Presentation Title, Arial Regular 29pt Sub title, Arial ...

Presentation Title, Arial Regular 29pt Sub title, Arial ...

Using ensembles in climate sensitivity and feedbacks Rob Colman Plus thanks to: Scott Power, Lawson Hanson, Jo Brown, Michael Grose1, Charmaine Franklin and others.. Climate Research Section, StS, Bureau of Meteorology 1 CSIRO Outline 1. Motivation: why should we care about climate sensitivity and

feedbacks? 2. Why is the spread so large and what do ensembles tell us? 3. Use of ensembles strengths and weaknesses 4. Where to in future? 5. Conclusions ECS: Equilibrium climate sensitivity ECS: equilibrium climate sensitivity AGCM coupled to mixed layer ocean, 2xCO2 Equilibrates in decades

Knowing ECS would drop Australian T 5-95% range T 5-95% range by a factor of ~3 From this to, say, this Grose et al., 2016: Limits to global and Australian temperature change this century based on expert judgment of climate sensitivity, Climate Dynamics 4 Potentially worth a few $$$...

Halving uncertainty in TCR by 2020 save $10T by end of century Estimates of ECS. history International assessments 1.5 to 4.5 C : National Academy of Science (US), 1979 1.5 to 4.5 C : IPCC 1st Assessment Report, 1990 1.5 to 4.5 C : IPCC 2nd Assessment Report, 1996 Jule Charney (1917-1981) (IPCC, 2001) 1.5 to 4.5 C : IPCC 3rd Assessment Report, 2001

2.0 to 4.5 C : IPCC 4th Assessment Report, 2007 (best estimate 3C)* 1.5 to 4.5 C : IPCC 5th Assessment Report, 2013** * >90% likely that >1.5 C ** Range of >66% likelihood. <5% chance of >6 C. CMIP5 model range 2.1 to 4.7 C IPCC ECS estimates are always based on ensembles3 Knutti and Hegerl (2008) AR5, Box 12.2 Figure 1 (2013) What does each ensemble give us and what are the strengths/weaknesses? 1. Multi-model ensemble (CMIPx):

'ensemble of opportunity' based on intense partially independent model development efforts, which are not focussed on sensitivity Ad hoc group, no systematic exploration of feedbacks and processes 2. Perturbed physics ensembles Carefully constructed sampling of parameter uncertainty Does not sample 'structural' differences Very host model dependent for unclear reasons

What does each ensemble give us and what are the strengths/weaknesses? 3. Sensitivity estimated from observational record Looks at real time changes Forcing poorly known, so many different estimates used 4. Estimates from paleo record Real world response

Forcing is different, long terms feedbacks (e.g ice sheets) come into play, paleo reconstructions differ What does each ensemble give us and what are the strengths/weaknesses? 5. Constraints from volcanic eruptions Forcing different, and poorly constrained Response rapid and transient, not involving deep ocean 6. Analogues/constraints from unforced variability

Possible to constrain feedbacks from observations 'Forcing' is different, and surface temperature pattern differs (although some overall similarities) What do ensembles of GCMs tell us about sensitivity and feedbacks? 1. They reveal critical physical processes and their relationships Blue = CMIP3 Red = CMIP5 Colman and Hanson, 2016: On the relative strength of radiative

feedbacks under climate variability and change, Climate Dynamics What do ensembles of GCMs tell us about sensitivity and feedbacks? 2. They reveal where the biggest uncertainties come from IPCC AR5 (2013) Chapter 9 Cloud feedbacks vary a lot, and are not well constrained: Short Wave cloud feedback in CMIP5 GCMs

Colman, Brown, Franklin, Hanson, Ye and Zelinka (2018) Evaluating cloud feedbacks and rapid responses in the ACCESS model (submitted) What do ensembles of GCMs tell us about sensitivity and feedbacks? 3. They suggest possible constraints Zhou et al., 2015 Interannual cloud feedback and climate change cloud feedbacks are correlated across GCMs Q: What model-for-model correspondences exist for cloud (and other)

feedbacks in CMIP5 between climate variability and change? A: A lot, as it turns out! Interannual = statistically significant correlation Each point is a CMIP5 model 15 Decadal Does all this correlation mean variability and ECS are linked? 6 CMIP5

5 ECS ECS (K) 4 3 2 1 R = 0.45

0 0.00 (a) 0.02 0.04 0.06 TS std , tropical (K) 0.08

Tropical decadal variability 0.10 Does this hold for earlier models?: no for CMIP3! 6 CMIP5 5 ECS (K) 4

3 2 1 R = 0.45 0 0.00 0.02 (a) 0.04

0.06 0.08 0.10 TS std , tropical (K) 6 CMIP3 5

Colman and Power, 2018: What can decadal variability tell us about climate feedbacks and sensitivity? Climate Dynamics ECS (K) 4 3 2 1 R = 0.00

CMIP6? Let's see 0 0.00 (b) 0.02 0.04 0.06 0.08

TS std , tropical (K) 0.10 0.12 Conclusions ECS a critical parameter for understanding the magnitude of projected regional and global climate change ECS 'range' traditionally estimated from an 'ensemble assessment of an ensemble of ensembles' Hasn't shifted much over >35years, but major advances have been made in the understanding of/confidence in the processes (feedbacks) controlling climate sensitivity.

Ensembles of many different types have been the key to progress so far, and will be the key to future progress. Diversity is good, a single world 'super model' would be a disaster. 18 Thank you 19 | AOGS 2013 Centre for Australian Weather and Climate Research

Recently Viewed Presentations

  • Tango Interactive Today for Education and Training Rice

    Tango Interactive Today for Education and Training Rice

    Shared SmartDesk system aimed at activities useful in special education with built in assessment Untested Quiz and Glossary Support Shared Simulations -- Fluid Flow and Planetary Motion JavaScript Shared Browser with Dynamic HTML -- "Pure" Client Event Shared Pointer added...
  • CPS PowerPoint Template New Plants - IBM Cognos User Group

    CPS PowerPoint Template New Plants - IBM Cognos User Group

    Through the various product modules, users are able to save tremendous time automating manual processes while adding a wealth of capabilities to extend the value of the out-of-the-box IBM Cognos software. What it is to us. Deployment between Development and...
  • psy.ku.dk

    psy.ku.dk

    Cand. psych. Helene Almind Jansen specialist i psykoterapi. Sexologisk Klinik - Psykiatrisk Center København, Rigshospitalet. Pc - København. Sexologisk klinik. Helene Almind Jansen. PC-København. Psykiatri. Indsæt hjælpelinjer til placering af objekter. 1. Højreklik uden for slidet og vælg . Gitter...
  • Figure 3. Alignment of parking brake after brake

    Figure 3. Alignment of parking brake after brake

    Link B is driving link where user applies force to engage brake. Link C is driving link where brake applies force on tire. Link D is coupler link. Four linkage joints are numbered from 1 to 4 counterclockwise. Washer between...
  • Hot-Potato vs. C

    Hot-Potato vs. C

    Hot-Potato vs. Cold-Potato Hot-Potato In hot-potato routing, an ISP hands off traffic to a downstream ISP as quickly as it can Hot-potato routing = route to closest exit point when there is more than one route to destination Cold-potato routing:...
  • Cultural Diffusion - Mrs. Brooks

    Cultural Diffusion - Mrs. Brooks

    Expansion Diffusion Expansion Diffusion: Idea that develops in an area and remains strong there while also spreading to other areas (McDonald's) 2. Relocation Diffusion Relocation Diffusion: Idea that develops in an area but moves to different places, leaving behind its...
  • SALIVARY GLANDS - Yola

    SALIVARY GLANDS - Yola

    Opening of parotid duct: It opens in the vestibule of the mouth opposite the upper second molar tooth. Surface anatomy: The duct is represented by the middle 1/3 of the line between a point on the lower border of tragus...
  • Unit 4 INDERJEET MANDER VIDHI PATEL NIVRITT KOHLI

    Unit 4 INDERJEET MANDER VIDHI PATEL NIVRITT KOHLI

    -Olmecas, Toltecas, Mayans, Aztecs all played-players hit a rubber ball with their hips, arms, or a bat to get it in the ring. Zapotecas and Otavalenos-Oaxaca is based on indigenous civilizations ...