Search

Lithostratigraphic modeling holds a vital role in mineral resource exploration and geological studies. In this study, we introduce a novel approach for automating pseudo-lithostratigraphic modeling in the deep subsurface, leveraging inversed geophysical properties. We propose a three-dimensional convolutional neural network with adaptive moment estimation (3D Adam-CNN) to achieve this objective. Our model employs 3D geophysical properties as input features for training, concurrently reconstructing a 3D geological model of the shallow subsurface for lithostratigraphic labeling purposes. To enhance the accuracy of pseudo-lithostratigraphic modeling during the model training phase, we redesign the 3D CNN framework, fine-tuning its parameters using the Adam optimizer. The Adam optimizer ensures controlled parameter updates with minimal memory overhead, rendering it particularly well-suited for convolutional learning involving huge 3D datasets with multi-dimensional features. To validate our proposed 3D Adam-CNN model, we compare the performance of our approach with 1D and 2D CNN models in the Qingniandian area of Heilongjiang Province, Northeastern China. By cross-matching the model’s predictions with manually modeled shallow subsurface lithostratigraphic distributions, we substantiate its reliability and accuracy. The 3D Adam-CNN model emerges as a robust and effective solution for lithostratigraphic modeling in the deep subsurface, utilizing geophysical properties.

Groundwater is closely related to hydrogeological structure and hydro-lithology, which mainly refers to the spatial distributions and properties of the environment where groundwater occurs. To analyze the constraints of hydrogeological structure and hydro-lithology on regional groundwater resources in the Eastern Henan Plain, China, we reconstructed the three-dimensional (3D) hierarchical models at two scales, hydrogeological structural models and hydro-lithological models, using hydrogeological cross-sections. First, the hydrogeological structural models of four aquifer groups, corresponding to four formations of the Quaternary in the study area, were reconstructed. Second, the hierarchical hydro-lithological model was built using SIS and IK estimation under the constraint of each aquifer group model space, respectively. Compared to global model, the variograms of hierarchical model captured more spatial characteristics of lithology in each aquifer group. The IK hierarchical model presents more continuities, clear boundaries, and realistic geometric shapes of the three lithologies, excluding the banding characteristics of the IK global model. The hierarchical SIS models reproduced the lithology distribution of each aquifer group and captured small changes in the lithology, with the smallest absolute percentage errors (APEs). Third, coupling the SIS hierarchical models and the groundwater levels, the groundwater resource in the study area was estimated to have a total volume of 1.2339 × 104 m3. The shallow groundwater in the study area is mainly concentrated in Hebi City and the Puyang basin of the Yellow River, and deep groundwater is mainly concentrated in the northern Anyang City and Hebi City. Finally, the possible quantities of shallow and deep groundwater recharges were estimated for future groundwater management decision in the study area. The hierarchical hydrogeological model, groundwater resource assessment, and possible groundwater recharge estimation can also provide a basis for groundwater vulnerability, groundwater extraction, and land subsidence assessment.