Borehole heat flow along the eastern flank of the Juan de Fuca Ridge, including effects of anisotropy and temperature dependence of sediment thermal conductivity

The thermal conductivities of 15 whole-round sediment samples collected dur ing Ocean Drilling Program (ODP) Leg 168 between 17 and 440 m below the sea floor on the eastern flank of the Juan de Fuca Ridge were tested to documen t their anisotropy and temperature dependence using the divided bar techniq ue. Tests over a temperature range of 5 degrees C to 60 degrees C reveal va riations in conductivity of up to +/-15%. The sign and amplitude of these v ariations depend on the thermal conductivity at laboratory (room) temperatu re (lambda(rt)): if lambda(rt) approximate to 0.8 W m(-1) K-1 (high porosit y), conductivity increases with temperature; if lambda(rt) approximate to 1 .2 W m(-1) K-1 (moderate porosity), conductivity does not change with tempe rature if lambda(rt) approximate to 1.6 W m(-1) K-1 (low porosity), conduct ivity decreases with temperature. This behavior results from a positive tem perature coefficient for seawater (lambda proportional to T) and a negative coefficient for rock matrix (lambda proportional to 1/T). A special sampli ng technique for unconsolidated sediments made it possible to measure horiz ontal (lambda(HOR)) and vertical (lambda(VER)) components of thermal conduc tivity independently and to determine a mean anisotropy value (lambda(HOR)/ lambda(VER)) Of 1.2 Corrections, which are <10% for anisotropy and <1% for temperature, were applied using in situ temperatures, shipboard line source thermal conductivities, porosities, and the geometric mean mixing model th at accounts for matrix and porewater constituent conductivities. On the bas is of these corrected conductivities and a harmonic averaging of values wei ghted according to the lithologic division of the sediment sections into cl ay-rich and sand-rich units, values of heat flow estimated from borehole te mperature measurements are lower than those previously estimated by similar to 10% except for one site where the section is dominated by sand. Shallow seafloor heat flow measured with gravity-driven probes may also need to be corrected for anisotropy, although the degree of anisotropy in the unconso lidated, high-porosity sediments within a few meters of the seafloor is poo rly constrained.