The alkaline-earth metal ion series comprises Be, Mg, Ca, Sr, Ba, and Ra. Calcium (Ca) and magnesium (Mg) are the most abundant alkaline-earth metal ions in nature and their interaction with the mineral surfaces of metal (hydr)oxides (e.g. ferrihydrite, Fh) affects the bioavailability, mobility, and geochemical cycling of many relevant ions. The adsorption of Ca2+ and Mg2+ ions to well-characterized freshly precipitated Fh has not been extensively measured yet in systems with a large variation in pH (5–10), ionic strength (0.01–1 M), and ion adsorption (0.002–2 μmol m−2). Nor have such adsorption data been interpreted with a surface complexation model that regards the structure of the adsorbed complexes and state-of-the-art insights into the surface structure of this nanomaterial. The primary adsorption data collected in this study (M2+/Fe) were scaled in a consistent manner to the surface area of Fh derived with a recently developed probe-ion methodology, before these data were interpreted with the charge distribution (CD) model, using MO/DFT/B3LYP/6-31+G** optimized hydrated geometries to obtain independently the CD coefficients. The pH-dependent adsorption behavior of Ca and Mg is rather similar. Both cations (M2+) form predominantly bidentate inner-sphere surface complexes ([tbnd](FeOH)2Δz0MΔz1), most possibly as a binuclear double corner (2C) complex according to EXAFS. This binding mechanism explains the relatively high H+/Ca2+ exchange ratio and the related pH-dependency of the Ca2+ adsorption. Modeling of the adsorption data reveals and quantifies the surface site heterogeneity of Fh, distinguishing high and low affinity sites for binding M2+ ions. The surface structure of Fh has been evaluated to rationalize this phenomenon and identify possible surface configurations. The increase in the [tbnd]FeOH-M2+ bond strength may be due to a redistribution of charge within specific sets of Fe1 polyhedra at the Fh surface that have in the underlying solid a set of common oxygen ions with an insufficient charge neutralization. According to our surface structural analysis of 2.2–2.8 nm particles, the [tbnd]FeOH site density involved (∼0.3 ± 0.1 nm−2) aggress with the surface site density of high affinity sites found by Ca2+ ion adsorption modeling (0.30 ± 0.03 nm−2). Extending our data analysis to literature data, comprising the full series of alkaline-earth ions, an increase in adsorption affinity with increase in the ionic radius of these cations was found, i.e. Be2+ < Mg2+ < Ca2+ ≈ Sr2+ < Ba2+ < Ra2+, which is opposite to the order of the Hofmeister series observed for other Fe-(hydr)oxides (e.g. hematite, goethite). The affinity trends have been evaluated with a model for ion adsorption energy developed in this study. This analysis suggests that a difference in the order of affinity (logK) can be explained by a different energy and/or entropy contribution of interfacial water in the binding of the metal ions. This points to a relatively strong binding of physisorbed water by Fh in comparison to other Fe-(hydr)oxides where the energy of ligand exchange dominates. For Fh, interfacial water is more strongly bound by high affinity sites than by low affinity sites, according to our model.