This blog is the first in a series of content provided by guest authors. Today, we feature a post from Dr Jan Zika, a specialist in sea level from the University of New South Wales, Sydney, who explores the topic of predicting coastal impacts and quantifying catastrophic risks with climate change and how global warming might drive devastating sea-level rise.
On the 3rd of June 2016 a weather system known as an East Coast Low, formed over the Tasman sea. The enormous strength of its winds, and their precise direction conspired to generate a sea-level event that battered the beaches of eastern Australia with tenacious ferocity. The event cost over AU$300M in the state of New South Wales alone. Images of Sydney’s beachside pools washing into the sea attest to the devastation.
Will such events happen more and more often around the globe with climate change? Are there other freak occurrences on the horizon that we can or cannot hope to foresee?
In this blog I will discuss risks on the horizon in terms of sea-level rise and coastal impacts. I will explore how their impacts might change. As is reask’s imperative, I will look at how foreseeable such risks are and will be in the future. It will turn out that the risk of sudden, localised and weather driven events devastating coastlines will pose a continued threat with climate change. But the risk of them occurring is likely to be broadly quantifiable and accurate (insofar as probabilistic forecasts can be). On the other hand, the risk of global average sea levels rising (relatively) rapidly over the remainder of the century is ‘possible’ and would be globally catastrophic. A translation of ‘possible’ into quantifiable risk is pure speculation.
The risk of devastating extreme sea level events: always possible, not substantially more probable.
East Coast Lows are cyclonic low pressure systems which form off the east coast of Australia. They bring strong easterly winds and lead to substantial rainfall. These systems are not easily forecast beyond the typical 10-15 day weather forecasting horizon. They are simulated by global climate models and so understanding about their general behaviour can be gleaned.
Research by Peppler and co-authors suggests that cyclonic systems will become less frequent along the east coast of Australia in the future. However, this suggestion is muted by a lack of statistical information about big storms (of the June 2016 variety).
Elsewhere, a plethora of diverse meteorological phenomena cause devastating sea-level events. In the tropics the obvious threat is from tropical cyclones. Predicting such events is, of course, challenging, but both dynamical and statistical modelling do deliver useful information in terms of risks. In addition, there is reasonably broad consensus that such storms will become more intense with global warming as the latent heat carried by air increases with warming.
In some regions sea-level events are driven by a coincidence of events. On the British coast for example a huge tidal range means storms can be devastating if they occur during high tide. The disastrous storm surge of 1953 was one such devastating coincidence. Events being catastrophic when coincident makes them harder to predict in isolation, but their probability perhaps easier to foresee. In this case we require statistical knowledge about two effectively linearly independent processes – tides and storm frequency.
As further attention is given to meteorological and climate change models and as machine-learning and other statistical methods are advanced – the threat of sea-level events will not diminish but the probability of their occurrence should become better known. An overriding additional aspect of these is the rate of background sea-level rise. The higher the background state the more catastrophic the impact of any particular event.
The risk of multiple metres of global sea level rise this century: some say possible, no one knows how probable.
Globally averaged sea levels have been rising at 3mm per year in the last two decades. Considering the devastation of 2016 (an event associated with metres of swell), residents of Australia’s east coast will not be troubled by the prospect of such a small rate of sea-level rise.
The picture of average sea-level rise does vary from place to place though due to a range of factors. Changes in ocean currents appear to be amplifying sea level rise along America’s east coast for example. More alarming is the fact that sea levels have varied by well over 100 m over past ice ages. Rates of rise and fall were up to 40-50 mm/year between each ice age. These variations were mostly driven by variations in the amount of solid ice in the earth system – the biggest contributor being Antarctica. We do not know how Antarctic ice will respond to global warming. As a result, projections of sea-level rise cover a broad ‘possible’ range. A study by deConto and Pollard for example suggests they could approach 2 m by 2100 and 16 m by 2500.
Sea-level rise of 2m by 2100 would eclipse almost all other factors. However, the deConto and Pollard study is based on one simple ice sheet model calibrated with sparse paleontological data. A report by the US Government attempted to place a statistical probability on such a risk. They concluded that there was a 0.01% chance of 1-2m of sea-level rise this century. However, this was not based on any statistical model or large ensemble – but merely on expert opinion.
Across the globe, the risk of sea-level events devastating property and infrastructure are real, but with sufficient expert attention and state-of-the-art tools, a clearer and clearer picture can be developed of the extent of that risk. On the other hand, as the climate warms, the risk of global sea levels rising by multiple metres is real and would be globally catastrophic. A quantification of the risk of such a catastrophic sea-level rise is so far illusive.