| |
|---|---|
 (794KB)
|
11,228,000 Yen (FY1998, 3,191,000 Yen)
Seasonal variations of CO2 emissions from the land biota and oceans are re-trieved by use of an inverse method. Concentrations of CO2, computed by an atmospheric chemical transport model (NIRE-CTM-93), are fitted to the observed CO2 Concentrations in terms of annual mean concentrations between 1984 to 1985, one-year period com, po-nents, half-year components, and a global trend averaged over the years 1979 to 1996. The emissions due to fossil fuel combustion, those from land biota consisting of thirteen land areas, and those from the ocean consisting of twelve oceanic areas, are considered. The problem is solved by a least squares method, using singular value decomposition. A prior estimate of the emissions due to fossil fuel combustion is used as a tight constraint. When the annual mean emission, the net flux from the land biota during the growing season, and the net flux from the ocean are loosely constrained to zero, the solution indicates that areas consist of two groups. That is, those areas with relatively reliable estimates and those without. No reliable estimates are obtained for Africa, tropical and South America, tropical Asia, the tropical and the South Atlantic Ocean, and the western tropical Pacific. A reasonable global budget is not obtained due to unreliable estimates. The amplitude of the seasonal variations over middle-latitude North America is found to be less than that of some tropical land areas.
Chemical transport model, Inverse Method, Carbon cycle