The electrical conductivity of soil is closely associated with various physical properties of the soil, and accurately establishing the interrelationship between them has long been a critical challenge limiting its widespread application. Traditional approaches in geotechnical engineering have relied on specific conduction mechanisms and simplifying assumptions to construct theoretical models for electrical conductivity. This paper adopts a different approach by using machine learning methods to predict the electrical conductivity of clay materials. A reliable dataset was generated using the quartet structure generation set to create random clay microstructures and calculate their electrical conductivity. Based on this dataset, machine learning methods such as least squares support vector machine and backpropagation neural networks outperform theoretical models in terms of prediction accuracy and resistance to interference, with a coefficient of determination (R2) exceeding 0.995 when unaffected by disturbances. The computation of Shapley values for input features aids in explicating the machine learning model. The results reveal that saturation is a key feature in predicting electrical conductivity, while porosity and constrained diameter are relatively less important. Finally, an already trained model is applied to the one-dimensional electroosmosis-surcharge preloading consolidation theory. The results of the calculations demonstrate that neglecting changes in soil electrical conductivity during electroosmosis can lead to an overestimation of the absolute values of anode excess pore water pressure and soil settlement.

Reliable subseasonal forecasts of high summer temperatures would be very valuable for society. Although state-of-the-art numerical weather prediction (NWP) models have become much better in representing the relevant sources of predictability like land and sea surface states, the subseasonal potential is not fully realized. Complexities arise because drivers depend on the state of other drivers and on interactions over multiple time scales. This study applies statistical modeling to ERA5 data, and explores how nine potential drivers, interacting on eight time scales, contribute to the subseasonal predictability of high summer temperatures in western and central Europe. Features and target temperatures are extracted with two variations of hierarchical clustering, and are fitted with a machine learning (ML) model based on random forests. Explainable AI methods show that the ML model agrees with physical understanding. Verification of the forecasts reveals that a large part of predictability comes from climate change, but that reliable and valuable subseasonal forecasts are possible in certain windows, like forecasting monthly warm anomalies with a lead time of 15 days. Contributions of each driver confirm that there is a transfer of predictability from the land and sea surface state to the atmosphere. The involved time scales depend on lead time and the forecast target. The explainable AI methods also reveal surprising driving features in sea surface temperature and 850 hPa temperature, and rank the contribution of snow cover above that of sea ice. Overall, this study demonstrates that complex statistical models, when made explainable, can complement research with NWP models, by diagnosing drivers that need further understanding and a correct numerical representation, for better future forecasts.