A scientific analysis of the reasons for the disaster that struck Uttarakhand, particularly the temple town. By R. RAMACHANDRAN
THE primary trigger for the Uttarakhand disaster following the very heavy rain during June 16-18 was the extremely unusual behaviour of the monsoon this year over north India. The incessant, heavy rainfall over three days, perhaps accompanied by a few cloudburst-type events (which cannot be confirmed), resulted in flash floods and associated landslides. The devastation all round in their wake has been huge but the largest impact has been at the temple town of Kedarnath, which was in the midst of the annual pilgrimage season, with tens of thousands of people thronging the town and the downstream region along the Mandakini river (Picture 1).
Rainfall measurements for June 16 and 17 at the Dehradun station, of 220 millimetres and 370 mm respectively, indicate the severity of the rain during these days in the region. Haridwar received 107 mm and 218 mm of rainfall on the two days. Uttarkashi received 122 mm and 207 mm. While Mukteshwar (altitude over 2,000 metres) received 237 mm and 183 mm respectively on June 17 and 18, Nainital on the same days received 176 mm and 170 mm. Though rainfall over a 24-hour period in different parts of Uttarakhand has greatly exceeded these figures in the past (on many occasions above 450-500 mm and once even 900 mm at Rajpur near Dehradun), prolonged heavy rainfall for nearly three days over a large area is perhaps unprecedented, and the cumulative effect, compounded by geophysical, meteorological and environmental factors, may be the reason for the enormity of the disaster.
More pertinently, these numbers do not give the actual quantitative picture of the very heavy rainfall in the higher reaches of the Himalayas (above 3,000 m) in Uttarakhand, where Kedarnath, Gangotri and Badrinath are located and where the impact has been most severe. This is because the rain gauge stations of the India Meteorological Department (IMD) are all located largely in the lower Himalayan reaches (below 2,000 m) and there are no stations in the higher reaches (above 3000 m). This is probably because snowfall data is regarded as more important than detailed rainfall data in these regions. As a result, there is no proper estimate of the rainfall in the affected regions.
What was peculiar about the monsoon this year? On June 14, the monsoon front was located over eastern India. In fact it was a trifle sluggish compared with the normal progress of the front (Map 1a). But within a day (Map 1b), the front advanced right across Uttar Pradesh and the western regions to cover the entire country by June 15, exactly a month ahead of its normal date of July 15. While the IMD had forecast a rapid advance with the announcement that the monsoon would strike Delhi before the normal scheduled date of June 30, its advance right across to the west just within a day was entirely unexpected.
This has never happened in the past, according to M. Rajeevan, Adviser in the Ministry of Earth Sciences (MoES). A system of westerly winds from the Arabian Sea had also been active during the same period and had covered Pakistan. It was a strong westerly system, and Rajeevan noted that it was similar to the system that stayed anchored over Pakistan in July 2010 and caused widespread flooding in Sindh, Punjab and Baluchistan. Of course, by July 2010 the south-west monsoon had covered entire India, but this time around the system had formed in June itself.
It was the interaction between the well-formed low-pressure system of the south-west monsoon from east to west and the upper air westerly trough running from north-west Rajasthan to the east that resulted in the heavy rainfall over Uttarakhand. In fact, the westerly system dragged the monsoon trough, which was anchored over Rajasthan and central India until then, towards the north across Haryana. A monsoon trough facilitates the movement of rainfall-causing low-pressure systems along its path. Its rapid movement northwards enabled the low-pressure system that was in the eastern part of the country to quickly traverse and locate itself over north-west India.
According to Rajeevan, while the phenomenon of the monsoon trough being dragged northwards by the advancing strong westerly trough is known to occur, the exact dynamics of interaction between the two systems is not well studied. Thus, as the press release of June 20 of the IMD noted, “North-west India became the zone of an unusual confluence of the two branches of the monsoon—the Arabian Sea branch and the Bay of Bengal branch. The geology and orography of [the Himalayan regions] of Uttarakhand and Himachal Pradesh resulted in the unprecedented impact in these two States.” While the IMD had issued warnings of widespread severe rainfall in the region soon after the observation of the advancing monsoon systems, the scale of impact could not be anticipated.
The peculiarity of the monsoon apart, the other interesting question is what geophysical dynamics channelled the major part of devastation along the Kedarnath valley and downstream of Kedarnath on the Mandakini. The region around Kedarnath is known to geologists to be prone to landslides. This is also clear from an early 1882 Geological Survey of India photograph of Kedarnath (Picture 2), which shows that the temple site is located not far away from the snouts of two mountain glaciers.
David Petley, an expert on landslides at the Department of Geography at Durham University, United Kingdom, has analysed the calamitous event at Kedarnath on the basis of images from the remote-sensing satellites of the Indian Space Research Organisation (ISRO) and the U.S. Landsat. He points out that the amount of debris and rubble below the glacier on the left side of the 1882 picture suggests that transportation of sediment and debris from the upper reaches was active even then, and adds that the steep slope that is visible would have aided rapid transportation. It should be borne in mind that the geology is still roughly the same (Picture 3).
It is evident from the post-event images of Kedarnath town around the temple that the massive destruction was the result of large-scale debris carried by the huge volume of water from the upper reaches above the town. One of the compounding factors was that the glacial regions above Kedarnath had received fresh and excess snowfall when heavy rainfall hit the region (Pictures 4a & b), according to scientists of the National Remote Sensing Centre (NRSC) of ISRO. Rainwater, with higher temperature, falling on the snow must have led to heavy snow melt and this runoff would have added to the rainwater runoff, resulting in a huge water flow that carried with it a huge debris flow, which struck the town with enormous ferocity. The snow cover has, in fact, increased in general subsequent to the extreme rainfall and flooding events (the satellite image on May 28/June 1 shows less snow cover). According to the NRSC scientists, the detailed dynamics of water flow due to snow melt caused by rain, particularly when snowfall is in excess, and the hydrology of it are not well understood.
The NRSC recently released excellent high-resolution pre-flood and post-flood images of the Kedarnath region (Pictures 5a & 5b) taken by ISRO’s remote-sensing satellites Cartosat-2A and Resourcesat-2. The NRSC, on the basis of remote-sensing images from Resourcesat-2, has carried out an inventory of the landslides that occurred between Kedarnath and Sonprayag, a distance of approximately 20 km on the Mandakini. According to the preliminary report, the study identified a total of 192 landslides in this Himalayan stretch (Picture 6). Many landslides were triggered in the glacial regions in the mountains above Kedarnath. The large-scale debris flows from above were the result of these massive landslides.
Actually, for Kedarnath it was a double whammy. The massive damage caused to Kedarnath town can be seen clearly in the post-flood image. Just as there was an unusual confluence of two monsoon streams up in the atmosphere, in the mountainous terrain around Kedarnath, too, there was a coincidental reinforcing of two massive debris flows from above, one from the north-western side of the Kedarnath temple and the other from the north-eastern side. Petley has analysed these images to arrive at a plausible scenario as to what caused the massive onslaught on the town, virtually flattening it. This flow cascaded further and caused heavy damage downstream as well. The NRSC scientists, too, in their analysis, have come roughly to the same general conclusion.
According to Petley, the two different but reinforcing events that caused the disaster were landslide-induced debris that came from the glaciated area in the north-east and a glacial-related flow that originated from the north-west glacier. From the images, one can distinctly identify the two flows.
Petley, from his analysis of the images, (http://blogs.agu.org/landslideblog/2013/06/27/new-high-resolution-images-of-kedarnath-the-cause -of-the-debris-flow-disaster-is-now -clear/), has inferred the following:
1. The flow from the north-east came down the margin of the glacier and spread out to strike the town.
2. The north-west flow descended from the other glacier to hit the town.
3. While a large part of the flow from the north-west passed the town on its west side, a part also struck it directly.
On the basis of the pattern of overlay of sediments and their nature, Petley concludes that the flow from the north-west occurred after the one from the north-east. According to him, the debris flow from the north-east was triggered by a large, 75 m wide, landslide caused by heavy rainfall high on the mountains, which then came down the steep slope about 500 m, gathering the debris in its path. The flow was initially channelled into a narrow gully formed by the glacier and on exiting it the flow spread out in the floodplains before striking the town over a large area. The steepness of the slope would have given the debris enormous velocity when it struck the town. The total length traversed by this debris flow is estimated to be about 1,200 m.
The event from the north-west was, however, quite different, points out Petley. The spot marked 1 in Picture 7 is a moraine, which had created a block for a basin to form, allowing the water to build up in it as a pool or a lake. This is what the local people call the Chorabari Tal, to which, in fact, pilgrims trek a few kilometres along the west side of the valley to have a dip. The Chorabari glacier has been retreating constantly in modern times, and according to D.P. Dobhal of the Wadia Institute of Himalayan Geology, it has retreated about 300 m since 1960.
“The effect of the retreat is to leave a moraine that can allow lakes to form, which can then collapse,” pointed out Petley in an e-mail message to Frontline. “In Kedarnath, this is exactly what happened. I am not sure when the lake basin formed—it may not have been in modern times—but this is a dangerous situation. Of equal concern is the trend towards more intense rainfall, especially if this occurs early in the year (that is, during snowmelt),” Petley added.
Wall of water
Eyewitness accounts say a huge wall of water swept the Kedarnath town in a flash. The spot marked 2 shows that the moraine had been breached by the rapidly building up water because of heavy rainfall and the water overtopping the moraine wall. The breach led to the sudden release of the impounded water and resulted in a massive wall of water sweeping across the Kedarnath valley and the town and causing a huge flash flood.
According to the NRSC scientists, this lake would have had a depth of about 15 m, and the event was not exactly a glacial lake outburst flood (GLOF), which occurs when a dam or moraine wall is breached because of the sheer pressure exerted by the stagnant glacial water and ice that it encloses. This was a case of lake flooding because of excessive rainfall and consequent overtopping of the moraine wall, which eventually breached.
The flow was so huge and forceful that it overtopped the moraine on the other side of the glacier as well, at the spot marked 3, resulting in three flows: one moving south-east to join the earlier debris flow from the north-east and enhancing it before turning southwards and striking the town. The third flow is a new channel that opened up, perhaps exploiting an existing old channel, because of the breach at spot 3. Heading down the slope towards the town at great velocity, it gathered sediment and debris en route and resulted in a much-widened flow closer to Kedarnath.
However, the bulk of the debris flow, as Picture 7 shows, moved southwards towards the town down the main channel on the south-western side, which is the normal channel for glacial water flow. The spot marked 4 shows heavy erosion due to the flow in the area, suggesting that the flow must have carried a huge volume of water. According to Petley, this flow must have carried the many huge boulders and rocks seen in the post-flood image of the temple town. Closer to the town, the flow spread before striking. As a result, the debris and water flow moved to the east side of the town as well, engulfing the town from both sides. According to this picture supplied by Petley, which others too are in general agreement with, Kedarnath was first pounded by an earlier debris flow from the north-east, then a later pounding by the flow from the north-west. Petley suggests that the latter flow must have been more efficient because of the preceding events and also because it struck the town from both the west and the east simultaneously.
The image also shows a dark patch just above Kedarnath on the north-eastern side (to the right of the spot marked 5) suggesting the formation of a new depression, which could have turned into a small-sized lake because of the heavy rainfall. It is also possible that water built up in this new depression, which would have been substantial, overtopped it and hit the town from the eastern side, enhancing the effect of the runoff and debris flow from the north-eastern side, an aspect that Petley has not considered.
Downstream of Kedarnath, the flow remained contained within the channel. As a result, there was massive erosion of the banks of the Mandakini (Pictures 8 a & b). Further, smaller villages downstream were also severely damaged, and some of them, such as Rambara, were totally destroyed (Pictures 9 a & b).
The damage caused to the Kedarnath region and downstream villages by the natural destruction resulting from unusual meteorological and geophysical processes was undoubtedly greatly enhanced because of the general environmental degradation caused by the massive and unregulated influx of pilgrims year after year, the haphazard development fuelled by tourist traffic, and the unplanned and poor construction of buildings and roads. Given the vulnerability of the region, the town itself has come up in a very dangerous location, points out Petley. Therefore, how much of the destruction in this event was actually man-made is a moot question.
Besides the challenges of disaster management on such a massive scale, the Uttarakhand floods have also thrown up a lot of scientific challenges in the detailed understanding of monsoon dynamics as well as in the geophysical processes of landslides and large-scale debris flow and the heavy damage they can inflict on life, property and the ecology of a region.