Multi-Parameter Coupling of Seismic, Drilling and Logging Data for Fine Prediction of Cambrian Gypsum-Salt Sequences

An accurate prediction of gypsum-salt cap rocks is crucial for safe drilling and efficient development of subsalt hydrocarbons in the Tarim Basin. However, due to the complex lithological assemblages within the Middle Cambrian sequences, the P-impedance of gypsum severely overlaps with that of the background carbonate rocks, making conventional seismic prediction based on a single impedance threshold inadequate. To address this issue, this paper proposes a multi-parameter coupling while-drilling prediction method that integrates seismic data, drilling parameters and geochemical logging. First, pre-stack P-impedance inversion is performed on amplitude-preserved seismic data to macroscopically delineate the depth interval of gypsum-salt development. Second, a weighted synthetic change-rate model incorporating drilling time (DT), weight on bit (WOB), and outlet conductivity is constructed to dynamically calibrate the top interface of the gypsum-salt layer while drilling. Finally, within the constrained depth interval, X-ray fluorescence (XRF) elemental inversion of cuttings is used to quantitatively calculate the contents of clay, dolomite, gypsum and calcite, achieving fine-scale lithofacies identification at the sublayer level. An empirical application to Well XSC1 in the Keping fault-uplift, Tarim Basin, demonstrates that compared with the conventional single P-impedance threshold method, the proposed approach effectively distinguishes gypsum, salt and carbonate wall rocks, and significantly reduces the prediction errors of individual salt beds and gypsum sublayers. Specifically, the mean absolute error (MAE) of total gypsum-salt content decreases from 24.56% to 13.73% (a reduction of 44.1%), the bias drops from 2.45% to 0.31%, and the root mean square error (RMSE) decreases from 33.63% to 21.55%. At a 1% content threshold, the F1 score for gypsum-salt identification increases from 0.63 to 0.82; the absolute depth error of the top and bottom interfaces is reduced by approximately 70%, and the relative thickness error shrinks from ±40% to within ±10%. Transferability of this multi-parameter coupling strategy to other salt-bearing basins in central-western China remains unvalidated and requires future validation and testing.