Observed records of Atlantic hurricane activity (e.g. Emanuel, 2007) show a strong correlation, on multi-year time-scales, between local tropical Atlantic sea surface temperatures (SSTs) and the Power Dissipation Index (PDI). PDI is an aggregate measure of Atlantic hurricane activity, combining frequency, intensity, and duration of hurricanes in a single index. Both Atlantic SSTs and PDI have risen sharply since the 1970s, and there is some evidence that PDI levels in recent years are higher than in the previous active Atlantic hurricane era in the 1950s and 60s. (Source: http://www.gfdl.noaa.gov/global-warming-and-hurricanes)
George,
Increasing surface water temperature is definitely a key control of Hurricane (and other tropical cyclone) occurrence. So increasing surface water temperatures could increase likelihood. However, another issue is upper atmospheric wind shear: which is predicted to increase. These winds disrupt hurricane systems.
As far as I'm aware the impact of climate change on hurricanes is still very much open. I think there is a difference between frequency and magnitude also (certainly for Europe) with models predicting a variety of outcomes.
Matt
What Matt says is right. However, it is the upper layer heat content (SST and ocean subsurface temperatures integrated to the level of mixed layer depth) which is a better index than SST. My argument is supported by numerous research studies that show a dramatic increase in mixed layer depth ahead of the hurricane (in north west quarter), storing more energy of the winds to be released subsequently for its intensification and propelling motion. The climate warming impact, probably lies, in this aspect.
In fact, proper statistical treatments of the detection function for Atlantic Basin hurricanes (i.e. - before radar, many hurricanes were not detected) shows that there is no statistically significant increase in hurricanes per decade going back to 1900. This is a myth that has been perpetuated by a few strong storms that made US landfall.
Hurricane frequencies vary considerably from decade to decade and are well known to have a 20 year "period" .
Our preliminary analysis (through 2009 - this work is continuing) using a correction for incompleteness yields the following:
1900-1950 = 55.5 +/- 5.4 storms per decade
1950-present = 61.6 +/- 9.1
Compare this to the raw counts (only detected storms)
1900-1950 = 40.4 +/- 3.9 storms per decade
1950-present = 61.6 +/- 9.1
to see that incompleteness in the pre-radar era is large.
The higher variance in the 1950-2009 counts is due to the fact that
the 50's were the beginning of the next hurricane max cycle which declined considerably in the 1970s.
The raw count data is here:
1950 69 (storms per decade)
1960 61
1970 49
1980 52
1990 64
2000 72
This counting analysis is only for systems that become hurricanes and does not include tropical storms.
The decade of the 2000s is not higher frequency than past decades - the percentage of storms classed as Cat 3+, however, is higher. More on that later.
Please note a point in last sentence of Dr. Botham's note. The category 3+ incidences have increased. It is not the total number, but the intensification of hurricanes, which is the likely impact of climate warming scenario. The ocean heat content increase in upper layers of ocean due to climate warming plausibly points to this happening.
A bit more:
1. Emanuel's PDI (Power Dissipation Index) is one way to measure total "hurricane power" - its not clear a) if the PDI can be reliably calibrated and b) if it can be applied consistently form system to system - nonetheless simple application suggests an effect. We are developing the DII (Dynamical Intensity Index) which is analogous to a dynamical mass, namely V^2 R where V is the maximum wind intensity and R is the radius of hurricane force winds. The available data on measured radii or reasonable estimated radii goes back to 1988. Preliminary results are consistent with the PDI - some of the stronger storms in the last 10 years have had very large hurricane force wind radii (up to 175 km in some cases) with max hurricane winds of 180 km/hr or a bit higher. For such a system V^2 R is about a factor of 10- 20 higher than average systems.
2. The 2005 Season (Katrina, Rita, Wilma) was driven primarily by an anomalously warm gulf of Mexico loop current. Variations in the temperature and depth of this loop current are not very well understood but in 2005 it was a maximum strength. The image below shows that Katrina's path and amplification was entirely driven by this and the loop current remained strong to amplify Rita and Wilma - whether this was at all coupled of more global Atlantic ocean properties is unclear.
3. Hurricanes amplify for TWO reasons and both are equally weighted:
a) Sea Surface Temperature (SST) - generally speaking, SST needs to be above 26 C and every degree above that significantly contains more "thermal power" to be tapped by the system.
b) BUT, the rising convective activity has to not be SHEARED off by upper level winds. It is the shearing that tends to kill or deflate most storms. The complicated dance between SST and the height and latitude of the subtropical jet stream produces the probability space in which hurricane amplification lives.
4. What our analysis is showing is that hurricane "spin-up times" are accelerating. We measure the "spin-up" time as the time it takes to go from central pressure of 1005 mb to if and when the hurricane gets to Cat 3 level. The decade of the 2000s for instance, sees significant decreases in spin up times compared to the 1960s. In general these spin up times or 3-5 days, more weighted toward 5 days - but recent years have seen some times as low as 50 hours (which is remarkable). But this still needs to be worked on in more detail - the preliminary results are tantalizing but need much more rigorous statistical treatment.
The problem is that no increasing trend has been evidenced for western Pacific, where the most frequent tropical cyclones occur each year. All the indicators, including landing and generating frequency, TC-induced precipitation amount, and extremely intense Typhoon, have not shown a significant upward trend during the past six decades. If apply the PDI index, we will not find an increasing trend in western Pacific Ocean.
Well we have worked on North Indian Ocean Cyclone data and correlated with SST and ENSO index. It is found that very little long term variations in the SST of Bay of Bengal, However, the intensity of Cyclones have increased about 20% in the cyclone months specially in May and November months. Seems the vertical wind shear favors the development and intensification of cyclones. The statistical analysis of more than 100 years data shows that there is no significant change in the frequency of cyclones in the Bay of Bengal and Arabian Sea. Study shows that ENSO has no significant impact on Cyclone activity over the Arabian Sea. During ENSO Frequency of cyclones diminishes in May and Nov. During ENSO Years the total cyclonic disturbances in the North Indian Ocean reduces
It is not the frequency but rather the power and destruction which will likely increase over the next decades, see the excellent article of Kerry Emmanuel a brillant scientist of MIT
https://www.nature.com/nature/journal/v436/n7051/pdf/nature03906.pdf
The surfaces of most tropical oceans have warmed by 0.25-0.5 degrees Celsius during the past several decades. Theory predicts that hurricane wind speeds should increase about 5% for every 1 degree Centigrade increase in tropical ocean temperature (Emanuel, 1987). Although there are controversies, it is difficult to believe that with global warming (increase of energy) the power of hurricane will stay the same.
Other article of interest
http://science.sciencemag.org/content/309/5742/1844.full
Finally, with the increase of coastal population, there will certainly be increase of damage and loss of life due to hurricane.
I note here that there are important differences in the behavior of Atlantic Basin Hurricanes, (some day I will publish my ongoing timestep trajectory analysis of literallly 10,000 measurements) compared to Pacific Typhoons. For Typhoons there does appear to be both an increase in frequency and definitely an increase in strength.
Our analysis of Atlantic Basin hurricanes, still ongoing, can offer this
1. There does not seem to be an increase in frequency
2. There does seem to be a shift in mean averaged location (or hurricane trajectory) farther to the South
3. If one defines a spin up time as the time it takes to go from 1000 mb to 970 mb - the data suggest that hurricanes are evolving faster
4. It does look, with significant error bars, that the fraction of hurricanes that develop to Cat 3+ is higher and this is consistent with Emanuel
The current research into the effects of climate change on tropical storms demonstrates not only the virtues and transparency of the scientific method at work, but rebuts the frequent suggestion that scientists fit their findings to a pre-determined agenda in support of climate change. In the case of storm frequency, there is no consensus and reputable scientists have two diametrically opposed theories about increasing frequencies of such events.
The background to these enquiries stems from a simple observation: extra heat in the air or the oceans is a form of energy, and storms are driven by such energy. What we do not know is whether we might see more storms as a result of extra energy or, as other researchers believe, the storms may grow more intense, but the number might actually diminish.
What do the records show? According to the Pew Centre, “Globally, there is an average of about 90 tropical storms a year”. The IPCC AR4 report (2007) says regarding global tropical storms: "There is no clear trend in the annual numbers [i.e. frequency] of tropical cyclones."
But this graph, also from the Pew Centre, shows a 40% increase in North Atlantic tropical storms over the historic maximum of the mid-1950, which at the time was considered extreme:
Indeed the El Nino data is important - there certainly is NO increase in El Nino and yes, the most recent event (starting Oct 15) is the third strongest on record, but the last most recent event was in 1997. So there is zero evidence suggesting that El Nino is systematically increase in response to warming oceans. The Maddej-Julian Oscillation (MJO) however, is another matter, and there may be some effects there.
The MJO is now recognized as likely the most important perturbation on the planet which a) directly effects Asian Monsoons and b) likely couples to weather patterns on the West Coast of the US
I also attach an interesting wavelet analysis of El Nino - not the approximate 40 year period of suppression that occurred in the mid 20th century: (note that this data only goes through 2010)
- Badges
- Science topic
- Similar topics
- Geoscience
- Atmospheric Sciences
- Climate Science
- Climate Change