IMPLICATIONS OF MULTIPLICITY IN KINETIC-PARAMETERS TO PETROLEUM-EXPLORATION - DISTRIBUTED ACTIVATION-ENERGY MODELS

A grid search procedure has been developed to estimate the kinetic parameters of kerogen decomposition represented by the distributed activation energy models (DAEM). This procedure uses approximate analytical solutions to the time-temperature integrals of DAEM. These approximations reduce the computational efforts involved in parameter estimation substantially and allow the rapid estimation and graphical representation of the three-dimensional error surface of the objective function over a wide range of kinetic parameters while retaining the true nonlinearity of the model. The existence of a number of sets of kinetic parameters which describe the observed data, generally within the limits of experimental accuracy, is also shown. These multiple sets of mean activation energy of the distribution and the corresponding preexponential factor correlated to a form referred to as the "compensation effect". This has been demonstrated by using published data on kerogen decomposition and experimental data. While these multiple sets of kinetic parameters can predict the laboratory data to similar levels of accuracy, they behave differently when extrapolated to geological heating rates. This then results in significantly different predictions for timing of onset of hydrocation generation. It is concluded that kinetic models used for the extrapolation from laboratory to geological heating rates are unlikely to provide any more guidance than a range of estimates for kerogen decomposition in the best case of known thermal history. This is shown to be due to lack of resolution arising from the analysis of the laboratory data which results in nonunique kinetic parameters. A kinetic model based on a distribution of activation energies initially developed for kerogen decomposition has been shown to predict length reduction of fission tracks in apatite (a mineral used as a thermal history indicator) due to thermal annealing in the laboratory. Compensation effect is observed in this case also.