Linear point and sound horizon as purely geometric standard rulers

The baryon acoustic oscillations feature (BAO) imprinted in the clustering correlation function is known to furnish us cosmic distance determinations that are independent of the cosmological-background model and the primordial perturbation parameters. These measurements can be accomplished rigorously by means of the purely geometric BAO methods. To date two different purely geometric BAO approaches have been proposed. The first exploits the linear-point standard ruler. The second, called correlation-function model-fitting, exploits the sound-horizon standard ruler. A key difference between them is that, when estimated from clustering data, the linear point makes use of a cosmological-model-independent procedure to extract the ratio of the ruler to the cosmic distance, while the correlation-function model-fitting relies on a phenomenological cosmological model for the correlation function. Nevertheless the two rulers need to be precisely defined independently of any specific observable (e.g., the BAO). We define the linear point and sound horizon and we fully characterize and compare the two rulers' cosmological-parameter dependence. We find that they are both geometrical (i.e., independent of the primordial cosmological parameters) within the required accuracy, and that they have the same parameter dependence for a wide range of parameter values. We estimate the rulers' best-fit values and errors, given the cosmological constraints obtained by the Planck Satellite team from their measurements of the cosmic microwave background temperature and polarization anisotropies. We do this for three different cosmological models encompassed by the purely geometric BAO methods. In each case we find that the relative errors of the two rulers coincide and they are insensitive to the assumed cosmological model. Interestingly both the linear point and the sound horizon shift by $0.5\ensuremath{\sigma}$ when we do not fix the spatial geometry to be flat in $\mathrm{\ensuremath{\Lambda}}\mathrm{CDM}$. This points toward a sensitivity of the rulers to different cosmological models when they are estimated from the cosmic microwave background.