Calculating glacier volume change from Space

This article on glacier volume change was written by Ethan Lee, from Newcastle University.

Introduction

Around the world, glaciers provide water for 2 billion people, however almost all glaciers are melting and shrinking reducing the amount of water available to them. We need to know how much glaciers are melting by and at what rate, because glacier melt contributes to sea level rise1, 2 and reduce water storage for the future. Glaciers will continue to melt and will increase in their rate of melt globally due to future climate change3, 4. But how do we know how much glacier volume has been lost?

Glacier mass change from 1961 to 2016 as estimated for different regions. The cumulative regional and global mass changes (in Gigatonnes, represented by the size of the bubbles) are shown. In orange, the locations of glaciers from RGI 6.0 are illustrated. Data source: Zemp et al. (2019)16. Credit: ESA/Zemp et al. https://climate.copernicus.eu/ESOTC/2019/glaciers-and-sea-level-rise

However, many glaciers are difficult to access, and only a few have long-term measurements. We can use satellite imagery, from the 1970s onwards, to map glacier extent change, and to work out how much snow and ice has been lost from glaciers each year. This includes both how much the glacier has shrunken, with the terminus of the glacier receding, but also how much the glacier has thinned, with surface of the glacier melting away (shown in the image below).

Photographs showing glacier thinning.
Panoramic view of West Rongbuk Glacier and Mount Everest, taken in 1921 (top) by Major E.O. Wheeler and in 2009 (bottom) by David Breashears. See the surface lowering from the glacial ice on the sides.

Using satellites to determine glacier mass and volume change

Satellites are used extensively to map and understand glacier changes in glaciological research. They have primarily been used to create inventories of glaciers globally6, 7, map glacier recession by mapping glacial depositional landforms to reconstruct past glacial extents8, 9, 10, and map glacier extents over time from old satellite images or aerial photography11, 12.

Mapping extents are great, and understanding how much glaciers have, and are, retreating by provides us with information on the state of glaciers. However, frontal retreat is not the entire story and understanding glacial change in three-dimensions allows us to understand more of the change being seen.

How do we determine three-dimensional glacier change?

Satellites are the answer. Satellites can give us the elevation of the ground they take an image of, which will include the surface elevation of the ice at that period of time. These are called Digital Elevation Models (DEMs). There are a number of global DEMs generated from these satellites.

Some of these DEMs are generated from overlapping satellite images (such as ASTER imagery). Other sensors use different techniques. The ICESat satellite, for example, measures ice surface elevation along a track, and repeated tracks show elevation change through time.

DEMs can also be generated from aerial photographs if they have stereo-pairs, this is if they have two (or more) overlapping images taken at the same time. There are a number of satellite platforms and arial systems that provide us with such information:

Satellite NameGround resolution (pixel size)Operation datesArea covered
SRTM30/90 mFebruary 2000 (single mission)Near global (56°S to 60°N)
ASTER GDEM30 mDecember 1999 – PresentNear global (83°S to 83°N)
CryoSat22.5 mApril 2010 – PresentArctic, Greenland, and Antarctica
ICEsat23 mSeptember 2018 – PresentArctic, Greenland, and Antarctica
HMA DEM8 mJanuary 2002 – November 2016High Mountain Asia
WorldDEM5 mJune 2010 – PresentGlobal
SPOT-6/71.5 mJune 2014 – PresentGlobal
Pléiades50 cm/2 mDecember 2011 – PresentGlobal
CORONA1.8/7.5 mFebruary 1962 – May 1972Limited
Key satellites used to generate Digital Elevation Models and some key DEM products available.
The Terra (EOS AM-1) satellite, operated by NASA, with the MODIS and ASTER sensors, is regularly used by glaciologists due to its affordability, wide swath (60km) and good resolution (15 m). Image credit: Wikipedia.
Satellite image of James Ross Island. Overlapping satellite images are used to make digital elevation models.
ASTER image of James Ross Island, taken 03 March 2009. Swath size is 60km. The plateau icefield, with a low-slope upper plateau surface, dominates the main part of the island.

DEMs of Difference

Using DEM data, if we have two DEMs that cover the same area that were taken at different dates, we can ‘difference’ these images to create a ‘image of difference.’  Differencing two DEMs is as simple as taking away the values from one image from the other. This is shown simply by the equation:

Where dh is the change in surface elevation, and ht2 and ht1 are the two DEMs taken at time1 and time2 . This will give the total elevation change between the two dates.

We want to then know the rate of this change, and how this rate has increased or decreased over time. Calculating the rate is dividing the total elevation change by the difference in time between the two DEMs. This is shown by the equation:

Where dh/dt is the elevation change over time, this is generally shown per year or per annum (yr-1 / a-1), and t is the length of time (in years) between the two DEM images. The outcome of this is an image showing where glacial ice is thinning (negative change) and places where ice is increasing in height (positive change). Examples of where you can find such data to view or to download and use are from Theia cartographica and from Hugonnet et al.2.

Glacier thinning for the European Alps
An example of an output from DEM differencing in the European Alps, showing dh/dt (m y-1). Image taken from Theia cartographica. Outlines of glacier are from the RGI 6.05

Calculating glacier volume change

Due to elevation change of the glaciers surface not providing much useful information, we convert elevation change into information that can be used to aid in comparisons. We can convert elevation change into volume change (km3), that can be used to understand any potential sea level rise from glacial melt. Volume change is shown in the equation below:

Where V is the volume loss you want to figure out for each cell, cell size is the ‘resolution’ of the DEM. Total volume loss can then be summed for each glacier.

Many scientists also show ice loss as ‘meters water equivalent (m w. e.)’ which is based on the density of ice. For example a value of -1.0 m w.e. per year is the same as an annual glacier-wide ice elevation loss of ~1.1 m per year, as the density of ice is 0.9 times the density of water.

DEMs of difference in the Himalaya

An example of using DEM differencing to understand glacier change is by King et al 13 in understanding mass loss of Himalayan glaciers since ~1974-2000, and 2000-~2015 in relation to glacial lakes.

King et al found that the rate of mass loss of Himalayan glaciers had change very little between these two time periods but had seen that after the year 2000 glaciers that had a glacier lake in front of them lost more glacial ice than those that did not. This can be seen in the figure below, where a glacial lake (blue outlines) is in front of a glacier there is substantially more melting (redder over the glacier).

Glacier thinning over the Khumbu region
Surface elevation change over the Central 2 (Khumbu) region over the period 2000-~2015. Figure from King et al. 13

Understanding glacier mass and volume changes before the satellite era

Of course, since satellites have only really been around and used in scientific research since 1972, when LANDSAT 1 was the first open-access satellite for scientific use, and the use of stereophotography from limited missions, we have been able to fully observe glaciers globally. But what about for times before 1972? How do we know if glaciers have been losing mass slower, faster, or at the same rate as they are currently? Luckily, we have other sources and methods that we can use to fill in this ‘lost time’.

Volume change using topographic maps

In countries where there have been extensive mapping expeditions to create topographic maps in regions where glaciers exist, we can use these to generate DEMs. This is very involved process that requires to use of a Geographic Information System (GIS) to transform a paper-based map into a raster image and DEM using contours (example of how this is done here).

Glacier volume change can also be reconstructed from topographic maps
A topographic map from 1909 of Mt Baker in Washington State clearly showing the Mazama and Coleman Glaciers (modern day Rainbow, Park, Boulder, Mazama, Roosevelt and Coleman Glaciers). Topographic map taken from the USGS topoView website.

When this done, we can then do a simple DEM differencing similar to the above. Another way we can understand mass loss since is by using the features glaciers have left to inform not only past glacial positions but thickness (or elevation).

Volume Change using glacial features

One such study by Lee et al.14 was able to push back mass change rates within the entire Himalaya from ~1970s, to when glaciers started to retreat from their Little Ice Age extents (400-700 years ago). The Little Ice Age was one of the last known periods where, almost globally, glaciers advanced.

To do this, Lee et al14 mapped the glacial moraines in front of glaciers and reconstructed their glacial extents to these moraines. Moraines being a mass of rock and sediment that is carried and deposited by a glacier at its edges or frontal position and look like small ridges. Using these moraines, it can be assumed that the moraine crest (the top of the moraine ridge) can represent the thickness of the glacier, and if we assume this, we can interpolate between other moraine crests to generate a flat glacier surface to represent its former surface elevation. With this, we can then do a difference between a modern DEM with the interpolated surface elevation of the glacier.

With this, Lee et al. determined that glaciers across the Himalaya have lost at least 40% of their area, and 390-586 km3 of their glacier volume, with a mass loss of -0.011 and -0.020 m w.e. yr-1. When compared to studies looking are more resent mass loss within the Himalaya 15, this is ten-times lower then present day mass loss rates. This provides important context on the state of glaciers and their response to modern day climate change when the longer time context is taken into account.

Glacier volume change from pre-satellite era can be reconstructued using moraines.
Example from the Langtang region of the Himalaya, illustrating geomorphological evidence comprising moraines and trimlines (A) used to delineate past glacier extent (B) and to reconstruct former glacier surfaces (C). Differencing of the reconstructed surface with a contemporary digital elevation model was used to quantify elevation change (D). The dataset analysis and preparation of this figure was made using ESRI ArcGIS software (v. 10.6). Figure is from Lee et al.14

Further Readings

About the author

Ethan Lee, Newcastle University

I am a glaciologist that focuses on glaciological changes from past and modern climatic change. I use remote sensing to monitor and map glaciers and glacial geomorphology in order to reconstruct glacier extents, thickness and dynamics. Past research has focused on glacial changes within the Himalaya since the Little Ice Age (400-700 years ago).

More recently my PhD has been on determining the timing, nature, and extent of palaeoglacier advances in the tropical Andes, Peru. To do this I have used surface exposure dating, geomorphology and glacier modelling (PISM).

References

1.         Zemp M, Huss M, Eckert N, Thibert E, Paul F, Nussbaumer SU, et al. Brief communication: Ad hoc estimation of glacier contributions to sea-level rise from the latest glaciological observations. The Cryosphere 2020, 14(3): 1043-1050.

2.         Hugonnet R, McNabb R, Berthier E, Menounos B, Nuth C, Girod L, et al. Accelerated global glacier mass loss in the early twenty-first century. Nature 2021, 592(7856): 726-731.

3.         Slater T, Hogg AE, Mottram R. Ice-sheet losses track high-end sea-level rise projections. Nature Climate Change 2020, 10(10): 879-881.

4.         Edwards TL, Nowicki S, Marzeion B, Hock R, Goelzer H, Seroussi H, et al. Projected land ice contributions to twenty-first-century sea level rise. Nature 2021, 593(7857): 74-82.

5.         RGI Consortium. Randolph Glacier Inventory – A Dataset of Global Glacier Outlines. Technical Report; 2017.

6.         Raup B, Racoviteanu A, Khalsa SJS, Helm C, Armstrong R, Arnaud Y. The GLIMS geospatial glacier database: A new tool for studying glacier change. Global and Planetary Change 2007, 56(1): 101-110.

7.         Pfeffer WT, Arendt AA, Bliss A, Bolch T, Cogley JG, Gardner AS, et al. The Randolph Glacier Inventory: a globally complete inventory of glaciers. Journal of Glaciology 2014, 60(221): 537-552.

8.         Pearce D, Ely J, Barr I, Boston C. Glacier Reconstruction. 2017.

9.         James WHM, Carrivick JL, Quincey DJ, Glasser NF. A geomorphology based reconstruction of ice volume distribution at the Last Glacial Maximum across the Southern Alps of New Zealand. Quaternary Science Reviews 2019, 219: 20-35.

10.       Lee E, Ross N, Henderson ACG, Russell AJ, Jamieson SSR, Fabel D. Palaeoglaciation in the Low Latitude, Low Elevation Tropical Andes, Northern Peru. Frontiers in Earth Science 2022, 10.

11.       Sidjak RW. Glacier mapping of the Illecillewaet icefield, British Columbia, Canada, using Landsat TM and digital elevation data. International Journal of Remote Sensing 1999, 20(2): 273-284.

12.       Silverio W, Jaquet J-M. Glacial cover mapping (1987–1996) of the Cordillera Blanca (Peru) using satellite imagery. Remote Sensing of Environment 2005, 95(3): 342-350.

13.       King O, Bhattacharya A, Bhambri R, Bolch T. Glacial lakes exacerbate Himalayan glacier mass loss. Scientific Reports 2019, 9(1): 18145.

14.       Lee E, Carrivick JL, Quincey DJ, Cook SJ, James WHM, Brown LE. Accelerated mass loss of Himalayan glaciers since the Little Ice Age. Scientific Reports 2021, 11(1): 24284.

15.       Maurer JM, Schaefer JM, Rupper S, Corley A. Acceleration of ice loss across the Himalayas over the past 40 years. Science Advances 2019, 5(6): eaav7266.

16.       Zemp, M., Huss, M., Thibert, E., Eckert, N., McNabb, R., Huber, J., Barandun, M., Machguth, H., Nussbaumer, S.U., Gärtner-Roer, I., 2019. Global glacier mass changes and their contributions to sea-level rise from 1961 to 2016. Nature 568, 382-386.

This site uses cookies. Find out more about this site’s cookies.