Agenda item

(1)       Samantha Howe (Environment Agency Coastal Risk Management) gave a presentation.  The accompanying slides are contained within the electronic papers on the KCC website.

(2)        Ms Howe began her presentation by saying that recently, 60,000 properties in Kent (residential and commercial) had been considered to be at risk of flooding from rivers and the sea.  This figure had decreased according to some studies. An updated figure would be produced once the latest modelling had taken place. 

(3)       Ms Howe then said that computer models were used to help understanding of the areas at risk. The Flood Map for Planning showed the present-day risk of flooding to land from main rivers and the sea without taking into account the presence of flood defences.   The map identified Flood Zone 3 areas where there was a 1% chance of fluvial flooding in any given year and a 0.5% chance of tidal flooding in any given year. Flood Zone 2 consisted of those areas which had a 0.1% chance of flooding in any given year.

(4)       The Environment Agency was also updating the Risk of Flooding from Rivers and the Sea map.  This data set had formerly been known as the “National Flood Risk Assessment” or NaFRA  Unlike the Flood Map for Planning, this map did take account of the presence and condition of flood defences and expressed the residual risk as “very low”, “low”, “medium” or “high risk.”

(5)       Ms Howe said that there were three flood risk models that covered the Kent coast. These were the North Kent coast from Erith to Seasalter, the East Kent coast from Seasalter to Hythe and Romney Marsh (Hythe to Fairlight in East Sussex).   The modelling method was consistent throughout these areas.

(6)       Scenario modelling had also been undertaken for both defended and undefended circumstances in the present day as well as the future, taking climate change into account and including increases in offshore wind speeds.

(7)        Ms Howe moved on to give an overview of the Kent Coastal models.   She said that the detailed model for the East Kent Coast area had been completed in 2018.   The area had not been completely covered before this date and the previous approach had been broad in scale.   The North Kent Coast model had been completed in 2013 but was now being reviewed in the light of the 2013 tidal surge which had taken place just afterwards and a new extreme sea level data set which had come about in 2015.  The Romney Marsh model was also being updated to allow breach modelling to be undertaken so that the implications of any such event could be fully understood.  This review also enabled the inclusion of Broomfield Sands defences, as well as further climate change runs.

(8)       Ms Howe then went on to use the East Kent model of 2018 as a case study. As this was such a large area, it had been split into two domains (Whitstable to Kingsdown and Kingsdown to Sandgate).  She said that no single modelling software package  was capable of collectively assessing the three components of wave transformation, wave overtopping and inundation.  They were therefore assessed separately before the results were brought together.  Wave and wind data was measured by using the Met Office’s WaveWatch.  The tide gauge at Dover, which had been gathered over 80 years, was used to model sea levels.  The Coastal Flood boundary dataset 2011 was used to inform still water levels at offshore boundaries. Once the water levels had been established, bathymetry data and modelling software was used to model the transformation of offshore wave conditions into those nearshore.  The model had been run for 500 events to derive the nearshore conditions at the toe of each defence. The wave overtopping discharges for defended scenarios were calculatedusing a tool called EurOtop (European Overtopping Manual), which used a large database of results from physical modelling tests at 150 locations to derive a prediction of wave overtopping at various types of defences.  The last stage was the creation of the inundation model which used a grid system representing the topography of the study area to represent the movement of flood water across the floodplain.  This was applied for still water level flooding and wave overtopping volumes and enabled the EA to establish which areas were most at risk.   

(9)       Ms Howe then set out the range of model scenarios that had analysed. These ranged from an Annual Exceedance Probability (AEP) of 20% to 0.1% for present day defended and undefended scenarios.  Climate Change scenarios of 0.5 and 0.1% had also been modelled for 2070 and 2115.   All modelling had been undertaken in line with national guidance. 

(10)     Ms Howe described the outcomes from the East Coast Model.  The EA was now able to identify the flood depths and levels, velocities, and the standard of each of the defences.  A forecasting tool had also been developed for each of the three coastal models. This had dramatically changed the information that would be shared with the public and other agencies during flooding events.     

(11)     Ms Howe went on to give examples of work carried out. The biggest tidal surge since 1953 had occurred in December 2013. The data collected during this event had been used to calibrate the results of the model.  She showed photographs and mapping for Sandwich and Margate taken during this period. 

(12)     Ms Howe said that as a result of the detailed modelling study in East Kent, 5,768 residential and 714 commercial properties had been removed from Flood Zone 3 whilst 432 residential and 126 commercial properties had been added to it.  An Area Benefitting from Defence (ABD) had now been created, including 2064 residential and 243 commercial properties. 

(13)     The reason for the great changes in categorisation had occurred because the previous modelling had been incomplete and had used broad scale methods.  The new model had used the latest techniques and data, resulting in a detailed and accurate understanding of the impact of nearshore water on the East Kent Coast.  It used topographical data to identify urban and rural areas and the impact that a flooding event would have on them.  Wave overtopping was now taken into account, which had not previously been the case.   The smaller grid size had also contributed to the detailed results. 

(14)     Ms Howe showed the model results for Sandwich demonstrating that at 0.5% AEP all Sandwich would be under threat in an undefended scenario.  The map also showed the parts of Sandwich under threat when all of the flood defences were in place.  The model also showed the Sandwich 200-year ABD. 

(15)     The model was also able to produce animations. An example of this for the Sandwich area showed how the flood propagated for a 200-year (0.5% AEP) defended scenario over 3 tidal cycles.

(16)     Ms Howe said that another model output was the forecoast tool which allowed the forecast information to be entered prior to and during an event so that the points of greatest expected impact could be identified.  This information would be shared with the public and would also enable evacuation to take place where this was appropriate. 

(17)     Ms Howe then identified the key considerations of the East Kent Coastal Model.   She said that it was used purely for coastal flooding from tides and waves.   A conservative approach was always taken to infiltration, by not taking it into account.  This also applied in respect of shingle beaches, surface water drainage and the sewage network.

(18)     Ms Howe briefly turned to the other two coastal models.  The North Kent Coast Model had recently been updated to include data collected during the December 2013 tidal surge and the most extreme dataset from 2015. The flood map and flood warning areas were due to be updated by the end of 2019, but it was not expected that these changes would make a major difference to the model.  A forecasting tool was also being developed.   

(19)     The Romney Marsh modelling had been updated in 2017 but further work was being carried out to model defence breaches at 14 locations and to incorporate the Broomhill Sands defences.  A forecasting tool was also being developed, in common with the other two areas.   

(20)     Ms Howe concluded her presentation by setting out the practical applications of coastal modelling.  It enabled an understanding of the areas at risk of flooding for both the responders and the wider public.   It assisted local authorities and developers in developing their Strategic Flood Risk Assessments and Flood Risk Assessments.  It also informed asset management and scheme development and was invaluable in incident response work.  The data was shared with multi-agency partners and was also freely available to the public.  

(21)     Ms Howe replied to a question from Mr Brazier by referring to the 2013 flooding. Coastal flooding had taken place from mid-December before the fluvial flooding on Christmas Eve.  In East Kent there were 22 models. Two of these were coastal and the rest fluvial.  The River Stour had been modelled up to Fordwich, where it converged with a fluvial model.   The necessary information existed and needed to be put together.   Coastal and Fluvial models had to be compiled in very different ways, but the local team had all the information available.   In those areas prone to both types of flooding, both sets of data were made available for flood response and strategic planning purposes. 

(22)     Mr Rayner asked how tidal streams were calculated in terms of tides coming through the Channel and also how the differentials were calculated between neap high water and spring high water in the light of the prevailing weather and seasons.   Ms Howe replied that bathymetry data was used in modelling offshore wave conditions to show the seabed as water was coming on shore.  She offered to send further detailed information to him. 

(23)     Ms Howe replied to a question from Mrs Blanford by saying that the models for wave transformation had been run some 500 times.  The wave component was taken from the national database.  This enabled winds and storms to be taken fully into account.  

(24)     Mr Rose asked whether urban development was taken into account. The Kent Fire and Rescue Service had to pump water somewhere when responding to a flooding incident and housing development was having an impact in this regard.   Ms Howe said that different roughness coefficients were applied to the model to represent both rural and urban areas.  This indicated where the water was flowing. In terms of planning applications, developers would request model data so that they could complete a flood risk assessment to demonstrate that the proposed development area was stable and flood resilient. 

(25)     Mrs Hurst said that she had concerns at Birchington where the windfarms were shifting sands around.  She believed this would have a dramatic effect on the models on that stretch of coast.  Ms Howe replied that the data for the models was taken at a particular point.   A static beach profile would therefore be inputted into the model.  The example given demonstrated the need for the models to be regularly reviewed. 

(26)     Ms Howe replied to a question from Mr Lake by saying that the climate change models projected forward for the next 100 years, taking account of the latest information.  The estimates contained within the Met Office’s UKCP18 document were being analysed by the Environment Agency at national level with a particular focus on its modelling implications. 

(27)     Ms Howe said that all data, including for climate change was freely available and that she would be able to provide it to any Member of the Committee on request.

(28)     RESOLVED that Samantha Howe be thanked for her presentation.