Search

Gravity interface inversion is a critical technique in delineating the substructure of basins, providing essential technological and data support for oil and gas exploration. Traditional gravity inversion approaches often encounter issues such as suboptimal local solutions and limited resolution. Moreover, conventional deep learning inversion methods typically require extensive time for empirical parameter adjustment, hindering the achievement of optimal training outcomes. By utilizing Bouguer gravity anomaly data, this research pioneers the application of the EfficientNetV2 network in predicting 3-D basement relief interfaces and variations in overburden density. The network employs a composite scaling technique to adaptively adjust its width, depth, and input resolution, thereby identifying the most effective network configuration. Concurrently, the innovative Fused-MBconv convolutional module efficiently achieves superior results with a reduced number of network parameters. Specifically, in the Poyang Lake Basin study in Jiangxi Province, China, the EfficientNetV2 model demonstrated enhanced accuracy in predicting density variations of the basement interface and overlying strata compared to traditional methodologies.

The three-pool and first-order model separates the mineralizable organic carbon into active, slow, and passive carbon pools. This paper used the model and decomposition curves of the soil organic carbon to fit the active pool and its decomposition rate, slow pool and its decomposition rate. The results showed that the size of the active pool from different profiles accounted for 2.09%-3.08% of the total soil organic carbon and the mean residue time was 3.57-17.21 days. And the size of the slow pool accounted for 3.19%-43.55% and the mean residue time was 1.12-4.94 years. Acid hydrolysis (6M HCl) was used to fractionate the passive organic carbon, which accounted for 50.83%-94.44% of the total soil organic carbon.