MIAMI PUBLIC WORKS STAFF TO PROFESSOR HAROLD WANLESS, C. 2008–091
THE build-up of greenhouse gases in the atmosphere is having a big impact on Earth’s water cycle. An increase in rainfall intensity has been recorded across many regions of the globe in recent decades. Essentially, rainfall intensity is a measure of how much water falls from the sky during a given period. Increases in the proportion of rain falling as heavy downpours have occurred in all regions of the US, with a 79 per cent increase in the northeastern states.2 Rainfall intensity is influenced by the rate of evaporation and the amount of water vapour in the atmosphere, as well as other factors. Greenhouse gases are a strong influence on rainfall intensity because they warm the oceans and atmosphere. A warmer ocean evaporates more readily, and a warmer atmosphere can hold more water vapour. When rainfall occurs under these warmer conditions, it is likely to be more intense.
Snow is a form of precipitation too. Today the lower atmosphere (where most of the greenhouse gases are) is warmer than it was previously, as is the sea. When warm, moisture-filled air meets colder air higher in the atmosphere, this can result in colossal snowfalls. Extreme snowfalls have been experienced in recent years in many parts of the globe.
Most drainage infrastructure has been built with historical rainfall intensity in mind. For example, cities in the tropics are equipped with relatively large drains, because tropical downpours can be far more intense than rainfall in temperate regions. So, changes in rainfall intensity can contribute to flooding by overwhelming drainage systems and levees built with less extreme conditions in mind. Storm surges can add to these flooding impacts in coastal cities.
In the summer of 2010, Queensland experienced its wettest December on record. Floods broke river-height records at more than 100 observation stations,3 and 78 per cent of the state was declared a disaster zone.4 At the time Queensland was experiencing these extremes, the surface temperature of the ocean surrounding northern Australia was the highest on record.5 The economic impact of the increased rainfall was unprecedented. More than 300,000 homes and businesses in the Brisbane–Ipswich area lost power, and mining infrastructure was put out of action for many months, forcing companies to declare force majeure. Total uninsured costs for the Queensland government alone were about A$5.6 billion. In order to help pay the repair bill, the Australian government instituted a one-off levy. The following year, Australians earning $50,000 or more per year paid up to 1.5 per cent extra income tax. There is some irony in this. At the time the flood levy was being mooted, Australia’s opposition Liberal Party was busy whipping up public fury at the new ‘carbon tax’ (in fact a fixed-price start to a carbon-trading scheme), which the Labor government had introduced to help reduce the impacts of climate change.
The Queensland floods of 2010–12 have been examined using newly developed methods to determine whether human-induced climate change exacerbated their intensity. The research reveals that the La Niña event then being experienced was by far the most important factor, although human-emitted greenhouse gases probably exerted a minor direct influence.6 La Niña is a phase in a climatic cycle that is marked by cooler than average sea surface temperatures in the central and western Pacific Ocean. It often brings increased rainfall to Australia. A more recent research paper, however, points out that La Niña events are themselves being influenced by human carbon pollution, because they accelerate the rate at which the land is warming relative to the oceans. As a result, La Niñas are projected to increase in frequency from the current average of once in 23 years, to one every 13 years. There is also a greater chance of them occurring hard on the heels of El Niño events, creating extreme swings in climate.7 So the human influence on floods such as those experienced in Queensland in 2010–12 may be felt more strongly in future.
Despite the increase in rainfall intensity experienced overall, shifting atmospheric circulation patterns are decreasing the amount of rain falling in some regions. Areas affected include Africa’s Sahel, America’s west, Australia’s south, and even flood-prone Queensland. Following the unprecedented 2010–12 floods, in March 2014, 80 per cent of Australia’s Sunshine State was drought declared, the greatest percentage ever in its history.8 While it’s true that Australia has always been a land of drought and flooding rains, climate change is influencing these events so that they become more extreme.
Other recent droughts with a documented climate-change impact include those that occurred in Texas and Oklahoma in 2011. Both states experienced their hottest summer since record keeping began in 1895, with many locations experiencing more than 100 days over 37.8°C. Rates of water loss, due in part to increased evaporation, were double the long-term average. The intense heat, combined with drought-depleted water resources, contributed to more than US$10 billion in direct losses in agriculture alone.9
When I wrote The Weather Makers, a convincing example of climate change’s impact on the hydrological cycle was the decrease in stream-flow into Perth’s dams up to 2002. A large decline in rainfall in southwestern Australia had most likely been caused by CO2 pollution and the ozone hole, and rising temperatures were robbing the soil of moisture. By 2001 this had led to a dire water shortage, forcing the Western Australian Water Corporation to plan to build a desalinisation plant. Completed in 2006 in Kwinana, a Perth suburb, the plant now supplies the city with 45 gigalitres (45 billion litres) of water per year. But the situation continued to deteriorate and a second plant, with an annual capacity of 100 gigalitres, was planned. Stage one is now complete, and stage two is under construction.
At the time these huge infrastructure investments were planned, there was a possibility that the scientists were wrong, and that the rain might return. The latest stream-flow figures for Perth’s dams, however, show how wise the investments were. In 2012, the dams received just two gigalitres of water, as against the average for the seven years following 2005 of around 70 gigalitres per year. Between 1911 and 1974 those same dams had received an average of 338 gigalitres per year.10 As of mid-2014, almost half of Perth’s water requirement is supplied by desalinisation.11
I was in Perth in 2011, when the hysteria about Australia’s carbon tax was at its height. The climate-change deniers were hard at work using the 2010 Queensland floods as ‘proof’ that the climate scientists were wrong in warning about water shortages. Much of the attack was aimed at me personally, so I was grateful when the previous head of the state’s Water Corporation thanked me, publicly, for helping raise the alarm about water shortages in Perth. Without such warnings, he said, the investments required for the desalinisation plants might not have been forthcoming, and Perth might have faced a full-blown water crisis.
Other changes in water availability predicted by climate scientists in 2005 have since been validated. In California, global warming is melting the snowpack earlier, robbing agriculture of water when it’s needed. In May 2014, authorities announced that the second smallest snowpack in the state’s recorded history had formed the previous year.12
In fact, as a new study shows, some parts of California’s mountains lifted as much as 15 millimetres during 2013–14 because the massive amount of lost snow is no longer weighing down the land, allowing the unburdened mountains to rise, a bit like an uncoiled spring.13 The snow losses saw the California Department of Water Resources set the 2014 water allocation at a mere 5 per cent of that requested by agricultural and other public water agencies—the smallest allocation in the department’s 54-year history.14
As weather historian Chris Burt wrote of the evolving situation:
The next six months are going to be a severe test of the state’s ability to manage . . . We will probably see (and already are) clashes between agricultural concerns and urban consumers. The specter of a horrific fire season also looms over all this. Since the drought of 1975–77 the population of California has almost doubled (from about 20 million to 38 million). The consumption of water resources by the agricultural industry has also dramatically increased.15
Recent research suggests that this may be just the beginning of a far larger water problem for the region. A study using data from tree rings (which reveal changes in rainfall going back 1000 years or more) combined with sophisticated computer models, indicates that California, along with much of the Great Plains and American Southwest, is at high risk of ‘mega droughts’ in coming decades.16 The tree rings reveal that droughts lasting 20 years or more—far more extreme than anything experienced since European settlement—struck the region 800–900 years ago. The computer models meanwhile indicate that conditions are right for even more severe droughts—lasting up to 50 years—to strike the region later this century. Cornell University’s Professor Tony Ault, co-author of the study, said of the findings: ‘In both the Southwest and Central Plains, we’re talking about levels of risk of 80 per cent of a 35-year-long drought by the end of the century, if climate change goes unmitigated.’17
When I wrote The Weather Makers, Africa’s Sahel region was in the grip of severe drought. A study by Australia’s scientific research organisation, CSIRO, had proposed that an important factor in the Sahel drought was ‘global dimming’ caused by particulate pollution produced by factories and motor vehicles in Europe and elsewhere.18 The unfolding story has become more complicated. Changing sea-surface temperatures have now been implicated, via a natural cycle known as the Atlantic Multidecadal Oscillation (AMO).19 The AMO describes cyclical changes in the surface temperature of the Atlantic Ocean and is associated with changes in thermohaline circulation (of which the Gulf Stream is part). The AMO’s warm phase is expected to peak in about 2020, which may bring temporary relief to the region’s water shortage.20 Another factor affecting the drought, however, is record-breaking high temperatures. In 2010, extreme temperature records were broken in Chad, Niger and Sudan, the new records all being between 47.1°C and 49.6°C.
One area where the global climate outlook has improved somewhat concerns hurricanes. By 2005 it was evident that the number of hurricanes in the North Atlantic was increasing. They are also becoming more intense and extending into regions, such as the south Atlantic, where they had not previously occurred. In the light of these trends, some scientists anticipated that the number of hurricanes worldwide would increase. But hurricanes are relatively rare weather events, making it difficult to get statistically significant samples over time and space, and other researchers felt that the data were not adequate to decide one way or the other.
Over the past decade the situation has become clearer. While the number of hurricanes in the North Atlantic has increased, the total number of hurricanes occurring worldwide has not, nor is it expected to. Hurricanes have, however, increased in severity, and they are occurring further from the equator. As the IPCC recently said in its fifth assessment report:
It is virtually certain that there has been an increase in the frequency and intensity of the strongest tropical cyclones in the North Atlantic since the 1970s. In the future, it is likely that the frequency of tropical cyclones globally will either decrease or remain unchanged, but there will be a likely increase in global mean tropical cyclone precipitation rates and maximum wind speed.21
The reason that the number of cyclones might decrease in future results from the relative influence of several consequences of global warming that work against each other when it comes to cyclone formation. The most influential is the way that the warming of the oceans and the land alters wind direction and speed. This can create wind shear, which tears into the embryonic cyclone vortex and dissipates it. Other factors affect the temperature difference between Earth’s surface and the upper troposphere (the top of the lowest layer of the atmosphere). The greater the difference, the more likely it is that a cyclone will form. The upper troposphere is warming faster than Earth’s surface because the greater evaporation caused by the warming is carrying increased amounts of water vapour high into the troposphere. When the water precipitates out in the upper troposphere, it releases latent heat energy. But the warming is also causing the troposphere to expand, providing a longer distance over which air can cool as it rises. At present, the latent heat energy provided by the extra water vapour more than offsets the cooling caused by the additional distance the air travels. The smaller temperature difference between the upper troposphere and Earth’s surface that results is likely to see fewer cyclones form. But other factors cause regional variations.
The US National Assessment Report summarised the situation for North America thus:
By late this century, models, on average, project an increase in the number of the strongest (Category 4 and 5) hurricanes. Models also project greater rainfall rates in hurricanes in a warmer climate, with increases of about 20 per cent averaged near the center of hurricanes.22
While a Category 1 hurricane may strip the leaves from trees, a Category 5 will uproot them. The extra rainfall, combined with the greater battering caused by rising sea levels, is likely to make future hurricanes a more severe threat to coastal cities. The kind of damage we’ll see more of was illustrated by Hurricane Sandy, which devastated the West Indies and the east coast of the US. With a diameter of 1770 kilometres, it was the largest Atlantic hurricane on record and the most destructive hurricane of the 2012 season. But, most significantly, with cumulative damages reaching US$68 billion, it was also the second most costly hurricane (after Katrina) in US history.
When it hit New Jersey, Sandy was only Category 1. Yet it drove a water surge of almost 4.3 metres at Battery Park, flooding large parts of Lower Manhattan. Subway flooding and prolonged interruption to the gas supply were just two of the impacts that left the city reeling. The extent of future damage to coastal infrastructure by hurricanes will be strongly influenced by the extent of sea-level rise, and there things are looking grim.
With just under 1°C of warming experienced to date, the world’s oceans are rising at the rate of 3.2 millimetres per annum. Sea-level rise has two components: ice melts, adding to the sea volume, and the expansion of water as it warms. The oceans are absorbing about 90 per cent of all the heat captured by the extra greenhouse gases in the atmosphere. Heat transfer into the ocean is reasonably well understood, and scientists are confident in their predictions of how the thermal expansion it creates will influence sea levels. By the end of the century the oceans will rise by between 11 and 43 centimetres due to heat transfer alone.
The other component—the water added by melting ice—is far harder to predict, as it depends on how the ice caps respond to the warming. This area of research is developing rapidly, with new findings being announced every year. A decade ago it was clear that the Arctic was in trouble. It was warming twice as fast as the planetary average, and the Arctic ice cap, which is sea-ice and therefore doesn’t affect the level of the sea as it melts, was vanishing fast. The Arctic ice does, however, help insulate the Greenland ice cap, and its melt waters are likely to add significantly to the level of the oceans. Year-on-year variation of the extent of Arctic ice is considerable, but a long-term trend of dramatic decline is now clear.23
As I feared in 2005, the rate of melt of the Arctic ice has proven to be of the runaway type. In the last decade the rate of ice loss over the Arctic has exceeded even the worst-case scenario of the climate models. We now expect to see the Arctic’s first ice-free summer in over a million years sometime between 2040 and 2050. An ice-free summer in the Arctic is defined as one with less than a million square kilometres of ice, as the ice around the edges of the Canadian islands melts less readily.
Until very recently, great uncertainty surrounded the state of the Antarctic ice cap. Scientists were having a hard time assessing the contribution of its ice melt to sea-level rise. Accordingly, the IPCC declined to include figures for total sea-level rise in the main body of its Third Assessment Report published in 2001.24 Climate-change denialists then deceptively used the raw figures to argue that the IPCC had revised the rate of future sea-level rise downwards. They also argued that there was no need for concern about sea-level rise because the extent of sea ice surrounding Antarctica was growing. Eighty per cent of Antarctic sea-ice melts away each year, then grows again.25 During 2008–11 its minimum extent was small (2.5–3.2 million square kilometres), but between 2012 and 2014 it averaged 3.6–3.9 million square kilometres. Its maximum extent is also variable, and during the winter of 2014 it reached 19.8 million square kilometres—its largest extent recorded since satellite monitoring began in 1979.
Nobody knows why the extent of Antarctic sea ice was so great in 2014, but among the factors that might be affecting it are the ozone hole (which influences wind that can push ice northwards, thereby increasing its extent), the warming of the atmosphere (enabling it to hold more water vapour, which then falls as snow, causing fresh water, which freezes more easily, to collect on the sea surface) and changes in ocean current circulation, which can bring cold seawater to parts of the surface. A recent study even suggests that changes in wave action might also be a factor.26 As with so much about the Antarctic, much research needs to be done before we have a clear picture about what is causing the variability.27
The story of the Antarctic land ice is much clearer, thanks to a series of discoveries published during the first half of 2014. In The Weather Makers, I said: ‘The increased precipitation occurring at the poles is expected to bring more snow to the high Antarctic ice cap, which might compensate for some of the ice being lost at the continent’s margins.’ Sadly, that compensation has proved illusory: it is now established beyond doubt that losses of land ice are occurring across all the major regions of Antarctica.
This new understanding comes courtesy of the European Space Agency’s Cryosat 2 spacecraft, which was launched in 2010 specifically to measure the thickness of polar ice. It uses a special kind of radar, known as a synthetic aperture interferometric radar altimeter, to chart the surface shape of ice sheets, and it concentrates particularly on their margins. The new assessment incorporates three years of data, 2010–13, and updates a synthesis of observations made by other satellites over the period 2005–10, providing the first accurate, continent-wide assessment of Antarctic ice.
The ice losses Cryosat 2 detected over Antarctica are gargantuan—in the order of 160 billion tonnes a year—the equivalent of two centimetres of snow off the entire surface of Antarctica. That’s twice as large a loss per year as when the continent was last surveyed in 2005, and sufficient to add almost half a millimetre of global sea-level rise annually.28
Most of the ice loss is occurring in West Antarctica, where the ice sheets have long been known to be less stable than those in East Antarctica. A decade ago I expressed fears that the West Antarctic Ice Sheet (WAIS) may destabilise and melt into the sea. Were this to occur, it would add about 4.8 metres to the level of the oceans.
The Pine Island Glacier (PIG) and the associated Thwaites Glacier are important elements of the WAIS. Between them, they hold enough water (as ice) to contribute a metre of sea-level rise to the world’s oceans. In 2014 a study revealed that the PIG is dead on it its feet.29 Due to a warming ocean, PIG is melting away from below, and nothing can save it. It’s important to understand that this finding does not result from computer modelling, but straight mechanics. The undersurface of the ice has melted to the point where the bedrock slopes back towards the glacier’s head. That means that the remaining ice will detach from the bedrock and slide into the sea, melting as it goes.
NASA recently analysed 40 years of observations of six big glaciers (including PIG and Thwaites) that drain into Amundsen Bay. It concluded that nothing now can stop them all melting away.30 Just how long it will take is a big unknown. If measured in centuries it would be a blessing, as the loss of the ice will add around 1.2 metres to the ocean level. At one estimate, during the next 20 years, a 20 per cent melt of ice is likely to add 3.5–10 millimetres to sea levels.31 As NASA’s Eric Rignot says, IPCC estimates of sea-level rise do not take these new data into account. As a result, we should expect sea-level rise to be at the high end of the range estimates for this century.32 The IPCC’s current anticipated range for sea-level rise, incidentally, is between 0.4 of a metre and one metre by 2100.
Another new study lends credibility to the idea that Antarctica’s coastal ice may melt more quickly than expected, with consequences for sea level. The researchers discovered that human pollution is causing a strengthening of westerly winds, and shifting them polewards as they circle Antarctica. The trend has been observed since the 1950s and is producing:
an intense warming of subsurface coastal waters that exceeds 2°C at 200–700 m depth . . . This analysis shows that anthropogenically induced wind changes can dramatically increase the temperature of ocean water at ice sheet grounding lines and at the base of floating ice shelves around Antarctica, with potentially significant ramifications for global sea-level rise.33
The link between winds and warming occurs through a phenomenon known as Elkman pumping, whereby the wind creates a surface stress on the ocean, which propagates a spiral of currents in the water below. This means that the westerly winds circling Antarctica can, at a certain depth, generate a southwards flowing current which draws in warmer water. These findings are so new that their full implications are as yet unexplored. But we would be wise to assume in the face of these cumulative findings that sea-level rise will be a greater problem—perhaps a far greater problem—and that it will occur sooner than we imagined.