The long awaited National Climate Change Policy that was launched in Islamabad last week called for the strengthening of “flood forecasting, drought monitoring and early warning systems in the country”. Given that Pakistan is now hit by disastrous flooding during the monsoon season on almost a yearly basis since 2010, the county desperately needs a comprehensive early warning system. In Bangladesh, experts claim that they have become “almost disaster proof” from the destructive floods that used to displace millions annually, thanks to the early warning systems set up in the country.
Dr Peter Webster, Professor in Earth and Atmospheric Sciences, Georgia Institute of Technology, is responsible for developing an operational flood-forecasting module for the Ganges and Brahmaputra Rivers in Bangladesh. This system has now been transferred successfully to the Bangladesh Flood Forecasting and Warning Centre. It is currently being run by an NGO since the government had trouble running it. The system successfully forecast the 2004, 2007 and 2008 major Brahmaputra floods. As Chief Scientist of the company Climate Forecast Applications Network based at Georgia Tech, Dr Webster has had plenty of experience in this field. He has led the development and implementation of operational forecasting systems for tropical cyclones in the North Indian Ocean, the North Atlantic, and the North Pacific Oceans.
Quite often, floods and tropical cyclones occur in South Asia with only a few days warning or no warning at all. During the last 20 years, however, weather and climate forecasting has improved substantially, enabling “extended range probabilistic forecasts” of weather hazards. “The system we set up in Bangladesh is used to forecast flood levels throughout the country. It is a 10 day forecast”, explained Dr Webster, who was visiting Islamabad last week. He gave a talk on ‘Hazard Prediction in South Asia’ at the National Library and planned to meet with officials from Pakistan’s Meteorology Department.
Dsr Webster is currently working on implementing a similar system in Pakistan, but explained that the system here would be different from the one in Bangladesh, because our Indus River system has its own complexities. He added that with data about inflows in Tarbela dam combined with rainfall forecasts from global sources (satellite data, ocean buoys, weather balloons etc), the system could be up and running by June this year. The system could be run from Pakistan or Georgia Tech in the USA.
He further explained: “A decade ago we had the opportunity to develop a flood forecasting system for Bangladesh and created a one to 10 day probabilistic scheme that was successful in predicting the 2004, 2007 and 2008 flooding of the Brahmaputra. Evaluation surveys found that the communities that used the forecasts were able to reduce losses of annual income. The flood forecasting and warning system developed for Bangladesh can act as a template for Pakistan”.
When the floods hit Bangladesh in 2007 and 2008, the communities were prepared. Thanks to the flood forecast and early warning system they had already stored their seeds, farming implements and harvest on higher ground. According to Dr Webster, “After the floods receded they could continue with their farming, although to a lesser degree. But they had their precious belongings and their resilience was not damaged. There was no downward spiral”. People all over Bangladesh have been trained on what to do when a flood is about to hit and an expanding cellphone network ensures that everyone gets the information on time. “You can’t stop floods, but you can get people out of the way”, he pointed out.
If we had the flood-forecasting module in place when heavy rains were falling in northern Pakistan in the summer of 2010, perhaps more water could have been released in Tarbela to allow for the extra flows that would be coming in. What about predictions of glacial melt swelling our river flows? “We have very little information about glacial melt but that is mainly background flow”, explained Dr Webster. In his view, the summer flooding in Pakistan is caused by erratic monsoon rainfall. While glaciers are retreating in the high mountains, the melt is not accelerating to the extent that it would cause major floods. “Glacial melt increases background flow, but it is not the main reason for flooding”.
Dr Webster has conducted “hind cast simulations of the 2010 and 2011 floods, again showing predictions 8-10 days in advance”. By studying fields of precipitation over Pakistan, Dr Webster has developed a new flood forecast model that calculates inundation in the delta. In his view, “the new hydro-meteorological forecast system portends considerable advantage for both flood warning and water resource management”. He added that: “the science is easy; getting the early warning system to translate into saving peoples lives is the hard part”. In the years to come, with hazards becoming more frequent and intense thanks to climate change, it is the societies that are the most adept at dealing with disasters who will manage to cope. We have to build up our resilience in Pakistan and a trustworthy flood forecasting system is the first step.