Southern Hemisphere Sea Surface Temperatures and
Southern Hemisphere Sea Surface Temperatures and Tropical Cyclone Activity: Implications for Climate Change John L McBride & Hamish Ramsay Centre for Australian Weather and Climate Research (CAWCR) This research is part of the IOCI project
Global Tropical Cyclones: 20 Years of Tracks (1985 – 2005)
Global Tropical Cyclones: 20 Years of Tracks (1985 – 2005)
North Indian
NorthWest Pacific
North East Pacific
North Atlantic
South Indian South Pacific
Monthly Sea Surface Temperature time- series NorthWest Pacific Ocean
North Atlantic Ocean
South East Indian Ocean
South West Pacific Ocean
In all tropical cyclone basins around the globe, sea surface temperatures have increased by approximately 0.5C over the past 30-40 years.
Tropical cyclone research community as a whole agrees with the assessment of the IPCC Report (2007): this sea surface temperature increase can be mainly attributed to anthropogenic climate change
Tropical cyclones form only over the tropical oceans
Energy source for tropical cyclones is the latent heat release in rainfall: heat is extracted by evaporation from the underlying warm tropical oceans
Tropical cyclones form only over the tropical oceans
Energy source for tropical cyclones is the latent heat release in rainfall: heat is extracted by evaporation from the underlying warm tropical oceans
Has there been an increase in tropical cyclone activity
associated with the sea surface temperature increase that has occurred?
Has there been an increase in tropical cyclone activity associated with the sea surface temperature increase that has occurred? For the North Atlantic Ocean YES:
There has been a statistically significant increase over the past 30 years in TC numbers, Proportion of intense TC’s, TC Power Dissipation Index PDI, etc
Best track
Satellite (Kossin et al GRL 2007)
For the other Ocean Basins: Not so clear Proportion intense cyclones
PDI
West Pacific and South Pacific PDI and proportion of intense cyclones
For the Indian Ocean (North and South) it is difficult to determine whether there has been a trend: Intensities are determined by infrared satellite algorithms: There has been a dramatic change in the satellite viewing angle.
Source: NOAA National Climate Data Centre, Asheville USA. Refs: Knapp and Kossin, 2007; Kossin et al (2007)
For the Indian Ocean (North and South) it is difficult to determine whether there has been a trend: Intensities are determined by infrared satellite algorithms: There has been a dramatic change in the satellite viewing angle.
South Indian Ocean PDI, with and w/out viewing angle correction
Source: NOAA National Climate Data Centre, Asheville USA. Refs: Knapp and Kossin, 2007; Kossin et al (2007)
Why can we not see a trend in the number of intense tropical cyclones? • •
•
Signal to noise ratio: Very large interannual variability in tropical cyclone activity Non homogeneous data: Observational techniques to detect the intensity of a cyclone depend fundamentally on interpretation of infrared satellite imagery. With changes in availability of satellite data, changes in spatial resolution, availability of multi-channel infrared imagery, training changes, technique development, availability of forecaster work-stations etc, there has been a continuous change in our ability to determine the intensity of a tropical cyclone. Theoretical models (Maximum Potential Intensity Theory) tells us that the signal should only be in the most intense cyclones -- Thus we need a very long time series to collect enough cases to determine a trend.
Why can we not see a trend in the number of intense tropical cyclones? • Physical association – Possibly there is no simple relationship between sea surface temperature and tropical cyclone activity
Tropical cyclone time series for South East Indian Ocean
Sea Surface Temperature time series for South East Indian Ocean.
Scatter-plot No of Tropical Cyclones versus Sea Surface Temperature South East Indian Ocean
Correlation = +0.42
Correlation of sea surface temperature at each gridpoint with Nino-3.4
For each calendar month
Correlation of sea surface temperature at each gridpoint with Nino-3.3
For each calendar month East Pacific warm area Boomerang pattern of opposite sign Indian Ocean response (same sign)
Seasonal evolution Boomerang sits over Australian region in August-SeptemberOctober (cold anomalies during El Nino)
East-Pacific anomaly has broken through in Jan-March (warm anomaly in El Nino)
Seasonal evolution Boomerang sits over Australian region in August-SeptemberOctober (cold anomalies during El Nino)
East-Pacific anomaly has broken through in Jan-March (warm anomaly in El Nino)
Seasonal evolution of the Boomerang pattern, discussed by McBride, Haylock, Nicholls: J. Clim 2003 -- identified as ENSO pattern.
Same for Power Dissipation Index (PDI)
Correlation with SST at each gridpoint
So,
What is happening here? Tropical cyclone activity in each region (of the Southern Hemisphere) is dominated, not by the local Sea Surface Temperature), rather by the planetary-scale sea surface temperature pattern of ENSO
So,
What is happening here?
Tropical cyclone activity ijn each region (of the Southern Hemisphere) is dominated, not by the local Sea Surface Temperature), rather it is dominated by the planetaryscale sea surface temperature pattern of ENSO East Pacific warm pool
Indian Ocean response
Boomerang pattern
Tropical cyclone activity, through --- Relative vorticity (increase of westerlies equatorward of monsoon trough) --- Upward motion (movement of Walker circulation)
ENSO
---- Possible vertical shear (smaller effect than relative vorticity and upward motion)
Sea Surface Temperature through ------Ocean dynamics (equatorially-trapped waves in the wave guide ----- Surface wind response effects surface fluxes
Thus, ENSO dominates the interannual variability of both: SST, Tropical cyclones
Thus relationships between (SST, Tropical cyclones) Simply reflects the fact that both are dominated by ENSO variations
This can lead to counter-intuitive effects:
Cyclone numbers (Jan-March) versus SST -- Negative association: more cyclones when sea is cold
Same for number of intense cyclones (L.E 955 hPa central pressure)…. More when sst is cold --- why: because it is dominated by ENSO, working through vorticity, vertical motion etc.
Summary:
Very clear signal: there is no in-situ relationship between Tropical cyclone activity and sea surface temperature in the Southern Hemisphere ocean basins:
Interannual variability of tropical cyclones in the Southern hemisphere is dominated by changes in relative vorticity and in upward vertical motion, related to Walker Cell response of ENSO
Implications for Climate Change: Hypothesis: The same physics should apply: The locations and intensity of regions of upward vertical motion across tropical Southern Hemisphere tropics, and location and magnitude of vorticity in the monsoon trough will determine the response of Southern hemisphere tropical cyclones to climate change.
Thank You
Global Tropical Cyclones: 20 Years of Tracks (1985 – 2005)
Global Tropical Cyclones: 20 Years of Tracks (1985 – 2005)
North Indian
NorthWest Pacific
North East Pacific
North Atlantic
South Indian South Pacific
Monthly Sea Surface Temperature time- series NorthWest Pacific Ocean
North Atlantic Ocean
South East Indian Ocean
South West Pacific Ocean
In all tropical cyclone basins around the globe, sea surface temperatures have increased by approximately 0.5C over the past 30-40 years.
Tropical cyclone research community as a whole agrees with the assessment of the IPCC Report (2007): this sea surface temperature increase can be mainly attributed to anthropogenic climate change
Tropical cyclones form only over the tropical oceans
Energy source for tropical cyclones is the latent heat release in rainfall: heat is extracted by evaporation from the underlying warm tropical oceans
Tropical cyclones form only over the tropical oceans
Energy source for tropical cyclones is the latent heat release in rainfall: heat is extracted by evaporation from the underlying warm tropical oceans
Has there been an increase in tropical cyclone activity
associated with the sea surface temperature increase that has occurred?
Has there been an increase in tropical cyclone activity associated with the sea surface temperature increase that has occurred? For the North Atlantic Ocean YES:
There has been a statistically significant increase over the past 30 years in TC numbers, Proportion of intense TC’s, TC Power Dissipation Index PDI, etc
Best track
Satellite (Kossin et al GRL 2007)
For the other Ocean Basins: Not so clear Proportion intense cyclones
PDI
West Pacific and South Pacific PDI and proportion of intense cyclones
For the Indian Ocean (North and South) it is difficult to determine whether there has been a trend: Intensities are determined by infrared satellite algorithms: There has been a dramatic change in the satellite viewing angle.
Source: NOAA National Climate Data Centre, Asheville USA. Refs: Knapp and Kossin, 2007; Kossin et al (2007)
For the Indian Ocean (North and South) it is difficult to determine whether there has been a trend: Intensities are determined by infrared satellite algorithms: There has been a dramatic change in the satellite viewing angle.
South Indian Ocean PDI, with and w/out viewing angle correction
Source: NOAA National Climate Data Centre, Asheville USA. Refs: Knapp and Kossin, 2007; Kossin et al (2007)
Why can we not see a trend in the number of intense tropical cyclones? • •
•
Signal to noise ratio: Very large interannual variability in tropical cyclone activity Non homogeneous data: Observational techniques to detect the intensity of a cyclone depend fundamentally on interpretation of infrared satellite imagery. With changes in availability of satellite data, changes in spatial resolution, availability of multi-channel infrared imagery, training changes, technique development, availability of forecaster work-stations etc, there has been a continuous change in our ability to determine the intensity of a tropical cyclone. Theoretical models (Maximum Potential Intensity Theory) tells us that the signal should only be in the most intense cyclones -- Thus we need a very long time series to collect enough cases to determine a trend.
Why can we not see a trend in the number of intense tropical cyclones? • Physical association – Possibly there is no simple relationship between sea surface temperature and tropical cyclone activity
Tropical cyclone time series for South East Indian Ocean
Sea Surface Temperature time series for South East Indian Ocean.
Scatter-plot No of Tropical Cyclones versus Sea Surface Temperature South East Indian Ocean
Correlation = +0.42
Correlation of sea surface temperature at each gridpoint with Nino-3.4
For each calendar month
Correlation of sea surface temperature at each gridpoint with Nino-3.3
For each calendar month East Pacific warm area Boomerang pattern of opposite sign Indian Ocean response (same sign)
Seasonal evolution Boomerang sits over Australian region in August-SeptemberOctober (cold anomalies during El Nino)
East-Pacific anomaly has broken through in Jan-March (warm anomaly in El Nino)
Seasonal evolution Boomerang sits over Australian region in August-SeptemberOctober (cold anomalies during El Nino)
East-Pacific anomaly has broken through in Jan-March (warm anomaly in El Nino)
Seasonal evolution of the Boomerang pattern, discussed by McBride, Haylock, Nicholls: J. Clim 2003 -- identified as ENSO pattern.
Same for Power Dissipation Index (PDI)
Correlation with SST at each gridpoint
So,
What is happening here? Tropical cyclone activity in each region (of the Southern Hemisphere) is dominated, not by the local Sea Surface Temperature), rather by the planetary-scale sea surface temperature pattern of ENSO
So,
What is happening here?
Tropical cyclone activity ijn each region (of the Southern Hemisphere) is dominated, not by the local Sea Surface Temperature), rather it is dominated by the planetaryscale sea surface temperature pattern of ENSO East Pacific warm pool
Indian Ocean response
Boomerang pattern
Tropical cyclone activity, through --- Relative vorticity (increase of westerlies equatorward of monsoon trough) --- Upward motion (movement of Walker circulation)
ENSO
---- Possible vertical shear (smaller effect than relative vorticity and upward motion)
Sea Surface Temperature through ------Ocean dynamics (equatorially-trapped waves in the wave guide ----- Surface wind response effects surface fluxes
Thus, ENSO dominates the interannual variability of both: SST, Tropical cyclones
Thus relationships between (SST, Tropical cyclones) Simply reflects the fact that both are dominated by ENSO variations
This can lead to counter-intuitive effects:
Cyclone numbers (Jan-March) versus SST -- Negative association: more cyclones when sea is cold
Same for number of intense cyclones (L.E 955 hPa central pressure)…. More when sst is cold --- why: because it is dominated by ENSO, working through vorticity, vertical motion etc.
Summary:
Very clear signal: there is no in-situ relationship between Tropical cyclone activity and sea surface temperature in the Southern Hemisphere ocean basins:
Interannual variability of tropical cyclones in the Southern hemisphere is dominated by changes in relative vorticity and in upward vertical motion, related to Walker Cell response of ENSO
Implications for Climate Change: Hypothesis: The same physics should apply: The locations and intensity of regions of upward vertical motion across tropical Southern Hemisphere tropics, and location and magnitude of vorticity in the monsoon trough will determine the response of Southern hemisphere tropical cyclones to climate change.
Thank You
