I enclose 4 representation forms in respect of various representations that I have on your Policy BEX9, The Spindlewood Drive site. This site is also the subject of a current Planning Application with Reference RR/2017/1705/P.

My first representation concerns the surface water drainage of the site through a SuDS system to an attenuation pond. I have submitted an 8 page report accompanied by 5 Figures. I have therefore attached a paper copy of that report to my representation and I would want them to be considered together:

1. No sufficient Appropriate Assessment has been lodged in respect of BEX9 where the attenuation pond is within 60 metres and the housing is within 175 metres of the Designated area and the pond has to be constructed substantially within ground water to which the Environment Agency have said no infiltration must be allowed. The Pevensey Levels are a Special Area Of Conservation (SAC), RAMSAR site and SSSI designated under the Habitats Directive.

The Court of Justice of the European Union has found that the Habitats Directive requires a full Appropriate Assessment of the effects on European sites listed in the Directive. Reference the case of People over Wind and Sweetman vs Coillte Teoranta in the Republic of Ireland.

2. When the attenuation pond water level is depleted after a few days without rainfall but plenty of wind to dry the pond through evaporation, as can occur at any time; but the ground water is still high the buoyancy thrust would be considerably greater than the dead weight remaining in the pond itself. The result with up to 3.8 tonnes of buoyancy thrust would be to lift and deform the geotextile membrane at the base of the pond over up to % of the pond's width. The initial silt trap at the pond would be eliminated and totally compromised. The next significant rainfall event would result in overtopping the ponds spillway and possibly backing up of run-off in the lower housing area and overland flow could occur. Untreated flood water would enter the Cole Stream and thence the Pevensey Levels.

A considerable depth of additional ballast would be required within the pond geotextile membrane to achieve a satisfactory equilibrium between weight and buoyancy. The effective invert level would become very close to era ADD. With the requisite weight of ballast and the pond full of water the complete pond structure would probably weigh more than 6 tonnes per square metre (counterweight, soil/gravel and water). Over parts of the pond area the bearing weight would be assymetrical. The pond structure would need to sit on a very firm base to disperse this weight to whatever the sub strata could bear. The bottom of the construction would therefore be well below the general water level of the watercourses in the Pevensey Levels (which the drainage board aim to keep at 1.0 to 1.3 metres above Ordnance datum) and below the water level of the Cole Stream.

The operation of pumping to a depth of below Ordnance Datum would be bound to disturb the hydraulic regime of the Cole Stream and the eastern Pevensey Levels generally once pumping reaches below the normal level of the watercourses in the Pevensey Levels. A back flow in the water courses of the eastern Levels is certain.

ATTACHED REPORT:

Second Report on an Insufficiently ballasted Attenuation Pond
(issued after study of the topographic survey)

Spindlewood Development site - SUDS - Bexhill

by Geoffrey P Lawson C. Eng. MICE. November 2018

Executive summary

1. The lowest level of the Pond field is about 2.65 metres above Ordnance Survey datum (AOD) of mean sea level. The field rises at a slope of ~4.5% to a height of 9.2 metres AOD (paragraphs 2- 3).
2. The Cole Stream which runs down the lower edge of the field flows full in the winter and overflows from time to time. There is evidence of a hydraulic gradient within the ground water at the end of the winter when ground water levels are replenished, of around 2.9% within the ground. The ground water level therefore rises to between 5.58M and 6.90M ADD at the top of the pond field where the actual ground level is 9.2 metres AOD. The winter ground water level would therefore be 2.3 metres below the surface which is consistent with British Geological Survey data. (paragraphs 7-10)
3. The pond with a base invert level of from 3.00 to 3.18 metres AOD would have to be constructed almost entirely within the ground water level. See Figures 2 and 3.
4. A 500mm clay lining above an impermeable geotextile membrane as proposed would not be sufficient to withstand the winter ground water buoyancy pressure. Construction in this manner as proposed would have the result that in the first instance of high ground water in the winter the pond base would fail, become deformed and the pond would lose capacity resulting in flooding of untreated water to the Pevensey Levels. This may only gradually be apparent. See figure 4
5. To properly ballast the pond with a sufficient clay and rock counterweight would require up to 5 Tonnes of counterweight per square metre of pond area in a deep excavation. The pond would require a very firm base to sit the pond on to avoid further movement and consequent deformation of the pond and tearing of the impermeable membrane due to low strength substrata. The excavation would require de-watering with consequent risks to the Pevensey levels from a very large quantity of silty water. (Paragraphs 30 - 31).
6. Comparison with the Ashridge Court site is entirely spurious on account of the relative size, very much lower level and nearness to the Designation area of the Pevensey Levels. (paragraphs 34-35)
7. I conclude that in this case and at this location the depth of construction required for the pond if properly ballasted will cause major disturbance to the water regime of at least the eastern part of the Pevensey Levels during construction;
8. the effects on the watercourses and fluvial ecosystem of the eastern part of the
Pevensey Levels SAC, RAMSAR and SSSI would be disastrous. No amount of
additional engineering could mitigate these effects sufficiently to achieve 'no
adverse effects on the Pevensey Levels'.
9. The development of the land adjacent to Spindlewood Drive should not be
considered any further. The application for Outline Planning permission should be
refused.

Existing ground levels.

1. This report on the Spindlwood Drive SuDs pond has been prepared after studying the topographic survey of the site at 1:500 scale carried out by Messrs Landtech Surveys on 13th September 2018 and the letter from Messrs Herrington Consultants dated 16th October 2018.
2. The topographic survey covers three traverses across the field in which the pond (or wetland) is to be located and one traverse of part of the upper field in which the lowest housing is to be located. In the field with the pond proposals one travers lies along the bank of the Cole Stream from a point 38 metres south west of the footbridge at Cooden Moat up to the line of the hedgerow between the two fields which is some 115 metres north east of the footbridge. Along the banks of the stream the ground height varies from 2.65 M AOD at the lower end to 4.32 M ADD at the higher end. At the footbridge the ground height is locally 2.36 M ADD.
3. The ground rises towards the north west in the pond field at about 4.5% (the gradient varies from ~ 4.46% to ~4. 7%) to the tree belt at the top of the field which separates it from the static Caravan site above. At this top end of the field the ground height varies from 8.1 M ADD to 9.2M ADD with the higher ground at the northern end.
4. A new analysis has been made in this report of the effects of these ground levels on the SUDS drainage system in the Spindlewood site and brief details are set out below. Information on the hydraulic conditions on land adjacent to the site has also been used and this is also described in this paper.
5. Figure 1 shows the relevant part of the layout plan of the Spindlewood site and the location of two cross sections through the site. The footpath between the footbridge on the Cole Stream at Cooden Moat and the caravan park is taken as a base line and cross section X-X. The positions of the other cross section where it crosses the Cole Stream is shown relative to the base line X-X, thus Y- Y is 90 metres away to the north. These sections are shown on Figure 1. Section X - X and Y - Y have been named to easily distinguish them from A - A and B - B used in the Herrington report. A convention has been adopted in this report to refer to the right-hand side of the pond as north and the left-hand side as south and to refer to the edge nearest the Cole stream as the front of the pond and the edge nearest the caravan site as the back of the pond.

6. Herrington's assumed Ground water level. At a meeting on 29th August 2018 the
Developer undertook to look at a 'worst case' scenario for the pond as an alternative to undertaking ground water monitoring over the winter period of 2018 - 2019. The winter ground water level is however assumed by Herrington to be ~3.0m AOD and flat. For some unexplained reason this water table estimate is lcm above the water level of 2.9 metres ADD found in the Cole Stream on the survey day (13th September 2018, at the end of a very dry summer). The existing height of the bank of the stream is shown in the survey to be ~4 metres AOD and the stream is known to flow full and occasionally over full in the winter. A flat water table would result in very little if any movement in the ground water. In 7. 1 of the Herrington letter variable green field run-off rates are shown in the event of rainstorms of different return periods ranging from 1 in 2 years to 1 in 100 years (i.e. of different lengths and intensity of rain storm). As well as overland runoff this implies differing hydraulic gradients to push the rainwater through clayey soil into the stream at differing rates. A ground water level of ~3 metres across the site is emphatically NOT a 'worst case'.

7. Winter ground water levels. The Cole Stream lies within a distinct valley with sides at a gradient of 4.5% to the North-west (within the site) at cross section Y - Y and about 3.1% to the South-east side towards Maple Walk. The ground water in the valley of the Cole Stream percolates slowly to the stream through silty clay soil. The fields act like a sponge during wet weather absorbing rainfall in the ground and only allowing run-off at a slow rate. For run-off to occur there must be a hydraulic gradient in the ground meaning that ground water levels further up the sides of the valley will be higher than the stream water level. This can easily happen without necessarily causing flooding below dependant on the hydraulic conductivity of the soil. The hydraulic conductivity of the clay overlying the site is tow and therefore in the winter the hydraulic gradient may be relatively steep.

8. Local evidence of the hydraulic gradient can be found on the eastern side of the stream in the woodland surrounding Cooden Moat and the woodland at the back of properties in Old Harrier Close which abuts the site to the east. These areas are waterlogged in winter and are consistent with a winter hydraulic gradient of at least 0.029 (or 2. 9%). Further evidence is found in Hazelwood Close immediately to the north seen in Figure 1 where the lower parts of the back gardens get flooded after rainfall in winter. This has reportedly occurred several times over the last 11 years (for example in the winter of 2013/2014 which was very wet for 5 consecutive months). At these times ground water enters a garage approximately 13 metres away from the stream and only some 80cm above the top of the bank. This suggests a winter hydraulic gradient of 0.06 (or 6%) which is relatively steep.

9. In this report it has been assumed that the lesser gradient is generally applicable and a winter hydraulic gradient of 0.029 (2.9%) has been used for this investigation. It is also observed that in wet winters the stream is running full or nearly full. This means that at section X-X the ground water level at the top of the field at the end of a wet winter will be 5.58M AOD (where the ground level is about 8. 1M). i.e. 2.5 metres below ground level.

10. At section Y - Y where the ground is about 8.85 M AOD the ground water level at the end of a wet winter will be about 6.9M AOD. i.e. 2.2 metres below ground level.

11. In summer the hydraulic gradient and ground water levels may be significantly lower.

12. The level of the attenuation pond. The water surface in the pond will vary with weather conditions but with the present arrangement as shown in Herrington report it will always basically be horizontal. The levels of the attenuation pond as now proposed are given in the following table.

13. Table 1 Attenuation pond section levels as proposed

Table 1 included which can be seen here:
http://www.rother.gov.uk/CHttpHandler.ashx?id=31276


14. There is a requirement for no infiltration to be allowed between the pond with only partially treated estate run-off water and the ground water The pond and its clay lining is proposed to be wrapped in an impermeable geotextile membrane. Herrington suggest lining the pond with 500mm of clay inside the impermeable geotextile membrane. Where the pond is below the winter ground water level there will be ground water buoyancy forces to be considered. This 500mm clay layer would take the invert of the deepest sections of the pond down to 2.5M ADD thus increasing the ground water buoyancy thrust on the base.

15. For the pond/wetland to contain reed beds and other suitable flora to treat the run-off water the invert in the low flow channels would be at a level of 3.18 metres allowing for a depth of growing medium and gravel below the maximum water depth of 500 mm. There is also the matter of a deeper section to allow for any residual silt to drop out before the exit via the flow control device. This invert level of 3.18 metres is well below the winter ground water level at this point.

16. Because the pond would be within the anticipated winter ground water level, ground water would be exerting an upward buoyancy thrust on the pond's base geotextile liner. In the winter this buoyancy thrust could be more than 3 tonnes per square metre at the back of the pond. This is due to the difference in level between the invert (3.0 metres) and the level where the reprofiled bank breaks above the anticipated winter ground water level of 6.4 metres AOD shown on figure 3, and the fact that a cubic metre of water requires a displacement of 1 tonne.

17. Figure 2 shows diagrammatically the section X -X which passes through the south side of the pond more or less on the line of the exiting footpath running between the footbridge at Cooden Moat and the Caravan site. The Figure 2 is prepared with an exaggerated vertical scale with a ratio of 10 vertical to 1 horizontal. This is typically done in Engineering diagrams to enable vertical features to be more easily appreciated. At this section the pond configuration is proposed to be comprised of marginal terraces which are designed to flood infrequently as shown in the 'Wetland layout plan' attached to the Herrington letter. The deeper sections of the pond do not feature in this southern part. These terraces will therefore dry out quite quickly. It should be remembered that wind especially, evaporates water from the pond surface more rapidly that it evaporates ground water for the simple reason that pond water is exposed whilst ground water is covered.

18. The pond normal level after only light rain or drizzle would be ~4.18M ADD (After a storm this might rise to 4.68M). Allowing for the soil and gravel growing medium and the 500 mm clay lining the pond base level and enclosing geotextile membrane would be 3.18 metres. In these circumstances the dead weight of the pond structure would be ~ 1.7 tonnes per square metre. The ground water buoyancy thrust could reach a force of 2.2 tonnes per square metre at the back of the pond where ground water levels would reach 5.4 M AOD as indicated on Figure 2. So here the pond's weight would be insufficient to counter the ground water force in a wet winter and the membrane would be deformed by ground water buoyancy forcing it up. About 40% of the ponds width is likely to be affected at this southern end of the pond.

19. Figure 3 shows the configuration of the pond at section Y-Y. Again this section is shown diagrammatically with the same exaggerated vertical scale. At Y - Y the ground level at the back of the pond is 8.2 M AOD and the winter ground water level is 6.4M AOD. The pond levels are as set out in Table 1 ranging from the extreme TWL of 5.18M to the Silt trap bottom of 3.00M AOD. Rainwater would enter the pond via the initial filter strip at 'A' and leave the system via the outfall and flow control device at 'B'.

20. At times when the pond water level is depleted after a few days without rainfall but plenty of wind to dry the pond through evaporation, as can occur at any time even in winter or early spring; but the ground water is still high the buoyancy thrust would be considerably greater than the dead weight remaining in the pond itself. The combination of a very small amount of residual water say 50mm, 300mm wet soil, 200mm gravel and 500mm of clay would weigh about 1.75 tonnes per square metre. The result with up to 3.8 tonnes of buoyancy thrust would be to lift and deform the geotextile membrane at the base of the pond over up to % of the pond's width.

21. The ground water buoyancy per square metre on the geotextile membrane with 500mm of enclosed clay lining is 3.2 Tonnes psm at the very back, 3.8 Tonnes psm at the silt trap, 2.6 Tonnes on the low flow channels and 2.8 Tonnes at the back of the final basin.

22. When not holding storm water and subject to the evaporating effect of wind the pond dead weight including the residual water and the 500mm clay liner would be 1.4 Tonnes psm at the very back, 2.0 Tonnes psm in the silt trap, ~1.70 Tonnes psm in the low flow channels and 2.25 tonnes in the final basin area including the water.

23. The result of the imbalance between dead weight and winter ground water buoyancy force at section Y - Y is shown in Figure 4. This figure is again diagrammatic with an exaggerated vertical scale as indicated. A new equilibrium between dead weight and buoyancy force would be established by nature. The effect would be to deform the geotextile membrane and push the base of the pond up substantially for up to % of the pond's width as shown in Figure 5.

24. Figure 5 shows that the overall capacity of the pond would be reduced to less than half, even considering the 500mm of freeboard in the design.

25. The overall effect of lifting and deformation of the geotextile membrane is shown in Figure 5. More than half of the pond area could be affected and the hydraulic capacity of the system would be severely reduced. Importantly the initial silt trap would be eliminated and totally compromised. The next significant rainfall event would result in overtopping the ponds spillway and possibly backing up of run-off in the lower housing area and overland flow could occur. Untreated flood water would enter the Cole Stream and thence the Pevensey Levels. In addition the heave and deformation of the membrane would create a strong likelihood of tearing of the membrane and deformation of the clay lining which might allow ground water to enter the pond further eroding its capacity.

26. An insufficiently ballasted pond would certainly lead to a failure of the pond structure as soon as the ground water level rose and the ground water pressure reached a buoyancy level exceeding the ballast weight within the geotextile liner. This is shown in Figure 5. The result would be an overtopping of the spillway and possibly the bund enclosing the pond and a failure due to backing up of run-off in the Swale. Such an event would be a disaster for the Cole Stream and the Pevensey Levels.

27. Ballasting the pond. Additional weight would need to be placed within the liner. This additional ballast could be a mixture of rocks and clay. The rocks would need to be kept away from the liner itself lest any sharp edges cause a perforation of the liner in time. The whole pond impermeable liner would need to be lined with clay with rocks in the middle. Two tonnes of this mixture would require more than 1 metre depth. This additional depth would of course give rise to further ground water displacement of a tonne. It is likely that 2.5 metres depth of the rock clay mixture would be required within the pond geotextile membrane to achieve a satisfactory equilibrium between weight and buoyancy. The effective invert level would become very close to zero AOD. To contain all of this within one layer of geotextile membrane would require in excess of 6750 square metres of membrane.

28. Weight of the pond when full. With the requisite weight of ballast and the pond full of water the complete pond structure would weigh more than 6 tonnes per square metre (counterweight, soil/gravel and up to 1 tonne of water per square metre). Over parts of the pond area the bearing weight would be assymetrical as there would be greater weight in some parts of the pond than in other areas. The pond structure would need to sit on a very firm base of consolidated granular material to disperse this weight to whatever the sub strata could bear. The bottom of the construction would therefore be well below the Ordnance datum. This is below the general water level of the watercourses in the Pevensey Levels (which the drainage board aim to keep at 1.0 to 1.3 metres above Ordnance datum) and below the water level of the Cole Stream.

29. De-watering. The construction of a lined pond could only be effectively done in dry conditions. But this has to be done well below ground water level. This would be
impossible without de-watering the site by pumping out the ground water and to avoid an endless hydraulic loop situation, a sheet piled and braced cofferdam would be required down to below the level of the base of construction.

30. Herrington Consulting have estimated that the pond should be 5750 square metres in overall area to allow for future anticipated climate change, intense rainfall and 'urban creep'. To de-water an area of 5750 sq. m. to a depth of ~ 6 metres would require the disposal of a very large volume of muddy and silty water (possibly as much as 20,000 cubic metres). It remains to be decided where this pumped water should be discharged to. A temporary silt trap as suggested by Aspect Ecology, to deal with 20,000 cubic metres of silty discharge water would itself be a major structure. Although a sheet piled cofferdam would keep most water out it would certainly leak and continuous pumping of sitty ingress water would be necessary throughout the construction period.

31. The operation of pumping to a depth of below Ordnance Datum would be bound to disturb the hydraulic regime of the Cote Stream and the eastern Pevensey Levels
generally once pumping reaches the invert level of the cofferdam which would be below the normal level of the watercourses in the Pevensey Levels. A back flow in the water courses of the eastern Levels is certain.

32. In other locations and circumstances this could be done, but in this case and at this location the effects on the watercourses and fluvial ecosystem of the eastern part of the Pevensey Levels SAC, RAMSAR and SSSI would be disastrous. No amount of additional engineering could mitigate these effects sufficiently to achieve 'no adverse effects on the Pevensey Levels'.

33. In the Swales section drawings attached to Herrington's letter the section from C2 to C3 rises at a gradient of 5% which is steeper than the CIRIA recommendations and terracing needed for housing would possibly cause spring lines to develop which may result in ground water contamination.

34. Comparison with the Ashridge Court site circumstances. The Ashridge court site Appeal has been allowed very recently. Comparison of the Ashridge Court site with the Spindlewood Drive site reveals that the two sites are very different in respect of the SuDs proposals. The Inspector's report notes at 35 'An Appropriate Assessment should be proportionate to the case'. In the case of Ashridge Court, the site is 1.5 ha in extent with a total of 31 dwellings only is proposed. The site is 430 metres from the nearest point of the Pevensey Levels designation area. The SuDs attenuation pond for the Ashridge Court development has an invert level of +23.44M AOD and ground water is not higher than +21.5M AGO (i.e. 2 metres below the invert). The pond will be about 20 metres above the Pevensey Levels.

35. In the case of the Spindlewood Drive site the area is 8.07 ha in extent and the number of dwellings proposed is 160. The site is only about 30 metres from the Pevensey Designation area and the SuDS attenuation pond is proposed with an invert level of 3.00 - 3.18 metres ADD. The winter ground water level is likely to be up to 6.9M AOD so the pond will have to be constructed on sloping ground almost entirely within the winter ground water. The counterweight and base of the pond will have to be constructed below ground water level in an area that wilt require de-watering by pumping. The Pevensey Levels are at a ground level of between 1 and 2 metres ADD so the pond invert is only 1 metre above the Levels. The risks to the water quality and fluvial ecosystems of the Pevensey Levels is a whole order of magnitude greater than is the case with the Ashridge site.

36. Conclusion. I have shown the following;
. If insufficiently ballasted the pond would fail due to deformation of the base within a very few years and possibly in the first winter period when ground water pressure at the back of the pond exceeded the ballast capacity;
. A sufficiently ballasted pond would require a very deep excavation;
. that there are substantial risks of contamination of ground water from failure in the
lining of the pond due to the size of the pond, any deformation of the base due to
differing weights and to ground water;
. the depth of construction required for the pond if properly ballasted will cause major disturbance to the water regime of at least the eastern part of the Pevensey Levels during construction;
. the silt from discharge water during de-watering is a potential major risk to the
Levels;
. For the regraded bank to withstand lateral water pressure will require a substantial
soil engineered retaining structure which in the event of failure of the pond base
would also be likely to fail;
. the pond will require a very firm base to distribute the weight of the structure plus
water to enable the subsoil to bear it without deformation or movement. This will
add to the construction depth required;
. There is a possibility of spring lines developing as a result of terracing of the ground in the upper part of the site. This could give rise to ground water contamination;
. The development of the land adjacent to Spindlewood Drive should not be
Considered any further. The application for Outline Planning permission should be refused.

Figures included:
Figure 1: http://www.rother.gov.uk/CHttpHandler.ashx?id=31057

Figure 2: http://www.rother.gov.uk/CHttpHandler.ashx?id=31058

Figure 3: http://www.rother.gov.uk/CHttpHandler.ashx?id=31059

Figure 4: http://www.rother.gov.uk/CHttpHandler.ashx?id=31060

Figure 5: http://www.rother.gov.uk/CHttpHandler.ashx?id=31061