Fieldwork report: Khumbu Glacier, 2015

By Scott Watson

Back in October, Owen King and I travelled to the Khumbu Glacier in Eastern Nepal, accompanied by two of our supervisors Duncan Quincey and Ann Rowan. The aim was to collect field data on how the glacier is thinning year-on-year, and to validate and improve our satellite remote sensing observations that have been the focus of our PhDs prior to fieldwork.


Location of our field site in the central Himalaya

The Khumbu Glacier is the highest glacier in the world and every year a small section of it becomes the home to Mount Everest Basecamp. Travel to the glacier involved a seven day walk from the nearest airstrip at the village of Lukla, which included two acclimatisation days to cover our ascent from 2,800 m to ~5,000 m. The highest elevation we reached (excluding a brisk jaunt up the small trekking peak Kala Patthar) was ~5,300 m at an unusually quiet Everest Basecamp, which was deserted following the recent earthquake.


Our campsite (left) and the lower debris-covered area of the Khumbu Glacier (below)

1Everest region background

It’s widely known that debris-covered Himalayan Glaciers in this region are losing ice mass year on year, although the presence of a thick layer of sand and rocky debris delays their response to climate change. The glaciers are currently out of equilibrium with climate, and will continue to thin irrespective of any contemporary slowdown in climatic warming. The debris cover, which is generally thickest at the terminus of the glaciers and becoming thinner at higher elevations, changes the spatial distribution of melt. Generally speaking, highest melt rates occur where the debris is thin or absent, whereas thick debris insulates the ice beneath.

Supraglacial ponds (i.e. existing on the surface) and ice cliffs are widespread on the low-gradient, debris-covered areas of the glaciers. Ice cliffs can range from several to tens of metres high, and ‘ponds’ can be over 100 m in diameter. Although data are limited, it is thought that ice cliffs and ponds contribute highly to overall melt at a glacier scale. At ice cliffs, bare ice is melted by incoming solar radiation, and ponds are similarly warmed and transmit this thermal energy to the ice below, or through conduits draining into the glacier.

The area of surface water ponding is increasing on the Khumbu Glacier and individual ponds are coalescing, which are likely to form a large glacial lake in coming decades.

Field data

The debris-covered area of the Khumbu Glacier is ~10 km long and the lower ~5 km is stagnant, whereas flow exceeds 60 m a-1 in the ice fall. Our work was predominantly in the lower 6 km where ponds are coalescing and large ice cliffs were present.

The high-altitude rugged topography, hazardous access, and unstable behaviour of the debris-covered glaciers means that field data area limited and hard-earned. Crevasses are minimal on the lower Khumbu Glacier but ice cliffs and topographic highs are constantly changing as the ice melts beneath. The hummocky topography was unstable and would often slump, making it hard to cover any nominal distance. Each hummock was essentially a rocky, ankle-twisting version of the Gladiators Travelator. Our Nepali guides were invaluable in this environment for route finding and helping carry our field equipment. A fine-resolution satellite image base map on a GPS device also made locating the cliffs and ponds considerably easier.

Ice cliffs: structure-from-motion and multi-view stereo (SfMMVS)

A photographic survey tailored to the requirements of the SfM-MVS workflow was conducted around each ice cliff selected for study, which basically requires photographing the environment from as many different locations as possible (i.e. 1- 2 hours clambering around the Khumbu Travelator). Specialist software (e.g. Agisoft Photoscan) is able to match the images and create a 3d point cloud representing the ice cliff and surrounding environment. Visible ground control points were distributed around each cliff and georeferenced with a dGPS before the photographic survey so the model could be scaled and georeferenced.

A preliminary example, which can be navigated in 3D is available at:


Example of a 3D ice cliff model

Pond surveys

The ponds forming on the surface of the glacier can broadly be split into two classes: those with active meltwater inflows (e.g. from adjacent ice cliffs), and those that are hydrologically isolated from any melting ice and thought to be relatively stable through time. My monitoring strategy involved deploying thermistor strings in a range of ponds to measure their thermal regime. These temperature loggers were deployed using an inflatable dinghy (christened HMS Khumbu), which was kindly borrowed from a fellow PhD student also working in Nepal. Most ponds were frozen at their surface by the end of the campaign, so access though ~10 cm of ice was required for logger retrieval (an ice screw/ ice axe combo worked well).

During our next field campaign, I plan to collect distributed depth measurements across a number of ponds to derive bathymetry, and hence determine the water storage volume of the ponds.


Deploying temperature loggers in a supraglacial pond


Scott Watson is a PhD student at the University of Leeds and a BSG Postgraduate Committee member. Scott’s fieldwork in 2015 was supported by a University Research Scholarship, the Royal Geographical Society, the British Society for Geomorphology, and water@leeds.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s