Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
ANALYSIS OF 20TH CENTURY ATLANTIC HURRICANE POTENTIAL INTENSITY AND TROPICAL CYCLONE ACTIVITY IN THE CMIP5 MODELS Suzana J. Camargo Lamont-Doherty Earth Observatory Columbia University Atlantic Sector Climate Variability over the Last Millennium and the Near Term Future Workshop LDEO, Palisades, NY, October 17, 2012 OUTLINE • Local and remote influences of Atlantic hurricane potential intensity • Tropical cyclone activity in the CMIP5 models LOCAL AND REMOTE INFLUENCES ON ATLANTIC HURRICANE POTENTIAL INTENSITY Suzana J. Camargo, Mingfang Ting and Yochanan Kushnir Lamont-Doherty Earth Observatory, Columbia University Thanks to Donna Lee, Naomi Naik and Cuihua Li ATLANTIC PDI (POWER DISSIPATION INDEX ~ V3MAX) AND TROPICAL SST Emanuel, 2005 20TH CENTURY NORTH ATLANTIC SST AND POTENTIAL INTENSITY (PI) PDI and SST PDI and relative SST Vecchi and Soden 2007 • Objective: • Contributions of natural variability and anthropogenic trend to North Atlantic potential intensity • CCM3 simulations forced with fixed SST • GOGA: global SST • TAGA: tropical Atlantic SST • 16 ensemble members, 1856-2006 • See description in Seager et al. (2005) • IOPOGA: Indo-Pacific SST • 8 ensemble members, 1856-2006 PI GOGA & REANALYSIS I CLIMATOLOGICAL ANNUAL MAXIMUM PI ANOMALY GOGA AND REANALYSIS ATLANTIC MAIN DEVELOPMENT REGION (MDR) PI GOGA, TAGA & IOPOGA PI AND RELATIVE SST: GOGA, TAGA & IOPOGA (a) Mean Anom. PI Tropical North Atlantic 6 GOGA TAGA IOPOGA Anom. PI (m/s) 4 2 0 -2 -4 -6 1860 1880 1900 1920 1940 1960 1980 2000 1980 2000 Year (b) Mean Anom. Relative SST Tropical North Atlantic GOGA TAGA IOPOGA Anom. Rel. SST (C) 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 1860 1880 1900 1920 1940 Year 1960 CLIMATE CHANGE AND INTERNAL VARIABILITY (AMV) INDICES (a) CC and AMV indices 0.5 AMV CC 0 -0.5 1900 1910 1920 1930 1940 1950 1960 (b) Regression SST and AMV 50N 1970 1980 1990 Ting et al., 2009 2000 3 2 -0.5 1900 1910 1920 1930 1940 1950 REGRESSION WITH SST 1960 1970 1980 1990 2000 (b) Regression SST and AMV 50N 3 2 1 0 0 -1 -2 50S 0E 50E 100E 150E 160W 110W 60W 10W -3 (c) Regression SST and CC 50N 3 2 1 0 0 -1 -2 50S 0E 50E 100E 150E 160W 110W 60W 10W -3 PI REGRESSION PATTERNS: CC (a) Anom. PI Reg. AMV & CC Anom. PI (m/s) 1 0 -1 -2 1900 1910 1920 1930 1940 1950 1960 Year 1970 1980 1990 2000 1980 1990 2000 1980 1990 2000 (b) Anom. PI Reg. AMV GOGA TAGA IOPOGA Anom. PI (m/s) 2 1 0 -1 -2 1900 1910 1920 1930 1940 1950 1960 Year 1970 (c) Anom. PI Reg. CC GOGA TAGA IOPOGA 2 Anom. PI (m/s) PI REGRESSIONS TIME-SERIES: GOGA TAGA IOPOGA 2 1 0 -1 -2 1900 1910 1920 1930 1940 1950 1960 Year 1970 SUMMARY • Remote SST reduces trend of North Atlantic PI (confirming Vecchi and Soden 2007). • Remote SST also slightly reduces AMV effect on PI in the North Atlantic. • Differences of PI for GOGA and TAGA not due to Atlantic extra-tropical SST. • Late 20th century PDI upward trend (Emanuel 2005) probably not dominated by climate change, but internal variability (AMV) as hinted in DelSole et al. 2010 with a small contribution of climate change. • Next step analysis of PI in the 21 st century in the CMIP5 simulations. • Camargo, Ting & Kushnir, Climate Dynamics, 2012 TROPICAL CYCLONE ACTIVITY IN THE CMIP5 MODELS Suzana J. Camargo Lamont-Doherty Earth Observatory Columbia University Thanks to Haibo Liu and Naomi Naik for the CMIP5 dataset! OBJECTIVES • Analyze the tropical cyclone (TC) activity in the CMIP5 models: • Globally • Atlantic • Storms in the models and environmental variables • Comparison with CMIP3 • Choice of models: availability of 6-hourly data! TRACKS OF TCS IN HISTORICAL RUNS GLOBAL NUMBER OF TCS PER YEAR GENESIS POTENTIAL INDEX GLOBAL NUMBER OF TCS FUTURE & PRESENT TRACKS ATLANTIC AND EASTERN NORTH PACIFIC ATLANTIC NUMBER OF TROPICAL CYCLONES MRI TC ACTIVITY – 5 ENSEMBLES NUMBER OF ATLANTIC TROPICAL CYCLONES FUTURE & PRESENT CLUSTER ANALYSIS: TRACKS ATLANTIC Observations Kossin, Camargo and Sitkowski, J. Climate 2010 Models models are given below. The percentages were calculating by giving each storm a cluster assignment, and repeating the procedure 100 times, randomizing the order in which the storms were entered in the cluster analysis. Percentages marked in bold show statistically significant differences for that cluster between the percentages in the historical and the correspondent scenario (RCP4.5 and RCP8.5) using a t-test TRACK CHANGES ATLANTIC: Model MPI Scenario Historical RCP4.5 RCP8.5 Historical RCP4.5 RCP8.5 K=1 0.33 0.35 0.37 0.32 0.33 0.29 K=2 0.27 0.31 0.30 0.32 0.31 0.30 K=3 0.17 0.20 0.18 0.23 0.25 0.27 • MPI: MRI• Increase: subtropical storms • Increase: eastern subtropical storms • Large Decrease: Deep tropics storms • MRI: • Decrease: eastern subtropical storms • Increase: western subtropical storms K=4 0.22 0.13 0.13 0.13 0.10 0.14 GPI CHANGES GPI DIFFERENCES – COMPARISON WITH CMIP3 22 CMIP3 models – June to November GPI multi-model differences Vecchi and Soden, 2007 7 CMIP5 models: Northern Hemisphere: ASO Southern Hemisphere: JFM ial index (GPI) climatology between the PI DIFFERENCES: COMPARISON WITH CMIP3 22 CMIP3 models – June to November PI multi-model differences Vecchi and Soden, 2007 7 CMIP5 models: Northern Hemisphere: ASO Southern Hemisphere: JFM VERTICAL WIND SHEAR DIFFERENCES: COMPARISON WITH CMIP3 22 CMIP3 models – June to November PI multi-model differences Vecchi and Soden, 2007 7 CMIP5 models: Northern Hemisphere: ASO Southern Hemisphere: JFM SUMMARY • TC activity in the models analyzed not very realistic yet. • Models have a low bias in TC counts, that improves with horizontal resolution. • No robust changes in TC frequency: globally or regionally. • Environmental changes: very similar to CMIP3 results • Need of downscaling (statistical, dynamical) and/or higher resolution runs • Submitted to J. Climate, CMIP5 MAPP North American Climate special issue.