The accuracy of the paleomagnetic method as evaluated from historic hawaiian lava flows

An analysis was made of the various levels at which errors enter in paleomagnetic studies of laval flows, using as controls lava flows from Hawaii which cooled in known fields. The largest intraflow angular variations occur between different individually oriented cores and are due to rotation of blocks after cooling, to intraflow magnetic field variations at the time of cooling, and to orientation errors; other sources of angular error have much smaller effects. The ultimate accuracy of a paleomagnetic study is limited by the magnetic anomalies existing over the old terrain where the lava flows cooled. Because of these ancient anomalies, little additional accuracy can be gained in volcanic regions like Hawaii by collecting more than six oriented samples per flow, a number which results in an accuracy at the 95 per cent confidence level of 3° to 5°. If these results are applied to older lavas on Hawaii, the low rates of secular variation previously noted from observatory data are found to persist back to 1750.