Groundwater levels data for the CEH River Lambourn Observatory at Boxford (01/02/2012 to 16/01/2015)

An existing gridded piezometer array originally installed by CEH in February 2012 was numbered 1-15. Supplemental piezometers to the same design specification were added in May 2013 to target observed temperature anomalies (locations 16-21). All locations comprise separate peat (P) and gravel (G) piezometers, with the exception of location 8 where the peat was too thin to complete an installation. Gravel piezometers were screened approximately 2.5-3.5 m below ground level (bgl) while peat piezometers were screened across the entire peat thickness. Peat piezometers in the pre-existing array were installed with the slotted screen extending above ground level, while bentonite was used to seal new piezometers with closed screens above ground level. Chalk (C) boreholes are also located at sites 3, 22 and 23; these were screened at 9.5-10.0, 8.0-9.0 and 5.0-6.0 m bgl, respectively.

At selected piezometers groundwater heads were monitored every 15 minutes using either In-Situ Level Troll 500sTM or SWS DiversTM installed to a consistent depth of 3 m bgl in gravel piezometers and to the base of the peat in peat piezometers. Groundwater heads were routinely checked at all piezometers by manually dipping observed water levels to quality control logged data susceptible to drift.

Piezometers were not anchored to bedrock, with the possible consequence that datums for water levels, taken at the top of piezometers, could move with the expansion and contraction of the peat due to saturation. Though it has been suggested that shallow (<0.5 m) peat generally possesses low compressibility, piezometer elevations were surveyed at periods of low (November 2013) and high (April 2014) water table level to ascertain any vertical movement. Surveys were carried out using Trimble 5600 DRTM total station and Trimble R8TM differential GPS (dGPS). Three total station setups were positioned with dGPS based on line of sight to fixed benchmark points, and to enable best coverage. Selected piezometers were included in the surveys based on line of sight to total station setups.

Results from the surveys showed differences could be grouped dependent on instrument setup. Mean differences for setups 1, 2, and 3 were 0.019, 0.000 and 0.006 m, respectively, which indicate systematic error and are likely due to inaccuracies in the Trimble R8TM GPS receiver due to obstructions, satellite geometry or atmospheric conditions. Variance around the means was not more than 0.003 m, within the accuracy for the Trimble 5600 DRTM total station. Hence, piezometer movement due to peat compressibility with saturation was discounted.