An efficient and accurate carbon cycle model for use in simple climate models
MetadataVis full innførsel
- CICERO Working Papers 
The aim of this publication is to document a simple model of the atmospheric CO2 concentration based on exogenous input of anthropogenic emission of CO2 and taking air-sea exchange and biospheric responses into account. The approach described by Joos et al (1996) is based on the application of a mixed layer pulse response function. The advantage of using a mixed layer pulse response function instead of an atmosphere pulse response function (Siegentaler and Oeschger, 1978; Oeschger and Heimann, 1983; Meier-Reimer and Hasselmann, 1987; Sarmiento et al., 1992) is that it is then possible to represent the non-linear effects of seawater chemistry. As long as the CO2 increase in the atmosphere is below approximately 50% of the pre-industrial level, the CO2 system behaves in a linear way and it is possible to represent the effects of anthropogenic emissions on the atmospheric concentrations by an atmospheric pulse response function. For CO2 perturbations beyond this level the non-linear effects of the seawater chemistry becomes important, and it thus becomes necessary to apply a mixed layer pulse response function to obtain accurate results. The approach described in this paper includes changes in CO2 uptake and release by terrestrial vegetation by CO2 fertilization, but does not take into account possible feedback mechanisms of climate change on the carbon cycle. Possible feedbacks include changes in CO2 solubility due to sea surface temperature (SST) changes and changes in vertical mixing by reduced deep-water formation in the North Atlantic. Joos et al. (1999a) have estimated the marine part of this to be of minor importance (approximately 4%) up to 2100, increasing to about 20% in 2500. Since damage caused by climate change could imply high costs, an accurate representation of the carbon cycle is very important in models that are to be used to estimate the costs of climate and evaluate different mitigation and/or adaptation strategies (Joos et al., 1999b). The carbon cycle model described in this paper is included in a simple model for scenario studies of changes in global climate (Fuglestvedt and Berntsen, 1999).