The role of stellar model input in correcting the asteroseismic scaling relations

Aims. This study investigates the variability of the theoretical correction factor, fI ν, used in red giant branch (RGB) scaling relations, arising from different assumptions in stellar model computations. Methods. Adopting a commonly used framework, we focused on a 1.0 Mastar and systematically varied seven input parameters: the reference solar mixture, the initial helium abundance, the inclusion of microscopic diffusion and mass loss, the method for calculating atmospheric opacity, the mixing-length parameter, and the boundary conditions. Each parameter was tested using two distinct but physically plausible values to mimic possible choices of different evolutionary codes. For each resulting stellar model, we computed the oscillation frequencies along the RGB and derived the large frequency spacing, I ν0. The correction factor fI ν was then calculated by comparing the derived I ν0 with that predicted by the uncorrected scaling relations. Results. We found substantial variability in fI ν across the different models. The variation ranged from approximately 1.3% in the lower RGB to about 3% at log ga =a 1.4. This level of variability is significant, as it corresponds to roughly half the values typically quoted in the literature and leads to a systematic change in derived masses from 5% to more than 10%. The most significant contribution to this variability came from the choice of atmospheric opacity calculation (approximately 1.2%), with a smaller contribution from the inclusion of microscopic diffusion (approximately 0.4%). Conclusions. These results indicate that the choice of the reference stellar model has a non-negligible impact on the calculation of correction factors applied to RGB star scaling relations.