Calibration of Old Well Log Data Using a Multiwell Environment

ABSTRACT This paper reviews the implementation of a special multiwell calibration technique to transform and normalise old well data within a petrophysical study. One problem in the evaluation of old wells is the porosity calibration of the Neutron counts. This can be achieved through a normalisation program on a single well basis if another porosity measurement such as the Sonic log is available in the same well. In the case of this study two such porosity-dependent measurements were not available and a new variable was needed to link old and new data. However, good log correlation in the field and the availability of a multiwell data base set-up allowed the use of log depth as this extra dimension. Old data is imported into new well data and is automatically tied to them. This is done by extracting from the Multiwell Data Base the formation tops of old and new wells and by depth-shifting the old well data by an amount equal to the difference between the formation top depths. Then an auto-correlation depthmatching program using comparable curves such as old and new Gamma Rays refines the tie between wells. Having linked the data through a common depth, a multiwell cross-plotting technique gives the best fit between old and new data. The best fit equation calibrates the old data and this is validated quickly against field trends using multiwell crossplots To account for the different sensitivity of the GR detectors as the technology of logging tools has evolved, the same calibration technique can be applied to rescale the GR data of the old wells. This re-scaling of the GR of old wells gives the same ranges as the GR of key wells allowing the use of multiwell histogram/ estimation programs and the use of similar interpretation parameters throughout the field. The old log data was successfully calibrated allowing their use as input to modem log evaluation packages. For each old well the multiwell evaluation set-up gave easy calibration against one or several recent nearby wells and it ensured quality checks of the results. The technique can be used to integrate older data into a study, giving the more detailed coverage needed for reservoir characterisation and simulation work.