Possible Effects of Global Warming on Tropical Cyclone Activity Johnny Chan Guy Carpenter Asia-Pacific Climate Impact Centre School of Energy and Environment City University of Hong Kong
Outline Background Relationship between global warming and frequency of intense tropical cyclone occurrence Variations of tropical cyclone characteristics in the western North Pacific Summary 1
Background Global warming leads to an increase in the temperature near the earth s surface (land and ocean) an increase in the amount of water vapour in the atmosphere due to an increase in ocean temperature and a higher atmospheric temperature capable of holding more water vapour No study has definitively demonstrated that the dynamic factors are modified by global warming (although some have suggested an increase in vertical wind shear). 2
Background Due to global warming, the thermodynamic factors have become more favourable for tropical cyclone formation and development. To determine whether global warming has an impact on the frequency of occurrence of tropical cyclones or of intense cyclones, we need to examine whether the thermodynamic factors are related to the variations on such frequencies. A good proxy of the thermodynamic factors is the Maximum Potential Intensity (MPI) 3
Background MPI = f(ocean temperature, outflow temperature, net amount of energy available for convection) Because MPI gives the maximum possible intensity, a higher value of MPI summed over the ocean basin and over a season should imply a more thermodynamically energetic atmosphere, and more TCs could reach higher intensities a season with a higher value of MPI should have more intense TCs if the dominant control is thermodynamic 4
Atlantic 6 68 5 Cat45 NCEP MPI 66 Number of Cat45 4 3 2 64 62 60 58 MP PI 1 56 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Year 54 5
Western North Pacific 12 10 8 Jul-Nov Cat45 NCEP MPI 74 72 70 Cat4 45 6 68 66 MP PI 4 2 64 62 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 60 Year 6
21-year running correlations with NCat45 Ocean Basin Period Correlation (best track) Atlantic 1960-2007 0.45 1970-2007 0.59 1980-2007 0.63 Correlation (Kossin et al. 2007) 1979-2006 0.61 0.61 Western North Pacific 1960-2007 -0.01 1970-2007 -0.06 1980-2007 -0.08 1981-2006 -0.13-0.36 Eastern North Pacific 1960-2007 0.29 1970-2007 0.35 1980-2007 0.34 South Indian Ocean 1981-2007 0.35 South Pacific 1981-2007 0.03 7
21-year running correlations of MPI with NCat45 21-year running correlations with NCat45 0.8 0.6 0.4 0.2 0.0-0.2-0.4 Atlantic WNP ENP -0.6 1980 1985 1990 1995 2000 2005 8
Summary Thermodynamic control on the frequency of intense TCs is important only in the Atlantic Estimating the effect of global warming on the frequency of intense TCs therefore must also assess such an effect on the dynamic processes. 9
Western North Pacific Tropical Cyclones 10
Number and Intensity 11
Annual Number of TCs and Intense TCs in the WNP Annual number of TSs and Cat45 TCs over the WNP 40 14 TS 10-yr filtered TS 12 35 Cat45 10-yr filtered Cat45 10 30 8 25 6 4 20 2 15 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 0 12
Webster et al. s (2005) Science paper 13
No. of Category 4 and 5 Typhoons 1975-89 1990-2004 Number 75 115 Percentage 32 42 14
No. of Category 4 and 5 Typhoons 1960-74 1975-89 1990-2004 Number 105 75 115 Percentage 37 32 42 15
3.0 2.5 2.0 1.5 ACE vs.. VORT, SHEAR and MSE Science, 311, 1713b, Tellus 2007 ACE VORT EOF1 SHEAR EOF1 MSE EOF2 1.0 0.5 0.0-0.5-1.0-1.5-2.0-2.5 0.58 0.72 0.67 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 16
Wavelet Analysis of Intense Typhoon Occurrence Frequency 2-7 yr 16-32 yr Period A1 Period B Period A2 17
Period A1 Period A2 Period B Ocean Temperature Anomalies 18
Vertical Wind Shear Period A1 minus Period B Period A2 minus Period B 19
Period A1 Period A2 Period B Frequency of Occurrence of Intense Typhoons 20
Difference in the Frequency of Occurrence of Intense Typhoons Proceedings, Royal Society A (2008) Period A1 minus Period B Period A2 minus Period B Blue shading: 95% Green shading: 90% 21
Track and Landfall Variations 22
No. of TCs Making Landfall in Japan and Korea Every 5-year period (1970-2004) No. of T ropical Cyclones Making Landfall in Japan and Korea 30 25 20 Number 15 10 5 0 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 Year 23
No. of Typhoons Making Landfall in East China Every 5-year period (1960-2005) 24
No. of Typhoons Making Landfall in South China Every 5-year period (1960-2005) 25
Standardizedanomalies standardized anomalies Standardized edanomalies standardized d standardized anomalies 4 3 2 1 0-1 -2 Variations of Landfall in Each Area at Various Oscillation Periods STC -3 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 4 3 2 1 0-1 -2 South China, Philippines and Vietnam East China Year MTC original 2-8 yr original 2-8 yr 8-16 yr 8-16 yr 16-32 yr 16-32 yr -3 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 Japan/Korea Year 2 1.5 1 0.5 0-0.5-1 -1.5 2 1.5 1 0.5 0-0.5-1 -1.5 26 standardized anomalies Standardizedanomalies Standardiz dizedanomalies standardiz zed anomalies standardized anomalies
TC occurrence anomalies 1977-88 Pattern 1 Pattern 2 1964-76 Pattern 3 1989-97 27 Patterns of TC occurrence anomalies
Summary No significant trend in any of the TC characteristics (number, intensity, track types, landfall locations) can be identified. In other words, TC activity in the western North Pacific does not follow the trend in the global increase in atmospheric or sea- surface temperature. Instead, all such characteristics go through large interannual and interdecadal variations. 28
Summary Such variations are very much related and apparently caused by similar variations in the planetary-scale atmospheric and oceanographic features that also do not have the same trend as the global increase in air temperature Unless the temporal variations of such features become linear, these TC characteristics are not expected to vary linearly with time. 29
Summary Even if the observed global warming has an effect, it is probably in the noise level relative to the large interdecadal variations and therefore is not detectable. 30