When the golf bursts its banks   January 1995

There is a glacier lake up in Rolwaling Himal above the settlement of Bed ing that is about to rupture. An earthquake or avalanche is likely to break the weak, moraine dam which props up the lake, and releases 71 million cubic metres glacial melt water.

If the water is released slowly, damage will belimited. But chances are that the marine collapse will be sudd en, in which case Nepal will experience a Glacier Lake Outburst Flood, or GLOF, which is greater than all the GLOF phenomena of recent decades.

The glacier lake is known as Tsho Rolpa, and is at the headwaters of the Rolwaling Khola in Dolakha District, northeast of Kathmandu, below the Gauri Shankar massif. Over the last four decades, glacial meltfrom the TrakrardingGlader has collected in a lake, held back by the loosely filled debris of the glacier´s end-moraine. The glacier is receding fast, andit is in the process of releasing a large volume of water which goes into the lake.

Satellite data, aerial observations and on-site investigations by the Water and Energy Commission Secretariat (supported by Japanese funding) show that the size of Tsho Rolpa has increased relentlessly. The surface area of the lake was about .23 sq km in 1959, while in 1993 it had expanded to 1.4 sqkm. The elongated lakeis today 3 km in length and has an average, width of half a kilometre. Towards themiddle, one measurement showed a depth of 132 m.

Like earthquakes, GLOFs are a fixture of Himalayan history. The evidence is found in erosion alongriver banks and breaches of terminal moraines. Pokhara Valley´s topography, in fact, was defined by debris brought down by a massive GLOF that occurred several centuries ago below Machhapuchare peak.

Even though it is a fixture of history, the awareness of GLOFs, and their potential to wreak havoc, is a recent phenomenon. Even at this late date, Nepal and Tibet/China are the only two countries in the Himalayan region that are doing any sdentific work in the field.

Chinese scientists studied the July 1981 GLOF from Zhangzangbo Lake on the Boqu River (Sun kosi- Bhote Kosi) in Tibet. In Nepal, the first GLOF event to draw scientific attention was when the Dig Tsho Lake on the Langmoche Glader (at the headwaters of the Dudh Kosi, Khumbu) burst in August 1985, destroying trails, foot-bridges, the Namche small-hydel project, and taking lives.

There have been many other GLOFs in just the last few decades. They include: release from Gelhaipuco Lake (Tibet) along the Arun River (Nepal), 1964; Phuchan Lake along the Tamur River (Nepal), 1980; the Arun basin (Tibet, west of Chomolongma), in 1968, 1969,1970,1982; along theSunKosi (Tibetand Nepal), 1935; on theLongda at the Trisuli source (Tibet), 1964; on the Nare Glacier (south slope of Ama Dablam peak), 1977; and Chubung GLOF at Ripimo Shar Glacier in Rolwaling, 1991.

Besides Tsho Rolpa, other glacier lakes that accelerated growth and bear watching are: the Thulagi Glacier Lake on the Marsyangdi Basin which has reached 2 km length today, up from .6 km in 1960; the Imja Glacier Lake which started forming around 1963 and today holds about 28 million cumof water; the Lower Barun Glacier Lake in the Arun Basin, which is not seen in the 1967 topographic map, but is today 0.6 sq km in area and contains 28 million cum.

To understand the devastation that abreak of Rolwaling´s Tsho Rolpa can lead to, the relatively modest 1981 GLOF on theSunKosi destroyed the diversion weir at the Sun Kosi hydro project, two bridges, and large sections of the Kathmandu-Lhasa Amiko Highway. For comparison´s sake, theDigTsho burst in theKhumbu involved release of no more than 10 million cu m water volume, compared to the 71 million cu m already impounded in Tsho Rolpa.

A sudden burst rather than a delayed release, would devastate downstream areas all the way from the Sherpa-inhabited Rolwaling through the entirestretchofmid-hillNepal,and possibly as far as the Tarai beyond Triboni where the three main tributaries of the great Sapta Kosi meet. Certainly, this would be one of the largest GLOF event of the last few decades.

Though time is short, itis possible tomitigate the impact of a Tsho Rolpa GLOF. The lake can be drained by a controlled breaching of the moraine, using massive water pumps, or tunneling under the dam. The most effective method seems to be siphoning of the lake, a method that is used in the Andes. Besides the high expense of massive siphoning structures, however, one has also to be careful thatdrainmgproceduresdonotthemselves trigger a GLOF due to some unprecedented event such as creation of a tidal wave when a glacier serac falls into the lake.

Many studies ha vebeen done on glacier lakes of theEastem Nepal and on TshoRolpa to indicate that the situation of Tsho Rolpa is dire. While continuing research, there should now be a concerted effort to raise fundsand gather expertise in order to drain the lake before it is too late.