1. How do regional aerosol emissions affect climate? Matthew Kasoar With: Apostolos Voulgarakis (Imperial), Drew Shindell (NASA GISS/Duke), Greg Faluvegi (NASA GISS), Jean-Francois Lamarque (NCAR), and Dill Shawki (Imperial)

2. Aerosol – climate interactions are complicated…  CTMs: Emissions  mixing ratios/RF  GCMs: mixing ratios/RF  Climate  CCMs: Emissions  Climate (but compounded uncertainty)  Greater scope for parametric model uncertainty  Greater scope for structural model uncertainty Precursor emissions Advection Oxidation Wet/dry deposition Nucleation/ Coagulation Wet/dry deposition RF Climate response Aerosol emissions Advection

3. SO 2, BC SO 2 SO 2, BC

4. SO 2

5. BC

6. N96L85 configuration (1.875°x1.25°, 85 levels up to 85km) 1° NEMO ocean CLASSIC aerosol scheme 100/200 year simulations, repeat year 2000 emissions Remove emission of interest from individual regions HADGEM3-AO Simulations with

7. BC vs. SO 2 TOA SW Temp

8. US vs. China SO 2 TOA SW Temp

9. India vs. China SO 2 TOA SW Temp

10. Dynamical changes in above SO 2 cases  Similarity in pattern of US and China responses corresponds to similar dynamical structure  Indian SO 2 removal produces different pattern China US India

11. 3X AS MANY MODELS Investigating US and China SO 2 removals with HadGEM3 1.875° x 1.25°; 85 vertical levels CESM1 2.5° x 1.875°; 30 vertical levels GISS-E2 2.5° x 2°; 40 vertical levels

12. HadGEM3 CESM1 GISS-E2 USCHINA

13. 1.57 TgHadGEM3 1.09 TgGISS-E2 Differences between HadGEM3 and GISS-E2 SO 2 dry deposition:  HadGEM3 = 406.2 kg(S)/s  GISS-E2 = 1399 kg(S)/s SO 2 dry oxidation:  HadGEM3 = 609.3 kg(S)/s  GISS-E2 = 681.9 kg(S)/s SO 2 wet removal/oxidation:  HadGEM3 = 1419 kg(S)/s  GISS-E2 = 742.5 kg(S)/s 0.64 TgHadGEM3 1.15 TgGISS-E2 SO 2 SO 4

14.  Present-day BC effects are negligible compared to SO 2 in HadGEM3/CLASSIC  Strong hemisphere-wide response to localised mid-latitude forcing  Mechanism of remote temperature responses  Which model is right?  Substantial model diversity in sensitivity to SO 2 emission reductions  Models vary in both aerosol quantity and radiative efficiency Conclusions Questions

15. THANK YOU

16. EXTRA SLIDES:

17. AERONET comparison  Biases in aerosol amount explain some of the discrepancies OBS HadGEM3 GISS-E2

18.  Biases in aerosol amount explain some of the discrepancies  But differences in radiative efficiency as well AERONET comparison OBS CESM1 GISS-E2

19. Map of regions China (NCAR/GISS) = 80E-120E, 20N-50N SA = 70E-90E, 10N-30N EA = 105E-145E, 20N-45N US = 70W-125W, 30N-50N NHML = 30N-60N SO 2, BC SO 2 SO 2, BC SO 2 SO 2, BC

20. US vs. EA SO 2 TOA SW Temp

21. India vs. EA SO 2 TOA SW Temp