Net CarbonGain

C-losses

C-fluxes Time

Gross-Primary-Productivity-GPP (= photosynthesis)

100%

Net-Primary-Productivity-NPP 50% (= plant growth)

Net-Ecosystem-Exchange-NEE Net-Ecosystem-Productivity-NEP (= change in litter & organic layer)

Net-Biome-Productivity-NBP <ii

(= change in charcoal & résiliant humus)

respiration of plants respiration of microbes, disturbance by trends (C02 & climate change)

episodic disturbancees (harvest & fire)

day year

10-year

>100 year accumulates organic material in the organic layer. This material is being decomposed to a level, that the local decomposers cannot digest this material any further, and it thus builds up an organic layer with increasing thickness. Groundfires will burn this layer and return it as C02 to the atmosphere, but groundfires will also produce charcoal with higher longevity. On a landscape level, the organic layer will probably be carbon neutral, although a specific plot shows a distinct rate of accumulation, depending on the time since the last fire. The only component, which accumulates at the landscape level is most likely charcoal, but also this component appears to undergo a decomposition process (Czimczik et al., 2000). Nevertheless, some of this carbon is stabilized in soils, and based on this fraction Schulze et al. (1999) calculated a rate of NBP in the order of 13-130 mmol m"2 y~'.

The accumulation of carbon in the organic layer after fire points at a basic problem of the simplified flux scheme of Schulze and Heimann (1998), namely, that intermediate pools exist at each level with different mean residence time, and depending on the level of spatial integration these pools may or may not average out. The problem is illustrated in Figure 7, where an inventory-process type approach (in contrast to eddy covariance flux measurements) was chosen to calculate the carbon sink capacity of European forest ecosystems. If NPP was plotted against C-mineralization, then NPP minus C-mineral-ization would represent NEP at the plot scale (Schulze et al.,

2000). Figure 7 shows a surprisingly large rate of C-sequestra-tion by plots ranging across Europe, and including deciduous and coniferous stands of different ages and sites. The range depicted for C-mineralization represents different methods of assessing C-mineralization. This large rate of NEP includes wood growth at each plot. However, on a landscape scale, wood would be harvested in managed European forests, and this compartment would thus be carbon neutral. If this fraction is removed, the rate of carbon sequestration, here termed NBP, would only be a small fraction of NEP measured at the plot scale. We are not sure, if this rate of NBP is permanent. Harrison et al. (2000) determined the mean residence time of the organic layer in these habitats, and found that this layer may be very short lived (L + F layer = 5 to 6 years), but also the following 0-5 cm layer (A + Oh horizon) was not very long lived and showed a mean residence lime of 35 years in France and 340 years in North Sweden, which is the same order of magnitude as that of the tree cover.

The implications of this observation are manifold. The "Kyoto forest" is planned to contain 0.5 ha plots, which will sequester carbon at the rate as it is shown in Figure 7a until harvest. However, if the "Kyoto forest" would extend across a landscape, the forest would most likely be carbon neutral, and only a very small fraction will be stabilized as "recalcitrant" soil organic matter. The other implication is that the flux measurement at the plot scale

Pinus sylvestris

Zotino, July 1996

Pinus sylvestris

Zotino, July 1996

FIGURE 6 Average daily courses of ecosystem C02 exchange of (a) Siberian pristine pine forest and natural regeneration after fire, (b) regrowth after logging, and (c) a bog, all located near Zotino, Central Yenisey River, July 1996 (Schulze et al„ 1999).

does not measure the long-term carbon sequestration, but mainly tree growth at the plot scale. Also, the change in C-pools at the plot scale does not represent the C-sequestration at the landscape scale, because some of the compartments remain at an average level as long as management is constant. However, changes in management may cause short-term variations in carbon pools, which do not reflect the long-term C-balance

AC = ACj biomass

The change in biomass takes into account carbon that has not yet been stabilized to a degree for which it could be counted as long-term C-sequestration. After tree death, this carbon would still decompose, and at a landscape level remain carbon neutral.

The change in carbon stocks of the biomass has a mean residence time related to the change in harvest by management, i.e., if harvest is delayed by 20 years (commitment period of the Kyoto protocol from 1990 to 2010) and timber demand is

CL CL Z

CL CL Z

FIGURE 7 (a) NPP in carbon units (NPPC) as related to C-mineraliza-tion and (b) litter fall as related to C-mineralization. The hatched areas in the top panel equals net ecosystem productivity (NEP) and in the bottom panel the hatched area equals net biome productivity (NBP). Abbreva-tions refer to study sites: boreal forest in Sweden Äheden, northern temperate forest of Denmark Sfogaby {Picea) and Gribskov (Fagus), temperate forest in Germany Wa/dstein (Picea) and Sc/iacht (Fagus), in Czech Republik Micetin (Picea) and Jezery (Fagus), in France Ai/bure Picea and Fagus, and montan mediterranean forest in Italy Co/lelongo (Fagus) and Monte di Mezzo (Picea) (Schulze et al., 2000).

C mineralisation [t C ha" yr ]

FIGURE 7 (a) NPP in carbon units (NPPC) as related to C-mineraliza-tion and (b) litter fall as related to C-mineralization. The hatched areas in the top panel equals net ecosystem productivity (NEP) and in the bottom panel the hatched area equals net biome productivity (NBP). Abbreva-tions refer to study sites: boreal forest in Sweden Äheden, northern temperate forest of Denmark Sfogaby {Picea) and Gribskov (Fagus), temperate forest in Germany Wa/dstein (Picea) and Sc/iacht (Fagus), in Czech Republik Micetin (Picea) and Jezery (Fagus), in France Ai/bure Picea and Fagus, and montan mediterranean forest in Italy Co/lelongo (Fagus) and Monte di Mezzo (Picea) (Schulze et al., 2000).

expectancy of the trees, then also this layer cannot be counted as a long-term carbon store.

This consideration shows that a terrestrial sink may be a very short term phenomenon. Indications that this is indeed the case come from flux measurements. The net carbon balance (NEP) of forests across Europe is not related to assimilation, which is highly adaptable, but determined by soil respiration, which is mainly a function of temperature and water availability (Vatentini et al., 2000).

0 0

Post a comment