EGU General Assembly 2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

The Transient Response of the Carbon-Cycle-Climate Continuum to CO2 Emissions is Pathway Dependent 

Alexander J. Winkler1,2, Ranga B. Myneni3, Markus Reichstein1, and Victor Brovkin2
Alexander J. Winkler et al.
  • 1Max-Planck-Institute for Biogeochemistry, Jena, Germany (
  • 2Max-Planck-Institute for Meteorology, Hamburg, Germany
  • 3Department of Earth and Environment, Boston University, United States of America

The prevailing understanding of the carbon-cycle response to anthropogenic CO2 emissions suggests that it depends only on the magnitude of this forcing, not on its timing. However, a recent study (Winkler et al., Earth System Dynamics, 2019) demonstrated that the same magnitude of CO2 forcing causes considerably different responses in various Earth system models when realized following different temporal trajectories. Because the modeling community focuses on concentration-driven runs that do not represent a fully-coupled carbon-cycle-climate continuum, and the experimental setups are mainly limited to exponential forcing timelines, the effect of different temporal trajectories of CO2 emissions in the system is under-explored. Together, this could lead to an incomplete notion of the carbon-cycle response to anthropogenic CO2 emissions.

We use the latest CMIP6 version of the Max-Planck-Institute Earth System Model (MPI-ESM1.2) with a fully-coupled carbon cycle to investigate the effect of emission timing in form of four drastically different pathways. All pathways emit an identical total of 1200 Pg C over 200 years, which is about the IPCC estimate to stay below 2 °K of warming, and the approximate amount needed to double the atmospheric CO2 concentration. The four pathways differ only in their CO2 emission rates, which include a constant, a negative parabolic (ramp-up/ramp-down), a linearly decreasing, and an exponentially increasing emission trajectory. These experiments are idealized, but designed not to exceed the observed maximum emission rates, and thus can be placed in the context of the observed system.

We find that the resulting atmospheric CO2 concentration, after all the carbon has been emitted, can vary as much as 100 ppm between the different pathways. The simulations show that for pathways, where the system is exposed to higher rates of CO2 emissions early in the forcing timeline, there is considerably less excess CO2 in the atmosphere at the end. These pathways also show an airborne fraction approaching zero in the final decades of the simulation. At this point, the carbon sinks have reached a strength that removes more carbon from the atmosphere than is emitted. In contrast, the exponentially increasing pathway with high CO2 emission rates in the last decades of the simulation, the pathway usually studied, shows a fairly stable airborne fraction. We propose a new general framework to estimate the atmospheric growth rate of CO2 not only as a function of the emission rate, but also include the aspect of time the system has been exposed to excess CO2 in the atmosphere. As a result, the transient temperature response is a function not only of the cumulative CO2 emissions, but also of the time the system was exposed to the excess CO2. We also apply this framework to other Earth system models and observational records of CO2 concentration and emissions.

The Earth system is currently in a phase of increasing, nearly exponential CO2 forcing. The impact of excess CO2 exposure time could become apparent as we approach the point of maximum CO2 emission rate, affecting the achievability of the climate targets.

How to cite: Winkler, A. J., Myneni, R. B., Reichstein, M., and Brovkin, V.: The Transient Response of the Carbon-Cycle-Climate Continuum to CO2 Emissions is Pathway Dependent , EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15392,, 2021.


Display file

Comments on the display

to access the discussion