Explanations for the topography flanking rift systems remain somewhat controversial, which mainly stems from variable assumptions concerning the flexural state of rifted continental lithosphere. A number of studies showed that rift-flank topography is significantly undercompensated implying that the topographical gradient is supported by the flexural strength of the lithosphere This hypothesis requires that flexural strength is maintained during extension and that elastic rebound triggered by tectonic unloading of the lithosphere during rifting will be regionally distributed with respect to the kinematically produced loads. The flexural rebound model predicts a proportional relation between the height of rift flanks and the depth of adjacent basins. This has been used to estimate the basin geometry and distribution of crustal extension responsible for the formation of rift basins. Applying the flexure rebound model to the western branch of the East African Rift shows that basin depth and rift-flank height in the majority of the basins could be explained by heaves of 3 - 10 km across the rift-bounding faults, predicting a maximum rift-flank topographical relief of 1 - 2 km, maximum basin depth of 7 - 8 km and flexural strength ranging between 21 - 38 km. A rheologically strong crust is supported by teleseismic and local earthquake hypocenters, which occur throughout a 5 - 40 km depth range suggesting that the high-angle border faults of the rift penetrate at least to the base of the crust.
In addition to flexural rebound by crustal unloading, thinning of the lithosphere adds heat to its base causing density variations that represent a buoyant load. Lateral temperature gradients cause small-scale convection in the mantle. The combination of lateral conduction and the advection of mantle material heat flux can cause considerable uplift in the case of passive rifting. It appears that it is only the low-viscosity material which is incorporated in the convective flow, and the more material is swept into the flow, the greater the uplift of the rift mountains is. Narrower rift zones and initially thicker lithosphere should both lead to greater uplift. For an active rifting mechanism, the amount of uplift depends largely on mantle advection. Alkaline volcanism due to decompression melting beneath the rift would add to thermal uplift of the flanks. Sills intruded into sedimentary strata, or magma underplated at the base of previously thinned crust can locally augment this uplift. Active rifting might be the reason for the southward propagation of rifting in East Africa. Nonetheless, it is also obvious that, while thermal processes add to rift-flank uplift, flexure is still important because it supports the rift-flank topography (i.e. freezes it in) even after lithosphere temperature has been reduced by cooling. A complicating factor is that lateral changes in the thermal structure of the lithosphere will have an immediate affect on its rheological behaviour, and ultimately its strength. The feedback between thermal and mechanical processes is therefore difficult to understand in a conceptual, qualitative way.
The role of erosion aiding rift-flank uplift in the East African Rift has not been studied in any detail so far. Estimates showed that the erosional denudation of the Livingstone escarpment of the Malawi Rift at the southern end of the western rift branch is lower than 200 m. It is therefore commonly believed that erosion is negligible and that flexural readjustment to erosion does not play a role. In the currently uplifting Rwenzori Mountains, however, deeply incised valleys, extreme rainfall and warm-based glaciers strongly suggest high erosion rates. Especially glaciation can dramatically increase erosion rates. In regions with abundant precipitation glaciation can create an "erosional buzzsaw" that removes the majority of rock mass above the glacial equilibrium-line altitude quickly. We think that the role of surface processes needs to be addressed in greater detail if we are to unravel the origins of uplifted rift flanks in general and that of the Rwenzori Mountains in particular.