Climate Change

Weekend cloudburst reiterates need to beef up early warning systems; is Uttarakhand on track

There have been positive and negative developments in Uttarakhand since 2013 as far as early warning systems are concerned; the key is to minimise the negative ones 

By Anand Sharma
Published: Tuesday 23 August 2022
Encroachments on water bodies and slopes are weak links in an early warning system. Photo: Megha Prakash
Encroachments on water bodies and slopes are weak links in an early warning system. Photo: Megha Prakash Encroachments on water bodies and slopes are weak links in an early warning system. Photo: Megha Prakash

It has been almost a decade since the terrible cloudburst at Kedarnath in which many people died. And yet, cloudbursts and flash floods continue to happen. The latest cloudburst to happen was in Raipur near Dehradun over the weekend. At least four people have died in the state and others are missing.

The 2013 weather system that caused the cloudburst was more powerful than the Raipur one. It also covered a greater area. The system that caused the Raipur cloudburst was more or less localised in Uttarakhand, mostly over Tehri and Dehradun district.


I had studied a research paper. I was also the in charge of the flood Met office in New Delhi. I knew that floods in Delhi were caused by the interaction between the westerlies and the easterlies. I was tracking the system and hence I was able to predict it. I was also upgrading the forecast consistently.

This shows that it is one thing to know about a system. But a lot also depends on constant monitoring and the forecaster’s gut feeling to sound out an early warning.

I had formulated the forecast in very simple language: If one was in the hills, they had to head to the plains. If one was in the plains, they were not to head to the hills. Delay and reschedule your journey to the Char Dham pilgrimage spots. There would be heavy to very heavy rains along with landslides. So all things kind of fell into place.

In the Raipur cloudburst, the news here in Uttarakhand is that the Met Department was not able to predict the system.

So there was no warning in the first place. Second, people these days are constructing structures over natural watercourses and slopes in the hills. The administration is not doing much to stop them.

There are flash floods in the hills, rather than floods. If there is even 5-6 centimetres of rain, the water mixes with rock, sediment and soil and comes crashing down the slopes.

If there are impediments in its way such as structures and buildings, where will the debris go? These structures have been built in the last two decades since the formation of Uttarakhand and most of them are unplanned constructions.

A decade later

There have been two-three changes since the tragedy of 2013. One is that structures are being built at places where there have been incidents in the past.

Read Uttarakhand may be staring at another disaster

For instance, people have told me that the Kedarnath Valley and its nearby areas, which were destroyed in 2013, are now again the site of several structures. This means people have not learnt any lessons.

Second, technology has improved since 2013. But a radar gives a warning only two hours prior to the actual occurrence of a cloudburst or flash flood.

For instance, the Kedarnath system could be predicted and warned about at least four days in advance. But sometimes, it may happen that a system is not strong initially. But then cumulonimbus clouds form and the cloudburst occurs within two hours.

This too can be caught by a radar. But suppose it happens at night. Who will be monitoring it then? For that, we need 24X7 monitoring like in the United States.

An early warning system has four components. The first is the risk and vulnerability analysis. One should know which areas are particularly vulnerable to flash floods and debris.

The second is weather monitoring and prediction. The third is that you may have predicted that a cloudburst or flash flood may occur. But you have to communicate it on time to the administration and the public.     

The fourth is government response. If you have allowed haphazard construction in vulnerable areas then quickly evacuate people.

These four components form links in a chain. If even one link is weak, your early warning system will fail. This is what happened in Kedarnath. There was prediction, monitoring and communication but no risk and vulnerability analysis and poor government response.

In the Raipur case, there was no early warning and there was haphazard construction. And it happened at 2 pm in the night.

These are the lessons that Uttarakhand must learn to improve its early warning systems.

Anand Sharma is the ex-Additional Director-General of IMD and is currently a visiting faculty at the Forest Research Institute, Dehradun

Views expressed are the author’s own and don’t necessarily reflect those of Down To Earth

Subscribe to Daily Newsletter :

Comments are moderated and will be published only after the site moderator’s approval. Please use a genuine email ID and provide your name. Selected comments may also be used in the ‘Letters’ section of the Down To Earth print edition.