The observed increase in atmospheric C02 concentration (e.g., Conway et al., 1994; Keeling et til, 1995), is a balance between anthropogenic inputs (such as fossil-fuel combustion and land-use change) and natural responses. The natural responses may arise directly from increasing atmospheric CO, concentration (e.g., extra dissolution in seawater) but may result from many other factors. An understanding of these natural responses, their sensitivities to human impacts, and their likely future trajectory forms a critical part of a well-founded projection of climate change and an attempt to manage that change. A first step is to quantify these responses via the net fluxes into various reservoirs. This has been the topic of considerable effort in the climate science and biogeo-chemistry communities particularly in the last decade. Summaries of the state of the science, really snapshots of evolving knowledge, are given, for example, by Schimel et al (1995). I should stress at the outset that quantifying current fluxes is an early step in a process of understanding and should be considered as a measurement of the underlying behavior somewhat akin to remote sensing. However, even at this step, divergence and controversy have been the rule rather than the exception. In Sec. 2 I will give one view of the current state of the most basic question in this area; the relative roles of oceanic and terrestrial fluxes in balancing the atmospheric CO, budget. I will review and compare several somewhat independent lines of evidence. I will present a synthesis of these lines of evidence, although I will not attempt to draw them into one overarching framework. I will focus on the role of large-scale constraints with either gross or zero spatial information.

If the solution of the long-term budget can be regarded as understanding the basic state of the global carbon cycle, then interannual variability is a clue to the sensitivity of that state. The task is to understand the processes that control year-to-year variations in flux but even the task of estimating these fluxes is at the limits of current capacity. In Sec. 3 I will compare estimates of this variability from models of the processes involved and atmospheric inferences to demonstrate this inconsistency. Finally, I will sketch a modeling framework that may help address this inconsistency.

0 0

Post a comment