19 October 2005The CRCES Workshop on Decadal Climate Variability Modelling the Variability of Midlatitude Storm Activity on Decadal to Century Time Scales Hans von Storch, Institute for Coastal Research, GKSS Irene Fischer-Bruns, Max Planck Institute for Meteorology Germany

19 October 2005The CRCES Workshop on Decadal Climate Variability Modelling the Variability of Midlatitude Storm Activity on Decadal to Century Time Scales Hans von Storch, Irene Fischer-Bruns The output of several multi-century simulations with a coupled Ocean-Atmosphere General Circulation Model is examined with respect to the variability of global storm activity in winter on time scales of decades and longer. The frequency of maximum wind speed events within a grid box, using the lower limits on the Beaufort wind speed scale of 8 and 10 Bft as thresholds, is taken as the characteristic parameter. Two historical climate runs with time-dependent forcing of the last five centuries, one control simulation, and three climate change experiments are considered. The storm frequency shows no trend until recently. Global maps for the industrially influenced period hardly differ from pre-industrial maps, even though significant temperature anomalies temporarily emerge in the historical runs. Two indicators describing the frequency and the regional shift of storm activity are determined. In historical times they are decoupled from temperature. Variations in solar and volcanic forcing in the historical simulations as well as in greenhouse gas concentrations for the industrially influenced period are not related to variations in storm activity. Also, anomalous temperature regimes like the Late Maunder Minimum are not associated with systematic storm conditions. In the climate change experiments, a poleward shift of storm activity is found in all three storm track regions. Over the North Atlantic and Southern Ocean, storm activity increases, while it decreases over the Pacific Ocean. In contrast to the historical runs, and with the exception of the North Pacific storm frequency index, the storm indices parallel the development of temperature, exceeding the 2 s-range of preindustrial variations in the early 21st century.

19 October 2005The CRCES Workshop on Decadal Climate Variability CRCES-IPRC Workshop on Decadal Climate VariabilityCRCES-IPRC Workshop on Decadal Climate Variability, Kona, HI, Variability on time scales of decades up to a century in a AOGCM simulation with realistic time-variable forcing Hans von Storch(1), Eduardo Zorita(1), Irene Fischer-Bruns(2) and Jian Liu(3) (1) Institute for Coastal Research, GKSS Research Center, Geesthacht, Germany (2) Max-Planck Institute for Meteorology Hamburg, Germany (3) Nanjing Institute of Geography and Limnology, CAS, Nanjing, P.R.China (4) Fidel González-Ruoco, Departamento de Astrofísica y CC de la Atmosfera, Universidad Complutense de Madrid, Spain Two multi-century simulations with a state-f-the-art AOGCM have been made using reconstructed solar, volcanic and GHG forcing. Significant variations in large scale air temperatures are generated, with a marked medieval warming, an extended Little Ice Age punctuated by extra cool periods during the Dalton and Late Maunder Minimum, and a fast increase of temperature during the last century. The results are presented and discussed; special emphasis is laid on the variations of the three large-scale storm tracks in the North Atlantic, North Pacific and the Southern Ocean, and in China.

19 October 2005The CRCES Workshop on Decadal Climate Variability Atmosphere-Ocean GCM ECHO-G Atmospheric Model ECHAM4 (T30) (~3.75°x 3.75°  ~300 km x 300 km) Ocean Model HOPE-G (T43) (~2.8°x 2.8°  ~200 km x 200 km) Model & Experiments 2 Historical Simulations (time dependent solar / volcanic / GHG forcing) 3 Future Climate Change Simulations (CMIP2, SRES A2/B2) (152 resp. 110 years) Control Simulation 1000 years (present day conditions)

19 October 2005The CRCES Workshop on Decadal Climate Variability

19 October 2005The CRCES Workshop on Decadal Climate Variability

19 October 2005The CRCES Workshop on Decadal Climate Variability The millennial run generates temperature variations considerably larger than MBH-type reconstructions. The simulated temperature variations are of a similar range as derived from NH summer dendro-data, from terrestrial boreholes and low- frequency proxy data.

19 October 2005The CRCES Workshop on Decadal Climate Variability Reconstructed and simulated winter temperature anomalies in eastern China (Liu et al., 2005)

19 October 2005The CRCES Workshop on Decadal Climate Variability vs Reconstruction from historical evidence, from Luterbacher et al. Late Maunder Minimum Model-based reconstuction

19 October 2005The CRCES Workshop on Decadal Climate Variability IPCC SRES A2 and B2 emissions marker scenarios A2 Business as usual External Forcing – Future Scenarios B2 Strong focus on environmental protection Lower emissions – less future warming

19 October 2005The CRCES Workshop on Decadal Climate Variability Applications so far Methodical analysis of performance of MBH method and MM05 “AHS” mechanism. (von Storch, H., E. Zorita, J. Jones, Y. Dimitriev, F. González-Rouco, and S. Tett, 2004: Reconstructing past climate from noisy data, Science 306, ; von Storch, H., and E. Zorita, 2005: Comment to "Hockey sticks, principal components and spurious significance" by S. McIntyre and R. McKitrick, Geophys. Res. Lett. (in press) doi: /2005GL022753) Simulation of Late Maunder Minimum – regional European or near-global phenomenon (Zorita, E., H. von Storch, F. González-Rouco, U. Cubasch, J. Luterbacher, S. Legutke, I. Fischer-Bruns and U. Schlese, 2004: Climate evolution in the last five centuries simulated by an atmosphere-ocean model: global temperatures, the North Atlantic Oscillation and the Late Maunder Minimum. Meteor. Z. 13, ) Comparison with multidecennial Chinese temperatures ( 刘 健, H. von Storch, 陈 星, E. Zorita, 郑景云, and 王苏民, 2005: 千年气候模拟与中国东部温度重 建序列的比较研究 (Comparison of simulated and reconstructed temperature in eastern China during the last 1000 years), Chinese Science Bulletin, in press) Low frequency variability in temperature modes and Extratropical storminess (Zorita, E., F. González-Rouco, H. von Storch, J.P. Montavez und F. Valero, 2005: Natural and anthropogenic modes of surface temperature variations in the last millennium, Geophys. Res. Letters 32, L08707 Fischer-Bruns, I., H. von Storch, F. González-Rouco and E. Zorita, 2005: Modelling the variability of midlatitude storm activity on decadal to century time scales. Clim. Dyn. DOI /s )

19 October 2005The CRCES Workshop on Decadal Climate Variability North Atlantic Storminess: Observational results 12 hourly pressure changes exceeding 16 hPa / 12h Lund Stock- holm Worsening of storminess in NAtl since a minimum in the 1960s consistent with NAO changes No significant changes over last 100 years (WASA, 1998) Different storm indicators from pressure readings 1780/  No evidence of long-term trend (Bärring & von Storch, GRL, 2005.)

19 October 2005The CRCES Workshop on Decadal Climate Variability Atmosphere-Ocean GCM ECHO-G Atmospheric Model ECHAM4 (T30) (~3.75°x 3.75°  ~300 km x 300 km) Ocean Model HOPE-G (T43) (~2.8°x 2.8°  ~200 km x 200 km) Model & Experiments Historical Simulation (time dependent forcing) 3 Future Climate Change Simulations 3 Future Climate Change Simulations (CMIP2, /B2) (152 resp. 110 years) (CMIP2, SRES A2/B2) (152 resp. 110 years) Control Simulation 1000 years Control Simulation 1000 years (present day conditions) (present day conditions)

19 October 2005The CRCES Workshop on Decadal Climate Variability diagnosed at every grid-point and at every time step stored every 30 minutes output every 12 hours Extreme wind speed events per season (  8 Bft, gales ) were counted Simulated data 10m maximum wind speed

19 October 2005The CRCES Workshop on Decadal Climate Variability Analysis of different model experiments with respect to gale frequency Determination of simple indicators describing storm activity Model Study Storm Intensity Index Storm Shift Index

19 October 2005The CRCES Workshop on Decadal Climate Variability Mean number of gale days (10m wind speed reaching at least 8 Bft) in the northern winter season DJF (left) and in the southern winter season JJA (right) for the pre- industrial period in the historical experiment (upper panels) and mean number of storm days (10m wind speed reaching at least 10 Bft, lower panels).

19 October 2005The CRCES Workshop on Decadal Climate Variability Mean number of gale days (10m wind speed reaching at least 8 Bft) in the northern winter season DJF (left) and in the southern winter season JJA (right) for the industrially influenced period in the historical experiment.

19 October 2005The CRCES Workshop on Decadal Climate Variability Mean number of gale days (10m wind speed reaching at least 8 Bft) in the northern winter season DJF (left) and in the southern winter season JJA (right) during the last 300 years of the control run of 1000 years length.

19 October 2005The CRCES Workshop on Decadal Climate Variability Mean number of gale days (10m wind speed reaching at least 8 Bft) in the northern winter season DJF (left) and in the southern winter season JJA (right) in the A2 climate change experiment.

19 October 2005The CRCES Workshop on Decadal Climate Variability industrial – pre-industrial Storm Intensity DJF JJA A2 – pre-industrial

19 October 2005The CRCES Workshop on Decadal Climate Variability Storm Intensity Mean number of gale-days averaged over time and region NH: no change SH: increase pre-ind ind A2 Storm Intensity Index Storm intensity index = plain storm count

19 October 2005The CRCES Workshop on Decadal Climate Variability Storm Intensity Mean number of galedays averaged over time and region NH: no change SH: increase pre-ind ind A2 N Atl: increase N Pac: decrease Storm Intensity Index pre-ind ind A2 (150E-90W) (90W-30E)

19 October 2005The CRCES Workshop on Decadal Climate Variability Storm Shift 26.5 % 23.4 % NAtl (DJF) NPac (DJF) SH (JJA) 11.2 % Leading EOFs pre-ind Pattern of slope coefficient in A2 y = x % y = 0.20x % y = 0.22x %

19 October 2005The CRCES Workshop on Decadal Climate Variability Storm Shift PCs obtained by projection onto EOF EOFs pre-ind NAtl (DJF) NPac (DJF) Storm Shift Index SH (JJA) pc 1 pc 4

19 October 2005The CRCES Workshop on Decadal Climate Variability Poleward shift (NE) Intensity: slight increase Poleward shift Intensity: decrease (11-yr running mean) Temp & Indices: No correlation in pre-industrial period N Atlantic Warming and Storms pc 1 pc 4 N Atlantic N Pacific Warming and Storms NAtl Temp Storm Intensity Index Storm Shift Indices NPacTemp Storm Intensity Index Storm Shift Index

19 October 2005The CRCES Workshop on Decadal Climate Variability Intensity: sharp increase Poleward shift Warming and Storms Southern Hemisphere SH Temp Storm Intensity Index Storm Shift Index (11-yr running mean)

19 October 2005The CRCES Workshop on Decadal Climate Variability Conclusions – long term simulation Historical runs done. Realistic sequence of warming and cooling. Variations larger than in multi-proxy regression-type reconstructions, but consistent with other reconstructions and some regional data. Hans von Storch

19 October 2005The CRCES Workshop on Decadal Climate Variability Conclusions - storms Analysis of long-term model data with respect to wind speed extremes; Determination of two simple indicators describing storm intensity and storm track location. The storm indices show no trends in the historical experiments, except for SH in 20th century. NH Temperature and Storm activity are uncorrelated in the pre- industrial period Future climate change scenario A2: - Parallel increase of storminess indices and temperatures. - Poleward shift of the region with maximum gale intensity for NAtlantic (NE), NPacific and SH; Storm intensity is constant over the NH as a whole, but increasing in the Atlantic region and decreasing in the Pacific - Increase of storm intensity for SH Irene Fischer- Bruns