19&0 1901 1982 19&3 1984 1985 1986 T&87 1988 1989 1990 1991 ^992 1993 1994 1 995 1 996 1997 T anomaly ----Û10 weighted T anomaly -O— Flux anomaly

Figure 3.6. Relationship between inferred COj flux anomalies and temperature anomalies north of 5 N for 1980-1997, The temperature anomalies and the Qjq-weighted temperature anomalies have been weighted by the net primary productivity for the region. Qjn is a measure of the temperature sensitivity of decomposition rates. {Taken from Randerson et al., 1999.)

a result consistent with My ne ni et al. (1997), whereas those in the fall are positively correlated with temperature, consistent with Gouldcn et aL (1998). As has been argued above, although the competition favors net uptake in the present climate, further increases in temperature may change the dynamics of the competition.

Maiepwan, 3amwmenHbifi aBTopcKnw npaBOM

At middle and low latitudes, the variations in precipitation are likely to be as important as the variations in temperature in causing interannual variations in photosynthesis and respiration. This is demonstrated by Goulden et al (19%), who suggest that the NH land sink may vary interannually by ±1 PgC/v because of seasonal climate perturbations.

3,7 The Human Dimension

In addition to burning fossil fuels, humans affect the carbon cycle via direct modification of the terrestrial biosphere. Deforestation, occurring now in the Tropics, generally replaces high-density biomass (forests) with low-density biomass (crops and pastures). If the mode of deforestation includes burning, a fraction of the above ground biomass carbon is released immediately to the atmosphere. The fraction released depends on the characteristics of the fuel - its size and moisture content - as well as on the repetition of the burns, and it ranges from <5% for large logs to 35% for multiple burns (Carvalho et al., 1998). The detritus left on site adds to the decomposition pool, and exposed soils are subject to rapid oxidation. The turnover time of carbon in the tropics is 5-10 years, and so deforestation contributes to a sustained release of CO? several years after the initial clearing.

Deforestation does not occur uniformly across a landscape but rather is concentrated in small areas (Skole and fucker, 1993). In Brazil, the distribution of the cleared areas shows that > 50% of the deforested plots arc > 100 ha and <15% of the deforested plots are >1000 ha (INPF, 1999). These plot sizes are large and represent the interests of commercial and large-scale enterprises. The analysis also shows that the deforestation rate in Brazil varies by a factor of 2 from year to year. The fluctuation is related to the Brazilian economy and land and timber prices. Hence, coarse-resolution (I x 1

ha) satellite imagery will likely overestimate the deforestation rate, and statistical extrapolation of a small-area survey to an ecosystem or a country is subject to bias. Our preliminary reanalysis of the deforestation source suggests that the annual CO2 release is smaller, by a factor of 2, than the Schimel et al. (1996) estimate; however, a firmer estimate cannot be made until the high-resolution satellite data for Africa, South Asia and Southeast Asia have been analyzed.

Analysis of the contemporary deforestation focuses on the Tropics. Over the past 100 years, the expansion of agriculture, mainly in the middle latitudes, has contributed a cumulative carbon source that is half the fossil fuel source. The cleared areas are now under management or are abandoned, so year-by-year carbon balance is now close to neutral in some areas or is showing slow accumulation in above- and belowground carbon in other areas (Post et al., 1997). The recovery is incomplete, so the present-day carbon inventory in mid-latitudes is still less than that during the prein dust rial era (DeFries et al., 1999). This recovery, when coupled with the inadvertent stimulation of photosynthesis via elevated CO? levels and nitrogen deposition, results in a net carbon uptake in the middle latitudes.

The preceding discussion highlights the fact that the magnitude and direction of the carbon flux are tied to the history of land use, which is driven by a variety of forces in addition to population pressures. Recent deforestation leads to a transient carbon

Maiepuan, 3auiML4©HHbii?i asTopcKUM npaBOM

source to the atmosphere, and abandonment of past clearings leads to transient carhon sequestration by the biosphere. Viewing the contemporary carbon budgets in short 5-or 10-year snapshots may thus obfuscate the causes of and thus the mitigation strategies for the atmospheric CO2 increase,

3.8 Implications for the Future

The natural carbon cycle involves large fluxes (^10* PgC/y) into and out of the biosphere, and large fluxes PgC/y) into and out of the oceans. At equilibrium, these fluxes cancel. The contemporary carbon sink I02 PgC/y) is a result of the incomplete cancellation of these opposing fluxes and represents a —1% enhancement of the CO2 fluxes out of the atmosphere. The large opposing fluxes are regulated by climate and are hence sensitive to climate perturbations.

In this chapter, we have examined the interannual variations of atmospheric CO? concentrations and have shown that the CO? sinks are climatically sensitive. There is currently carbon absorption by the high-latitude biosphere because photosynthesis enhancement is greater than the respiration enhancement. It is not likely that this sink will continue with global warming. Our reliance on the biosphere can lead to carbon sources at one time and carhon sinks at another. How atmospheric CO2 levels, and climate, w ill evolve will depend on the delicate competition between direct human-induced and indirect climate-induced alterations of the biosphere. Strategies for deliberate manipulation of the carbon cycle must take into account the evolving climatic envelope.

0 0

Post a comment