Paleomagnetism can provide quantitative, empirical data on basement heating by cooling lava flows, thereby constraining theoretical model of heat transport processes. We used paleomagnetism to study heating in a Frenchman Springs (FS) lava by the overlying cooling Roza flow. The flows belong to the Miocene Columbia river Basalt (CRB) groups, and they recorded very different primary paleomagnetic directions: FS has normal polarity, and Roza is transitional. The solid base of Roza typically rests on a few meters of the brecciated and vesicular top of FS, which is otherwise solid. Thermal demagnetization of FS specimens was usually successful at separating the overprinting due to Roza from the primary FS direction and for establishing limits on the unblocking temperatures of the overprinting. The unblocking temperatures agree well between sites separated by up to 70 km, irrespective of Roza thickness at the sites, which vary from 35--62 m. In specimens from 0.3 to 1 m below the contact, the overprinting was unblocked in the laboratory between 600 and 300 ¿C and close to 200 ¿C at 4-m depth. No overprinting was apparent below 6 m. The influence of longer heating times in the field on the unblocking temperatures was apparent below 6 m. The influence of longer heating times in the field on the unblocking temperatures was estimated by viscous remanence acquisition at elevated temperatures.

The results suggest typical reductions of the unblocking temperatures by a few tens of degrees. Our observations imply much less heating of the basement than predicted by simple conductive thermal models. According for the low conductivity breccia in the contact zone results in a better agreement with the experimental temperature profile, but unrealistically low conductivities are needed to sufficiently reduce the absolute temperature. The observed heating is effectively explained by postulating a wet basement at the time of Roza extrusion, as well as groundwater, to maintain a low-temperature isotherm (~100 ¿C) a few meters below the contact. The presence of water during the time of Roza extrusion and some other CRB flows has been suggested by field observations. ¿ American Geophysical Union 1988