The Environment Agency plainly thinks that Aggregate Industries’ consultants can accurately predict the maximum winter water table and base of any quarry at Straitgate Farm to cm precision – across some 55 sloping and geologically faulted acres – with just six data points.
It is of course nonsense. Professor Brassington is clear:
...the MWWT is not a representation of the maximum water table readings for the proposed quarry site 4.16
Dr Rutter was also concerned about AI’s groundwater predictions at Straitgate:
This surface is only a model of reality, and may not represent actual groundwater levels across the site…
There are many private water supplies reliant on the aquifer underlying Straitgate Farm. The EA has a statutory duty to protect and improve the water environment through effective management of water resources and by reductions in pollution. The EA obviously has a short memory about the difficulties of modelling groundwater levels in the Otter Valley:
The Permian and Triassic sandstone aquifer in the Otter Valley is one of Devon's largest groundwater sources, supplying drinking water to 200,000 people.
 |
The National Rivers Authority, the EA’s predecessor, produced a four volume Environmental Study of the River Otter in 1994, based on their network of observation boreholes.
Volume 4 of the report accepts that:
The Otter Valley Triassic aquifer is complex. As a result there is considerable spatial variation in groundwater level response to natural recharge.
Water level responses are influenced by many factors. Those of greatest significance include: * proximity to areas of groundwater discharge; * proximity to the river, where changes in stage are reflected as groundwater level fluctuations; * proximity to the coast, where tidal responses are similarly reflected; * proximity to abstraction points; * thickness of the unsaturated zone, and the presence of low permeability material in the sequence...
But the bombshell is in Volume 2:
A substantial amount of work in this area has already been done by MRM Consultants (1989), who developed a ground water model of the sherwood sandstone aquifer in the Otter Valley. The model was used to assess the likely impact of certain abstraction policies on ground water levels in the aquifer, and resultant changes in the flow regime of the Otter between Fenny Bridges and Dotton. Calibration of the model was difficult to achieve, and modelled ground water levels were sometimes as much as 15m adrift from estimated ground water level after calibration. These discrepancies probably arise from the adopted assumption that the aquifer permeability is constant across the whole catchment (ie that the aquifer is homogeneous and isotropic). In practice, permeability can be highly variable, substantially affecting local transmissivities and rates of water table recession and drawdown. 3.1.6
So, the best that these consultants could produce for the EA – using records collected over almost two decades – was a model of groundwater levels "as much as 15m adrift."
It’s perhaps not surprising. After all, as this EA technical report makes clear:
A model is a tool, not a substitute for reality.
So, why then has the EA so much faith in AI’s model of maximum groundwater levels at Straitgate Farm – given that it was produced by consultants Amec Foster Wheeler (now Wood Group) who were looking to accommodate their client’s desire to maximise the recoverable resource, given that it was produced from groundwater levels recorded from just six boreholes only 12 months after they were drilled?
Why too – in an effort to support AI’s assertions that those groundwater levels at Straitgate can be relied on to be the highest ever – does the EA rely on groundwater levels from an assortment of boreholes on Woodbury Common and elsewhere, an assortment of boreholes up to 93m deep (at Straitgate they're 6-12m deep), including ones in different geology, and one going no further back than 2008? Why does the EA choose to ignore actual evidence from the site? Why does the EA choose to ignore higher baseflows in previous years from one of the streams emanating from Straitgate? Why does the EA choose to ignore groundwater levels from June 1990 in one borehole that shows water levels almost 3m higher than AI’s best guess close to where 'Drillers were surprised they struck water so quickly' in June 2017?
It’s a mystery – particularly as AFW has been so evasive over telling us, DCC and the EA the accuracy of the MWWT. No model can be 100% accurate. This paper says:
…it is vital that the uncertainties of the modelled GWLs are quantified so that land managers and scientists interpreting predictions can determine which features reflect statistically significant variation in groundwater processes rather than model errors.
It was back in January 2016 that we first asked DCC:
Since AI now intends to dig right down to the maximum water table, perhaps you could ask Amec to confirm the specific level of accuracy (in +/- m) to which their maximum groundwater contours are mapped?
AFW were asked again in 2017, but still refused to answer – even to the EA.
In February 2018, the EA asked again:
We recommend that a description of tolerance levels is therefore requested again from the applicant to support the planning application and to provide clarity in advance of the Planning Committee.
If any answer has subsequently been provided that adequately deals with tolerance levels, none has yet been made public. Why have AI’s consultants been so reluctant to provide an answer? Is it the risk that their groundwater predictions would look no more reliable than a chocolate teapot?
In any case, AFW’s model of groundwater levels factors nothing in for climate change. BGS – in "Groundwater and climate change" – forecasts that during this century:
annual rainfall will remain about the same but winter rainfall will increase by 10 to 30% over most of the country
Producing:
increases in average winter [groundwater] levels, by the 2050s under a high greenhouse gas emissions scenario...
Perhaps it's for these reasons that Prof Brassington wrote:
If Devon County Council are minded to grant permission for these proposals, despite the evidence given in this report, they are asked to ensure that an unquarried buffer of at least 3 m is left above the maximum water table to minimise the negative impacts. It has already been demonstrated that the MWWT surface cannot be modelled accurately, neither can the MWWT be reliably adhered to during the operational phase. 6.14
It's important to get this right, particularly for those relying on the site for their drinking water.
It's important for AI too. They wouldn't want to be looking at scenes like this in the future:
Excavator swimming up the river toward its place of birth to mate for the first time: pic.twitter.com/W9KstxM898
— Persian Rose 🇺🇸🇮🇷🏳️ (@PersianRose1) June 24, 2019
