2. PDRMIP: Two thoughts Land response and timescales

4. «The message»
Multiple studies have found that the precipitation response to a perturbation, such as an increase in greenhouse gas concentrations, may usefully be divided into two broad categories with different scaling properties: a fast change, associated with an initial radiative and thermal reorganization of the atmospheric column combined with rapid adjustments to cloud fractions and microphysics, and a slower, response-driven change, associated with an altered global surface temperature. The slow change mainly scales with surface temperature, which in turn scales closely with top-of-atmosphere radiative forcing, while the rapid adjustments scale with the amount of additional energy absorbed through the atmospheric column. [Samset et al. 2017, in press]

5. Global response

6. Ocean response

7. Land response

8. Land response

9. «Instantaneous». «Rapid». «Slow»
«Instantaneous»? «Rapid»? «Slow»? What are the timescales of dP response?

10. The slow response…

11. To deal with this… …average output from several models, and do some smoothing. In the following: Data from CAM5, NorESM, HadGEM2, IPSL

12. CO2x2

13. CO2x2

14. SO4x5

15. SO4x5

16. «Rapid response»: 150 days?
CO2x CH4x Sol+2%
BCx SO4x5
CO2x CH4x Sol+2%
BCx SO4x5
«Rapid response»: 150 days?
Decoupling of fSST and fully coupled evolution more or less immediately

17. CO2x2

18. CO2x2

19. BCx10

20. BCx10

21. «Instantaneous»: 10-15 days?
CO2x CH4x Sol+2%
BCx SO4x5
«Instantaneous»: days?
Coherence length of 20 days for our model setup

22. «Instantaneous»: 10-15 days?
CO2x CH4x Sol+2%
BCx SO4x5
«Instantaneous»: days?
Coherence length of 20 days for our model setup

24. «Rapid response»: 150 days?
CO2x CH4x Sol+2%
BCx SO4x5
«Rapid response»: 150 days?
Decoupling of fSST and fully coupled evolution more or less immediately