Observing and quantifying variability in natural processes and the impact of human interventions is of primary importance for three reasons:

• It provides a key diagnostic of how climatic factors and societal dynamics affect exchange fluxes and provides information that is needed to develop and validate comprehensive process-based carbon cycle models that include human dynamics.

• Several terrestrial and oceanic carbon cycle components effectively contain a longer-term "memory"—that is, their present state is partly a result of past natural and human perturbations. For example, enhanced fire frequencies or management strategies during a drier past period will be reflected in the age structure of a present forest, or the carbon content in particular oceanic deepwater masses will reflect climate-driven variations in deepwater formation in the past.

• Climate variability often tends to mask the slow, longer-term signals in the carbon cycle that are of primary interest; for example, many studies have shown that the uptake rates of the terrestrial biosphere or the ocean are highly variable. In this context, climate variability constitutes "noise" from which the signals must be distinguished. Indeed, a significant fraction of the uncertainty in the global budget in Table 2.1 can be traced to incomplete quantification of possible climate perturbations in the pertinent observations.

On a global scale, climate-driven variability can be inferred from atmospheric time series of CO2 and associated variables, such as 13C/12C and O2/N2 ratios. Continental or ocean basin—scale variations can be detected on timescales of up to several years by means of top-down inversions of atmospheric CO2 concentration measurements (Heimann et al., Chapter 8, this volume). It is difficult, however, to separate the terrestrial signals from the ocean signals using these large-scale approaches (Greenblatt and Sarmiento, Chapter 13, this volume). On land and in the oceans, local-scale direct observations of variability exist from in situ flux measurements at a few time-series stations during the past few years. On a regional scale, however, an observational gap exists. The present atmospheric network is not dense enough to resolve carbon sources and sinks on regional scales by atmospheric inversion, while the upscaling of measured in situ oceanic and terrestrial carbon fluxes is extremely difficult. On land, the large heterogeneity of terrestrial ecosystems and complex atmospheric transport patterns resulting from topography make it difficult to scale up local measurements. In the oceans, the complex interplay between physical and biological controls on sea surface pCO2 and the dynamics of air-sea exchange complicate the extrapolation of the in situ observations.

Interannual variability in climate leads to large changes in atmospheric temperature and rainfall patterns, as well as changes in ocean surface temperatures and circulation. All of these changes can have dramatic effects on biological productivity and complex effects on CO2 exchanges with the atmosphere on land and in the oceans. For example, during El Niño events, the warming of ocean surface waters and the reduction in biological productivity in the Equatorial Pacific Ocean should lead to enhanced out-gassing of CO2. Since the upwelling of carbon-rich deepwaters, which release CO2 to the atmosphere, is reduced during El Niños, however, the net effect is a significant reduction in the outgassing of CO2 in this region. On the other hand, warmer temperatures and anomalous rainfall patterns during El Niños can lead to increased terrestrial biosphere respiration, forest fires, and droughts. The timing of these effects and the teleconnections between them have a direct impact on atmospheric CO2 concentrations that is still not completely understood.

Overall, the general consensus is that the interannual variability in air-sea CO2 fluxes is smaller than that of terrestrial CO2 fluxes, but the exact amplitude and spatial distributions remain uncertain (Greenblatt and Sarmiento, Chapter 13, this volume). To better determine these signals, an expanded network of time-series CO2 measurements must be maintained for the atmosphere, oceans, and land systems. These measurements, together with intensive process studies, a better use of satellite data, atmospheric observations, and rigorously validated models, will help us better understand the current global carbon cycle and how it is evolving over time.

0 0

Post a comment