Due to decreases in seawater pH resulting from ocean acidification, permeable calcium carbonate reef sands are predicted to be net dissolving by 2050. However, the rate of dissolution and factors that control this rate remain poorly understood. Experiments performed in benthic chambers predict that reefs will become net dissolving when the aragonite saturation state (Ωa) in sea water falls below ∼3, as underlying reef sediments start net dissolution due to lower saturation states in the pore water. We used flow-through reactors to investigate the rate of dissolution at various Ωa at the pore scale. The sediment became net dissolving at Ωa = 1.68–2.25, which is significantly greater than 1. This indicates that the bulk pore water does not represent conditions at the site of dissolution, and dissolution probably occurs in microniches inside porous sand grains. Measured dissolution rates were much higher under oxic conditions than anoxic conditions, but were not affected by the addition of carbonic anhydrase. Analysis of δ13C-CO2 produced in the flow-through reactors revealed a bias in the conventional alkalinity anomaly method under anoxic conditions, showing that some of the CO2 attributed to metabolism by may actually be derived from carbonate dissolution. This deviation likely originates from alkalinity consumption by fermentation, which masks the alkalinity generated by dissolution. Therefore, dissolution rates determined by alkalinity changes in reef sands with anaerobic metabolisms may underestimate actual values.