Application of the coherent thermodynamic model [W. Holzapfel and S. Klotz, J. Chem. Phys. 155, 024506 (2021)] for H2O ice Ih to the more detailed data for D2O ice Ih provides better insight into the contributions from quenched proton disorder and offers a new basis for understanding the apparent differences between the data for thermal expansion measured with neutron diffraction on polycrystalline samples [A. Fortes, Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 74, 196 (2018) and A. Fortes, Phys. Chem. Chem. Phys 21, 8264 (2019)] and macroscopic dilatation measurements on single crystals [D. Buckingham et al., Phys. Rev. Lett. 121, 185505 (2018)]. The comparison points to contributions from defects effecting the two techniques in different ways. The uncertainties in thermodynamic data due to the contributions from proton disorder and additional defects are compared with the "reference data" [R. Feistel and W. Wagner, J. Phys. Chem. Ref. Data 35, 1021 (2006)] for H2O ice Ih.