Physically meaningful signal extraction from sensors deployed in extreme environments requires a combination of attenuation of confounding inputs and the removal of their residual using decorrelation techniques. In space applications where the resources for physical attenuation are limited, there is a necessity to apply the most effective post-processing analysis available. This paper describes the extraction of the seismic signal from an MEMS microseismometer to be deployed on the surface of Mars. The signal processing, which covers the full bandwidth 1 × 10 -5 Hz to 40 Hz, uses a novel application of sensor fusion through an indirect Kalman Filter in combination with a thermal model of the microseismometer to remove the aseismic contribution of temperature over the frequency range. Owing to the full-band decorrelation, the analysis (based on pre-landing testing in analogous scenarios) produces both a characterization of the microseismomter and a signal processing approach for information retrieval on Mars, along with other planetary and terrestrial planetary deployments.