Baseline Scotland sampling

Baseline Scotland was a joint project between BGS and the Scottish Environment Protection Agency (SEPA), which provided new groundwater chemistry data for Scotland. The new data provided essential information to sustainably manage Scotland’s water environment, as required by Scottish, UK and European water legislation, such as the Water Framework Directive. Before Baseline Scotland, very little reliable groundwater chemistry data existed, particularly for trace elements.

Systematic groundwater sampling surveys were carried out across all the major bedrock aquifers in Scotland between 2005 and 2011, as well as many superficial aquifers. Standardised sampling techniques were used to characterise the groundwater chemistry in detail and investigate:

• how long it had been since the groundwater was recharged from rainfall
• how the groundwater has interacted with the aquifer rocks
• whether there is any groundwater pollution

Spatial coverage

Before the Baseline Scotland project began, there was little information on natural groundwater chemistry in Scotland. Existing data was mainly old, of variable quality, limited (often to a few major ions only). It was skewed to areas of groundwater contamination, particularly that related to mining, and to the more productive Scottish aquifers, such as the Devonian aquifers of Fife, Strathmore and Morayshire, and the Permian aquifer in Dumfries.

Since the start of the Baseline Scotland project, 268 new groundwater samples have been collected from aquifers across Scotland. A further 479 samples collected during other projects since 2001, using equally rigorous sampling techniques to the Baseline Scotland samples, were incorporated into the dataset.

Baseline sampling

The sampling surveys collected new data on groundwater chemistry, stable isotopes and residence times. Rigorous well-head measurements (of parameters including dissolved oxygen, redox potential, SEC and bicarbonate) were combined with collecting samples for analysis of:

• major, minor and trace ions
• stable isotopes (δH2, δ¹⁸O and δ¹³C)
• dissolved organic carbon
• CFC (chlorofluorocarbon), SF₆ (sulfur hexafluoride), and other dissolved gases such as CH₄ (methane) where possible

Sample site selection

The criteria for site selection during dedicated Baseline Scotland sampling surveys were:

• a suitable number of samples from each of the studied aquifer types, according to the distribution and areal extent of the aquifers
• samples distributed as evenly as possible over each of the aquifer types studied
• samples that are were, as far as possible, a representative of groundwater in the studied aquifers

The primary criteria for including additional data from earlier studies was the demonstration of suitable robust sampling and analysis procedure. The distribution of baseline samples collated to 2010 is shown in Figure 1.

The sites for all samples considered in the Baseline Scotland project were chosen to be representative of groundwater in the local area. Sources that are very poorly constructed, such as open or uncased boreholes in leaking chambers, and which are close to obvious and significant sources of contamination, like unsecured slurry stores, were avoided.

Each of the sample sites was assessed at the time of sampling, in particular in terms of the source construction and the surrounding land use, to identify any potential sources of contamination that could affect the chemistry of sampled groundwater and the risk of contamination.

Sample sources

Across Scotland as a whole, most of the groundwater samples were collected from boreholes. However, in some regions and particularly in upland areas, springs formed a significant proportion of the sources sampled. In all areas, shallow, large diameter wells were typically the least common source type. In the Midland Valley, a small proportion of samples were collected from mine adits, mine shafts or boreholes intercepting mine workings.

In most cases, the sampled boreholes were pumped regularly in the weeks leading up to sampling. Where boreholes were not pumping on arrival, they were pumped for at least 10 minutes to allow purging before sample collection. The groundwater samples were therefore believed to be as representative of in situ groundwater as possible. Pumped groundwater samples represent the compositions of water entering the borehole over its open-hole section. As such, the sample may represent a mixture of waters with different chemistry, especially where the borehole screen extends over more than one fracture inflow.

Springs and wells typically tap shallow groundwater and have relatively small catchments, so are less likely than boreholes to represent a mixture of groundwaters from different depths. Because springs are constantly flowing, they are naturally purged and the sampled groundwater is therefore likely to be representative of groundwater in the surrounding aquifer. The main reason Baseline Scotland samples were not often collected from shallow wells is that they often form poor sampling points because they are difficult to purge. Wells were only sampled where they were in daily use and the sample obtained was thus likely to be representative of groundwater in situ in the aquifer.

Sampling and analysis procedure

At each sample site, field measurements were made of:

• pH
• dissolved oxygen (DO)
• redox potential (Eh)
• water temperature
• specific electrical conductance (SEC)
• alkalinity

Where possible, pH, DO and Eh were measured in an in-line flow cell to minimise atmospheric contamination and parameters were monitored (typically for at least 10 to 15 minutes) until stable readings were obtained. Where not possible, measurements of water as close as possible to the pump outlet were made in a bucket within one to two minutes of abstraction.

Water samples were collected for subsequent laboratory analysis. Samples for major- and trace-element analysis were filtered through 0.45 μm filters and collected in polyethylene bottles rinsed with sample water before collection. Four filtered aliquots were collected at each site: two were acidified to 1 per v/v with Aristar HNO₃, one for analysis of major cations, total sulfur and silicon by inductively coupled plasma-optical emission spectroscopy (ICP-OES), and the other for a large range of trace elements by inductively coupled plasma mass spectrometry (ICP-MS). A third aliquot was acidified to 1 per cent v/v with Aristar HCl for analysis of arsenic by atomic fluorescence spectrometry (AFS) with hydride generation, to be used if necessary as a check on ICP-MS arsenic analyses. A fourth aliquot was left unacidified for analysis of anions by ion chromatography (Cl, NO3-N, Br, F) and automated colorimetry (NO2-N, NH4-N).

Samples for dissolved organic carbon (DOC) analysis were also collected in chromic-acid-washed glass bottles, after filtration using silver-impregnated 0.45 μm filters. DOC was measured by carbon analyser.

At most of the sample sites, additional water samples were collected in glass bottles for stable isotopic analyses (δ²H, δ¹⁸O and δ¹³C).

Where it was possible to collect an air-free sample direct from the source, samples were also collected for CFC and SF₆ analysis in glass bottles submerged under flowing groundwater to prevent atmospheric contamination.

Most analyses are carried out at the BGS Laboratories, except for some ICP-MS analyses that were carried out by ACME laboratories ion Vancouver, Canada. Charge imbalances were calculated for each sample and reported on, and were typically less than 3 per cent, giving confidence that the analytical results for the major species of all samples were reliable. The BGS laboratories ran a series of QA checks, including analysis of certified standards, to ensure that all analyses were within their prescribed limits.

Analyses of total sulfur are expressed as SO₄ and alkalinity as HCO₃. Analyses of δ²H, δ¹⁸O are expressed as per mil deviations relative to Vienna standard mean ocean water (VSMOW) and δ¹³C relative to Vienna Pee Dee belemnite (VPDB).