Rapid adjustments to catchment sediment yield following a catastrophic rock-ice avalanche and debris flow, Uttarakhand, India

On 7th February 2021 a massive rock-ice avalanche originating from a mountain ridge in Chamoli District, Uttarakhand, Indian Himalaya, transformed into a fast-moving and catastrophic debris flow which travelled along the Rishiganga, Dhauliganga, and Alaknanda rivers. The flow killed hundreds of people, destroyed or damaged mature and under-construction hydropower projects, and caused severe modification to the channel and wider valley floor landscape, including the destabilising of steep valley sides. Once the flood subsided, rapid post-event analysis revealed that sediments deposited by the debris flow were more than 20 m thick in places, and that the flow was capable of transporting boulders exceeding 20 m in diameter.

The next 12 months are a crucial period for this river system because this is when we predict that newly deposited sediments will be eroded and transported in vast quantities, and we believe that most of this activity will occur within a distance of around 50 km from the avalanche source, which includes four hydropower facilities and riverside settlements and infrastructure. This 're-activation' of sediments left behind by the flood has implications for local hydropower operators, who need to anticipate these elevated sediment loads and plan accordingly to reduce the risk of blockage to dam outlets and tunnels, avoid reduced discharge capacity, and damage to mechanical equipment. In addition, there is a high risk of further valley flank instability as this new drape of sediment is removed and banks that were undercut by the initial flow become more unstable, or undercutting is initiated in new areas. We also anticipate that sediment deposition could also present a hazard where these deposits intersect with valley floor energy and transport infrastructure.

To urgently predict rates and patterns of post-flood channel modification we will use a computer model that is capable of simulating river flow and the erosion, transport, and deposition of sediment. We will run this model for an initial period of one year (including the summer monsoon, which brings an order-of-magnitude increase in river discharge) and we will generate critical summary datasets that can be rapidly communicated to in-country end users. We already have access to most of the data that we require to set up and run the model, and project partners are well-placed to provide missing data that we need to perform initial runs and perform regular checks on model performance.

The work will be carried out by an international team comprised of experts in extreme floods and numerical flood modelling, the hydrology of high mountain landscapes, and community adaptation to (rapid) environmental change. The team includes researchers from the UK, India, Canada and the USA with a collective track record of delivering high quality science to inform real-world decision-making.

Follow-on work will broaden the scope of the work to look at sediment transport and deposition over a much larger area: analysis of satellite imagery shows that the initial sediment plume generated by the flood travelled >150 km in ~24 h and we anticipate that annual re-activation of flood sediment will have significant impacts on the hazard posed by this extreme event.