Optimization of gamma‐ray logging techniques for uranium

In gamma‐ray logging for uranium exploration and evaluation, many factors affect the accuracy of the final processed log. This paper presents the results of a study of various logging and data processing parameters over which the logger can exercise control. These include probe velocity, sampling interval, detector length, and, in the case of analog recording, ratemeter time constant and chart paper depth scale. Numerous replications of gamma‐ray logs past the same sequence of radioactive zones varying one logging parameter at a time, combined with theoretical considerations, lead to some useful conclusions regarding the effects of those parameters on the accuracy and resolution of the logs, and the efficacy of deconvolution techniques under various conditions. In general, digital recording is preferable to analog recording because, with the latter, the accuracy of the computed grade‐thickness product can suffer as a result of the extra steps of converting count rate to an analog chart record and back again to digital form for processing. In the case of digital gamma‐ray logging for uranium, it appears that it is less important to log slowly for good statistical accuracy than it is to use a small sampling or digitization interval. Under the high count‐rate conditions encountered in uranium logging, the factors limiting the accuracy of the log are likely to be detector length (which limits spatial resolution) and geologic noise (e.g., inhomogeneities in the rock); logging more slowly does not remedy these problems. In cases where the recorded gamma‐ray flux is low, such as in lithologic correlation, it may still be necessary to log very slowly to achieve acceptable statistical reliability.