1. Steve Rumbold Keith Shine, Lesley Gray Charlotte Pascoe (CASE, RAL) Picture: http://cloud1.arc.nasa.gov/crystalface/ Strat Hour - July 05, 2006 The University of Reading Solar cycle effects on stratospheric temperature

2. Effect of 11 year Solar Cycle Two main ways to affect stratospheric temperatures –Direct effect of flux change –Effect of resultant Ozone changes 0.1 – 0.15 % on total solar irradiance (TSI) However, large percentage effect in UV Insert Ref. Data: www.ngdc.noaa.gov/stp/SOLAR/

3. How can the solar cycle influence the stratosphere? Large percentage effect in UV Effects shortwave heating (O 2, O 3 ) Percentage change in spectral flux from solar min to max [data: Lean et al, 1997]. 1. UV effect 2. Ozone effect Percentage change in ozone from solar min to max determined from observations (annual mean) [data: SAGE satellite, Randel, 2005]. Change in UV has effect on O 3 concentrations This in turn has effect on heating Magnitude and distribution of change varies between studies

4. Approach Description of model Blue = inputs, Pink = model, Yellow = outputs Determine radiative component of solar cycle effect The model [Forster and Shine, 1997]: Narrow band radiation code Fixed Dynamical Heating (FDH) i.The radiative response can be isolated. ii.High spectral resolution radiative calculations can be performed. iii.A large number of separate calculations can be run quickly. This has been used instead of a General Circulation Model (GCM) for three main reasons:

5. Spectral sensitivity experiment (from Lean et al. 1997)

6. Black line = Lean et al. 1997 Grey = Lean 2005 range (1 Comparison of classic data with new study

8. Test of influence of Mg II line To asses the importance of the Mg II line, three experiments were conducted with the FDH model: i.Using actual solar cycle UV change (Black). ii.Same as i. but without Mg II line (Blue). iii.Just Mg II line change (Red). The Mg II line has moderate importance to total solar change. ~1/6 of total effect at 1 mbar. ii. and iii. sum linearly to reproduce i. Oct, Equator Investigating Mg II line Temperature change due to the spectrally resolved irradiance changes. Note: weird units [ K / (10nm) ]

9. LW SW NET Change in radiative heating (max minus min) [ K / day ] (or greater) (or less) Results from idealised ozone change experiment 6% 56 km +5 km -5 km 1/e Minimum = Climatology – 0.5 x change Maximum = Climatology + 0.5 x change

10. Temperature change (max minus min) [ K ] (Oct)

11. A solar maximum minus minimum experiment Irradiance changes only Ozone changes only Combining the two Compare to other studies

12. Irradiance Changes Only Minimum = solar neutral spectra – 0.5 x change Maximum = solar neutral spectra + 0.5 x change

13. Ozone changes solar maximum minus minimum [ % ] (annual mean) SAGE I/II Regression – Randel (2005) GISS – Shindell et al. (1999)Imperial College – Haigh (1994)

14. Resulting FDH temperature changes (solar maximum minus minimum [ K ] ) Shibata and Kodera, 2005 Using GISS ozone Shibata and Kodera, 2005 Using Imperial ozone Reading FDHM Using SAGE I/II ozone

15. ERA-40 Reanalysis Crooks & Gray (2005) Reading FDHM (SAGE ozone changes + UV irradiance changes) Temperature change ( solar maximum minus minimum [ K ] ) (annual mean) 0.5 K 1.8 K 0.5 K

16. Ozone effect vs. irradiance effect Bold line 50/50. Dotted lines irradiance wins. Solid lines ozone wins