Search

Abstract Climate cycles fundamentally control surface processes that affect the distribution of water and sediment, and their associated loads, across the Earth's surface. Here, we use a geodynamic model to examine how water loading can affect mantle melt generation in continental rift settings covered by deep lakes. Our modeling results suggest that lake level fluctuations can modulate the timing and rate of mantle melting. A rapid lake level drop of 800 m has the potential to increase mantle melt volumes by enhancing mantle upwelling beneath the rift, whereas a lake level rise can lead to a reduction of mantle melting. The volume of melt produced driven by lake level fluctuations is also dependent on crustal rheology, extension rate, mantle potential temperature, and lithosphere thickness. Our study identifies the importance of water loading for controlling rift processes, while also demonstrating critical links between changing climate, rift evolution and mantle melting.

Abstract We present the Malawi Active Fault Database (MAFD), an open‐access (https://doi.org/10.5281/zenodo.5507190) geospatial database of 113 fault traces in Malawi and neighboring Tanzania and Mozambique. Malawi is located within the East African Rift's (EAR) Western Branch where active fault identification is challenging because chronostratigraphic data are rare, and/or faults are buried and so do not have a surface expression. The MAFD therefore includes any fault that has evidence for displacement during Cenozoic East African rifting or is buried beneath the rift valley and is favorably oriented to the regional stresses. To identify such faults, we consider a multidisciplinary data set: high‐resolution digital elevation models, previous geological mapping, field observations, seismic reflection surveys from offshore Lake Malawi, and aeromagnetic and gravity data. The MAFD includes faults throughout Malawi, where seismic risk is increasing because of population growth and its seismically vulnerable building stock. We also investigate the database as a sample of the normal fault population in an incipient continental rift. We cannot reject the null hypothesis that the distribution of fault lengths in the MAFD is described by a power law, which is consistent with Malawi's relatively thick seismogenic layer (30–40 km), low (