Abstract

In a 2008 paper, using satellite-derived wind speed estimates from tropical cyclones over the 25-yr period 1981–2006, we showed the strongest tropical cyclones getting stronger. We related the increasing intensity to rising ocean temperatures consistent with theory. Oceans have continued to warm since that paper was published, so the intensity of the strongest cyclones should have continued upward as well. Here I show that this is the case, with increases in the upper-quantile intensities of global tropical cyclones amounting to between 3.5% and 4.5% in the period 2007–19 relative to the earlier base period (1981–2006). All basins individually show upward intensity trends for at least one upper quantile considered, with the North Atlantic and western North Pacific basins showing the steepest and most consistent trends across the quantiles.

Science is validated through predictions. Using satellite-derived wind speed estimates from tropical cyclones over the 25-yr period between 1981 and 2006, Elsner et al. (2008) showed that the strongest tropical cyclones worldwide were getting stronger. They related this increase in tropical cyclone intensity to a corresponding rise in ocean temperatures consistent with theory (Emanuel 1988). The oceans have continued to warm since that paper was published, so I would anticipate that the upward trend in the intensity of the strongest tropical cyclones has continued.

To check on this, I consider all global tropical cyclones with wind speeds of at least 33 m s‒1 occurring over the six basins. I divide the data into two epochs: the period 1981–2006 corresponding to the years used in Elsner et al. (2008) and the period 2007–19 corresponding to the 13 years following that publication. The data are 3-hourly estimates of the wind speed maximum made by forecast operational centers and compiled by IBTrACS (Knapp et al. 2010, 2018). Note that the “data” used in Elsner et al. (2008) are regression-estimated intensities (wind speed maxima) from satellite imagery, so the wind speed and trend magnitudes are not directly comparable due to the “regression-to-the-mean” effect. However, to be consistent with the earlier work I use the single highest wind speed over the lifetime of each tropical cyclone. I compute quantile highest wind speeds over the two distinct epochs and quantify the change in these quantile wind speeds in terms of the percentage increase from the earlier base period (1981–2006).

I note that the 75th-percentile highest wind speed for the set of global tropical cyclones having lifetime highest wind speeds of at least 33 m s‒1 has increased by 4% from 61.7 m s‒1 during the earlier period to 64.3 m s‒1 during the later period. Further, I note that the 90th-percentile wind speed has increased by 3.6% and the 95th-percentile wind speed has increased by 4.3% from 72 m s‒1 during the earlier period to 75.2 m s‒1 over the later period. All six basins individually show upward intensity trends for at least one upper quantile, with the North Atlantic and western North Pacific basins showing the steepest and also the most consistent trends across the spread of quantiles examined (Fig. 1).

Fig. 1.

Percentile wind speeds in two distinct epochs by major tropical cyclone basins. The epoch years [chosen based on the pre- and post-Elsner et al. (2008) study] and the number of tropical cyclones with lifetime highest wind speeds of at least 33 m s‒1 (n) are shown at the top of the left and right columns for each basin. The wind speeds corresponding to the 75th, 90th, and 95th percentiles are given in the respective columns below each epoch. The changes are noted by the slope of the line segments colored by percentiles (75th in light gray, 90th in gray, and 95th in black). The vertical scales are the same across the basin.

Fig. 1.

Percentile wind speeds in two distinct epochs by major tropical cyclone basins. The epoch years [chosen based on the pre- and post-Elsner et al. (2008) study] and the number of tropical cyclones with lifetime highest wind speeds of at least 33 m s‒1 (n) are shown at the top of the left and right columns for each basin. The wind speeds corresponding to the 75th, 90th, and 95th percentiles are given in the respective columns below each epoch. The changes are noted by the slope of the line segments colored by percentiles (75th in light gray, 90th in gray, and 95th in black). The vertical scales are the same across the basin.

For example, the 95th-percentile wind speed for the set of western North Pacific typhoons increased by 6.4% from 74.6 m s‒1 during the earlier period to 79.7 m s‒1 during the later period. While the 95th-percentile wind speed for the set of North Atlantic hurricanes increased by 4.8% from 72 m s‒1 during the earlier period to 75.6 m s‒1 during the later period. The largest percentage increase across the basins and quantiles is 7.4% for the 75th-percentile wind speed over the western North Pacific. The largest percentage increase in the Southern Hemisphere is 5.4% for the 75th-percentile wind speed over the South Pacific. The interbasin differences in percentage increases are not large and no basin shows a significant change (at the α = 0.001 level) in the number of cyclones reaching at least 33 m s‒1 between the two epochs, although the western North Pacific had three fewer typhoons per year on average during the latest epoch.

Wind speed values from the IBTrACS dataset are based on estimates made by operational centers. The centers rely on satellite images and algorithms that identify and quantify specific features in the images (Velden et al. 2006). Changes in satellite technology likely contribute to the observed trends presented here. Kossin et al. (2020) indicate that the observed upward trend in the fraction of all tropical cyclone reports of at least 33 m s‒1 (hurricane intensity in the Atlantic basin) that are at least 50 m s‒1 (major hurricane intensity in the Atlantic basin) globally over the period 1979–2017 might be overestimated by 50% due to the inhomogeneity of satellite data. The overestimate is based on comparing the IBTrACS wind speeds with wind speeds estimated from the Hurricane Satellite (HURSAT) dataset (Knapp and Kossin 2007). The HURSAT wind speeds result from applying an automated Dvorak technique to degraded satellite images (8-km resolution). However, I note the IBTrACS fraction of strong tropical cyclones (wind speeds of at least 50 m s‒1) is 80% of the HURSAT fraction during the first half of this nearly 40-yr period but increases to 93% of the HURSAT fraction during the second half of the period (Table 1 of Kossin et al. 2020). This increase in percent matching is evidence that the IBTrACS wind speed estimates are now more homogeneous, so the overestimation is likely lower and perhaps much lower starting with 2007. In fact, over the most recent 13-yr period (since 2007) when available satellite technology is stable, I find an upward trend in the global 90th-percentile highest wind speed of 3.6% from 69.5 m s‒1 during the first half of the period (2007–13) to 72 m s‒1 during the later period (2014–19).

This postpublication analysis shows a continued increase in the intensity of the strongest hurricanes worldwide, as first identified in Elsner et al. (2008), using data only through 2006. The results were anticipated given the continued heating of the tropical oceans (Cheng et al. 2018), although other factors like tropopause temperature changes and changes in the amount of wind shear also play a role, making it difficult to anticipate near-future changes on the time scale of a few years.

References

References
Cheng
,
L. J.
,
G. J.
Wang
,
J. P.
Abraham
, and
G.
Huang
,
2018
:
Decadal ocean heat redistribution since the late 1990s and its association with key climate modes
.
Climate
,
6
,
91
, https://doi.org/10.3390/cli6040091.
Elsner
,
J. B.
,
J. P.
Kossin
, and
T. H.
Jagger
,
2008
:
The increasing intensity of the strongest tropical cyclones
.
Nature
,
455
,
92
95
, https://doi.org/10.1038/nature07234
Emanuel
,
K. A.
,
1988
:
The maximum intensity of hurricanes
.
J. Atmos. Sci.
,
45
,
1143
1155
, https://doi.org/10.1175/1520-0469(1988)045<1143:TMIOH>2.0.CO;2.
Knapp
,
K. R.
, and
J. P.
Kossin
,
2007
:
New global tropical cyclone data set from ISCCP B1 geostationary satellite observations
.
J. Appl. Remote Sens.
,
1
,
013505
, https://doi.org/10.1117/1.2712816.
Knapp
,
K. R.
,
M. C.
Kruk
,
D. H.
Levinson
,
H. J.
Diamond
, and
C. J.
Neumann
,
2010
:
The International Best Track Archive for Climate Stewardship (IBTrACS): Unifying tropical cyclone best track data
.
Bull. Amer. Meteor. Soc.
,
91
,
363
376
, https://doi.org/10.1175/2009BAMS2755.1.
Knapp
,
K. R.
,
H. J.
Diamond
,
J. P.
Kossin
,
M. C.
Kruk
, and
C. J.
Schreck
,
2018
: International Best Track Archive for Climate Stewardship (IBTrACS) Project, version 4 [WP, NP, NA, SI, NI, SP].
NOAA/National Centers for Environmental Information
, accessed 1 March 2020, https://doi.org/10.25921/82ty-9e16.
Kossin
,
J. P.
,
K. R.
Knapp
,
T. L.
Olander
, and
C. S.
Velden
,
2020
:
Global increase in major tropical cyclone exceedance probability over the past four decades
.
Proc. Natl. Acad. Sci. USA
,
117
,
11975
11980
, https://doi.org/10.1073/pnas.1920849117.
Velden
,
C.
, and et al
,
2006
:
The Dvorak tropical cyclone intensity estimation technique: A satellite-based method that has endured for over 30 years
.
Bull. Amer. Meteor. Soc.
,
87
,
1195
1210
, https://doi.org/10.1175/BAMS-87-9-1195.
For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).