Sensitivity of Gravity Anomalies to Mantle Thermal Models at Mid-Ocean Ridges

At mid-ocean ridges (MORs), accounting for plate cooling-related long-wavelength gravity anomalies is crucial for accurately estimating variations in crustal thickness and distributions of density anomalies within crust and mantle. Either the classic 1-D plate cooling model (PCM) or numerical models incorporating complex heat advection-conduction and melting processes are commonly employed for this purpose. However, the differences in gravity anomalies predicted by these thermal models have rarely been investigated. In this study, we quantitatively assess the sensitivity of gravity anomalies to various mantle thermal models, considering the influence of mantle rheology-related heat advection and latent heat of melting (LHM). Our results indicate that the PCM systematically overestimates the mantle temperature, predicting more pronounced negative gravity anomalies near the ridge axis compared to numerical models, with peak deviations exceeding 10 mGal and increasing as spreading rates decrease. The comparatively more positive gravity anomalies in numerical models are attributed to efficiency of rheology-related heat advection and LHM. Heat advection generally decreases as spreading rates decrease, resulting in more positive gravity anomalies which modulated by mantle rheologies. In contrast, LHM contributes to less positive gravity anomalies at slower spreading rates due to lower degrees of melting. Specifically, effect of LHM that commonly ignored in gravity modeling plays a more significant role than rheology-related heat advection in influencing axial gravity anomalies at intermediate- to fast-spreading rates. Our systematic work distinguishes the differences in gravity anomalies predicted by various mantle thermal models, providing insights for interpretation of local gravity anomalies at global MORs.