Natural disasters and global warming

The automatic assumption is now that weather-related natural disasters are linked to climate change. Politicians and administrators are quick to blame global warming, partly todistract attention from human incompetence. Two recent examples illustrate the point.

The terrible floods in Tabasco, Mexico have left 80% of the province under water. 800,000 people have lost their homes. All crops have been lost. Many places are flooded to a depth of 2m and it will be weeks before the water recedes.

40cm of rain fell in three days, almost twice the monthly average for October. The president of Mexico was quick to blame global warming for the deluge. But this part of the country is low-lying and has had very severe floods in the past. The severity of this episode may have arisen as much from the mismanagement of the local hydro-electric power plants as from climate change.

The Californian wildfires destroyed about 600,000 acres of woodland. The cost of repair will be over $1bn. State authorities blamed the strong Santa Ana winds (possibly connected to climate change), high temperatures, and the prevailing drought. The reality is more complex: American fire losses are tending to rise, but the California fires were more to do with poor forest management practices over the last decades than ‘global warming’. Parts of California have had very little water this year, but North America regularly suffers from water shortages and this year’s drought is no worse than at several other times in the last hundred years.

Climate change will almost certainly bring far higher temperatures, more drought, greater numbers of extreme rainstorms, rising sea levels and increased winds. But in the case of some recent catastrophes, the evidence to link the disaster to global warming is thin or debatable. Politicians and administrators worried that blame will be attached to them for past inaction are far too ready to shift responsibility to an amorphous global force. News media are very willing to fall unquestioningly in line. This sloppiness gives easy targets to the global warming deniers and makes generating international action more difficult, not less.

***

Mexican floods
The Mexican state of Tabasco is largely low-lying, marshy, and prone to flood. Several substantial rivers cross the state, draining the highlands to the south. Four major hydro-electric plants dam the important Grijalva River. The Penitas dam, built in 1987, is the furthest downstream.

In 1999, very heavy rains lasting much of September and October filled this dam. Then, as now, the storms were said to be the heaviest ever recorded but I could find no mention of climate change in the government’s assessment of the weather eight years ago. Eventually, the flood gates of the Penitas dam had to be opened, putting the town of Villahermosa in Tabasco province under water to a depth of several metres. A large fraction of the inhabitants of the state were flooded out of their homes.

When emergency discharges need to be made to stop breaches or over-topping of a dam, the land downstream is likely to be saturated already. The surge of water will often cause flooding. The Penitas dam designers had predicted this problem in 1987 and insisted on measures to maintain the flow of water downstream and to allow for controlled local flooding. The New York Times of 25 October 1999 said ‘the state government mapped the key drainage areas in 1987 and prohibited development in them’. The newspaper reported that the bans were ignored by local authorities and developers.

After the floods of 1999, money was allocated to flood protection works downstream from the Penitas dam. Reports in the wake of the 2007 disaster consistently say that these funds have not been properly spent.

When President Calderon visited the affected area last week he is reported as making the following statement: ‘I can assure [local people] that the origin and cause of this catastrophe is enormous climate change’. The rains of late October 2007 were indeed intense: about 40cm fell in three days, far more than is expected in the month as a whole. It is probable that this would have caused major floods anyway. But the presence of the dams and the use of the flood plain for development and roads almost certainly made the situation far worse.

As the science writer Fred Pearce wrote in a report in 2001 for the World Wildlife Fund, the job of a hydro-electric dam is to generate power. The engineers want the dam to be full so that next year’s dry season does not reduce the amount of electricity that can be generated. Severe conflict is inevitable between the urge to hold the valuable water for power generation and let it out to maintain spare capacity for holding future rainfall.

So even though October often brings heavy rains to this part of Mexico – and precipitation this year was predicted to be particularly heavy because of ocean temperatures – the dam operators would have been reluctant to let water out gradually prior to the crisis brought on by the torrential storm of 29 October-1 November. When the power company was ultimately forced to release water in a hurry, the land downstream was largely unprotected by the investments promised after the 1999 inundation. The further development of the flood plain since then made the eventual flooding worse than it otherwise might have been.

Climate change may or may not be responsible for the peculiar intensity of this year’s rain. But President Calderon is wrong to suggest that the only cause of the disaster was global warming. Human failings, and the constant disputes between the state and federal governments, exaggerated the scale of the floods. ‘Climate change’ is simply too easy an excuse. The people of Mexico are convinced of the reality of climate change (see Carbon Commentary Newsletter #2 and its article on the HSBC global opinion survey) but in this case they seem to have ignored their president’s glib assessment and blamed the dam and the electricity company that operates it.

Californian wildfires
A long period of drought made southern Californian woodlands vulnerable to fire. Very little rain has fallen in this part of the US this year and temperatures have been relatively high – though not record-breaking. Several thousand fires, mostly very small, are quenched in California every year, but the size and rapid spread of this year’s outbreak was particularly striking. Nevertheless, to put the issue in perspective, the fires in 2003 actually burned a larger area.

About 600,000 acres of woodland were lost, about 1% of the Californian forest total. The losses amount to about 6% of the 2007 area burnt in the US in 2007. This year is likely to be the worst on record for the US woodland losses, somewhat worse than the previous record of 2005.

The attention paid to the fires by the international media was largely driven by the extremely large number of people who were evacuated from their homes. The proximity of some of the fires to the homes of celebrities encouraged the coverage. Horrifying though the images were, the south-west United States is always susceptible to woodland fire, particularly at times of fierce Santa Ana winds, which fan any fire that starts and which may have contributed this year by blowing down live power cables.

Some commentators saw the fires as evidence of global warming. Though the intensity of the drought will have contributed to the rapid growth of these outbreaks, the US frequently sees drought. Chart 1 shows the pattern over the last hundred years, demonstrating that the 1930s were at least as dry as the last few years.

Though forest fire acreage may be increasing, the cause may not be global warming. Forest management techniques in the US strongly encourage the trend. The following quotation from an April 2007 US Environmental Protection Agency document makes the problem clear. It refers to the need to clear out substantial accumulations of dead wood in a Californian national forest. It says that ‘fire exclusion’ – the policy of stopping any fires, however small – is raising the risk of catastrophic burning:

A century of fire exclusion in the project area has resulted in overcrowded forest conditions and the loss of wet meadow habitat. Overcrowded conditions in forested areas have reduced tree vigor, reduced the proportion of hardwoods in forested areas and promoted the spread of root diseases. In the absence of fire, an understory of shrubs and small trees has developed which can act as a fuel ladder and carry fire into the forest canopy resulting in the loss of forest habitat. The lack of fire has resulted in accumulations of ground fuels which also increases the likelihood of flames reaching the canopy layer.

Until recently, US policy has generally been to stop any fires at all times in woodland areas. This has had the paradoxical effect of increasing major fire risk. Tree density is increased, dead undergrowth acts as tinder and sources of water are gradually overgrown by vegetation, reducing their effectiveness at restraining fire. The evidence is increasingly strong that the current policy is responsible for at least a large part of the increase in forest fires.

Increased suburban development close to forested areas exacerbates the problem. Impermeable surfaces increase the speed of run-off and reduce average soil moisture levels. Trees become drier. So although climate change may be responsible for higher temperatures in California and possibly create faster Santa Ana winds, the true reason for this year’s fire outbreaks and their impact on homes is more to do with forest management policy and the spread of housing into dry forest areas. A sensible policy, as in Mexico and elsewhere, would be to accept that dangerous events are going to happen and plan housing and infrastructure development with unusual weather events in mind. As we see in the UK with our continued willingness to build in river floodplains, this is an easy policy to recommend, but more difficult to implement.

Global climate-related disasters in 2007
Mexican floods and Californian fires have more complex causes than ‘climate change’. They probably arise from the admixture of increasingly energetic weather systems and greater human intervention in the natural environment, as people demand better roads, larger houses, and more living space.

Across the world, this year has seen a large increase in natural disasters that appear to come, at least in part, from climate change. The United Nations said on 2 November that 140m have been affected by floods this year, of which about 100m are in China. The total is already four times last year’s figure.

Another UN agency dealing with humanitarian relief has launched 15 ‘flash’ appeals this year for emergency aid to a crisis in a developing country. Almost all of these appeals have been partly as a result of a climate-related event. In 2005, only six weather-driven appeals were made. This is probably a trend, but one cannot be sure. The new broom at the head of this UN agency may be more willing to ask for money, or it may be simply that humanitarian disasters are becoming more visible as a result of TV and the internet. The odds are quite high that the world is already seeing the impact of greater extremes of weather, but easy generalisation is dangerous.

Chart 2 shows the number and cost of ‘billion dollar’ weather disasters in the US. Is the number rising, or is it stable since 1990? No one can be completely sure, and human beings tend to need certainty before they will act. Unfortunately, the evidence on climate-related disasters is not quite at the required level of undeniability yet. Another bad excuse to do nothing.