Another post, another tale of groundwater modelling problems – this time PZ2016/001, PZ2016/002, PZ2017/04 and PZ2017/05.
Firstly, PZ2016/001 and PZ2016/002: Why are levels in these two piezometers important? Because they formed the basis of a meeting between Aggregate Industries and the Environment Agency last year, a meeting where AI tried to convince the EA of the merits of its seasonal working scheme – in particular how much unsaturated material would be left in situ during summer working down to the maximum winter water table.
It was agreed following the meeting that AI would
1. Produce a contour map showing the difference between the grid and the recorded maximum water levels across the site
In order to provide clarification to point 1. above, the actual recorded maximum water levels to date for piezometers 2016/001 and 2016/002 (142.1mAOD and 143.1mAOD respectively) were uploaded to a copy of the existing MWWT model… When compared to the original ‘grid’ this effectively pulled the contours down in the central parts of the site revealing the extent of unsaturated mineral that would be left below the working base. A contour map has been produced…
That contour map was the subject of the previous post And these contours are a joke too – with errors the height of houses!
At the meeting, AI acknowledged that:
the maximum water levels recorded at piezometers 2016/001 and 2016/002 may not actually be representative of maximum levels as these piezometers did not exist in the wet years of 2013 and 2014.
But then tried to persuade the EA that:
…the data does not suggest that the highest water level [at 2016/001] will be higher than 142.1m AOD… Groundwater level at 2017/002 [sic] seems to be even less responsive and is even less likely to approach the ‘apparently’ conservative elevation of the MWWT at this location.
It was also discussed at the meeting on the 10th August that although piezometers 2016/001 and 2016/002 had only been reading for a short period they followed the same pattern of little difference between summer and winter as did the 'dry' western part of the site. This suggests that groundwater in the middle of the site is still draining rapidly eastwards to where it backs up against the Otter Sandstone and drains to the lower water table. As such it is likely the water level in these piezometers may never reach substantially higher levels.
Has this been the case? Of course not.
This year – which as we have already posted was unexceptional for rainfall – PZ2016/001 reached 142.64m AOD and PZ2016/002 143.98m AOD. This may be below the MWWT, but given that the winter and spring of 2013/2014 saw 35% more rainfall in the South West than 2018, who knows where these levels might reach – certainly not AI, as they have already clearly demonstrated.
But it’s not just AI that’s been found out. Last November, in correspondence with the EA following a request for additional information, AI's water consultants Amec Foster Wheeler – despite not yet having the benefit of borehole data – were trying to convince the Agency that in the northwest of the site "very little (if any) water level fluctuation is observed":
It is worth reiterating that groundwater levels vary very little in the northwest of the site and the BSPB is effectively dry for much of the time. Groundwater levels do not vary much here because water drains rapidly through the formation to where it accumulates in the east and south east (upgradient of the fault) where large thicknesses of unsaturated material exist to accommodate this flow.
PZ2017/04 and PZ2017/05 – shown on the map below – have now provided a limited amount of groundwater data. Does it back up Amec’s assertions? Of course not. PZ2017/04 has so far shown a groundwater variation of 1.34m and PZ2017/05 1.88m – which is plainly neither very little fluctuation nor material that is effectively dry for much of the time.
As we’ve said before, AI and Amec appear prepared to say anything in order to win approval.
