I M P A C T S OF A C I D P R E C I P I T A T I O N ON

CONIFEROUS FOREST ECOSYSTEMS

G U N N A R ABRAHAMSEN, R I C H A R D HORNTVEDT and BJORN TVEITE

Norwegian Forest Research Institute,
1432 d s - N L H , Norway

(Received 26 March, 1976)

Abstract. This paper summarizes the results from current studies in Norway. One main approach is the
application of artificial acid 'rain' and of lime to field plots and lysimeters.
Application during two growth seasons of 50 mm mo-~ of 'rain water' of pH 3 to a podzol soil increased
the acidity of the humus and decreased the base saturation. The reduction in base saturation was mainly
due to leaching of Ca and Mg.
Laboratory experiments revealed that decomposition of pine needles was not affected by any acid 'rain'
treatment of the field plots. Liming slightly retarded the decomposition.
No nitrification occurred in unlimed soils (pH 4.4-4.1). Liming increased nitrification.
The soil enchytraeid (Ofigochaeta) fauna was not much affected by the acidification.
Germination of spruce seeds in acidified mineral soil was negatively affected when soil pH was 4.0 or
lower. Seedling establishment was even more sensitive to increasing soil acidity.
Analysis of throughfall and stemflow water in southernmost Norway reveals that the total deposition of
H2SO 4 beneath spruce and pine is approximately two times the deposition in open terrain. A large part of
this increase is probably due to dry deposition. Increased acidity of the rain seems to increase the leaching
of cations from the tree crowns.
Tree-ring analysis of spruce (Picea abies (L.) Karst.) and pine (Pinus sylvestris L.) has been based on
comparisons between regions differently stressed by acid precipitation and also between sites presumed to
differ in sensitivity to acidification. No effect that can be related to acid precipitation has yet been detected
on diameter growth.

1. Introduction

Southern N o r w a y receives acidic pollutants with air and precipitation far in excess o f
what could be expected from the emissions within the region. The monthly mean p H of
the precipitation varies between 4 and 5.
The deposition o f acids represents a possible threat to forest and fresh-water
ecosystems. A n increasing number o f rivers and lakes are becoming too acidic for trout
(Jensen and Snekvik, 1972). Occasional mass-dying of trout has been observed.
Estimates of an anticipated reduction in forest growth have been given (Dahl and Skre,
1971). Lysimeter studies have indicated a d r a m a t i c increase in the leaching of C a from
soil when the percolating water p H drops below 3 (Overrein, 1972).
The great concern over these problems in N o r w a y has resulted in the establishment
o f a large nationwide project n a m e d ' A c i d Precipitation - Effects on F o r e s t and Fish'
(Overrein, 1976). The aim of the forest-research portion of this project is primarily to
study the effects o f acid precipitation on forest growth and development, secondly to
study the effects on main processes influencing forest growth and development. The
present paper summarizes the main approaches, and the results obtained so far, without
comprehensive discussion.

Water, Air, and Soil Pollution 8 (1977) 57-73. All Rights Reserved
Copyright © 1977 by D. Reidel Publishing Company, Dordreeht-Holland
 58 GUNNAR ABRAHAMSEN E T AL.

2. Effects on Soll

The effects of acid precipitation on soil properties and processes are undoubtedly very
complicated. Experiments with simplified systems in the laboratory must therefore be
controlled by field experiments where the complex interaction of climate, organisms,
and organic and inorganic matter is largely undisturbed. Because most field
experiments have to be pursued over a long period of time, these experiments have been
given priority in the starting phase of the project.
The soil studies include investigations on soil fauna, decomposition and soil
chemistry.

2.1. EXPERIMENTAL DESIGN

The soil studies are based on the application of lime and of simulated acid rain, to field
plots and lysimeters. The aim of these experiments is to obtain information on the effect
of acid precipitation on soil properties, soil processes, and plant growth. At present,
results from one plot experiment and one lysimeter study, both started in 1972, are
available. These particular experiments were located in a semipodzolic soil
(intermediate profile between podzol and brown earth), developed in a sandy,
glacifluvial sediment. Soil chemical characteristics appear in Table I. The experimental
area, which is situated approximately 40 km north of Oslo, was clear cut about 1960
and reforested with Pinus contorta Douglas, in 1965. In 1974 the trees were between 2
and 4 m high. The ground flora is completely dominated by Deschampsiaflexuosa (L.)
Trin.
TABLE I
Soil chemical characteristics of the field experiments

Soil Soil Soil chemical properties (~ + s)

horizon depth
cm Loss on PH(H~O) Cation exchange Base N in % of oven-
ignition capacity saturation dry material
% meq/100 g %
A0 0-ca. 3 28+10 4.4+0.1 39+13 25+3 1.27
A2 3-6 7.3+0.9 4.4_+0.1 17+ 2 15+4 0.14
B 9-12 4.3+0.4 4.7+0.1 11+ 1 7+2 0.06

Three by five meter plots were treated with lime and 'rain' of various acidity
(Table II). 'Rain water' quality was adjusted with H2SO4. The simulated rain was
applied in addition to the natural precipitation, and was applied once a month in the
non-frozen period of the year.
The lysimeters are 29.5 cm wide and 45 cm high fiberglass cylinders. Each is filled
with an undisturbed soil monolith according to the procedure described by Overrein
(1968). Treatments appear in Table II.
The leachate from the lysimeters was collected in 5-1 plastic containers. Each week,
100 ml of the leachate was analyzed for pH, conductivity and color.
 IMPACTS OF ACID PRECIPITATION ON CONIFEROUS FOREST ECOSYSTEMS 59

T A B L E II
Treatment of the field plot (P) and the lysimeter (L) experiment. Lime was applied as ground limestone.
Experimental design: Complete randomization with 3 replicates in the plot experiment and 2 in the
lysimeter study

Liming No Irrigation
irrigation
25 m m mo -1 50 m m m o -1
pH 5.7 pH 4 pH 3 pH 5.7 pH 4 pH 3

N o lime P P L P L P L P P P
1500 kg CaO ha -1 P
3000 kg CaO ha -1 P P P
6000 kg CaO ha -1 P

The remaining leachate was stored in the field and preserved by adding HC1
(10 ml 1-1). At the end of each month, I 1 of the preserved leachate for the month was
used for chemical analysis.

2.2. SOILZOOLOGY
Soil samples collected from the field experiment in October 1974 were extracted for
enchytraeids (Oligochaeta) which is one of the most abundant groups of animals in
coniferous forest soils. The procedures used in sampling, extraction, etc. are described
elsewhere (Abrahamsen, 1972). Other groups of soil animals were not considered.
Statistical analyses did not reveal any significant influence of the various treatments
on the total abundance of enchytraeids. The three main species, however, seem to
respond differently to the various treatments. Cognettia sphagnetorum (Vejdovsky,
1877) is a dominant species in acidic raw-humus habitats (e.g. Abrahamsen, 1972).
Populations of this organism were significantly reduced by liming, but not by tile 'acid
rain'. Enchytronia parva Nielsen and Christensen, 1959 is confined to more fertile
soils (Abrahamsen, 1972). This organism increased in numbers with increasing acidity
of the rain at a high lime concentration and decreased with increasing acidity at a low
lime concentration (significant interaction at 1% level). The abundance of the third
main species, Enchytraeus norvegicus Abrahamsen, 1972 has not been significantly
influenced by any treatment.

2.3. SOIL MICROBIOLOGY

Studies on microbiological processes have been focused on the decomposition of

organic matter, nitrification and nitrogen fixation. No results are yet available for
nitrogen fixation.
Decomposition has been studied on needles of Pinus contorta (Ishac and Hovland,
1976). The needles were collected from the field in 1973 and 1974 in the month of
November. They were incubated at 15 ° and 25°C on moistened glass-wool in conical
flasks. Weight loss was measured after 90 days.
Decomposition was significantly increased by increasing temperature from 15 o to
60 GUNNAR ABRAHAMSEN ETAL.

40

309+_12 31.3±0.3

J • el
30- • o I o
D • o•
• • o•
nJo o•
o • •0
o• eo
u,o oo
go Iv
24.6+-2.6
oo •e
so • I
m ol ,.g o I
...o
• oq

i o l ~o o•
20-
i o q ~o o•
ooq ~ o e•
o ol J o go
oo so
ooq J • • •
oo Jo
o o l ) •O i ,
P booo
.:.:, ,;.; ....
~•l • •
• ol
o• t o,
o o l Io, oe
eo Jo~
o o l =oq • •
o• =0,
e o l pol oo
10-
• ol ~o~ • •
o• pol
• ol ~o~ go
• • oo ~ol
Do i ~eq go
• • ~ol
Dol bo~ go
• • ~ol
J o i ~o~ oo
• • ,o~
~o~ • •
• • o q
Dol ~o~ oo
• • o• ,oq
J I i ~o~ go
go o q
Dol
0 go • (
oO oe
Do e ~ol oo
m,a o,• o.e

1 1.8 3 3.5 4 initial pH

I // = // i //0 f>
1 1.85 505 6.45 Final pH

Fig. 1. Effect of pH on the decomposition(weight loss after 105 days) of pine needles (Pinus contorta).
pH of the substrate was adjusted by dilute H2SO4.

25°C, significantly decreased by liming (3000kg C a • ha -~) and not significantly

influenced by the acidity (pH 5.7 and pH 3) of the 'rain'.
Decomposition was also examined on needles from the control plots. The needles
were moistened with various dilutions of H2SO4. The experiment showed that the
decomposition rate (weight loss) was significantly (P < 0.05) increased when the initial
pH was increased from 1.8 to 3.5 (Figure 1). No difference was found between pH 3.5
and 4. At pH 1 no decomposition took place.
Humus samples collected in October 1974 from the field plot experiment (Table II)
were used for nitrification experiments in the laboratory (Hovland and Ishac, 1975).
The experiment was carried out using a perfusion technique with the addition of
(NH4)2SO 4. No nitrification occurred in humus from the unlimed plots. Liming on the
other hand significantly increased the nitrification. The amount of nitrate formed was
almost proportional to the quantity of lime added to the soil. No interaction was found
between liming and pH. The experiment indicates that in this particular soil (Table I),
the formation of nitrate is of small ecological importance probably due to the higb
acidity.
 IMPACTS OF ACID PRECIPITATIONON CONIFEROUS FOREST ECOSYSTEMS 61

2.4. SOIL CHEMISTRY

Soil cores sampled from the plot experiment (Table II) in October 1974, and the
leachate of the lysimeters were analyzed for chemical properties by standard methods.
Soil reaction and the amount of soluble and exchangeable nutrients in the soil of the
various experimental plots are given in Tables III and IV. Table III shows the main
effect of liming independently of the treatment with acid 'rain'. This can be given since
acid 'rain' had no effect on soil characteristics at the lowest rain level where lime had
been applied. Liming had on the other hand highly significant effects on soil chemical
properties. The decrease in K by increased liming must be explained by ion exchange of
the two elements. The decrease in the amount of soluble and exchangeable Mn is on the
other hand explained by a transition to water-insoluble Mn compounds.
Effects of 'rain' acidity were only observed at the highest 'rain' level where no lime
was applied (Table IV). Rain of pH 3 applied at a rate of 50 mm mo -1 resulted in a
significant reduction (P<0.01) of the degree of base saturation. The interaction
between the amount and the acidity of the 'rain' was also highly significant (P < 0.005).
The decrease in base saturation at the highest 'rain' level results from the decreasing

TABLE III
Effect of liming (ground limestone) on soil chemical properties. Data from the lowest 'rain' level only
(25 mm mo -1)

Soil layer and Kg CaO ha -x

depth cm
0 1500 3000 6000

PH(H20)
Humus ca. 3 cm 4.27 5.92 6.54 6.65
Mineral soil 0 - 3 cm 4.31 4.75 4.85 5.14
Mineral soil 6 - 9 cm 4.68 4.71 4.70 4.92

K meq/100 g
Humus ca. 3 cm 1.15 0.87 0.73 0.85
Mineral soil 0 - 3 cm 0.13 0.12 0.11 0.10
Mineral soil 6 - 9 cm 0.06 0.06 0.06 0.05

Ca meq/100 g
Humus ca. 3 cm 4.8 24.1 40.6 57.6
Mineral soil 0 - 3 cm 1.2 4.1 5.0 7.1
Mineral soil 6 - 9 cm 0.4 0.8 0.6 1.1

Mn meq/100 g
Humus ca. 3 cm 1.07 0.58 0.50 0.53
Mineral soil 0 - 3 em 0.19 0.16 0.I 1 0.11
Mineral soil 6 - 9 cm 0.04 0.05 0.04 0.03

Base saturation %
Humus ca. 3 cm 22 77 100 100
Mineral soil 0 - 3 crn 12 26 31 40
Mineral soil 6 - 9 cm 7 9 8 12
62 GUNNAR ABRAHAMSENE T AL.

TABLE IV
The effect of the quantity and acidity of the 'rain' on soil chemical properties. Data from unlimed plots only

25 mm mo -x 50 mm mo -a
pH 5.7 pH 4 pH 3 pH 5.7 pH 4 pH 3

pH(H20)
Humus ca. 3 cm 4.4 4.1 4.3 4.4 4.6 4.1
Mineral soil 0 - 3 cm 4.3 4.2 4,4 4.5 4.4 4.2
Mineral soil 6 - 9 cm 4.8 4.6 4,7 4.7 4.7 4.7

N a meq/100 g
Humus ca. 3 cm 0.15 0.17 0.14 0.11 0.10 0.09
Mineral soil 0 - 3 cm 0.05 0.05 0.04 0.06 0.05 0.05
Mineral soil 6 - 9 cm 0.04 0.04 0.04 0.04 0.04 0.04

K meq/100 g
Humus ca. 3 cm 1.14 1.32 1.00 0.99 0.85 0.87
Mineral soil 0 - 3 cm 0.13 0.13 0.13 0.14 0.16 0.13
Mineral soil 6 - 9 cm 0.06 0.07 0.07 0.06 0.07 0.06

Ca meq/100 g
Humus ca. 3 cm 4.3 5,4 4.7 6.1 4.9 2.8
Mineral soil 0 - 3 cm 1.3 0.9 1.3 1.2 1.2 0.8
Mineral soil 6 - 9 cm 0.5 0.3 0.4 0.6 0.4 0.4

Mg meq/100 g
Humus ca. 3 cm 1.38 1.45 1.15 1.55 1.12 0.81
Mineral soil 0 - 3 cm 0.39 0.30 0.30 0.39 0.34 0.23
Mineral soil 6 - 9 cm 0.12 0.06 0.13 0.11 0.09 0.09

Mn meq/100 g
Humus ca. 3 cm 1.20 0.87 1.16 1.31 0.94 0.73
Mineral soil 0 - 3 cm 0.20 0.15 0.22 0.22 0.21 0.19
Mineral soil 6 - 9 cm 0.03 0.04 0.04 0.06 0.04 0.03

Base saturation %
Humus ca. 3 cm 24 19 23 25 27 18
Mineral soil 0 - 3 cm 14 9 13 14 11 8
Mineral soil 6 - 9 cm 9 5 7 7 6 6

content of soluble and exchangeable cations that occur in the soil when the pH of the
'rain' drops from 4 to 3. The decrease in the amount of the individual cations was not,
however, large enough to be statistically significant.
No effect due to the acidity of the 'rain' was found on soil chemical properties in the
field plots treated with 25 mm 'rain' per month. However, acid 'rain' applied at the
same rate to the lysimeters significantly increased the leaching of Ca and Mg (Table V).
The figures given in the table are relative, meaning that the amount of Ca or Mg
leached in 1973 at pH 5.7 is designated 1.0 and the other values are designated relative
to this one. Increased leaching by increased acidity of the 'rain' has not been observed
for other cations, nitrate or organic N; nor has the acidity of the leachate been
 IMPACTS OF ACID PRECIPITATION ON CONIFEROUS FOREST ECOSYSTEMS 63

TABLE V
The effect of acid 'rain' on the relative leaching of Ca ana Mg. 'Rain'
quantity: 25 m m mo -1. Leachingin 1973, pH 5.7 is designatea 1.0.
Differences in leaching due to different treatments are significant at the
0.1% level

Element Year pH of'rain'

5.7 4

Ca 1973 1.0 1.4 1.4

1974 1.3 2.9 4.5
Mg 1973 1.0 1.2 1.4
1974 1.2 1.9 2.4

influenced by the treatment. Other lysimeter experiments in operation indicate,

however, very significant increases in the leaching of most cations when the amount of
'rain' increases and especially when it becomes more acidic than pH 3.

3. Effects on Trees

3.1. WASHING AND LEACHING OF TREE CROWNS

When penetrating tree crowns, the chemical composition of the rain is altered, mainly
due to the washing of absorbed deposits ('dry' deposits), and the leaching of excreted
metabolites.
The chemical composition of incident rainfall, throughfall and to a lesser degree,
stemflow has been studied during three summer and autumn periods in Birkenes, South
Norway, and one period in Mfilselv, North Norway.
Throughfall collectors were placed halfway between the stem and the perimeter of
the crown projection. Incident rainfall was measured in open terrain. Stemflow was
collected by means of a collar nailed to the tree.
The results (Table VI) show that the deposition of most chemical substances was
higher in South Norway than in North Norway. Note, however, that the sampling
period is shorter and the amount of precipitation less in North Norway. Further, the
throughfall enrichment of sulphate, Ca, and K was greater in South Norway than in
North Norway. In South Norway, the tree crowns seem to have absorbed nitrate and
ammonium from rainwater. Throughfall beneath spruce and pine, in contrast to birch,
contained more strong acid than did incident rainfall. In North Norway, the amounts of
strong acid were negligible.
The concentrations of the different substances were in general negatively but not
significantly correlated with the amount of rainfall or throughfall. In incident rainfall,
chloride, Na, and Mg (mainly derived from sea salt) constituted one group of correlated
ions, and ammonium, nitrate, Ca, sulphate, and strong acid constituted another. In
throughfall, chloride, Na, Mg, ammonium, nitrate, Ca, sulphate, and K were correlated.
64 GUNNAR ABRAHAMSEN E T AL.

o~

"0 <~

0"~

0 0 0 0

0 0 0 0
g

0 0 0 0

o~

D e~
~ ¢¢~ .~- t~

~o
~o z
0 0 0 0
lall.

,.. T' I
N~ o 0"! 0 0 0 0
-~ z ko ~ 0.3 P--
0

0
.~.
?°
RR~? Z
0

8 zzzz, 9999
0
O,-t~
I
0

0 M ~mm0
~8
0
O~ ~
< ~.~ m ~
 IMPA~2TSOF ACID PRECIPITATION ON CONIFEROUS FOREST ECOSYSTEMS 65

The lack of correlation between strong acid and the other ions in throughfall was most
consistent beneath birch.
When the net removal of chemical substances from the tree crowns (throughfall
minus incident rainfall values) in North Norway and South Norway is compared to
values obtained from southern Sweden and from Germany (Table VII), a certain
pattern appears. The removal of chloride and Na reflects the distance from the coast to
the study areas. The removal of K, Mg, Ca, and, most distinguishably, S shows a
considerable increase from north to south. The increase in S and Ca is probably caused
by a greater dry deposition, cf. the wet deposition pattern in Europe (Od6n, 1971;
Royal Ministry for Foreign Affairs and Royal Ministry of Agriculture, 1971).
The amount of stemflow increased in the order: spruce < pine < birch (Table VIII).
The concentration of strong acid, sulphate, chloride, Ca, and Mg was roughly two
times greater in stemflow than in throughfall. Na and K differed less. In birch the
stem bark or epiphytes absorbed a considerable amount of nitrate and ammonium. It is
further noteworthy that while birch throughfall was less acidic than incident rainfall,
stemflow was more acidic.
A conclusive interpretation of the results appears to be impossible at present. It is
probable that a larger part of the throughfall enrichment in chloride, sulphate, Ca, Na,
and H ions in Birkenes is derived from dry deposits than from leached metabolites.
Further, H ions from dry and/or wet deposits might replace other cations in the tree
crowns.

3.2. GERMINATIONAND ESTABLISHMENT

The results presented here are based on experiments carried out with seeds of spruce
(Picea abies (L.) Karst.) and pine (Pinus sylvestris L.) in artificially acidified mineral
soil (Teigen, 1975).
Soils of different pH were obtained by percolating mineral soil with 1500 mm water
of different pH values. Additional soil pH levels were obtained by liming. The soil used
was derived from a podzolized morainic soil with a low content of plant nutrients
(cation exchange capacity 10 meq/100 g and base saturation 2%).
Seeds from one open pollinated tree of each species were transferred to the acidified
soils. An adequate moisture regime was obtained by adding distilled water. The number
of germinated seeds and established seedlings was recorded after 7 weeks. Seeds were
defined to have germinated when the emerged embryo was at least 3 mm long.
Seedlings were defined as established when the primary needles were developed.
The experiments with pine covered a soil pH range of 4.0 to 4.6. No effects upon
germination or establishment were found within this range.
The experiments with spruce covered a soil pH range of 3.8 to 5.6. Significant effects
of soil pH upon both germination and establishment were demonstrated (Figures 2 and
3). Germination seems to have a rather broad optimum around pH 4.8. Establishment
seems to have a more narrow optimum around pH 4.9. About 80% of the seeds did not
develop normal seedlings at pH 3.8. A large number of these plants had developed
66 GUNNAR ABRAHAMSEN E T AL.

e~

..=
O O O

_=
• . .

..=

O
O • t'--. e q.
.o
8

~ 0q. ~.

~.o.

gz ©
O O O

• . ,
Z
~a
t~
>

0 Z

T_
0 ¢'4 --

E ~C
OC
8 tt~ t r

I I I

o
o
E
<
 IMPACTS OF ACID PRECIPITATION ON CONIFEROUS FOREST ECOSYSTEMS 67

%
100-
O

80-

O O

60- ©
O

~" = - g 9 ~ , 0 6 " 205, 54 " ( p H ~ 2l tS.(pH} 2

40- R 2 = 0, ~05 s = 7, /~

20-

// pH
a~ 4.0. . 4.2
. . .
4.4 4.6 4.8 5 io '
5.2 5 ~4 s'.6 5~
Fig. 2. Relation between soil reaction and percentage of spruce seed germination.

%
100- o
o
0 O

co o

80- o %
~/3 O

o ~ °
60- o

40- o

20- :/ co
(13
O
= -[ [gg,(}q :

I~ 2 = f 7 7 7 /
%[q,{q' [pill-

S = 10,78
52,811 ([)I[) Z

// 3[8 410 412 414 4.6 418 5[0 5[2 5~1 5~5 5~8pH
Fig. 3. Relation between soil reaction and percentage of spruce seedlings established.

roots, but the roots did not penetrate into the soil. Variations in the nutrient content of
the soils had no apparent effect upon germination or establishment.

3.3. TREE GROWTH

Possible effects of acid precipitation upon tree growth are partly being studied in field
experiments with simulated acid rain and partly by analyzing past tree growth in
different regions and on different sites. At present only preliminary results from tree
ring analyses are available.
Tree ring analyses have been used in studies of air pollution effects around local
emission sources (e.g. Pollanschiitz, 1971, Sundberg, 1974). As far as we know the only
attempt to use this approach in analyses of regional acidification effects has been made
by Jonsson and Sundberg (1972).
68 GUNNAR ABRAHAMSEN E T AL.

Our approach and models lean heavily upon the Swedish study of Jonsson and
Sundberg (1972). Tree ring development is compared: (1) between regions presumed or
known to have different inputs of acid precipitation, and (2) within regions between
sites supposed to differ in sensitivity owing to soil properties. Increment cores of spruce
(Picea abies (L.) Karst.) and pine (Pinus sylvestris L.) are obtained from the National
Forest Survey which annually collects information about the Norwegian forests
according to a systematic clustered sampling scheme. Sample plot information is used
to stratify the material into site or regional groups.

Diff.Iogloir Difference series

Sorlandet - Qstlandet

+0.100I

- 0.10

SPRUCE
Logtoir
1.20
S6rlandet - 4 8 3 plots
1.10 ~ . . ,A~ 7 . . ,'-" ~
/.~'~
, A
.--. ~stlandet - 2 2 4 plots

1.00 ~V~ ~.~ " ~~ ~ ~. u ~'..\ • ~,J\~ ,

0.90
O.8O
, l l l l l iiI p , l i l l l l l l l l l l l i , , i l , , I , i ~ ,i'l' '' 'iJ'
1930 1940 1950 1960 1970
Fig. 4. Tree ring developmentof spruce in differentregions and a differenceseries betweenthe regions.

It is necessary in the models used to have a reference period before the acidification
effects started. At this stage of the study, the period from the year 1927 and onwards
has mainly been investigated. Trees of the same species covering the whole investiga-
tion period are put together within sample plots to form an average plot tree-ring series.
In the models adopted, the factors influencing growth are supposed to act mul-
tiplicatively. Additive models are obtained by using logarithmic ring width as the
dependent variable.
Plots within the same region or site group are combined to form average tree ring
series for the separate groups. Two groups are further compared by forming a
difference series which shows the relative tree ring development of the groups. The
difference series are analyzed for possible trend changes.
The analyses are complicated by a number of factors. Ideally there should be a
negligible interaction between group and growth year. This is probably not the case in
 IMPACTS OF ACID PRECIPITATION O N CONIFEROUS FOREST ECOSYSTEMS 69

comparisons between geographically separated regions or in comparisons between

regions which are extreme in ecological factors. Autocorrelation in the time series is
another factor which decreases the sensitivity of the analyses. Other complicating
factors are possible differences in cutting practices and intensity, stand history, and the
normal time-trend between regions or site groups that are compared. The effect of these
factors must be evaluated in addition to the time series analyses.

Diff. logl0ir Difference series

Sorlandet - Qstlandet
+ 0.100 I

- 0.10

PINE
Log 10i~
1.20
Sorlandet - 817 plots
1.10 .... Ostlandet - 163 plots

1.00

0.90
0.80 i'°
!
'~i
0.70
,,,,,,, ...... i . . . . ~. . . . l , , , , I .... i ......... ~,,
1930 1940 1950 1960 1970
Fig. 5. Tree ring developmentof pine in differentregions and a differenceseries between the regions.

Figures 4 and 5 show the average tree ring development in pine and spruce in two
geographic regions - Sorlandet and Ostlandet, or South Norway and East Norway
respectively - as well as the difference series between the regions. Sorlandet receives
more acid precipitation and is also supposed to be more sensitive to acidification due to
shallow soils.
The main features of ring-width development are similar in the two regions pointing
back to common components of climatic growth factors and time trends. The difference
series reveal some fluctuations in the relative tree ring development. Interactions
between region and year seem to be a main component of these fluctuations according
to preliminary analyses.
These interactions decrease the sensitivity of analyses of trend changes. The year
1950 has been used as a starting point for possible acidification effects in the Swedish
study. There are, however, no indications of a relatively slower growth within Sorlandet
70 GUNNAR ABRAHAMSEN ET AL.

after this year. The region Ostlandet is, on the other hand, also influenced by acid pre-
cipitation and does not give an ideal reference.
The data from one region (Sorlandet) have been more fully analyzed in an attempt to
elucidate the possible effect of different productivity (the site-class concept in forest
terminology), vegetation type, water regime, soil depth, and tree species in relation to
the acidification problem. Although the analyses are not completed, the difference series

Diff.logloir Difference series

poor t y p e - rich type

- 0.10

°I
-0,20

-0.30
SPRUCE
Vaceinium myrt. - 117 plots
Loglo ir • --. Herb - 45 plots
1.30

1.20
• ~
•-./ " ~
",..../o - o / \
\/\ /
1.10

1.00

0.90

0.80
ii
0.70

1930 1940 1950 1960 1970

Fig. 6. Tree ring developmentof spruce within differentvegetationtypes and a differenceseries between
the types. Stands from Sorlandet (South Norway) below 300 m altitude.

between extreme groups do not support the hypotheses that less productive sites, poor
vegetation types, ombrogen sites, or shallow soils should be more sensitive to acidifica-
tion. Neither does an analysis of the development of spruce relative to pine support the
hypothesis that spruce is more sensitive to acidification when growing on poor sites.
Figure 6 shows as an example the tree ring development of spruce within two groups of
vegetation types (a Vaccinium myrtillus type and a herb-rich type) from stands within
Sorlandet and below 300 malt. Figure 7 shows finally the development of pine on two
groups of soil depths (<20 cm and >70 cm) within the same region and height zone.
The reason for the lack of relationship between tree ring development and the
hypotheses about acid precipitation effects can be many-fold. The effects may be
 IMPACTS OF ACID PRECIPITATION ON CONIFEROUS FOREST ECOSYSTEMS 71

Diff. log lo Jr Difference series

* 0.10 , shallow soil - deep soil

- 0.10

-0.20

PINE
L o g lo i r
1.20 I I
/'\ ~Soil depth < 20cm - 209plots
, • / ~ °--° Soil depth > 70 cm - 41 p l o t s

1.10

1.00
k. \, ,k'a. ,I,'/
V
/ ..,fl
"
I",', '. /,.4.,"~ ,

0.90

0.80

0.70
i

0.60
,ll~l,tlllltl iiiiIiitl tlllllll,~ll,lllll, 11
1930 1940 1950 1960 1970

Fig. 7. Tree ring dev~opment of pine on different so~ depth groups and a difference series between the
groups. Stands ~om Sod~d~ (Sou~ Norway) below 300 m aRitude.

masked by other factors which are not satisfactorily isolated by the approach. The
hypotheses about possible effects are based upon meagre facts and may be false. More
data may be needed. It also may be that no large effects of acid precipitation upon tree
growth have occurred.

4. Summary and Conclusion

The present paper summarizes the current results of ongoing studies in Norway on the
impact of acid precipitation and dry deposits on coniferous forest ecosystems. These
studies include investigations on soil, leaching from tree crowns, germination, establish-
ment of seedlings and tree ring analyses.
The soil studies are entirely based on simulated acid rainfall (pH range 5.7 to 3.0)
and the application of lime, in field plots and lysimeters. Results from these experiments
are at present available from an experiment running since September 1972. The results
concern soil fauna, decomposition of organic matter, and soil chemistry.
The abundance of one important group of the soil fauna - the Enchytraeidae, has
been studied. Two species had not been influenced by the acid rain. One species
72 GUNNAR ABRAHAMSENETAL.

decreased in abundance with increasing acidity of the 'rain' at a low lime level and
increased in abundance at a high lime level.
Laboratory experiments with pine needles (Pinus contorta) from the field plots have
not revealed any influence of acid 'rain' (pH range 5.7 to 3.0) on decomposition. Liming
on the other hand, has reduced the decomposition rate significantly.
During the period from September 1972 to October 1974 significant reduction in the
degree of base saturation has occurred in the topmost layer of the soil of the plots
where the highest amounts of acid have been applied. This reduction is reflected by the
increased leaching of Ca and Mg in the lysimeter experiments.
Leaching from tree crowns has been studied in Birkenes, South Norway, being
heavily exposed to long transported air pollutants, and in M~dselv, North Norway,
which is supposed to serve as a control area.
Both the amount of various chemical components in throughfall and the ratio
between concentration in throughfall and in incident rainfall are significantly higher in
Birkenes than in M~tlselv. Conclusive interpretation of these results is impossible at
present. However, the high concentrations in Birkenes may derive from dry deposition,
the leaching of metabolites and the exchange of ions between plant tissue and deposited
pollutants.
Green-house experiments with spruce (Picea abies) on artificially leached (with dilute
H2SO4) mineral soil indicate that germination and especially establishment are
negatively affected when soil reaction drops below pH 4.0 to 4.2.
Regions supposed to differ in exposure to acid deposition have been compared with
regard to tree ring development. So far no differences have been found that can be
related to the pollution. Nor have clear effects of acid precipitation been observed on
sites supposed to be most sensitive to acidification (poor vegetation types, shallow soils,
etc.).
The experiments described in the present paper indicate that severe negative
influences on soil and organisms are not to be expected over a period of a few years
when the acidity of the rain is above ca. pH 4. Since the acidity of the precipitation in
southern Norway varies greatly between pH 3.5 and 5.7 with a prevailing pH of ca. 4.5
(1974) it seems likely that possible effects on tree growth up to now have been too small
to be detected by tree ring analysis. However, longer exposure time and increased
acidity of the precipitation may be a threat to soil conditions, the health of vegetation,
and finally to forest production.

Acknowledgment
SNSF contribution FA 1/75.

References
Abrahamsen, G.: 1972, Pedobiologia 12, 26.
Dahl, E. and Skre, O.: 1971, in Konferens om avsvavling, Stockholm 11. november 1969. Nordforsk, Mil-
j6v~dssekretariatet, Publikation 1971: 1, pp. 27-40.
 IMPACTS OF ACID PRECIPITATION ON CONIFEROUS FOREST ECOSYSTEMS 73

Hovland, J. and Ishae, Y. Z.: 1975, 'Effects of Simulated Acid Precipitation and Liming on Nitrification in
Forest Soil', SNSF-prosjektet, IR 14/75, 15 pp.
Ishac, Y. Z. and Hovland, J.: 1976, 'Effects of Simulated Acid Precipitation and Liming on Pine Litter
Decomposition', SNSF-prosjektet, IR 24/76, 20 pp.
Jensen, K. W. and Snekvik, E.: 1972,Ambio 1,223.
Jonsson, B. and Sundberg, R.: 1972, 'Has the Acidification by Atmospheric Pollution Caused a Growth
Reduction in Swedish Forests?, A Comparison between Regions with Different Soil Properties',
Skogsh6gskolan, Inst. f6r skogsproduktion. Rapporter och Uppsatser, Nr. 20, 48 pp.
Nihlgfird, B.: 1970, Oikos 21, 208.
Odin, S.: 1971, in Mysterud, I. (ed.). Forurensning og biologisk miljovern, Universitetsforlaget, Oslo 1971,
pp. 63-98.
Overrein, L. N.: 1968, Soil Sei. 106, 280.
Overrein, L. N.: 1972, Ambio 1, 145.
Overrein, L. N.: 1976, this volume.
Pollanschfitz, J.: 1971, Mitt. forstl. BundVersAnst. Wien 91, 153.
Royal Ministry for Foreign Affairs and Royal Ministry of Agriculture: 1971, 'Air Pollution across
National Boundaries. The Impact on the Environment of Sulfur in Air and Precipitation. Sweden's Case
Study for the United Nations Conference on the Human Environment', Stockholm 1971. 96 pp.
Sundberg, R.: 1974, in Pratt, J. W. (ed.). Statistical and Mathematical Aspects of Pollution Problems, New
York, pp. 167-175.
Teigen, O.: 1975, 'Spire og etableringsforsok reed gran og furu i kunstig forsuret mineraljord', SNSF-
prosjektet, IR 10/75, 36 pp.
Ulrich, B.: 1968, Allg. Forstz. 23, 815.