3
IN the last five monsoon seasons, Pakistan has seen five floods. That means each year since 2010 has brought a massive flood that has affected the lives of millions of people in each case. Each flood has been caused by unusual rains. And the rainfall-producing storm systems in at least three of these years — from 2010 till 2012 — have been studied very carefully by a group of meteorologists who argue in their latest paper that these storms are not normal monsoon systems.
For the last couple of months, I have immersed myself in a careful reading of papers from the cutting edge of meteorological research being carried out on Pakistan’s flood-producing storms and the anomalous weather patterns behind them. The first results from my readings appear in the latest edition of Herald, this newspaper’s sister publication, as two long format reports. In this column, let me give a quick synopsis.
Scientific studies of the Indian monsoon, on whose western edge we sit, began in the late 19th century by the British colonial government following a deadly drought that led to mass starvation across the subcontinent. The Indian Met Department (IMD) was created in the aftermath of that event, and its first two directors general — Messrs Blanford and Walker — were the first people to begin detailed scientific observations of the annual cycle of rains that sustained life in such critical ways across British India.
From the earliest observations, two different directions emerged. Blanford searched for a land-based link between the monsoon system and the weather patterns that produced the winter snowfalls in the Himalayas. Walker on the other hand, was the first to discern a link between a peculiar seesaw-like variation in atmospheric pressure between the Indian and Pacific Oceans. This seesaw mechanism came to be connected later with the phenomenon known as El Nino.
This year, all Met Departments gave their flood alert barely 48 hours ahead of the flood peak — insufficient time in which to organise a response.
As the volume of meteorological measurements being undertaken increased over the decades of the 20th century, the two lines of inquiry initiated by these two men began to bear fruit. For most of the 20th century, meteorologists focused on the El Nino link and growing volumes of research output added more flesh to the proposition that the great Southern Oscillation, as the pressure reversal brought about by El Nino was called, holds the secret to forecasting the monsoon rains in India.
But forecasting the monsoon rains with any meaningful exactitude remained elusive, although it became possible to say with some measure of probabilistic certitude how wet the forthcoming wet season might be. Yet despite the increasingly technological sophistication of the data and the statistical models being used for meteorological observations, the IMD was wrong more often than it was right in its forecasts.
This year, for instance, the IMD forecasted a dry monsoon season, with below average rains and even the possibility of drought. A newly launched private weather forecasting service in India came to the same conclusion, citing the appearance of El Nino in the Pacific in April, saying that a drought is likely across North India with monsoons being far below average.
The CEO of the private service was even quoted, in July, saying if there are no rains in July then the monsoon will most likely fail. The drought was going to be particularly intense in the northwest, according to both of these services.
The Pakistan Meteorological Department (PMD) also acknowledged the appearance of El Nino this year in April, but hedged its forecast, saying “irregular rains” are likely this season.
The rains were certainly irregular. When they came they were so heavy as to submerge the provincial capital of the very region that the IMD and the private forecaster were saying would be at the epicentre of a drought. In fact, a report by Deutsche Bank, widely cited in the Indian media, looked in detail at the IMD’s forecasts over the last 20 years and found that they have been correct in only four years. A coin toss is more likely to give you a more accurate result.
The story has been largely the same every year since 2010. Why are the Met Departments having such a hard time detecting these storms with any meaningful lead time? This year, all Met Departments gave their flood alert barely 48 hours ahead of the flood peak, grossly insufficient time in which to organise a response.
Part of the answer takes us back to the differences between the approaches taken by Blanford and Walker. New research is finding out that the El Nino connection might be overstated, that the Indian monsoon interacts with weather systems in the Eurasian landmass as well as more distant systems. It’s also telling us that the very structure of the monsoon system appears to be changing, producing rainfall patterns that are entirely anomalous.
While details are contained in the longer report, for now it’s enough to say the following. Flood forecasting in an era of climate change is a crucial priority for us now. Five floods in five years are enough of a hint that our climate is changing in crucial ways. We cannot reverse the process, nor did we create it. But we must adapt to it, and adaptation begins by upgrading our forecasting abilities so we can have some lead time in which to prepare our response.
The technology to do this exists, and was offered to Pakistan last year by the World Bank, but the authorities were too busy in other matters to pay any attention. Instead each year’s floods have amplified calls within Pakistan for building more dams and barrages and other hydrological infrastructure to be used for management of floodwaters. But more infrastructure for flood management is pointless in the absence of longer and more reliable forecasts.
The writer is a member of staff.
Twitter: @khurramhusain
Published in Dawn, October 9th, 2014
