Oecologia (1991) 87:198-207

Oecologia

9 Springer-Verlag 1991

Carbon and nitrogen isotope ratios in different compartments of a healthy and a declining Picea abies forest in the Fichtelgebirge, NE Bavaria G. Gebauer and E.-D. Schulze Lehrstuhl Pflanzen6kologie,Universit~itBayreuth,Postfach 101251, W-8580 Bayreuth,Federal Republic of Germany Received October 10, 1990 / Acceptedin revisedform February 5, 1991

Summary. Natural carbon and nitrogen isotope ratios were measured in different compartments (needles and twigs of different ages and crown positions, litter, understorey vegetation, roots and soils of different horizons) on 5 plots of a healthy and on 8 plots of a declining Norway spruce ( P i c e a abies (L.) Karst.) forest in the Fichtelgebirge (NE Bavaria, Germany), which has recently been described in detail (Oren etal. 1988a; Schulze et al. 1989). The 613C values of needles did not differ between sites or change consistently with needle age, but did decrease from the sun- to the shade-crown. This result confirms earlier conclusions from gas exchange measurements that gaseous air pollutants did no long-lasting damage in an area where such damage was expected. Twigs (613C between - 25.3 and - 27.8%o) were significantly less depleted in 13C than needles (613C between -27.3 and -29.1%o), and c5x3C in twigs increased consistently with age. The 61SN values of needles ranged between - 2.5 and - 4.1%o and varied according to stand and age. In young needles 615N decreased with needle age, but remained constant or increased in needles that were 2 or 3 years old. Needles from the healthy site were more depleted in 15N than those from the declining site. The difference between sites was greater in old needles than in young ones. This differentiation presumably reflects an earlier onset of nitrogen reallocation in needles of the declining stand. 615N values in twigs were more negative than in needles ( - 3 . 5 to -5.2%0) and showed age- and stand-dependent trends that were similar to the needles. 61SN values of roots and soil samples increased at both stands with soil depth from - 3 . 5 in the organic layer to + 4%0 in the mineral soil. The 6XSN values of roots from the mineral soil were different from those of twigs and needles. Roots from the shallower organic layer had values similar to twigs and needles. Thus, the bulk of the assimilated nitrogen was presumably taken up by the roots from the organic layer. The problem of separation of ammonium or niOffprint requests to: G. Gebauer

trate use by roots from different soil horizons is discussed.

Key words: 6'3C 615N-Nitrogen assimilation Forest decline P i c e a abies - Stable isotopes

Anthropogenic nitrogen depositions affect forest health on acidic soils. Atmospheric depositions lead to an imbalance between the availabilities of nitrogen and cations. This imbalance is especially detrimental to plant productivity, because nitrogen stimulates growth even at a low supply of other base cations (Oren et al. 1988 b; Schulze 1989). A complicated system emerges in which the assimilation of nitrogen deposited by anthropogenic processes disturbs the normal balance among carbon uptake, allocation and growth (Schulze et al. 1989). To further investigate this interaction between nitrogen assimilation and the growth processes driven by carbon assimilation we used the natural abundances of stable isotopes of carbon (613C) and nitrogen (615N) as indicators of long-term plant responses to disturbances in the natural supplies of carbon and nitrogen. The 61BC value of a leaf is determined by the carbon isotope ratio of atmospheric COz, by the metabolic pathway of photosynthesis, and by leaf conductance (O'Leary 1981; Farquhar et al. 1982). It responds to changes in the diffusive pathway and in the photosynthetic system. Damage to the pathway of CO2 uptake will affect the 613C value irrespective of whether this change was caused by poor nutrition or atmospheric pollution. The 615N value of a plant sample is primarily determined by the isotope ratio of the nitrogen source. The isotope ratio of the various nitrogen sources varies depending on the isotopic fractionation processes that are associated with soil nitrogen transformations (Mariotti et al. 1982; Shearer and Kohl 1986). Therefore, inorganic nitrogen compounds of different chemical nature, from different soil types and soil depths or from

199

the atmosphere, are expected to have different isotope ratios. Presently it is unclear from which soil compartments trees draw their nitrogen supply, whether they use ammonium or nitrate, and to what extent uptake through the canopy contributes to tree nitrogen nutrition. Thus, the natural 513C and ~tSN values contain information on the supply and use of carbon and nitrogen within a plant as well as within the ecosystem. This study was designed to investigate natural variations of 613C and 6 ~SN values in different compartments of a declining and a healthy spruce forest (Schulze et al. 1989). In addition to examining the specific effects of anthropogenic nitrogen deposition on 6~5N values, we investigated how 6aSN values change within different compartments of a tree or an ecosystem at a more general level. The latter information is especially needed since, in contrast to the abundant literature on changes of ~ ~aC with environmental parameters (Rundel et al. 1989), relatively little is known about the natural changes of fia5N. It is expected that different plant compartments and metabolites vary with respect to their nitrogen isotope ratios in a way analogous to the variations of carbon isotope ratios within plants (Benner et al. 1987; Brugnoli et al. 1988; O'Leary 1981; Winkler et al. 1978; Ziegler 1979). Material and methods

Plant material and soil samples Two 30-year-old Picea abies (L.) Karst. stands in the Fichtelgebirge (NE Bavaria, Germany) were chosen for investigation (Oren et al. 1988a). The stands occur on the same parent rock material, phyllite, and are exposed to the same climate. One stand is apparently healthy and shows no visible symptoms of forest decline. The other stand shows a high degree of variation in symptoms ranging fi'om trees with full crowns and green needles to trees with a large fraction of yellow needles and needle loss. Therefore, the health status of the trees at the latter stand ranges from apparently similar to the healthy stand to apparently different from the healthy stand. The declining stand grows on a podzol, while the healthy stand grows on a podzolic cambisol with higher calcium and magnesium and lower aluminum concentrations in the soil solution than found at the declining site (Kaupenjohann 1989). For further details of the site see Oren et al. 1988a. Plots of 80 m 2 (5 at the healthy, 8 at the declining site) were randomly selected for this investigation and for a series of other investigations (Oren et al. 1988a; Schulze et al. 1989). The plots are distributed over an area of about 4 km 2 (healthy stand) or 3.2 km 2 (declining stand), respectively. The distance between the two stands is about 15 km. In October 1987 needles and twigs of four different age classes (current-year to 3-year-old) were collected for this study from three canopy levels (upper sun-crown, lower sun-crown and shade-crown) of one co-dominant tree per plot randomly selected from the other co-dominant trees of each plot. These trees had consistently green needles only at the healthy site. At the declining site some trees had yellow needles among the older needle age classes, a condition associated with magnesium deficiency (Oren et al. 1988b). In October 1988 sampling of needles and twigs from the lower sun-crown of the same trees was repeated and expanded to all needle age classes present on each twig sampled. Freshly fallen litter was collected from 1-m 2 litter traps under each sampling tree. Understorey vegetation (Vaccinium myrtillus L., Deschampsia flexuosa (L.) Trin. and mosses) was found on the surface of soil cores from four plots at the declining site and all species from each plot were combined for further analyses. Root

and soil samples were taken from three soil cores per sample tree and were separated into two classes based on mineral soil horizons (Ao- 5 cm, A5-15 era) and two classes based on organic layers (O~f,

Oh). Plant samples were carefully cleaned in deionized water. Mineral soil was sieved (mesh 3.15 mm). All samples were dried at 105 ~ C, ground in a bali mill (Retsch Schwingmfihle M M 2) and stored in a desiccator.

Analytical methods" Isotope ratios and nitrogen contents were measured with a system combining an elemental analyzer (HERAEUS CHN-O Rapid) for Dumas combustion of the samples, a F I N N I G A N MAT Trapping Box HT for automatic cryo-purification of the combustion products, and a F I N N I G A N M A T delta D gas isotope mass spectrometer with a dual inlet system. Standard gases were calibrated with respect to international standards (N2 in the air and COz in PeeDee belemnite) by use of the reference substances N1 and N2 for nitrogen isotope ratios and NBS 16 to 20 for carbon isotope ratios provided by the IAEA, Vienna. In the following sections isotope ratios are denoted as 3 values, calculated according to the following equation :

~x= (Rsample/RStandard 1)" 1000 [%0] -

-

8X is the isotope ratio of carbon and nitrogen in delta units relative to the international standards, and Rs,mple and Rs~ana,ra are the laC/12C or 15N/14N ratios of the samples and standards, respectively (Rundel et al. 1989). The upper limit for the quantitative combustion of plant and soil samples in the elemental analyzer was 40 mg per sample. The lower limit for linearity of the measured 51SN values was 17.8 gmol eq nitrogen (=8.9 Ixmol Nz gas) per sample. Therefore, a total nitrogen concentration of about 0.4 mmol eq per g dry substance was needed for the measurement of the ~lSN value after a single combustion. For samples with lower total nitrogen concentrations sample combustion was repeated up to five times and the resulting Nz gas was collected in the Trapping Box system before measure-

I

I

I

Y = 0.563 x + 0.128

10.0-

I

/

= .

8.0t:L

-.

6.0-

4.o13.

2.0

I

I

I

I

5

10

15

20

N 2 gas [ pmol ]

Fig. 1. Example of a six-point calibration curve used for nitrogen quantification by pressure measurement of the N2 gas trapped from the elemental analyzer combustion products. The positive y intercept of the regression equation represents the pressure also found for " b l a n k " measurements

200

Table 1. F values, probabilities (P), and degrees of freedom (df), for relations between 6x3C, 6~5N, or nitrogen concentrations (N) and different compartments of the trees. P values are listed for main effects independent of their significance, but for interactions P values are listed only if P < 0 . 0 5 ; n.s.: not significant

c~13C

c]15N

N

df

F

P

E

P

F

P

1.88 101.40 17.96 n.s.

0.1752 0.0001 0.0001

19.70 43.12 1.04 n.s.

0.0001 0.0001 0.3568

1.16 363.53 1.43 4.16

0.2841 0.0001 0.2447 0.0443

1 1 2 1

1.58 6.90 7.42 n.s. n.s.

0.2168 0.0030 0.0002

11.47 0.63 36.71 9.67 2.91

0.0014 0.5366 0.0001 0.0001 0.0106

3.24 0.50 17.84

0.0784 0.6121 0.0001 n.s. n.s.

1 2 3 3 6

yr 0/1 yr 1/2 yr 2/3

5.56 2.69 6.66 14.17

0.0030 0.1270 0.0241 0.0027

42.89 78.74 26.19 0.2/

0.0001 0.0001 0.0001 0.6493

3.21 0.56 0.03 9.14

0.0306 0.4639 0.8658 0.0077

3 1 1 1

yr 0/1 yr 1/2 yr 2/3

3.63 0.00 0.96 13.41

0.0178 1.000 0.3383 0.0016

8.92 18.57 7.20 19.18

0.0001 0.0002 0.0121 0.0002

24.34 3.81 10.26 26.94

0.0001 0.0609 0.0034 0.0001

3 1 1 1

site position age age*site age*position

0.52 11.64 82.01

10.26 0.59 34.34 4.59 n.s.

0.0025 0.5598 0.0001 0.0043

0.93 2.09 126.00 n.s. 3.99

0.3405 0.1354 0.0001

4.55

0.4748 0.0001 0.0001 n.s. 0.0004

0.00t

1 2 3 3 6

age, healthy site yr 0/1 yr 1/2 yr 2/3

46.86 42.73 0.00 22.74

0.0001 0.0001 1.0000 0.0005

37.78 91.26 0.68 12.10

0.0001 0.0001 0.4210 0.0029

52.34 25.10 16.64 9.68

0.0001 0.0001 0.0008 0.0064

3 l 1 1

yr 0/1 yr 1/2 yr 2/3

45.94 27.19 0.36 66.93

0.0001 0.0001 0.5575 0.0001

15.20 49.90 0.01 41.26

0.0001 0.0001 0.9201 0.0001

88.41 41.84 69.57 29.47

0.0001 0.0001 0.0001 0.0001

3 1 1 1

0.07 1.74

0.7945 0.1821

0.41 63.70

0.5366 0.0001

0.17 12.37

0.6861 0.0001

1 3

0.88 0.03 5.96 n.s.

0.3724 0.8694 0.0372

25.98 37.40 5.01 n.s.

0.0006 0.0002 0.0520

23.52 0.08 0.67 48.35

0.0009 0.7812 0.4347 0.0001

1 1 1 3

All data site organ position site*organ Needles site position age age*site age*position age, healthy site

age, declining site

Twigs

age, declining site

Roots site horizon Olf/Oh Oh/A5 A5/A15 horizon*organ

ment of the isotope ratio. There was no such problem for carbon isotope measurements. Reproducibility of the isotope ratio measurement was controlled by daily measures of acetanilide ( M E R C K Chemikalien, Darmstadt, Germany). Mean values and standard deviations of these measurements were -1.03_+0.12%o for c515N o for 6 13C (n=85). (n=105) and -30.76_+0.13%o Nitrogen concentrations of plant and soil samples were calculated from partial pressure measurements of pure N2 gas trapped from the elemental analyzer combustion products in a distinct volume of the F I N N I G A N M A T Trapping Box system. Acetanilide with a constant nitrogen concentration of 10.36% was used for a daily six-point calibration (see Fig. 1). Tenfold analysis of the nitrogen standard in the range between 7.5 and 20 gmol N2 gas resulted in a maximal deviation of _+3.5%. The dimension mmol eq N g-1 was used because N is present in organic material in a heterogenous mixture of substances in which more than one atom of N may be present with different molarity.

Statistical methods All data of needles, twigs a n d roots were analysed in a repeatedmeasures experimental design using the SAS statistical pagkage (SAS Institute 1985). Data of litter samples were analysed-b~"a Student's t-test. Regressions were calculated using ~he SPSS-PC statistical package (Norusis 1987).

Results The main effects on 613C and 615N values and on N c o n c e n t r a t i o n o f site, c r o w n p o s i t i o n , o r g a n a n d o r g a n a g e a r e s u m m a r i z e d i n T a b l e 1. A l l d a t a i n d i c a t e d t h a t

201

Upper sun-crown I

I

I

Lower I

I

i

i

i

9

healthy

o

declining Oct.1987 Oct.1988

U nderstorey vegetation

Shade-crown

sun-crown i

l

|

i

i

i

i

I

I

I

!

I

I

!

I

-27-

-28" "3

-29. IZD

-30" -31"

/ 1

I

I

T

I

I

I

I

I

l

I

I

I

I

I

I

I

I

l

I

I

|

--3-

=

_z.. -5-

---~__~

1.000.75E z

050-

0.251

I

|

I

I

l

l

I

I

I

l

6E

o

iS

n m

I

_--

/

2

E .9 all

I

I

1

I

I

I

I

t

1

1

I

I

I

0

I

2

3

0

I

2

3

~

5

6

0

2

Needle age [ years]

Fig. 2. Mean (_+1 SE) 61aC and 61SN values, nitrogen concentrations and biomass of needles from different age classes and canopy levels of trees growing at the healthy (n=5) and the declining site (n=8) and of understorey vegetation collected at the declining site (n = 4)

organ had a significant effect, while site had significant effects on c515N values, but not on nitrogen concentrations or 613C. Crown position had a significant effect only on 613C values. For needles and twigs, the 6'3C values changed significantly with crown position and age, while the effects of site and age were significant for the a~SN values. N concentration of needles and twigs was only affected by age. In roots, site and horizon

had no effect on 613C, but 6*SN and N concentration of roots changed significantly with soil layer. The data presented in Fig. 2 provide more detail of the effects of needle age on 6 a3C and a 15N. In the upper sun-crown the average 613C value of -27.74%0 ( S E = 0.09, n = 5 2 ) corresponds to a calculated intercellular CO~ concentration c~ of 230 gl 1-1 (Farquhar et al. 1982) and this agrees with the c~ measured by gas exchange

202 Upper sun-crown

Lower i

i

sun-crown

I

I

Shade-crown 1

i

I

I

I

I

I

I

I

l

-26-27 o

O

-2s

V'D

-29 -30-

-31"

9

healthy

o

declining Oct.1987 0ct.1988

.... f

I

1

I

I

I

I

I

I1

I

!

i

I

I

I

I

I

I

I

i

I

I

i

|

-3Z v,-

-Z* ~

-5-

9r,'= 1.00-

m

o')

~0.75E

s z

=--

m

4

o.2sI-,

Twig age [years]

Fig. 3. Mean (_+1 SE) 613C and 6ZSN values and nitrogen concentrations in twigs from different age classes and canopy levels of trees growing at the healthy (n = 5) and the declining site (n = 8)

( Z i m m e r m a n n 1990). In the shade, the average 513C value decreased to - 28.75%o (SE = 0.11, n = 51), which corresponds to a calculated q of 250 ppm. The greater c~ value in the shade m a y be due to combined effects of low light, high air humidity and increasing CO2 close to the ground ( Z i m m e r m a n n 1990). The 613C value decreased to -31.2%o in the herbaceous layer (Fig. 2). Although the overall 613C value was significantly affected by needle age, this effect was not consistent between needle age classes. It probably represents yearly variations of 613C which are consistent in all organs. There is no consistent trend which would separate the sites,

but rather a specific environmental effect of each year of needle production. Thus, needles produced during the years 1987 and 1985 (age classes 0 and 2) produced biomass o f lower 613C than those produced in 1986 or 1984 (age classes I and 3). These year by year changes are inverse to the measured rate of CO2 assimilation (Schulze et al. 1990). 61SN values of needles ranged between - 2 . 5 and -4.1%o and changes were more complicated than those of 613C (Fig. 2). Effects of site and age were significant and additionally there was a significant age-dependent position effect of the 615N values in needles (Table 1).

203

Fichtelgebirge, 0ct.1987

Picea abies I

I

I

-2S-

0

-26

I

9

-

I

i

I

9 healthy

I

to.c,o,no

~

I

--2 oo

--3 %

~ eo

g o

._g- 2 7 -

o

o e O ~

o

Z

_ o/st - , - , 9

kO ,r,-

O

,.,,,p'~ 0

t~

~o - 2 8

Z Oo-o O~LID

o'D

o

-5

O0

fo

o o

-29

9

y-- 0.950 x +0.205

/

r = 0.84

T ~

o I

-30

I

-29

I

-28

I

-26

y_-1002,.,.o,2

I

-6

r : 0.93

I

I

I

I

I

-5

-4

-3

-2

-I

I

-27

i~ 13Cneedle s [ ~

"/r

o]

6 15Nneedles [ % o ]

Fig. 4. Correlations between 613C or 61SN values in needles and twigs from three canopy levels of trees growing at the healthy and the declining site

While 5tSN values decreased from current-year to 2or 3-year-old needles at the healthy site and remained constant in older needles, the declining site showed initially a less pronounced decrease, and in contrast to the healthy site, a significant increase of 515N values of old needles. Thus, the difference between sites was greater in old needles than in young ones. The same result was obtained in both years measurements were made. On both stands needles from the shade-crown were about 0.3%0 more depleted of ~SN than needles from the upper sun-crown. The 5~SN value of the herbaceous ground vegetation (Fig. 2) was -4.6%0 and thus depleted in 15N compared to needles. Needle nitrogen concentrations were not different between sites, but decreased with needle age due to increases in needle weight (Fig. 2). The N concentration of plants in the herbaceous layer was only slightly higher than in needles. Twigs had 513C values between -25.3 and -27.8%o and thus were significantly less depleted in ~3C than needles (Fig. 3). In contrast to needles, 513C of twigs increased consistently with age, although this change was not significant from current-year to 1-year-old needles (Table 1). The average difference between the 513C values of needles and their respective twigs was 1.2%o in current-year samples and increased to 2%0 in 3-yearold samples. Like needles, the &13C values of twigs decreased significantly from the sun-crown to the shadecrown (Fig. 3, Table 1). The 515N values of twigs ranged between - 3 . 5 and -5.2%0 and thus, in contrast to 5~3C values, were significantly lower than in needles (Fig. 3, Table 1). The age dependency was more pronounced than in needles and

similar at both sites. 6 lsN decreased significantly from current-year to 1-year-old twigs and increased again from 2nd- to 3rd-year twigs. Twig N concentrations were not different between sites or crown positions, but decreased significantly with age. The relation between 513C or 515N in needles and twigs is positive and linear (Fig. 4). Twigs were always more enriched with laC, but contained less lSN than needles. Over the whole range of data the 515N values of needles and twigs are proportional with a constant offset of -1.06%o (SE=0.10, n = 154). The slope of the regression equation is 1.002 (SE=0.03, n=154) and is not significantly different from 1. The slope of the regression for the 513C values is 0.95 (SE=0.05, n = 155) and thus is slightly different from 1, and the error of the y intercept (+0.205) is high (SE = ] .39, n = 155). This observation is related to the different age-dependency of 513C in needles and twigs (see Figs. 2 and 3). In contrast to the variation in 5~3C within trees, which was larger than between trees, the 515N values showed larger variation between trees than within trees (not shown in Fig. 4). The nitrogen isotope ratios and N concentrations of litter samples were similar to those found for old age classes of living needles at both sites and were significantly different between sites ( P < 0.05, Fig. 5). For soil samples at both sites there was a significant increase of the 5 1 5 N values with soil depth from - 3 to +4%0, which was accompanied by a decrease in soil nitrogen (Fig. 5). Correlation between N concentrations and 515N values of soil samples was linear and negative (Fig. 6). In parallel to soil samples the 51SN values of

204 Roots I

I

I

I

-26" o

~-27 rD

(,

r ~=

-28 Litter

.

m

Soil 1

I

I

I

I

I

6

9 healthy o declining

2 O

z

tO

0

tr

-2 -4 I

I

[

I

f I

I

I

I

qo *7, O~ O"

2O 0.75E E

Z I

I

0if

Oh A0-s As-is

Organic layer Mineral soil

0If

I

Oh

I

I

Fig. 5. Mean (+1 SE) 613C and c~15Nvalues and nitrogen concentrations of soil and root samples from four soil horizons and in litter samples collected at the healthy (n = 5) and the declining site (n = 8)

A0-s

Organic layer Mineral soil

roots increased with depth in the soil (Fig. 5). Thus, the 61SN of roots is positively correlated with the 6~5N of the soil (Fig. 6), but there is a slope of 0.488, which is significantly different from a 1:1 relationship ( P < 0.001). In the uppermost organic layer (Oil horizon) roots had a 61SN value of -3.5%0, which was lower by -0.5%o than the soil, and this difference increased with soil depth to -4.2%0 in the lowermost root horizon. The 615N values of roots in the uppermost organic layer ranged between the 615N values found in needles and twigs. The N concentration of roots decreased with soil depth. In contrast to 61SN, 6~3C values of roots did not differ between soil horizons (Fig. 5), The ~13C values of roots from both sites were similar to those found in twigs.

Discussion The carbon isotope ratios of needles, twigs and roots of Picea abies in the Fichtelgebirge are in the typical range for plants with Ca photosynthesis (O'Leary 1988) and are similar to those found for other Norway spruce samples from the same habitats (Oren et al. unpublished work). There were no significant differences in ~13C between healthy and declining stands, indicating that gaseous air pollutants, such as SO2, 03 and NO2, had no differentiating or long-lasting effect (Lange et al. 1989; Zimmermann et al. 1988). Significant differences in fi13C were found between twigs and needles, as known from other woody plant species, and have previously been explained by differences in chemical composition

205 Fichtelgebirge, 0ct.1987

Picea abies I

4

I~

~

[

'

'

I

!

I

t e healthy

'

Io

/

declining

/

/ /

/

//

y = 0.488 x -2.116

4

/ //

r = 0.90 / o

o< 0 I/j ~0

-2 -4

o

o

/

oP. o

////

/

/

/

/ O

o<

/ /

OO

%

00~

=.,=d

0

z

LC)

y =-5.278 x + 3.782

ee

_ ~,,.

; = 091

-"

9

o

o

-2

2

--/, o

I

l

0.25

I

I

I

0.50 0.75 1.00 1.25

Total Nsoil [ mmol eq g'ldw ]

I

-2

I

I

I

0

2

4

615 Nsoil [ % o ]

Fig. 6. Correlations between nitrogen concentrations and c~15Nvalues of soil samples and between 615N values of soil and root samples from four soil horizons collected at the healthy and the declining site (O'Leary 1981). Lipide, proteins and other secondary carbon metabolites are above-average 13C depleted (Winkler et el. 1978) and contribute largely to the total carbon of photosynthetic tissues. In contrast, cellulose and hemicellulose are below-average 13C depleted. These substances contribute largely to the total carbon of nonphotosynthetic tissues (Benner et al. 1987; Tieszen and Boutton 1989). In twigs, the increase of 613C values with increasing age can be explained by two possible effects. The ratio of wood to bark biomass increases with twig age. According to the different chemical composition of wood and bark, this changing ratio could cause an apparent increase of 13C. In this study bark and wood were not analysed separately. In addition, respiration may influence the isotopic composition of non-photosynthetic tissues (O'Leary 1988). Therefore, the ~13C values of twig cuttings may increase with the time of respiratory activity. The decreasing 613C values of needles and twigs from the sun to the shade canopy is probably related to the c~13C values of COz available for assimilation at the bottom of the canopy and to the increasing level of CO2 in the atmosphere close to the ground. CO2 at the bottom of the canopy would be expected to be more ~3C depleted than at the top of the canopy due to 13C discrimination during soil respiration (Vogel 1978; Medina and Minchin 1980; Schleser and Jayasekera 1985). More important, leaf intercellular CO2 concentrations increase along the decreasing light gradient from the top to the bottom of the canopy, suggesting a greater 13C discrimination during COz assimilation in the shade-crown (Francey et el. 1985; Ehleringer et el. 1986). The calculated and measured internal CO2 concentrations support this interpretation.

The 615N values of this study are at the lower end of the nitrogen isotope ratios known from other habitats (e.g. Shearer and Kohl 1986; Vitousek et el. 1989; Virginia and Delwiche 1982). The 15N abundance in forest soils is, in general, lower than in agricultural soils (Riga et el. 1971) and also the lSN abundance in trees is known to be below average (Virginia and Delwiche 1982). Low nitrogen losses in forest ecosystems are considered to be mainly responsible for this observation since all major pathways of nitrogen loss (denitrification, ammonia volatilization and nitrate leaching) cause a 1sN enrichment of the remaining nitrogen (Shearer and Kohl 1986). In addition, gaseous, aerosolic and liquid anthropogenic nitrogen depositions may have influenced the bulk 615N values of this study similar to the way sulphur depositions influence the bulk sulphur isotope ratio of ecosystems (Krouse 1989). Anthropogenic nitrogen depositions contribute significantly to the nitrogen nutrition of all compartments of Norway spruce (Schulze and Gebauer 1989), but the sum of nitrogen depositions was found to be similar for the two stands investigated (Eiden 1989). Therefore, the effects of 61SN values by site, organ and organ age observed in this study are considered to be primarily the result of isotopic discrimination in certain steps of the nitrogen cycle. 615N of soil samples increased with soil depth, which is typical for undisturbed, non-cultivated soils (Mariotti et el. 1980). This pattern reflects an isotopic discrimination of 15N during mineralization of soil nitrogen (Nadelhoffer and Fry 1988) as indicated by the decrease in total soil nitrogen. Root nitrogen was always depleted in l SN compared to soil nitrogen, also indicating discrimination associated with mineralization and nitrification (Shearer et al. 1974). The correlation between 61SN in soil and roots indicates that roots preferentially used

206 nitrogen from the soil horizons they were in for their own biomass production. The bulk of the nitrogen in the above-ground biomass, however, appears to have been taken up f r o m the organic soil layer. This was reflected in the isotopic composition of roots in the mineral soil being significantly different from the nitrogen isotope ratio of above-ground biomass. The highest root density is in the uppermost organic layer of these spruce stands (Schneider et al. 1989). The increasing difference of 615N in roots and soil with increasing soil depth m a y have two reasons: (1) Roots in the mineral soil are composed of nitrogen which originates partly from the surrounding mineral soil and partly f r o m the organic layer or (2) roots in the organic layer and in the mineral soil use different mineral nitrogen forms with different 615 N values. It was suggested that at the low p H of these sites, roots f r o m the organic layer m a y use a m m o n i u m more than nitrate, but in the mineral soil nitrate rather than a m m o n i u m (Schulze 1989). A m m o n i u m in the soil solution should have higher ~15N values than nitrate (Shearer et al. 1974). A predominant use of a m m o n i u m would support the hypothesis that nitrate leaching to the ground-water and the imbalance of magnesium and nitrogen during decline is related to a m m o n i u m rather than nitrate uptake. Twigs were more depleted of 15N than needles. Similar gradients between w o o d y tissues and leaves have been found for the 3~SN values of other tree species (Virginia et al. 1989). The gradient is opposite to that found for the 613C values of the same samples. Compounds in plant tissues that are the most depleted in ~SN should be products from the end of a metabolic chain, similar to the pattern observed in above-average laC-depleted secondary carbon metabolites (see above). The nitrogen in twigs and other sink tissues for organic N m a y therefore preferentially originate f r o m compounds recycled f r o m senescing compartments. Decreases in 61SN values from current-year to 2- or 3-year-old needles reflect the fact that younger needles are more anabolically active than older needles, and thus represent a greater sink for assimilated nitrogen. Conversely, increases in 615N values in older needles m a y reflect the onset of catabolic activity and a subsequent export of N-containing compounds. The agedependent differentiation of 615N values in needles from the healthy and the declining stand presumably reflects reduced anabolic activity and an earlier onset of catabolic decay in needles of the declining stand. Furthermore, it leads to significantly different nitrogen isotope ratios in the litter shed f r o m trees of the two stands. Based on these different nitrogen isotope ratios in the litter a further differentiation of the 6~SN values in the soil of the two stands can be expected in the future. The results of this study confirm that the carbon metabolism of the healthy and the declining stand was not affected by decline symptoms. The 6~SN values, however, were affected by decline s y m p t o m s and this allows insight into important metabolic steps during the assimilation of nitrogen f r o m the soil, its allocation within the plant, and its return back to the soil.

Acknowledgements. Skilful technical assistance in mass spectrometry by Andrea Beran and valuable help in sampling and sample preparation by Petra Dietrich, Martina St6cker and Dr. Reiner Zimmermann, in statistical analysis by Dr. Hermann Heilmeier, and in improvement of the English by Prof. R.K. Monson is gratefully acknowledged.

References Benner R, Fogel ML, Sprague EK, Hodson RE (1987) Depletion of 13C in lignin and its implications for stable carbon isotope studies. Nature 329:708-710 Brugnoli E, Hubick KT, Caemmerer S von, Wong SC, Farquhar GD (1988) Correlation between the carbon isotope discrimination in leaf starch and sugars of C3 plants and the ratio of intercellular and atmospheric partial pressure of carbon dioxide. Plant Physiol 88 : 1418-1424 Ehleringer JR, Field CB, Lin ZF, Kuo CY (1986) Leaf carbon isotope and mineral composition in subtropical plants along an irradience cline. Oecologia 70: 520-526 Eiden R (1989) Air pollution and deposition. Ecol Stud 77:57-100 Farquhar GD, O'Leary MH, Berry JA (1982) On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Aust J Plant Physiol 9:121-137 Francey R J, Gifford RM, Sharkey TD, Weir B (1985) Physiological influences on carbon discrimination in huon pine (Lagarostrobosfranklinii). Oecologia 66:211-218 Kaupenjohann M (1989) Effects of acid rain on soil chemistry and nutrient availability in the soil. Ecol Stud 77 : 297-341 Krouse HR (1989) Sulfur isotope studies of the pedosphere and biosphere. Ecot Stud 68:424-444 Lange OL, Heber U, Schulze ED, Ziegler H (1989) Atmospheric pollutants and plant metabolism. Ecol Stud 77:238-273 Mariotti A, Pierre D, Vedy JC, Bruckert S, Guillemot J (1980) The abundance of natural nitrogen 15 in the organic matter of soils along an altitudinal gradient (Chablais, Haute Savoie, France). Catena 7: 293-300 Mariotti A, Mariotti F, Champigny ML, Amarger N, Moyse A (1982) Nitrogen isotope fractionation associated with nitrate reductase activity and uptake of NO~- by pearl millet. Plant Physiol 69 : 880-884 Medina E, Minchin P (1980) Stratification of 6t3C values of leaves in Amazonian rain forest. Oecologia 45 : 377-378 Nadelhoffer KJ, Fry B (1988) Controls on natural nitrogen-15 and carbon-13 abundances in forest soil organic matter. Soil Sci Soc Am J 52:1633-1640 Norusis M (1987) SPSS/PC+. For the IBM PC/XT/AT. Chicago O'Leary M (198l) Carbon isotope fractionation in plants. Phytochem 20 : 553-567 O'Leary M (1988) Carbon isotopes in phytosynthesis. BioSci 38 : 328-336 Oren R, Schulze ED, Werk KS, Meyer J, Schneider BU, Heilmeier H (1988a) Performance of two Picea abies (L.) Karst. stands at different stages of decline. I. Carbon relations and stand growth. Oecologia 75:25 37 Oren R, Schulze ED, Werk KS, Meyer J (1988b) Performance of two Picea abies (L.) Karst. stands at different stages of decline. VII. Nutrient relations and growth. Oecologia 77:163 173 Riga A, Van Praag HJ, Brigode N (1971) Rapport isotopique naturel de l'azote dans quelques sols forestiers et agricoles de belgique soumis a divers traitements culturaux. Geoderma 6:213 222 Rundel PW, Ehleringer JR, Nagy KA (eds) (1989) Stable isotopes in ecological research. Ecol Stud 68 SAS Institute Inc. (1985) SAS User's Guide: Statistics, Version 5 Edition. Cary, NC Schleser GH, Jayasekera R (1985) 6~3C-variations of leaves in for-

207 ests as an indication of reassimilated CO2 from the soil. Oecologia 65: 536-542 Schneider BU, Meyer J, Schulze ED, Zech W (1989) Root and mycorrhizal development in healthy and declining Norway spruce stands. Ecol Stud 77:370-391 Schulze ED (1989) Air pollution and forest decline in a spruce (Picea abies) forest. Science 244:776-783 Schulze ED, Gebauer G (1989) Aufnahme, Abgabe und Umsatz yon Stickoxiden, NH2 und Nitrat bei Waldb/iumen, insbesondere der Fichte. In: Reuther M, Kirchner M (eds) 1. Statusseminar der PBWU zum Forschungsschwerpunkt ,,Waldsch~iden", GSF M iinchen, pp 119-133 Schulze ED, Lange OL, Oren R (eds) (1989) Forest decline and air pollution. A study of spruce (Picea abies) on acid soils. Ecol Stud 77 Schulze ED, McCracken I, Zimmermann R, Benecke U (1990) Performance of two Picea abies (L.) Karst. stands at different stages of decline. VIII. Photosynthesis and nutrition under polluted and clean air conditions of Germany and New Zealand. Oecologia 82:158-161 Shearer G, Kohl DH (1986) N2-fixation in field settings: estimations based on natural lSN abundance. Aust J Plant Physiol 13 : 699-756 Shearer G, Duffy J, Kohl DH, Commoner B (1974) A steady-state model of isotopic fractionation accompanying nitrogen transformations in soil. Soil Sci Soc Amer Proc 38:315-322 Tieszen LL, Boutton TW (1989) Stable carbon isotopes in terrestrial ecosystem research. Ecol Stud 68:167-195

Virginia RA, Delwiche CC (1982) Natural 15N abundance of presumed N2-fixing and non-N2-fixingplants from selected ecosystems. Oecologia 54:317-325 Virginia RA, Jarrell WM, Rundel PW, Shearer G, Kohl DH (1989) The use of variation in the natural abundance of 15N to assess symbiotic nitrogen fixation by woody plants. Ecol Stud 68 : 375394 Vitousek PM, Shearer G, Kohl DH (1989) Foliar 15N natural abundance in Hawaiian rainforest: patterns and possible mechanisms. Oecologia 78 : 383-388 Vogel JC (1978) Recycling of carbon in a forest environment. Oecol Plant 13 : 89-94 Winkler F J, Wirth E, Eatzko E, Schmidt HL, Hoppe W, Wimmer P (1978) Influence of growth conditions and development on c~3C values in different organs and constituents of wheat, oat and maize. Z Pflanzenphysiol 87:255-263 Ziegler H (1979) Diskriminierung yon Kohlenstoff- und Wasserstoffisotopen: Zusammenhfinge mit dem Photosynthesemechanismus und den Standortsbedingungen. Ber Deutsch Bot Ges 92:169-184 Zimmermann R (1990) Photosynthese und Transpiration von Picea abies (L.) Karst. bei unterschiedlichem Ern/ihrungszustand im Fichtelgebirge (Nordostbayern). Dissertation, Bayreuth Zimmermann R, Oren R, Schulze ED, Werk KS (1988) Performance of two Picea abies (L.) Karst. stands at different stages of decline. II. Photosynthesis and leaf conductance. Oecologia 76:513-518