Watch this brief introductory video, made by Time for Geography with Bethan Davies and Simon Cook, about glacier mass balance. This is suitable for GCSE and A-level students.
This more in-depth lecture (29 minutes) introduces the concept of glacier mass balance, and then goes on to discuss how glaciers worldwide are shrinking, and what this means for global water resources and sea level rise. This lecture is aimed at A-Level and Post-16 students.
This lecture by Bethan Davies, from the Reading Climate Festival in November 2020, goes into more detail about global glacier change (1 hr).
How do we build a glacier? We start with a snowflake. Snow, over time, is compressed into firn, and then into glacier ice.
Snow falls in cold regions, such as mountain tops or in polar regions. In glaciology, snow refers to material that has not changed since it fell1.
Snow is very light and fluffy, and has a very low density. If the snow is wetter, it will have an increased density. Snowflakes have a hexagonal structure, and fallen snow has a significant amount of air in it.
Firn is usually defined as snow that is at least one year old and has therefore survived one melt season, without being transformed to glacier ice.
Firn is transformed gradually to glacier ice as density increases with depth, as older snow is buried by newer snow falling on top of it. Year after year, successive accumulation layers are built up. In the accumulation zone of a glacier, density therefore increases with depth; the rate depends on the local climate and rate of accumulation1. Firn transforms to glacier ice at a density of 830 kg m-3.
New snow (immediately after falling, calm conditions
Damp new snow
Very wet snow and firn
Typical densities (kg m-3). From Cuffey and Paterson, 2010.
Firn transforms to glacier ice in 3-5 years in the temperate Upper Seward Glacier in the St Elias Mountains near the Alaska-Yukon border. Firn becomes ice at a depth of about 13 m1. At sites like this with rapid snow accumulation, the depth of a firn layer, and the load on it, increases rapidly with depth.
However, in cold, dry East Antarctica, firn becomes ice at a depth of 64 m at Byrd and 95 m at Vostok. 280 years are needed at Byrd, and 2500 at Vostok. Low temperatures slow the transformation. Temperatures at Vostok, the coldest region of Earth, are 30°C lower than Byrd, which explains the slower increase in density. In addition, slow accumulation gives slow burial, and a small load each year; the increase in density takes much longer.
Typically, the transformation of firn to ice takes 100-300 years, and a depth of 50 – 80 m1.
Firn becomes glacier ice when the interconnecting air or water-filled passageways between the grains are sealed off (“pore closure”)1. Air is isolated in separate bubbles. This occurs at a density of 830 kg m-3. The air space between particles is reduced, bonds form between them, and the particles grow larger. This is a process known as sintering. Increasing pressure compresses the bubbles, placing the enclosed air under pressure and increasing the density of the ice2.
Fresh snowflakes, which have a complex shape, have a large surface area. Over time and under pressure, the surface area is reduced, the surface is smoothed, and the total surface area is reduced. Fresh, complex snowflakes are transformed into rounded particles.
The transformation of firn to ice is much faster where there is melting and refreezing2. Meltwater can percolate downwards, infilling porespaces, and the displaced air escapes upwards. If the snow is under 0°C, the water will freeze, producing areas of compact ice. This will produce high density ice much more rapidly than in colder regions without melting.
The density of pure glacier ice is usually taken as 917 kg m-3. This strictly is only true at 0°C and in the upper layers of ice sheets and mountain glaciers; the density may be greater at the mid-depth ranges in polar ice sheets, where there are no bubbles and temperatures are -20°C to -40°C1.
Below 4 km of ice, such as at the centre of the East Antarctic Ice Sheet, the pressure would increase the density to 921 kg m-3.
Bubbles are common in glacier ice. Bubbles can contain liquid water or atmospheric gases, making them very useful for ice core research. The air in the bubble largely reflects the atmospheric concentrations when the ice formed1. In polar environments, bubbles in the ice occupy about 10% of the volume when firn turns to ice.
With greater depth in polar ice sheets, bubbles shrink as the overlying ice increases. The gas pressure within the bubbles therefore increases, and at certain depths, the gas attains a dissociation pressure. The bubbles begin to disappear as the gas molecules form clathrate hydrates1. This process takes thousands of years.
Glacier ice contains various impurities in tiny amounts. By most scales, glacier ice is a very pure solid-earth material, because the processes leading to snowfall on a glacier – evaporation, condensation, precipitation – act as a natural distillation system1.
However, glaciers can contain impurities. The dirtiest glaciers are mountain glaciers, where debris can fall directly onto the ice surface. On ice sheets and glaciers, dust and other debris may blow onto the ice surface.
Debris on the ice surface can affect the absorption of energy at the ice surface, and lead to increased or decreased melting.
Layers in the ice
Glaciers are composed of sedimentary layers in their accumulation zones, formed of annual layers of snowfall. These layers are initially parallel to the glacier surface. This is the primary stratification in structural glaciology.
In temperate and subpolar settings, the annual sedimentary layers consist of alternating thick layers of bubble-rich ice, which originated as winter snow, and thin layers of clear ice, which are the refrozen meltwater from the summer melt season.
Debris horizons may form, when summer melting concentrates debris (such as rockfall and wind-blown dust) on the ice surface.
In cold polar regions, annual layering forms instead by seasonal variation of snow metamorphism and wind deposition1.
Blue glacier ice
Glacier ice is blue because the longer visible wavelengths are absorbed. The more energetic, blue, wavelengths are scattered back2. The effect is greatest with deep, basal ice, which is bubble free and has large crystals. The blue colour tends therefore to be most intense in the calls of calved icebergs or fresh fractures.
Rough, weathered ice and fresh snow will appear white because preferential absorption does not occur.
The USGS has an excellent resource on the mass balance of Lemon Creek Glacier, a World Reference Glacier, and the other USGS Benchmark Glaciers.
This has resulted in a publication showing a reanalysis of the USGS Benchmark Glaciers (O’Neel et al., 2019). Point data were collected at each glacier over many years. These point datasets allow glacier-wide mass balance to be calculated. The full datasets are rather complicated, probably too much so for post-16 education, but the “Glacier-wide solutions” spreadsheets could be used to calculate glacier mass balance from annual winter and summer balances.
There are annual mass balance reports, and these are presented with some clear graphics showing cumulative glacier mass balance.
WGMS Glacier Browser
The WGMS have produced a browser where you can view the mass balance records of different reference glaciers. The map is based on ArcGIS Online and allows students to explore reference glaciers worldwide.
The numbers in the circules highlight the number of types of measurement and glaciers in each area. As you zoom in, you can click through the individual glaciers and see a graph of mass balance observations, surges, and front variations over time.
This resource allows you to explore observations on glaciers worldwide, and examine the dataset easily to see if glaciers really are receding.
Raw mass balance data
These data are all available in the following publication:
WGMS (2020, updated, and earlier reports). Global Glacier Change Bulletin No. 3 (2016-2017). Zemp, M., Gärtner-Roer, I., Nussbaumer, S. U., Bannwart, J., Rastner, P., Paul, F., and Hoelzle, M. (eds.), ISC(WDS)/IUGG(IACS)/UNEP/UNESCO/WMO, World Glacier Monitoring Service, Zurich, Switzerland, 274 pp., publication based on database version: doi:10.5904/wgms-fog-2019-12.
Case study: Bahia del Diablo, Vega Island, Antarctic Peninsula
An example of a student exercise could be to look at the Mass Balance Point data for a single year for a single glacier from the WGMS dataset, and plot elevation against balance for each point. Students could then plot a graph of elevation against mass balance and get the mass balance gradient through time. These are plotted in the WGMS Bulletin for comparison.
An example could be Glaciar Bahia del Diablo on Vega Island on the Antarctic Peninsula. This is a reference glacier that has been monitored since 2009.
Glaciar del Diablo is on the northern side of the island, and is a land-terminating glacier.
Using the mass balance point data from the WGMS, students could attempt to plot the net balance for each point over certain years.
OGGM Glacier Simulator
You can explore more about glacier mass balance using the OGGM Glacier Simulator.
This is an interactive web application that allows you to learn about how glaciers flow, shrink and grow, and what parameters influence their size.
WGMS 2020. Global Glacier Change Bulletin No. 3 (2016-2017). Zemp, M., Gärtner-Roer, I., Nussbaumer, S. U., Bannwart, J., Rastner, P., Paul, F., and Hoelzle, M. (eds.), ISC(WDS)/IUGG(IACS)/UNEP/UNESCO/WMO, World Glacier Monitoring Service, Zurich, Switzerland, 274 pp., publication based on database version: doi:10.5904/wgms-fog-2019-12. PDF (20 MB)
The Mass balance of a glacier can be thought of as the health of a glacier. Mass balance is the total sum of all the accumulation (snow, ice, freezing rain) and melt or ice loss (from calving icebergs, melting, sublimation) across the entire glacier.
If glaciers have a mass balance that is in equilibrium with climate, then the inputs are equal to the outputs. The glacier remains the same size, and does not grow or shrink. Ice continues to be transferred from the top of the glacier (the accumulation zone) to the bottom of the glacier (ablation zone).
If the amount of melting across the glacier increases, then the glacier will have a negative mass balance, and the glacier will shrink.
If the amount of snow or ice that the glacier receives increases but the amount of melt stays the same, then the glacier will grow. The glacier will have a positive mass balance.
This page draws from the excellent review of glacier melt by Hock (2005) and the equally informative chapter on ‘Snow, Ice and Climate’ (Chapter 2) in Glaciers and Glaciation by Benn and Evans (2010). I highly recommend these resources if you want to dive deeper into the processes of glacier melting and the surface energy balance.
What is the surface
The balance of energy at the glacier surface control melting.
The term surface energy balance is specifically used to describe the balance between all surface energy inputs and outputs over a given interval of time. By calculating the surface energy balance, glaciologists can work out the change in temperature at a glacier surface and rate of glacier melt1-4.
The surface energy balance is the sum of all energy fluxes at the glacier surface, and is represented by the simple equation:
QM = SW + LW + QH + QE +
where QM is the energy available for melting SW is shortwave radiation, LW is longwave radiation, QH is sensible heat transfer, QE is latent heat transfer, QR is the energy supplied by rain.
This page will explain each component of the energy balance equation in order. But first, we need to understand how snow/ice is melted.
The melting of snow and
For snow or ice to melt, its temperature must first be increased from sub-freezing (i.e. temperatures lower than 0°C) to the melting point, which is 0°C. Once at 0°C, further energy surplus will cause melting3.
For this reason, an air temperature of 0°C or warmer does not automatically result in melting. It may heat the glacier surface, but unless it raises the temperature of snow/ice to the melting point, no change in state will arise3.
The opposite transpires when there is an energy deficit at the glacier surface. This deficit either cools the ice further or forms new ice by freezing surface water or through the condensation of water vapour3,4.
The energy needed to change state
The relationship between energy input (in the form of heat) and the change in temperature of a material is known as the specific heat capacity. The specific heat capacity is different for different materials. To warm 1 gram of ice at –10°C by 1°C requires an energy input of 2 joules3,4.
The melting of snow and ice requires an energy input of 334 joules per gram3,4. Conversely, 334 joules are released when water is frozen. The energy released during this change of state (as well as the consumption of energy in melting) is known as latent heat.
The energy required for evaporation, and that released by condensation, is much higher than for melting/freezing, being 2500 joules per gram.
The surface energy
Now that we have the basics of the heating and melting of snow and ice, let’s break down the main components of the surface energy balance1-3.
QM = SW+ LW +
QH + QE + QR
We’ll start with the radiation emitted by the sun, otherwise called ‘shortwave’ radiation as it occurs at short wavelengths (0.2 – 4 micrometres). The shortwave radiation receipt varies from glacier to glacier depending on several factors.
First, is latitude. The sun’s energy is focused over a much smaller area at the equator than at the poles, owing to the higher angle of the sun’s rays in the low latitudes. Therefore, more shortwave radiation is available per unit area for heating the Earth’s surface at the equator than at the poles3,4.
Some solar radiation is scattered by air/water molecules and other particles in the atmosphere as it moves towards Earth’s surface3,4. Because the distance between the sun and the poles is greater than between the sun and the equator, more energy is lost by the time it reaches the high latitudes, leaving less energy available for melting.
Shortwave radiation receipt tends to be highest at low latitude glaciers in high altitude mountain ranges (e.g. the Andes and Himalayas) where the sun angle is high and the thin, relatively cloudless air at high altitude limits the amount of solar energy lost by scattering5-7. In contrast, shortwave radiation receipts are often lower for mid- or high latitude glaciers in persistently cloudy (e.g. coastal) areas, such as Patagonia or Alaska.
Second, the local slope
gradient and aspect impact the angle at which solar rays meet the land surface
and, therefore, the amount of shortwave radiation received2,8. The
terrain around a glacier may also play an important role, by shading the ice
from direct solar radiation8,9.
Time of day
Shortwave radiation receipt also varies as the sun angle changes throughout the day, with a peak at midday when the sun angle is highest.
The albedo effect
Only some of the
shortwave radiation that reaches Earth’s surface is absorbed. A portion is also
reflected. This is known as albedo. The energy available for melting,
therefore, is the difference between incoming and outgoing radiation3,4.
The amount of energy reflected depends on the surface material10. Clean ice or snow reflect more solar radiation than dirty or debris-covered ice, leaving less energy for melting.
QM = SW +LW+ QH +
QE + QR
Longwave radiation, so-called because it occurs at the ‘long’ wavelengths of 4 to 120 micrometres, is emitted from both the land surface and atmosphere. Glaciers also emit longwave radiation, so the net energy available for melting is the difference between that received and that emitted by ice3,4.
The main sources of incoming longwave radiation are the valley walls that surround glaciers and water vapour (as well as CO2 and ozone) in the atmosphere11,12. Heat emitted from warm rockwalls increase the energy available for melting, particularly around the glacier sides. As water vapour absorbs and emits longwave radiation, it is most important where the air is humid, and weather cloudy3,4.
For most glaciers, the most important components of the surface energy are the combined flux of shortwave (solar) and longwave radiation, which may supply over 75% of the energy for melting1,3,12.
QM = SW + LW +QH+
QE + QR
Sensible heat is the thermal energy passed directly from one material to another, in this case, from the atmosphere to a glacier (or vice versa). It is called sensible heat as it we can sense or feel it (a cold breeze, for example).
The amount of sensible heat transferred from the atmosphere to a glacier depends on the temperature gradient near the glacier surface (i.e. the difference in temperature between the ice and air above it) and wind speed3,12.
The higher the
temperature gradient and the faster the wind speed, the more sensible heat is
transferred to a glacier.
There are several examples of such weather conditions. The first are Föhn winds, which are dry, strong winds that blow down the leeside of mountains, warming the air, and glacier surface, as they descend. The second are valley winds, which are warm, low-level winds that are drawn up alpine valleys as the air over mountain ranges heats up during the day.
QM = SW + LW + QH +QE+ QR
Latent heat is the energy consumed or released
during a change of state at the ice surface; condensation (vapour to
liquid), evaporation (liquid to vapour), deposition (vapour to
solid) and sublimation (solid to vapour).
Like sensible heat above,
these processes depend on wind speed over the glacier and the humidity of the
air at and above the ice. For this reason, sensible and latent heat transfers are
often grouped together and termed the turbulent fluxes3,13.
Where the air is above
the glacier is more humid than at the surface, evaporation or sublimation will
occur. In contrast, where the air is above the glacier is less humid (drier)
than at the surface, condensation or deposition will occur. This change in
state, releases energy for warming or melting ice.
Winds are important in
latent heat transfer as they can stir up the air at the ice surface (which is
often humid) and mix it with the air above.
Latent heat fluxes are important at high-altitude mountain glaciers that experience cold and dry weather conditions, where sublimation occurs14. However, they can also be important in the energy balance of maritime glaciers15.
The energy supplied by rain
QM = SW + LW + QH + QE +QR
Rainfall that falls on
the surface of a glacier, although generally a minor component of the overall
energy balance (except for short periods, e.g. during warm fronts or storms16),
can supply energy for melting.
Rain will cool to the temperature of the surface snow or ice and ultimately, freeze. This releases latent heat that contributes to heating or melting ice. If the glacier surface is already at the melting point, then rain will add to melting directly.
How is glacier melt
So, we have covered the
main components of glacier surface energy balance, but how is glacier melt
actually modelled? Well, there are two main types of model for this purpose3.
First, are point models, where the energy balance is estimated at a single point on a glacier surface. Usually, this is at the site of a weather station.
Second, are distributed models, where the energy balance is estimated across an area1,2,3. This method has become more common in recent years as satellite datasets (such as digital elevation models) and computer power have improved.
Distributed models are used to investigate how the individual energy balance components influence ice melting over different parts of a glacier. Detailed information of this nature is needed to better understand how glacier ablation trends will react to climate and weather changes, and to predict how glacier mass balance will change in the future.
Why study glacier energy balance?
To sum up – why should we care about the surface energy balance of glaciers and ice sheets?
Well, mainly because it allows glaciologists to understand current – and, therefore, predict future – trends in glacier melting.
At the global scale, this information can help us estimate the glacier contribution to sea-level rise with climate change.
At the regional scale, this information helps to predict river discharge and geomorphic activity downstream of mountain glaciers, where huge numbers of people rely on glacial freshwater for drinking, the irrigation of crops, and hydroelectric power. It is also important for the forecasting of floods and the safety of downstream communities.
 Arnold, N.S.,
Willis, I.C., Sharp, M.J., Richards, K.S. and Lawson, W.J., 1996. A distributed
surface energy-balance model for a small valley glacier. I. Development and
testing for Haut Glacier d’Arolla, Valais, Switzerland. Journal of
Glaciology, 42, 77-89.
 Hock, R. and
Holmgren, B., 2005. A distributed surface energy-balance model for complex
topography and its application to Storglaciären, Sweden. Journal of Glaciology, 51,
 Hock, R., 2005.
Glacier melt: a review of processes and their modelling. Progress in Physical
Geography, 29, 362-391.
 Benn, D.I., and
Evans, D.J.A., 2010. Glaciers and Glaciation. Hodder-Arnold,
 Benn, D.I.,
Wiseman, S. and Hands, K.A., 2001. Growth and drainage of supraglacial lakes on
debris-mantled Ngozumpa Glacier, Khumbu Himal, Nepal. Journal of
Glaciology, 47, 626-638.
 Mölg, T., Hardy,
D.R. and Kaser, G., 2003. Solar‐radiation‐maintained glacier recession on
Kilimanjaro drawn from combined ice‐radiation
geometry modeling. Journal of Geophysical Research: Atmospheres, 108
 Pellicciotti, F.,
Helbing, J., Rivera, A., Favier, V., Corripio, J., Araos, J., Sicart, J.E. and
Carenzo, M., 2008. A study of the energy balance and melt regime on Juncal
Norte Glacier, semi‐arid
Andes of central Chile, using melt models of different complexity. Hydrological
Processes, 22, 3980-3997.
 Arnold, N.S., Rees,
W.G., Hodson, A.J. and Kohler, J., 2006. Topographic controls on the surface
energy balance of a high Arctic valley glacier. Journal of Geophysical
Research: Earth Surface, 111 (F2).
 Olson, M. and
Rupper, S., 2019. Impacts of topographic shading on direct solar radiation for
valley glaciers in complex topography. The Cryosphere, 13,
 Paterson, W.S.B.,
1994. Physics of glaciers. Butterworth-Heinemann.
 Brock, B.W.,
Willis, I.C., Sharp, M.J. and Arnold, N.S., 2000. Modelling seasonal and
spatial variations in the surface energy balance of Haut Glacier d’Arolla,
Switzerland. Annals of Glaciology, 31, 53-62.
 Oerlemans, J. and
Klok, E.J., 2002. Energy balance of a glacier surface: analysis of automatic weather
station data from the Morteratschgletscher, Switzerland. Arctic,
Antarctic, and Alpine Research, 34, 477-485.
 Morris, E.M.,
1989. Turbulent transfer over snow and ice. Journal of Hydrology, 105,
 Cullen, N.J.,
Mölg, T., Kaser, G., Steffen, K. and Hardy, D.R., 2007. Energy-balance model
validation on the top of Kilimanjaro, Tanzania, using eddy covariance
data. Annals of Glaciology, 46, 227-233.
 Conway, J.P. and
Cullen, N.J., 2013. Constraining turbulent heat flux parameterization over a
temperate maritime glacier in New Zealand. Annals of Glaciology, 54,
 Hay, J.E. and Fitzharris, B.B., 1988. A comparison of the energy-balance and bulk-aerodynamic approaches for estimating glacier melt. Journal of Glaciology, 34, 145-153.
Rock and sediment debris often cover part or all of a glacier’s surface, where it plays an important role in surface energy balance and the rate of glacier ablation.
between debris thickness and glacier melting
Measurements taken at the surface of glaciers show there to be a strong relationship between the thickness of debris and the rate of ice melting1,2.
Where debris covering
the ice surface is thin, the rate of melting rises. The rate of melting
continues to rise until the debris layer reaches around 2 cm thick. Where debris
cover is thicker than 2 cm, the rate of melting falls exponentially1.
This relationship is
explained by both the albedo effect and the insulation effect.
Albedo (usually denoted by the Greek
letter ‘α’) is the
term used for the proportion of incoming shortwave (solar) radiation that is
reflected by a surface3. The albedo of a surface can be determined using
the following simple equation:
α = SWout / SWin
where α is albedo, SWout is outgoing shortwave radiation (i.e. the amount reflected by a surface), and SWin is incoming shortwave radiation (i.e. the amount received by a surface).
Slightly dirty ice
from: Paterson (1994)
The albedo of dirty and debris-covered ice (i.e. the parts of a glacier’s surface littered with bare rock and sediment) is lower than that of clean ice or snow (see table above) and, as a consequence, they absorb more incoming shortwave radiation4. This increases the amount of energy that is available for melting.
The second effect of debris on ice surface melting is insulation. Surface debris forms a barrier between the glacier and the atmosphere, reducing the amount of energy that reaches the ice surface and, therefore, insulating the ice from melting3.
Which effect is most
That depends – as you may have guessed – on the thickness of debris at the glacier surface. The albedo effect has a greater influence on ablation rates where the debris cover is sparse or absent, whereas the insulation effect is more important where the debris cover is thick3.
Calculating the impact of surface debris on ice melt
The influence of surface
debris on ice melt can be assessed by calculating how much heat is transferred vertically
through a debris layer to the top of the glacier. This heat transfer is known
as the conductive heat flux (Qc) and can be estimated by
a simple equation5:
Qc = k (Ts
– Ti) / hd
where k is the thermal conductivity of the debris layer (i.e. the ability of the debris to conduct heat, which varies depending on rock type), Tsand Ti are the temperature at the top and base of the debris layer, and hd is the debris layer thickness.
The above equation is handy
as it provides a simple method of calculating the conductive heat flux, which
can then be used to calculate ice melt rates (using the equation below) beneath
debris cover. However, it makes several assumptions.
Most importantly, the equation assumes that the change in temperature between the top and base of the debris layer is linear (i.e. that the temperature changes at an even rate).
In nature, however, this temperature gradient is rarely stable, but is always changing in response to fluctuations in the receipt of energy at the surface6. In short, it is non-linear. This means that the above equation may not always give a reliable estimate of the passage of heat through a debris layer and, thus, melt rate.
However, as the difference
in temperature between the top and base of a debris layer is linear when
averaged out over the course of a day, it is possible to get around this
problem by using daily mean surface temperature data (which can be accessed from
local weather stations) to calculate ice melt rates6.
Using daily mean surface temperatures, the equation above gives the daily average heat flux through a debris layer, which can be used to calculate ice melt rate (M) using the following simple equation:
M = M/Lf(M
where M is the energy flux available for melting (i.e. the daily average heat flux from above) and Lfis latent heat given off by melting.
Why should we care
about the effect of debris on glacier ablation?
Many of the world’s alpine glaciers are covered by debris to some extent7, and this debris (as explained above) affects the rate of ice melting1,2,5. This, in turn, impacts the overall mass balance of glaciers, as well as the landforms produced at ice margins7.
Understanding the relationships between surface debris and glacier melting is also important for accurately predicting how debris-covered glaciers in regions such the Himalayas, Andes, and Southern Alps of New Zealand, will react to climate change, and whether changes in the patterns of ice melting will threaten communities living downstream (e.g. flooding)8.
 Østrem, G. 1959.
Ice melting under a thin layer of moraine, and the existence of ice cores in
moraine ridges. Geografiska Annaler Series A, 41, 228–230.
 Kayastha, R.B.,
Takeuchi, Y., Nakawo, M. and Ageta, Y., 2000. Practical prediction of ice
melting beneath various thickness of debris cover on Khumbu Glacier, Nepal,
using a positive degree-day factor, IAHS-AISH P, 264, 71-81.
 Benn, D.I., and
Evans, D.J.A., 2010. Glaciers and Glaciation. Hodder-Arnold,
 Paterson, W.S.B.,
1994. Physics of glaciers. Butterworth-Heinemann.
 Nakawo, M. and
Young, G.J., 1981. Field experiments to determine the effect of a debris layer
on ablation of glacier ice. Annals of Glaciology, 2, 85-91.
 Nicholson, L. and
Benn, D.I., 2006. Calculating ice melt beneath a debris layer using
meteorological data. Journal of Glaciology, 52, 463-470.
 Benn, D.I.,
Kirkbride, M.P., Owen, L.A. and Brazier, V., 2003. Glaciated valley
landsystems. In Evans, D.J.A. (ed.) Glacial Landsystems, pp.
 Benn, D.I., Bolch, T., Hands, K., Gulley, J., Luckman, A., Nicholson, L.I., Quincey, D., Thompson, S., Toumi, R. and Wiseman, S., 2012. Response of debris-covered glaciers in the Mount Everest region to recent warming, and implications for outburst flood hazards. Earth-Science Reviews, 114, 156-174.
Most (99.5%) of the permanent ice volume in the world is locked up in ice sheets and glaciers. The Antarctic Ice Sheet is the largest store of frozen freshwater; it would raise sea levels by 57.9 m (its “sea level equivalent”, or SLE) on full melting (BedMachine). The Antarctic Ice Sheet covers 8.3% of the Earth’s land surface.
The Greenland Ice Sheet has a sea level equivalent ice volume of 7.42 m, and covers 1.2% of the global land surface (BedMachine).
Finally, glaciers and ice caps have a sea level equivalent ice volume of 0.32 m, covering just 0.5% of the global land surface (Figure 1). There is a nice illustration of this here.
Global glaciers (in yellow) and ice sheets (white). From IPCC AR5
Figure 1. Global land ice. Glaciers are highlighted in yellow, ice shelves in green, ice sheets in white.
Other sources of global ice
There are also small amounts of ice stored in the ground in permafrost regions, frozen lakes and rivers, seasonal snow cover, and so on.
Sea ice (frozen sea water) and ice shelves (frozen floating extensions of land ice; green on Figure 1 above) do not have a “sea level equivalent” of ice volume as they are already floating, so would not raise sea levels on full melting.
The Greenland Ice Sheet has been losing mass for over 20 years. The most recent estimates suggest that the Greenland Ice Sheet from 2012 to 2016 had a negative mass balance, losing 247 ± 15 Gigatonnes (Gt) per year of ice volume, contributing 0.69 ± 0.04 mm per year to sea level rise. The mass balance of Greenland has been increasingly negative since 1995, and it is now equivalent to the global contribution to sea level rise from glaciers and ice caps (Figure 2).
Figure 2. Cumulative ice mass loss from Greenland ice sheet 1992–2012 (from IPCC AR5).
Driven by changes in surface mass balance
These changes have largely been driven by changes in surface mass balance. While in Greenland 60% of mass loss is through ice discharge across the grounding line to the ocean (as icebergs or melting in the ocean), 40% of mass loss is from surface melt. Increases in surface melt (ablation) are largely responsible for the increasing melting of Greenland .
On June 15, 2016, the Advanced Land Imager (ALI) on NASA’s Earth Observing-1 satellite acquired a natural-color image of an area just inland from the coast of southwestern Greenland (120 kilometers southeast of Ilulisat and 500 kilometers north-northeast of Nuuk). From Wikimedia Commons
Figure 3. Surface meltwater on the Greenland Ice Sheet.
The estimates of Greenland Ice Sheet mass balance above include the peripheral glaciers surrounding the larger ice sheet. These peripheral glaciers account for around 15-20% of the total mass imbalance of the ice sheet[2, 4].
These increases in surface melt and mass losses from Greenland are due to recent increases in winter and summer air temperatures, with increases in the size of the ice sheet ablation area (the area with net melting over one year). This is associated with changes in the surface albedo, as ice has a lower albedo than white snow, exacerbating melt. Overall, this is leading to a lowering of the Greenland Ice Sheet surface elevation (Figure 4), and a decrease in ice volume.
Acceleration in outlet glaciers
Ice discharge from the major outlet glaciers of the Greenland Ice Sheet has also increased, with glaciers accelerating in western Greenland (e.g. Jakobshavn Isbrae, JI) (Figure 4). This faster ice flow leads to these outlet glaciers discharging more ice volume to the ocean as icebergs than is replaced by snow, so the outlet glaciers are also thinning, as can be seen by the red on the figure below.
Figure 4. Average rates of surface elevation change (dh/dt) through time (2010-2017) for the Greenland and Antarctic Ice Sheets.
Antarctic Ice Sheet
Antarctic Ice Sheet ice volume
The best estimates of Antarctic volume come from BEDMAP2 . BEDMAP2 provides us with a detailed map of the base of the ice sheet, derived mostly from radar data. There are three ice sheets in Antarctica, each with their own unique characteristics. They are the larger East Antarctic Ice Sheet (EAIS), with an SLE of 53.3 m, the West Antarctic Ice Sheet (WAIS), with an SLE of 4.3 m, and the Antarctic Peninsula Ice Sheet (APIS) with an SLE of 0.2 m.
Surface elevation of the Greenland and Antarctic ice sheets (IPCC AR5)
Figure 5. BEDMAP2 (Fretwell et al., 2013; IPCC AR5).
Antarctica surface mass balance
It is very cold in Antarctica, with very limited surface melt . There is abundant accumulation in the coastal parts of Antarctica, especially western West Antarctica and on the APIS. The figure below shows where surface mass balance is highest; reds and yellows indicate far more snowfall than is lost through surface melting. It is cold and dry in the centre of the East Antarctic Ice Sheet, with very little snowfall or surface melt.
The average ice-sheet integrated surface mass balance of Antarctica is +2418 ± 181 Gt yr-1 .
Figure 6. Mean (1979–2010) surface mass balance [mm w.e. y−1]. 
Changes in Antarctic mass balance
Most mass loss in Antarctica is driven through ocean melting and iceberg calving[7, 8]. This ice discharge to the ocean through the grounding line is increasing as outlet ice streams are accelerating and grounding lines are retreating (see here). Thus increased ice flow in Antarctica accounts for almost all recent increases in mass losses.
The sea level rise contribution from Antarctica was 0.49 – 0.73 mm yr-1from 2012-2017, mostly from the APIS and WAIS and due to acceleration of outlet glaciers in Amundsen Sea Embayment (e.g. Pine Island Glacier/Thwaites Glacier) (Figure 4; 7).
Ice streams of Antarctica with Pine Island Glacier and Thwaites glacier highlighted.
Figure 7. Location of Pine Island and Thwaites Glacier in Antarctica, with ice velocity from Rignot et al. 2011
Including ice gained and lost through all mechanisms, the current mass balance of Antarctica from 1992 to 2017 was:
EAIS: +5 ± 46 Gt yr-1
WAIS: –94 ± 27 Gt yr-1
APIS: –20 ± 15 Gt yr-1
Total Antarctic Ice Sheet: -109 ± 56 Gt yr-1
Antarctic Ice Sheet mass balance changed from 2012 to 2017 to -219 ± 43 Gt yr-1  . Mass losses from West Antarctica are driving most of the total mass losses from Antarctica, with the mass balance of East Antarctica showing negligible changes .
The amount of ice contained in global glaciers and ice caps is mapped by the Randolph Glacier Inventory[9, 10]. This inventory uses satellite imagery and a formalised methodology to organise researchers working on mapping glaciers and glacier change. The Randolph Glacier inventory estimates that there are 198,000 glaciers worldwide (Figure 9); however, this is an arbitrary number as it depends on:
Subdivision of glaciers and mapping of ice divides
Accuracy of the digital elevation model used
Minimum area threshold; it is hard to map glaciers smaller than 0.2 km2 and so this is usually set as a minimum area threshold. There could be up to 400,000 glaciers if small glacierettes are included (but they only account for 1.4% of glacierised area).
Bamber et al. 2018
Figure 9. Global glaciers (yellow) and their area (pie charts) [2, 10].
The RGI estimates a total glacierised area of: 726,000 km2
Subantarctic and Antarctic: 132,900 km2
Arctic Canada North: 104,900 km2
Asia: 62,606 km2
Low latitudes: 2346km2
44 % is in Arctic regions, 18% in Antarctic & Subantarctic.
Global glacier ice volume
An estimate of global ice volume in glaciers and ice caps remains a “grand challenge” in glaciology; there are few glaciers with direct measurement by radar . Bed topography and thus ice thickness is usually then estimated, either by volume-area scaling [12, 13], inversions of ice surface slope and velocity [14, 15], or from numerical modelling of ice flow .
Our best current estimate of global glacier ice volume is:
170 x 103 ± 21 x 103 km3 (moutain glaciers & ice caps outside Greenland & Antarctica)
Geomorphological evidence of glacier extent (LIA/sig. advances)
Automated and manual mapping from satellite imagery
Limit realistically of mapping glaciers min. 0.2 km2
Mass loss can also be quantified from analysis of glacier surface elevation change (dh/dt)[18, 19] using digital elevation model differencing, satellite gravimetry or altimetry, and in-situ surface mass balance measurements .
The figure below shows the current best estimates of ice volumes lost from Antarctica and Greenland from 2012-2016 (taken from Bamber et al. 2018) and from glaciers around the world. Bamber et al. 2018 do not provide an individual assessment of ice volume lost from each area, so here I have plotted ice volumes lost from 2003-2009 from Gardner et al. 2013. Each region corresponds to those mapped out in Figure 9 and glacier outlines are from GLIMS and the Randalph Glacier Inventory.
Note that peripheral glaciers around Greenland and Antarctica are included in the assessment for the ice sheets (cf. Bamber et al. 2018). These glaciers are however changing rapidly, and indeed account for a large portion of the overall change.
These data, recently compiled by Bamber et al. 2018, give a global estimate of mass loss from glaciers of -227 ± 31 Gt yr-1 (2012-2016). This does not include losses from peripheral glaciers around Greenland and Antarctica, which are included in the ice sheet mass balance assessments.
Figure 11. Global glacier melt (IPCC AR5)
This has led the World Glacier Monitoring Service (WGMS) to state: “rates of early 21st-century mass loss are without precedent on a global scale, at least for the time period observed and probably also for recorded history” .
This global melt is a challenge for society. While the sea level rise from glaciers is ultimately constrained by their small ice volume globally, they remain important as sources of freshwater ; their melting poses new hazards to mountain communities[23-25], and they remain important for local economies .
Mass loss is accelerating (Figure 12), with changes in ocean melt driving recession in Antarctica, increased ice discharge and surface melt driving changes in Greenland, and negative surface mass balances largely driving glacier recession worldwide. Losses from Greenland are now the most significant contributor to global sea level rise (this includes the peripheral glaciers around the ice sheet), recently overtaking glaciers as the largest contributor.
A glacier is a pile of snow and ice. In cold regions (either towards the poles or at high altitudes), more snow falls (accumulates) than melts (ablates) in the summer season. If the snowpack starts to remain over the summer months, it will gradually build up into a glacier over a period of years.
Unnamed Glacier, Ulu Peninsula, James Ross Island. Small valley glacier.
The key input to a glacier is precipitation. This can be “solid precipitation” (snow, hail, freezing rain) and rain1. Further sources of accumulation can include wind-blown snow, avalanching and hoar frost. These inputs together make up the surface accumulation on a glacier.
The Glacier as a System. Inputs are largely from precipitation, and also from wind-blown snow and avalanches. The glacier loses mass (ablates) mainly by the processes of calving and surface and subaqueous melt. After Cogley et al., 2011.
In general, glaciers receive more mass in their upper reaches and lose more mass in their lower reaches. The part of the glacier that receives more mass by accumulation than it loses by ablation is the accumulationzone.
Heavy snowfall over Monte San Valentín (4058 m asl) and in the accumulation zone of the North Patagonian Icefield. Photo: Murray Foubister Wikimedia Commons.
Formation of glacial ice
Over time, the snowfall (by far the most important input to a glacier) is gradually compressed and compacted by the weight of further snowfall on top it. The beautiful pointy edges of the snowflake gradually lose their tips and shape, becoming first granular ice, thenfirn, and finally glacial ice.
Layers of ice on Davies Dome Glacier, James Ross Island, Antarctic Peninsula.
The processes of transformation from snow to ice include partial melting, refreezing and fusing. The rate of transformation varies according to climate (temperature and precipitation regimes). The image below is from an ice core. Note the summer and winter layers in the ice. You can also no longer see the individual crystals that make up the glacier ice at this depth.
This 19 cm long of GISP2 ice core from 1855 m depth shows annual layers in the ice. This section contains 11 annual layers with summer layers (arrowed) sandwiched between darker winter layers. From the US National Oceanic and Atmospheric Administration, Wikimedia Commons.
Glacier ice is a crystalline material, and the crystal size and depth varies with the history of the ice.
As ice flows downhill, it either reaches warmer climates, or it reaches the ocean. This causes various processes of melt, or ablation, to occur. In a land-terminating glacier (a glacier that ends on dry land), the main processes of ablation will be surface melt, because air temperatures generally increase as you lose altitude. This meltwater runs off the glacier and forms a number of rivers that typically drain the glacier.
Meltwater stream on Mendenhall Glacier, Alaska. From: Gillfoto, Wikimedia Commons
This surface meltwater may runoff as surface runoff (as shown above; this is a supraglacial meltwater stream on the surface of the glacier), or it may make its way to the bed of the glacier through cracks in the ice (see the figure below). The water at the glacier bed eventually makes it way to the margin of the glacier, where it exits as a meltwater stream.
Meltwater propagates to the glacier bed through crevasses and moulins
Glaciers that reach the sea or terminate in a lake (Marine-terminating and lacustrine-terminating respectively) additionally will calve icebergs and melt underwater. In large parts of Antarctica, melting underneath the base of floating ice shelves and calving from the margin of the glaciers dominate over surface melt.
Upsala Glacier, from the Southern Patagonian Ice Field, terminates in a large lake. Note the calved icebergs drifting out across the lake. Credit: NASA
The lower part of the glacier generally loses more mass from ablation than it receives from accumulation. This part of the glacier is the ablation zone.
Small tidewater (marine-terminating) glaciers calving into Croft Bay, Antarctic Peninsula
Equilibrium line altitude
Most glaciers receive more inputs and accumulation in their upper reaches, and lose more mass by ablation in their lower reaches. The Equilibrium Line Altitude (ELA) marks the area of the glacier separating the accumulation zone from the ablation zone, and were annual accumulation and ablation are equal2.
Equilibrium line altitudes in a hypothetical glacier
Glaciers as a system
Glacier ice is actually a viscous fluid, which flows and deforms under its own weight. Glaciers can therefore be thought of as systems, which receive snow and ice, flow downslope, and melt. Snow and ice are stored in the glacier until they melt as the glacier reaches lower elevations. This concept is explored in more detail in the Introduction to Glacier Mass Balance page and the pages on Glacier Flow.
In the European Alps and North America, most glaciers receive snowfall throughout the winter, and the main glacier ablation occurs in the summer. The Mass Balance, the balance of accumulation and ablation, is usually therefore positive in the winter and negative in the summer3. These glaciers, which receive more snow in winter and less in summer, are known as Winter Accumulation Type Glaciers. These glaciers form the majority of the world’s glaciers4.
In contrast, in places like the Himalaya, the monsoon brings more precipitation in the summer and less in the relatively cold, dry winter. These glaciers therefore receive more accumulation in the summer, and are known as Summer Accumulation Type Glaciers.
A new paper with a whole host of authors has just been published in Nature (IMBIE Team, 2018). It provides a new estimate of mass balance of the entire Antarctic Ice Sheet over the last 25 years, the longest and most thorough estimate of this to date.
This article argues that the Antarctic Peninsula, the smallest ice sheet in Antarctica, has lost an average of 20 Gigatonnes (Gt) of ice per year over the 25 year study. This increased during the study and especially since the year 2000. The West Antarctic Ice Sheet lost 53±29 Gt yr-1 from 1992-1997, but this accelerated to 159±26 Gt yr-1 from 2012-2017. The East Antarctic Ice Sheet is more stable, with small gains (with large errors) over the study period. Continue reading →
Is Antarctica currently losing or gaining mass? Will this massive ice sheet grow or shrink in the future? And what effect will increased snowfall have over coming centuries? In order to answer these questions, we must analyse the surface mass balance of the Antarctic Ice Sheet.
First, let’s introduce some definitions.
Mass balance is the sum of all processes of accumulation and ablation, including those at the ice surface and at the bed, but does not include mass changes due to ice flow1. See this page (Introduction to Glacier Mass Balance) for more information.
Surface mass balance is the net balance between the processes of accumulation and ablation on a glacier’s surface (it does not include dynamic mass loss and basal melting)1.
Climatic mass balance includes surface mass balance and internal accumulation1.
Ice dynamical changes may include changes to ice discharge and acceleration or deceleration of flow, which can lead to dynamic thinning or thickening, ice-shelf collapse, marine ice sheet instability, and other factors resulting in changes in the glacier beyond surface mass balance.
Surface mass balance
Surface mass balance varies extensively over Antarctica. The Antarctic Peninsula has the highest accumulation rates (up to 1500 mm per year), followed by coastal West Antarctica, which has around 1000 mm accumulation per year2. Compare this with the interior of the Antarctic Ice Sheet, where it is dry and cold; here accumulation can be less than 25 mm per year.
Surface mass balance estimates are constantly improving as scientists gain better understandings of glacio-isostatic adjustment, improve glacier modelling techniques and gain access to higher resolution satellite datasets over longer timescales3. Surface mass balance estimates therefore tend to improve over time, but are subject to large uncertainties4. For this reason, there tends to be differences between the results of different techniques used to measure surface mass balance. Surface mass balance of the grounded Antarctic Ice Sheet is currently estimated at ~2000 gigatonnes per year2, 5, 6, and it is subject to large variability across the ice sheet and through time.
Total mass balance
The figure below shows some recent estimates for total mass balance (including basal processes) over Antarctica7. Each box is bounded by the time interval studied and the uncertainties identified.
Overall, a recent estimate puts Antarctic net mass balance at -71 ± 53 gigatonnes per year8, so just negative over the 19 year survey. Mass losses are increasing in West Antarctica and the Antarctic Peninsula. The mass balance of West Antarctica is dominated by dynamic losses from the Amundsen Sea sector, and dynamic gains from the Kamb Ice Stream8. From the period 2005-2010, Shepherd et al. (2012) estimate the mass balance of the entire Antarctic Ice Sheet to be -81 ± 37 gigatonnes per year8.
An unweighted average of recent estimates suggests that Antarctica moved from a weakly negative mass balance in the 1990s to a faster rate of mass loss at a rate of between -45 and -120 gigatonnes per year7. Larger dynamic losses in West Antarctica are being partially offset by increases in accumulation over East Antarctica.
The total mass balance of Antarctica was recently updated here.
Accelerating total mass losses from Antarctica
The GRACE (Gravity Recovery and Climate Experiment) satellite gravity mission shows that total mass loss in Antarctica is accelerating over time. They found that total mass loss increased by 26 ± 14 gigatonnes per year from 2002 to 20099. Rignot et al. (2011) found a smaller acceleration of 14.5±2 gigatonnes per year from 1993-20115, but this change is still three times larger than that found for mountain glaciers and ice caps.
Surface mass balance of Antarctica in the past
How has the surface mass balance of Antarctica changed in the past? Firn and ice-core records can hold the key to providing a longer perspective on surface mass balance than is currently available from satellite records. Frezzotti et al. used 67 of these cores to reconstruct surface mass balance over the last 800 years. They found that current surface mass balance is not exceptionally high compared with the last 800 years10. Periods of high accumulation occurred in the past, in the 1370s and 1610s AD, but there has been an increase of 10% in snow accumulation in some coastal regions since 1850 – a fact that agrees with independent work on the Antarctic Peninsula11.
Surface mass balance of Antarctica in the future
Climate models predict that, for a generally warmer climate, snowfall will increase over Antarctica7. Surface melt will increase around the more northerly Antarctic Peninsula, and dynamic changes such as increased ice discharge12, ice-shelf collapse and grounding line recession13, and marine ice-sheet instability are likely to offset any increases in precipitation7. However, if no dynamical ice response is assumed, then increases in snowfall over the entire continent of 6-16% to 2100 AD and 8-25% to 2200 AD are likely to result in a drop in sea level of 20-43 mm in 2100 and 73-163 in 2200, compared with today14. However, it is more likely that the Greenland and Antarctic ice sheets will lose mass over the next century, with rapid coastal changes, increases in ice flow and ice-shelf collapse all likely4. As a result of these complex expected changes, there are a number of uncertainties in past, present and future ice sheet mass balance.
2. Lenaerts, J.T.M., van den Broeke, M.R., van de Berg, W.J., van Meijgaard, E., & Kuipers Munneke, P. A new, high-resolution surface mass balance map of Antarctica (1979–2010) based on regional atmospheric climate modeling. Geophysical Research Letters.39, L04501 (2012).
4. Alley, R.B., Spencer, M.K., & Anandakrishnan, S. Ice-sheet mass balance: assessment, attribution and prognosis. Annals of Glaciology.46, 1-7 (2007).
5. Rignot, E., Velicogna, I., Van den Broeke, M., Monaghan, A., & Lenaerts, J. Acceleration of the contribution of the Greenland and Antarctic ice sheets to sea level rise. Geophysical Research Letters.38, (2011).
6. Agosta, C., Favier, V., Krinner, G., Gallée, H., Fettweis, X., & Genthon, C. High-resolution modelling of the Antarctic surface mass balance, application for the twentieth, twenty first and twenty second centuries. Climate Dynamics. 41, 3247-3260 (2013).
8. Shepherd, A., Ivins, E.R., A, G., Barletta, V.R., Bentley, M.J., Bettadpur, S., Briggs, K.H., Bromwich, D.H., Forsberg, R., Galin, N., Horwath, M., Jacobs, S., Joughin, I., King, M.A., Lenaerts, J.T.M., Li, J., Ligtenberg, S.R.M., Luckman, A., Luthcke, S.B., McMillan, M., Meister, R., Milne, G., Mouginot, J., Muir, A., Nicolas, J.P., Paden, J., Payne, A.J., Pritchard, H., Rignot, E., Rott, H., Sørensen, L.S., Scambos, T.A., Scheuchl, B., Schrama, E.J.O., Smith, B., Sundal, A.V., van Angelen, J.H., van de Berg, W.J., van den Broeke, M.R., Vaughan, D.G., Velicogna, I., Wahr, J., Whitehouse, P.L., Wingham, D.J., Yi, D., Young, D., & Zwally, H.J. A Reconciled Estimate of Ice-Sheet Mass Balance. Science.338, 1183-1189 (2012).
9. Velicogna, I. Increasing rates of ice mass loss from the Greenland and Antarctic ice sheets revealed by GRACE. Geophysical Research Letters.36, (2009).
10. Frezzotti, M., Scarchilli, C., Becagli, S., Proposito, M., & Urbini, S. A synthesis of the Antarctic surface mass balance during the last 800 yr. The Cryosphere.7, 303-319 (2013).
11. Thomas, E.R., Marshall, G.J., & McConnell, J.R. A doubling in snow accumulation in the western Antarctic Peninsula since 1850. Geophysical Research Letters.35, L01706 (2008).
12. Winkelmann, R., Levermann, A., Martin, M.A., & Frieler, K. Increased future ice discharge from Antarctica owing to higher snowfall. Nature.492, 239-243 (2012).
13. Barrand, N.E., Hindmarsh, R.C.A., Arthern, R., Williams, C.R., Mouginot, J., Scheuchl, B., Rignot, E., Ligtenberg, S.R.M., van den Broeke, M.R., Edwards, T.L., Cook, A.J., & Simonsen, S.B. Computing the volume response of the Antarctic Peninsula Ice Sheet to warming scenarios to 2200. Journal of Glaciology.59, 397-409 (2013).
14. Ligtenberg, S.R.M., Berg, W.J., Broeke, M.R., Rae, J.G.L., & Meijgaard, E. Future surface mass balance of the Antarctic ice sheet and its influence on sea level change, simulated by a regional atmospheric climate model. Climate Dynamics.41, 867-884 (2013).