The all-important thin blue line


Seeing the Earth from space gives us a unique perspective on the eggshell-thin layer of the atmosphere – the warming blanket of breathable gases on which all life on Earth depends. Credit: NASA


What happens in the atmosphere has a direct effect on our everyday lives and on every scale. We have local weather forecasts that keep us informed about short-term atmospheric conditions, all the way up to the global-scale climate projections which tell us about future, long-term changes in the climate system as a whole: the atmosphere operates at all scales: from the microscopic to the global; from seconds to millennia.  It is one of the most important and most complex of the climate model components and is intrinsically linked to each of the other elements of the climate system.


So what is the atmosphere made up of?

In dry air, 78% (by volume) is made up of nitrogen, 21% oxygen and 1% argon. These numbers are rounded up, which hides the much smaller contribution from CO2 (less than 0.04%) and other trace gases such as neon, helium, methane and krypton.

But don’t let their comparatively small volume deceive you: these trace elements – especially CO2 and methane – are some of the biggest players in global climate change, both over geological timescales and over more rapid transitions.


The atmosphere is a rapidly changing reservoir in comparison to the other components. Mixing within a hemisphere occurs over a matter of weeks, between hemispheres in the order of about a year, but the different gases themselves can have vastly different residence times, from weeks to thousands of years.

The layers of the Earth’s atmosphere. Credit: / Panacea.



Lenticular clouds over the mountains of Antarctica.
As the atmosphere moves across the land surface it encounters obstacles. Lenticular clouds like these form as fast-moving, moist air is forced over a topographic barrier such as a mountain chain, creating an atmospheric wave in its wake where the moisture condenses as clouds. Photo: /  Dmytro Pylypenko.

The atmospheric model used in NorESM is the Community Atmosphere Model (CAM) from the Community Earth System Model (CESM), but NorESM has its own aerosol chemistry, physics and cloud modules developed by the Oslo NorESM community.

The horizontal grid is based on a typical bipolar latitude-longitude grid. The vertical coordinate is more complicated and uses changes in pressure with height.

Closest to the ground, the model uses “terrain-following”  sigma coordinates, which as the name suggests, follow the slopes and shapes of the surface. Conforming to the natural terrain eliminates issues that other coordinate systems can have with matching to the land surface topography. It also allows the scientists to increase the resolution near the ground, where they may wish to look more closely at boundary layer processes such as turbulence or daily temperature changes. The upper atmosphere of the model is purely pressure based, forming layers that are smoother and easier for modelling upper atmosphere processes such as radiative transfers – and again, the resolution here can be increased. Between these two is a hybrid layer combining aspects of the terrain-following sigma and pressure coordinates. Since pressure is itself one of the governing physical properties in the atmospheric equations, keeping the coordinates in this normalised pressure format makes it much easier to use in further mathematical calculations for other properties and processes.


Diagram of the vertical structure of the atmosphere in NorESM, using CAM. Near the land surface the atmospheric layers conform to the topography, but as you move higher, this influence is reduced, until you reach the upper atmosphere, where the layers are determined by pressure alone. From the NCAR CAM website.




The Earth’s atmosphere interacts with radiation from the sun in many ways –scattering, absorbing and reflecting off particles and surfaces. The physics of the atmosphere is largely driven by these interactions. There are a lot of processes which occur, but most are summarised in the figure below. This is from the IPCC 5th Assessment Report, 2013, and breaks down the global energy budget into a number of different processes.


Global mean energy budget under present-day climate conditions. Numbers state magnitudes of the individual energy fluxes in W m–2, adjusted within their uncertainty ranges to close the energy budgets. Numbers in parentheses attached to the energy fluxes cover the range of values in line with observational constraints. From IPCC 5th Assessment Report 2013. (TOA = top of atmosphere).


Many of the NorESM modellers are based in Bergen, in western Norway. Here, in the summer, the sun only just dips below the horizon for a little while during the night – which leaves plenty of time for long, colourful sunsets in the fjords. Credit: Elizabeth Farmer.

You’re probably more aware of some of the physical processes going on in the atmosphere than you realise.

It’s actually the scattering of this incoming radiation that gives us the blue sky, the colourful sunsets and the greys and whites of clouds.

Particles in the atmosphere that are smaller than the wavelength of the incoming radiation (so things like gas molecules), cause a type of scattering known as Rayleigh scattering.  This affects shorter wavelengths the most – the wavelengths at the blue end of the spectrum.  Sunlight travels the least distance through the atmosphere when it’s directly overhead, so during the day it’s only the blue end of the spectrum that is affected by this scatter. This results in a diffuse blue colour of the sky. As the sun sinks and the angle of the sun decreases, the sunlight has to travel through more of the atmosphere to reach us. This means that more and more of the shorter wavelengths are scattered away, with the only unscattered light left being towards the longer, redder wavelengths – hence the colourful sunsets.

When the atmosphere is full of particles that are larger and closer to the wavelength of the incoming radiation (like water droplets, for example), a different form of scattering occurs – Mie scattering.  In this case all wavelengths tend to be scattered in a similar manner.  Water droplets in clouds create Mie scattering. Since the incoming radiation is scattered equally across all wavelengths of visible light, we see the clouds as grey or white.


– Radiation budget –

The Sun’s rays strike the Earth at different angles depending on latitude, affecting the amount of energy received in these areas. Image taken from the University of Bergen’s “Causes of Climate Change” MOOC.


The balance of incoming and outgoing radiation is central to the climate system, and it’s in the atmosphere that this budget is determined.


The amount of radiation received by the Earth at the top of the atmosphere depends on the latitude: close to the equator, solar radiation is at its strongest, with the Sun directly overhead, but as you move poleward, the Sun’s rays strike at an angle and the incoming radiation is spread across a larger area. The result: more energy per unit area in the tropics compared to the high latitudes.

To balance the incoming short-wave radiation, the Earth’s atmosphere emits infrared long-wave thermal radiation. The amount of thermal radiation emitted by an object depends on its temperature – so for the atmosphere, the discrepancy between the poles and the equator is again evident – the cold poles and cold cloud tops emit far less than the arid and cloudless areas of the tropics.  This is very clear in the NASA visualisation below.


NASA visualisation of the long-wave radiation emitted from the Earth.The warmest areas are the biggest emitters, showing up in oranges and yellows, while the colder regions, such as the poles and the cloud tops, emit far less and are in blues. From NASA visualisations website.


Video from the University of Bergen’s “Causes of Climate Change” MOOC.

But the radiation emitted to space by the atmosphere does not decrease with latitude at the same rate as the absorption of the incoming solar radiation – so the atmosphere ends up with a net surplus of heat retained in the tropics (up to around 40°), but a net loss of energy to space in the high latitudes (poleward of around 40°).

This latitudinal gradient in the annual mean radiation must be balanced. This is achieved by the poleward transfer of energy within the climate system via atmospheric and ocean heat transport. This transport massively cuts down the latitudinal contrast between the equator and the poles – if this was shut down, we’d have even more extreme temperatures in the two regions.


Video from the University of Bergen’s “Causes of Climate Change” MOOC.

So, which one plays the greatest role in transporting heat from the low latitudes to the high? Well, total heat transport is difficult to measure, but if we look at meteorological measurements and satellite observations we can get a good idea of the atmospheric contribution of heat to this movement, and then back-track the contribution from the ocean.

What we find is that at around 30° the atmosphere and ocean are roughly equal, but poleward of this latitude, it is the atmosphere that dominates. The ocean on the other hand peaks at around 20° and actually contributes very little to the transport of heat into the high latitudes.

– Albedo –

The radiation budget is complicated by the process of albedo. This is a measure of how reflective or bright a surface is, and how much it reflects or absorbs incoming radiation. It is measured as the ratio of reflected radiation to incoming radiation received by the surface, ranging from one, which would be a smooth, white surface that reflects everything, to zero – a black body that absorbs everything.

So, sunlight hitting a light coloured surface will generally be largely reflected (high albedo), whilst darker surfaces tend to absorb more radiation (low albedo).  The current average albedo of the surface of the Earth is about 0.3.


This image shows how much solar radiation is reflected by the Earth’s surface across the entire planet. Areas coloured red show the brightest, most reflective regions; yellows and greens are intermediate values; and blues and violets show relatively dark surfaces. White indicates no data were available, and no albedo data are provided over the oceans. Image produced from the MODIS instrument and downloaded from NASA Visualisations.

Land use obviously plays a huge role in determining a region’s albedo. One of the classic and simplest examples of a positive climate feedback is looking at the effect of changing the balance of ice/snow and water: ice, and particularly snow, have very high albedos in contrast to water which is much darker – so ice and snow reflect and water absorbs. At its most simplistic, if you increase the area of water over the area of ice, you increase the amount of energy being absorbed, creating warming, which helps to melt more ice, which leads to more radiation-absorbing dark water, and so the feedback cycle continues. The reverse is also true – more ice leads to more reflection, which helps to cool the climate, leading to more ice.

And it’s not just at the Earth’s surface – cloud albedo it also important, but this is a complex process that can both help to cool (by reflecting incoming solar radiation) and warm (trapping outgoing thermal radiation) the atmosphere, depending on the processes involved.

Incoming solar radiation and outgoing thermal radiation can go through a series of different albedo-related processes depending on the surfaces they interact with on the way in and out of the atmosphere. These continuous processes are a big contributor to keeping heat within the atmosphere.

The strength of the albedo effect is dependent on the strength of the energy reaching the system – so changing the snow cover in the tropics, or an area with a lot of clear skies, will give a stronger feedback than the equivalent change at the poles, since these regions receive far more incoming radiation.

– Water vapour and lapse rate –

The amount of water vapour in air varies hugely and has a central role in governing the thermodynamics of the atmosphere.

Simply put, warm air holds more moisture. The amount of water that can be held in the atmosphere actually increases exponentially with increasing temperature. Why does this matter? Well, not only does lots of water vapour lead to precipitation, but water vapour is also the most powerful naturally occurring greenhouse gas. This relationship with temperature creates a fundamentally important positive feedback – and it’s estimated that the presence of water vapour in the atmosphere makes the Earth about twice as sensitive to changes in radiative forcing than it would be without.

Condensed water vapour. Colourbox image.
Condensed water vapour. Photo: / aaron007

But water vapour does not work alone. The lapse rate feedback is also part of the equation and actually works to reduce this effect. The lapse rate is the rate at which the temperature decreases with height. An increase in surface temperatures is compensated for at the top of the atmosphere, by more heat being radiated out into space. The efficiency of the greenhouse effect is driven by the temperature difference between the surface and the part of the atmosphere which radiates out to space – so the weaker the difference, the weaker the effect of greenhouse gases. Since the effects of water vapour are felt most strongly in the upper troposphere, an increase in water vapour here leads to warming of the upper atmosphere and a more efficient loss of energy to space, reducing the temperature difference between the surface and the upper atmosphere and thereby the efficiency of the greenhouse effect, and inducing a negative feedback which partly counteracts the effect of the positive feedback from the water vapour.


– Clouds –

Clouds may seem like transient, immaterial and at times even inconvenient things from a human perspective, but they play a much more important role in the Earth’s climate system than you might expect – and not just because they generate our rain supply. Covering almost 70% of the Earth’s surface at any one time, they mediate the energy budget, influencing both the incoming solar radiation and the thermal outgoing radiation, as well as setting up a number of complex feedbacks.

Clouds can act as absorbers of outgoing thermal radiation. This reduces the amount that is emitted out to space, effectively holding more heat within the atmosphere and producing a warming effect. On the other hand, they can also be very good reflectors of incoming solar radiation, limiting the amount that reaches the surface and thereby producing a cooling effect.

The balance of these feedbacks is determined by the types and amount of cloudiness and their optical properties, as well as their distribution and location. Anything that alters these aspects will alter this balance. All in all though, modellers estimate that when you combine the countless complicated relationships, the total cloud feedback is slightly positive.

Clouds cover up to 70% of the Earth’s surface at any given time. NASA image.

Because clouds play such a vital but complex role in the larger climate system, one of the key challenges for climate modellers is to accurately simulate cloud formation. For this reason, NorESM has developed its own modules for clouds, including aerosol physics and chemistry, since these supply the condensation nuclei essential for droplet formation.



The combination of radiative forcings, the process of convection (warm air rises), the spin of the Earth on its axis (the Coriolis effect) and the physical law of the conservation of energy (no energy is lost from the system, it’s just converted into other forms of energy) result in movement. The atmosphere is in constant motion – and at much higher rates than the ocean. And much like the ocean, the atmosphere acts at a range of scales, from the minute to the global.

– Convection –

Convection is a process we’re all familiar with in some form or another – we rely on convection to boil our water and heat our homes.

It’s the process whereby a liquid or gas – in this case, the atmosphere –  is warmed from below and cooled from above. In the atmosphere, a warm parcel of air, just like a warm parcel of ocean, becomes less dense and so rises.  As this parcel rises, cooler air takes its place from below, creating a continuous loop of motion. And this motion doesn’t stop until the heat difference between the warm and cold areas disappears.

Convection is fundamental driver in creating motion in both the atmosphere and the ocean. One obvious example of convection in the atmosphere is cumulonimbus or thunder clouds. Extremely strong convective updrafts suck warm air up to huge heights (over 12,000 m or even higher) – they can get so high that the top reaches the inversion layer at the top of the troposphere, where strong shear winds create the classic flat topped “anvil” shape.

Convection occurs across a huge range of scales, from the warm air rising from your heater, to the global atmospheric circulation cells.

Prairie Storm Clouds ominous weather Saskatchewan Canada
Towering cumulonimbus clouds fan out as the convected air meets the top of the troposphere. Photo:

– Differences in warming: water vs. land –

Water is more transparent to radiation than land, allowing radiation to penetrate deeper into bodies of water. With its higher specific heat, the water warms slowly and its mobility allows circulation to spread this heat across large areas and depths.  The higher latent heat of water encourages evaporation. In contrast the land surface is opaque to radiation so it cannot reach any depth.  The incoming radiation is absorbed at the surface and then rapidly re-emitted as thermal radiation, warming the air sitting directly above. The lower specific and latent heat allows the land to warm and cool much faster and drives less evaporation.


In the continental interiors, the rapid warming and cooling of the land masses creates large thermal extremes, whilst the slower response of the oceans allows the oceans to act like a heat source or sink such that maritime environments tend to experience a more moderate climate.

These differences in the rates and heat capacity of the land and the ocean set up localised convection circulation in the atmosphere, driven by the density difference of the air over the land and ocean.  During the day, the land warms rapidly driving convection, pulling cool moist water across from the sea (daytime sea breeze), whilst at night, the higher heat capacity of the ocean and the rapid cooling of the land surface switches this relationship resulting in a land breeze moving offshore.

Maspalomas Dunes and coast of Atlantic ocean on Gran Canaria, Canary islands, Spain
The very different thermal properties of the ocean and the land help create localised convection cells which often result in alternating onshore and offshore winds during the day versus during the night. Photo:

This contrast in the warming of the oceans versus land can also occur on a seasonal timescale and at a continental scale. In the summer, the landmasses warm faster than the ocean, creating low pressure from the rising air over the land and relatively higher pressure over the ocean. This in turn generates a movement of moist air from the ocean onto the land – creating the Asian, African and Australian monsoons. In winter, the circulation switches direction, as the colder continent becomes dominated by high pressure and the sea dominated by lower – so the wind direction moves offshore.

– Large-scale circulation –

Localised processes combine to create a large-scale, global atmospheric circulation which interacts with the ocean, land and ice components.


A basic schematic of the large-scale patterns of atmospheric circulation. Adapted from material from the University of Bergen’s “Causes of Climate Change” MOOC.


Have a look at the simple schematic above – it shows some of the main features of the global patterns of circulation. But things really start to come to life in the model simulation below. This shows the circulation of atmospheric water vapour across the globe and really highlights the dynamic nature of the atmosphere. We’ll run through a few of the main features below the simulation. White areas are clouds, pink shows evaporation and blue precipitation.



Water transport in the atmosphere Sept-Nov 2014. Data from ECMWF. Animation by Mats Bensen, NORCE and the Bjerknes Centre for Climate Research.


Just like the smaller-scale convection cells, the global-scale atmospheric circulation is driven by the temperature differences between the warm and cold regions.

At the equator, the air warms and begins to rise, reaching up to the troposphere before beginning to move poleward. At the Earth’s surface, this rising air is replaced by an equator-ward movement of cooler air, creating a convection loop. If the Earth wasn’t rotating, this cell would directly connect the equator and the poles, but the motion of the Earth makes such a cell unstable and air begins to sink again at around 30° latitude.

The rising limb at the equator and the descending limb here create the “Hadley cells”. The warm air rising at the surface is packed with moisture and this condenses as the air rises and cools, producing the huge amounts of rainfall typical of the lush green tropics. In contrast, the descending limb is made up of much drier, colder air, leaving the areas at the surface starved of rainfall – it’s no coincidence that most of the Earth’s deserts fall within this region.

As the Earth rotates, the Coriolis force deflects the air that is  flowing towards the equator, shifting it to the right in the northern hemisphere and to the left in the southern, and generating the easterly trade winds of the tropics. These trade winds meet just outside the equatorial belt, in what is known as the Intertropical Convergence Zone or ITCZ. This equatorial flow is very clear in the model simulation.

Green tropics with abundant cloud cover and rainfall, and the cloudless, dry desert areas of the subtropics.

The mid-latitudes are associated with strong westerly winds, which flow wave-like around the planet. These bands of westerlies are often unstable, creating areas of high and low pressure, features which anyone living in these latitudes will be very well aware of – it’s these flowing and shifting pressure systems that dominate the surface weather patterns here. These are the atmospheric equivalent to the ocean eddies and are responsible for much of the movement of heat, moisture and energy.

Pressure and wind direction are closely connected. In the northern hemisphere high pressure areas have a clockwise wind rotation, and low pressures a counter-clockwise wind direction – and the opposite in the southern hemisphere. Winter precipitation in the mid-latitudes is often driven by low pressure cyclones – in the northern hemisphere these tend to track in a common north-easterly direction. In the model simulation this storm track is clearly visible as it’s often associated with regions of intense rainfall as the water vapour in the warm tropical air condenses as it moves poleward – we see this in the streaks of white and blue. You’ll also notice how the UK and Norway often face the brunt of these rainfall events – so we can thank these systems for the 2 meters of rainfall we see in Bergen each year – or more, if the low pressures really home in on us!

A classic low pressure system moving north-east along the North Atlantic storm track and bringing moisture up from the tropics.


In terms of the latitudinal movement of air masses, the mid-latitudes are dominated by the Ferrell cells. Weaker than the Hadley cells, these have an ascending limb on the poleward side and a descending limb on the equator-ward side –the opposite of the Hadley cells. They contribute very little to the total atmospheric heat transport, in comparison with the cyclones and anticyclones.

The transition between the Hadley cells and the Ferrell cells creates energetic high altitude westerly winds, known as tropospheric jet streams. These jet streams meander and shift and can sometimes interact with the lower level westerlies, guiding and shifting them. Whether the jet is flowing longitudinally north or south of our position can have a marked impact on the weather we experience at the surface, with northward meanders bringing warmth far up into the mid-latitudes, and southward meanders bringing cold polar air further south.



There are several chemical cycles and processes occurring in the atmosphere that the earth system models try to capture – including ozone and the cycling of carbon, nitrogen and sulphur. There’s not enough space here to go into each of these, but they are connected in part to one of the most fundamental controls on atmospheric chemistry – aerosols, and we’ll have a quick look at that here.

– Aerosols –

Aerosols are one of the key areas of study in climate modelling – in fact, NorESM modellers have developed new modules just to tackle their complexities.

So what is an aerosol? Aerosols are basically very small particles, either liquid or solid, that are suspended in the air. These different aerosols scatter or absorb radiation (the so-called “direct” effects) and can change the formation of clouds and their optical properties (their “indirect” effects). So, despite their short lifetime in the atmosphere – usually only a few weeks – they can have a significant effect on localised weather and larger-scale climate. For this reason, it is essential for our climate models to be able to properly represent these particles: we need to know about their size and composition, as well as their concentrations.

They can have natural or man-made sources:

Natural sources:e.g.  sea salt, soil dust, forest fire particles, volcanoes, biological processes or organic processes from vegetation.

Man-made sources: e.g. soot particles and sulphate produced from combustion processes – note: many of which also produce CO2.


aerosols-earthclim Projected climate changes with NorESM from pre-industrial 1850 to the period 1980-2010, resulting from CO2 (top) and aerosols (bottom). To left, change in ground temperature (2 m) and to right, change in precipitation. White areas indicate where change is too small to be distinguished from the internal natural variability (at 5 % significance level). Credit: Trond Iversen
Comparing the effects of CO2 and aerosol increases. Projected climate changes with NorESM from pre-industrial 1850 to the period 1980-2010, resulting from CO2 (top) and aerosols (bottom). To left, change in ground temperature (2 m) and to right, change in precipitation. White areas indicate where change is too small to be distinguished from the internal natural variability (at 5 % significance level). Credit: Trond Iversen.


There are two classes of aerosols:

Primary particles: added to the atmosphere directly, e.g.  as sea-salt particles being released as sea spray evaporates, or as dust particles are entrained as winds blow over dry areas. These particles are generally relatively large, usually between 0.1 and 100 microns in size.

Secondary particles: produced in the atmosphere itself through the conversion of precursor gases and chemical reactions. These includes sulphates and some other organic particles. And they’re generally tiny – so tiny that they are really hard to actually measure, seriously limiting our ability to study and understand how they form and grow. We know they grow as gases condense onto them, but this process is so important that NorESM has developed whole new mathematical procedures to represent this as well as possible in the model.


And they have different types of effects:

Direct effects: scattering and absorbing radiation. e.g. pure sulphate particles reflect nearly all the radiation, versus soot which absorbs radiation.

Overview of interactions between aerosols and solar radiation and their impact on climate. The left panels show the instantaneous radiative effects of aerosols, while the right panels show their overall impact after the climate system has responded to their radiative effects. FAQ 7.2 Figure 1 from the IPCC 5th Assessment.

Indirect effects: changing the optical properties of clouds or their role in precipitation.

Overview of aerosol-cloud interactions and their impact on climate. Panels (a) and (b) represent a clean and polluted low-level cloud respectively. FAQ 7.2 Figure 2 from the IPCC 5th Assessment.


Beautiful nature of Kamchatka: eruption Tolbachik Volcano. Russia, Far East, Kamchatka Peninsula.
Volcanoes are a natural source of aerosols – which can have both direct and indirect effects on the climate. Photo: / Alexander Piragis.

Volcanoes are an example of a natural source of aerosols. Explosive eruptions can blast huge amounts of dust particles and aerosols up into the stratosphere, over 20 km above the surface, and once there they can hang around  for many months and have a significant effect on climate.

Although they are often associated with large amounts of soot and dust, which would act to warm the atmosphere and reduce the albedo of areas where it lands, their dominant aerosol product is actually sulphates, which increase the brightness of clouds, helping to reflect sunlight and actually inducing cooling.

Huge eruptions like Tambora in 1815 – when the ejecta reached 50 km up into the stratosphere  – had such a dramtic effect on the climate by reducing the amount of incoming solar radiation, that the year 1816 was known as the year without a summer. Such events are also known as volcanic winters.