Ecological Archives E091-028-A1

Kristiina Karhu, Hannu Fritze, Kai Hämäläinen, Pekka Vanhala, Högne Jungner, Markku Oinonen, Eloni Sonninen, Mikko Tuomi, Peter Spetz, Veikko Kitunen, and Jari Liski. 2010. Temperature sensitivity of soil carbon fractions in boreal forest soil. Ecology 91:370–376.

Appendix A. Description of litter chemical characteristics, SOC changes over the incubation period, respiration rate of the combined POM and MOM fraction, and the 14C activity of the younger SOC fraction in mineral soil.

Litter chemical characteristics

Litter at sites similar to our study sites is of rather poor quality, which is typical for boreal forests (Table A1). The nitrogen (N) concentration of needle, branch or fine root litter (0.3 to 0.86 %) classifies as low or intermediate in a global context (cf. Parton et al. 2007), and the N concentration was still considerably lower in stem wood (0.06 to 0.11 %). The organic chemical composition of litter was typical for coniferous trees, characterized by a low concentration (less than 26 %) of compounds soluble to a polar or non-polar solvent (cf. Moore et al. 1999, Tuomi et al. 2009).

TABLE A1. Chemical quality of litter at boreal forest sites similar to those studied. Extractives refer to compounds soluble in a polar or non-polar solvent.

Litter component

N (%)

 

Cellulose (%)

 

Lignin (%)

 

Extractives (%)

 

References (N; organic chem. composition)

 

Mean

SD

Mean

SD

Mean

SD

Mean

SD

 

Scots pine

                 

Foliage

0.50

0.31

49

2

25

3

26

5

n = 37,  Berg et al. (1991); n = 46, Berg et al. (1991)

Branches and coarse roots

0.33

n.a.

47

3

43

1

10

3

n = 25, Vavrova et al. (2008)

Fine roots

0.30

0.08

57

4

23

3

19

5

n = 13,  Berg et al. (1991

Stem

0.06

n.a.

69

1

26

2

4

1

Finer and Kaunisto (2000); n = 2, Hakkila 1989

                   

Norway spruce

                 

Foliage

0.86

0.49

47

0

37

2

15

3

n = 29,  Berg et al. (1991); n = 4, Berg et al. (1991)

Branches and coarse roots

0.30

n.a.

61

3

35

3

3

0

Sah and Ilvesniemi (2008); n = 2, Hakkila 1989

Fine roots

0.44

0.12

55

6

25

4

20

5

n = 11,  Berg et al. (1991)

Stem

0.11

n.a.

69

2

30

2

1

0

Kaakinen et al. (2007); n = 2, Hakkila 1989

 

SOC changes over the incubation period

The production rate of CO2 decreased substantially in the organic layer samples over the 18-month incubation period (Fig. A1). This rate decreased also in the topmost 0–15 cm mineral soil layer samples, whereas in the deeper mineral soil layer samples (15–30 cm) the rate remained at the same level over the entire experiment. After the incubation, the organic layer samples had about 83 % of the original SOC remaining, while the topmost mineral soil layer samples had 90 to 92 % and the deeper mineral soil layer samples 94 to 95 % (Fig. A2). These trends indicate that, firstly, the organic layer samples contained larger amounts of labile SOC compounds than the mineral soil samples did and, secondly, the amounts of labile SOC decreased towards the deeper layer inside mineral soil.

The quality of SOC changed also over the incubation period. In the organic layer samples, the concentration of Klason lignin increased while the concentration of acid hydrolysable SOC decreased (Fig. A3a). The concentration of ethanol soluble SOC did not change, and the concentration of water soluble SOC remained unchanged at the spruce site but increased at the pine site (Fig. A3b). Still, the quality of water soluble SOC decreased in the organic soil layers of both sites, as indicated by a decreased concentration of sugar monomers and increased concentration of lignin in the water soluble SOC fraction (Fig. A4). In the mineral soil layers, the changes in the quality of SOC were slow (see, e.g., the trends of particulate SOM and mineral-associated SOM in Fig. A5). This is in agreement with the CO2 production measurements (Fig. A1) which showed that the mineral soil layers contained more stable SOC than in the organic layers did.

FigA1
 
   FIG. A1. CO2 production of the soil samples over the incubation period (closed symbols represent the spruce site and open symbols the pine site; triangles represent the organic soil layer, dots the 0–15 cm mineral soil layer and diamonds the 15–30 cm mineral soil layer.

 

 

FigA2
 
   FIG. A2. The fraction of the original SOC remaining in the soil samples over the incubation period (closed symbols represent the spruce site and open symbols the pine site; triangles represent the organic soil layer, dots the 0–15 cm mineral soil layer and diamonds the 15–30 cm mineral soil layer.

FigA3
 
   FIG. A3. The quality SOC in the soil samples of the organic soil layers over the incubation period, (a) Klason lignin (diamonds) and acid hydrolysable fraction (dots), (b) water soluble fraction (diamonds) and ethanol soluble fraction (dots). Closed symbols represent the spruce site and open symbols the pine site.

FigA4
 
   FIG. A4. The quality of water soluble SOC fraction in the soil samples of the organic soil layers over the incubation period, sugar monomers (diamonds) and lignin (dots). Closed symbols represent the spruce site and open symbols the pine site.

Respiration rate of the combined POM and MOM fraction, and the 14C activity of the younger SOC fraction in mineral soil

Estimates for the respiration rate of the combined POM and MOM fraction were needed to calculate the 14C activity of the younger SOC fraction in mineral soil (Eq. 1). We estimated these rates based on the mass loss of the combined POM and MOM fraction over the entire incubation period (Fig. A5). We applied these estimates for the periods of collecting CO2 to measure the 14C activity of the respired carbon because these mass loss rates did not change significantly over the incubation period.

The mass loss rate of the combined POM and MOM fraction was very low (Fig. A5). Consequently, the respiration rate estimates of this combined fraction were associated with a substantial uncertainty. To account for this uncertainty in a Monte Carlo analysis, we derived a probability distribution of the estimates (Fig. A6) based on the standard error of the slope of the linear regression (SEb) fitted to the mass loss measurements (Fig. A5). The 95 % confidence limits of the slope were calculated as CLb = z0.95 * SEb, where z0.95 is the 95 % fractile of Student's t distribution (2.571, n = 7). To account for physical limitations, we restricted the respiration rate estimates of the combined POM and MOM fraction between zero and the rate of the total respiration. The Monte Carlo analysis showed that despite of the large uncertainty, the estimates for the respiration rate of the combined POM and MOM fraction were reliable enough to calculate the 14C activity of the younger SOC fraction in each mineral soil layer (Fig. A7) and detect statistically significant differences in temperature sensitivity of decomposition between the younger and older SOC fractions (Fig. 1).

We accounted for the uncertainty in the respiration rate of the combined POM and MOM fraction (Fig. A6) in a Monte Carlo analysis together with uncertainty in every other factor that affected the 14C activity of the younger SOC fraction and calculated probability densities for the 14C activity estimates (Fig. A7). We used these probability densities, plus probability densities of every other factor that affected the estimates for the Q10 values of the SOC fractions, to estimate 95 % confidence limits for estimates of the Q10 values (Fig. 1).

FigA5
 
   FIG. A5. Mass of carbon in the combined POM and MOM fraction of the mineral soil layers over the incubation period (closed dots 0–15 cm mineral soil layer of the spruce site, closed diamonds 15–30 cm layer of the spruce site, open dots 0–15 cm layer of the pine site, open diamonds 15–30 cm layer of the pine site). The slope estimates and their standard errors, which we used to derive probability densities of the estimates for the mass loss of this combined fraction (Fig. A6), were 1.09 ± 0.73, 0.79 ± 0.51, 1.22 ± 1.45, 0.86 ± 0.88
µg C g‑1 soil d.w. day-1, respectively.

FigA6
 
   FIG. A6. Probability densities of the estimates for the total respiration at the end of the incubation (solid lines) and the respiration of the combined POM and MOM fraction (dashed lines) in the mineral soil samples (blue lines 0–15 cm mineral soil layer of the spruce site, green lines 15–30 cm layer of the spruce site, red lines 0–15 cm layer of the pine site, grey lines 15–30 cm layer of the pine site). For the 15–30 cm mineral soil layer of the spruce site (the green lines), the mean estimate of the POM and MOM respiration rate was higher than the total respiration rate. This discrepancy is a result of the low POM and MOM concentration in this soil layer (Fig. A5) and, consequently, an imprecise estimate for the mass loss of the combined POM and MOM fraction. The larger uncertainty of this estimate is reflected in the wider probability density for the 14C activity of the younger fraction in this soil layer (Fig. A7).

FigA7
 

   FIG. A7. Probability densities of the estimates for the 14C activity of the younger SOC fraction in the mineral soil layers.


LITERATURE CITED

Berg, B., H. Booltink, A. Breymeyer, A. Ewertsson, A. Gallardo, B. Holm, M.-B. Johansson, S. Koivuoja, V. Meentemeyer, P. Nyman, J. Olofsson, A.-S. Petterson, A. Reurslag, H. Staaf, I. Staaf and L. Uba. 1991. Data on needle litter decomposition and soil climate as well as site characteristics for some coniferous forest sites, Part II, Decomposition data. Report 42, Swedish University of Agricultural Sciences, Department of Ecology and Environmental Research, Uppsala, Sweden.

Finér, L., and S. Kaunisto. 2000. Variation in stemwood nutrient concentrations in scots pine growing on peatland. Scandinavian Journal of Forest Research 15:424–432.

Hakkila, P. 1989. Utilization of residual forest biomass. Springer-Verlag, Berlin, Germany.

Kaakinen, S., P. Saranpaa, and E. Vapaavuori. 2007. Effects of growth differences due to geographic location and N-fertilisation on wood chemistry of Norway spruce. Trees - structure and functions 21:131–139.

Moore, T. R., J. A. Trofymow, B. Taylor, C. Prescott, C. Camire, L. Duschene, J. Fyles, L. Kozak, M. Kranabetter, I. Morrison, M. Siltanen, S. Smith, B. Titus, S. Visser, R. Wein, and S. Zoltai. 1999. Rates of litter decomposition in Canadian forests. Global Change Biology. 5:75–82.

Parton , W., W. L. Silver, I. C. Burke, L. Grassens, M. E. Harmon, W. S.Currie, J. Y. King, E. C. Adair, L. A. Brandt, S. C. Hart, and B. Fasth. 2007. Global-scale similarities in nitrogen release patterns during long-term decomposition. Science 315:361–364.

Sah, S. P., and H. Ilvesniemi. 2007. Interspecific variation and impact of clear-cutting on natural 15N abundance and N concentration in the needle-to-soil continuum of a boreal conifer forest. Plant, Soil and Environment 53:329–339.

Tuomi, M., T. Thum, H. Järvinen, S. Fronzek, B. Berg, M. Harmon, J. A. Trofymow, S. Sevanto. and J. Liski. 2009. Leaf litter decomposition - Estimates of global variability based on Yasso07 model. Ecological Modelling. In press.

Vavrova, P., T. Penttilä, and R. Laiho. 2009. Decomposition of Scots pine fine woody debris in boreal conditions: Implications for estimating carbon pools and fluxes. Forest Ecology and Management 257:401–412.


[Back to E091-028]