Is climate change making tropical cyclones worse?
The US National Oceanic and Atmospheric Administration (US NOAA) concludes the following:
“It is premature to conclude that human activities… have already had a detectable impact on Atlantic hurricane or global tropical cyclone activity. That said, human activities may have already caused changes that are not yet detectable.”
It’s a surprising conclusion if we base our conclusion on anecdotal and intuitive evidence, or the media. Yet whilst tropical cyclones seem to have become increasingly strong, the statistical analysis that attempts to find significant trends in the noise of meteorology is one of the thorniest issues for scientists. In this post, we’ll examine the recent studies, which – through computer modelling and historical data manipulation – highlight both the key and unknown influences on the tropical cyclone activity.
Sea surface temperatures (SSTs)
Due to the fact that tropical cyclones (the encompassing term for the identical phenomena of hurricanes, cyclones and typhoons) can only form over tropical surface water in excess of 26oC, it would make sense that as sea surface temperatures (SSTs) warm due to climate change, the frequency of tropical cyclones would increase. This should be particularly true for the Atlantic which has seen a sharp increase since the 1970s in SSTs. However, as the eminent cyclone scientist Kerry Emanuel explains in a 2013 paper:
“Sea surface temperature plays no well-defined direct role in any existing theory of tropical cyclones, even though it is widely used in empirical relationships for genesis and storm intensity and by forecasters trying to predict genesis and intensity change. Physically, the atmosphere responds to the ocean mostly through radiative fluxes, which depend, among other things, on the ocean temperature relative to the temperature at various levels in the atmosphere, and on … near-surface wind speed [amongst other differences in saturation and temperature]. It is the difference that matters, not the absolute value of the sea surface temperature.”
A 2015 analysis also emphasizes that absolute SSTs are an erroneous measure to use in modelling. This study highlights that the difference in temperature between tropical SSTs and the basin SSTs at the latitudes for cyclogenesis – called relative SST – is what drives the intensity of rainfall, rather than absolute SST. Similarly, stratospheric cooling due to climate change, and changes in aerosols at the tropopause may influence cyclone activity. Moreover, attributing SST change to anthropogenic climate change is a risky business: in the Atlantic basin, a climate cycle known as the Atlantic Multidecadal Oscillation operates over a 60-80-year period. We are currently at its peak warmth, perhaps explaining the 2017 hyperactive season; this influence must also be factored into whatever models are constructed, with a reminder that this change is natural.
Historical data trends
Analysing historical data to understand whether there are any long-term shifts in cyclonic activity would appear the best way to empirically demonstrate a trend. An upward trend does exist in raw hurricane count in the Atlantic from 1880 to 2005, albeit not overly strong. The problem, however, is the word ‘raw’: before the advent of satellite monitoring, records of tropical cyclones derives from ship logbooks. The density of ships across the Atlantic in the early parts of these records was much sparser than today’s density, therefore underrepresenting the true number of non-landfalling storms. When the US NOAA attempted to homogenize the data to account for these differences, between 1878 and 2006, the increase was only 1.60 storms per century, statistically indistinguishable from 0.
Nonetheless, historical data can be useful: a clear, surprising and statistically significant trend is that tropical cyclones have migrated towards the poles over time. The study that concluded this plotted the point at which tropical cyclones reached their lifetime maximum intensity (LMI), negating the problems of data inhomogeneity associated with frequency. The study concludes a poleward migration of 53km per decade in the Northern Hemisphere, and 62 km per decade in the Southern Hemisphere.
A small caveat exists, however: the trend was not evident in the Atlantic Basin, although both Pacific Basins and the South Indian Basin show strong changes. The uncertainty increases when scientists try to explain the phenomenon: changes in stratospheric aerosol concentrations, decreases in wind shear, and hotter SSTs further northward are all potential explanations but no consensus has been reached since the study’s publication in 2014.
A second way to look at the issue is to ask: has the hurricane season lengthened?
A 2008 study claims that, yes, it has lengthened – although this comes with high uncertainty – but not necessarily due to global warming, citing localized SST variability and climate variability. A 2015 study argues no: perhaps, the North Pacific is lengthening, but the South Pacific season has shortened, a trend repeated in other basins. No consensus has been reached, despite anecdotal evidence from unusual cyclonic activity outside their typical seasons: Hurricane Alex and Typhoon Pali both spawned in January in 2016; Hurricane Arlene in April 2017.
Such occurrences underscore the difficulties inherent to meteorology: climate and weather are separate systems, and trying to find a pattern within the noise created by the large variability of weather – and then extrapolating it to oceanwide basins – is a nearly impossible task to complete with the statistically-required precision to yield a definitive trend. This is one reason why the poleward migration study is so fascinating, in that the trend the analysis yields is robust and geographically significant.
So, why don’t models predict an increase in total cyclonic activity? The Southwest Atlantic Basin can provide a glimpse into the factors inhibiting tropical cyclone production: simply too much wind shear exists to allow tropical cyclones to develop, with only one hurricane (Catarina in 2004) having struck Brazil in recent memory. For a tropical depression to intensify into a tropical cyclone, the warm core (powered by the latent heat of evaporation, as the water vapour from the ocean condenses) must be positioned over the centre of lowest air pressure at sea level. Without this mechanism of intensification via latent heat, the depression will simply dissipate. Horizontal wind shear greater than 10 ms-1 inhibits intensification, and some models suggest that if climate change increases wind speed, tropical cyclones will be inhibited.
the above map shows the recorded tracks of global tropical cyclonic activity between 1985 and 2005; notice how there is a lone hurricane (Catarina) in the South Atlantic, due to the higher wind shear present there in comparison to the density of cyclone in other basins
This hypothesis goes some way to explain the seemingly contradictory conclusion that, rather than increasing, the global occurrence of TC will in fact decrease as a result of a changing climate. Studies suggest that more tropical depressions may form, but fewer may become full-blown tropical cyclones. However, climate change also has a sting in its tail: the frequency and intensity of Category 4 and 5 tropical cyclones on the Saffir-Simpson scale are both projected to increase. One model cites this at a 24% increase in frequency, and 35% increase in the duration, of Category 4 and 5 cyclones. Meanwhile, hurricane intensity is projected to increase by between 2 and 11% by the end of the century in the Atlantic basin.
Rainfall, and stalling tropical cyclones
It’s also worth pondering how significant increases in rainfall could be, given such extremes are imprinted on the American conscience following the incredible volume of water that Hurricane Harvey deluged Texas with, with a record 1540 mm peak, as Hurricane Harvey stalled over Texas as a Category 4 hurricane, making landfall 3 times in 4 days, allowing it to maintain its intensity by sucking up more seawater.
Kerry Emmanuel authored a modelling study involving hundreds of synthetic hurricanes to see whether climate change could make hurricanes more likely to stall. The answer is yes: hurricanes are more likely to stall at the coast, and the percentage of storms with rapid intensification at the coast could increase dramatically. It’s tempting to call his report prophetic, but we must remember that it was meteorological chance that Harvey was detained by a high-pressure system to the north. It’s also easy to forget that Hurricane Harvey was the first major (Cat. 3 and above) hurricane to make landfall on US soil since Hurricane Wilma in 2005.
Crucially, what Emmanuel’s 2015 study calls into question is that perhaps we’re looking at the relationship between cyclonic activity and climate change the wrong way. The central reason for more damaging cyclones is not necessarily anthropogenic climate change, but instead increasing urbanization in the Gulf of Mexico and the Caribbean, and anthropogenic sea level rise. This will be exacerbated to an even greater extent in Bangladesh, the Philippines, and low-lying Pacific islands. Imagine what a one-metre increase in sea level would do to a storm surge in Bangladesh, even if the cyclone remained identical in strength
the extent of the flooding from rainfall after Hurricane Harvey
So, at the end of the century, will climate change have made tropical cyclones worse? Category 4 and 5 storms might have become slightly more intense with more rainfall, perhaps lying closer to the coast, perhaps persisting for a bit longer. There will be fewer smaller hurricanes, balancing out any statistical increase in total activity. Hurricanes might occur a bit outside their typical seasons; they might migrate further away from the equator. Their effects will be undoubtedly exacerbated by higher sea levels, but the tropical cyclones themselves won’t be culpable for the increased destruction.
It’s the cardinal sin of meteorology to attribute a single weather event to the climate. But, if reports from media about climate-exacerbated hurricanes kindle climate activism in the USA in the wake of Harvey, Irma, Jose and Maria, then turning a blind eye to their premature conclusions is no bad thing.
References: inline citations are directly hyperlinked to original sources
All pictures are sourced from the public domain, except where stated otherwise