Andreas Schmidt Mumm
Institut fuer Geologische Wissenschaften und Geiseltalmuseum, Domstr. 5, D-06108 Halle Wittenberg
The most important prerequisite for a successfull reconstruction of the thermal history of a given geological unit is the detailed knowledge of the genetic relationships of the diagenetic mineral assemblages. Conventional thin section petrography, backed up by cathodoluminescence and, in the case of investigations of higher hydrocarbons, also fluorescnce microscopy are applied to obtain the required petrogenetic information (BURRUSS 1987). The first step in studying the fluid inclusion assemblages is to determine the genetic relationships of inclusion generations to their (diagenetic) host minerals, as outlined by for example in GOLDSTEIN & REYNOLDS (1994). Up to this point the procedure is a relatively easy to follow routine. The first problems may arise when analysing fluid inclusions in diagenetic settings associated with evaporitic environments, as the low temperature behaviour of the aqueous brines - due to their commonly unsual and complex composition - is often very difficult to interpret and marked by metastabilities and irreversible/irreproducable phase changes. The analyst will have to develope measuring routines suitable for each particular assemblage of inclusions to take credit for hydrate, ice and clathrate melting patterns but also for effects of salting out or supersaturation. This may even include heating up to final homogenization of the inclusions between induvidual freezing runs. It may take up to several days to establish a routine that allows reliable and reproducable low temperature measurments. The importance of the accurate determination and interpretation of the low temperature phase changes arises from the interpretation of the homogenization temperature(s), e.g. derivation of trapping temperatures from homogenization temperatures. While the effect of salinity on the pressure correction of homogenization temperatures can be accounted for by assuming an NaCl saturated solution without introducing too much error, the possible presence of dissolved gases may introduce a much larger error. In the case of a gas saturated aqueous inclusion a pressure correction may even be unnecessary as Th equals the trapping temperature (GOLDSTEIN & REYNOLDS, 1994).
Another prerequisite for the reliable interpretation of fluid inclusion data is the consistency of data within a given inclusion assemblage. It should be stated, that within a given inclusion assemblage, the microthermometric results should have only minimal variation - however, this rarely happens. In most cases the results of measurements will spread considerably which confronts the analyst with the problem of a conclusive interpretation. Inclusions do undergo post-trappment changes which may cause homogenization temperatures to spread. A simple method of determining whether an inclusion assemblage underwent partial re-equilibration resulting in differential volume adjustment, is a plot of homogenization temperatures versus inclusion size. If a correlation of the two is obvious, a post-trapping volume adjustment is likely to have ocurred, as this correlation expresses the more readily adjustment of large inclusions compared to smaller ones due to the lower volume/surface ratio of the latter. In an undisturbed setting the distribution should be random. However, the determination of a post-formational re-equillibration of an inclusion assemblage can be helpfull for interpreting the data in the geological framework. In many cases fluid inclusions in diagenetic minerals, especially in cements and porefills, are rare and small and thus the obtained data set may comprise only few measurements which in addition may spread over a considerable temperature interval. How can such data sets be sensibly interpreted to derive trapping conditions? Several borderline cases have to be considered here. If the post-trapping P-T path was characterized by isothermal uplift (clockwise path) the maximum measured Th can be considered to represent maximum temperatures of trapping. If the post-trapping P-T path was marked by isobaric cooling (counter-clockwise path) the effect of re-equilibration would be that of lowering homogenization temperatures e.g. increase densities. This path can lead to a wide spread of homogenization temperatures. Once a reliable set of temperature data has been obtained for the fluid inclusions the temperatures of formation of the respective hosting diagenetic minerals are determined. The internal geothermal gradient that prevailed during diagenesis is given by the temperature difference between two samples divided by the covered depth interval. A factor that has to be considered here is the post-formational compaction of the considered rock unit, which will artificially increase the thermal gradient, especially towards depth. In a homogeneous sequence of sedimentary rocks, the vertical gradient should therefore be slightly curved towards higher values at depth. In heterogeneous rock sequences, however, the gradient varies according to the different thermal conductivity and variation of compaction of different rock types. Further complications in the course of determination of internal gradients are faults that offset the the gradients. Local influx of hot or cold fluids along permeable layers or structures tends to shift the temperatures to higher or lower values through fluid mixing.
However, in most cases these disturbances can be identified through detailed inclusion analysis and correlation of the different diagenentic minerals. If different sets of diagenetic mineral assemblages can be identified, the thermal history of a basin sequence can be cronologically reconstructed. Inclusions in the different minerals are not only distinguished upon their volumetric properties but also by systematic variation of composition. In many studied cases, different thermal gradients could be clearly established for individual phases of mineral formation, leading to a complex picture of the diagenetic evolution of the studied area. The final target of reconstruction of the thermal gradients may be the determination of the time of activity of a respective fluid system. This can be approached by extrapolating the overall gradient of a distinct fluid-thermal event to a constructed 0m/10 C palaeo-surface. The stratigraphic unit that is intersected at this point should indicate the interval during which the system was active. The detailed and precise reconstruction of the thermal history of sedimentary basins is fundamental to the understanding of generation, migration anf fixation of natural oil and gas. The analysis of fluid inclusions is a unique method to decipher the often complex patterns of the thermal history of sedimentary.