CONCENTRATIONS AND POOLS OF HEAVY METALS IN ...miljobarometern.stockholm.se/content/docs/mg/linde.pdfdetermine heavy metal contents, the soils were digested in boiling 7 M HNO 3 (p.a.)
Post on 02-Apr-2021
7 Views
Preview:
Transcript
CONCENTRATIONS AND POOLS OF HEAVY METALS IN URBANSOILS IN STOCKHOLM, SWEDEN
MATS LINDE∗, HELENA BENGTSSON and INGRID ÖBORNDepartment of Soil Sciences, Swedish University of Agricultural Sciences, Uppsala, Sweden
(∗ author for correspondence, e-mail: Mats.Linde@mv.slu.se)
(Received 30 April 2000; accepted 15 December 2000)
Abstract. The concentrations of heavy metals (Cd, Cr, Cu, Hg, Ni, Pb and Zn) and arsenic (As)were surveyed and the metal pools estimated in soils in Stockholm Municipality. The sampling siteswere distributed all over the entire municipality with a higher sampling density in the city centre.Soils were sampled to a maximum depth of 25 to 60 cm. Soil texture, total-C content, electricalconductivity and pH were analysed. Heavy metal concentrations were determined after wet digestionwith boiling 7 M HNO3.
The results showed a wide range in heavy metal concentrations, as well as in other soil prop-erties. The city centre soils constituted a rather homogeneous group whereas outside this area nogeographical zones could be distinguished. These soils were grouped based on present land use, i.e.undisturbed soils, public parks, wasteland (mainly former industrial areas), and roadside soils. Thecity centre and wasteland soils generally had enhanced heavy metal concentrations to at least 30 cmdepth compared to park soils outside the city centre and rural (arable) soils in the region, which wereused to estimate background levels. For example, the mean Hg concentration was 0.9 (max 3.3) mgkg−1 soil at 0–5 cm and 1.0 (max 2.9) at 30 cm depth in the city centre soils, while the backgroundlevel was 0,04 mg kg−1. Corresponding values for Pb were 104 (max 444) and 135 (max 339) mgkg−1, at 0–5 and 30 cm, respectively, while the background level was 17 mg kg−1.
The average soil pools (0–30 cm depth) of Cu, Pb and Zn were 21, 38 and 58 g m−2 respectively,which for Pb was 3–4 times higher and for Cu and Zn 1.5–2 times higher than the background level.The total amount of accumulated metals (down to 30 cm) in the city centre soils (4.5∗10 6 m2 publicgardens and green areas) was estimated at 80, 1.1, 120 and 40 t for Cu, Hg, Pb and Zn, respectively.The study showed (1) that from a metal contamination point of view, more homogeneous soil groupswere obtained based on present land use than on geographic distance to the city centre, (2) theimportance of establishing a background level in order to quantify the degree of contamination, and(3) soil samples has to be taken below the surface layer (and deeper than 30 cm) in order to quantifythe accumulated metal pools in urban soils.
Keywords: accumulation, background levels, cadmium, chromium, contaminated, copper, heavymetals, lead, mercury, nickel, urban soils, zinc
1. Introduction
Soils in urban environments differ in several respects from soils in other environ-ments (Craul, 1985; Hollis, 1991). Urban soils have a highly variable and oftenunknown history as a result of differences in land use, transfer between sites andmixing in connection with excavations, addition of new soil materials etc. Untilnow, systematic quantitative knowledge regarding metal contents in urban soils has
Water, Air, and Soil Pollution: Focus 1: 83–101, 2001.© 2001 Kluwer Academic Publishers. Printed in the Netherlands.
84 M. LINDE, H. BENGTSSON AND I. ÖBORN
been limited (Thornton, 1991). One of the main sources of heavy metals in soils isthe parent material from which they are derived (Ross, 1994), and there is a verywide range in heavy metal concentrations in different types of parent material. Inorder to assess the extent to which soils in the urban environment have been alteredby anthropogenic influence it is important to establish a baseline by which changesmay be assessed (Paterson et al., 1996).
This study is a part of the Swedish Environmental Protection Agency researchprogramme entitled ‘Metals in Urban Areas’ with the overall objective to obtainknowledge about environmental consequences of the metal accumulation in urbanareas. The municipality of Stockholm was used as a test case. Even though Stock-holm lacks heavy metal emitting industries, analysis of aquatic sediments in theStockholm region has shown increasing concentrations of heavy metals with de-creasing distance from the city centre (Sternbeck and Östlund, 2001). The samegeographical pattern has been demonstrated for Cu, Pb and Zn concentrations intopsoils (0–5 cm) in the Stockholm region (Berglund et al., 1994). Hg was notincluded in the latter study. In a similar study in Aberdeen in NE Scotland, acity also lacking heavy industrial emissions, enhanced levels of Cu, Pb and Znwere found along roads and in city parks as compared to rural soils having thesame parent material (Paterson et al., 1996). Culbard et al. (1988) have showna correlation between urbanisation (i.e. high population density) and metal con-centrations (Cu, Pb, Zn) in garden soils (areas with ‘hot spots’ excluded). In thereferred studies, heavy metal concentrations in the surface layer were measuredand discussed related to the potential health risks (Culbard et al., 1988; Berglundet al., 1994). However, if the aim is to assess the potential risks of metal leachingto ground and surface waters, or to estimate the metal accumulation is urban soilsover a longer period of time, metal concentrations in subsurface layers have to beconsidered. Our hypotheses is that heavy metals accumulated in subsurface layersare a considerable part of the anthropogenic soil pool.
The aim of this study was to determine the soil heavy metal (Cd, Cr, Cu, Hg,Ni, Pb and Zn) and arsenic (As) concentrations in Stockholm Municipality, and torelate the concentrations to the distance to the city centre and various types of landuse. The amount of metals per unit surface area and the total accumulated metalpool in the soil in the city centre were calculated, and related to the estimatedannual inputs.
2. Materials and Methods
2.1. SELECTION OF SAMPLING SITES
The sampling sites were selected in order to cover the entire municipality (187km2 of land area) and to provide a satisfactory geographical representation. Themunicipality was divided into three zones based on the distance to the centre of
HEAVY METALS IN URBAN SOILS IN STOCKHOLM 85
Stockholm (defined as the square Sergels torg); 0–3 km, 3–9 km and more than9 km. Sampling sites were distributed among the zones with the aim of ensuringthat each zone had an approximately equal number of sites. This sampling schemeresulted (as intended) in a higher sampling density in the central zone than in theouter parts of the municipality. Within each zone, we also tried to locate samplingsites to get a variation with regard to historical and present land use. The exactlocations of the sampling sites were determined based on practical considerationswith regard to factors such as sampling accessibility. The sampling was carried outin May 1996.
2.2. SAMPLING TECHNIQUE AND PREPARATION
The surface layer (0–5 cm) was sampled with a steel cylinder (� 7 cm) in order toobtain samples with a defined volume from which bulk density could be determ-ined. At most locations, the sampling below 5 cm depth was carried out with asteel corer. At some sites, existing pits were used for sampling. At each site, 5–10subsamples were collected within a maximum area of 10 m2. Maximum samplingdepth varied between 25 and 60 cm depending on what was technically possible.The subsurface soil material was divided into 1–4 layers based on depth and visualinspection of the soil cores at each site. At two locations it was only possible tosample the surface layer (0–5 cm). Since sampling down to 30 cm was possible atmost of the sites, the layers including 30 cm depth (denoted 30 cm) were used forcomparing the subsurface soil material at different sampling sites. The subsampleswere combined to one composite sample for each layer. The samples were air driedat approx. 30 ◦C, aggregates were crushed with a wooden pestle and the fraction >2 mm was removed by using a steel sieve.
2.3. SOIL ANALYSES
Soil texture was determined with a method that combines wet sieving and sed-imentation. To remove carbonates and organic matter, the soil was treated witha few drops of 1 M HCl, boiled with H2O2 and dispersed with sodium hexa-metaphosphate prior to particle size analysis. The sand fraction (0.2–2.0 mm) wasseparated by sieving and the particle size distribution of fractions < 0.2 mm wasdetermined with a hydrometer (van Reeuwijk, 1993). The soil content of totalC was determined on finely ground samples by total elemental analysis (Leco).In soils with a pH ≥ 7, the CaCO3-content was determined by treating a soilsample with hydrochloric acid (6 M) and measure the released amount of CO2
(International Organization of Standardization, 1995). Electrical conductivity (EC)and pH were measured on air-dried samples in a soil deionized water suspension(soil:water, 1:2.5 by volume) (Standardiseringskommissionen i Sverige, 1994). Todetermine heavy metal contents, the soils were digested in boiling 7 M HNO3 (p.a.)(Swedish Standards Institute, 1997). Concentrations of Cr, Cu, Ni, Pb, and Znwere determined by AAS (acetylene air-flame) and Cd by AAS-graphite furnace
86 M. LINDE, H. BENGTSSON AND I. ÖBORN
technique (Perkin Elmer Zeeman 3030, HGA-600 graphite furnace) using additioncalibration. Arsenic was analysed by AAS using the hydride method and mercurywas determined with a cold vapour AAS-technique after reduction with SnCl2.
2.4. STATISTICAL ANALYSES
The box plot command in SYSTAT (1996) was used to present descriptive statist-ics. The data were also analysed statistically by one-way analysis of variance, usingsoil type as a factor (General Linear Model, SYSTAT, 1996). The significance levelwas p < 0.05.
2.5. ESTIMATION OF METAL POOLS
Total metal pools in soil that was not hardened or developed (i.e. built on) werecalculated based on the amount of metal per unit surface area of green ground incentral Stockholm. In the calculations, all soil layers down to 30 cm depth wereincluded. When calculating the amount of metals per unit surface area we used themeasured bulk density for the 0–5 cm layer. For layers below 5 cm the bulk density,ρb, was estimated based on the soil organic matter content and an estimate of soilporosity as follows:
ρb = m∗(1-n)∗(Vom + Vms)−1,
where m = total soil weight, n = porosity, Vom = volume of organic matter and Vms
= volume of mineral matter. Organic matter content by weight (OM) was calculatedwith the following equation: OM = (Ctot-CCO3) ∗ 0,58−1. Porosity (n) was giventhe nominal value of 0.5 if OM was 8% or more and 0.4 if OM was less than 8%.Furthermore Vom = m ∗ OM ∗ 1.2−1 and Vms = m ∗ (100-OM) ∗ 2.65−1 wherethe parameters 1.2 and 2.65 are the density in g cm−3 of solid organic matter andmineral matter, respectively.
The total green area in the Stockholm City centre was estimated based oninformation from Stockholm Municipality.
3. Results and Discussion
3.1. METAL CONCENTRATIONS IN DIFFERENT SOIL TYPES IN THE
STOCKHOLM MUNICIPALITY
Metal concentrations in samples from the 0–5 cm layer and from 30 cm are presen-ted for all sampling sites (Figure 1). The sampled material as a whole showed awide variation with regard to both metal concentrations and other soil properties.This variation is most likely due to our effort to get a representation of varioustypes of land use within the zones, but it also reflects the great heterogeneity inurban soils in Stockholm Muncipality.
HEAVY METALS IN URBAN SOILS IN STOCKHOLM 87
3.2. CLASSIFICATION INTO DIFFERENT SOIL TYPES
The zonal divisions used when selecting sampling sites were found to be of ratherlimited value as a basis for analysing the results. Outside of the immediate citycentre, there was no clear relationship between distance to the centre of Stock-holm, i.e. Sergelstorg, and type of settlement, land use or vegetation. The variablelandscape and a conscious effort to preserve large green areas as the city has ex-panded has resulted in a situation where areas with natural forest and grassland andnewly established buildings occur in fairly central locations at the same time asolder residential areas, parks and industrial areas can be found further out from thecentre. This is probably the reason why there is no systematic relationship betweendistance to the centre of Stockholm and metal content in soils except for the highconcentrations in soils of the city centre (Figure 1).
Sampling sites outside the central zone (city centre) were grouped based onpresent land use, i.e. undisturbed soils (areas with semi-natural vegetation), pub-lic parks (green areas excluding those with semi-natural vegetation), waste land(mainly former industrial areas) and roadside soils (sites situated within a fewmetres from the roadway). Sites in the city centre were treated as a single group.The city centre represents the central parts of Stockholm which are nowadaysdominated by office buildings and stores, etc. All of the sampling sites in this groupwere situated in public parks within areas with lawns or other types of vegetation.The group could be expected to include the areas that have been exposed to urbanconditions – and thus to metal loads typical of urban environments – for the longestperiod of time.
3.3. PROPERTIES AND METAL CONCENTRATIONS OF THE VARIOUS SOIL
TYPES
Metal concentrations, pH, EC, total C and clay content of samples from the 0–5cm layer and from 30 cm are given for each soil type (Table I). For comparison,corresponding data for arable soils (soils with > 40 % clay excluded) from theStockholm region, i.e. the Stockholm, Uppsala and Södermanland counties areincluded (Table I). These data were calculated from the database originating froman arable soil survey, which was presented by Eriksson et al. (1997; 1999). Metaldata for the different soil types are also presented in Figure 2.
Concentrations of Cd, Cu, Hg, Pb and Zn in the city centre soils were higherthan the average for arable soils in the region (Table I). The mean for Hg (0.86 mgkg−1, 0-5 cm; 1.00 mg kg−1, 30 cm) was about 20 times larger than the mean forthe arable soils (0.04 mg kg−1, 0–20 cm). Comparisons between the different soiltypes (all horizons included) showed that the city centre soils had significantly (p< 0.05) enhanced levels of Cd, Hg, Pb and Zn compared to the undisturbed soilsand the park soils (only Hg and Pb). In the surface samples (0–5 cm) no significantdifferences could be observed between these soil groups, which indicated that theenhanced metal concentrations observed at 30 cm could be due to historical rather
88 M. LINDE, H. BENGTSSON AND I. ÖBORN
Figure 1. Sampling sites (a) and concentrations (mg kg−1) of heavy metals in soil at the 0–5 cm layerand the layer including 30 cm depth (30 cm) extracted with 7 M HNO3; As (b), Cd (c), Cr (d), Cu(e), Hg (f), Ni (g), Pb (h), Zn (i). In (a) capital letters at sampling sites refer to different soil types; C= City centre, P = park, R = road side, U = undisturbed, W = waste land. In (b)-(j) the concentrationat each sampling site is presented as bars and figures. The left bar shows the concentration at 0–5 cmand the right bar the concentration at 30 cm.
HEAVY METALS IN URBAN SOILS IN STOCKHOLM 89
Figure 1. Continued.
90 M. LINDE, H. BENGTSSON AND I. ÖBORN
Figure 1. Continued.
HEAVY METALS IN URBAN SOILS IN STOCKHOLM 91
TAB
LE
I
Con
cent
rati
ons
(mg
kg−1
dry
wei
ght)
ofA
s,C
d,C
r,C
u,H
g,N
i,P
ban
dZ
nin
soil
extr
acte
dw
ith
HN
O3
(7M
),pH
,ele
ctri
cco
nduc
tivit
y(E
C),
tota
lCan
dcl
ay(<
2µ
m)
cont
ent(
byw
eigh
t)
Soi
ltyp
e/D
epth
EC
CC
lay
As
Cd
Cr
Cu
Hg
Ni
Pb
Zn
pHµ
Scm
−1%
%m
gkg
−1
Cit
yce
ntre
soil
s,0–
5cm
mea
n(n
=14
)6.
332
85.
922
5.4
0.43
2747
0.86
9.0
104
157
std
0.5
134
2.0
113.
30.
206.
925
0.96
2.3
110
96
min
4.8
128
3.1
112.
00.
1918
120.
105.
614
34
max
6.9
568
9.8
4510
.20.
7939
105
3.3
14.4
444
408
30cm
mea
n(n
=14
)6.
722
63.
326
5.4
0.33
2767
1.00
10.6
135
193
std
0.6
116
1.7
45.
40.
148
480.
924.
812
213
1
min
5.9
580.
510
1.8
0.10
1414
0.03
3.5
1637
max
8.0
464
6.1
4423
.40.
5842
153
2.9
21.1
339
527
Und
istu
rbed
soil
s,0–
5cm
mea
n(n
=7)
5.7
171
11.4
nd6.
80.
2749
230.
1214
.445
76
std
1.1
4913
.96.
10.
1043
100.
155.
938
33
min
3.9
107
2.1
2.8
0.14
237.
30.
036.
114
20
max
6.6
227
41.7
20.0
0.42
145
350.
4421
.312
210
9
30cm
mea
n(n
=7)
6.2
170
3.0
nd3.
70.
1635
230.
0416
.019
80
std
1.1
105
2.9
2.2
0.09
1511
0.03
8.3
9.1
32
min
4.3
200.
61.
00.
026
2.5
0.01
2.1
2.4
18
max
7.3
333
9.4
8.0
0.25
5237
0.08
27.8
3211
2
92 M. LINDE, H. BENGTSSON AND I. ÖBORN
TAB
LE
I
Con
tinu
ed
Soi
ltyp
e/D
epth
EC
CC
lay
As
Cd
Cr
Cu
Hg
Ni
Pb
Zn
pHµ
Scm
−1%
%m
gkg
−1
Park
soil
s,0–
5cm
mea
n(n
=8)
6.3
327
7.2
485.
10.
3535
300.
4617
.430
144
std
0.6
142
3.3
2.3
0.27
1611
0.61
7.2
1314
7
min
5.3
824.
91.
90.
1210
150.
037.
413
35
max
7.1
457
14.4
8.9
0.94
6645
1.46
31.0
4950
2
30cm
mea
n(n
=7)
6.9
188
2.4
nd4.
30.
1640
300.
1021
.725
101
std
1.0
104
0.7
0.7
0.05
6.4
3.8
0.06
5.1
6.6
23
min
5.2
611.
43.
30.
1129
260.
0213
.218
53
max
8.0
304
3.3
5.6
0.23
4836
0.18
26.2
3512
6
Was
tela
ndso
ils,
0–5
cm
mea
n(n
=6)
7.0
296
6.8
2812
.30.
5444
287
1.22
14.9
254
407
std
0.8
124
7.8
612
.70.
2937
512
2.60
4.0
503
317
min
5.5
109
0.7
243.
80.
2816
260.
028.
213
106
max
7.9
476
22.1
3237
.31.
1111
913
156.
5219
.312
7996
5
30cm
mea
n(n
=5)
7.1
328
3.0
4226
.20.
3336
142
0.16
18.1
155
134
std
1.2
187
2.3
2746
.40.
3814
219
0.11
7.2
253
46
min
5.0
183
0.7
112.
90.
0820
230.
0612
.521
96
max
8.0
647
6.8
6010
91.
0050
533
0.35
30.0
606
212
HEAVY METALS IN URBAN SOILS IN STOCKHOLM 93
TAB
LE
I
Con
tinu
ed
Soi
ltyp
e/D
epth
EC
CC
lay
As
Cd
Cr
Cu
Hg
Ni
Pb
Zn
pHµ
Scm
−1%
%m
gkg
−1
Roa
dsid
eso
ils,
0–5
cm
mea
n(n
=7)
6.8
234
3.4
182.
50.
3725
270.
1411
.410
012
6
std
0.4
132
1.0
60.
70.
345.
816
0.16
2.6
150
89
min
6.4
119
1.8
141.
50.
0616
100.
017.
94.
632
max
7.6
504
4.7
223.
30.
9931
530.
4815
.142
424
9
30cm
mea
n(n
=7)
7.5
258
2.0
223.
40.
3436
280.
0818
.726
146
std
0.5
500.
89
0.3
0.21
4.6
3.5
0.07
3.7
1872
min
6.5
184
1.1
162.
80.
1431
230.
0312
.311
83
max
8.1
330
2.9
293.
70.
6643
320.
2123
.564
271
All
sam
ples
,0–5
cm
mea
n(n
=42
)6.
428
16.
824
b6.
10.
4034
710.
5912
.810
117
1
std
0.8
133
6.7
126.
20.
2524
201
1.16
5.4
208
174
min
3.9
820.
511
1.5
0.06
107.
30.
015.
64.
620
max
7.9
568
4248
37.3
1.11
145
1315
6.52
31.0
1279
965
30cm
mea
n(n
=40
)6.
922
82.
829
c7.
20.
2733
550.
4115
.879
142
std
0.9
119
1.8
1516
.90.
1911
850.
066.
912
394
min
4.3
200.
511
1.0
0.02
6.3
2.5
0.01
2.1
2.4
18
max
8.1
647
9.4
6010
91.
0052
533
2.89
30.0
606
527
94 M. LINDE, H. BENGTSSON AND I. ÖBORN
TAB
LE
I
Con
tinu
ed
Soi
ltyp
e/D
epth
EC
CC
lay
As
Cd
Cr
Cu
Hg
Ni
Pb
Zn
pHµ
Scm
−1%
%m
gkg
−1
Ara
ble
soil
saSt
ockh
olm
regi
on
mea
n(n
=22
6)6.
32.
628
3.39
0.24
2918
.80.
0415
.317
.472
.0
std
0.5
2.6
9.2
1.67
0.09
8.3
8.1
0.05
5.5
6.8
23.7
min
4.9
1.1
2.5
0.05
0.05
5.2
2.2
0.01
3.0
4.5
11.6
max
7.9
20.7
3911
.55
0.82
4760
.50.
5930
.974
.713
9
aT
hepl
ough
laye
r(0
–20
cm)
(Eri
ksso
net
al.,
1997
;199
9),b
n=
14,c
n=
13.
HEAVY METALS IN URBAN SOILS IN STOCKHOLM 95
Figure 2. Concentrations (mg kg−1) of As (a,b), Cd (c,d), Cr (e,f), Cu (g,h), Hg (i,j), Ni (k,l), Pb (m,n)and Zn (o,p) (7 M HNO3) at 0–5 and the horizon including 30 cm depth (30 cm) of the different soiltypes; A = all soils (n = 42(40 at 30 cm)), C = city centre (n = 14), P = park (n = 8(7)), R = road side(n = 7), U = undisturbed (n = 7) and W = waste land (n = 6(5)). The centre horizontal line marks themedian. The box edges and lines represent quartiles. Outside and far outside values are plotted with∗ and ◦ (SYSTAT, 1996).
than more recent loads. The concentrations of Cr and Ni were smaller (althoughnot significantly) in the city centre soil than in any other soil group reflecting thefact that Stockholm has never been a heavily industrialised city. The reason for thesmaller concentrations is most probably that the concentration of these metals isstrongly correlated with the clay content (Andersson, 1977) and most city centresoils are built up from coarser soil material than the rural (arable) soils used asreference level.
The roadside soils outside the city centre did not differ (p < 0.05) in metal levelsfrom the undisturbed soils or park soils. This finding was unexpected and might bedue to a mixture of older and more recently established roads as well as roads withdifferent traffic intensities. The Pb concentration in the roadside soils (0–5 cm)
96 M. LINDE, H. BENGTSSON AND I. ÖBORN
Figure 2. Continued.
varied from 5 to 424 mg kg−1 (mean 100) which can be compared with 14 to 444(mean 104) in the city centre (0–5 cm) or 5 to 75 (mean 18) in the arable soils(Table I).
Waste land soils were, as expected, the group diverging most from the othersoil types with regard to metal contents. The soil samples in this group showedstrongly elevated metal levels and the mean concentrations for all metals studiedwere greater than the average values of arable soils in the region (Table I). The As,Cd, Cu, Hg, Pb, and Zn, concentrations were higher (p < 0.05) in the waste landsoils than in the undisturbed, park land (except Hg) and roadside (except Cd) soils(Table I). A statistical comparison between the waste land soils and the city centresoils showed that the former soil group had greater concentrations of As and Cu,whereas the Cd, Hg, Pb, and Zn levels did not differ significantly (p < 0.05).
HEAVY METALS IN URBAN SOILS IN STOCKHOLM 97
Figure 2. Continued.
3.4. SOIL METAL POOL IN THE CENTRE
The City centre sites represent the group exposed to urban conditions for the longestperiod of time. Soils at these sampling sites seem to be clearly affected by past con-ditions in the urban environment. This group is also relatively uniform regardingvegetation and land use. In terms of land area it is the dominant type of green,non-hardened ground in central Stockholm. Data from these sampling sites cantherefore be used to calculate the amount of metals that has accumulated due tourban activities in the soils in central Stockholm. It is worth noting, however, thatsuch a calculation does not take into account the large amount of various metals thathas accumulated in extremely contaminated sites, i.e. former industrial grounds, orin land which today is covered by roads, buildings etc. The individual amounts ofthe various metals per m2 down to 5 or 30 cm depth, in soil in the city centre werecalculated (Table II). Corresponding values for park soils and arable soils in theregion are also given. Since the city centre soils were all sampled in parks, the soil
98 M. LINDE, H. BENGTSSON AND I. ÖBORN
Figure 2. Continued.
properties and management practices could be assumed to be similar for the twosoil groups which enables a direct comparison.
The calculations showed that the quantity of Hg in the upper 30 cm soil layerin the city centre (0.29 g per m2) was 6 times larger than in the park soils (0.05 gper m2) (Table II). The soil pool (per m2) of Pb was 3 times larger in the city centrethan in the park soils and the pools of Cd, Cu and Zn were 1.5–2 times higher. Thesoil pools (0–5 cm) of all metals, except Hg, were at the same level in the parksoils as in the arable soils. This indicates that the metal levels in the park soils canbe used as a reference value in comparison with the city centre soils. Compared tometal concentrations found in the geological parent material, the metal contents inthe arable soils as well as the park soils can be expected to be enhanced throughdeposition of long-distance transported metals and supply of soil conditioners andchemical fertilisers (Eriksson et al., 1997).
The differences between the metal contents in the park soils and the city centresoils were calculated to provide a rough estimate of the amounts of metals accumu-
HEAVY METALS IN URBAN SOILS IN STOCKHOLM 99
TABLE II
Calculated average pool (g m−2) of As, Cd, Cr, Cu, Hg, Ni, Pb, and Zn down to 5 cmdepth in arable soil (δb = 1.3 g cm−1) (Eriksson et al., 1997; Eriksson et al., 1999) anddown to 5 and 30 cm depth in public gardens and green areas in (City centre soil) andoutside (Park soil) central Stockholm. The total pool of the metals (kg) down to 30 cm inpublic gardens and green areas in central Stockholm (4.5∗106 m2) has been estimated
Metal Arable soil Park soil City centre soil
Average pool Average pool Average pool Total pool
g m−2 g m−2 g m−2 103 kg
0–5 cm 0–5 cma 0–30 cmb 0–5 cmc 0–30 cmc 0–30 cm
As 0.20 0.20 1.55 0.21 1.65 7.4
Cd 0.02 0.01 0.08 0.02 0.13 0.6
Cr 1.85 1.45 13.4 1.11 9.50 42.8
Cu 1.22 1.18 11.9 1.91 20.7 93,2
Hg 0.003 0.02 0.05 0.03 0.29 1.3
Ni 0.99 0.72 6.90 0.36 3.54 15.9
Pb 1.16 1.19 11.9 4.15 38.3 172
Zn 4.68 6.43 39.5 6.29 57.9 260
a n = 8, b n = 7, c n = 14.
lated in the city centre soils due to local emissions. These figures on anthropogenicinput to the soils were multiplied with the estimated total area of public gardens andgreen areas within the city centre (4.5∗106 m2; pers com Fritzon) in order to get arough quantitative estimate of the metal accumulation. These calculations showedthat 1.1 t of Hg have accumulated in city centre soils due to local urban activities.The total accumulated Pb was about 120 t and the corresponding quantities of Cuand Zn were 40 and 80 t respectively.
Sörme et al. (2001) and Bergbäck (1998) have estimated the annual emissionsof Cu and Zn to the Stockholm soil environment to be 4 and 15 t, respectively.This means that the total accumulated quantities in the city centre soils wouldbe the result of 10 yr accumulation for Cu and 6 yr for Zn, which seems ratherunlikely. Although, the metals might spread over a larger area within the StockholmMunicipality, this indicates that the Cu and Zn estimated to be released from thetechnosphere only to a smaller extent ends up in the city centre soils.
4. Conclusions
The soils in Stockholm Muncipality showed a wide range in heavy metal con-centrations. The city centre soils constituted a rather homogeneous group whereasoutside this area no distinct geographical zones could be distinguished, and we
100 M. LINDE, H. BENGTSSON AND I. ÖBORN
found it more useful to group the soils based on land use, i.e. undisturbed soils,public parks, waste land and roadside soils.
In order to quantify heavy metal accumulation in urban soils it is of vital im-portance to establish the background levels originating from the parent material.Rural (arable) soils in the region as well as park soils outside the city centre werefound to be useful. The enhancement was highest for Hg, the concentrations ofwhich were at least 20 times higher in the city centre and waste land soils than thebackground level. The contents of Pb and Zn were also enhanced in the two soiltypes, as were As, Cd and Cu in the wasteland soils.
Both the city centre and wasteland soils were contaminated to at least 30 cmdepth. This showed the importance of deep sampling, most desirable to layers hav-ing background concentrations, in order to quantify the anthropogenic soil pool.This would also be of great importance in risk assessments of potential metalmobility to the ground water, or related to changes in land use or excavations.The quantity of metals accumulated down to 30 cm in the city centre soils werecalculated at 40, 1.1, 120 and 80 t for Cu, Hg, Pb and Zn, respectively.
Acknowledgements
We would like to thank Gunilla Lundberg for carefully analysing the heavy metals,Per Ola Fritzon for data about green areas in Stockholm and Magnus Åsman forhelp with constructing the maps. The Swedish Environmental Protection Agencyfunded the study.
References
Andersson, A.: 1977, Heavy Metals in Swedish Soils: On their Retention, Distribution and Amounts,Swedish J. Agric. Res. 7, 7–20.
Berglund, M., Fahlgren, L., Freland, M. and Vahter, M.: 1994, Metaller i mark i Stockholms inner-stad och kranskommuner – förekomst and hälsorisker för barn. IMM-rapport 2/94 Karolinskainstitutet, Stockholm, p. 48 (in Swedish).
Bergbäck, B.: 1998, Metaller i Stockholm. Kunskapssammanställningar av metallflöden via olikaverksamheter i Stockholm, Naturvårdsverket, Rapport 4952 (in Swedish).
Craul, P. J.: 1985, A Description of Urban Soils and heir desired characteristic, Journal ofArboriculture 11(11), 330–339.
Culbart, E. B., Thorton I., Watt J., Wheatley, S., Moorcroft, S. and Thompson, M.: 1988, Metalcontamination in British Dusts and Soils, J. Environ. Qual. 17, 226–234.
Eriksson, J., Andersson, A. and Andersson, R.: 1997, Current status of Swedish Arable Soil,,Naturvårdsverket, Rapport 4778, p. 59 (in Swedish with English Summary).
Eriksson, J., Andersson, A. and Andersson, R.: 1999, Texture of Agricultural Topsoils in Sweden,Naturvårdsverket, Rapport 4955, p. 26 (in Swedish with English Summary).
International Organization of Standardization: 1995, Soil quality – Determination of carbonatecontent – Volumetric method. International Organization of Standardization ISO 10 693.
HEAVY METALS IN URBAN SOILS IN STOCKHOLM 101
Hollis, J. M.: 1991, The classification of soils in urban areas, in P. Bullock and P. J. Gregory (eds),Soils in the Urban Environment, Blackwell Scientific Publications, Oxford, p. 1–4.
Paterson, E., Sanka, M. and Clark, L.: 1996, Urban soils as pollutant sinks – a case study fromAberdeen, Scotland, Applied Geochemistry 11, 129–131.
Ross, S.: 1994, Sources and Forms of Potentially Toxic Metals in Soil-Plant Systems, in Ross, S.(ed.), Toxic Metals in Soil-Plant Systems, John Wiley and Sons Ltd, Chichester, p. 3–25.
Swedish Standards Institute: 1997, Soil analysis – Determination of trace elements in soils –Extraction with nitric acid. Swedish Standards Institute SS 02 83 11 (in Swedish).
Standardiseringskommissionen i Sverige: 1994, Markundersökningar - Bestämning av pH. Stand-ardiseringskommissionen i Sverige SS-ISO 10390 (in Swedish).
Sternbeck, J. and Östlund P.: 2001, Metals in sediments from the Stockholm region: Geographicalpollution patterns and time trends, Water, Air and Soil Pollution: Focus 1(3–4), 151–165.
SYSTAT for Windows: 1996, Statistics, Version 6.0.1 (ed.) Evanstone, IL: SYSTAT, Inc.Sörme, L., Berbäck B. and Lohm U.: 2001, Goods in the anthroposphere as a metal emission source-a
case study of Stockholm, Sweden, Water, Air and Soil Pollution 1(3–4), 213–228.Thornton, I.: 1991, Metal contamination of soils in urban areas, in P. Bullock and P. J. Gregory (eds),
Soils in the Urban Environment, Blackwell Scientific Publications, Oxford, p. 47–75.Van Reeuwijk, L. P. (ed.): 1993, Procedures of Soil Analysis. 4th (ed.) International Soil Reference
and Information Centre, Technical paper no. 9.
Personal Communication
Fritzon, P.O.: 1998. Stockholm City Council, Box 8311, 10420 Stockholm,tfn. +46(0)8 – 508 261 13.
top related