Models of the velocity structure of volcanoes can help define possible magma pathways and contribute to calculating more accurate earthquake locations, which can help with monitoring volcanic activity. However, shear-wave velocity of volcanoes is difficult to determine from traditional seismic techniques, such as local earthquake tomography (LET) or refraction/reflection surveys. Here we use the recently developed technique of noise cross correlation of continuous seismic data to investigate the subsurface shear-wave velocity structure of the Tongariro Volcanic Centre (TgVC) of New Zealand, focusing on the active Ruapehu and Tongariro Volcanoes. We observe both the fundamental and first higher-order modes of Rayleigh and Love waves within our noise dataset, made from stacks of 15min cross-correlation functions. We manually pick group velocity dispersion curves from over 1900 correlation functions, of which we consider 1373 to be high quality. We subsequently invert a subset of the fundamental mode Rayleigh- and Love-wave dispersion curves both independently and jointly for one dimensional shear-wave velocity (Vs) profiles at Ruapehu and Tongariro Volcanoes. Vs increases very slowly at a rate of approximately 0.2km/s per km depth beneath Ruapehu, suggesting that progressive hydrothermal alteration mitigates the effects of compaction driven velocity increases. At Tongariro, we observe larger Vs increases with depth, which we interpret as different layers within Tongariro's volcanic system above altered basement greywacke. Slow Vs, on the order of 1–2km/s, are compatible with P-wave velocities (using a Vp/Vs ratio of 1.7) from existing velocity profiles of areas within the TgVC, and the observations of worldwide studies of shallow volcanic systems that used ambient noise cross-correlation methods.Most of the measured group velocities of fundamental mode Love-waves across the TgVC are 0.1–0.4km/s slower than those of fundamental mode Rayleigh-waves in the frequency range of 0.25–1Hz. First-higher mode Love-waves are similarly slower than first-higher mode Rayleigh waves. This is incompatible with synthetic dispersion curves we calculate using isotropic, layered velocity models appropriate for Ruapehu and Tongariro, in which Love waves travel more quickly than Rayleigh waves of the same period. The Love-Rayleigh discrepancy is likely due to structures such as dykes or cracks in the vertical plane having increased influence on surface-wave propagation. However, several measurements at Ruapehu have Love-wave group velocities that are faster than Rayleigh-wave group velocities. The differences between the Love- and Rayleigh-wave dispersion curves also vary with the azimuth of the interstation path across Ruapehu and Tongariro Volcanoes. Significant azimuthal dependence of both Love and Rayleigh-wave velocities are also observed. This suggests azimuthal anisotropy within the volcanic structures, which coupled with radial anisotropy, makes the Vs structures of Ruapehu and Tongariro Volcanoes anisotropic with orthorhombic or lower order symmetry. We suggest that further work to determine three-dimensional volcanic structures should include provisions for such anisotropy.