We used a diagenetic model to test the hypothesis that manganese-rich layers in gas hydrate-bearing Arctic Ocean sediments are reliable time markers for interglacial periods. In the model, diagenesis is fuelled by two sources of reactive carbon: particulate organic carbon settling to the sediment surface, and methane diffusing up from deep gas hydrate deposits. The model includes oxidation of organic carbon and soluble reduced manganese by oxygen supplied continuously from an invariant bottom-water oxygen reservoir; reduction of particulate manganese by hydrogen sulfide generated through anaerobic methane oxidation; transport of dissolved oxygen and manganese by diffusion; and advective transport of particulate components by burial. Particulate organic matter and particulate manganese are only supplied to the sediment during interglacials. Sulfate reduction is not modeled explicitly; instead, the effect of anaerobic methane oxidation on Mn reduction is simulated at the lower boundary of the model by prescribing that particulate manganese is reduced there to soluble Mn(II). The soluble reduced Mn then diffuses upward and is oxidatively precipitated to Mn(IV) by downward diffusing oxygen. The upward flux of soluble Mn(II) is thus a function of the rate at which particulate manganese is advected into the Mn-reduction layer at the bottom of the model; it is not synchronous with events at the sediment–water interface. Model runs reveal that, under idealized but realistic conditions for the Arctic Ocean, oxidation of upward-diffusing Mn(II) generates post-depositional manganese enrichments that cannot readily be distinguished from the manganese-rich sediment layers that accumulate during interglacials. This compromises the use of manganese-rich layers as proxies for interglacial periods. In contrast, manganese-rich layers may be used as first-order markers of interglacial periods in sediments where gas hydrates or other forms of reactive carbon are absent.< Réduire