Some Problems in Thermal History Studies

Abstract The conversion of organic matter to oil and gas is determined by time and temperature. Estimation of a thermal history allows the timing and extent of oil and gas generation to be estimated by applying a model of organic maturation. A logical first step in constraining a thermal history is determination of the present day thermal state. The analysis can then be extended into the past through a mathematical model constrained by paleothermal indicators. The preferred way to characterize the thermal state of a sedimentary basin is by heat flow, because this removes most of the vertical and lateral variation in temperature and temperature gradients caused by vertical and lateral changes in lithology and thermal conductivity. Heat flow is not measured, but is estimated from measurements or estimates of temperature and thermal conductivity. Present day temperature can be measured in a drillhole with an accuracy better than 2°C, but continuous high-precision temperature logs in deep drillholes are scarce. Bottom-hole temperatures (BHTs) measured in geophysical logging runs are frequently the only type of temperature data available. Single BHTs are suspect, but temperature can probably be estimated from large suites of BHT data with an accuracy of about 5 °C, by applying methods which average random errors. Thermal conductivity can be measured in the laboratory, or estimated on the basis of lithology. Laboratory measurements must be corrected for in-situ conditions, and probably yield estimates of in-situ thermal conductivity that are no better than ±10%, on the average. Estimates made on the basis of lithology are likely to be in error by as much as ±30 to 40%. There is an established hierarchy for the accuracy of heat flow estimations that ranges from ±5% to ±50%. Simple models are used to illustrate how the thermal regime in a sedimentary basin is affected by heat conduction, radiogenic heat production, groundwater movement, solid-state advection (sedimentation or erosion), and transient effects. Lateral changes in thermal conductivity can result in lateral changes in the geothermal gradient as large as 100%. Lateral changes in radiogenic heat production can similarly produce lateral changes in the geothermal gradient as large as 25%, and may be particularly important in areas such as fold-thrust belts where radiogenic heat production varies widely. Heat transfer by advection is proportional to the velocity, density, specific heat, and temperature of the moving fluid. Upwelling groundwater increases the temperature gradient near the surface; downwelling groundwater decreases the near-surface geothermal gradient. Lateral variations in heat flow on the order of ±20 to 40 mW/m2 in some sedimentary basins may be due to topographically-driven groundwater flow. Erosion leads to higher near-surface heat flow, sedimentation leads to lower near-surface heat flow. The magnitude of surface heat flow depression or elevation associated with erosion or sedimentation is dependent on the erosion/sedimentation rate, and the thermal conductivity of the sediments eroded/deposited. Methods for assessing thermal maturity can be divide into those based on organic reactions and those based on inorganic reactions, or radioactive decay. Paleothermal indicators can be calibrated from laboratory measurements or assumed thermal histories; neither method is completely satisfactory. Different paleothermal indicators record different types of information. Vitrinite reflectance is probably the most common organic method, and is sensitive to the maximum paleotemperature. Problems with vitrinite data include mis-identification of vitrinite, reworking, possible sample bias, a dependence of reflectance on lithology, oxidation, and a lack of inter-laboratory agreement. The most common inorganic method is based on apatite fission track analysis. Apatite fission track analysis provides information about the thermal history subsequent to the last cooling event through approximately 70 to 125°C, including information about the timing of this event. An empirical relationship that relates the fractional reduction in mean fission-track length to time and temperature has been formulated from laboratory data, but recent field data call the validity of this relationship into question.