Automated lithology prediction from PGNAA and other geophysical logs

Different methods of lithology predictions from geophysical data have been developed in the last 15 years. The geophysical logs used for predicting lithology are the conventional logs: sonic, neutron–neutron, γ (total natural-γ) and density (backscattered γ–γ). The prompt gamma neutron activation analysis (PGNAA) is another established geophysical logging technique for in situ element analysis of rocks in boreholes. The work described in this paper was carried out to investigate the application of PGNAA to the lithology interpretation. The data interpretation was conducted using the automatic interpretation program LogTrans based on statistical analysis. Limited test suggests that PGNAA logging data can be used to predict the lithology. A success rate of 73% for lithology prediction was achieved from PGNAA logging data only. It can also be used in conjunction with the conventional geophysical logs to enhance the lithology prediction.