Wednesday, 11 February 2015

Storm Surge 2013 : One Year On - Part One : Modelling the Surge

This is the first post of a four week mini-series looking back at the storm surge of 5 December 2013. The surge caused extensive flooding along the East Coast of the UK but our focus has been on the area immediately around the Humber, and you can read our reaction shortly after the storm surge in this older post. Over the next few weeks we will be discussing the research that has been ongoing since the event, how it affected and continues to affect local residents and businesses, the community resilience that has been built and finally we consider the damage done to Spurn Point and its potential future.

This week the focus will be on a paper recently published by myself, colleagues at the University of Hull, the Association of British Ports (ABP) and the local Environment Agency (EA), which stemmed directly from the storm surge. The paper is free to view until 28 March 2015, after which you will require a subscription to Estuarine, Coastal and Shelf Science to view.

Hull's flood defences overtopping on 5 December 2013 (by @tom_coulthard)

Estuaries are very complex environments. There is a lot going on, beginning with the inputs of often several rivers, and the sea in the form of tidal flows. The relative influence of these on when and where the water and sediment moves in the estuary depends on the tidal cycle and the discharge levels of the rivers. It is a to and fro tug of war between these for influence within the estuary.

If that was not complex enough, there are secondary flows within the estuary. River water is fresh and sea water is salty, making the two flows a different density along with water that is mixture of the two in between. The two water types are often different temperatures too, again resulting in different densities and inducing flows from more dense to less dense regions. All of the flows are influenced by Coriolis forces, the deflection of water flow caused by the rotation of the Earth. The shape of an estuary also influences flow, and in combination with the influences above, estuaries like the Humber often show two channels along the bottom - one resulting from tides coming in and one from tides going out. Finally, overlain on these are the winds, waves and pressure influences of the weather.

This makes estuaries very complex and turbulent, and this turbulence can form a layer of thick sediment laden water to form along the bed - this basal mud layer clings to the bottom and effectively lubricates water flows along the estuary and shields the bed from erosion and deposition.  The salinity of the water also causes fine sediment to clump together in a process called flocculation which makes them behave like larger sediment particles.

It is commonly thought that to model the processes in an estuary then you need to account for all of these processes, but doing so is incredibly computationally expensive. It is possible to do, but even on expensive and powerful machines it often takes several days to model a single tidal cycle. Trying to use them to predict the future of an estuary several decades in the future would be almost impossible. Our approach was to use a simpler model, CAESAR-Lisflood, which has been widely used for a similar purpose on rivers for over a decade, to try and model the Humber Estuary successfully without all of this detail.

Animation showing the CAESAR-Lisflood model simulating the 2013 storm surge and associated flooding.

It was during this process when the storm surge struck and the focus of our research switched. We had already tested the model's ability to reproduce tidal flows - rapidly and at small timescales - so we soon tried applying the data recorded by ABP during the surge. This showed that the model could also reproduce the location and extents of the flooding on that night. This was using the latest information on the Humber's flood defences provided by the EA. The quickness of the model to process the data would make it suitable for producing numerous possible scenarios based on live and forecast data, and potentially help predict the extent of future flooding before it occurs.

This work is ongoing. Next week I will highlight how local residents and businesses were affected by the flooding, as discussed at the Humber Conference of December last year. If you wish to view this paper you can do so here.

Skinner, C. J., Coulthard, T. J., Parsons, D. R., Ramirez, J. A., Mullen, L., and Manson, S., 2015. Simulating tidal and storm surge hydraulics with a simple 2D inertia based model, in the Humber Estuary, UK. Estuarine, Coastal and Shelf Science. 155, 126-136 doi:10.1016/j.ecss.2015.01.019

No comments:

Post a Comment