Technical Details: Water Level and River Slope

[Note (October 2016):  the page on the original web site was written before the 2014 URS report which investigated the variation in water level along the River Itchen; this is a place holder for a new page which will be written in the light of that report]

chain_mapThe likelihood of water levels reaching a particular height during a flood (often expressed as a “return period”) has been calculated using the time series of measurements at Dock Head.  In order to apply these results in assessing flood risk in St Denys, we need to know not just the height of the local land, but also how water levels at different positions along the River Itchen differ from those measured at Dock Head.

Measurements and Modelling by URS (2012 Report)

As part of a study for Southampton City Council (URS, 2012, Southampton Coastal Flood and Erosion Risk Management Strategy) the environmental consultants URS Infrastructure & Environment UK Ltd measured water levels  at four positions along the River Itchen, shown by the red dots on the map. These were Crosshouse Hard (referred to as “Itchen Bridge” by URS), south of the railway line boardwalk (“Northam Bridge”), near Pettinger Gardens (“Cobden Bridge”), and “Woodmill” boathouse.  The data were obtained from around 22nd September to 28th October 2010.

The water levels measured showed little change in height between the different locations, except for  Woodmill where levels were lower.  However URS considered that the measurements confirmed the results from a hydrodynamic model which predicted that water levels above Northam Bridge would be 10cm or more greater than those at Dock Head.


Water levels at various points along the River Itchen calculated by a hydrodynamic model. The key to the “chainage” is shown in the map above (URS, 2012).

Water levels on 10th March 2008

During the exceptional high water of 10th March 2008, observational estimates of the water level were obtained near Priory Hard, at a Park Bench in Riverside Park, and near Saltmead. Near Priory Hard the water was level with the threshold of a riverside shed which was later surveyed (from an OS benchmark and by GPS) to be 5.76m above Chart Datum. The water level at the river wall just south of Saltmead was also estimated to be about 5.76m. The water level at a bench in Riverside Park was estimated from a photograph to have been 5.68m and this was probably before the peak level, and therefore an under-estimate. These observations (shown in green on the graph) suggested a “river slope” greater than the hydrodynamic modelling (blue) but possibly less than would be derived from the Tide Table (red).


Observations of water level on 10th March 2008 (green), the ABP Tide Table correction for spring tides at Woodmill (red), and water levels derived from the URS hydrodynamic model (blue).

Measurements by URS (2014 Report)

A further study by URS investigated water level variations in the River Itchen (Appendix E in Flood Risk Interactions, St Denys Southampton, URS for the Southampton City Council,  July 2014). Measurements were obtained at Woodmill and near Priory Road Hard from 3rd December 2013 to 1st March 2014, a particularly stormy period including a flood event on 14 February 2014. The mean difference between the peak of the tide at Dock Head and the tidal peak at St Denys (which occurs slight later) was 6 cm and the maximum was 15 cm.  The values for Woodmill were 8 cm and 24 cm.

The variation of water level differences with air pressure, wind speed, rainfall, and river flow were examined but URS found no significant correlation except perhaps a “slight correlation” with air pressure. However their published graphs do show that the largest water level differences occur when the air pressure is low – a time when the risk of storm surges is greater.  The reason for this is not known.

Based on the URS data predicted water level values for Dock Head  have been increased by 10cm when assessing flood risk for St Denys. However if atmospheric pressure is below 980 mb it is possible that a larger value, possibly 20 cm might be more appropriate.