In situ thermal measurements were made using the Adara APC temperature tool and the DVTP (Davis-Villinger Temperature Tool; Davis et al., 1997). The in situ temperature instruments are described below and the resulting data are listed in the site chapters.
The Adara tool consists of electronic components, including battery packs and a data logger, and a platinum resistance temperature device calibrated over a temperature range of 0º-30ºC. A photograph of Adara components can be found in Fisher and Becker (1993). The Adara thermal tool fits directly into the cutting shoe on the APC and, therefore, can be used during regular piston coring. The thermal time constant of the cutting shoe assembly into which the Adara tool is inserted is ~2-3 min. To obtain thermal measurements, the only modification required to the normal APC procedures is that the corer be held in place for at least 10 min after firing. During this time the Adara tool tracks the thermal equilibration of the sediments.
The Adara tool logs data on a microprocessor contained within the instrument. Following deployment, the data are downloaded for processing on PC computers. The tool can be preprogrammed to record temperatures at a range of sampling rates. Sampling rates of 5 s were used during Leg 180.
A typical APC measurement consists of a mudline temperature record lasting 10 min for the first deployment at each borehole, a pulse of frictional heating when the piston is fired, a period of thermal decay that is monitored for 10 min, and a frictional pulse upon removal of the corer. Prior to reduction and drift corrections, the nominal accuracy of Adara temperature data is estimated at 0.1ºC.
At depths below which the Adara tool could no longer be used, the DVTP was deployed. The probe is conical and has two thermistors, one located 1 cm from the tip of the probe and the other 12 cm above the tip. A third thermistor, referred to as the internal thermistor, is located in the electronics package. Thermistor sensitivity is 10-3 ºC in an operating range of -5º to 20ºC, and the total operating range is -5º to 100ºC. The thermistors were calibrated at the factory and on the laboratory bench before installation in the probe. In addition to the thermistors, the probe contains an accelerometer sensitive to 0.98 m·s-2. Both peak and mean acceleration are recorded by the logger. The accelerometer data are used to track disturbances to the instrument package during the equilibration interval.
In a DVTP deployment, mudline temperatures are measured for 10 min on the first run within each hole before descent into the hole for a 10-min equilibration interval at the bottom. Mudline temperatures are also collected for at least 2 min on ascent. Data collected during Leg 164 indicate that the time constants are probably 1 min for the probe tip thermistor and ~2 min for the thermistor at 12 cm from the tip (Shipboard Scientific Party, 1996b). Data from the probe tip thermistor were used for estimation of in situ temperatures.
Data reduction procedures are similar for all the temperature tools. The transient thermal decay curves for marine thermal probes are known to be a function of the geometry of the probes and the thermal properties of the probe and the sediments (Bullard, 1954; Horai and von Herzen, 1985). Analysis of data requires the construction of synthetic decay curves calculated on the basis of tool geometry, sampling interval, and tool and sediment thermal properties. For the DVTP tool, thermal decay data are analyzed by comparison to computed type curves using the software program CONEFIT developed by Davis et al. (1997). However, it is never possible to obtain a perfect match between the synthetic curves and the data because (1) the probe never reaches thermal equilibrium during the penetration period; (2) contrary to theory, the frictional pulse upon insertion is never instantaneous; and (3) temperature data are sampled at discrete intervals, meaning that the exact time of penetration is always uncertain. Thus, both the effective penetration time and equilibrium temperature must be estimated by applying a fitting procedure, which involves shifting the synthetic curves in time to obtain a match with the recorded data. The data collected more than 20-50 s beyond penetration usually provide a reliable estimate of equilibrium temperature. The thermal decay curves for the measurements were initially fit aboard ship assuming formation thermal conductivity of 1.0 W·m-1·ºC-1 for all of the data. Thermal conductivities from core measurement (see "Thermal Conductivity") were used for final estimation of in situ temperatures and for calculation of heat flow. Laboratory thermal conductivity results were corrected to approximate in situ conditions assuming an increase in conductivity of 0.5% per 1000 m of water depth and a decrease in conductivity of 0.25% per degree of temperature increase (Ratcliffe, 1960). These corrections do not account for any underestimation of thermal conductivity caused by elastic rebound during recovery of the cores.