The National Aeronautics and Space Administration’s (NASAs) Interior exploration using Seismic Investigations, Geodesy and Heat Transport (InSight) lander successfully touched down on Mars in November 2018, and, for the first time, a seismometer was deployed on the surface of the planet. The seismic recordings reveal diurnal and seasonal changes of the broadband noise level that are consistent with variations of the local atmospheric conditions. The seismic data include a variety of spectral peaks, which are interpreted as wind-excited, mechanical resonances of the lander, resonances of the subsurface, or artifacts produced in the measurement system. Understanding the origin of these signals is critical for the detection and characterization of marsquakes as well as for studies investigating the ambient noise. We identify the major spectral peaks up to 9 Hz, corresponding to the frequency range the most relevant to observed marsquakes. We track the variations in frequency, amplitude, and polarization of these peaks over the duration of the mission so far. The majority of these peaks can readily be classified as measurement artifacts or lander resonances (lander modes), of which the latter have a temperature-dependent peak frequency and a wind-sensitive amplitude. Of particular interest is a prominent resonance at 2.4 Hz, which is used to discriminate between seismic events and local noise and is possibly produced by a subsurface structure. In contrast to the lander modes, the 2.4 Hz resonance has distinctly different features: (1) a broad and stable spectral shape, slightly shifted on each component; (2) predominantly vertical energy; (3) temperature-independent peak frequency; (4) comparatively weak amplification by local winds, though there is a slow change in the diurnal and seasonal amplitude; and (5) excitation during all seismic events that excite this frequency band. Based on these observations, we suggest that the 2.4 Hz resonance is the only mode below 9 Hz that could be related to a local ground structure.