Low Risk Emissions Corridors for Safeguarding the Atlantic Thermohaline Circulation

Transcription

1 Presentation to the Expert Workshop on "Greenhouse Gas Emissions and Abrupt Climate Change" Paris, France 30 Sept 1 Oct 2004 Low Risk Emissions Corridors for Safeguarding the Atlantic Thermohaline Circulation Thomas Bruckner* and Kirsten Zickfeld** *Institute for Energy Engineering Technical University of Berlin, Germany **Potsdam Institute for Climate Impact Research Potsdam, Germany *Correspondence: bruckner@iet.tu-berlin.de Website:

2 2 Content Introduction stability of the North Atlantic thermohaline circulation (THC) Integrated assessment of the thermohaline circulation with dimrise model overview dynamic THC module reduced-form climate module aggregate economic module Model application schemes and first results cost-effectiveness analysis Tolerable Windows Approach (TWA)

3 North Atlantic thermohaline circulation 3 Surface Deep Bottom Salinity > 36 Salinity < 34 Deep Water Formation Source: S. Rahmstorf, Nature (2002)

4 Short-term business-as-usual evolution 4 Key sources of uncertainty initial THC overturning climate sensitivity hydrological sensitivity m init T 2xCO2 h Taking h: F(t) = h. T NH (t) Simulated strength of the Atlantic overturning relative to the mean with future-forcing under the IS92a scenario 1 Sv = 1 Sverdrup = 10 6 m 3 /s Source: IPCC, TAR, WGI (2001) F(t) is change of freshwater flux into the Atlantic, north of 50 N T NH (t) is atmospheric temperature change in the northern hemisphere

5 Long-term sensitivity analysis for hydrological sensitivity h 5 ppm Carbon dioxide concentration Low h Global mean temperature C High h Control Control Sv High h Low h THC overturning Source: Rahmstorf and Ganopolski, Climatic Change (1999)

6 Integrated Assessment of the thermohaline circulation 6 Integrated assessment model dimrise dynamic integrated model of regular impacts and singular events Components dynamic model of the Atlantic overturning reduced-form multi-gas climate model aggregate model of the world economy Features dynamic, fully coupled, computationally fast GAMS model able to derive least-cost emissions paths and emissions corridors

7 Dynamic model of the Atlantic overturning 7 dynamic four-box interhemispheric extension of the seminal Stommel model calibrated against results of the CLIMBER 2 climate model of intermediate complexity T 1 T 2 T 3 T 4 = = = = m ( T V 4 T 1 ) + λ 1 ( T 1 T 1 ) 1 m ( T V 3 T 2 ) + λ2 ( T 2 T 2 ) 2 m ( T V 1 T 3 ) + λ 3 ( T 3 T 3 ) 3 m ( T V 2 T 4 ) 4 m k ρ2 ρ1 ρ 0 S 1 S 2 S 3 S m S 0 F 1 = ( S 4 S 1 ) V 1 V 2 V 1 m S 0 F 2 = ( S 3 S 2 ) V 3 V m S 0 ( F 1 F 2 ) = ( S 1 S 3 ) m = ( S 2 S 4 ) V 4 ( ) = = k [ β ( S 2 S 1 ) ( α ( T 2 T 1 ))] V 3 Equilibrium model (red curve) CLIMBER 2 (solid curve) Dynamic Box-Model (dashed curve) Source: Zickfeld, Slawig, and Rahmstorf, Ocean Dynamics 54, 8-26 (2004)

8 Dynamic THC model : hydrological sensitivity h 8 Response of the Atlantic overturning for different values of the hydrological sensitivity h given in Sv C -1 Source: Zickfeld, Slawig, and Rahmstorf, Ocean Dynamics 54, 8-26 (2004)

9 Dynamic THC model : rate of temperature change sensitivity 9 Response of the Atlantic overturning for different rates of temperature increase measured in C century -1 (hydrological sensitivity h = Sv C -1 ) Stability diagram of the THC for different values of the hydrological sensitivity The stable (unstable) domains are located to the left (right) of the respective curves 'SS' indicates the stability curve from Stocker and Schmittner (1997) Source: Zickfeld, Slawig and Rahmstorf, Ocean Dynamics 54, 8-26 (2004)

10 Reduced-form multi-gas climate model 10 ICLIPS climate module (ICM) Component of the ICLIPS (Integrated Assessment of Climate Protection Strategies) suite CO 2 -cycle: differential-impulse-response representation of the 3-dim Hamburg Model of the Oceanic Carbon Cycle (HAMOCC) Climate system: differential-impulse-response representation of ECHAM 3 Non-CO 2 greenhouse gas atmospheric chemistry for CH 4, N 2 O, halocarbons, SF 6 and aerosols according to MAGICC (Wigley et al ) Source: Bruckner et al., Climatic Change 56, (2003)

11 Aggregate model of the world economy 11 Economic relationships contained within DICE Dynamic Integrated model of Climate and the Economy (Nordhaus 1989, Nordhaus 1992, Nordhaus and Boyer 2000) Ramsey-type intertemporal optimal growth model with endogenous investment decisions and capital accumulation cycle Cobb-Douglas production function with exogenous technological change applied to assess emissions mitigation costs in terms of global welfare losses computationally fast GAMS model well-known and widely used in the integrated assessment community conceptual model used for proof-of-concept application to be replaced by a sophisticated multi-regional model

12 Aggregate model of the world economy 12 Global welfare: discounted flow of global utility CO 2 -emissions: = 1 W t U ( c( t), L( t)) t (1+ i) E( t) = [1 µ ( t)] σ ( t) Q( A( t), K( t), L( t)) active emissions reduction output dependence Percentage output loss due to active emissions mitigation: Q Q = b µ ( t) / 1 b 2 Control variables: emissions control level µ(t) and per capita consumption c(t)

13 Framework of the Tolerable Windows Approach (TWA) 13 Prescription of explicit normative "guard-rails" that cover both intolerable climate impacts socio-economically unacceptable mitigation side-effects Scientific analysis of the relevant and interconnected elements of the Earth system, including: ecosystems, the climate system, socio-economic systems Calculation of the set of admissible policy paths by applying a suitable integrated assessment model Selection of a specific policy path by applying quantitative optimization, by referring to qualitative arguments, and/or by relaxing normative constraints after public consultation

14 Normative guard-rails (constraints) 14 Climate guard-rail: prevention of a THC collapse Atlantic overturning m( t) mmin = 10 Sv Socio-economic guard-rail: acceptable emissions mitigation burden maximum percentage welfare loss relative to the non-intervention case RC WRC W W RC l max maximum increase in the emissions control level 0 & µ ( t) & µ max

15 Model application schemes 15 Overarching goal: preservation of the THC Cost-effectiveness analysis Tolerable windows approach Min WRC W W RC s.t. m( t) m min m( t) m WRC W W RC min l 0 & µ ( t) & µ max max Least-cost emissions paths Emissions corridors

16 First results 16 Model calibration Standard Worst case climate sensitivity T 2xCO2 2.5 C 4.5 C hydrological sensitivity h 0.03 Sv C Sv C -1 Normative guard-rails Default values Atlantic overturning overall welfare loss rate of change in the emissions control level m min I max µ max 10 Sv 2.0 % 1.33 %-Pts/year

17 Emissions corridors for standard conditions CO 2 Emissions [GtC/yr] Upper boundary Lower boundary Path maximizing emissions in 2040 Path maximizing emissions in 2100 Path maximizing emissions in 2160 BAU emissions path Calendar year

18 18 Variation of the climate sensitivity T 2xCO2 CO 2 Emissions [GtC/yr] Upper boundary for T 2xCO2 =2.5 C Upper boundary for T 2xCO2 =3.5 C Upper boundary for T 2xCO2 =4.5 C Cost-effective path for T =2.5 C 2xCO2 Cost-effective path for T 2xCO2 =3.5 C Cost-effective path for T 2xCO2 =4.5 C Lower boundary Calendar year

19 Variation of the hydrological sensitivity h CO 2 Emissions [GtC/yr] Upper boundary for h =0.03 Sv/ C 2 Upper boundary for h =0.04 Sv/ C 2 Upper boundary for h =0.05 Sv/ C 2 Cost-effective path for h = Sv/ C 2 Cost-effective path for h =0.05 Sv/ C 2 Lower boundary Calendar year

20 20 Variation of the admissible welfare loss l max 45 CO 2 Emissions [GtC/yr] Upper boundary for l max = 4% Upper boundary for l max = 2% Upper boundary for l = 1% max Upper boundary for l max = 0.2% Lower boundary for l max = 1-4% Lower boundary for l = 0.2% max Cost-effective emissions path Calendar year

21 Variation of the max emissions reduction rate µ max 21 CO 2 Emissions [GtC/yr] Upper boundary for µ max = % Lower boundary for µ max = 2.5% Lower boundary for µ = 2% max Lower boundary for µ = 1.33% max Lower boundary for µ = 1% max Lower boundary for µ max = 0.5% Cost-effective path Calendar year

22 Variation of the admissible welfare loss l max Worst case emissions corridors and least-cost path CO 2 Emissions [GtC/yr] Upper boundary for l max = 4% Upper boundary for l max = 2% Upper boundary for l max = 1% Upper boundary for l max = 0.5% Lower boundary for l max = 1-4% Lower boundary for l max = 0.5% Cost-effective emissions path BAU emissions path Calendar year

23 Variation of the max emissions reduction rate µ max Worst case emissions corridors and least-cost path CO 2 Emissions [GtC/yr] Upper boundary for µ max = 2.5% Upper boundary for µ max = 1.33% Upper boundary for µ max = 0.5% Lower boundary for µ max = 2.5% Lower boundary for µ max = 1.33% Lower boundary for µ max = 0.5% Cost-effective path BAU emissions path Calendar year

24 24 Conclusions dimrise a fully-coupled dynamic integrated assessment model for investigating THC instability suitable for deriving cost-effective emissions paths and emissions corridors (proof-of-concept) best-guess conditions cost-effective emissions path does not deviate from the business-as-usual emissions "comfortable" emissions corridors sensitive to uncertain climate and hydrological sensitivities worst-case conditions (moderate) business-as-usual path transgresses the upper corridor boundary within the next two decades if future world-wide emissions mitigation capabilities remain low

25 25 Thank you for your attention Contact: Dr. Thomas Bruckner Institute for Energy Engineering Technical University of Berlin Marchstrasse 18 D Berlin Germany Tel.: ++49/30/ WWW:

26 Selected references 26 Petschel-Held, G, H-J Schellnhuber, T Bruckner, F.L Tóth, K Hasselmann: The Tolerable Windows Approach: Theoretical and Methodological Foundations, Climatic Change 41, (1999). Bruckner, T, G Petschel-Held, F.L Tóth, H-M Füssel, C Helm, M Leimbach, H J Schellnhuber: Climate Change Decision-Support and the Tolerable Windows Approach. Environmental Modeling and Assessment 4, (1999). Tóth, F.L, T Bruckner, H-M Füssel, M Leimbach, G Petschel-Held, H-J Schellnhuber: Exploring Options for Global Climate Policy: A New Analytical Framework, Environment 44/5, (2002). Tóth, F.L., T Bruckner, H-M Füssel, M Leimbach, G Petschel-Held: Integrated Assessment of Long-Term Climate Policies: Part 1 - Model Presentation, Climatic Change 56, (2003). Tóth, F.L, T Bruckner, H-M Füssel, M Leimbach, G Petschel-Held, Integrated Assessment of Long-Term Climate Policies: Part 2 - Model Results and Uncertainty Analysis, Climatic Change 56, (2003). Bruckner, T, G Petschel-Held, M Leimbach, F.L Tóth: Methodological Aspects of the Tolerable Windows Approach, Climatic Change 56, (2003). Bruckner, T, G Hooss, H-M Füssel, K Hasselmann: Climate System Modeling in the Framework of the Tolerable Windows Approach: The ICLIPS Climate Model, Climatic Change 56, (2003). Zickfeld, K, T Bruckner: Reducing the Risk of Abrupt Climate Change: Emissions Corridors Preserving the Atlantic Thermohaline Circulation, Integrated Assessment 4, (2003). E Kriegler, T Bruckner: Sensitivity Analysis of Emissions Corridors for the 21 st Century, Climatic Change 66, (2004). T. Bruckner, K. Zickfeld: Low Risk Emissions Corridors for Safeguarding the Atlantic Thermohaline Circulation, Expert Workshop on Greenhouse Gas Emissions and Abrupt Climate Change, September 30, 2004, Paris. Internet Proceedings: