Letters to Nature

Nature 393, 155-158 (14 May 1998) | doi:10.1038/30218; Received 8 May 1997; Accepted 16 March 1998

Possible triggering of Heinrich events by ice-load-induced earthquakes

A. G. Hunt1 & P. E. Malin1

  1. Division of Earth and Ocean Sciences, Nicholas School of the Environment, Duke University, Box 90235, Durham, North Carolina 27708-0235, USA

Correspondence to: P. E. Malin1 Correspondence and requests for materials should be addressed to P.E.M. (e-mail: Email: pem@vaino.geo.duke.edu.)

North Atlantic sediments dating from the last ice age contain layers of rock fragments from northeastern Canada (so-called Heinrich layers)1. Like modern iceberg-borne sediments from Greenland, these layers have been attributed to ice-rafting episodes1, 2, 3. Six Heinrich layers have been documented and correlated with climate changes4, 5, 6, 7, 8. The layers, which are several centimetres thick, contain negligible amounts of foraminifera (which accumulate at a few millimetres per century), implying that they were deposited over just a few years. These ice-rafting Heinrich events are separated by progressively shorter intervals from about 40 to 6 kyr (ref. 9), and it has been suggested10 that they are related to the Milankovitch cycles in the Earth's orbital parameters11. Alternatively, they may be generated by forcing mechanisms arising from the internal dynamics of the Laurentide ice sheet12. Here we suggest the possibility that the Heinrich events were precipitated by ice-load-induced earthquakes, analogous to those produced by reservoir water loads13. We suggest that near its edge, the Laurentide ice sheet sheared the Earth's crust, inducing repeated failure that released the ice rafts. This region (along Canada's northeastern seaboard) shows evidence of both current14,15 and past seismic activity owing to postglacial rebound. Our model accounts for the intervals between both the Heinrich events and the evidence of palaeoseismicity, and can be tested by studying local sedimentary relationships.