Analysis: the 10th March 2008 Event ( part 2)

On 10th March 2008 one of the highest water levels yet recorded for Southampton resulted in partial flooding of roads near Priory Road Hard and at Woodmill.  Examining what caused the high water on that day will perhaps help us anticipate potential flood events in future.  Part 1 of this analysis considered  the effect of local atmospheric pressure and and a larger scale storm surge. In Part 2 we will consider the effect of changes in the timing of the tide and also the duration of the surge.

The rate at which the tide can move along the coast is determined by the water depth.  If the water level is raised (due to the effects of low pressure and the accompanying storm surge) the tide is able to travel along the coast quicker and the tide will arrive at Southampton sooner than is shown in the Tide Table.

Why did the residual water level appear to vary rapidly?

resid_tcorr_1to0at14In Part 1 of our analysis of the 10th March event we noted that the difference between observed and predicted water levels varied rapidly during the event. These rapid variations in the apparent magnitude of the storm surge were caused by the tide arriving earlier than predicted.

The graph shows in blue the observed difference between the water level and the predicted tide.  If we assume that the tide time was about 1 hour early at 6 in the morning but that this difference had disappeared by one hour after high tide, we obtain the residual values shown by the red line. The match is not perfect but does suggest that changes in the timing of the tide can produce variations of similar nature to those observed.

Why it is important to understand changes in tidal timing

Because the timing of the tide can change, it is often misleading to try to predict how high the water will be at high tide by checking the water levels earlier in the day.

hm_predictFor example, the graph shows the tide table prediction of water level on the 10th March 2008 in light blue and the observed water level in dark blue.

Marked in red on the graph is what would have happened if someone had checked the ABP reading of the Dock Head tide gauge at 7am on that morning. They would have seen that the tide was about 1.4m higher than expected, and they might have thought there would be 6.2m of water at high tide and probably panicked!

In contrast, someone checking a bit later at 9am (shown in green) would only have predicted a 5.2m tide! Why was this? At 7am the tide had come in early and was much higher than expected. However 9am was during the period of “young flood stand” so difference was not so large.  In the event, the actual high water level was between these predictions, 5.6m.

These variations in the difference between observed and predicted tides caused by changes in timing are often evident in the Storm Surge Prediction for Portsmouth. The times of high tide are marked by circles and those are the only times at which the true value of the storm surge can be estimated.

Did the flood on 10th March reach full potential?

Using the land height data and assuming the height of the water in the river was about 5.6m (as measured at Dock Head) we can predict the area of Priory Road near the Public Hard that would have been flooded. This is shown in the diagram with the predicted water depth shown in cm.  For orientation, selected house numbers are shown in red.  Note that in reality the public hard area and many of the riverside gardens were also flooded.

The predicted flooding of the road is similar to, but slightly overestimates, the actual flooded area. For example there was flood water in the gutter outside No. 84 Priory Road but it was not quite deep enough to reach the pavement. This over estimate of the flood is despite evidence that the water level in the river sloped and at St Denys was 15 to 20cm higher than Dock Head – which should have caused a greater area to flood.

The over-estimate may be because there on this occasion the highest water only last perhaps around 10 minutes. There may not have been time for the flood to reach maximum depth before the tide began to recede. A fall of 0.3m immediately after the time of first High Water without further raise at the predicted time of second High Water is noted in the ABP Tide Table as sometimes associated with strong south-westerly gales, as occurred on that day.