Compound flooding from tropical cyclone-induced sea surge and precipitation in Sri Lanka (C-FLOOD)

Lead Research Organisation: Plymouth University


Coastal communities in the north and east of Sri Lanka (SL) face significantly greater risk of coastal flooding from storm surges associated with seasonal cyclones than those in the rest of the country. These storm surges are essentially local elevations in sea level caused by the weather system which subsequently inundate the land. Storm surges are caused by a combination of: (1) low atmospheric pressure 'lifting' the sea surface (barometric tide) (2) frictional drag of the wind blowing over the sea causing a slope in the water surface (wind stress) and (3) breaking waves transferring their momentum into the water column (wave setup). Hazard maps, to indicate predicted storm surge inundations around the Sri Lankan coastline, were produced by Prof Wijetunge in association with the Disaster Management Centre. The computer models on which these hazard maps were based, were limited to describing only the barometric tide and the wind stress. Subsequent advancements in understanding mean that wave setup can now be included. It is critical to do so, because the wave setup effect may contribute 40% of the surge in some locations i.e. some communities may face a more grave risk than hitherto realised.

The situation is potentially much worse than this however, as scientists are beginning to understand the interaction of storm surges with severe rainfall events which almost always accompany the cyclones in the Indian Ocean region. The mechanism for this so-called compound flooding is that rivers swollen from heavy rainfall are prevented from effectively discharging to the sea due to storm surges coming inland.

To protect against flooding events in the west we are familiar with flood defence structures; Sri Lanka has no such hard-engineered structures. However, they do have natural protective features such as mangrove forests and salt marshes. The potential benefits of mangroves in particular have received some attention since the devastating Boxing Day tsunami of 2004, though the intentional implementation in formal coastal schemes is still in its infancy. Prof Taylor recently received funding from the Global Challenge Research Fund to investigate how design codes might incorporate their effects.

Against this backdrop, the C-FLOOD project will produce a new generation of compound flood hazard maps, based upon state-of-the-art computer modelling that will consider all the storm surge components and the rainfall effect. It will also consider a variety of climate change scenarios which will influence flooding due to predicted rising sea levels. This will be done by Prof Wijetunge at the University of Peradeniya in SL, and Dr Jayaratne at the University of East London, with their related expertise. Furthermore, the protective effects of the natural vegetation will be included in the modelling and maps, by conducting experiments at the University of Plymouth's COAST Laboratory. This will be undertaken by Dr Raby (Plymouth) and Prof. Taylor (University of Western Australia). The C-FLOOD project will focus on three communities that are deemed most vulnerable due to their geography and levels of poverty (associated with the past military conflict). The project team will work with community members in addition to local and regional leaders/administrators to maximise the benefits and uptake of the new hazard maps. Individual localised hazards will also be captured in comprehensive multi-hazard maps for the communities. Dr Kitagawa from the University of East London and Mr Ranawaka of the Coast Conservation & Coastal Resource Management Department have past experience of such activities and will be overseeing these critical aspects.

The final outcome will be improved predictions of flooding inundation, with engagement of the selected communities, leading to improved resilience to compound flooding. The hazard map production techniques and flood impact mitigation methods could then be implemented across other vulnerable communities.

Planned Impact

The principal beneficiaries are SL coastal communities, planners and technical agencies. The outputs from the compound flooding models will produce more reliable and accurate inundation hazard maps. These will improve preparedness and provide information for the development of improved evacuation strategies. Guidance on the use of local coastal protection measures will also provide valuable information for planners. All these aspects will enable better targeting of resources for DRR, reducing loss of life from these natural hazards and thereby creating more resilient coastal communities. Other beneficiaries are:
SL project partners: the DMC who are interested in the compound flood hazard maps as well as aspects of the education and community outreach; the CCS who will gain input for the National Adaptation Plan for Climate Change Impacts in Sri Lanka; the DoM who will benefit from the compound flood forecasting model and knowledge sharing; the CC&CRMD who are interested in the survey of coastal features and the assessment of the non-structural measures; and Lanka Hydraulics Institute will benefit from adopting the model in their consultancy work.
UK project partners: the EA have stated they would like an insight into a novel multi-hazard model and low-cost, sustainable coastal protection approaches; B&V are interested in innovative flood risk solutions; HRW and WSA have interests in mangrove restoration.
In the longer term, outputs will contribute to planning for DRR and resilience in communities both within and without SL. This will be delivered through uptake of the C-FLOOD methodology by our project partners.
The academic impact will be assured through standard routes e.g. high impact factor journals will be targeted for publication, and presentations in appropriate international conferences.


10 25 50
Description Developed a methodology for conducting river flood modelling for areas lacking high precision elevation data through utilization of streamburning of digital elevation models. Mapping of stream networks can improve hydrologic simulations but quality control is needed on those created by citizen science (e.g., OpenStreetMap).

We could not find any strong tropical cyclones affecting the study area in our analysis of weather and climate model output. This limitation (likely due to limited model resolution) makes it difficult to account for the influence of climate change on tropical cyclones, and so should be the focus of further research.

Progress with the research goals of the programme
- Set up a hydrologic model to simulate river runoff for the watersheds emptying into the study area (Valaichchenai lagoon, eastern Sri Lanka) on a high performance computing (HPC) cluster.
- Conducted a comprehensive analysis of weather and climate model output to identify tropical cyclones affecting the study area, to provide weather input data for the compound flood modeling. As an alternative approach, we have also set up the Tropical Cyclone Wind Statistical Estimation Tool (TCWiSE), which uses statistical modelling to simulate wind and rainfall from tropical cyclones.
- Conducted sensitivity testing of HEC-RAS2D and DELFT3D hydrodynamic and wave models for Kalkudah sites, related to bed roughness, grid-spacing etc..
- Recent further analysis of experimental data acquired in WP2 has demonstrated how debris dam formations, arising from storm surges, affect the drag coefficient associated with obstacles in the flow.
Exploitation Route A high precision digital elevation model has been produced for the Valaichchenai lagoon, eastern Sri Lanka. This quality controlled stream network for eastern Sri Lanka can be used for hydrological modelling in this area.
Sectors Environment