Fault Friction

Fault friction research aims to understand the physical processes governing slip on geological faults, crucial for earthquake prediction and hazard assessment. Current research heavily utilizes machine learning, particularly deep learning architectures like Physics-Informed Neural Networks (PINNs), Long-Short Term Memory (LSTM) networks, and Convolutional Neural Networks (CNNs), to analyze both laboratory experiments and field observations (e.g., seismic noise and GPS data) to improve models of fault behavior, including slow slip events. These advancements offer improved forecasting capabilities for earthquake timing and magnitude, potentially leading to more accurate seismic hazard maps and improved early warning systems.