Document 8R7Kv7Y80ZvM7XyjrMzJXrjvo
a.
tyv*
1:
Pf p t? > 3
Ecology of Heavy Metals -- a Regional and Historical Study
By Ake Riililing and Germund Tyler
Department of Plant Ecology, University of Lund, Sweden
C4j>
ABSTRACT
Samples of Hypiuim cupressiformc, collected during 1968--69 in north eastern Gotaland, the low-land areas and the ridge areas of Skane (southern Sweden) have been analysed for Pb, Zn, Cu, Ni, and Cr. The concentration of these metals were significantly lower in the north-eastern area than in the areas of Skane, particularly in Pb, Ni, and Cr. Between the two areas in SUune differences were only established for Pb and Cu, with the largest concentra tions in the humid ridge areas. The differences in the concentrations of these heavy metals between north-eastern Gotaland and Skane will be explained by the geographical position with respect to the large industrial regions of Europe.
Samples of the same species, collected in Skane 1870--1943 hare also been analysed. Significant increases during the decades around 1900 were demon strated for Cu and Zn. In the concentrations of Ni a rapid rise has occurred since about 1920, when this inetal was introduced in the world production. The content of Ni in the samples from 1969 is more than twice as large as in the samples collected before 1920. The historical trends in the concentra tions of Pb have earlier been demonstrated (RUh l in g & Ty l e r 1968 a). There is little doubt that the rise in the concentrations of these metals is an effect of a larger air-borne supply, originating front human activity.
INTRODUCTION
The present paper is a continuation of a previous study (Ru h l in g & Ty l e r 1968 a, in the following abbreviated RT). In Ibis study a distinct regional decrease towards the north-east in southern and central Swe den was established as lo lead concentration of three different mosses. Particularly large concentrations were measured in the mosl humid parts of south-western Gotaland, indicating lliat a considerable part of the lead which is brought down by the rain, al least in Ibis area, originates from areas oulside Sweden.
The aims of the present paper is lo measure the local, regional and
Bot. Kotiser, vol. 122,1909
N36942
ECOLOGY OF HEAVY METALS
249
historical differences in the concentrations of certain heavy metals, as they are reflected in a common moss species.
MATERIALS AND METHODS
As the material of this study the moss Hgpnam cupressiforme was chosen. In 21 localities, situated more than 300 m from roads along a S\V--NE transect across Skanc (cf. Fig. 1), three separate samples were collected on each locality in February 1969 and analysed. For comparison samples from Ostergotland and adjacent areas, as well as from Denmark (Sjtelland) and the very humid south-western slopes of Sydsvenska Hoglandet, collected in 1968 and previously analysed for lead (cf. RT) were analysed for other heavy metals (Zn, Cu, Ni, Cr, and Co). The same elements were also determined in a historical material from Skane, collected in 1870--1943. The analyses were performed by atomic absorption spectrophotometry (cf. RT). The metal con centrations of the samples are always calculated as ppm dry matter. Statistical calculations have been performed according to Sx e d e c o r (1961).
THE TRANSECT ACROSS SKlNE
The position of the 21 localities along the transect line is illustrated in Figs. 1 and 2, cf. also Table 1. The transect is drawn from the low-land plain J 7 Bot. Notiser, vot. 122,19G9
a,-
M #
DUP050314056
.I-
250
Ak e k Oh l in g a n ij g e r mu n d t y l e r
I- y :.i
_ Fig. 2. The transect across Sk&ne. Mean annual precipitation, concentration of Cu and Pb in Hypnum cupressiforme and elevation above sea-level of the sampling points (1--21).
E Mahno (localities 1--3), across the ridge of RomeleSsen (4--5), the plain of Vombsankan (6--9), the ridge of Linderodsasen (10--15) and the plain of Kristianstadslatten (16--21). Consequently, widely different soils are repre sented, and the humidity is considerably higher on the ridges than in the low-land areas. The bedrock of the ridges (localities 4--5, 10--15), is pre dominantly igneous Archaean gneiss and the soils are more or less podzolised Archaean moraines poor in calcium. The mean annual precipitation is di stinctly higher than on the adjacent plains. The plains are developed on
Bot. Notiser, vol. 122.1009
DU P050314057
ECOLOGY OF HEAVY METALS
251
Tabic 1. The concentration of heavy metals in Hypnum cuprcssiformc from tire southwest--northeastern transect across SkSne. Median values (n = 3), calculated
as ppm dry matter.
-- -------Locality
^ ""
Element Pb ------ ---
Cu | Zn
Ni
Cr Co
1. Torup, 4.5 km S Bara cli................... 2. Ekholmssjon, 4.5 km. SE Hyby ch. 3. Hackeberga, 2 km SSE Geiutrp----4. 300 m SW Romeleklint, 3.5 km SW
Veberod ............................................... 5. 0.5 km SSW HusagArd, 3 km SSW
Veberod ............................................... 6. Skogmollnn, 3.5 km SE Veberod . .. 7. 1 km ESE Vomb cli............................ 8. 1.3 km NW S Asum ch.....................
9. 0.5 km W Vollsjo ch..........................
10. 1.5 km N Franninge ch.....................
11. 2.5 km SSE Langarod ch................. 12. 3.0 km SW Huarod ch..................... 13. 2.5 km NW Huarod ch..................... 14. 2.2 km SW 0. Sonnarstov ch.......... 15. 1.1 km SW 0. Sonnarstov ch.......... 1G. 2 km SW Everod ch* ..................... 17. 2.5 km E Everod ch. . ...................... 18. 4 km NNE Vittskovle ch................... 19. Horna, 3 km NNW Ahus ................. 20. Fjalkinge backe,2 km N Fjiilkinge ch.
21. 2.5 km NNE Valje............................
68 120 67
112
120 107 116 116 83 112 136 117 156 151 103
85 76 57 92 95 100
12.7 99 19.2 126 9.8 102
21.2 149
17.0 112 13.5 135 12.5 93 15.0 102 11.8 115 16.3 107 16.6 148 13.9 98 15.8 101 16.4 100 13.7 100
9.8 93 10.1 88 11.2 90 11.7 99 11.4 108 12.5 119
11.3 6.5 13.3 9.5 8.6 5.5
12.5 9.7
8.8 7.5 10.2 7.5
7.8 5.8 8.6 9.8 9.2 14.0 7.6 6.5 7.7 5.6 9.8 8.0 9.4 7.7 11.8 6.2 8.3 8.2 8.6 8.6 6.5 5.5 9.9 8.9 7.4 7.5 6.8 8.1 6.7 5.3
1.5 1.3 1.7
1.2
1.3 2.7 1.3 2.5 3.0 1.6 1.5 2.8 1.9 1.4 1.6 1.2 1.0 2.4 1.3 1.0 1.3
Cretaceous rocks, chiefly limestone (localities 1--3, 6--8, 16--19), Silurian slate (9) or Archaean rocks (20--21). The low-land soils are variable, in the south-west (localities 1--3) rather clayey moraines but otherwise mostly glacio-fluvial sands, on Kristianstadslatten partly very calcareous.
The mean annual precipitation for the period 1962--66, calculated from SMHI (1964--1968) is between 700--800 mm on the ridges, 600--700 mm on the south-western plain and in Vombsankan, and 500--550 mm on Kristianstadslatten (cf. Fig. 2, where the altitude of the localities is also given).
The results of the analyses are compiled in Table 1 as the medians from each locality. Any significant differences between the ridges and the low-land areas have only been established with Pb and Cu. The lead concentration ranges between 103--156 ppm (mean 124) on the ridges, and 57--100 ppm (mean 87) on Kristianstadslatten, correspond ing to the precipitation of 725--775 mm and 500--590 mm, respec tively. The values for the south-western plain and Vombsankan are intermediate both in lead and iii mean annual precipitation. A quite similar trend may be demonstrated for Cu (cf. Fig. 2). The copper concentration on the ridges averages 16.3 ppm, on the plains 12.4 ppm.
Bot. Notiscr, vol. 122,1909
;,
V
DUPO 50314058
4r -.A; 252 AKE KCh LING AND GERMUN'D TYLER
'$'
i.'y?'
Fig. 3. Correlation between Pb and Cu in the samples of the transect across Skane. Open circles=low-land areas, filled circles--ridge areas.
The correlation between Pb and Cu is evident from Fig. 2 and illustrated in Fig. 3. The coefficients of correlation between the concentrations of the metals have been compiled in the following table.
Pb Zn Cu Ni
Zn.............
__ _,, __
Cu.............
0.56 __ __
Ni ............ .......... 0.29 0.33 0.61
__
Cr............ ........... 0.29 0.16 0.24 0.45
Rather good positive correlations were obtained between Cu and Ni, Cu and Zn, as well as between Cu and Pb.
Bot. Kotiser, vol. 122. JDG9
j ( t.
\
*
3?' "-is?'
DUPO 50314059
*
4
4-
?
V-
!
i
)
'f
ECOLOGY OF HEAVY METALS
I
253
I
I Fig. 4. The distribution of metal concentrations (ppm dry matter) in samples from I 1968--69. I=Ostergotland and adjacent areas, lI=Sk&ne, low-land areas, III --Skine, ! ridge areas. Means and limits of means, probability 95 /o, are indicated.
Bot. Notiser, vol, 122,1969
!
IT
:
4 *
DUP050314060
I
# "
i
254
Ak e r Uh l in g a n d g e r mu n d t y l e r
Table 2. Slatislical calculations on the regional differences in 1968--69. Means (ppm) and significance of the differences between the means. Area I=Ostergotland and adjacent areas, Il=Sk&ne, low-land areas, 111= SkSne, ridge areas. Level of
significance: =p < 0.05, **=p < 0.01, ***=p < 0.001.
Area
Means, ppm i 1 II III
l-vnlues and levels of significance
I--II
I--HI
ii--m
Pb .......... Cu .......... Zn .......... Ni ...... Cr ..........
49 93 124 12.4 12.4 16.3 95 105 114 6.1 8.8 9.6 4.5 7.9 7.9
4*4 8.49
0.00 4 2.21 *4* 4.33
4.74
444 10.4
3.94 * 2.64 444 5.40 444 5.30
444 4.10 444 5.36
1.38
0.17
0.00
REGIONAL COMPARISONS
23 samples of Hypmtm cupressiforme from different parts of Oster-
gotland and adjacent areas have also been, analysed for heavy metals.
This group is designated I in Fig. 4. For comparison the samples from
the transect across Skane have been divided into two groups, one group >> (II) comprising the low-land plains (localities 1--3, 6--9, 16--21) and
one group (III) comprising the ridge areas (localities 4--5, 10--15).
Statistical calculations have been compiled in Table 2.
Significant differences (***) between the means of lead were re
corded for all areas, with the lowest mean (49 ppm) from the north eastern area, the highest mean from the humid ridge areas in Skane
F:
(124 ppm) and with the mean of the low-land areas of Skane in an intermediate position (93 ppm). Also for Ni and Cr the means of the north-eastern area are significantly lower than the corresponding means of the two areas in Skane but no differences exist between the latter.
f,
fi
The copper means of the north-eastern area and the low-land area are
identical (12.4 ppm), whereas the mean of the ridge area is significantly
higher (16.3 ppm). The observed differences in the means of Zn are
not fully secured.
In general the concentrations of these melals are highest on the
ridges in Skane, lowest in the north-eastern, area. In this area Zn takes
the largest share of the melals analysed, but on the humid ridges Zn
is surpassed by Pb. As a comparison, a few samples from Denmark
(Sjaelland) and Halland have also been analysed. The samples from
Sjaelland are higher in Pb and Cu than the samples from the corre-
Not. Kotiser, vol. 122,1000
r.
5'
r'W.' -f
DU P050314061
ECOLOGY OF HEAVY METALS
255
Table 3. Statistical calculations on the historical changes. Means (ppm) and significance of the differences between the means (cf. Table 2).
Means, ppm 1870/99 1900/19 1920/43
1969
t-values and levels of significance
1870/99-- 1900/19-- 1920/-J3-- 1870/99--
1900/19 1920/13
1969
1969
Cu 10.2 Zn 66 Ni 4.0 Cr 5.8
13.0 13.5 13.9 94 103 109
3.7 5.5 9.1 5.5 6.5 7.7
1.73 * 2.23
0.55
0.27
* 0.30 0.30 2.74
*** 0.70 0.71 5.49
**8 $$$ 4.12 10.2
*** 10.7
1.15 1.36 1.74
spending low-land areas in Skane, but no differences were measured in Zn, Ni and Cr. The four samples from the very humid Halland area are very high in Pb (mean 164 ppm), rather high in Cu (15.7 ppm), but the concentrations of Zn, Ni and Cr are similar to the north eastern area.
MEASUREMENTS OF HISTORICAL CHANGES
In previous papers (Ru h l in g & Ty l e r 1968 a, 1968 b, 1969) changes in the lead concentrations of mosses during the last 100 years have been recorded and discussed. In order to establish if any changes in other heavy metals have occurred during this period, we have also analysed samples of Hypnum cupressiforme, collected in Skane 1870-- 1943. This historical material has arbitrarily been divided into three groups: 1870--99 (n=15), 1900--19 (n=15) and 1920--43 (n=13). As a fourth group the figures in Table 1 from the transect across Skane in 1969 (n=21) have been used. The means and the significance of the differences between the means of the periods have been compiled in Table 3.
The copper concentrations increase between the periods 1870--99 and 1900--19, whereas no changes have been measured in the 20th century (Fig. 5). A large increase in the concentrations of Zn has occurred since the 19lh century, chiefly confined to the decades around 1900. For Ni and Cr any corresponding change between the first and the second period has not been measured. Between 1900--19 and 1920-- 43 there is a rise in the concentrations of Ni and the figures from 1969 indicate that this rise has continued and accelerated. In the case of Cr the apparent increase of the means and the medians is not statistic ally significant, partly owing to two aberrant values in period 1870--99
Bot. Kotfser, \ol. 122,19G9
`4>: X
"'V'
DUP050314062
256
AKE RtfHLING AND GERMUND TYLER
Fig. 5. Changes in the concentrations of Zn, Cu, Ni, and Cr in samples from Skane during the last century. The entire curve connects the means and the broken curve the limits of the means, probability ho /o. The position of the medians are indicated
by the black symbols, the number of samples in each class by the height of the columns.
and 1969, respectively. But half of the values from 1870--99 are below the lowest value from 1969 (cf. Fig. 5).
DISCUSSION As distinguished from vascular plants, mosses have little or no pos
sibilities to utilize minerals directly from the ground. The minerals must be collected either through a concentration of the minerals in the precipitation or supplied as dust (cf., e.g., Ta mm 1953, Sv en s s o n & Lid e n 1965). Through the action of mineral or organic acids the elementary constituents of the dust particles may be released and absorbed by the moss carpet. The share of metals supplied as dust or as dry deposition will be larger in the vicinity of the source. The in direct influence of the substrate cannot be neglected, because the main
Bot. Nollser, vot. 122, 1209
?
V
DUP05Q314063
ECOLOGY OF HEAVY METALS
257
part of the dust originates from the close vicinity. However, the local mineral substrate seems to be less important as a source of the metals included in this study than the supply from the atmosphere. The deposition of dust particles, originating from eolic erosion, must be larger in the low-land areas than on Che ridges in Skftne. If this deposi tion were of any importance to the supply and accumulation of these minor metals by the mosses, the low-land areas would exhibit the largest concentrations, but this is not the case with any of the metals.
The recent acidification of the precipitation (cf. EaiKSSON & Od en in Miljovardsforskning 1967, Od en 1968} could theoretically have in creased the weathering and the release of minerals from the dust par ticles, deposited in the moss carpets. But as no change in the concentra tions of Cu and Zn lias been demonstrated during the 20th century, the large recent increases in Pb and Ni cannot be due to increased weathe ring. A substantial increment in the supply of Ni and Pb from the atmosphere, originating from human activities must be the obvious reason.
The soils of north-eastern Gotaland and the ridge areas of Slcane are both predominantly Archaean moraines. Nevertheless, the concentra tions of all minor metals are considerably lower in the north-eastern area (cf. Fig. 4). The single probable reason for Ihese differences must be a corresponding difference in the atmospheric supply of these metals.
The very considerable rise in the lead concentrations, recorded since about 1950, can be connected to a very special source of pollution -- the lead petrol -- which does not apply to other metals. But a recent increase has been established for Ni, first recognized as the difference in the means of the periods 1900--19 and 1920--43, and the increase has accelerated between 1920--43 and 1969. This trend is correlated with the world production of Ni, which was negligible before 1920 -- less than 0.01 million tons a year, corresponding to 0.36 million tons in 1966 (Statistisk Arsbok for Sverige 1920--1968).
The rise observed in the nickel concentration of mosses will be the effect of industrial losses. The differences in the contents of Ni between north-eastern Gotaland and the low-land areas of Skane (cf. Fig. 4) will be explained by the geographical position with respect to the large industrial regions of Europe. But unlike lead, no difference in Ni be tween the low-land and the ridge areas of Sk&ne has been observed. Evidently, a larger precipitation will not achieve any increase in the case of Ni. Will this indicate that a large share of this metal, tran sported as atmospheric pollution, is supplied as dry deposition
Bot. Notlscr, t o I, 122, 1900
.Jus?:. . .
258
AKE ROHLING AND GERMUND TYLER
and therefore more uniformly distributed between the ridge and the low-land areas?
For Zn and Cu an increase has been established between the periods 1870--99 and 1900--19, but not later. However, it is very probable that this rise may have started already before the first period (as previously demonstrated for Pb; ef. Fig. 8 in RT), but sufficient material from earlier days was not available. The rise in zinc concentration is con siderable, amounting to about 30 ppm between the first two periods, whereas the corresponding figures for Cu is only about 3 ppm. The production of these metals did not differ substantially, but the probable
reason for this much larger supply of Zn will be the greater volatility of this metal. Metallurgic processes as well as the increased combustion of coal may both have been of importance as sources of pollution. During the 20th century more refined processes may have reduced the relative emission, balancing the increases in the production of the metals and keeping the total emission on a more constant level.
CONCLUSIONS
Significant differences in the recent concentrations of Pb, Cu, Zn, Ni, and Cr in Hypnum cupressiforme have been established between north-eastern Gotaland and the ridge areas as well as the low-land areas of Skane, the only exception being Cu between the north-eastern area and the low-land areas. Between the two areas in Skane significant differences exist in Pb and Cu, but not in the other metals. Where differences have been demonstrated, the highest means occur in the ridge areas of Skane, the lowest means in north-eastern Gotaland.
Significant increases during the last century have been established for Cu, Zn and Ni, as previously for Pb. The rise in the concentrations of Ni has been very marked during the last decades and corresponds to the introduction of this metal in the world production. For Zn and Cu there has also been an increase in the concentrations, considerable in the case of Zn, but chiefly confined to the decades around 1900. There is little doubt that the rise in the concentrations of these metals is an effect of a larger air-borne supply, originating from industrial activity.
ACKNOWLEDGEMENTS
This investigation was performed in the laboratories of the Department of Plant Ecology, University of Lund. The authors are indepied to Laborator
Boi. Notiscr, voh 122,19G9
l f
'y
!
DUPO 50314065
is.
ECOLOGY OK HEAVY METALS
2o<)
Nil s Maj .m mt, Head of the department, for putting the facilities of the department to our disposal ancl for advise concerning the manuscript. Part of the laboratory work was performed by Miss Le v a Ho i.m.
LITERATURE CITEIt
MiljdvAicIsforsliidng. 1967. I. Forskningsomrfidct Statens offcultiga utredningur 1967. No. 43. -- Stockholm.
OnisN, S. 19CS. Nederbdrdcns forsurning -- ell allvnrligl miljohot. -- In: Hotel mot miljon, -- Stockholm.
UOu l in g , A. & Ty l eis , G. 1968 a. An ecological approach lo the lead problem. -- Bot. Notiser 121: 321--342.
-- 1968 b. Ekologiska synpunkter pfi blyproblemel. -- Svcriges Natur 59:238--240. -- 1969. Bly i bladmossor. -- Forskning och Framsteg 69:10--11. SMHI. 1964--1968. Nederborden i Sverige. Svcriges Meteorologiska och Hydrologiska
Institnt. Arsholi 44, 2.1--48, 2.t. Stockholm. Sn ed ec o r , G. 1961. Slalisticul methods. -- Ames. Stalistisk Arsbok for Sverige. 1920--1968 (7--55(. -- Stockholm. SvENS-SON, G. K. & Lid e n , K. 1965. The quantitative accumulation of MZr+otNb and
1,0Ba+140La in carpets of forest moss. -- Health Physics II: 1033--1042. Ta mm, C. O. 1953. Growth, yield and nutrition in carpels of a forest moss (Ilylo-
comiurn splendens). -- Meddelanden Statens Skogsforskningsinsl. 43: 1--140.
!
(
{
:t S; I\
Hot. !SriMK(!r. vol. 122, ISJIKJ
V S.C .-ti-
. $T4
DUP050314066