The unusually large rainfall from this year's monsoon has caused the most catastrophic flooding in Pakistan for 80 years, with the UN estimating that around a fifth of the country is under water. Perhaps 20 million people are homeless, along a path of destruction over 600 miles long.
Meanwhile, the UN is warning of impending food shortages, with the World Bank estimating that $1bn (£640m) worth of crops have been ruined, whilst cholera is reported to have broken out in the North-West Frontier Province, where flooding has been most intense.
This is a crisis of the very first order. What is particularly disturbing about it is that, despite long-range forecasts of the likelihood of heavy summer rain, appropriate warnings appear not to have been issued to local authorities and communities who might have made the necessary preparations. The worst of the rains may now be over, but this is not simply a retrospective, academic issue. Heavy monsoon precipitation – potentially exacerbated by any intensification of the El Niño/La Niña cycle, and also by the effects of global warming (which means extremes of rainfall are also a growing world-wide trend) – has increased in frequency in Pakistan and western India in recent years.
In July 2005, for instance, Mumbai was deluged by almost 950mm (37in) of rain in just one day, and more than 1,000 people were killed in floods in the state of Maharashtra. Last year, deadly flash floods hit north-western Pakistan. Karachi was also flooded. The danger is that Pakistan, and the Indian subcontinent in general, will become the focus of much more regular catastrophic flooding, with all the concomitant problems this would bring for an unstable state lurching from crisis to crisis and at the centre of a "campaign against terrorism".
This is therefore not just a question of trying to mitigate natural hazards better, but one with profound implications for geopolitics and inter-national security.
How can better flood warnings and prevention measures be a part of the solution? On the warnings point, there have been major advancements since the 1990s in the international exchange of weather forecast data, with regional computer models used widely by authorities across Asia. Such exchanges have been encouraged by the UN and bilateral government agreements, often as part of development assistance programmes, leading to greater accuracy in regional computer models, such as those of the UK Met Office.
In practice, however, there is great sensitivity in many countries about long-term warnings being issued publicly, especially in international media, due to the anxiety and potential mayhem this could cause. This is understandable, and it underlines the need for an effective and responsible structure within governments to evaluate risk, to do scenario planning, and to issue authoritative warnings based on all the available information.
Especially once heavy rain is imminent, much better short-term warnings could help to inform local communities about the likely duration of precipitation and flooding; such warnings are now much more reliable because the interaction between flood waters, soil and vegetation is better understood. Yet whereas some countries, for instance China, have generally issued effective short-term warnings, few – if any – were apparently given by the Pakistani authorities in advance of this most recent disaster.
The Chinese example is a good one for other countries in the region to aspire towards. Local communities in China receive detailed short-term forecasts before floods – allowing sufficient time for community workers to move people, especially the elderly and sick, to higher ground. Organisational support, with army and government collaboration, is also provided for "clean-up" operations afterwards. Similar arrangements have also been established in Bangladesh. The need for such short-term warnings is especially clear in developing countries, as there is often reluctance and fear on the part of many (especially very poor) communities of moving and leaving behind their possessions, including livestock. If monsoon flooding does become a perennial concern in Pakistan in the future, what may be inevitable is migration away from particularly flood-prone areas towards urban zones. Similar movement to urban areas has taken place in Africa, albeit as a result of drought, over the past three years. As floods or drought diminish (as now in West Africa), reversing this migration will only happen slowly, if at all.
As for prevention, some argue that nothing could have been done to mitigate the flooding. But the unusually large rainfall from the monsoon, which fell farther north and west in Pakistan this year than normal, has exposed the country's lack of investment in water infrastructure, including dams, and the removal of forest cover which allowed rain-water to drain into rivers more quickly.
Aside from the very real issue of authorities being weighed down by security and other social problems, one of the key issues here appears to be the embezzlement and mis-spending of public funds for flood prevention. Syed Adil Gilani, head of Transparency International's Pakistan office, says that 60-70 per cent of the 85bn Pakistan rupees (about £640m) spent by the country's Federal Flood Commission has been embezzled since the commission's inception in 1977.
Whatever the truth of this claim, the Pakistani government, the military, and the international community will now need to consider carefully what can be done to introduce cost-effective flood prevention and resilient flooding infrastructure, so as to help avoid a repeat of the current disaster in which, tragically, many rescue workers have struggled to deliver aid because of washed-out bridges and roads, with communication lines destroyed.
In the context of more variable monsoons and more catastrophic flooding, the introduction of better warning systems and enhanced infrastructure is essential. If the international community fails in this mission, societies and governments will be unable to respond to the devastating combination of changing environmental stresses, a growing population and geopolitical instability.
Lord Hunt is a visiting professor at Delft University and former director-general of the UK Met Office