London Atlas: Materials and methods III: data analysis

From MediaWiki
Jump to navigation Jump to search
Ferreira, A, Johnson, C C, Appleton, J D, Flight, D M A, Lister, T R, Knights, K V, Ander, L, Scheib, C, Scheib, A, Cave, M, Wragg, J, Fordyce, F and Lawley, R. 2017. London Region Atlas of Topsoil Geochemistry. British Geological Survey.

The subsets SEEN and LOND

The London Region Topsoil Dataset (LRD) is composed of the LOND and SEEN data subsets, which were collected using two different sampling densities (LOND 1 per 0.25 km2, SEEN 1 per 2 km2), and broadly correspond to two regions, which are very different in terms of population density and historical record of human occupancy. About 95% of LOND and SEEN samples are located respectively inside and outside of the GLA. LOND samples collected outside GLA (307 out of 6801) are grouped into four urban clusters located to the east of the GLA, while SEEN samples collected inside GLA (80 out of 1599) are mainly located in the northern and north-eastern London outskirts, always close to the GLA border (Figure 2). Given the different sampling densities between LOND and SEEN subsets, LRD statistics are necessarily biased towards the values observed inside the GLA, thus not adequately representing the area of the London Region Atlas of Topsoil Geochemistry. The effect of this bias is particularly important for elements for which concentrations are expected to be much higher (or much lower) inside the GLA, such as those commonly classified as anthropogenic.

Accordingly, it was decided to include statistics and graphs of LOND and SEEN subsets separately in addition to the statistics and graphs of the London Region Topsoil Dataset as a whole (LRD). By showing the summary statistics and distribution of the two subsets (SEEN and LOND), and taking into account that they mainly represent the area outside and inside the GLA respectively, an unbiased view of topsoil geochemistry within the built up area of London relative to the outskirts is obtained. This enables a better understanding of soil quality in the LRA area, the distinctive anthropogenic geochemical signal observed within and outside the GLA becoming particularly clear.

The term urban domain in Appleton et al. (2013)[1], one of the key references of the present atlas, is used as in Ander et al. (2011)[2] and Ander et al. (2013)[3], while rural domain includes both rural and semi-urban areas defined in the same document. This urban domain definition is based on a urbanisation index after the Generalised Land Use Database (GLUD) Statistics for England, 2005 (Department for Communities and Local Government, 2007[4]), a document from the Office of National Statistics (ONS). This definition is such that its complementary rural includes large open-space areas within central London, such as Richmond Park and Wimbledon Common in south-west London. For the present atlas, urban and rural are rather used to define the sampling survey type (LOND and SEEN, respectively) or when referring to the 'GLA area' or 'central London' (urban) relative to 'outwith GLA area' or 'London outskirts' (rural).


Univariate statistics and graphics

A preliminary outline of the 46 variables (44 elements, LOI, pH) in the LRD dataset and in the two subsets (SEEN and LOND) is given in Table 9. This was built in Microsoft Excel®, and includes some of the most commonly used distribution measures reporting univariate descriptive statistics, namely percentiles and some non-parametric statistics in attempting to obtain robust statistics, less dependent on outliers (Reimann et al., 2008)[5]. It includes the lower limit of detection (LLD), 9 percentiles (or quantiles, Q02, Q05, Q10, Q25, Q75, Q90, Q95, Q98, Q99), the minimum (Min) and the maximum (Max); measures of central tendency, namely, the median (Mdn), the geometric mean (GM) and the arithmetic mean (AM); some measures of dispersion, namely, the interquartile range (IQR), the median absolute deviation (MAD), the standard deviation (SD) and the geometric standard deviation (GSD); two other measures of dispersion, the coefficient of variation (CV%) and the robust coefficient of variation (CVR%), expressed in percent, are also shown, as they have the advantage of being independent of the magnitude of the data (Reimann et al., 2008[5]). The CV% is defined as the SD divided by the AM, while the CVR% is defined as the MAD divided by the Mdn. The MAD is a robust equivalent of the SD measuring the average deviation from a central value, in this case the median. Powers, also a measure of dispersion, are defined as the decimal logarithm of the ratio between the Max and the Min, thus showing the orders of magnitude of the variation. Finally, the skewness is also shown, both for normal (Skew) and log-transformed (SkewLOG) data. Skewness is a measure of asymmetry of the distribution, indicating if the tails on both sides of the AM (or the GM) balance out or not. This statistic, however, must always be evaluated together with other information, such as a histogram or other graphical representation of the distribution, as has been done for the present work. These measures of asymmetry can be helpful in deciding whether to use a linear or a logarithmic scale in the graphical representation of element concentrations, and, together with the descriptive graphics, can be used to decide the most appropriate parametric statistics, in case these are required.


Table 9 Summary table providing a selection of statistical parameters to describe analytes determined on topsoils from the London region. Number of samples is LRD = 8400 (7928 for LOI and 7929 for pH), SEEN = 1599 (1128 for LOI and pH) and LOND = 6801 (6800 for LOI). For further explanation go to Univariate statistics and graphics
Analytes
units
Dataset LLD

Min

Q02

Q05

Q10

Q25

Mdn

Q75

Q90

Q95

Q98

Q99

Max

IQR

MAD

CVR%

AM

SD

CV%

GM

GSD

Powers

Skew

Skew
LOG

Al2O3

%

LRD 0.2

0.8

3.0

3.9

4.7

5.9

7.6

10.0

12.0

13.2

14.4

15.3

25.5

4.1

2.8

37.1

8.0

2.9

35.9

7.5

1.47

1.50

0.52

-0.71

SEEN 0.2

0.8

2.3

3.0

3.8

5.7

8.2

10.7

12.2

13.4

14.6

15.6

19.9

4.9

3.7

45.2

8.2

3.3

39.7

7.4

1.62

1.40

0.09

-1.11

LOND 0.2

0.8

3.4

4.3

4.9

5.9

7.5

9.7

11.9

13.2

14.4

15.2

25.5

3.8

2.7'

35.6'

8.0

2.8

34.9

7.5

1.43

1.50

0.66

-0.44

CaO

%

LRD 0.05

<0.05

0.27

0.37

0.50

0.74

1.18

2.23

4.31

6.95

15.45

23.26

48.97

1.49

0.84

71.3

2.30

3.97

172.7

1.34

2.51

3.09

5.58

0.62

SEEN 0.05

<0.05

0.11

0.18

0.29

0.55

0.80

1.44

5.75

13.14

28.04

32.48

47.47

0.89

0.49

61.4

2.64

5.97

226.0

0.99

3.33

3.07

4.20

0.84

LOND 0.05

0.22

0.34

0.43

0.56

0.81

1.31

2.31

4.20

6.15

12.84

18.54

48.97

1.49

0.91

69.1

2.22

3.33

149.9

1.44

2.29

2.35

5.86

0.76

Fe2O3

%

LRD 0.01

0.13

1.56

2.09

2.48

3.10

3.80

4.70

5.52

6.09

6.66

7.14

15.59

1.60

1.17

30.7

3.93

1.27

32.4

3.71

1.43

2.08

0.87

-1.42

SEEN 0.01

0.13

0.91

1.51

2.00

2.80

3.65

4.53

5.25

5.90

6.62

7.31

15.59

1.73

1.29

35.3

3.70

1.42

38.3

3.39

1.59

2.08

1.06

-1.83

LOND 0.01

0.15

1.81

2.24

2.56

3.15

3.82

4.76

5.57

6.11

6.66

7.12

15.37

1.61

1.14'

29.7'

3.98

1.23

30.9

3.79

1.39

2.00

0.85

-0.89

K2O

%

LRD 0.01

0.12

0.57

0.74

0.88

1.08

1.32

1.71

2.07

2.28

2.48

2.58

3.47

0.63

0.43'

32.6'

1.41

0.47

33.2

1.33

1.42

1.46

0.52

-0.72

SEEN 0.01

0.15

0.48

0.63

0.79

1.05

1.36

1.83

2.15

2.32

2.51

2.65

3.47

0.78

0.55'

40.3'

1.43

0.52

36.6

1.32

1.52

1.36

0.26

-1.01

LOND 0.01

0.12

0.63

0.78

0.90

1.09

1.32

1.68

2.06

2.27

2.46

2.57

3.33

0.59

0.42'

31.4'

1.40

0.45

32.2

1.33

1.40

1.44

0.59

-0.57

MgO

%

LRD 0.3

<0.3

0.3

0.4

0.4

0.6

0.8

1.1

1.4

1.6

2.0

2.2

4.6

0.5

0.3'

37.1'

0.9

0.4

48.5

0.8

1.59

1.36

1.65

-0.16

SEEN 0.3

<0.3

<0.3

0.3

0.5

0.7

0.8

1.0

1.2

1.4

1.8

2.0

4.2

0.3

0.3'

37.1'

0.9

0.3

40.6

0.8

1.53

1.32

1.78

-0.82

LOND 0.3

<0.3

0.3

0.4

0.4

0.6

0.8

1.1

1.4

1.7

2.0

2.3

4.6

0.5

0.3'

37.1'

0.9

0.4

50.2

0.8

1.61

1.36

1.62

-0.05

Analytes
units
Dataset LLD

Min

Q02

Q05

Q10

Q25

Mdn

Q75

Q90

Q95

Q98

Q99

Max

IQR

MAD

CVR%

AM

SD

CV%

GM

GSD

Powers

Skew

Skew
LOG

MnO

%

LRD 0.005

<0.005

0.013

0.022

0.030

0.043

0.056

0.075

0.108

0.147

0.204

0.264

0.697

0.032

0.022

39.6

0.066

0.047

70.7

0.056

1.79

2.24

4.01

-0.35

SEEN 0.005

<0.005

0.008

0.016

0.025

0.043

0.071

0.108

0.162

0.205

0.261

0.290

0.573

0.065

0.047

66.9

0.085

0.062

73.6

0.065

2.21

2.16

1.98

-0.74

LOND 0.005

<0.005

0.016

0.024

0.031

0.043

0.055

0.070

0.093

0.120

0.170

0.226

0.697

0.027

0.019

35.2

0.062

0.041

66.6

0.054

1.67

2.24

5.23

-0.27

Na2O

%

LRD 0.3

<0.3

<0.3

<0.3

0.3

0.3

0.4

0.5

0.6

0.7

0.8

0.9

10.0

0.2

0.1

34.2

0.4

0.2

47.5

0.4

1.39

1.70

21.43

0.24

SEEN 0.3

<0.3

<0.3

<0.3

<0.3

0.3

0.4

0.6

0.7

0.7

0.8

0.9

1.1

0.3

0.1

37.1

0.4

0.2

37.8

0.4

1.47

0.74

0.65

-0.12

LOND 0.3

<0.3

<0.3

0.3

0.3

0.3

0.4

0.5

0.6

0.7

0.8

0.9

10.0

0.2

0.1

34.2

0.4

0.2

49.5

0.4

1.38

1.70

23.28

0.40

P2O5

%

LRD 0.05

<0.05

0.12

0.15

0.18

0.22

0.30

0.40

0.54

0.66

0.88

1.09

4.49

0.18

0.12

39.5

0.34

0.22

64.3

0.30

1.60

2.05

5.82

0.34

SEEN 0.05

<0.05

0.09

0.12

0.14

0.18

0.24

0.30

0.39

0.47

0.59

0.77

3.26

0.12

0.09

37.0

0.26

0.17

64.3

0.24

1.57

1.91

8.38

0.00

LOND 0.05

<0.05

0.13

0.16

0.19

0.24

0.31

0.42

0.56

0.69

0.91

1.13

4.49

0.18

0.13

43.1

0.36

0.23

62.7

0.32

1.58

2.05

5.67

0.50

SiO2

%

LRD 0.1

4.6

34.9

48.0

54.2

60.2

66.6

73.3

79.2

82.7

86.4

89.1

100.0

13.1

9.8

14.7

66.1

11.4

17.3

64.8

1.25

1.34

-1.04

-3.48

SEEN 0.1

5.9

20.6

39.0

52.0

61.6

69.7

76.8

83.4

86.9

90.4

93.5

100.0

15.2

11.3

16.2

67.7

14.7

21.6

65.3

1.37

1.23

-1.37

-3.34

LOND 0.1

4.6

39.3

49.1

54.5

59.9

66.0

72.6

78.1

81.4

84.6

87.1

100.0

12.7

9.3

14.2

65.7

10.5

16.0

64.7

1.22

1.34

-0.90

-3.17

TiO2

%

LRD 0.01

0.17

0.33

0.37

0.41

0.47

0.57

0.70

0.82

0.88

0.93

0.97

1.18

0.23

0.16

28.3

0.59

0.16

26.5

0.57

1.31

0.83

0.44

-0.21

SEEN 0.01

0.18

0.29

0.34

0.38

0.49

0.63

0.75

0.86

0.90

0.95

0.98

1.06

0.26

0.19

30.6

0.62

0.17

28.1

0.59

1.36

0.78

-0.05

-0.73

LOND 0.01

0.17

0.34

0.38

0.42

0.47

0.56

0.68

0.81

0.87

0.93

0.96

1.18

0.21

0.15

26.2

0.59

0.15

25.9

0.57

1.29

0.83

0.57

-0.04

Analytes
units
Dataset LLD

Min

Q02

Q05

Q10

Q25

Mdn

Q75

Q90

Q95

Q98

Q99

Max

IQR

MAD

CVR%

AM

SD

CV%

GM

GSD

Powers

Skew

Skew
LOG

Ag

mg/kg

LRD 0.5

<0.5

<0.5

<0.5

<0.5

0.5

0.5

0.5

1.0

2.2

6.2

11.5

268.8

0.0

0.0

0.0

1.1

5.8

552.0

0.6

1.95

2.95

30.49

3.35

SEEN 0.5

<0.5

<0.5

<0.5

<0.5

0.5

0.5

0.5

0.5

1.4

4.2

7.3

237.2

0.0

0.0

0.0

0.9

6.2

681.6

0.5

1.74

2.90

34.46

4.35

LOND 0.5

<0.5

<0.5

<0.5

<0.5

0.5

0.5

0.5

1.1

2.5

6.6

12.4

268.8

0.0

0.0

0.0

1.1

5.7

525.9

0.6

1.99

2.95

29.22

3.19

As

mg/kg

LRD 2.4

<2.4

6.8

8.6

10.0

12.2

14.8

18.2

22.9

27.5

36.9

46.5

160.9

6.0

4.3

29.1

16.3

8.2

50.7

15.0

1.46

1.91

5.42

0.31

SEEN 2.4

<2.4

4.5

6.0

7.6

10.2

12.7

15.2

17.8

20.4

25.4

29.3

111.8

5.0

3.7

29.2

13.1

5.6

43.1

12.1

1.48

1.75

5.36

-0.65

LOND 2.4

<2.4

8.1

9.5

10.7

12.8

15.4

18.8

24.0

28.5

38.8

48.8

160.9

6.0

4.3

27.9

17.0

8.6

50.4

15.8

1.43

1.91

5.46

0.76

Ba

mg/kg

LRD 1

139

229

257

283

324

371

417

490

577

738

885

3475

93

70

18.8

389

140

36.0

374

1.30

1.40

6.48

1.26

SEEN 1

139

198

228

248

298

340

376

405

422

443

518

1850

78

58

17.0

339

91

26.7

330

1.24

1.12

6.88

0.40

LOND 1

143

242

271

293

331

380

429

512

602

763

920

3475

98

73

19.1

401

147

36.6

386

1.30

1.39

6.43

1.41

Bi

mg/kg

LRD 0.3

<0.3

<0.3

<0.3

<0.3

<0.3

<0.3

0.6

1.2

2.2

4.2

7.5

70.5

0.4

0.0

0.0

0.8

2.5

338.5

0.4

2.42

2.55

14.43

1.72

SEEN 0.3

<0.3

<0.3

<0.3

<0.3

<0.3

<0.3

0.4

0.7

1.0

1.9

3.2

59.3

0.2

0.0

0.0

0.5

1.9

381.4

0.3

1.94

2.47

23.74

2.17

LOND 0.3

<0.3

<0.3

<0.3

<0.3

<0.3

<0.3

0.6

1.4

2.4

4.7

8.1

70.5

0.4

0.0

0.0

0.8

2.7

328.9

0.4

2.51

2.55

13.41

1.61

Br

mg/kg

LRD 0.8

1.1

5.3

6.4

7.4

9.2

11.4

14.5

18.5

22.1

27.4

33.6

241.1

5.3

3.9

33.8

12.7

6.9

54.5

11.6

1.48

2.34

8.76

0.34

SEEN 0.8

1.1

4.2

5.5

6.4

7.9

9.5

11.6

14.6

17.7

23.3

28.2

94.8

3.7

2.7

28.1

10.4

5.6

53.5

9.6

1.47

1.94

6.35

0.28

LOND 0.8

1.5

5.7

6.8

7.8

9.7

12.0

15.0

19.0

22.5

28.5

34.1

241.1

5.3

3.9

32.1

13.2

7.1

53.7

12.1

1.47

2.21

9.22

0.42

Analytes
units
Dataset LLD

Min

Q02

Q05

Q10

Q25

Mdn

Q75

Q90

Q95

Q98

Q99

Max

IQR

MAD

CVR%

AM

SD

CV%

GM

GSD

Powers

Skew

Skew
LOG

Cd

mg/kg

LRD 0.5

<0.5

<0.5

<0.5

<0.5

<0.5

0.6

0.8

1.3

1.9

3.8

6.8

165.2

0.5

0.3

49.4

0.9

3.5

375.2

0.6

1.95

2.74

30.07

1.81

SEEN 0.5

<0.5

<0.5

<0.5

<0.5

<0.5

0.5

0.6

0.9

1.3

2.1

3.4

135.8

0.3

0.3

59.3

0.8

4.7

605.7

0.5

1.81

2.66

25.08

2.81

LOND 0.5

<0.5

<0.5

<0.5

<0.5

0.5

0.6

0.8

1.4

2.1

4.2

7.3

165.2

0.3

0.3

49.4

1.0

3.1

325.4

0.6

1.95

2.74

31.43

1.72

Ce

mg/kg

LRD 1

16

29

33

37

43

50

59

68

74

85

101

238

16

11.9

23.7

51.8

14.2

27.5

50.1

1.29

1.17

1.93

0.11

SEEN 1

16

24

29

33

44

55

65

75

86

106

117

165

21

14.8

27.0

55.7

18.3

32.8

52.8

1.39

1.01

1.04

-0.38

LOND 1

18

31

34

38

43

50

57

65

70

79

91

238

14

10.4

20.8

50.9

12.9

25.4

49.5

1.26

1.12

2.27

0.27

Co

mg/kg

LRD 1.5

<1.5

3.6

5.3

6.7

9.0

11.4

14.4

18.0

20.8

26.0

31.6

85.2

5.4

3.9

33.8

12.2

5.5

45.5

11.1

1.57

1.93

2.63

-0.90

SEEN 1.5

<1.5

2.1

3.4

4.9

8.1

11.4

14.6

19.3

23.7

29.1

36.4

69.9

6.5

4.9

42.9

12.1

6.6

55.0

10.4

1.82

1.84

1.95

-1.04

LOND 1.5

<1.5

4.3

5.8

7.1

9.2

11.4

14.3

17.7

20.2

24.7

30.6

85.2

5.1

3.7

32.5

12.2

5.2

43.0

11.3

1.50

1.93

2.91

-0.59

Cr

mg/kg

LRD 3

9

38

46

52

61

73

88

104

116

140

172

2094

27

19

26.4

78

45

58.3

73

1.38

2.37

20.75

0.79

SEEN 3

9

32

39

47

61

74

88

104

118

133

155

718

27

19

26.0

77

35

45.3

72

1.42

1.90

8.39

-0.22

LOND 3

15

41

47

52

61

72

89

104

115

142

177

2094

28

19

26.8

78

48

60.8

74

1.37

2.14

21.51

1.15

Cs

mg/kg

LRD 1

1.0

1.0

1.5

2.0

2.0

3.0

4.0

5.0

6.0

6.9

7.0

11.2

2.0

1.5

49.4

3.2

1.4

44.5

2.9

1.55

1.05

1.12

-0.14

SEEN 1

1.0

1.0

2.0

2.0

2.0

3.0

4.7

5.8

6.9

7.0

8.0

11.2

2.7

1.5

49.4

3.5

1.6

45.0

3.2

1.58

1.05

0.92

-0.18

LOND 1

1.0

1.0

1.0

2.0

2.0

3.0

4.0

5.0

6.0

6.0

7.0

11.0

2.0

1.5

49.4

3.1

1.3

43.6

2.8

1.54

1.04

1.14

-0.15

Analytes
units
Dataset LLD

Min

Q02

Q05

Q10

Q25

Mdn

Q75

Q90

Q95

Q98

Q99

Max

IQR

MAD

CVR%

AM

SD

CV%

GM

GSD

Powers

Skew

Skew
LOG

Cu

mg/kg

LRD 1.3

3

10

13

17

24

38

67

114

167

285

433

5326

43

25.7

67.7

63.5

132

207.8

41.7

2.23

3.25

19.83

0.71

SEEN 1.3

3

7

9

11

17

22

29

41

55

96

180

1055

13

8.3

38.4

29.2

54

186.2

22.1

1.83

2.55

12.97

1.16

LOND 1.3

3

13

16

20

28

45

75

126

184

315

461

5326

47

29.7

65.9

71.6

143

199.9

48.4

2.15

3.25

18.91

0.76

Ga

mg/kg

LRD 1.0

1.0

6.5

7.2

7.9

9.1

10.6

12.9

15.1

16.3

18.0

18.9

44.1

3.8

2.7

25.2

11.1

2.9

25.9

10.8

1.29

1.64

0.92

-0.04

SEEN 1.0

3.1

5.8

6.3

7.2

9.0

11.1

13.5

14.9

16.2

17.7

18.9

34.7

4.5

3.4

30.7

11.2

3.1

27.7

10.8

1.34

1.05

0.48

-0.43

LOND 1.0

1.0

6.9

7.5

8.0

9.1

10.6

12.7

15.1

16.4

18.1

18.9

44.1

3.6

2.5

23.8

11.1

2.8

25.5

10.8

1.28

1.64

1.04

0.11

Ge

mg/kg

LRD 0.5

<0.5

<0.5

0.5

0.7

1.0

1.4

2.1

3.1

4.1

5.9

7.6

38.7

1.1

0.7

53.0

1.8

1.6

91.4

1.4

1.92

2.11

7.17

0.06

SEEN 0.5

<0.5

<0.5

<0.5

0.5

0.8

1.1

1.4

1.7

1.9

2.4

2.9

38.7

0.6

0.4

40.4

1.2

1.2

99.9

1.0

1.65

2.11

23.26

-0.26

LOND 0.5

<0.5

<0.5

0.5

0.7

1.1

1.5

2.3

3.4

4.3

6.3

8.1

32.7

1.2

0.7

49.4

1.9

1.7

87.6

1.5

1.93

2.04

6.06

-0.03

Hf

mg/kg

LRD 1.0

1.1

3.8

4.5

5.2

6.2

7.4

9.1

11.3

12.8

14.3

15.8

40.7

2.9

2.1

28.0

7.9

2.7

34.5

7.5

1.38

1.57

2.02

-0.03

SEEN 1.0

1.2

3.7

4.7

5.9

7.6

9.5

11.7

13.5

14.5

16.5

18.7

31.2

4.1

3.1

32.8

9.7

3.2

33.5

9.1

1.43

1.41

0.94

-0.86

LOND 1.0

1.1

3.8

4.5

5.1

6.0

7.2

8.6

10.1

11.5

13.4

14.6

40.7

2.6

1.9

26.8

7.5

2.4

32.1

7.2

1.34

1.57

2.58

0.03

I

mg/kg

LRD 0.5

<0.5

1.2

1.5

1.8

2.4

3.1

4.2

6.6

8.8

12.3

16.7

79.9

1.8

1.2

38.3

3.9

3.4

86.5

3.3

1.72

2.30

7.12

0.65

SEEN 0.5

<0.5

1.1

1.4

1.8

2.7

3.9

5.7

8.2

10.0

12.6

15.5

32.9

3.0

2.1

53.2

4.6

3.1

68.1

3.9

1.82

1.92

3.03

-0.12

LOND 0.5

<0.5

1.2

1.5

1.8

2.4

3.0

3.9

5.8

8.2

12.2

16.8

79.9

1.5

1.0

34.6

3.7

3.4

91.2

3.2

1.68

2.30

8.02

0.87

Analytes
units
Dataset LLD

Min

Q02

Q05

Q10

Q25

Mdn

Q75

Q90

Q95

Q98

Q99

Max

IQR

MAD

CVR%

AM

SD

CV%

GM

GSD

Powers

Skew

Skew
LOG

La

mg/kg

LRD 1

3

12

14

16

20

25

30

36

40

49

59

134

10

7.4

29.7

25.9

10

37.0

24.4

1.41

1.65

2.41

-0.19

SEEN 1

3

9

12

16

22

28

34

41

47

59

73

134

12

8.9

31.8

29.0

12

41.5

26.7

1.53

1.65

1.94

-0.80

LOND 1

3

12

14

16

20

24

29

34

38

45

55

130

9

7.4

30.9

25.1

9

34.7

23.9

1.37

1.64

2.49

0.00

Mo

mg/kg

LRD 0.2

<0.2

0.4

0.6

0.7

1.0

1.4

2.0

2.8

3.6

5.5

7.5

561.2

1.0

0.7

53.0

1.9

8.1

432.6

1.4

1.83

3.75

52.74

0.56

SEEN 0.2

<0.2

0.3

0.4

0.5

0.7

0.9

1.2

1.5

1.8

2.6

3.8

192.2

0.5

0.3

32.9

1.1

4.8

424.7

0.9

1.70

3.28

38.41

0.48

LOND 0.2

<0.2

0.5

0.7

0.8

1.1

1.5

2.1

3.0

3.8

5.9

7.8

561.2

1.0

0.7

49.4

2.0

8.7

424.7

1.6

1.77

3.75

51.32

0.79

Nb

mg/kg

LRD 1.0

4.7

8.3

9.3

9.9

11.0

12.6

15.1

16.9

17.8

18.5

19.0

146.7

4.1

2.8

22.4

13.1

3.1

23.5

12.8

1.23

1.49

10.08

0.01

SEEN 1.0

4.7

7.2

8.2

9.2

11.3

14.2

16.4

17.7

18.3

18.8

19.1

23.4

5.1

3.7

26.1

13.8

3.2

23.3

13.3

1.29

0.70

-0.35

-0.86

LOND 1.0

5.3

8.8

9.4

10.0

11.0

12.5

14.6

16.6

17.5

18.4

18.8

146.7

3.6

2.5

20.2

12.9

3.0

23.4

12.6

1.22

1.44

13.26

0.36

Nd

mg/kg

LRD 4.0

<4.0

8.0

10.4

12.9

17.0

21.8

26.9

32.5

36.7

45.5

54.8

172.7

9.9

7.3

33.3

22.7

9.7

42.5

21.0

1.49

1.69

3.05

-0.47

SEEN 4.0

<4.0

5.8

8.4

11.4

17.8

24.2

30.7

36.7

41.7

55.1

71.2

172.7

12.9

9.6

39.8

25.1

12.6

50.3

22.3

1.66

1.69

3.08

-0.83

LOND 4.0

<4.0

8.6

11.0

13.2

16.9

21.3

26.1

31.1

34.7

42.0

51.4

122.8

9.2

6.7

31.3

22.2

8.7

39.4

20.7

1.45

1.55

2.70

-0.34

Ni

mg/kg

LRD 1.3

<1.3

8.0

11.3

13.9

18.7

24.6

32.2

41.3

49.6

62.3

78.3

505.6

13.5

9.8

39.8

27.2

16.8

61.8

24.2

1.62

2.62

8.77

-0.40

SEEN 1.3

<1.3

3.7

6.6

9.6

15.2

21.1

30.0

38.7

45.6

56.8

74.6

469.4

14.8

10.4

49.2

24.1

20.0

82.7

20.1

1.86

2.59

10.41

-0.73

LOND 1.3

2.3

10.0

12.7

15.1

19.5

25.4

32.6

41.8

50.2

63.7

79.0

505.6

13.1

9.3

36.8

27.9

15.9

57.0

25.3

1.54

2.34

8.05

0.16

Analytes
units
Dataset LLD

Min

Q02

Q05

Q10

Q25

Mdn

Q75

Q90

Q95

Q98

Q99

Max

IQR

MAD

CVR%

AM

SD

CV%

GM

GSD

Powers

Skew

Skew
LOG

Pb

mg/kg

LRD 1.3

10

32

38

46

70

138

284

531

775

1222

1668

25206

214

122

88.

251

477

190.3

149

2.59

3.40

22.98

0.46

SEEN 1.3

10

25

30

34

40

55

81

126

184

325

484

1914

41

25

45.8

79

105

133.4

61

1.82

2.28

8.95

1.39

LOND 1.3

11

39

49

60

94

175

332

593

845

1323

1864

25206

238

145

83.0

291

520

178.4

184

2.45

3.36

21.74

0.42

Rb

mg/kg

LRD 1.0

6.5

25.5

33.1

38.4

45.9

56.5

72.5

89.5

98.0

107.4

113.5

157.2

26.6

18.7

33.1

60.3

20.1

33.4

57.0

1.41

1.38

0.64

-0.50

SEEN 1.0

6.5

21.1

27.3

35.4

47.2

62.8

80.0

95.7

103.0

111.0

114.1

157.2

32.8

24.2

38.5

63.8

23.0

36.1

59.1

1.52

1.38

0.21

-1.05

LOND 1.0

8.9

28.4

34.4

38.9

45.8

55.8

70.8

87.7

96.4

106.5

112.9

148.6

25.0

17.3

31.1

59.5

19.3

32.4

56.5

1.39

1.22

0.77

-0.26

Sb

mg/kg

LRD 0.5

<0.5

0.5

0.7

0.9

1.3

2.4

4.5

8.2

12.7

22.0

33.1

612.3

3.2

1.9

80.3

4.5

13.2

294.8

2.6

2.48

3.18

26.04

0.67

SEEN 0.5

<0.5

<0.5

0.5

0.6

0.8

1.0

1.4

2.2

3.1

5.7

9.9

47.1

0.6

0.4

44.5

1.4

2.1

148.4

1.1

1.81

2.07

11.40

1.42

LOND 0.5

<0.5

0.7

1.0

1.2

1.7

2.9

5.2

9.3

14.2

24.9

38.2

612.3

3.5

2.1

71.6

5.2

14.6

279.9

3.2

2.34

3.18

23.91

0.76

Sc

mg/kg

LRD 3.0

<3.0

<3.0

3.0

4.0

5.8

7.8

10.2

12.6

14.0

15.7

16.9

35.6

4.4

3.3

41.8

8.1

3.4

41.5

7.4

1.60

1.25

0.61

-0.83

SEEN 3.0

<3.0

<3.0

<3.0

<3.0

4.6

7.3

10.0

12.3

13.7

15.6

17.3

35.6

5.4

4.0

54.8

7.5

3.8

50.6

6.4

1.81

1.25

0.65

-0.65

LOND 3.0

<3.0

<3.0

3.5

4.5

6.0

7.9

10.3

12.6

14.1

15.7

16.8

33.3

4.3

3.1

39.4

8.3

3.2

39.2

7.6

1.54

1.22

0.66

-0.73

Se

mg/kg

LRD 0.2

<0.2

0.2

0.2

0.3

0.4

0.5

0.7

1.0

1.2

1.7

2.3

19.6

0.3

0.3

59.3

0.6

0.6

92.6

0.5

1.72

2.29

14.13

0.09

SEEN 0.2

<0.2

<0.2

0.2

0.2

0.3

0.4

0.6

0.7

0.9

1.3

1.7

15.4

0.3

0.1

37.1

0.5

0.5

101.4

0.4

1.69

2.19

18.10

0.14

LOND 0.2

<0.2

0.2

0.2

0.3

0.4

0.6

0.8

1.0

1.3

1.8

2.3

19.6

0.4

0.3

49.4

0.7

0.6

90.1

0.6

1.71

2.29

13.75

0.09

Analytes
units
Dataset LLD

Min

Q02

Q05

Q10

Q25

Mdn

Q75

Q90

Q95

Q98

Q99

Max

IQR

MAD

CVR%

AM

SD

CV%

GM

GSD

Powers

Skew

Skew
LOG

Sn

mg/kg

LRD 0.5

0.9

2.6

3.1

3.7

5.6

10.9

22.6

44.2

67.2

124.4

169.1

1041.5

17.0

9.6

88.4

21.5

40.0

186.2

12.1

2.64

3.06

9.24

0.62

SEEN 0.5

1.2

2.2

2.5

2.7

3.4

4.5

7.3

11.9

18.9

38.1

71.8

466.5

3.9

2.2

49.4

8.1

19.7

243.5

5.3

2.00

2.59

13.61

1.80

LOND 0.5

0.9

3.2

4.0

4.7

7.3

13.6

25.7

50.0

75.9

133.5

181.1

1041.5

18.4

11.4

83.9

24.6

42.8

173.9

14.6

2.52

3.06

8.83

0.59

Sr

mg/kg

LRD 1

11

29

38

45

57

73

93

125

153

217

272

601

36

25.2

34.5

82.2

44.4

54.0

74.2

1.55

1.74

3.24

0.34

SEEN 1

11

24

29

35

48

63

80

115

171

277

330

576

32

22.2

35.3

75.1

54.9

73

64.3

1.68

1.72

3.53

0.71

LOND 1

12

33

41

48

60

76

96

126

152

198

254

601

36

25.2

33.2

83.9

41.4

49.3

76.8

1.50

1.70

3.09

0.34

Th

mg/kg

LRD 0.7

<0.7

3.1

3.8

4.4

5.4

6.8

8.4

9.5

10.1

10.6

11.0

456.5

3.0

2.2

32.7

7.0

5.3

76.1

6.6

1.38

2.88

73.02

-0.69

SEEN 0.7

<0.7

2.2

3.1

3.9

5.7

7.7

9.1

10.0

10.4

10.8

11.0

15.7

3.4

2.4

30.8

7.3

2.3

31.6

6.8

1.49

1.42

-0.49

-1.49

LOND 0.7

<0.7

3.3

4.0

4.5

5.4

6.6

8.2

9.4

9.9

10.6

11.0

456.5

2.8

2.1

31.4

6.9

5.8

83.9

6.5

1.35

2.88

69.56

-0.29

U

mg/kg

LRD 0.5

<0.5

1.1

1.4

1.6

1.9

2.2

2.6

2.9

3.1

3.4

3.6

11.5

0.7

0.5

24.7

2.2

0.6

26.1

2.2

1.34

1.46

0.88

-1.55

SEEN 0.5

<0.5

1.1

1.3

1.5

1.9

2.3

2.6

2.9

3.1

3.2

3.4

7.3

0.7

0.5

23.6

2.2

0.6

24.9

2.2

1.31

1.26

0.43

-1.00

LOND 0.5

<0.5

1.1

1.4

1.6

1.9

2.2

2.6

2.9

3.2

3.4

3.6

11.5

0.7

0.5

24.2

2.2

0.6

26.4

2.1

1.34

1.46

0.96

-1.65

V

mg/kg

LRD 3

9

36

44

50

61

75

98

122

135

151

163

531

37

25.2

33.6

81.4

29.7

36.4

76.5

1.43

1.77

1.41

-0.24

SEEN 3

9

25

33

41

54

71

91

113

128

146

161

352

37

26.7

37.6

74.5

30.1

40.4

68.6

1.52

1.59

1.24

-0.59

LOND 3

16

40

47

53

62

76

100

124

137

151

163

531

38

25.2

33.2

83.1

29.3

35.3

78.4

1.40

1.52

1.50

0.04

Analytes
units
Dataset LLD

Min

Q02

Q05

Q10

Q25

Mdn

Q75

Q90

Q95

Q98

Q99

Max

IQR

MAD

CVR%

AM

SD

CV%

GM

GSD

Powers

Skew

Skew
LOG

W

mg/kg

LRD 0.6

<0.6

<0.6

0.8

1.1

1.6

2.1

2.6

3.3

3.9

5.6

8.4

316.8

1.0

0.7

35.3

2.4

5.8

241.8

2.0

1.70

2.90

37.19

0.21

SEEN 0.6

<0.6

<0.6

<0.6

0.8

1.6

2.4

3.0

3.5

4.0

4.8

5.5

184.6

1.4

1.0

43.2

2.5

4.8

193.1

2.0

1.83

2.66

35.32

-0.70

LOND 0.6

<0.6

<0.6

0.9

1.1

1.6

2.0

2.5

3.2

3.9

6.2

8.7

316.8

0.9

0.7

37.1

2.4

6.1

252.4

2.0

1.67

2.90

36.98

0.55

Y

mg/kg

LRD 1

3

9

12

14

17

21

25

29

32

42

50

158

8.0

5.9

28.2

21.6

8.5

39.3

20.3

1.40

1.72

4.16

0.01

SEEN 1

3

7

9

12

18

23

28

32

39

51

69

158

10.5

7.4

32.2

23.5

11.3

48.1

21.4

1.55

1.72

3.47

-0.41

LOND 1

5

10

12

14

17

20

24

28

30

37

48

133

7.0

5.9

29.7

21.1

7.6

35.9

20.1

1.35

1.42

4.24

0.18

Zn

mg/kg

LRD 1.3

<1.3

27

44

58

81

130

225

371

521

801

1120

10095

144

89

68.4

195

269

138.0

137

2.21

4.00

13.90

0.14

SEEN 1.3

<1.3

18

28

42

60

77

98

136

176

254

446

2505

38

27

34.7

94

120

127.8

76

1.84

3.40

11.83

-0.42

LOND 1.3

<1.3

35

51

65

95

152

252

412

563

842

1214

10095

157

102

67.3

219

288

131.7

158

2.15

4.00

13.57

0.15

Zr

mg/kg

LRD 1

35

153

182

204

240

288

350

424

478

536

580

1488

110

79

27.3

304

99

32.4

290

1.36

1.63

2.02

-0.14

SEEN 1

48

137

182

219

286

351

434

504

537

596

646

1062

148

107

30.4

360

115

32.1

340

1.43

1.34

0.60

-1.14

LOND 1

35

156

182

202

235

277

330

391

437

496

552

1488

95

70

25.2

291

89

30.7

279

1.32

1.63

2.71

0.05

LOI

%

LRD 0.01

0.7

3.2

3.9

4.5

5.7

7.1

8.9

10.9

12.5

15.5

18.3

72.1

3.2

2.4

33.4

7.6

3.3

43.8

7.1

1.45

2.01

4.27

0.10

SEEN 0.01

1.2

2.8

3.3

3.8

4.8

6.0

8.0

10.6

12.8

15.9

18.7

72.1

3.2

2.1

34.6

6.9

4.3

61.7

6.2

1.53

1.78

6.31

0.72

LOND 0.01

0.7

3.3

4.0

4.7

5.8

7.3

9.1

11.0

12.5

15.5

18.3

58.7

3.3

2.4

32.5

7.8

3.2

40.6

7.3

1.43

1.92

3.46

0.02

pH

log[H+] mol/L

LRD 0.01

2.8

3.4

4.0

4.6

5.5

6.5

7.0

7.3

7.4

7.5

7.6

8.4

1.5

0.9

14.1

6.2

1.1

17.3

6.1

1.22

0.48

-0.96

-1.38

SEEN 0.01

2.9

3.2

3.6

4.3

5.0

6.2

7.1

7.4

7.5

7.5

7.6

7.8

2.1

1.5

24.5

5.9

1.2

20.6

5.8

1.26

0.44

-0.53

-0.95

LOND 0.01

2.8

3.5

4.1

4.7

5.6

6.6

7.0

7.3

7.4

7.5

7.6

8.4

1.4

0.8

12.9

6.2

1.0

16.6

6.1

1.21

0.48

-1.03

-1.45



Following Table 9, a set of box-plots of the 44 elements are shown in Figure 7, Figure 8, and Figure 9 for LRD, SEEN and LOND respectively. These were generated running R (Rx64 3.1.0) in RStudio (Version 0.98.977) with an R script derived mainly from functions and scripts in StatDA R package (Reimann et al., 2008[5]). The elements are ordered by abundance in the SEEN subset. Each figure is composed of two graphics, one (on the left) showing the 10 most abundant elements (expressed as oxides in wt%) and another (on the right) showing the less abundant elements (in mg/kg); Mn is included in both graphics allowing a visual calibration of the concentration scale (Y axis). The concentration scales of these three figures are log-decimal, thus allowing visualisation of elements with orders of magnitude difference in content on the same graphic. This also has advantage when showing elements with a highly positive skewed distribution.



Figure 7 Box-plot of element concentrations in topsoil in the LRD dataset (8400 samples), shown by order of abundance in the SEEN subset. The 10 most abundant elements (expressed as oxides in wt%) on the left and the less abundant (in mg/kg) on the right. Manganese is shown as MnO and as Mn for reference between both graphs. (P929864).


Figure 8 Box-plot of element concentrations in topsoil in the SEEN subset (1599 samples), shown by order of abundance in the SEEN subset. The 10 most abundant elements (expressed as oxides in wt%) on the left and the less abundant (in mg/kg) on the right. Manganese is shown as MnO and as Mn for reference between both graphs. (P929865).


Figure 9 Box-plot of element concentrations in topsoil in the LOND subset (6801 samples), shown by order of abundance in the SEEN subset. The 10 most abundant elements (expressed as oxides in wt%) on the left and the less abundant (in mg/kg) on the right. Manganese is shown as MnO and as Mn for reference between both graphs. (P929866).


Adjacent to each element map, a univariate table of statistics (e.g.: Figure 10) and a set of statistical graphics are shown. These were also generated running R (Rx64 3.1.0) in RStudio (Version 0.98.977) using an R script derived from functions and scripts from StatDA[6] R package (Reimann et al., 2008[5]).

Where appropriate the LRD dataset, and SEEN and LOND subsets are represented in black/white, dark green and dark red respectively. Quaternary, Palaeogene and Cretaceous geological time periods are represented by soft yellow, orange and lime green colours respectively.

Graphics shown are (i) a histogram of LRD data distribution with the density lines of the SEEN and LOND subsets; (ii) a Tukey boxplot of the LRD, SEEN and LOND dataset distributions and (iii) a cumulative probability plot of the LRD, SEEN and LOND datasets (e.g.: Figure 11); (iv) the LOND and SEEN topsoil parameter concentrations over each geological unit (e.g.: Figure 12); (v) the LRD topsoil parameter concentrations over each geological unit (e.g.: Figure 13 and Figure 14).

The histogram, is used to show the LRD topsoil parameter distribution. The SEEN and LOND subsets are shown as density traces, as these allow overlap of the three data distributions (LRD, SEEN, LOND), thus directly comparing their shapes, independently of the number of soil samples. The Tukey boxplots (Tukey, 1977[7]) are built around the median (line dividing the box in two parts); the box representing the interquartile range (Q3-Q1=IQR), that is, 50% of the data, from the lower (Q1) to the upper (Q3) quartiles; the lowest end of the whiskers is 1.5 times the length of the box (IQR) starting from the lower quartile, while the highest one is 1.5 times the IQR starting from the upper quartile; in this graphical representation, data below and above the whiskers (plotted with the symbol +) are considered as outliers. The boxplots shown also include a notch around the median. This gives an approximate indication as to whether the medians are different or not. If notches of different boxplots do not overlap, there is some evidence of a statistical difference between the medians. The cumulative probability (CP) plots (Sinclair, 1976[8]) are built with a non-linear probability scale on the Y axis (from >0 to <100%). The quartiles are close together near the median (probability = 50%) and stretch out symmetrically moving away from the median. If the variable follows a normal (or lognormal) distribution, then its values (or log transformed values) fall in a straight line from the bottom left to the top right of the graph; inflexion points on the curve suggest that the distribution is made up of multiple data populations. CP plots are a good complement to the previous graphs in understanding and comparing the distribution of several datasets.

The Tukey style of boxplots used to display the LRD, SEEN and LOND datasets were used to show element concentrations over each geological parent material unit also. In the LOND and SEEN comparison, two boxplots (LOND above SEEN) are shown, together with the number of soil samples collected over each geological unit (n, on the right of the graphic) (Figure 12). For some parent material classes (e.g.: LOND over Gault Fm.), the boxplot is not shown as there are no soil samples over this rock type (Table 7). The LRD geological parent material comparison is presented in two formats. The upper set of boxplots show topsoil element concentrations over each geological unit (Figure 13), whereas the lower set of boxplots show topsoil element concentrations classified by parent material geological period (Figure 14). The number of soil samples in each geological time zone is shown in brackets and the boxplots are coloured according to the geological time period. The last boxplot refers to LRD and is shown for reference.


Figure 10 Example of a univariate table of statistics for LRD, SEEN and LOND shown next to the respective map. The case for Al2O3. These and other statistical parameters are shown in Table 9 for other elements also. For further explanation go to Univariate statistics and graphics (P929875).


Figure 11 Example of a histogram, a Tukey boxplot and a cumulative probability plot shown for LRD, SEEN and LOND next to the respective map. The case for Al2O3. For further explanation go to Here (P929874).


Figure 12 Example of a LOND and SEEN topsoil parameter concentrations over each geological unit shown next to the respective map. The case for Al2O3. For further explanation go to Here (P929876).


Figure 13 Example of the LRD topsoil parameter concentrations over each geological unit (parent material) shown next to the respective map. The case for Al2O3. For further explanation go to Here (P929877).
Figure 14 Example of the LRD topsoil parameter concentrations over each geological time period shown next to the respective map. The case for Al2O3. For further explanation go to Here (P929878).

Compositional data analysis (CoDA)

Geochemical data have an intrinsic compositional nature (Aitchison, 1986[9]), as the sum of all parts in a sample necessarily sum up to a constant (frequently 1, 100% or 1 000 000 mg/kg). These datasets are called compositional data or closed data. For closed data the concentration obtained for one part (element) does not vary independently from the others, thus the information is not absolute but only relative. As pointed out by Pearson (1897)[10], applying classical methods of statistical data analysis to this type of dataset may lead to wrong results, as for example, in the form of spurious correlations when performing a bivariate analysis. This problem was first tackled by Aitchison (1986)[9], who found that compositional data is better encompassed, not in the usual Euclidean space (on which classical statistical methods have been developed), but in the Aitchison geometry on the simplex. As the statistical evaluation and interpretation of a certain element concentration should take into account the remaining parts, the statistical methods and techniques developed in this context are mainly based on three different logratio transformations. The alr - additive logratio and the clr – centred logratio transformation, both proposed by Aitchison (1986)[9], and the ilr – isometric logratio transformation (Egozcue et al., 2003[11]). The idea of these transformations is that the data should be transformed into the correct geometry first, after which classical methods can be fully applied.

However these transformations lead to dimensionless values, which constitute a serious drawback when the main task is to document and to study the spatial distribution of element concentrations (Reimann et al., 2012[12]). This is particularly evident in the production of geochemical atlases where the primary aim is to present single element distribution maps and associated statistics (Reimann et al., 2014[13]).

As this work is a geochemical atlas, the uni-element maps and associated statistics are shown as prevoiusly explained, as the absolute values are of interest. However, readers, especially earth science practitioners, are advised to take into account that the geochemical dataset presented here is clearly a case of compositional data. Thus a conflict with the CoDA approach may exist, particularly for the statistics typically used to compare variables such as the measures of dispersion, despite of some extenuatory reasoning as referred by Filzmoser et al. (2009)[14] and Reimann et al. (2014)[13].

As pointed out above, uni-element and bivariate statistics are not among the best practices when statistically analysing compositions, because these datasets are multivariate in nature. Still, the center or compositional mean (equation 4.1 in van den Boogaart and Tolosana-Delgado, 2013[15]) is meaningful in the CoDA context, namely with respect to translation and scaling operations, referred as perturbation and powering in CoDA (van den Boogaart and Tolosana-Delgado, 2013[15]). This central tendency CoDA statistic is computed for the LRD dataset and SEEN and LOND subsets after the line command mean(x), with x being a composition, in compositions[16] R package (Table 10); the composition is based on all the 44 chemical elements and in all the topsoil samples of this atlas (8400 for LRD, 1599 for SEEN and 6801 for LOND). The compositional mean was computed after converting the trace elements to percent (%, i.e.: concentration in mg/kg / 10000); the obtained values were then back transformed to the original units (% for the 10 major and mg/kg for the 34 less abundant elements).

By computing the compositional mean in LOND and in SEEN, a preliminary overview of the geochemical signatures in urban versus rural areas can be obtained. The comparison of LOND and SEEN is better achieved after the (Log(LOND/SEEN) in Table 10, which gives an indication of element enrichment (positive) / depletion (negative) in the urban environment relative to the rural environment, and the degree (distance to zero) of this enrichment / depletion; the elements are ordered from the highest to the lowest Log(LOND/SEEN) value in order to facilitate comparison of the elements. Pb (closely followed by Sb, Sn, Cu, Zn) is the element showing the highest relative enrichment in LOND, contrary to Hf (closely followed by I, Zr, MnO) which shows the highest relative depletion in LOND; in this sense the major Al2O3 and K2O are the elements showing the lowest enrichment or depletion in LOND relative to SEEN as their Log(LOND/SEEN) values are the closest to zero among all elements. This sort of approach, i.e., the use of ratios to estimate enrichment / depletion (enrichment factors, EFs), has been used previously to examine urban data relative to rural data (e.g.: Fordyce et al., 2005[17]; Flight and Scheib, 2011[18], among many others). This and other indexes, such as the Igeo-index (Müller, G, 1979[19]), have been widely used in searching for anthropogenic impact on the surface environment since their introduction in the 1970s (e.g.: Chester and Stoner, 1973[20]; Zoller et al., 1974[21]).


Table 10 Compositional mean based on the concentration of 44 chemical elements in 8400, 1599 and 6801 topsoil samples respectively for LRD, SEEN and LOND. Log(LOND/SEEN) corresponds to the logarithm of column 6 (ratio between the LOND and the SEEN compositional mean values). Elements are ordered from the highest to lowest log(LOND/SEEN) value.

--Chemical element--

--Units--

---LRD---

---SEEN---

---LOND---

--LOND/SEEN--

--Log(LOND/SEEN)--

Pb

mg/kg

183.9

75.4

226.7

3.01

0.478

Sb

mg/kg

3.20

1.37

3.90

2.86

0.456

Sn

mg/kg

14.92

6.59

18.08

2.74

0.438

Cu

mg/kg

51.50

27.35

59.74

2.18

0.339

Zn

mg/kg

169.6

93.8

194.8

2.08

0.318

Mo

mg/kg

1.73

1.10

1.93

1.75

0.243

Ge

mg/kg

1.75

1.25

1.89

1.51

0.179

CaO

%

1.66

1.23

1.78

1.44

0.160

Cd

mg/kg

0.75

0.57

0.80

1.39

0.143

P2O5

%

0.37

0.29

0.40

1.35

0.131

Se

mg/kg

0.67

0.53

0.70

1.33

0.125

As

mg/kg

18.62

15.11

19.55

1.29

0.112

Br

mg/kg

14.35

11.91

14.99

1.26

0.100

Bi

mg/kg

0.46

0.38

0.48

1.25

0.098

Ni

mg/kg

29.97

25.00

31.26

1.25

0.097

Sr

mg/kg

91.70

79.72

94.73

1.19

0.075

Sc

mg/kg

9.10

7.95

9.39

1.18

0.072

Ba

mg/kg

462.6

409.2

476.0

1.16

0.066

V

mg/kg

94.47

85.04

96.80

1.14

0.056

Fe2O3

%

4.59

4.20

4.68

1.11

0.047

Co

mg/kg

13.71

12.86

13.91

1.08

0.034

Ag

mg/kg

0.70

0.67

0.71

1.06

0.027

Cr

mg/kg

90.53

89.00

90.86

1.02

0.009

Na2O

%

0.51

0.50

0.51

1.02

0.006

Al2O3

%

9.26

9.23

9.26

1.00

0.002

K2O

%

1.642

1.640

1.642

1.00

0.001

U

mg/kg

2.66

2.67

2.65

1.00

-0.002

Ga

mg/kg

13.31

13.37

13.29

0.99

-0.002

SiO2

%

80.1

80.9

79.8

0.99

-0.006

W

mg/kg

2.45

2.51

2.43

0.97

-0.014

MgO

%

0.96

0.98

0.95

0.97

-0.015

Th

mg/kg

8.14

8.45

8.07

0.95

-0.020

Rb

mg/kg

70.48

73.30

69.81

0.95

-0.021

TiO2

%

0.71

0.74

0.70

0.95

-0.023

Nb

mg/kg

15.77

16.53

15.59

0.94

-0.026

Y

mg/kg

25.14

26.50

24.83

0.94

-0.028

Ce

mg/kg

61.91

65.50

61.07

0.93

-0.030

Nd

mg/kg

26.06

27.78

25.67

0.92

-0.034

La

mg/kg

30.15

33.11

29.48

0.89

-0.050

Cs

mg/kg

3.55

3.98

3.45

0.87

-0.062

MnO

%

0.069

0.081

0.067

0.83

-0.080

Zr

mg/kg

358.1

421.2

344.5

0.82

-0.087

I

mg/kg

4.06

4.79

3.90

0.81

-0.089

Hf

mg/kg

9.34

11.39

8.91

0.78

-0.107

References

  1. Appleton, J D, Johnson, C C, Ander, E L, and Flight, D M A. 2013. Geogenic signatures detectable in topsoils of urban and rural domains in the London region, UK, using parent material classified data. Applied Geochemistry, Vol. 39, 169–180. DOI 10.1016/j.apgeochem.2013.07.010
  2. Ander, E L, Cave, M R, Johnson, C C, and Palumbo-Roe, B. 2011. Normal background concentrations of contaminants in the soils of England. Available data and data exploration. British Geological Survey Commissioned Report, CR/11/145.
  3. Ander, E L, Johnson, C C, Cave, M R, Palumbo-Roe, B, Nathanail, C P, and Lark, R M. 2013. Methodology for the determination of normal background concentrations of contaminants in English soil. Science of The Total Environment, Vol. 454-455, 604-618. 10.1016/j.scitotenv.2013.03.005
  4. Communities and Local Government. 2007. Generalised Land Use Database Statistics for England 2005. Department for Communities and Local Government. Product Code 06CSRG04342. February 2007
  5. 5.0 5.1 5.2 5.3 Reimann, C, Filzmoser, P, Garrett, R G, and Dutter, R. 2008. Statistical data analysis explained. Applied environmental statistics with R. John Wiley & Sons Ltd., Chichester, England. https://www.statistik.tuwien.ac.at/StatDA/R-scripts/
  6. StatDA: Statistical Analysis for Environmental Data https://cran.r-project.org/web/packages/StatDA/index.html
  7. Tukey, J W. 1977. Exploratory data analysis. Addison-Wesley, Reading, Massachusetts, USA. 506pp.
  8. Sinclair, A J. 1976. Applications of probability graphs in mineral exploration. Special Volume 4. Association of Exploration Geochemists. Toronto, Canada. 95pp.
  9. 9.0 9.1 9.2 Aitchison, J. 1986. The statistical analysis of compositional data. Monographs on statistics and applied probability. London: Chapman & Hall (Reprinted in 2003 with additional material by The Blackburn Press), 416 pp.
  10. Pearson, K. 1897. Mathematical contributions to the theory of evolution. On a form of spurious correlation which may arise when indices are used in the measurement of organs. Proceedings of the Royal Society of London, Vol. LX, 489–502.
  11. Egozcue, J J, Pawlowsky-Glahn, V, Mateu-Figueras, G, and Barcelo-Vidal, C. 2003. Isometric logratio transformations for compositional data analysis. Mathematical Geology, Vol. 35(3), 279–300.
  12. Reimann, C, Filzmoser, P, Fabian, K, Hron, K, Birke, M, Demetriades, A, Dinelli, E, Ladenberger, A, and the GEMAS Project Team. 2012. The concept of compositional data analysis in practice — total major element concentrations in agricultural and grazing land soils of Europe. Science of the Total Environment, Vol. 426, 196–210
  13. 13.0 13.1 Reimann, C, Birke, M, Demetriades, A, Filzmoser, P, and O'Connor, P (eds.). 2014. Chemistry of Europe's agricultural soils. Part A: methodology and interpretation of the GEMAS dataset. Geologisches Jahrbuch, B 102, 528 pp., 358 figs., 86 Tables, 1 DVD; Hannover, Germany.
  14. Filzmoser, P, Hron, K, and Reimann, C. 2009. Univariate statistical analysis of environmental (compositional) data: problems and possibilities. Science of The Total Environment, Vol. 407, Issue 23, 6100–6108. DOI 10.1016/j.scitotenv.2009.08.008
  15. 15.0 15.1 van den Boogaart, K G, and Tolosana-Delgado, R. 2013. Analyzing compositional data with R. Use R! Springer-Verlag Berlin Heidelberg. DOI 10.1007/978-3-642-36809-7 1
  16. van den Boogaart, K. G. & Tolosana-Delgado, R. (2008). Compositions: a unified R package to analyze compositional data. Computers and Geosciences, 34(4), 320–338.
  17. Fordyce, F M, Brown, S E, Ander, E L, Rawlins, B G, O’Donnell, K E, Lister, T R, Breward, N, and Johnson, C C. 2005. GSUE: urban geochemical mapping in Great Britain. Geochemistry: Exploration, Environment, Analysis. Vol. 5(4), 325-336 (12).
  18. Flight, D M A, and Scheib, A J, 2011. Soil geochemical baselines in UK urban centres: the G-BASE project. In: Johnson, C, Demetriades, A, Locutura, J, and Ottesen, R T (eds.). Mapping the chemical environment of urban areas. Wiley-Blackwell, Oxford, pp. 186-206.
  19. Müller, G, 1979. Schwermetalle in den Sedimenten des Rheins - Veränderungen seit 1971. Umschau 79, 778–783.
  20. Chester, R, and Stoner, J H. 1973. Pb in particulates from the lower atmosphere of the eastern Atlantic. Nature 245, 27–28.
  21. Zoller, W H, Gladney, E S, and Duce, R A. 1974. Atmospheric concentrations and sources of trace metals at the South Pole. Science, Vol. 183(4121), 198–200