Difference between revisions of "OR/15/048 Regional hydrogeochemistry"

From Earthwise
Jump to navigation Jump to search
[checked revision][checked revision]
Line 1,409: Line 1,409:
 
Values of δ<sup>13</sup>C are an index of the evolution of the dissolved inorganic carbon system (DIC) (Clark and Fritz, 1997<ref name="Clark 1997">CLARK, I, and FRITZ, P. 1997. ''Environmental Isotopes in Hydrogeology''. (Boca Raton, USA: Lewise Publishers.)</ref>; Darling et al., 2005<ref name="Darling 2003"></ref>). Groundwater in calcareous sedimentary terrain acquires DIC through reaction with soil CO<sub>2</sub> and reaction with carbonate minerals in the soil and aquifer. The δ<sup>13</sup>C composition of DIC is fundamentally governed by interaction between soil CO<sub>2</sub> (δ<sup>13</sup>C ~ -26‰) and carbonate minerals (δ<sup>13</sup>C ~ 0‰), resulting in a δ<sup>13</sup>C-DIC value of around ~ -13‰, but further modified depending on the nature of the carbonate system. If open (typical of unconfined aquifer conditions), further exchange with soil-derived CO<sub>2</sub> will result in δ<sup>13</sup>C-DIC values <-13‰. If closed (typical of confined conditions), δ<sup>13</sup>C-DIC will start to evolve towards the rock composition, resulting in δ<sup>13</sup>C-DIC values rising progressively above -13‰. The compositions in the Corallian groundwaters range between -18.0 and -12.0‰ with an average of -15‰. Most of the sites are therefore relatively depleted in δ<sup>13</sup>C, which suggests a relatively immature DIC system of generally limited residence time. This conclusion supports the δ<sup>18</sup>O and δ<sup>2</sup>H results insofar as they suggest no significant modification (i.e. mixing with old groundwater) since recharge. However the most δ<sup>13</sup>C enriched sites tend to be close to the valley floor and probably signify slightly older waters. (Figure 54)
 
Values of δ<sup>13</sup>C are an index of the evolution of the dissolved inorganic carbon system (DIC) (Clark and Fritz, 1997<ref name="Clark 1997">CLARK, I, and FRITZ, P. 1997. ''Environmental Isotopes in Hydrogeology''. (Boca Raton, USA: Lewise Publishers.)</ref>; Darling et al., 2005<ref name="Darling 2003"></ref>). Groundwater in calcareous sedimentary terrain acquires DIC through reaction with soil CO<sub>2</sub> and reaction with carbonate minerals in the soil and aquifer. The δ<sup>13</sup>C composition of DIC is fundamentally governed by interaction between soil CO<sub>2</sub> (δ<sup>13</sup>C ~ -26‰) and carbonate minerals (δ<sup>13</sup>C ~ 0‰), resulting in a δ<sup>13</sup>C-DIC value of around ~ -13‰, but further modified depending on the nature of the carbonate system. If open (typical of unconfined aquifer conditions), further exchange with soil-derived CO<sub>2</sub> will result in δ<sup>13</sup>C-DIC values <-13‰. If closed (typical of confined conditions), δ<sup>13</sup>C-DIC will start to evolve towards the rock composition, resulting in δ<sup>13</sup>C-DIC values rising progressively above -13‰. The compositions in the Corallian groundwaters range between -18.0 and -12.0‰ with an average of -15‰. Most of the sites are therefore relatively depleted in δ<sup>13</sup>C, which suggests a relatively immature DIC system of generally limited residence time. This conclusion supports the δ<sup>18</sup>O and δ<sup>2</sup>H results insofar as they suggest no significant modification (i.e. mixing with old groundwater) since recharge. However the most δ<sup>13</sup>C enriched sites tend to be close to the valley floor and probably signify slightly older waters. (Figure 54)
  
[[Image:OR15048 fig51.jpg|thumb|center| 400px| '''Figure 51''' O and H stable isotopic composition of the Corallian aquifer. WMWL = world meteroric water line (Craig, 1961<ref name="Craig 1961">  CRAIG, H. 1961. Isotopic variations in natural waters. ''Science'', Vol. 133, 1702–1703    </ref>).]]
+
[[Image:OR15048fig51.jpg|thumb|center| 400px| '''Figure 51''' O and H stable isotopic composition of the Corallian aquifer. WMWL = world meteroric water line (Craig, 1961<ref name="Craig 1961">  CRAIG, H. 1961. Isotopic variations in natural waters. ''Science'', Vol. 133, 1702–1703    </ref>).]]
  
 
==Chemical variations with depth==
 
==Chemical variations with depth==
Line 1,419: Line 1,419:
 
In many cases the major ions follow similar trends over the period they were measured. For example at one site about 2km south-east of Helmsley [SE 63 82], Ca, Cl, and Mg all increase gradually, but steadily, over a period of 10 years (Figure 55). At the same site this is mirrored by a general decrease in the HCO<sub>3</sub> concentration (Figure 56) over the same time period. There is no obvious reason for these changes.
 
In many cases the major ions follow similar trends over the period they were measured. For example at one site about 2km south-east of Helmsley [SE 63 82], Ca, Cl, and Mg all increase gradually, but steadily, over a period of 10 years (Figure 55). At the same site this is mirrored by a general decrease in the HCO<sub>3</sub> concentration (Figure 56) over the same time period. There is no obvious reason for these changes.
  
[[Image:OR15048 fig52.jpg|thumb|center| 400px| '''Figure 52''' Regional variation of δ<sup>18</sup>O in the Corallian aquifer.]]
+
[[Image:OR15048fig52.jpg|thumb|center| 400px| '''Figure 52''' Regional variation of δ<sup>18</sup>O in the Corallian aquifer.]]
[[Image:OR15048 fig53.jpg|thumb|center| 400px| '''Figure 53''' Regional variation of δ<sup>2</sup>H in the Corallian aquifer.]]
+
[[Image:OR15048fig53.jpg|thumb|center| 400px| '''Figure 53''' Regional variation of δ<sup>2</sup>H in the Corallian aquifer.]]
  
 
At the same site near Helmsley there is an increase in NO<sub>3</sub>-N over a period of six years. This is not an unusual trend in areas dominated by agriculture, as NO<sub>3</sub>-N is a common diffuse pollutant derived from agricultural practices. At this site the concentration increases from values around the drinking water limit (11.3 mg L<sup>-1</sup> NO<sub>3</sub>-N)at an average of 1.7 mg L<sup>-1</sup> each year, to twice the drinking water limit over a period of seven years (Figure 57).
 
At the same site near Helmsley there is an increase in NO<sub>3</sub>-N over a period of six years. This is not an unusual trend in areas dominated by agriculture, as NO<sub>3</sub>-N is a common diffuse pollutant derived from agricultural practices. At this site the concentration increases from values around the drinking water limit (11.3 mg L<sup>-1</sup> NO<sub>3</sub>-N)at an average of 1.7 mg L<sup>-1</sup> each year, to twice the drinking water limit over a period of seven years (Figure 57).
  
[[Image:OR15048 fig54.jpg|thumb|center| 400px| '''Figure 54''' Regional variation of δ<sup>13</sup>C in the Corallian aquifer.]]
+
[[Image:OR15048fig54.jpg|thumb|center| 400px| '''Figure 54''' Regional variation of δ<sup>13</sup>C in the Corallian aquifer.]]
[[Image:OR15048 fig55.jpg|thumb|center| 400px| '''Figure 55''' Temporal changes in Ca, Mg, and Cl at a site near Helmsley.]]
+
[[Image:OR15048fig55.jpg|thumb|center| 400px| '''Figure 55''' Temporal changes in Ca, Mg, and Cl at a site near Helmsley.]]
  
 
Another four of the EA’s monitoring sites show similar NO<sub>3</sub>-N trends, and are distributed throughout the Vale of Pickering. This finding is consistent with previous reports which document the high concentrations of NO<sub>3</sub>-N and PO<sub>4</sub><sup>3-</sup> in the region (EA, 2009<ref name="EA 2009">EA. 2009. River Basin Management Plan Humber River Basin District, Environment Agency.</ref>; Natural England, 2015<ref name="Natural 2015">        NATURAL ENGLAND. 2015. National Character Area Profile: 26. Vale of Pickering www.gov.uk/natural-england.  </ref>).
 
Another four of the EA’s monitoring sites show similar NO<sub>3</sub>-N trends, and are distributed throughout the Vale of Pickering. This finding is consistent with previous reports which document the high concentrations of NO<sub>3</sub>-N and PO<sub>4</sub><sup>3-</sup> in the region (EA, 2009<ref name="EA 2009">EA. 2009. River Basin Management Plan Humber River Basin District, Environment Agency.</ref>; Natural England, 2015<ref name="Natural 2015">        NATURAL ENGLAND. 2015. National Character Area Profile: 26. Vale of Pickering www.gov.uk/natural-england.  </ref>).
Line 1,431: Line 1,431:
 
A more unusual temporal variation was observed in groundwaters at a site near Kirbymoorside [SE 70 87]. Concentrations of the major elements were relatively constant between 1995 and 1998, most elements increased until 1999, where they stayed relatively constant, with the exception of three lower values recorded on 9/11/99, 17/10/03 and 20/10/05. The exceptions to this are Na and K, which mirror this trend (Figure 58 and Figure 59). There is no clear reason for this trend.
 
A more unusual temporal variation was observed in groundwaters at a site near Kirbymoorside [SE 70 87]. Concentrations of the major elements were relatively constant between 1995 and 1998, most elements increased until 1999, where they stayed relatively constant, with the exception of three lower values recorded on 9/11/99, 17/10/03 and 20/10/05. The exceptions to this are Na and K, which mirror this trend (Figure 58 and Figure 59). There is no clear reason for this trend.
  
[[Image:OR15048 fig56.jpg|thumb|center| 400px| '''Figure 56''' Temporal changes in HCO<sub>3</sub> at a site near Helmsley.]]
+
[[Image:OR15048fig56.jpg|thumb|center| 400px| '''Figure 56''' Temporal changes in HCO<sub>3</sub> at a site near Helmsley.]]
[[Image:OR15048 fig57.jpg|thumb|center| 400px| '''Figure 57''' Temporal changes in NO<sub>3</sub>–N at a site near Helmsley.]]
+
[[Image:OR15048fig57.jpg|thumb|center| 400px| '''Figure 57''' Temporal changes in NO<sub>3</sub>–N at a site near Helmsley.]]
[[Image:OR15048 fig58.jpg|thumb|center| 400px| '''Figure 58''' Temporal changes in Ca, Cl, SO<sub>4</sub>, HCO<sub>3</sub> at a site near Kirkbymoorside.]]
+
[[Image:OR15048fig58.jpg|thumb|center| 400px| '''Figure 58''' Temporal changes in Ca, Cl, SO<sub>4</sub>, HCO<sub>3</sub> at a site near Kirkbymoorside.]]
[[Image:OR15048 fig59.jpg|thumb|center| 400px| '''Figure 59''' Temporal changes in Mg, Na, K and NO<sub>3</sub>–N at a site near Kirkbymoorside.]]
+
[[Image:OR15048fig59.jpg|thumb|center| 400px| '''Figure 59''' Temporal changes in Mg, Na, K and NO<sub>3</sub>–N at a site near Kirkbymoorside.]]
  
 
==References==
 
==References==
  
 
[[category:OR/15/048 Baseline groundwater chemistry: the Corallian of the Vale of Pickering, Yorkshire | 08]]
 
[[category:OR/15/048 Baseline groundwater chemistry: the Corallian of the Vale of Pickering, Yorkshire | 08]]

Revision as of 15:12, 11 September 2015

Bearcock, J M, Smedley, P L and Milne, C J. 2015. Baseline groundwater chemistry: the Corallian of the Vale of Pickering, Yorkshire. British Geological Survey Internal Report, OR/15/048.

Corallian strata are dominated by calcite, and it is likely that groundwater chemical compositions will be dominated by interactions with this mineral. The data set obtained from the study of the Corallian aquifer of the Vale of Pickering is discussed below in terms of data distribution (i.e. summary statistics) as well as spatial distribution.

Major constituents

Statistical data for major ions, field-determined parameters and stable isotopes are summarised below (Table 5, Figure 8 to Figure 10). In addition spatial variations are presented in this section. It should be noted that where an element was not analysed in the EA sample, only the 24 new BGS analyses are presented. The summary statistics reported are derived from a range of methods, depending on the proportion of non-detects (see Data availability and analytical methodology). To produce the box plots in this report (Figure 6), ROS statistics have been applied using functions available in the NADA package in R.

The main mineral in the Corallian aquifer is calcite (Table 5, Figure 8).Calcium and HCO3 dominate the aqueous chemistry in all samples (Figure 8). There is little variation between sites, which demonstrates the homogeneity of the groundwaters from the Corallian aquifer, despite the variable proportions of calcite and quartz in different formations. When calcium carbonate minerals are present in rocks or soils at levels of 1% or more they tend to dominate the aqueous geochemistry, as demonstrated here.

Table 51 Statistical summary of field-determined parameters, major ions, and stable isotope compositions.
units

n

n(c) min mean max

P5

P25 P50 P75 P90 P95
Temp ºC 24

0

8.70 11.1 15.6 8.90 9.70 10.7 12.1 13.3 15.2
pH 25

0

6.96 7.36 7.95 7.02 7.20 7.30 7.48 7.76 7.87
Eh mV 23

0

203

379

469 243 350 395 434 444 451
DO mg L-1 24

1

<0.1 6.56 12.7 0.56 5.14 6.33 9.24 9.81 11.5
SEC µs/cm 24

0

341

622

1020 361 485 616 754 780 781
δ'2'H 23

0

-57.6 -53.8 -47.0 -57.1 -56.3 -54.0 -52.0 -51.0 -49.5
δ'18'O 23

0

-8.61 -8.11 -7.28 -8.57 -8.39 -8.24 -7.93 -7.62 -7.42
δ'13'C 23

0

-18.0 -15.0 -12.1 -16.6 -15.8 -15.2 -14.0 -13.8 -13.0
Ca mg L-1 25

0

48.8

101

161 52.7 82.3 108 124 134 137
Mg mg L-1 25

0

2.12 7.72 19.4 2.53 4.58 6.91 8.80 14.3 17.1
Na mg L-1 25

0

8.71 13.6 29.7 8.77 10.4 11.4 15.9 19.8 21.5
K mg L-1 25

0

0.660 2.18 15.5 0.720 1.31 1.60 2.01 2.73 3.07
Cl mg L-1 25

0

13.2 33.4 63.8 15.9 23.0 29.5 41.0 56.3 56.5
'SO'4 mg L-1 25

0

25.3 51.0 102 25.9 37.2 47.2 57.7 77.0 78.9
'HCO'3 mg L-1 25

0

114

220

324 136 177 221 260 291 305
NO'3'-N mg L-1 25

0

0.0860 7.75 24.4 1.50 3.85 5.70 10.6 16.0 22.6
Si mg L-1 25

0

2.91 4.10 7.50 3.43 3.58 3.95 4.26 4.65 5.29

DO: dissolved oxygen; SEC: specific electrical conductance; DOC: dissolved organic carbon;
P: percentile; n(c): number censored; min and max are observed values

Figure 8 Piper digram of Corallian groundwaters in the Vale of Pickering.

The temperatures of the groundwaters range from 8.7 to 15.6ºC, with a 5th to 95th percentile range of 8.9 to 15.2 ºC and a median of 10.7 ºC (Table 5). This is fairly typical of modern recharge in UK aquifers. Temperature can be a good indicator of depth, and therefore to some extent indicates residence time. There are no depth data to verify this, however. The temperature tends to be higher around the margins of the confined aquifer (Figure 11), which suggests that the groundwaters in these areas are deeper, and therefore older.

The pH of the groundwaters has a range of 6.96 to 7.95, a 5th to 95th percentile range of 7.02 to 7.87 and a median of 7.3 (Table 5). This range is similar to other limestone aquifers in the UK, including the Chalk, which typically has a narrow range and median values between 7.1 and 7.3 (Shand et al., 2007[1]). Such a small range is typical of unconfined groundwaters that are dominated by carbonate equilibrium reactions, again suggesting that CaCO3 is the mineral with the most influence over the aqueous geochemistry of the Corallian aquifer. The lowest pH values are found in the areas around Pickering and Malton, while the higher values are found near Scarborough and to the west of Pickering (Figure 12). There is no clear reason for this distribution; however there is little variation between the maximum and minimum values. It is very likely therefore that the entire range of pH values represents the range of baseline compositions.

The measured dissolved oxygen (DO) has a range of <1 to 12.7 mg L-1, a 5th to 95th percentile range of 0.56 to 11.1 mg L-1 and a median of 6.33 mg L-1 (Table 5). This represents a large range up to DO saturated conditions. The Eh values range from 203 to 469 mV, with a 5th to 95th percentile range of 243–451 mV, and a median of 395 mV (Table 5). This represents a relatively narrow range, dominated by oxidised water. This is typical of unconfined groundwaters. The DO content of the waters is generally lowest in the east of the area (Figure 13), and around the edge of the zone confined by Jurassic Clays. The higher values are found in the west. The spatial distribution of Eh is similar with the lowest concentrations generally found in the east and around the periphery of the confining Jurassic Clay (Figure 14). The highest values are found in the north-west portion of the Corallian aquifer. The most north-westerly samples also have the highest pH values. The interdependence of Eh and pH are the important factors which determine the solubility of minerals and speciation of chemical constituents (Levinson, 1974[2]).

Figure 9 Box plot of major ions in the Corallian aquifer.
Figure 10 Cumulative-probability plots of major ions in the Corallian aquifer.
Figure 11 Regional variation of temperature within the Corallian aquifer.
Figure 12 Regional variation of pH within the Corallian aquifer.

The specific electrical conductance (SEC) has a range of 341 to 1020 mS cm-1, a 5th to 95th percentile range of 361 to 781 mS cm-1 and a median of 616 mS cm-1 (Table 5). This range is similar to other limestone aquifers in the UK (Shand et al., 2007[1]). There is a cluster of lower SEC values in the north-west of the area, which corresponds to the highest DO and EH values. The highest SEC values are found in the south (Figure 15). There is no obvious reason why this distribution exists.

Figure 13 Regional variation of dissolved oxygen in the Corallian aquifer.
Figure 14 Regional variation of Eh in the Corallian aquifer.
Figure 15 Regional variation of specific electrical conductance (SEC) in the Corallian aquifer.

The major-ion data are presented as box plots in Figure 9 and cumulative probability distribution plots in Figure 10. Both these graphical methods are useful for summarising and characterising geochemical data distributions (Shand et al., 2007[1]).

Box plots are effective at showing data ranges within a population, and for identifying outliers. This means that they can be used to identify concentrations that could be outside a typical baseline range. Cumulative-probability plots are also an efficient way to identify distinct populations and to define anomalies. In Figure 10 the x axis (concentration) is logarithmic, meaning that a log normal distribution will plot as a straight line. A bimodal, or multimodal distribution would plot as a curve (Shand et al., 2007[1]), and may be used to differentiate baseline concentrations from anomalous populations. However, natural reactions, including redox reactions, sorption, and denitrification can introduce anomalies, which are of entirely natural origin (Shand et al., 2007[1]).

Figure 9 and Figure 10 present the ranges and outliers for the major ions. The range of concentrations for each element generally spans less than one order of magnitude. The exception to this is NO3-N, (around 2 orders of magnitude). This is a relatively narrow range of NO3-N. As NO3-N is redox sensitive it is common for concentrations to range over five orders of magnitude (Shand et al., 2007[1]). In such situations many data are left-censored as denitrification has taken place in reducing environments. The data presented here do not contain any censored data, although the lowest value is just 0.086 mg L-1. Most of the groundwaters sampled have nitrate concentrations consistent with oxidising conditions, although the lowest suggest that conditions are reducing in some. The box plot of K shows there is an upper outlier (1.2 mg L-1), which is shown by the cumulative-probability plot to be separate to the main population. This sample would warrant further investigation as it is unlikely to be representative of a baseline range. For the remaining major ions the cumulative-probability plots display steep curves, and the box plots have few outliers, indicating a near log-normal distribution. It is likely that these represent natural baseline ranges. Each element will be discussed individually, below.

Calcium has a concentration range of 48.8 to 161 mg L-1, a 5th to 95th percentile range of 52.7 to 137 mg L-1 and a median of 108 mg L-1 (Table 5). The cations in the groundwaters are dominated by Ca (Figure 5). The shape of the cumulative-probability plot is similar to those produced for groundwaters in Chalk (Shand et al., 2007[1]). By contrast the Cotswold Oolite and the Corallian of Oxfordshire and Wiltshire, which are more geologically similar to the Corallian of the Vale of Pickering, show a strong negative skew. This is due to the presence of deeper groundwaters which have exchanged Ca for Na in ion-exchange reactions. The lack of such a skew indicates that the groundwaters sampled as part of this study are all fresh shallow groundwaters. The Ca is derived from dissolution reactions with the carbonate minerals.

Most samples were at equilibrium, or supersaturated, with respect to calcite, which is why the distribution of these Ca data is similar to that of the Chalk aquifers. In addition, there are no clear trends in the spatial distribution of Ca (Figure 16).

Magnesium has a concentration range of 2.12 to 19.4 mg L-1, a 5th to 95th percentile range of 2.53 to 17.1 mg L-1 and a median of 6.91 mg L-1 (Table 5). This range is similar to that found in the Cotswold Oolite and the Corallian of Oxfordshire and Wiltshire, as well as the Chalk aquifers. The cumulative-probablity plot shows that the distribution of Mg concentrations is log normal. The lowest concentrations of Mg are all found in the north-west corner of the study area (Figure 17); the highest concentrations are in the northern portion. There is scant information regarding the mineralogy of the Vale of Pickering in the available memoirs, but it is most likely that the source of Mg is within calcite or clays. It is unlikely, however, that the source is dolomite, as all the waters analysed in this study are undersaturated with respect to it.

Sodium has a concentration range of 8.71 to 29.7 mg L-1, a 5th to 95th percentile range of 8.77 to 21.5 mg L-1 and a median of 11.4 mg L-1. This range is relatively narrow as Na in many UK groundwaters ranges over two orders of magnitude (Shand et al., 2007[1]). The data presented here lack the positive skew seen in data from comparable groundwaters of the Cotswold Oolite and Corallian of Oxfordshire and Wiltshire. While the highest concentration is found close to the coast, there is no trend of Na decreasing away from the sea, which suggests that the influence of marine aerosols or saline intrusion is limited (Figure 18). It is likely that many of the lower concentrations represent evapotranspired rainwater. The remaining Na is probably sourced from percolation through overlying soils, and minor proportions of feldspars or clays that may be present in the gritty layers of the Corallian aquifer.

Figure 16 Regional Variation of Ca in the Corallian aquifer.

Potassium has a concentration range of 0.660 to 15.5 mg L-1, a 5th to 95th percentile range of 0.720 to 3.07 mg L-1 and a median of 1.60 mg L-1. This range is comparable to those found in the groundwaters of Cotswold Oolite and Corallian of Oxfordshire and Wiltshire (Shand et al., 2007[1]). There is no obvious spatial trend (Figure 19), but there are very low K concentrations in the groundwaters in the area to the north-east of Pickering, which may represent concentrated rainfall. This is a wooded area of high ground (Figure 2), whereas the rest of the study area is predominantly arable and horticulture, and may receive K inputs from fertilisers and soil enhancement. The highest concentration is an outlier and is found at a farm site in the Howardian Hills (Figure 10). This is likely caused by agricultural contamination. The relatively high concentrations found in groundwaters around Scarborough may be derived from marine aerosols, but this is not a dominant source of K.

Figure 17 Regional variation of Mg in the Corallian aquifer.
Figure 18 Regional variation of Na in the Corallian aquifer.

Chloride has a concentration range of 13.2 to 63.8 mg L-1, a 5th to 95th percentile range of 15.9 to 56.5 mg L-1 and a median of 29.5 mg L-1. This is also a narrow range although the cumulative probability plot is very similar to that of the Corallian of Oxfordshire and Wiltshire (Shand et al., 2007[1]). The lower end of the range is consistent with rainfall impacted by evapotranspiration. Higher values could be derived from fertilizers, although the higher values are not extreme and contamination appears to be minor. There is a good correlation between Na and Cl (r2 = 0.64). There is no simple clear spatial trend (Figure 20) and no clear relationships with the proximity to the sea suggesting that marine aerosols are not a significant influence. The lowest values are found around the Howardian Hills and the North York Moors. The highest values are found in groundwaters near Scarborough, and commonly also from sites along the foot of the slopes which surround the Vale of Pickering.

Sulphate has a concentration range of 25.3 to 102 mg L-1, a 5th to 95th percentile range of 25.9 to 78.9 mg L-1 and a median of 47.2 mg L-1. The cumulative probability plot is similar to that of the groundwaters of the Corallian of Oxfordshire and Wiltshire (Shand et al., 2007[1]). The highest concentrations of SO4 are generally found near Scarborough, in the Howardian Hills, and around the foot slopes of the North York Moors (Figure 21). Lower concentrations of SO4 are found in the higher ground of the North York Moors, and the west of the region. The groundwaters mostly saturated or supersaturated with respect to barite.

Alkalinity as HCO3 has a concentration range of 114 to 324 mg L-1, a 5th to 95th percentile range of 136 to 305 mg L-1 and a median of 221 mg L-1. This is a very narrow range and the cumulative probability plot is very similar to those of the Cotswold Oolite and the Corallian of Oxfordshire and Wiltshire (Shand et al., 2007[1]). The aquifer mineralogy is dominated by CaCO3; hence the anion chemistry is dominated by HCO3, as shown on the Piper Plot (Figure 8). There is a strong correlation between HCO3 and Ca (r2 = 0.74). While there are no clear spatial trends of HCO3 (Figure 22) the distribution is similar to that of Ca (Figure 16).

Figure 19 Regional variation of K in the Corallian aquifer.
Figure 20 Regional variation of Cl in the Corallian aquifer.
Figure 21 Regional variation of SO4 in the Corallian aquifer.

Nitrate as N has a concentration range of 0.086 to 24.4 mg L-1, a 5th to 95th percentile range of 1.5 to 22.6 mg L-1 and a median of 5.7 mg L-1. The cumulative probability curve is relatively shallow as a result of a large concentration range, probably caused by varying human inputs. The drinking water limit of NO3-N is 11.3 mg L-1, and one fifth of the analyses exceed this value, suggesting significant human inputs. There are no distinct spatial trends (Figure 23), but all the concentrations > 7mg L-1 are found in groundwaters sampled within 500 m of farm land. The concentration of NO3-N in groundwaters is dependent on the redox conditions as well as availability of NO3 in the environment. Figure 24 demonstrates how larger NO3-N concentrations occur in oxic conditions (higher Eh values).

Figure 22 Regional variation of HCO3 in the Corallian aquifer.
Figure 23 Regional variation of NO3-N in the Corallian aquifer.

Silicon has a concentration range of 2.91 to 7.5 mg L-1, a 5th to 95th percentile range of 3.43 to 5.29 mg L-1 and a median of 3.95 mg L-1. The near-vertical line presented on the cumulative probability curve indicates that Si rapidly attains saturation. Despite the abundance of Si in the Earth’s crust (it is the second most abundant element) the concentrations of Si are relatively low, owing to low solubility of silicate minerals (Hem, 1992[3]). The most soluble of the SiO2 polymorphs is amorphous SiO2 and the least soluble is quartz. From modelled saturation indices, quartz is generally close to equilibrium, with one site being supersaturated. In contrast, all sites were undersaturated with respect to amorphous SiO2. There is no particular spatial trend (Figure 25), which probably indicates there are variable concentrations of Si present in the aquifer.

Figure 24 Relationship of NO3-N and Eh.
Figure 25 Regional variation of Si in the Corallian aquifer.

Minor and trace constituents

Summary statistical data (Table 6, Figure 26 and Figure 27) and spatial variations (Figure 28 to Figure 50) for minor and trace constituents are presented in this section. Where all the determinations are below the detection limit, the element is not discussed here (see Table 6).

Phosphorus

Phosphorus has a concentration range of <0.02 to 0.043 mg L-1, and 17 of the 24 analyses are below the detection limit (Table 6). The concentrations of P are generally lower than those found in the groundwaters of the Cotswold Oolite and the Corallian of Oxfordshire and Wiltshire, which ranged from <0.02 to 0.12 mg L-1 (Cobbing et al., 2004[4]; Neumann et al., 2003[5]). There are no spatial trends evident in the Vale of Pickering (Figure 28), which is probably a function of the large number of censored data.

Halogen elements

Fluoride has a concentration range of <0.25 to 0.311 mg L-1, a 5th to 95th percentile range of 0.024 to 0.252 mg L-1, and a median of 0.06 mg L-1 (Table 6). All these values are well within the drinking water limit of 1.5 mg L-1 (The Water Supply Regulations, 2010[6]). One of the samples contained F below the detection limit of 0.25 mg L-1. However this detection limit is higher than some of the measured values, of which the minimum is 0.033 mg L-1. The F range in the Corallian of the Vale of Pickering is much smaller than those of the comparable groundwaters found in the Cotswold Oolite and the Corallian of Oxfordshire and Wiltshire, which are between <0.05 to 4.8 mg L-1, and 0.05 to 1.98 mg L-1 respectively (Cobbing et al., 2004[4]; Neumann et al., 2003[5]). In these the higher concentrations were the result of ion exchange.

Figure 26 Box plot for the minor and trace elements in the Vale of Pickering Corallian.
Table 6 Statistical summary of data for minor and trace elements.
(P= percentile; n(c) = number censored; min and max are observed values)
units

n

n(c) min mean max

P5

P25 P50 P75 P90 P95
Ag mg L-1 25

25

<0.05 <0.05

Al

mg L-1 25

1

<1

9.52

89

1

1

3

4

29

42

As mg L-1 24

20

<0.5

0.7

Au mg L-1 24

24

<0.05 <0.05

B

mg L-1 25

1

<100

16

42

7

10

14

19

26

31

Ba mg L-1 25

0

28.6 60.4 102 35.5

45

59.7 68.9 89.4 89.8
Be mg L-1 25

25

<0.05 <0.05

Bi

mg L-1 24

24

<0.05 <0.05

Br

mg L-1 25

0

0.042 0.0998 0.179 0.042 0.081 0.096 0.117 0.15 0.17
Cd mg L-1 25

25

<0.05 <0.05
Ce mg L-1 24

14

<0.01 0.11
Co mg L-1 25

19

<0.02

0.1

Cr

mg L-1 25

25

<0.5 <0.5
Cs mg L-1 24

21

<0.01 0.02
Cu mg L-1 24

0

0.8

2.39

9

0.8

1

1.9

2.8

3.6

5

Dy mg L-1 24

17

<0.01 0.03

Er

mg L-1 24

22

<0.01 0.02
Eu mg L-1 24

23

<0.01 0.01

F

mg L-1 25

1

<0.25 0.0756 0.311 0.024 0.04 0.06 0.073 0.103 0.252
Fe mg L-1 24

14

<5

842
Ga mg L-1 24

24

<0.05 <0.05
Gd mg L-1 24

18

<0.01 0.05
Ge mg L-1 24

23

<0.05 0.07
Hf mg L-1 24

24

<0.02 <0.02
Hg mg L-1 24

20

<0.1

0.1

Ho mg L-1 24

23

<0.01 0.01

In

mg L-1 24

23

<0.01 0.01

Ir

mg L-1 24

24

<0.05 <0.05
La mg L-1 24

2

<0.01 0.0146 0.07 0.01 0.01 0.01 0.02 0.02

Li

mg L-1 24

0

1

3.75 12.5

1.3

1.7

2.9

4.85

6.2

8.1

Lu mg L-1 24

24

<0.01 <0.01
Mn mg L-1 25

4

<0.05 5.71 109 0.18 0.38 2.18 4.13 12.5
Mo mg L-1 24

0

0.1

0.221

0.7

0.1

0.1

0.2

0.2

0.4

0.5

Nb mg L-1 24

23

<0.01 0.01
Nd mg L-1 24

2

<0.01 0.02 0.14 0.01 0.01 0.01 0.05 0.05

Ni

mg L-1 25

15

<0.2

1.2

Os mg L-1 24

24

<0.05 <0.05

P

mg L-1 24

17

<0.02 0.043
Pb mg L-1 25

4

<0.1 0.206

1.1

0.1

0.1

0.2

0.4

0.4

Pd mg L-1 24

24

<0.2 <0.2

Pr

mg L-1 24

20

<0.01 0.02

Pt

mg L-1 24

24

<0.01 <0.01
Rb mg L-1 24

0

0.12 0.779 3.22

0.3

0.42 0.59 1.01 1.16 1.38
Re mg L-1 24

5

<0.01 0.016 0.07 0.01 0.01 0.01 0.03 0.035
Rh mg L-1 24

7

<0.01 0.0608 0.16 0.05

0.1

0.15 0.15
Ru mg L-1 24

20

<0.05 0.08
Sb mg L-1 25

22

<0.05 0.14
Sc mg L-1 24

0

1

1.04

2

1

1

1

1

1

1

Se mg L-1 24

3

<0.5 0.754

1.2

0.6

0.8

0.9

1.1

1.1

Sm mg L-1 24

23

<0.02 0.03
Sn mg L-1 24

18

<0.05 0.12

Sr

mg L-1 24

0

78.9

191

447

106

125

155

216

377

403

Ta mg L-1 24

24

<0.02 <0.02
Tb mg L-1 24

23

<0.01 0.01
Te mg L-1 24

24

<0.05 <0.05
Th mg L-1 24

24

<0.05 <0.05
units

n

n(c) min mean max

P5

P25 P50 P75 P90 P95

Ti

mg L-1 24

24

<10 <10

Tl

mg L-1 24

10

<0.01 0.0112 0.03 0.01 0.01 0.01 0.02
Tm mg L-1 24

24

<0.01 <0.01

U

mg L-1 24

0

0.06 0.225 0.51 0.08 0.12 0.21 0.25

0.4

0.41

V

mg L-1 25

5

<0.2 0.283

0.5

0.2

0.3

0.3

0.4

0.5

W

mg L-1 24

24

<0.02 <0.02

Y

mg L-1 24

0

0.01 0.0367 0.19 0.01 0.02 0.03 0.04 0.06 0.07
Yb mg L-1 24

22

<0.01 0.01
Zn mg L-1 25

1

<5

10.2 63.2

0.7

1.8

4.8

14

32.4 33.8

Zr

mg L-1 24

23

<0.02 0.06
Figure 27 Cumulative probability plots for selected trace elements in the groundwater from the Vale of Pickering Corallian aquifer.

In general, the highest F concentrations are found in the areas around the periphery of the valley floor (Figure 29) and the lowest concentrations are found in groundwaters sampled higher up the valley sides, towards the north and west of the region. Fluoride concentrations in rainfall are generally low, so the dominant source of F tends to be mineralogical. Unfortunately the lack of mineralogical information makes it difficult to draw firm conclusions, although CaF2, the most common F-rich mineral, is the most likely source.

Figure 28 Regional variation of P in the Corallian aquifer.
Figure 29 Regional variation of F in the Corallian aquifer.

Bromide has a concentration range of 0.042 to 0.179 mg L-1, a 5th to 95th percentile range of 0.042 to 0.17 mg L-1, and a median of 0.096 mg L-1 (Table 6). This is a smaller range than that of either the Cotswold Oolite and the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]; Neumann et al., 2003[5]), for which Br concentrations are up to an order of magnitude greater. Bromide behaves in a similar way to Cl, and the two elements have a moderate-strong correlation (r2 = 0.59) and a similar spatial distribution (Figure 30). High concentrations of Cl and Br are found around Scarborough, and to the north-west of Malton, while the lowest concentrations are found north of the Howardian Hills.

Edmunds et al. (1989)[7] noted that because of the similar geochemical behaviour of Cl and Br the ratio Cl/Br is usually very similar to that of sea water (Cl/Br = 288). They suggested that deviation from this ratio would have diagnostic value in interpreting the origin of Br. The Cl/Br ratios of 25 samples in this study range from 207 to 604. Most calculated ratios (n=18) are greater than the seawater value of 288, which indicates Br depletion or Cl enrichment. If Cl/Br values far exceed 288, it is likely that there has been a dilution effect from a halite source, either a natural geological source or from agriculture or industry (Edmunds et al., 1989[7]). The most enriched sites are generally surrounded by farmland or are industrial, suggesting that there is some NaCl contamination. Where the Cl/Br values are moderately higher than that of seawater, the deviation from a seawater average is most likely caused naturally on infiltration (Edmunds, 1996[8]).

Alkaline earth elements

Barium has a concentration range of 28.6 to 102 mg L-1, a 5th to 95th percentile range of 35.5 to 89.8 mg L-1, and a median of 59.7 mg L-1 (Table 6). These concentrations are much higher than those found in the geologically similar aquifers of the Cotswold Oolite and the Corallian of Oxfordshire and Wiltshire, which have ranges of 3.46 to 14.0 mg L-1, and 3.23 to 50.19 mg L-1, respectively. There are however few diffuse anthropogenic sources of Ba, and the higher concentrations found here likely reflect natural regional variations in Ba.. The groundwaters in the study area are mostly saturated or supersaturated with respect to barite, which is the limiting control on Ba. Where the SO4 concentration is low (<40 mg L-1), Ba ranges between 38 and 102 mg L-1. There is no distinct spatial distribution of Ba (Figure 31), which likely represents natural variability as well as barite solubility.

Figure 30 Regional variation of Br in the Corallian aquifer.

Strontium has a concentration range of 78.9 to 447 mg L-1, a 5th to 95th percentile range of 106 to 403 mg L-1, and a median of 155 mg L-1 (Table 6). This range is about half that of the geologically similar Corallian of Oxfordshire and Wiltshire, which ranges from 88.6 to 1860 mg L-1 (Cobbing et al., 2004[4]). Strontium is geochemically similar to Ca, and can be present at high concentrations in Ca-bearing minerals owing to the similarities in the ionic radius of Ca and Sr. In fresh groundwater there are generally no solubility controls on Sr, and it typically increases with residence time. Increases in the Sr/Ca ratio likewise reflect increasing residence time (Shand et al., 2007[1]). The Sr/Ca ratio of the data presented here has a moderate-strong relationship with the Sr concentrations (r2 = 0.64). The spatial distribution of Sr in the Vale of Pickering is similar to that of Ca (Figure 32), and there is a moderate relationship between the two elements (r2 = 0.38). These values indicate that, while there is no information available on the aquifer mineralogy and chemistry, is it most likely that the Sr concentrations presented here, represent baseline concentrations.

Alkali metals

The cumulative-probability distributions (Figure 27) for the alkali metals show a small range of concentrations. The distributions of Li and Rb are very similar (Figure 33 and Figure 34).

Lithium has a concentration range of 1 to 12.5 µg L-1, a 5th to 95th percentile range of 1.3 to 8.1 µg L-1, and a median of 2.9 µg L-1 (Table 6). This range is narrower than either of those observed in the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]) and the Cotswold Oolite (Neumann et al., 2003[5]), which range up to 78.3 and 48 mg L-1 respectively. The concentration of Li in groundwaters is a combination of availability and residence time, so it would be expected that the baseline vary greatly between groundwater systems. Lithium-rich minerals such as clays are likely to be rare in the Corallian limestone aquifer, as reflected by the low concentrations. The highest concentrations are found in the south of the area, and near to the boundary of the Corallian outcrop and the valley bottom clays (Figure 33). This may indicate interaction with the clays, but could also be related to residence times, as the lower concentrations are generally found in groundwaters sampled from higher up on valley slopes.

Figure 31 Regional variation of Ba in the Corallian aquifer.
Figure 32 Regional variation of Sr in the Corallian aquifer.

Rubidium has a concentration range of 0.12 to 3.22 µg L-1, a 5th to 95th percentile range of 0.3 to 1.38 µg L-1, and a median of 0.59 µg L-1 (Table 6). This range is much smaller than that found in the Corallian of Oxfordshire and Wiltshire, which has a minimum of 0.22 and a maximum of 23.4 mg L-1. Rubidium displays similar geochemical behaviour to K, and the two elements are strongly correlated (r2 = 0.79), the processes controlling K likely also affect Rb. The spatial variation of Rb is similar to that of K, with the highest concentrations found in the Howardian Hills, and lower concentrations to the west of Scarborough (Figure 34). The observed maxima for both Rb and K occur in the same sample. The sample was taken from a farm, and may therefore represent contamination derived from fertilizer or other soil enhancement materials. It is likely that concentrations up to the 95th percentile represent baseline.

Caesium has a concentration range of <0.01 to 0.02 µg L-1, of 24 analyses 21 are censored reflecting its scarcity in natural groundwaters (Table 6). These values most likely represent baseline concentrations.

Iron and manganese

Iron and Mn are both redox- and pH-sensitive. Iron is mobilised as dissolved Fe(II) under moderately reducing and acidic conditions. Under oxidising conditions, Fe(III) prevails. This forms insoluble Fe oxyhydroxides at circum-neutral pH (Drever, 1997[9]). Fe and Mn display a similar cumulative probability curve, which is relatively shallow, but skewed at the upper end (Figure 27).

Iron has a concentration range of <5 to 842 µg L-1, (Table 6) of 24 analyses, 14 are censored. This is a similar range to the Cotswold Oolite (Neumann et al., 2003[5]), but the Corallian of Oxfordshire and Wiltshire ranges from <5 to 1770 mg L-1 (Cobbing et al., 2004[4]). The highest Fe concentration (842 mg L-1) is found in the groundwater with the lowest Eh value (203 mV). This is the only sample which exceeds the drinking water standard for Fe of 200 mg L-1 (The Water Supply Regulations, 2010[10]), and would require treatment if used for potable supply. This sample probably represents contamination as the Fe concentration is five times higher than the next most concentrated sample. The Eh is weakly correlated with Fe (r2 = 0.26), and indicates that the distribution of dissolved Fe is controlled by the oxidation state. The spatial variation is indistinct (Figure 35), although there are some spatial relationships with the Eh. The most oxidised waters are found to the north-west of the region, where the Fe concentrations are below detection limits.

Figure 33 Regional Variation of Li in the Corallian aquifer.
Figure 34 Regional Variation of Rb in the Corallian aquifer.

Manganese has a concentration range of <0.05 to 109 µg L-1, a 95th percentile of 12.5 µg L-1, and a median of 0.38 µg L-1 (Table 6). Of 25 analyses, four are censored. This is a larger range of concentrations than is found in the Cotswold Oolite (Neumann et al., 2003[5]), where the Mn ranges from <2 to 18 mg L-1, but a much smaller range than that of the groundwaters of the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]), where the Mn ranges from <2 to 466 mg L-1. Most of the concentrations of Mn are very low (out of 25 analyses 23 are <4.2 mg L-1), while the sample containing the highest Mn concentration also has the highest Fe concentration. This sample could represent contamination, and the 95th percentile is a good approximation for a baseline range. This sample exceeds the drinking water limit of 50 mg L-1 for Mn and would require treatment for drinking (The Water Supply Regulations, 2010[10]). The spatial trends are indistinct (Figure 36), and are dissimilar to both Fe (Figure 35) and Eh (Figure 14).

Rare earth elements

The rare earth elements (REE) are here, as commonly, considered as a group together with yttrium due to their similar behaviour in the environment (Cornell, 1993[11]). The REEs are most commonly below detection limits in groundwaters in the UK (Shand et al., 2007[1]). Yttrium, La and Nd are found in this study to be detectable in most groundwaters analysed. The remaining REEs are detectable in up to 40% of the samples collected. Where detectable the concentrations are generally low; the maximum REE concentration is 0.14 mg L-1 (Nd) and Y has a concentration of 0.19 mg L-1 (Table 6). As the REEs behave geochemically so similarly the spatial variation is very similar for each element and the highest concentration of each element is found at the same site. Figure 37 presents an example of the spatial distribution of the REEs. This presents Ce, which is below detection limits in 14 of the 24 samples, and shows a typical distribution. There are no clear spatial trends and it is likely that these data represent baseline concentrations.

Figure 35 Regional variation of Fe in the Corallian aquifer.
Figure 36 Regional variation of Mn in the Corallian aquifer.
Figure 37 Regional variation of Ce in the Corallian aquifer.

Other trace elements

Aluminium has a concentration range of <1 to 89 µg L-1, a 5th to 95th percentile range of 7 to 31 µg L-1, and a median of 14 µg L-1 (Table 6). This range is very similar to that of the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]). While Al is one of the most abundant elements on Earth, with an average crustal and soil abundance of 6 to 7 wt%, it is classed as a trace element in natural waters because the solubility of Al minerals is limited at circum-neutral pH values (pH 6–8). All the groundwaters in the Corallian of the Vale of Pickering fell into this pH range; hence Al concentrations are low. There is no distinct spatial distribution of Al (Figure 38), and as there is no evidence that the Al concentrations in the groundwaters are of anthropogenic origin, the 5th to 95th percentile range is taken to represent baseline compositions.

Figure 38 Regional Variation of Al in the Corallian aquifer.

Arsenic has a concentration range of <0.5 to 0.7 µg L-1, of the 24 analyses 20 were below detection limits (Table 6). The measurable concentrations were found close to Scarborough, and in the Howardian Hills. This concentration range is slightly smaller than that found in the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]) and Cotswold Oolite (Neumann et al., 2003[5]), and it is likely it represents baseline concentrations.

Boron has a concentration range of <100 to 42 µg L-1, a 5th to 95th percentile range of 7 to 31 µg L-1, and a median of 14 µg L-1 (Table 6). These are very low concentrations when compared to those found in aquifers hosted within geologically similar rocks (Cobbing et al., 2004[4]; Neumann et al., 2003[5]). One sample has a high detection limit, which is greater than the maximum measured concentration. It is likely that this value falls within the range of the rest of the samples. There are no clear spatial patterns, however the lowest concentrations are generally found higher up the valley sides (Figure 39).

Cobalt has a concentration range of <0.02 to 0.1 µg L-1. Of the 25 analyses 19 were below detection limits (Table 6). This range is similar to that found in the Cotswold Oolite (Neumann et al., 2003[5]), while the concentrations found in the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]) is an order of magnitude larger, ranging from <0.02 to 1.96 mg L-1. Cobalt mobility is strongly limited at neutral pH values, and the concentrations represented here most likely represent baseline concentrations. Most of the detectable Co is found in groundwaters sampled from the west of the study area, with one sample containing detectable Co near Scarborough (Figure 40)

Figure 39 Regional Variation of B in the Corallian aquifer.
Figure 40 Regional Variation of Co in the Corallian aquifer.

Copper has a concentration range of 0.8 to 9 µg L-1, a 5th to 95th percentile range of 0.8 to 5 µg L-1, and a median of 1.9 µg L-1 (Table 6). This range is smaller than those observed in the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]) and Cotswold Oolite (Neumann et al., 2003[5]), which range from 0.5 to 304 mg L-1 and 0.1 to 115 mg L-1 respectively. Copper occurs naturally as native metal, or in sulphide ore deposits, but is also a very commonly used metal in industrialised countries, and in pipework in water supply infrastructure. Anthropogenic inputs are to be expected in young groundwaters (Shand et al., 2007[1]). Where the groundwater is oxidising, Cu is most soluble under acidic conditions. As the pH increases, Cu can adsorb to organic matter or Fe and Mn oxyhydroxides (Shand et al., 2007[1]). There is a poor correlation between pH and Cu (r2=0.02), and there is no clear spatial distribution (Figure 41). The distribution of Cu could be influenced by anthropogenic activities.

Figure 41 Regional Variation of Cu in the Corallian aquifer.

Molybdenum has a concentration range of 0.1 to 0.7 µg L-1, a 5th to 95th percentile range of 0.1 to 0.5 µg L-1, and a median of 0.2 µg L-1 (Table 6). In comparison the Mo is at or below detection limits (<0.1) in the majority of grounwaters from the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]) and Cotswold Oolite (Neumann et al., 2003[5]). The concentration range is much narrower than that observed in the groundwaters in these similar aquifers, which have maximum Mo concentrations of 5.2 and 1.8 µg L-1, respectively. It is likely that the Mo concentrations measured in the Vale of Pickering Corallian represent baseline concentrations. The highest Mo concentrations are found in groundwaters in the south of the area, around the Howardian Hills (Figure 42). Lower concentrations are observed in the groundwaters from the hillslopes in the north of the region.

Nickel has a concentration range of <0.2 to 1.2 µg L-1, of the 25 samples 15 were below the detection limit (Table 6). There is no clear spatial trend of Ni in the Vale of Pickering, as groundwaters with measurable Ni are in close proximity to those where Ni is not detectable. It is therefore possible that the highest concentrations are associated with anthropogenic influences. Nickel behaves in a similar way to Co, and Ni is commonly detectable in groundwaters where Co is also present, hence the spatial variation (Figure 5.36) is similar to that of Co (Figure 40).

Lead has a concentration range of <0.1 to 1.1 µg L-1, a 5th to 95th percentile range of <0.1 to 0.4 µg L-1, and a median of 0.1 µg L-1 (Table 6). Four of the 25 analyses were below the detection limit. This range is slightly smaller than other geologically similar aquifers where the maximum Pb values are 5.4 µg L-1 (Cotswold Oolite (Neumann et al., 2003[5])) and 3.9 µg L-1 (Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4])). Lead can be found in the natural environment in sulphide minerals and metal oxides, and as a trace constituent in K feldspars (Shand et al., 2007[1]). However the baseline of Pb is likely to be anthropogenically influenced in built up areas. The relatively low Pb concentrations may reflect the mainly rural surroundings and the range presented here most likely represents baseline concentrations. In general the highest concentrations are found in groundwaters in the east of the area (Figure 44), with two outliers evident towards the south of the outcrop in the Howardian Hills.

Figure 42 Regional variation of Mo in the Corallian aquifer.
Figure 43 Regional variation of Ni in the Corallian aquifer.

Rhenium has a concentration range of <0.01 to 0.07 µg L-1, a 5th to 95th percentile range of <0.01 to 0.035 µg L-1, and a median of 0.01 µg L-1 (Table 6). Five of the 24 analyses were below the detection limit. These values are slightly high when compared to other similar UK aquifers: there was a maximum concentration of 0.02 µg L-1 in groundwaters in the Corallian aquifer of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]). In general the highest concentrations in the Vale of Pickering are found in groundwaters around the periphery of the flat valley floor, with lower concentrations found in groundwater samples taken higher up the valley slopes, especially in the north and west of the region (Figure 45). The generally low Re concentrations reflect the limited use and rarity of this element.

Figure 44 Regional variation of Pb in the Corallian aquifer.
Figure 45 Regional variation of Re in the Corallian aquifer.

Rhodium has a concentration range of <0.01 to 0.16 µg L-1, a 5th to 95th percentile range of <0.01 to 0.015 µg L-1, and a median of 0.05 µg L-1 (Table 6). Seven of the 24 analyses were below the detection limit. These values are higher than those recorded in other similar UK aquifers. Rhodium in the groundwaters of Cotswold Oolite (Neumann et al., 2003[5]) and the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]) did not exceed 0.02 µg L-1. The highest Rh concentrations in the Vale of Pickering are found in groundwaters in the north-west of the region, while the lowest concentrations are found in the south (Figure 46).

Ruthenium has a concentration range of <0.05 to 0.08 µg L-1, of the 24 analyses 20 were below detection limits (Table 6). The measurable concentrations were found in close proximity to each other in the north-west of the region (Figure 47). There was no measurable Ru (<0.05 µg L-1) in the groundwaters of the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]) and the Cotswold Oolite (Neumann et al., 2003[5]). It is therefore possible that the measured concentrations found in the Vale of Pickering represent values outside baseline groundwater concentrations.

Antimony has a concentration range of <0.05 to 0.14 µg L-1, of the 24 analyses 22 were below detection limits (Table 6).

Scandium has a concentration range of 1 to 2 µg L-1: one value was 2 µg L-1, while the remaining 23 analyses were 1 µg L-1 (Table 6). There is insufficient variation within these data to comment on any spatial trends, although for comparison Sc concentrations in the Cotswold Oolite (Neumann et al., 2003[5]) groundwaters range up to 5.94 µg L-1.

Selenium has a concentration range of <0.5 to 1.2 µg L-1, a 5th to 95th percentile range of <0.5 to 1.1 µg L-1, and a median of 0.8 µg L-1 (Table 6). The range of values is typical for UK aquifers (Shand et al., 2007[1]), Se concentrations tend to be low in natural waters, rarely exceeding 1 µg L-1(Hem, 1992[3]). Selenium is a relatively rare element, mobilised in oxidising waters, but immobile under reducing conditions (Hem, 1992[3]). The spatial variation is indistinct (Figure 48), although there are some spatial relationships with the Fe concentrations. The highest concentrations of Se are found in groundwaters where the Fe is low (Figure 35), indicating the redox control on this element. The dominance of agriculture in this area may provide a source of Se to groundwaters.

Figure 46 Regional variation of Rh in the Corallian aquifer.

Tin has a concentration range of <0.05 to 0.12 µg L-1, of the 24 analyses 18 are below the detection limit (Table 6). The use of Sn as a coating for corrosion prevention and on steel containers for food storage means that Sn can be widespread in the urban environment, especially around landfill sites. However it is generally low in groundwaters (Shand et al., 2007[1]).

Figure 47 Regional variation of Ru in the Corallian aquifer.
Figure 48 Regional variation of Se in the Corallian aquifer.

There is no clear spatial distribution, and the low concentrations are typical of the largely arable environment.

Thallium has a concentration range of <0.01 to 0.03 µg L-1, a 5th to 95th percentile range of <0.001 to 0.02 µg L-1, and a median of 0.01 µg L-1 (Table 6). Of the 24 analyses, 10 were below the detection limit. There are no distinct spatial trends, in part owing to the small range of values. This distribution is similar to that found in the groundwaters from the similar aquifers of Cotswold Oolite (Neumann et al., 2003[5]) and the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]).

Uranium has a concentration range of 0.06 to 0.51 µg L-1, a 5th to 95th percentile range of 0.08 to 0.41 µg L-1, and a median of 0.21 µg L-1 (Table 6). This range is relatively small, as groundwaters from the similar aquifers of Cotswold Oolite and the Corallian of Oxfordshire and Wiltshire have maximum concentrations that are over three times the maximum value reported here. There are no clear spatial distributions, although there is a cluster of lower concentrations around the north-west of the region. The higher concentrations tend to be close to the edge of the flat valley floor with the maximum concentrations in the south of the region (Figure 49). Uranium mobilisation is strongly controlled by its redox state. It is usually only the oxidised form (U(VI)) that occurs significantly in solution (Smedley et al., 2006[12]). Where there are reducing conditions, the concentration of U is generally low. There is some spatial correlation of U to Eh, as the lowest U concentrations occur where the Eh is the highest (Figure 14). Uranium occurs naturally in soils and rock-forming minerals, and it is likely that the source of U in these waters is predominantly natural, although fertilizers are a known possible additional source. All of the determinations are well within the WHO provisional drinking water guideline value of 30 µg L-1.

Zinc has a concentration range of <5 to 63.2 µg L-1, a 5th to 95th percentile range of 0.7 to 33.8 µg L-1, and a median of 4.8 µg L-1 (Table 6). There is one value below the detection limit, however, this detection limit is higher than most measured values within this data set (Figure 50). This is a relatively small range when compared to the groundwaters from the similar aquifers of the Cotswold Oolite (Neumann et al., 2003[5]) and the Corallian of Oxfordshire and Wiltshire (Cobbing et al., 2004[4]) which have maximum concentrations of 133 and 289 µg L-1 respectively. The solubilities of some Zn minerals are high, meaning that Zn can be widespread in groundwaters, and present in higher concentrations than many other transition metals. In addition, Zn in groundwaters can be derived from anthropogenic sources such as road dust, landfill leachate, and urban industry. There is no clear spatial trend (Figure 50), and this coupled with the low concentration range suggests that the values presented here represent baseline concentrations.

Figure 49 Regional variation of U in the Corallian aquifer.
Figure 50 Regional variation of Zn in the Corallian aquifer.

Isotopic compositions and tracers

Stable oxygen and hydrogen isotope ratios are extremely useful as tracers of physical processes in groundwater and can help to define recharge condition (Darling and Talbot, 2003[13]). The rainfall in north-east England is depleted in the heavier isotopes of O and H relative to western parts of the country in response to the Rayleigh fractionation effect: the rain which falls first, over the west of the country, is relatively enriched isotopically (Darling and Talbot, 2003[13]). Measured compositions of δ18O lie in the range -8.61 to -7.28‰ with an average of -8.11 ‰ (Table 6). The δ2H has a range of -57.6 to -47‰, with an average of -53.8‰. These values are consistent with compositions of recently recharged (Holocene) groundwaters analysed from north-east England (Darling et al., 2003[13]). The relationship between δ18O and δ2H (Figure 51) shows a general correspondence with the world meteoric water line, indicating that the groundwaters represent recharged regional modern rainfall. There are no clear spatial trends (Figure 52 and Figure 53), which would be expected if the waters were being affected by local factors such as evaporation or mixing.

Values of δ13C are an index of the evolution of the dissolved inorganic carbon system (DIC) (Clark and Fritz, 1997[14]; Darling et al., 2005[13]). Groundwater in calcareous sedimentary terrain acquires DIC through reaction with soil CO2 and reaction with carbonate minerals in the soil and aquifer. The δ13C composition of DIC is fundamentally governed by interaction between soil CO213C ~ -26‰) and carbonate minerals (δ13C ~ 0‰), resulting in a δ13C-DIC value of around ~ -13‰, but further modified depending on the nature of the carbonate system. If open (typical of unconfined aquifer conditions), further exchange with soil-derived CO2 will result in δ13C-DIC values <-13‰. If closed (typical of confined conditions), δ13C-DIC will start to evolve towards the rock composition, resulting in δ13C-DIC values rising progressively above -13‰. The compositions in the Corallian groundwaters range between -18.0 and -12.0‰ with an average of -15‰. Most of the sites are therefore relatively depleted in δ13C, which suggests a relatively immature DIC system of generally limited residence time. This conclusion supports the δ18O and δ2H results insofar as they suggest no significant modification (i.e. mixing with old groundwater) since recharge. However the most δ13C enriched sites tend to be close to the valley floor and probably signify slightly older waters. (Figure 54)

Figure 51 O and H stable isotopic composition of the Corallian aquifer. WMWL = world meteroric water line (Craig, 1961[15]).

Chemical variations with depth

At the time of writing, no data could be found on groundwater chemical variations with depth.

Temporal variations

Data are available from the EA database for eleven sites for a period of five to 10 years. These span from 1995 to 2006 to coincide with the new BGS samples presented here. Generally sites have been sampled twice a year, although in most cases there are only sufficient data for the major ions, temperature and pH. There are generally variations in temperature and pH, but these follow no distinct trend. This is unsurprising given the dynamic nature of these parameters.

In many cases the major ions follow similar trends over the period they were measured. For example at one site about 2km south-east of Helmsley [SE 63 82], Ca, Cl, and Mg all increase gradually, but steadily, over a period of 10 years (Figure 55). At the same site this is mirrored by a general decrease in the HCO3 concentration (Figure 56) over the same time period. There is no obvious reason for these changes.

Figure 52 Regional variation of δ18O in the Corallian aquifer.
Figure 53 Regional variation of δ2H in the Corallian aquifer.

At the same site near Helmsley there is an increase in NO3-N over a period of six years. This is not an unusual trend in areas dominated by agriculture, as NO3-N is a common diffuse pollutant derived from agricultural practices. At this site the concentration increases from values around the drinking water limit (11.3 mg L-1 NO3-N)at an average of 1.7 mg L-1 each year, to twice the drinking water limit over a period of seven years (Figure 57).

Figure 54 Regional variation of δ13C in the Corallian aquifer.
Figure 55 Temporal changes in Ca, Mg, and Cl at a site near Helmsley.

Another four of the EA’s monitoring sites show similar NO3-N trends, and are distributed throughout the Vale of Pickering. This finding is consistent with previous reports which document the high concentrations of NO3-N and PO43- in the region (EA, 2009[16]; Natural England, 2015[17]).

A more unusual temporal variation was observed in groundwaters at a site near Kirbymoorside [SE 70 87]. Concentrations of the major elements were relatively constant between 1995 and 1998, most elements increased until 1999, where they stayed relatively constant, with the exception of three lower values recorded on 9/11/99, 17/10/03 and 20/10/05. The exceptions to this are Na and K, which mirror this trend (Figure 58 and Figure 59). There is no clear reason for this trend.

Figure 56 Temporal changes in HCO3 at a site near Helmsley.
Figure 57 Temporal changes in NO3–N at a site near Helmsley.
Figure 58 Temporal changes in Ca, Cl, SO4, HCO3 at a site near Kirkbymoorside.
Figure 59 Temporal changes in Mg, Na, K and NO3–N at a site near Kirkbymoorside.

References

  1. 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 1.10 1.11 1.12 1.13 1.14 1.15 1.16 1.17 1.18 SHAND, P, EDMUNDS, W M, LAWRENCE, A R, SMEDLEY, P L, and BURKE, S. 2007. The natural (baseline) quality of groundwater in England and Wales. British Geological Survey & Environment Agency, RR/07/06 & NC/99/74/24 (Keyworth and Solihull). Cite error: Invalid <ref> tag; name "Shand 2007" defined multiple times with different content Cite error: Invalid <ref> tag; name "Shand 2007" defined multiple times with different content
  2. LEVINSON, A A. 1974. Introduction to Exploration Geochemistry. (Calgary: Applied Publishing Ltd.)
  3. 3.0 3.1 3.2 HEM, J D. 1992. Study and Interpretation of the Chemical Characteristics of Natural Water (Third Edition edition). US Geological Survey Water-Supply Paper 2254. (Washington: United StatesGovernment Printing Office.)
  4. 4.00 4.01 4.02 4.03 4.04 4.05 4.06 4.07 4.08 4.09 4.10 4.11 4.12 4.13 4.14 4.15 4.16 4.17 4.18 COBBING, J, MOREAU, M, SHAND, P, and LANCASTER, A. 2004. The Corallian of Oxfordshire and Wiltshire. British Geological Survey and Environment Agency, BGS Report CR/04/262N; Environment Agency Report NC/99/74/14 (Keyworth and Solihull). Cite error: Invalid <ref> tag; name "Cobbing 2004" defined multiple times with different content
  5. 5.00 5.01 5.02 5.03 5.04 5.05 5.06 5.07 5.08 5.09 5.10 5.11 5.12 5.13 5.14 5.15 5.16 NEUMANN, I, BROWN, S, SMEDLEY, P L, and BESIEN, T. 2003. The Great and Inferior Oolite of the Cotswolds District. British Geological Survey and Environment Agency, BGS Report CR/03/202N; Environment Agency Report NC/99/74/7 (Keyworth and Solihull).
  6. THE WATER SUPPLY REGULATIONS. 2010. Statutory Instrument 2010 No. 991.
  7. 7.0 7.1 EDMUNDS, W M, COOK, J M, KINNIBURGH, D G, MILES, D L, and TRAFFORD, J M. 1989. Trace element occurrence in British groundwaters. British Geological Survey, Research Report SD/89/3 (Keyworth, Nottingham).
  8. EDMUNDS, W M. 1996. Bromine geochemistry of British groundwaters. Mineralogical Magazine, Vol. 60, 275–284.
  9. DREVER, J I. 1997. The Geochemistry of Natural Waters: Surface and Groundwater Environments, 3rd Edition. (Upper Saddle River: Prentice-Hall, Inc.)
  10. 10.0 10.1 THE WATER SUPPLY REGULATIONS. 2010. Statutory Instrument 2010 No. 991.
  11. CORNELL, D H. 1993. Rare-earths from supernova to superconductor. Pure and Applied Chemistry, Vol. 65, 2453–2464.
  12. SMEDLEY, P L, SMITH, B, ABESSER, C, and LAPWORTH, D J. 2006. Uranium occurence and behaviour in British groundwater. British Geological Survey, BGS Report CR/06/050N.
  13. 13.0 13.1 13.2 13.3 DARLING, W G, BATH, A H, and TALBOT, J C. 2003. The O & H stable isotope composition of fresh waters in the British Isles. 2. Surface waters and groundwater. Hydrology and Earth System Sciences, Vol. 7, 183–195.
  14. CLARK, I, and FRITZ, P. 1997. Environmental Isotopes in Hydrogeology. (Boca Raton, USA: Lewise Publishers.)
  15. CRAIG, H. 1961. Isotopic variations in natural waters. Science, Vol. 133, 1702–1703
  16. EA. 2009. River Basin Management Plan Humber River Basin District, Environment Agency.
  17. NATURAL ENGLAND. 2015. National Character Area Profile: 26. Vale of Pickering www.gov.uk/natural-england.