We use cookies to make your experience of our website better. You can delete or block cookies, but some parts of this website won't work without them. By using this website you accept our use of cookies.

River Clyde Flood Management Strategy, Scotland

Glasgow City Council commissioned the River Clyde Flood Management Strategy to develop a long term plan to address potential flood risk along the River Clyde.

A key aspect of the study was the development of a catchment-wide modelling approach, including a broad scale model for the upper catchment and a detailed hydrodynamic model extending over a 56km reach covering the main areas at risk of fluvial and tidal flooding.

The Clyde catchment extends mainly in the Southern Uplands, but it also includes areas in Central Scotland with the River Kelvin and River Leven sub-catchments. The total catchment is nearly 3,900km2 at Greenock with a number of large tributaries (Kelvin, Cart and Leven) joining the Clyde downstream of Glasgow city centre. The total catchment is 1,700km2 upstream of Bothwell and 2,000km2 upstream of the City centre.

Upper catchment broad scale modelling

A broad scale model was developed to represent the upper reaches of the River Clyde catchment in order to better understand hydrological processes and to assess the viability of flood management options based on catchment storage.

The model provided a hydrological representation of the River Clyde system routing flow through a simplified depiction of the channel geometry.

It was constructed as a model for the Flood Modeller 1D solver made of inflow boundaries (recorded flow at gauging stations and FEH rainfall-runoff units at ungauged locations) and flow routing units (Muskingum-Cunge).

The model was verified against three historical events including the largest event on record (December 1994, equivalent to a 50 year to 100 year return period).

Hydrodynamic modelling

The 1D model previously developed was comprehensively upgraded to integrate up-to-date topographical survey information. This included LiDAR data providing a 2m digital terrain model as well as multibeam and single beam bathymetry survey data, providing extremely detailed river bed levels.

The model was robust and within generally accepted tolerance. It was verified against 12 events illustrating a range of fluvial and tidal conditions.

The roughness coefficient of the river channel throughout the upstream fluvial reach ranged from 0.025 to 0.048. A value of 0.019 was used for all sections downstream of the tidal weir, where the river is tidally influenced and channelised.

The roughness of the floodplain ranged between 0.035 and 0.120 based on an assessment of the land type/use based on local maps and knowledge of the River Clyde areas (urban areas were represented by a roughness value of 0.120). Comprehensive sensitivity analyses to channel roughness were also carried out.

Model boundary conditions were extensively reviewed, with particular attention paid to considering the joint probability of extreme flows and extreme tides. The analysis was based on previous results and further analysis of recent historic records.

The design tidal boundary at Greenock was adopted following extensive sensitivity analyses. It was based on a spring tide recorded event which exhibits a double peak shape and leads to a significant difference of water level between Greenock and Glasgow.

The double peak (or shoulder) shape results from the shallow water effect when the tide enters the Estuary and has been found to be critical to the way the tide propagates upstream. It can increase the tide gradient, and the higher the gradient the higher the water levels in Glasgow.

It was demonstrated that for a given peak level in Greenock, water levels in Glasgow could be between 0.2m to 1.5m higher depending on the tide shape, amplitude and wind conditions. This was demonstrated by the use of the model and by the analysis of historical records. The choice of the boundary was therefore critical.

The design boundary adopted resulted in a difference of design water level between Greenock and Glasgow Broomielaw of 1m for a 200 year return period. This was conservative but consistent with historic records. By taking into account the effect of wind, using our software, the difference could reach 1.5m which was historically the highest difference.

Flood risk mapping

A series of flood maps were also developed as part of the flood management strategy – these can also be used for development control and planning by Glasgow City Council and adjacent local authorities.

Floodplain maps are an essential tool for land use planning in flood-prone areas. Our 1D solver enables the generation of probability-depth floodplain maps which can make it easier to evaluate flood risk. Probability-depth floodplain maps assist in understanding and communicating the uncertainties involved in flood risk modelling.

Tools available within the our software can help users develop flood maps and allow floodplain managers to better understand flood mechanisms. Our range of analysis methods can also be used to help target funds to apply more detailed and costly modelling and mapping only in higher risk locations.

Privacy & Terms