We have long argued – as have others, including Professor Brassington and Dr Rutter – that Aggregate Industries' model of the maximum groundwater levels – the MWWT, the base of any quarry, denoted by the contours below, originally guesstimated across 55 sloping acres from 'maximum' water levels recorded in just 6 boreholes – is not accurate. How on earth could it be?
Meanwhile, in the land of fairy tales, Aggregate Industries’ consultants – Amec Foster Wheeler now Wood – have over many years been peddling the story that the MWWT – the determinant of the recoverable resource and hence the proposal’s monetary worth – was generated by a clever and super accurate computer algorithm, NOT by the hand of some easily persuadable human.
Lots of people were taken in by this story, including the Environment Agency.
Documents were produced on the subject, following a meeting we had with the Environment Agency in September 2017, when Aggregate Industries was asked to supply:
A description of the tolerance levels and interpolation method used to produce the ‘Maximum Winter Water Table’ grid
At this point, consultants had every chance to come back and say we used a sharp pencil and our immense brain power. But they didn’t. Consultants said:
Four alternative grid interpolation methods were selected: Simple linear Kriging; A Radial Basis Function; Inverse Distance to a Power; Triangulation with linear interpolation. Each of these methods represents an exact interpolator in which the input data point will be preserved and respected by the interpolation algorithm…. The grid determined by Radial Basis Function was therefore adopted as being a more conservative preferred method for estimating the maximum water table distribution.
Clearly, consultants accepted they were estimating, but no indication of the accuracy of this estimation was supplied, so in February 2018, the Environment Agency asked again. The consultants came back with more nonsense:
The description of how the MWWT grid was calculated explained that two of the interpolation methods used produced realistic results (in line with expected hydrogeological behaviour and professional judgement). One method, ‘Radial Basis’ produced a grid that was generally higher than the other grid prepared by a Kriging method. As the MWWT was subsequently based on the higher of the two interpolated surfaces (i.e. Radial Basis interpolation) this could be considered to represent the upper limit of interpretation, with the Kriging interpolation representing the lower limit of interpretation and therefore the difference between these two realistic grids could be considered as reflection of the ‘tolerance’.
It was a bogus argument. The Environment Agency asked more questions, and were told:
The generally higher Radial basis grid is subtracted from the generally lower Kriging basis grid to give an indication of “tolerance”….The Radial grid formed the basis for the final composite MWWT (shown in Figure 5), which also underwent final adjustment (upwards) to take further account of hydrogeological professional judgement.
More gibberish, but the Environment Agency was satisfied. In June 2018 it wrote:
We have reviewed the document and we are satisfied that it answers our questions about the derivation of the Maximum Winter Water Table grid.
But all the radial basis and kriging talk was baloney. A mathematician has analysed the MWWT, and has concluded "it is not a radial basis function, per se":
If a sensible, well conditioned parameter and an appropriate generating function are chosen, a radial basis function will be a well behaved mathematical surface with continuous, smoothly differentiable contours between interpolation points... The only conclusion is that AI have defined their MWWT by hand.
And indeed, if we look through the myriad of documents, we find a map showing "hand contoured groundwater levels for February 2014 (mAOD)" submitted with the 2015 application, and again thereafter:
If we overlay these hand drawn contours over the MWWT (before recent changes) we find miraculous agreement over large areas. Check out the agreement along the 141mAOD contour, the 145m contour, the 148m contour, the 150m contour; the hand drawn contours are grey, the MWWT contours are blue.
No mathematical model, no matter how clever, radial basis, kriging or otherwise, could produce such agreement from just – as it was originally – 6 numbers. A similar overlay with the MWWT recently supplied (and recently adjusted for elevated groundwater levels along the eastern boundary) shows corresponding agreement – see below – to areas not recently tweaked.
This means that consultants have not only been spinning a long and convoluted yarn to the Environment Agency, but have used whatever professional judgement they have – which may of course be biased, compromised, subject to errors or simply lacking – to guesstimate the levels that their fee-paying client may dig down to; in other words, deciding exactly how much money Aggregate Industries could lay its hands on. Anybody smell a massive conflict of interest?
And knowing that the MWWT has been drawn by hand, on the back of a proverbial fag packet for all we know, means there is certainly NO basis for claiming cm accuracy of these contours, and EVERY basis for not trusting them, and EVERY need to add a very large safety margin above them to accommodate "professional judgement".
Who can trust a single word Aggregate Industries' consultants say any more?
