Organic Chlorine in Deciduous and Coniferous Forest Soils in Southern Sweden

Abstract

The concentration of organic chlorine (Clorg) was determined in soil samples collected in the O-horizon in forest soils in southern Sweden to describe differences among stand types and elucidate the influence of chloride (Clinorg), pH, and organic carbon. The samples were collected within the Swedish National Survey of Forest Soils and Vegetation, with moraine as the dominating soil type and with granite and gneiss as bed-rock. The concentration of organic chlorine (Clorg) found in the deciduous forest soils was significantly lower than that in the coniferous forest soils. The Clorg increased with Clinorg, organic carbon content, and decreasing pH but was most strongly correlated to Clinorg. Crosswise comparisons among the variables showed that the influence of Clinorg_ overrode the influence of pH and organic carbon. We concluded that the major driving force in the formation of Clorg in soils is deposition of Clinorg and that the difference among deciduous and coniferous stands is due primarily to a higher input of Clinorg in the canopy of coniferous trees as a result of a larger wet and dry deposition in comferous compared with deciduous forests. The concentration of Clorg decreased significantly from June to August. This follows the pattern of chloride deposition in the region and gives further evidence that Clinorg is a driving force in the formation of Clorg in soil. The results of the present study gives further evidence that the turnover of Clorg is closely related to the turnover of Clinorg and that it is necessary to include Clorg in the assessment of the biogeochemical cycling of chlorine in the environment.

Full text

1
0038-075C/03/16805-000–000 May 2003
Soil Science Vol.168, No. 5
Copyright © 2003 by Lippincott Williams & Wilkins, Inc. Printed in U.S.A.
YEARS of research have revealed that substan-
tial amounts of naturally formed organic
matter-bound chlorine (Clorg) are present in the
terrestrial environment (Asplund and Grimvall,
1991; Field et al., 1995; Gribble, 1996; Harper,
1985; Hjelm, 1996). Furthermore, it has become
evident that Clorg is as abundant in soil organic
matter as organic phosphorus (Asplund and
Grimvall, 1991; Hjelm, 1996) and that Clorg is
formed during the degradation of organic matter
in soil (e.g.,de Jong et al., 1994; Hoekstra and de
Leer, 1993; Johansson, 1996; Myneni, 2002;
Öberg,1998).It has also been found that Clorg on
a large scale is related to the amount of organic
matter (Asplund and Grimvall, 1991), but recent
studies suggest that the deposition of chloride
(Clinorg) overrides the influence of organic mat-
ter in soils when the organic matter content is
sufficient ( Johansson et al.,2003; Johansson et al.,
in press; Öberg and Sandén, in press).
In addition, a previous study suggested that
the occurrence and formation of Clorg increases
with decreasing pH (Öberg et al., 1996a). Most
studies on the occurrence and formation of Clorg
have been conducted in coniferous forest soils,
and only a few,unreplicated samples are available
from other types of forest soils (Asplund and
Grimvall, 1991). These analyses indicate that the
concentration of Clorg is higher in coniferous
than in deciduous forest soils. Coniferous forest
soils generally have a lower pH, a higher Clinorg
ORGANIC CHLORINE IN DECIDUOUS AND CONIFEROUS FOREST
SOILS IN SOUTHERN SWEDEN
Emma Johansson, Per Sandén, and Gunilla Öberg
ITUF, Campus Norrköping, Linköping University, SE-60174 Norrköping, Sweden.
Dr. Öberg is corresponding author. E-mail: gunob@ituf.liu.se
Received Sept. 23, 2002; accepted Jan. 28, 2003.
DOI: 10.1097/01.ss.0000070909.55992.91
The concentration of organic chlorine (Clorg) was determined in soil
samples collected in the O-horizon in forest soils in southern Sweden to
describe differences among stand types and elucidate the influence of
chloride (Clinorg), pH, and organic carbon. The samples were collected
within the Swedish National Survey of Forest Soils and Vegetation, with
moraine the dominating soil type and with granite and gneiss as bedrock.
The concentration of organic chlorine (Clorg) found in the deciduous for-
est soils was significantly lower than that in the coniferous forest soils. The
Clorg increased with Clinorg, organic carbon content, and decreasing pH
but was most strongly correlated to Clinorg. Crosswise comparisons among
the variables showed that the influence of Clinorg overrode the influence of
pH and organic carbon. We concluded that the major driving force in the
formation of Clorg in soils is deposition of Clinorg and that the difference
among deciduous and coniferous stands is due primarily to a higher in-
put of Clinorg in the canopy of coniferous trees as a result of a larger wet
and dry deposition in coniferous compared with deciduous forests.
The concentration of Clorg decreased significantly from June to Au-
gust. This follows the pattern of chloride deposition in the region and
gives further evidence that Clinorg is a driving force in the formation of
Clorg in soil. The results of the present study gives further evidence that
the turnover of Clorg is closely related to the turnover of Clinorg and that
it is necessary to include Clorg in the assessment of the biogeochemical
cycling of chlorine in the environment. (Soil Science 2003;168:000–000)
Key words: Organic chlorine, chloride, organochlorine, biogeochem-
istry of chlorine, chlorine cycle.
SS-05/03-126891-Johansson 4/7/03 5:29 PM Page 1

input, and often a higher carbon content in the
top layer. This suggests that the concentration as
well as the amount of Clorg ought to be larger in
coniferous forest soils compared with deciduous
forest soils.
A study of decomposing spruce needles indi-
cated that the formation of Clorg in soil varies
seasonally, with maximum levels occurring dur-
ing the fall and minimum levels during the early
summer (Öberg et al., 1996b). This suggests that
the concentration as well as the amount of or-
ganic chlorine in soil ought to exhibit the same
type of pattern.
The aim of the present study is to investigate
if the occurrence of Clorg (i) is higher in conifer-
ous forest soils compared with deciduous forest
soil (ii) varies seasonally with a minimum during
the summer and a maximum during the fall and
(iii) is related to the occurrence of Clinorg,soil or-
ganic matter and pH.
MATERIALS AND METHODS
Terminology
The method used to detect Clorg actually
measures the sum of chlorine, bromide and io-
dine but does not distinguish between the differ-
ent halogens. Since chlorine is by far the most
abundant of these halogens in the soil environ-
ment (Hägg, 1979), the mass estimates are based
on the assumption that chlorine dominates in the
samples. Hence, we use the term chlorine to de-
note the sum of the halogens, and the discussion
is limited to that specific halogen. However, it
should be noted that chlorine has a lower mole-
cular weight than the other halogens,and, there-
fore,if halogens other than chlorine are present in
considerable amounts, the method will result in
the underestimation of the mass of organic halo-
gens in the analyzed samples.
The Swedish National Survey of Forest Soils
and Vegetation
The soil samples used in the present study
were collected by the Swedish University of
Agricultural Sciences within the Swedish Na-
tional Survey of Forest Soils and Vegetation
(Ståndortkarteringen: Ranneby, 1987). The sur-
vey monitors the state of the Swedish forests at
23,000 systematically laid rectangular sites (side
lengths: 300–1800 m) with respect to variables of
importance for forestry as well as environmental
changes in soil and vegetation. Circular sampling
areas (7–10-m radii) are placed on the corners
and in the middle of the sides of the rectangular
sites.The first round of the survey was conducted
during 1983 to 1987 when samples were col-
lected from one or several of the sampling areas
within each site until all sampling areas had been
sampled, i.e., there is no pseudoreplication.After
collection, the soil samples were stored at room
temperature in cotton bags for a maximum of 1
week before transportation to the laboratory,
where they were dried in a chamber at 35 C un-
til they reached a constant weight. Thereafter,the
samples were coarsely milled (Retsch, AS200
Digit) and sieved (2-mm-mesh size) before
analysis, resulting in two portions of the material:
particles smaller or larger than 2 mm.
Outline of the Present Study
The unreplicated samples analyzed in the pre-
sent study were collected from 1985 to 1987 in an
area covering approximately 270 km 70 km in
the southern part of Sweden (572–574N,
1159– 1639E; Fig. 1,Table 1).Because of re-
source limitations, it was not possible to analyze
all 856 samples available from the area.Therefore,
199 samples were chosen such that they would
cover the area evenly and so that each of the three
years would be represented over the entire area.
The study was designed to permit an analysis of
the spatial distribution of Clorg and Clinorg in a re-
gion reaching from Kattegatt on the west coast to
the Baltic sea on the east coast of Sweden ( Jo-
hansson et al., 2003). Since samples were col-
2JOHANSSON, SANDÉN, AND ÖBERG SOIL SCIENCE
Fig. 1. The region in southern Sweden (572–574N,
1159–1639E;) where the 199 samples that were
subjected to analysis in the present study were col-
lected by the Swedish National Forest inventory in
1985–1987.
SS-05/03-126891-Johansson 4/7/03 5:29 PM Page 2

lected in mature coniferous forest soils (both pure
and mixed pine and spruce stands) as well as in
deciduous forest soils, the data set was also suit-
able for a comparative study of deciduous and
coniferous forest soils.
There were four types of stands where sam-
ples were taken: pine (Pinus sylvestris, n 55),
spruce (Picea abies, n 74), mixed coniferous for-
est stands (n36),and mixed deciduous stands (n
34; Table 2). The different tree types were dis-
tributed evenly over the sample area. The ages of
the stands in the sampled areas ranged from 7 to
175 years, with a median age of 65 years.
The humic layer (O-horizon) reached to a
maximum depth of 20 cm, and the fraction 2
mm was chosen for the analyses.
The parent material in the region is domi-
nated by granite and gneiss. The soil types at the
sites where the samples were collected were clas-
sified as moraine (n134), peat (n29), sedi-
mentary (n25), and flat rock (n11). The
stand types were distributed evenly among the
soil types, except that the deciduous stands were
not represented among the flat rock sites.Most of
the sites were classified as fresh (n115) or
damp-to-fresh (n65),with a smaller number of
damp (n11) or dry (n5); only one site was
categorized as very dry or wet. Deciduous stands
were represented among all types except the dry
and very dry sites.
Chemical Analyses
Total carbon content was determined as a
part of the survey by the Swedish Agricultural
University, Uppsala, Sweden, using a Leco CNS-
1000 (Leco Corp., USA) through dry combus-
tion of the samples. Known amounts (0.03–0.5 g
d.w.) were combusted, with the carbon trans-
formed to carbon dioxide. The carbon dioxide
formed was detected with an IR detector,and the
whole system was flushed with He gas to screen
N2from the air.
Clinorg and Clorg (TOX) were determined at
the department of Water and Environmental
Studies in Linköping, Sweden. The Clorg content
was determined according to Asplund et al.
(1994).Briefly, 20 mg of dried and sieved soil was
shaken with 20 mL of an acidic nitrate solution
(0.2 MKNO3,0.02 MHNO3) for approximately
1 hour.This was done to induce ion exchange for
halides that could otherwise interfere with the
analysis. The samples were then filtered (polycar-
bonate,0.45 m),and the filter with the filter cake
was incinerated in a stream of oxygen.The halides
formed were titrated micro-coulometrically with
silver ions.The detection limit was less than 1 g
Clorg g1soil (d.w.).
The Clinorg was determined by mixing soil (5
g) with an acidic nitrate solution (50 mL, 0.2 M
KNO3,0.02 M HNO3). The soil extract was fil-
tered (Munktell nr 3) and washed with the acidic
solution to a final volume of approximately 70
mL.The filtrate was analyzed for its chloride con-
tent through potentiometric titration (ABU 80
Autobyrette and TTT80 Titrator, Radiometer,
Denmark). In short, 15 mL of soil filtrate was
mixed with 15 mL of carrier electrolyte (1 M
KNO3and 0.19 MHNO3) and titrated with 0.01
MAgNO3to the endpoint given by titration of a
standard portion (0.1 mL) of 0.1 MNaCl. The
Clinorg in the samples was then calculated accord-
ing to the standard procedure (Swedish standard,
SS 02 81 36, 1981). The detection limit was ap-
proximately 0.35 mg Clinorg L1,which corre-
sponds to about 10 g Clinorg g1soil (d.w.).
Statistics
Since the variables studied deviated strongly
from a normal distribution nonparametric tests,
VOL. 168 ~ NO. 5 ORGANIC CHLORINE IN FOREST SOIL 3
TABLE 1
Number of soil samples collected in various stands from
May–October 1985–1987 across southern Sweden
(5703–5769N, 1199–1665E). Each sample
was obtained through pooling 6–8 sub-samples
from the specific site.None of the sites was sampled twice
Month Year
1985 1986 1987
May 16 9 10
June 7 24 12
July 2429
August 7 8 6
September 21 16 4
October 12 7 5
TABLE 2
Number of soil samples collected in different stand types
during the sampling period (1985–1987) across southern
Sweden (5703–5769N, 1199–1665E). One pooled
sample of 6–8 sub-samples per stand was used for the
study, hence, the number of samples equals the
number of stands for each stand type
Year Deciduous Coniferous Total
Pine Spruce Mixed
1985 12 21 19 13 65
1986 12 17 32 7 68
1987 10 17 23 16 66
Sum 34 55 74 36 199
SS-05/03-126891-Johansson 4/7/03 5:29 PM Page 3

the Wilcoxon signed rank test and the 2-tailed
Kendall’s tau,were used for significance and cor-
relation, respectively, and a 5% significance level
was used if not otherwise stated.The data are pre-
sented as median values, and the variability is
given as minimum and maximum values.
RESULTS
Differences among the Stand Types
The median concentration of Clorg in the de-
ciduous forest soils was 188 g g1d.w., which
was significantly lower than the concentration in
the coniferous forest soils, which had a median of
340 g g1(Fig. 2, P0.001). The concentra-
tion of Clinorg was also significantly lower in the
deciduous stands compared with the coniferous
stands (median 34 and 77 g g1d.w., respec-
tively).No age dependence could be found when
controlling for stand type (Fig. 3, rdec 0.12,
P0.32; rcon 0.1, P0.074). The pH was
significantly higher (median 4.66 and 3.85, re-
spectively) and the organic carbon content was
significantly lower (median 19 and 33%, respec-
tively) in the deciduous compared with the
coniferous stands.The concentration of Clorg did
not differ significantly between the morain and
the other soil types.
The concentration of Clorg increased signifi-
cantly with increasing Clinorg,organic carbon
content and decreasing pH, with the strongest
correlation to Clinorg (Fig. 4, Table 3). The corre-
lation between the explanatory variables is con-
siderable,and in all cases significant.Constructing
a linear model with these three variables as ex-
planatory gave a coefficient of determination (r2)
of 0.53 for a model with Clinorg,0.56 when
adding organic carbon to the model, and 0.57
when all three variables were added.
Seasonal Variation
As the samples collected in the deciduous
forest soils differed significantly from the conifer-
ous forest stands with respect to all the variables
studied, these samples were excluded from the
seasonal analyses to investigate if possible patterns
would appear more clearly. The variation de-
creased throughout this operation, but the pat-
terns did not change substantially. Therefore,the
entire data set from the 3-year pereiod was com-
piled and divided according to the sampling
month.
The concentration of Clorg was significantly
lower in August than in May (P0.019) and
June (P0.04),with a median in August of 290
g Clorg g1(d.w.), compared with 330 and 420
g Clorg g1(d.w.) in May and June, respectively
(Fig.5). The concentration of organic chlorine in
September and October seemed to be higher
than in August, but the difference was not statis-
tically significant (P0.19 and P0.39).A pat-
tern similar to that of Clorg seemes to apply also
to Clinorg (lower concentrations in summer than
in spring and fall), but the concentrations varied
less among the months than was found for the
Clorg concentration, and the observed pattern
cannot be statistically supported.
DISCUSSION
The results of the present study suggest that
the concentration of Clorg is higher in coniferous
forest soils than in deciduous forest soils. The
Clorg increased with Clinorg,organic matter con-
tent, and decreasing pH. Previous studies have
shown that the chlorine-to-carbon ratio varies
less than the concentration of Clorg (e.g.,Asplund
and Grimvall, 1991) and that the concentration
of Clorg increases with decreasing pH (Öberg et
al., 1996a). The results of the present study sug-
gest that the influence of Clinorg overrides the in-
fluence of organic matter and pH. These variables
co-vary strongly in the field situation, which
makes it difficult to clarify the underlying pro-
cesses causing the observed patterns. However,
even if there are still some pieces lacking in the
4JOHANSSON, SANDÉN, AND ÖBERG SOIL SCIENCE
Fig. 2. Concentration of organically bound chlorine
(Clorg; black box) and chloride (Clinorg; white box) in soil
samples collected in deciduous (n 34) and conifer-
ous (n 165) forest stands across southern Sweden
(5703–5769N, 1199–1665E;). The box plots
show median 1st and 3rd quartiles. Whiskers show min-
imum and maximum if they are within 1.5 box lengths.
Data outside these values are shown as outliers.
SS-05/03-126891-Johansson 4/7/03 5:29 PM Page 4

VOL. 168 ~ NO. 5 ORGANIC CHLORINE IN FOREST SOIL 5
Fig. 3. Concentration of Clorg as a function of stand age in soil samples collected in deciduous forest stands (trian-
gles) and coniferous forest stands (squares) in a region in southern Sweden (572–574N, 1159–
1639E;).
Fig. 4. Concentration of Clorg as a function of Clinorg; organic carbon and pH age in soil samples collected in 199 for-
est stands in a region in southern Sweden (572–574N, 1159–1639E;).
SS-05/03-126891-Johansson 4/7/03 5:29 PM Page 5

puzzle, the results of the present study, in combi-
nation with previously conducted studies,make it
possible to formulate a hypothesis that explains
the observed patterns: That is, Clorg is formed
during degradation of organic matter through
(enzymatic) formation of reactive chlorine. The
two main substrates for this process are organic
matter and Clinorg;hence its dependence on these
two variables. In the pH-interval 3–5, the (enzy-
matic) process increases with decreasing pH (As-
plund et al., 1994; Asplund and Grimvall, 1991),
hence its dependence on pH. The patterns ob-
served in the present study imply that Clinorg is,in
general, the limiting factor for the formation of
Clorg in Swedish soils.
Influence of Forest Type
Clinorg in soil derives mainly from dry and
wet deposition. Unfortunately, Clinorg deposition
data are not available for the specific sites that are
monitored in the Swedish National Survey of
Forest Soils and Vegetation. However, it is well
documented that the dry deposition is consider-
ably higher in conifer stands. The significantly
higher concentration of Clinorg in the conifer
stands of the present study (Fig. 2) suggests
strongly that the deposition of Clinorg was higher
in the conifer stands than in the deciduous stands.
Hence, our conclusion is that increasing deposi-
tion of Clinorg renders increasing concentrations
of Clorg in soil and that the difference among de-
ciduous and coniferous stands is due primarily to
a higher input of Clinorg in the canopy of conif-
erous trees as a result of a larger wet and dry de-
position of Clinorg in coniferous as compared with
deciduous forests.
Previous studies indicate strongly that both
the formation of Clorg (consumption of Clinorg)
and the mineralization (release of Clinorg) of Clorg
take place in soil ( Johansson et al., 2001; Öberg
et al., 1998; Öberg et al.,1996b). It is not yet pos-
sible to distinguish these processes, and, conse-
6JOHANSSON, SANDÉN, AND ÖBERG SOIL SCIENCE
TABLE 3
Correlation between Clorg,Cl
inorg,organic carbon content,
and pH. p0.001 in all cases
Clinorg organic carbon pH
Clorg 0.48 0.4 0.38
Clinorg 0.31 0.31
Organic carbon 0.38
Fig. 5. Variation from May (5) to October (10) in the concentration of Clorg and Clinorg in 199 forest soil samples
across southern Sweden (5703–5769N, 1199–1665E).
SS-05/03-126891-Johansson 4/7/03 5:29 PM Page 6

quently, only net changes may be studied. The
results of the present study indicate that the oc-
currence of Clorg is related to the deposition of
Clinorg,which suggests that the net-formation of
Clorg is enhanced by Clinorg.A recently concluded
laboratory study gives further credence to this
conclusion (Öberg and Sandén, in press). Thus,it
seems likely that the observed differences among
the coniferous stands, compared with the decid-
uous stands,are caused by a higher formation rate
of Clorg in coniferous forest soil compared with
deciduous forest soil.
Seasonal Variation
A previous study of decomposing spruce
needles suggests that the concentration of Clorg
varies on a seasonal basis with minimum levels
during early summer and higher concentrations
in spring and especially during late fall (Öberg et
al.,1996b). These results are supported somewhat
by the observations made in the present study
since it gives further evidence that the concen-
tration of Clorg varies seasonally (Fig. 5). In line
with the above reasoning on the relation between
the deposition of Clinorg and the concentration of
Clorg in soil,a possible explanation of the seasonal
variation in the concentration of Clorg is that the
deposition of Clinorg peaks during the winter in
Sweden and declines during the summer (Granat,
1987). It is interesting to note that the seasonal
variation in the concentration of Clinorg is weaker
than the variation of Clorg.The large variation
among replicates implies that only a smaller part
of the variation in the concentration of Clorg in
soil is explained by seasonal variation and that
other variables have a stronger influence.
A few comments on the seasonal variation
may be of interest. A significant decrease in the
Clorg concentration was observed in the present
study from June to August.The same pattern was
observed in a previous study of decomposing
spruce litter at a field sites 3north of the region
where the present study was conducted. A de-
crease in Clorg concentration can be caused by
mineralization, leaching, or volatilization. During
the period June to August, the transport of water
through soil is limited in both regions because the
increased evapotranspiration renders a permanent
moisture deficiency (Raab and Vedin, 1995).
Thus, leaching is not a very plausible reason for
the observed decrease.This suggests that mineral-
ization of Clorg or volatilization of the same in-
creases during this period. It is known that chlo-
rinated volatiles such as chloromethane and
chloroform are formed in soil (Harper, 1985;
Harper, 1995; Harper et al., 1988; Laturnus et al.,
1995; Watling and Harper, 1998). However, the
underlying processes that have so far been identi-
fied suggest that chloromethane is formed by a
methylation of chloride rather than by a fragmen-
tation of chlorinated organic matter. This suggests
that the formation of chlorinated volatiles con-
sumes Clinorg rather than decreases the Clorg con-
tent of soil organic matter. Hence, mineralization
of chlorinated organic matter (release of Clinorg) is
the most plausible explanation for the decrease
observed during the summer period.
The litter bag study conducted previously
(Öberg et al., 1996b) suggested a decrease in the
concentration of Clorg from May to June,whereas
no change (or an increase) was observed during
the corresponding period in the present study.The
former study focused on the litter,and the present
study comprises the O horizon,i.e.,the part of the
soil that is below the litter layer.The flux of water
as well as the transport of organic matter increases
during the spring floods, which is stronger further
north in the country. This suggests that leaching is
a plausible cause of the decrease in the Clorg con-
centration observed in the litter in the former
study and the lack of decrease (or even an increase)
in the O-horizon of in the present study.This sug-
gests that the processes that favor mineralization of
Clorg override those that favor formation of Clorg
during the summer period.
The litter bag study (Öberg et al., 1996b)
suggested that the concentration of Clorg in-
creases from August to October.A similar pattern
was discerned in the present study but could not
be supported statistically.Hence,the results of the
present study can be used neither to corroborate
nor to reject the pattern suggested by the previ-
ous study.
A Comment on Possible Interference with the
Geochemical Cycling of Clinorg
The present study strengthens previous find-
ings that suggest that the Clorg storage in Scandi-
navian soils is considerably larger than the storage
of Clinorg and that the formation of Clorg increases
with increasing deposition of Clinorg (Johansson
et al., 2001; Rodstedth et al., 2002). Thus, it is
clear that the turnover of Clorg interferes strongly
with the Clinorg cycle.
Clinorg is highly mobile as it is highly soluble
and, consequently, follows the water movement.
In the Scandinavian region, as in the temperate
region in general, the major movement of water
is downwards inasmuch as precipitation is greater
than evapotranspiration. The annual deposition
VOL. 168 ~ NO. 5 ORGANIC CHLORINE IN FOREST SOIL 7
SS-05/03-126891-Johansson 4/7/03 5:29 PM Page 7

of Clinorg in the investigated region varies from
around 30 kg Clinorg ha1on the West Coast to
less than 5 kg Clinorg ha1on the east coast
(Granat, 1987). If Clinorg is conserved in soil, the
amount transported through the soil will equal
the amount transported to surface water.
Clorg can be expected to be less mobile than
the inorganic fraction, because the major part of
the Clorg is bound to high-molecular-weight,
humic-like structures (Hjelm, 1996; Myneni,
2002). The Clorg present in soil originates from
wet and dry deposition,litter fall,an intrinsic for-
mation within the soil, and, occasionally, local
anthropogenic sources (Öberg and Grøn, 1998).
It is recognized that Clorg is lost from the system
through mineralization and leaching (Rodstedth
et al., 2002; Öberg et al., 1996a; Öberg et al.,
1996b; Öberg and Sandén, in press).If formation
exceeds mineralization, and the leaching is of
considerable magnitude,the soil will act as a sink
of Clinorg.On the other hand, if the mineraliza-
tion exceeds formation and leaching,the soil will
act as a source.Thus,the cycling of Clinorg may be
much more complex than the commonly de-
scribed geochemical cycle suggests. Hence, the
geochemical cycle of Clinorg ought to be ex-
tended and combined with the turnover of Clorg
to form the biogeochemical cycle of chlorine.
ACKNOWLEDGMENTS
As described in the paper, part of the data set
used in this publication was made available by the
Swedish National Survey of Forest Soils and Veg-
etation performed by the Department of Forests
Soils, Swedish University of Agricultural Sci-
ences, Sweden, which we greatfully acknowl-
edge. The authors of the present paper are solely
responsible for the interpretation of data. We are
also indebted to Evert Carlsson, SLU, for kind
assistance with information concerning details
around the sampling procedures. We also thank
Cecilia Toomväli, Britt Aurén, both at Swedish
Geotechnical Institute, Linköping, and Lena
Lundman at the Department of Water and Envi-
ronmental Studies in Linköping for help with the
chloride measurements. The authors also thank
the Swedish Council for Forestry and Agricul-
tural Research for financial support. Finally, we
thank our anonymous referees for helpful and
constructive comments.
REFERENCES
Asplund, G., and A.Grimvall. 1991.Organohalogens in
nature, more widespread than previously assumed.
Environ. Sci. Technol. 25:1346–1350.
Asplund, G.,A. Grimvall,and S.Jonsson.1994. Determi-
nation of total and leachable amounts of organohalo-
gens in soil. Chemosphere 28:1467–1475.
de Jong, E., J.A. Field, H.-A.Spinnler, J. B. P.A.Wijn-
berg, and J. A. M. de Bont. 1994. Significant bio-
genesis of chlorinated aromatics by fungi in natural
environments. Appl. Environ. Microbiol. 60:264–
270.
Field,J.A., F. J.M.Verhagen,and E.de Jong.1995.Nat-
ural organohalogen production by basidiomycetes.
TIBTECH 13:451–456.
Granat, L. 1987. Luft och nederbördskemiska station-
snätet inom PMK. 3475, Naturvårdsverket, Solna.
Gribble, G. W., 1996. Naturally Occurring Organo-
halogen Compounds — A Comprehensive Survey.
Progress in the Chemistry of Organic Natural
Products. Springer,New York.
Hägg, G. 1979. Allmän och oorganisk kemi. Almqvist
& Wiksell Förlag AB, Uppsala,Sweden.
Harper,D. B.1985.Halomethane from halide ion — A
highly efficient fungal conversion of environmen-
tal significance. Nature 315:55–57.
Harper, D. B. 1995.The contribution of natural halo-
genation processes to the atmospheric halomethane
burden.In Naturally-Produced Organohalogens.A.
Grimvall and E. W. B. de Leer (eds.). Kluwer Aca-
demic Publishers, Dordrecht, The Netherlands,pp.
21–34.
Harper, D. B., J. T. Kennedy, and J. T. G. Hamilton.
1988. Chloromethane biosynthesis in poroid fungi.
Phytochemistry (Oxf.). 27:3147–3153.
Hjelm, O., 1996. Organohalogens in coniferous forest
soil. PhD dissertation, Linköping Studies in Arts
and Science Nr 139, Linköping University,
Linköping, Sweden.
Hoekstra, E., and E.W. B. de Leer. 1993. Natural pro-
duction of chlorinated organic compounds in soil.
In Contaminated Soil. F.Arendt, G.J. Annokkée, R.
Bosman, and W.J. van den Brink (eds.). Kluwer
Academic Publishers, Dordrecht, The Netherlands,
pp. 215–224.
Johansson, C., 1996.Halogenated structures in natural
organic matter. PhD dissertation, Linköping Stud-
ies in Arts and Science Nr 140, Linköping Univer-
sity, Linköping, Sweden.
Johansson,E., G.Ebenå, P.Sandén, T.Svensson,and G.
Öberg. 2001. Organic chlorine and chloride in
Swedish spruce forest soil: Influence of nitrogen.
Geoderma 10:1–13.
Johansson, E., Z. Hu, P. Sandén, X. B. Zhang, and G.
Öberg. Turnover of organic chlorine in submerged
paddy soil; a case study in the Meicun area, south-
east China. in press.
Johansson,E.,P.Sandén,P.and G.Öberg. 2003.Spatial
patterns of organically bound chlorine and chlo-
ride in Swedish forest soil. Chemosphere, in press.
Laturnus, F., G.Mehrtens,and C.Grøn.1995. Haloper-
oxidase-like activity in spruce forest soil – A source
of volatile halogenated organic compounds?
Chemosphere 31:3709–3719.
8JOHANSSON, SANDÉN, AND ÖBERG SOIL SCIENCE
AQ1
AQ2
SS-05/03-126891-Johansson 4/7/03 5:29 PM Page 8

Myneni, S. C. B.2002. Formation of stable chlorinated
hydrocarbons in weathering plant material. Science
295:1039–1041.
Öberg, G. 1998. Chloride and organic chlorine in soil.
Acta Hydrochem. Hydrobiol. 26:137–144.
Öberg,G.,and C. Grøn.1998. Sources of organic halo-
gens in spruce forest soil. Environ. Sci. Technol.
32:1573–1579.
Öberg,G.,and P.Sandén. Retention of chloride in soil
and cycling of organic matter bound chlorine. (in
press).
Öberg, G., I. Börjesson,and B.Samuelsson.1996a. Net
change in organically bound halogens in relation to
soil pH.Water Air Soil Pollut.89:351–361,
Öberg, G., E. Nordlund, and B. Berg. 1996b. In situ
formation of organically bound halogens during
decomposition of Norway spruce litter — Effects
of fertilization. Can. J. For. Res. 26:1040–1048.
Öberg, G., C. Johansen, and C. Grøn. 1998. Organic
halogens in spruce forest throughfall. Chemosphere
36:1689–1701.
Raab, B., and H.Vedin (eds.). 1995. Klimat, sjöar och
vattendrag. Nationalatlas. Bokförlaget Bra Böcker,
Höganäs, Sweden.
Ranneby,B. 1987.Designing a new national forest sur-
vey for Sweden. Forestalia Suecica, p. 177.
Rodstedth, M., C. Ståhlberg, P. Sandén, and G. Öberg.
2002. Chloride imbalances in soil lysimeters.
Chemosphere (in press).
Swedish standard,SS 02 81 36. 1981.Determination of
chloride content in water — potentiometric titra-
tion.
Watling, R., and D. B. Harper. 1998. Chloromethane
production by wood-rotting fungi and an estimate
of the global flux to the atmosphere. Mycol. Res.
102:769–787.
VOL. 168 ~ NO. 5 ORGANIC CHLORINE IN FOREST SOIL 9
AQ1”Can this be completed now?”
AQ2”Can this be completed now?”
AQ3”Can this be completed now?”
AQ4”Can this be completed now?”
AQ5”Where is this published? Please indicate name / organization of publisher.”
AQ3 AQ4
AQ5
SS-05/03-126891-Johansson 4/7/03 5:29 PM Page 9