Aggregate Industries faces more water trouble at Straitgate Farm, and potentially further delays.
Not only does the Environment Agency want further information from the company - see post below - but new boreholes drilled last year show that AI’s model of the maximum winter water table (MWWT) has failed before any quarrying has even started.
So much for all those assurances. In the latest Hydrogeological Assessment - the one that got whitewashed - AI’s consultants promised that:
Groundwater levels would be lower than this elevation for the vast majority of the time, with this level only being reached in extremely wet conditions. 4.2.17
Monitoring over the exceptionally wet winters of 2013 and 2014 allow this surface to be defined with confidence. 6.2.2
More recently, after the EA’s objection, the same consultants again pushed the idea that the MWWT:
builds in a conservatism 2.2.2
But it does no such thing. We’ve had a relatively dry few months in the South West. The most recent groundwater monitoring data for Straitgate Farm, up to and including 8 February 2018, confirms this:
In their objection last year, the EA advised AI:
The accuracy of the maximum winter water level grid may benefit from additional piezometers
A number of additional boreholes were duly drilled - see the 2017 locations on the map below.
The monitoring data from these boreholes show the groundwater level at PZ2017/02 and PZ2017/03 to have recently reached 134.62m and 138.06m respectively.
However, the MWWT at PZ2017/02 and PZ2017/03 has been modelled to 135m and 137m respectively.
In other words, the estimated - because that’s all it is - MWWT level around these points - the level that AI proposes to quarry down to - fails to even allow for normal rainfall, let alone maximum events.
In fact, the MWWT at PZ2017/03 has been modelled ONE METRE BELOW current levels.
PZ2017/03 is at the location of "Test Hole 4". Last year we posted about "Test Hole 4"; about how AI’s own report talked about "the closeness of the water table to ground level in this area". Borehole levels now confirm this - and expose the flaws in the troubled MWWT model; it was only last year that we posted about problems elsewhere too: that AI’s seasonal working scheme can't work as described.
So, what would these 2017 piezometers be telling us if the rainfall was as high as it was during the winter of 2013/14, which was 182% of the 1981-2010 average? How far below water would the MWWT guesstimate be then?
As the EA warned earlier this month - these events are happening with increasing regularity:
Met Office records show that since 1910 there have been 17 record breaking rainfall months or seasons – with 9 of them since 2000.
We’ve warned about all this numerous times before; click on the groundwater label.
The MWWT 'surface' - the base of the proposed quarry - was modelled from just 6 'maximum' water level points across the site. We’ve argued for a long time that this surface cannot be accurate. AI’s consultants have so far refused to confirm the level of accuracy (in +/- m) for this surface - see post below. This issue is especially important given the number of local people and businesses relying on the area for their drinking water and given that AI’s unorthodox seasonal scheme does not propose to leave a 1m buffer of unquarried material above the maximum water table to protect these supplies - the typical minimum requirement elsewhere where such supplies are at risk.
The base of any quarry as proposed is obviously too low, and allows nothing for a margin of safety.
There’s been a post on the right hand side of this blog for some time: The importance of leaving 1 metre...; it’s as relevant now as when it was written in 2015:
Leaving 1m unquarried above the maximum water table should be a precautionary given; 1m allows for a margin of error and a margin of safety, because the maximum water table is not known with accuracy; faulting across the site is not known with accuracy; excavators would not dig with accuracy; future climate is not known with accuracy; future land use pollutants are not known with accuracy.
