Monitoring the impacts of methane extraction on the stability of stratification in Lake Kivu
Tomy Doda1,2, Ange Mugisha3, Dieudonné Tuyisabyimbabazi3, Eric R. Mudakikwa3, François Darchambeau4, Martin Schmid1
Affiliations: 1Surface Waters - Research and Management, Eawag - Swiss Federal Institute of Aquatic Science and Technology, Kastanienbaum, Switzerland; 2Faculty of Geoscience and Environment, Institute of Earth Surface Dynamics, University of Lausanne, Lausanne, Switzerland; 3Environment Analytics and Lake Kivu Monitoring Division, Rwanda Environment Management Authority (REMA), Kigali, Rwanda; 4ContourGlobal/KivuWatt Ltd., Kigali, Rwanda
Presentation type: Talk
Presentation time: Friday 08:30 - 08:45, Room S150
Programme No: 6.8.1
Abstract
Lake Kivu is a large East African rift lake renowned for the high amounts of gases (~60 km3 of methane and ~300 km3 of CO2) dissolved in its deep waters. These gases represent an enormous risk, as their hypothetical eruption would far exceed the catastrophic events that occurred at the Cameroonian crater lakes Monoun and Nyos. However, the dissolved methane is also a valuable resource, which has been exploited at commercial scale since 2016 by Kivuwatt, and further expansion of the extraction is foreseen by the governments of Rwanda and the Democratic Republic of the Congo. While stable density stratification currently prevents gas release, the lake's vertical density structure may evolve, especially due to the water flows induced by the methane extraction activities. Our study aims to assess how methane extraction has modified the density stratification in the lake since 2016, and how these changes compare to the natural changes observed before 2016 as well as to previous model predictions. We analyzed 2524 Conductivity-Temperature-Depth (CTD) profiles collected between 2008 and 2022 by monitoring programs of the Rwanda Environment Management Authority (REMA) and KivuWatt. The results show that the reinjection of degassed water following methane extraction formed a new saline layer just above the main density gradient and that the combination of extraction and injection water flows led to a slow drawdown of that gradient. These changes qualitatively confirm the model predictions, although they seem to proceed faster than expected, highlighting the need for continued monitoring and evaluation.