The Southern Rocky Mountains (SRM) have the highest regional elevations, but a comparable crustal thickness to that of the Great Plains. Isolating the contributions of the crust and upper mantle in supporting the topography of the SRM has been a topic of considerable controversy for the past two decades, leading to contrasting hypotheses. In order to constrain the relative contributions from the crust and mantle, reliable models of the physical state of the crust and upper mantle are necessary. Here we employ a recently developed multiobservable Bayesian inversion approach to study the present‐day physical state (i.e., Vs,Vp, density, Vp/Vs, crustal thickness, and temperature) of the crust in a region encompassing the western‐central United States. We image highly heterogeneous temperature and compositional structures, which combine in a nonintuitive way with total crustal thickness to create the observed topography and control other geophysical signatures. In the SRM, topography is supported by a combination of crustal and mantle contributions. South of 38°N, mantle sources (both static and dynamic) constitute the main support to the high elevations, whereas to the north of this latitude, crustal structure provides the main contribution. In the Colorado Plateau, Arizona transition zone, and Basin and Range provinces, most of the topography seems to be supported by mantle sources (within the lithospheric mantle and/or dynamic). At middle to lower crustal depths beneath the SRM, we image high temperatures, high Vp/Vs values, anomalously low crustal velocities, and low densities. These observations correlate with high electrical conductivities and the location of recent volcanism, suggesting the occurrence of melt/fluids. We do not find low Vp/Vs values at middle/low crustal depths beneath the SRM that may suggest felsic, weak lithologies.