Low-Frequency Distributed Acoustic Sensing reveals transient flow and heat-transfer regimes during geothermal injection

High-resolution diagnostics of flow and heat transfer in deep geothermal wells are commonly constrained by sparse downhole instrumentation and production-logging tools that require well intervention, interrupting steady-state operations. Here, we demonstrate continuous flow profiling and convection-regime identification (pump-driven forced convection vs. buoyancy-driven natural convection) using fiber-optic sensing methods. We used a combination of the Low-Frequency Distributed Acoustic Sensing (LF-DAS) approach, which measures axial strain-rate changes along a fiber, and Distributed Temperature Sensing (DTS) in a 4.1 km MD (Measured Depth) geothermal injection well within Munich, Germany.During shut-in, LF-DAS reveals three persistent, depth-localized natural convection cells, characterized by distinct strain-rate patterns and coincident with rapid warming with depth, as observed in the DTS data. In quasi-steady-state injection, operationally occurring temperature changes as small as 10 mK/min in the injected fluid induce thermo-mechanical deformation along the fiber. These downward-propagating thermal fronts initially reflect pump-driven forced convection and enable flow profiling based on advective heat transport. From a depth of 3580 m MD, these fronts are blurred by the onset of buoyancy-driven natural convection. LF-DAS allows estimation of the plume-shedding frequency, plume height, travel distance, and velocity, all related to the temperature gradient measured with DTS. The accuracy of a threshold criterion for the onset of buoyancy-driven flow based on the temperature gradient is currently limited by the precision of the reference DTS measurement. Across all operational states, values range from 39 to 44 °C/km of true vertical depth.These findings show that fiber-optic sensing can detect fluid-flow pathways, convection behavior, and regime changes without well intervention, thereby improving continuous monitoring and reservoir characterization for sustainable geothermal operation.