Hurricane Irene’s destructive trail re-opened the still fresh wounds of 2005’s Katrina – and brought home the realisation that it’s time to face the new high-hazard reality and get prepared.
By Douglas Fox

The hurricane churning off the east coast of North America seems destined for the mid-Atlantic. Then a cold front descending out of Canada nudges the Category 2 storm northwest instead – setting it on a worst-case course for New York City.

New York Harbour has often sheltered the city, dissipating energy from violent gales that start at sea. But now it plays an opposite role: it turns an otherwise moderate hurricane into a disaster. As the eye of the storm passes over Staten Island, the 150 km/h counter-clockwise winds shove 500 million tons of seawater directly into the harbour. The narrowing shorelines and shallowing sea bottom cause the mass of water to build. By the time the storm surge washes over the shores of Brooklyn, Queens and Manhattan, it towers 3 to 5 metres high.

Water flows through New York’s financial district and reaches 3 kilometres southern Brooklyn and Queens, flooding 4 600 km of roads. Impromptu rivers gush into subway stations and pour through hundreds of sidewalk gratings.

In Manhattan, the lower levels of Penn Station and Grand Central fill with water. The subway floods within 40 minutes – paralysing the city’s chief form of public transportation. Three of the four vehicular tunnels linking Manhattan to the outer boroughs and New Jersey also flood, submerging hundreds of cars stranded in traffic jams during evacuation. A million people lose electricity and phone service as floods shut down 10 power plants and the emergency generators powering cellphone towers.

While this scenario may sound like yet another apocalypse-in-New York summer blockbuster, it was produced using calculations from the US Army Corps of Engineers – and it’s been given serious attention from government planners. That 1995 Army Corps report and a 2006 analysis by the US Department of Homeland Security predict that a Category 4 hurricane scoring a direct hit on New York City would inflict R3,5 trillion worth of damage – quadruple that wrought by Category 5 Hurricane Katrina in 2005.

A third study, released this September by New York state, predicts that an even milder, Category 1 hurricane or winter nor’easter could inundate the city’s subway and cause R400 billion in losses. Experts don’t consider such disastrous flooding a mere possibility; they believe it’s a certainty – a onein- 100-year event. Sea level rise will upgrade it to a one-in-35- year event by 2080.

“We’ve been very, very lucky because we haven’t had that (direct hit),” says Cynthia Rosenzweig, a climate-impact scientist at the Nasa Goddard Institute for Space Studies in New York who has helped guide the city’s storm- and climate-planning effort. “But the potential vulnerability for that is very high.”

Every region of the US is subject to catastrophic storms of one type or another. Although the severe floods and tornadoes that devastated large swaths of the country in recent months surprised many people, there’s no reason they should have. Annual losses from natural hazards have increased severalfold over time – costing the nation R4 trillion in crops and property since 1960. Americans are turning even routine storms into full-blown disasters by settling where they strike. Then, when vulnerable infrastructure is swept away, people have exhibited a steadfast commitment to rebuilding it.

“There are more people living in what we might consider to be high-hazard areas,” says Susan Cutter, a disaster scientist at the University of South Carolina in Columbia. These include coastal areas, floodplains and places especially prone to tornadoes and landslides. By 2040, 70 per cent of the US population – which should then number 400 million – is expected to concentrate in 11 megaregions, seven of which occupy coastal counties.

If New York – part of the Northeast megaregion – suffers a direct hit, workers will spend weeks pumping several billion litres of brackish water out of its severe storm warning subway and train tunnels. The salt will corrode power lines, transformers and thousands of brakes and switches that control the trains. Some subsystems could take a year or more to restore.

To avoid such a scenario, New York state recommends the city invest well over R700 million a year in storm protections. City planners are already experimenting with dozens of low-tech fixes, says Adam Freed, deputy director of the Mayor’s Office of Long- Term Planning and Sustainability. These include raising subway vents above sidewalks, installing barriers several centimetres high around subway entrances and using porous roadways. They’ve also considered building lips around rooftops to slow the percolation of water into streets and sewers, because every centimetre of rain that falls on New York translates to about 2 billion litres of storm water that must be managed.

Some observers, such as Malcolm Bowman, an oceanographer at the State University of New York at Stony Brook, have even suggested that four massive barriers be built across the waterways surrounding the city. The arms would swing shut during severe storms – much like those of the Maeslantkering, a barrier that protects the Port of Rotterdam from surges in the North Sea.

Canton, a town of 2 500 on the upper Mississippi River in Missouri, has been at the centre of an increasingly high-stakes environmental wager for years now. In the summer of 1993, a high-pressure system stalled over the southeastern US, forcing the jet stream, laden with moist air from the Gulf of Mexico, to the north, where it collided with cold air from Canada. As a result, rainstorms drenched the upper Midwest. Many towns received two to six times the normal amount of rainfall for June and July.

The Mississippi River crested at 4,2 metres above official flood levels in Canton, overtopping several local levees.

That year, more than 1 000 levees ruptured or overflowed along the Mississippi and Missouri rivers. Seventy towns, including Canton, flooded. The water stayed high for six months.

According to US government statistics, the flood that Canton experienced in 1993 was a freak, one-in-500-year event – not something that would happen again soon. That estimate came from analysing the 140-year historical record – calculating the frequency of floods of various magnitudes and extrapolating the curve out to events at a scale never seen before.

If only it were that simple. Canton suffered another 500-year flood in 2008, a 70-year flood in 2001, and 10-year floods in 1996, 1998 and again in early 2011. Plenty of towns across the region have suffered similar events.

“We’re witnessing higher and higher floods over time,” says Robert Criss, a hydrogeologist at Washington University in St Louis. “We are seeing higher and far more frequent floods than government estimators say we should.”

The data are too noisy to chalk that trend up to increased rainfall. Instead, official statistics may underestimate the severity of floods in this region because records are too short to reveal the full variability of the climate. “We have no idea what Mother Nature is capable of dishing out,” Criss says.

People ten d to view earthquakes and hurricanes as the most damaging natural disasters – but a steady rain could do far worse. In the winter of 1861 to 1862, California experienced a series of rainstorms lasting 45 days. The Central Valley, the large, flat plain running down the middle of the state, became a shallow lake that lingered for months. Newspapers described people travelling the streets of Sacramento in boats.

A team of 40 scientists recently modelled the effects of such a roughly 500-year storm if it were to strike California today. “There’s no way that the magnitude of the storm and the subsequent flooding could be contained by the existing flood structures,” says Justin Ferris, a hydrologist at the US Geological Survey’s California Water Science Centre in Sacramento. “Such a flood would be devastating.”

Ferris and others estimate that a 500- by 30-kilometre swath of the Central Valley would flood. Waterlogged soil would trigger hundreds of landslides. While the USGS considers a magnitude 7,8 earthquake along the San Andreas fault an equally likely event, the California storm would cost nearly three times as much – R3 trillion in direct damages.

Plenty could be done to soften the impact of massive downpours, but it will mean undoing 150 years of misguided policy. Engineers have progressively walled in the upper Mississippi and lower Missouri rivers as they straightened them for ship navigation – in some places decreasing the rivers’ width by two-thirds since 1875. This reduced their ability to expand during floods. Compared with a century ago, an equivalent amount of water flowing down the upper Mississippi River now causes the water to rise 3 to 4 metres higher.

In Chesterfield Valley, Missouri, malls and homes worth several trillion rand have been constructed in the past decade on land that was underwater in 1993 – requiring the government to build up levees. Breaking that cycle, Criss says, will mean putting an end to misleading flood-risk statistics and the artificially cheap federal flood insurance that goes hand in hand with them. “It enables people to obtain financing for very economically damaging projects,” he says. “It puts the taxpayer on the short end of the stick.”

You might say the Army Corps of Engineers took a small step in the right direction on 3 May, 2011. That’s when it dynamited a 3-km section of levee on the Mississippi River to divert water onto farmland and save the town of Cairo, Illinois, from flooding. But the ensuing wall of water inflicted long-term damage – scouring away topsoil, gouging gullies 2,5 metres deep and dumping sand.

A better approach is to plan for flooding by building lower levees that are designed to overflow, allowing the farmland to flood more often – and more gently. “We need a system that uses farmland for floodwater storage,” Criss says. “It will help the environment, and the farmers can be compensated for that, just like we compensate them for letting land lie fallow.” Such a system would pay for itself by reducing rebuilding costs; flood damages currently total R7 billion to R70 billion a year in the US.

Floo d an d hurricane risk can at least be predicted: it is heavily influenced by topography, and the storms and floodwaters can be tracked for days in advance. But severe tornadoes, like the ones that tore across the central and eastern US in 2011, pose a very different challenge.

The tornado that ripped through Joplin, Missouri, on 22 May, a 5 on the enhanced Fujita (EF) scale, existed for just 38 minutes. During that time it ploughed a path 1,2 km wide through town, destroying nearly 7 000 homes and tossing pick-ups 200 metres. Many people, such as Pizza Hut manager Christopher Lucas, reacted just in time: he crowded 19 people into his walkin freezer. They survived, although Lucas – sucked out of the freezer as he held the door shut with a bungee cord – was among the 150 who died.

The tornado seemingly could have struck anywhere. The moist air that flowed from the Gulf of Mexico that day created a whole herd of potentially tornadospawning thunderstorms, from Oklahoma to Minnesota. The Joplin tornado affected just a few square kilometres of that vast area – yet it did so with overwhelming fury. Its winds, over 300 km/h, hammered buildings with four times the pressure that a Category 2 hurricane with 150 km/h winds would have exerted.

It’s possible to build a house to withstand 150 km/h winds, providing partial protection against some weaker tornadoes. But 300 km/h winds? “It’s just not practical to design the entire building to withstand those kinds of pressures,” says Ernst Kiesling, an engineer at Texas Tech University’s Wind Science and Engineering Research Centre. “It would be too expensive.” Even if houses can’t be protected from EF5 tornadoes, Kiesling has spent decades looking for ways to save the lives of the people inside them.

After a tornado killed 26 people and destroyed hundreds of homes in Lubbock, Texas, in 1970, Kiesling and his colleagues noticed a curious thing: even in buildings that were blown apart, an interior bathroom or closet was sometimes left intact. It gave Kiesling an idea: convert a small, windowless room in the house into a tornado shelter that could survive 400 km/h winds.

Many of the fatalities in tornadoes occur when people are struck by projectile debris – the Joplin tornado, for example, drove all four legs of a chair through the walls of one house. So Kiesling’s team tested their shelters with a gun that fired 50 x 100 mm posts at 150 km/h (the speed at which they would be propelled by 400 km/h winds). They settled on steel-reinforced plywood to make the structures puncture-proof. Such shelters can now be installed.

Researchers are working on other technical solutions to tornado protection – for example, radar that provides earlier warnings. But societal trends continually work against even the best of efforts. “For any intensity of tornado, you’re more likely to be killed if you’re in a mobile home than in a permanent home – 15 to 20 times more likely,” says Harold Brooks of the US National Severe Storms Laboratory. Unfortunately, the proportion of Americans living in mobile homes has tripled since 1970 (and it is highest – about 15 per cent – in the tornado- prone Southeast).

In this sense, the problems posed by tornadoes do bear a resemblance to those of hurricanes, floods and other severe storms. Some steps needed to minimise losses, such as stricter building codes and sturdier infrastructure, are well-known. They require greater investment. Other solutions, just as important, involve choosing what not to protect. Instead of applying brute force, they’ll mean removing the perverse incentives that encourage people to build in high-risk areas. Because for those who find themselves in harm’s way, even modest storms can be super.

Catergory 3 Hurricane
1 Heat Engine
Hurricanes are massive dynamos powered by the evaporation of water and its subsequent re-condensation into clouds and rain. Each litre of water that evaporates and condenses carries about as much thermal energy into the atmosphere as that contained in about a quarter cup of petrol.

2 Hotspot
Hurricanes spawn over patches of ocean where the surface water has warmed to at least 27 degrees down to a depth of at least 50 metres. Ocean water evaporates in these hotspots and the moist air rises.

3 Updraft
The rising of warm, moist air creates a lowpressure zone, pulling in more air from nearby areas, which also moistens and rises. The continued updraught is fed by condensation of evaporated water into clouds and rain. The condensation dumps energy back into the air, warming it and making it more buoyant.

4 Spiral
Surrounding air flows into the low-pressure zone in a spiral pattern. This inward-spiralling air forms the hurricanefs destructive winds. The direction of the spiral is determined by the Coriolis effect . a byproduct of the Earthfs rotation.

5 Sinking
Air is ejected from the top of the storm at an altitude of about 12 000 metres. This cooled, dried air sinks through the eye of the storm or else flows out and sinks in the outer bands of the storm, forming areas without rain.

6 Magnitude
At its peak, a hurricane can dump 20 cubic kilometres of rain a day and unleash thermal power (freed by condensation of that water) at a rate of 6 x 1014 watts – equal to 200 times the amount of electricity generated by humans worldwide. Only about 0,25 per cent of this power is converted to wind.

7 Effects
In addition to destructive winds, hurricanes can pile up sea surges higher than 6 metres (which are responsible for most deaths). Even after winds dissipate inland, rain can cause flooding for days - as happened with Hurricane Mitch, whose floods and mudslides killed nearly 20 000 people in Central America in 1998.

Ef4 Tornado
1 Supercell
Tornado-spawning thunderstorms, called supercells, arise where a current of low, warm, moist air travelling north from the Gulf of Mexico flows underneath a higher, cooler mass of air travelling east. Shear from these opposing winds causes the entire supercell to rotate slowly.

2 Updraught
The low, moist air is warmed by sunlight, making it increasingly buoyant. The moist air breaks through the cooler air above and rises. As it does so, vapour condenses into water droplets – dumping the heat of condensation back into the rising air, warming it and further feeding the updraught that will ultimately power the tornado.

3 Downdraught
The updraught is counterbalanced by a downdraft of sinking air, which is cooled by rain. This cool, sinking air next to warm, rising air produces a pressure gradient in the bottom 1 000 metres of the atmosphere, lending a spiralling motion to the updraught – which will become the tornado.

4 Stretching
The supercell travels northeast, towing the updraught like a leash. Stretching causes the updraught to narrow. As this occurs, its spiralling winds accelerate – much like a spinning ice skater speeding up as she pulls in her arms and legs. A violent tornado is born.

5 Pressure
The tornado’s updraft creates a low-pressure zone at its core. That pressure differential relative to the surrounding air is roughly equal to that of a Category 5 hurricane – except that in the case of the tornado, the differential exists within a kilometre rather than 150 kilometres – helping fuel ferocious winds that can far exceed those of a hurricane.

6 Effects
Supercells don’t just spawn tornadoes. They can also produce powerful down-draughting wind bursts kilometres from the tornado. These bursts can sometimes fl ip mobile homes. Supercells can also produce dangerous lightning and hail as large as golf balls or grapefruit.

Flood
Prevent future damage

Improved statistics
Flood-risk statistics for the upper Mississippi River seriously understate the potential for flooding. In St Louis and dozens of smaller cities and towns, a one-in-200-year flood may actually occur every 50 years. Correcting those risk numbers could raise flood insurance rates, curb development in high-risk areas, and ultimately reduce federal government flood insurance payouts.

Coastal buffers
During the 2004 Indian Ocean tsunami, mangrove trees as little as 100 metres deep blunted tsunami flow pressure by up to 90 per cent. Development has substantially reduced mangroves that naturally occur along the US Gulf Coast from Texas to Florida; restoring them could help protect coastlines against hurricane storm surges.

Proactive mapping
Coastal cities need to retool zoning laws to incorporate the 60 to 120 centimetres of sea-level rise expected by 2100. Jonathan Overpeck at the University of Arizona has produced elevation maps of 180 coastal US cities – on average, almost 10 per cent of their area will be inundated by 1 metre of sea-level rise – with larger areas at risk of floods or extreme tides.

Handheld warnings
Quicker flood warnings could help people react in time. The US Geological Survey is developing smartphone apps that track users’ locations and warn them if a flash flood is headed their way. The apps could become available within five years in flash-flood-prone areas.

Floating structures
In areas where floods are a problem but wind, waves and strong currents are not, one solution is to build infrastructure that floats. In New York City, for example, several oil-fired power plants on floating barges provide electricity during peak demand. These “power barges” can move upriver during storms and then restore power after the surge has passed.

Cost of inaction
There were 1 764 US presidential disaster declarations due to natural hazards – such as severe weather and wildfire – between 1960 and 2009, costing the country R1 trillion.

Visit www.popularmechanics.co.za to watch a retrospective video showing Hurricane Katrina as captured by satellites. Also watch a video that takes you up close and personal with Hurricane Katrina.