducted in the Sir Kingsley Dunham Palaeoclimate Laboratory and the Stable Isotope Laboratory, both housed in the Department of Earth Sciences at Durham University. The Uranium-series dating is being conducted at the University of Bristol except in the case of stalagmites provided by our Yok Balum Cave collaborators which are being dated by Dr Yemane Asmerom at the University of New Mexico.

The resolution of the record depends on the growth rate of the samples chosen, but approximately 5 to 15 samples per annual layer is our goal. Sampling involves sectioning the stalagmite, polishing it, and then micro-milling powders along the central growth axis.

The stalagmite YokG
Far left, photograph of stalagmite YOK-G from Yok Balum Cave in Belize. Left, photograph of stalagmite TC-99-4(M) from Conch Bar Caves, Turks and Caicos Island. These stalagmites are ideal for a hurricane reconstruction as they have excellent preservation, are annually layered, and were actively growing at time of collection. YOK-G has the added benefit of being incredibly fast growing (1 mm per year on average).


Conservation and Cave Management

We are determined to balance the scientific needs of our research with the other values that communities and other researchers place on the caves that we use. There is a huge amount of research taking place around the world in these sensitive environments and we see ourselves as ambassadors for sustainable cave research. We are respectful to the endeavours of other scientists and the economic and cultural value of the cave to local communities. We have also made all possible arrangements to minimise disruption to cave ecology, wildlife and cave aesthetics. We intend to remove stalagmites only when we were certain of their appropriateness after long term screening. We also intend to make use of stalagmites that have already been collected by others.