2012-01-18 01:20:32What are climate models - and why should we trust them?
climatesight
Kate
climatesight@live...
140.193.251.86

Here is a blog post I have been working on, that we planned to publish on SkS. Please comment and review. Thanks! -Kate

This is a climate model:

T = [(1-a)S/(4es)]1/4

An extremely simplified climate model, that is. It's one line long, and is at the heart of every computer model of global warming. Using basic thermodynamics, it calculates the temperature of the Earth based on incoming sunlight and the reflectivity of the surface. The model is zero-dimensional, treating the Earth as a point mass at a fixed time. It doesn't consider the greenhouse effect, ocean currents, nutrient cycles, volcanoes, or pollution.

If you fix these deficiencies, the model becomes more and more complex. You have to use multivariable calculus and parameters inferred from data. You have to repeat the calculations over and over for different parts of the Earth. Eventually the model is too complex to solve using pencil, paper and a pocket calculator. It's necessary to program the equations into a computer, and that's what climate scientists have been doing ever since computers were invented.

A pixellated Earth

Today's most sophisticated climate models are called GCMs, which stands for General Circulation Model or Global Climate Model, depending on who you talk to. On average, they are about 500 000 lines of computer code long, and mainly written in Fortran, a scientific programming language. Despite the huge jump in complexity, GCMs have much in common with the one-line climate model above: they're just a lot of basic physics equations put together.

Computers are great for doing a lot of calculations very quickly, but they have a disadvantage: computers are discrete, while the real world is continuous. To understand the term "discrete", think about a digital photo. It's composed of a finite number of pixels, which you can see if you zoom in far enough. The existence of these indivisible pixels, with clear boundaries between them, makes digital photos discrete. But the real world doesn't work this way. If you look at the subject of your photo with your own eyes, it's not pixellated, no matter how close you get - even if you look at it through a microscope. The real world is continuous (unless you're working at the quantum level!)

Similarly, the surface of the world isn't actually split up into three-dimensional cells (you can think of them as cubes, even though they're usually wedge-shaped) where every climate variable - temperature, pressure, precipitation, clouds - is exactly the same everywhere in that cell. Unfortunately, that's how scientists have to represent the world in climate models, because that's the only way computers work. The same strategy is used for the fourth dimension, time, with discrete "timesteps" in the model, indicating how often calculations are repeated.

It would be fine if the cells could be really tiny - like a high-resolution digital photo that looks continuous even though it's discrete - but doing calculations on cells that small would take so much computer power that the model would run slower than real time. As it is, the cubes are on the order of 100 km wide in most GCMs, and timesteps are on the order of hours to minutes, depending on the calculation. That might seem huge, but it's about as good as you can get on today's supercomputers. Remember that doubling the resolution of the model won't just double the running time - in fact, the running time will increase by a factor of sixteen (one doubling for each dimension).

Cracking the code

A climate model is actually a collection of models - typically an atmosphere model, an ocean model, a land model, and a sea ice model. Some GCMs split up the sub-models (let's call them components) a bit differently, but that's the most common arrangement.

Each component represents a staggering amount of complex, specialized processes. Here are just a few examples from the Community Earth System Model, developed at the National Center for Atmospheric Research in Boulder, Colorado:

  • Atmosphere: sea salt suspended in the air, three-dimensional wind velocity, the wavelengths of incoming sunlight
  • Ocean: phytoplankton, the iron cycle, the movement of tides
  • Land: soil hydrology, forest fires, air conditioning in cities
  • Sea Ice: pollution trapped within the ice, melt ponds, the age of different parts of the ice

Each component is developed independently, and as a result, they are highly encapsulated (bundled separately in the source code). However, the real world is not encapsulated - the land and ocean and air are very interconnected. Some central code is necessary to tie everything together. This piece of code is called the coupler, and it has two main purposes:

  1. Pass data between the components. This can get complicated if the components don't all use the same grid (system of splitting the Earth up into cells).
  2. Control the main loop, or "time stepping loop", which tells the components to perform their calculations in a certain order, once per time step.

Show time

When it's time to run the model, you might expect that scientists initialize the components with data collected from the real world. Actually, it's more convenient to "spin up" the model: start with a dark, stationary Earth, turn the Sun on, start the Earth spinning, and wait until the atmosphere and ocean settle down into equilibrium. The resulting data fits perfectly into the cells, and matches up really nicely with observations. It fits within the bounds of the real climate, and could easily pass for real weather.

Scientists feed input files into the model, which contain the values of certain parameters, particularly agents that can cause climate change. These include the concentration of greenhouse gases, the intensity of sunlight, the amount of deforestation, and volcanoes that should erupt during the simulation. It's also possible to give the model a different map to change the arrangement of continents. Through these input files, it's possible to recreate the climate from just about any period of the Earth's lifespan: the Jurassic Period, the last Ice Age, the present day...and even what the future might look like, depending on what we do (or don't do) about global warming.

The highest resolution GCMs, on the fastest supercomputers, can simulate about 1 year for every day of real time. If you're willing to sacrifice some complexity and go down to a lower resolution, you can speed things up considerably, and simulate millennia of climate change in a reasonable amount of time. For this reason, it's useful to have a hierarchy of climate models with varying degrees of complexity.

As the model runs, every cell outputs the values of different variables (such as atmospheric pressure, ocean salinity, or forest cover) into a file, once per time step. The model can average these variables based on space and time, and calculate changes in the data. When the model is finished running, visualization software converts the rows and columns of numbers into more digestible maps and graphs. For example, this model output shows temperature change over the next century, depending on how many greenhouse gases we emit:

Predicting the past

So how do we know the models are working? Should we trust the predictions they make for the future? It's not reasonable to wait for a hundred years to see if the predictions come true, so scientists have come up with a different test: tell the models to predict the past. For example, give the model the observed conditions of the year 1900, run it forward to 2000, and see if the climate it recreates matches up with observations from the real world.

This 20th-century run is one of many standard tests to verify that a GCM can accurately mimic the real world. It's also common to recreate the last ice age, and compare the output to data from ice cores. To further test a GCM's robustness, scientists will vary the values of parameters (quantities inferred from data rather than derived from first principles), to make sure the output doesn't hinge on one particular approximation.

Climate models aren't perfect, but they are doing remarkably well. They pass the tests of predicting the past, and go even further. For example, scientists don't know what causes El Niño, a phenomenon in the Pacific Ocean that affects weather worldwide. There are some hypotheses on what oceanic conditions can lead to an El Niño event, but nobody knows what the actual trigger is. Consequently, there's no way to program El Niños into a GCM. But they show up anyway - the models spontaneously generate their own El Niños, somehow using the basic principles of fluid dynamics to simulate a phenomenon that remains fundamentally mysterious to us.

In some areas, the models are having trouble. Certain wind currents are notoriously difficult to simulate, and calculating regional climates requires an unaffordably high resolution. Phenomena that scientists can't yet quantify, like the processes by which glaciers melt, or the self-reinforcing cycles of thawing permafrost, are also poorly represented. However, not knowing everything about the climate does't mean scientists know nothing. Incomplete knowledge does not imply nonexistent knowledge - you don't need to understand calculus to be able to say with confidence that 9 x 3 = 27.

Also, history has shown us that when climate models make mistakes, they tend to underestimate the scale of the global warming problem. Take the Arctic sea ice: just a few years ago, GCMs were predicting it would completely melt around 2100. Now, the estimate has been revised to 2030.

Answering the big questions

At the end of the day, GCMs are the best prediction tools we have. If they converge on an outcome, it would be silly to bet against them. However, the big questions, like "Is human activity warming the planet?", don't even require a model. The only things you need to answer those questions are a few fundamental physics and chemistry equations that we've known about for over a century.

You could take climate models right out of the picture, and the answer wouldn't change. Scientists would still be telling us that the Earth is warming, humans are causing it, and the consequences will be severe - unless we take action to stop it.

2012-01-18 12:28:58
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.102.70

Nice job Kate, very interesting.  One thing I often hear from 'skeptics' is that GCMs do a poor job modelling paleoclimate changes.  Any comments on that?

2012-01-18 13:04:22Great post, Kate
John Cook

john@skepticalscience...
124.186.107.58

Nicely written, very understandable and strong conclusion. 

I've resized the image to 570 pixels wide so it doesn't break the SkS design.

One suggestion - what about including one or two of those schematics from your poster of the structure of models to visualise the different model components?

2012-01-18 16:04:55
climatesight
Kate
climatesight@live...
74.216.64.2

Thank you both for your input. I added a few paragraphs to address both of your comments. Also, I put "width=100%" tags inside the images so they will fit nicely inside whatever

they are part of.

 

Here is the new version; publish it whenever you see fit.


This is a climate model:

T = [(1-α)S/(4εσ)]1/4

An extremely simplified climate model, that is. It's one line long, and is at the heart of every computer model of global warming. Using basic thermodynamics, it calculates the temperature of the Earth based on incoming sunlight and the reflectivity of the surface. The model is zero-dimensional, treating the Earth as a point mass at a fixed time. It doesn't consider the greenhouse effect, ocean currents, nutrient cycles, volcanoes, or pollution.

If you fix these deficiencies, the model becomes more and more complex. You have to use multivariable calculus and parameters inferred from data. You have to repeat the calculations over and over for different parts of the Earth. Eventually the model is too complex to solve using pencil, paper and a pocket calculator. It's necessary to program the equations into a computer, and that's what climate scientists have been doing ever since computers were invented.

A pixellated Earth

Today's most sophisticated climate models are called GCMs, which stands for General Circulation Model or Global Climate Model, depending on who you talk to. On average, they are about 500 000 lines of computer code long, and mainly written in Fortran, a scientific programming language. Despite the huge jump in complexity, GCMs have much in common with the one-line climate model above: they're just a lot of basic physics equations put together.

Computers are great for doing a lot of calculations very quickly, but they have a disadvantage: computers are discrete, while the real world is continuous. To understand the term "discrete", think about a digital photo. It's composed of a finite number of pixels, which you can see if you zoom in far enough. The existence of these indivisible pixels, with clear boundaries between them, makes digital photos discrete. But the real world doesn't work this way. If you look at the subject of your photo with your own eyes, it's not pixellated, no matter how close you get - even if you look at it through a microscope. The real world is continuous (unless you're working at the quantum level!)

Similarly, the surface of the world isn't actually split up into three-dimensional cells (you can think of them as cubes, even though they're usually wedge-shaped) where every climate variable - temperature, pressure, precipitation, clouds - is exactly the same everywhere in that cell. Unfortunately, that's how scientists have to represent the world in climate models, because that's the only way computers work. The same strategy is used for the fourth dimension, time, with discrete "timesteps" in the model, indicating how often calculations are repeated.

It would be fine if the cells could be really tiny - like a high-resolution digital photo that looks continuous even though it's discrete - but doing calculations on cells that small would take so much computer power that the model would run slower than real time. As it is, the cubes are on the order of 100 km wide in most GCMs, and timesteps are on the order of hours to minutes, depending on the calculation. That might seem huge, but it's about as good as you can get on today's supercomputers. Remember that doubling the resolution of the model won't just double the running time - in fact, the running time will increase by a factor of sixteen (one doubling for each dimension).

Cracking the code

A climate model is actually a collection of models - typically an atmosphere model, an ocean model, a land model, and a sea ice model. Some GCMs split up the sub-models (let's call them components) a bit differently, but that's the most common arrangement.

Each component represents a staggering amount of complex, specialized processes. Here are just a few examples from the Community Earth System Model, developed at the National Center for Atmospheric Research in Boulder, Colorado:

  • Atmosphere: sea salt suspended in the air, three-dimensional wind velocity, the wavelengths of incoming sunlight
  • Ocean: phytoplankton, the iron cycle, the movement of tides
  • Land: soil hydrology, forest fires, air conditioning in cities
  • Sea Ice: pollution trapped within the ice, melt ponds, the age of different parts of the ice

Each component is developed independently, and as a result, they are highly encapsulated (bundled separately in the source code). However, the real world is not encapsulated - the land and ocean and air are very interconnected. Some central code is necessary to tie everything together. This piece of code is called the coupler, and it has two main purposes:

  1. Pass data between the components. This can get complicated if the components don't all use the same grid (system of splitting the Earth up into cells).
  2. Control the main loop, or "time stepping loop", which tells the components to perform their calculations in a certain order, once per time step.

For example, take a look at the IPSL (Institut Pierre Simon Laplace) climate model architecture. In the diagram below, each bubble represents an encapsulated piece of code, and the number of lines in this code is roughly proportional to the bubble's area. Arrows represent data transfer, and the colour of each arrow shows where the data originated:

We can see that IPSL's major components are atmosphere, land, and ocean (which also contains sea ice). The atmosphere is the most complex model, and land is the least. While both the atmosphere and the ocean use the coupler for data transfer, the land model does not - it's simpler just to connect it directly to the atmosphere, since it uses the same grid, and doesn't have to share data with any other component.

You can see diagrams like this for seven different GCMs, as well as a comparison of their different approaches to software architecture, in this summary of my research.

Show time

When it's time to run the model, you might expect that scientists initialize the components with data collected from the real world. Actually, it's more convenient to "spin up" the model: start with a dark, stationary Earth, turn the Sun on, start the Earth spinning, and wait until the atmosphere and ocean settle down into equilibrium. The resulting data fits perfectly into the cells, and matches up really nicely with observations. It fits within the bounds of the real climate, and could easily pass for real weather.

Scientists feed input files into the model, which contain the values of certain parameters, particularly agents that can cause climate change. These include the concentration of greenhouse gases, the intensity of sunlight, the amount of deforestation, and volcanoes that should erupt during the simulation. It's also possible to give the model a different map to change the arrangement of continents. Through these input files, it's possible to recreate the climate from just about any period of the Earth's lifespan: the Jurassic Period, the last Ice Age, the present day...and even what the future might look like, depending on what we do (or don't do) about global warming.

The highest resolution GCMs, on the fastest supercomputers, can simulate about 1 year for every day of real time. If you're willing to sacrifice some complexity and go down to a lower resolution, you can speed things up considerably, and simulate millennia of climate change in a reasonable amount of time. For this reason, it's useful to have a hierarchy of climate models with varying degrees of complexity.

As the model runs, every cell outputs the values of different variables (such as atmospheric pressure, ocean salinity, or forest cover) into a file, once per time step. The model can average these variables based on space and time, and calculate changes in the data. When the model is finished running, visualization software converts the rows and columns of numbers into more digestible maps and graphs. For example, this model output shows temperature change over the next century, depending on how many greenhouse gases we emit:

Predicting the past

So how do we know the models are working? Should we trust the predictions they make for the future? It's not reasonable to wait for a hundred years to see if the predictions come true, so scientists have come up with a different test: tell the models to predict the past. For example, give the model the observed conditions of the year 1900, run it forward to 2000, and see if the climate it recreates matches up with observations from the real world.

This 20th-century run is one of many standard tests to verify that a GCM can accurately mimic the real world. It's also common to recreate the last ice age, and compare the output to data from ice cores. While GCMs can travel even further back in time - for example, to recreate the climate that dinosaurs experienced - proxy data is so sparse and uncertain that you can't really test these simulations. In fact, much of the scientific knowledge about pre-Ice Age climates actually comes from models!

Climate models aren't perfect, but they are doing remarkably well. They pass the tests of predicting the past, and go even further. For example, scientists don't know what causes El Niño, a phenomenon in the Pacific Ocean that affects weather worldwide. There are some hypotheses on what oceanic conditions can lead to an El Niño event, but nobody knows what the actual trigger is. Consequently, there's no way to program El Niños into a GCM. But they show up anyway - the models spontaneously generate their own El Niños, somehow using the basic principles of fluid dynamics to simulate a phenomenon that remains fundamentally mysterious to us.

In some areas, the models are having trouble. Certain wind currents are notoriously difficult to simulate, and calculating regional climates requires an unaffordably high resolution. Phenomena that scientists can't yet quantify, like the processes by which glaciers melt, or the self-reinforcing cycles of thawing permafrost, are also poorly represented. However, not knowing everything about the climate doesn't mean scientists know nothing. Incomplete knowledge does not imply nonexistent knowledge - you don't need to understand calculus to be able to say with confidence that 9 x 3 = 27.

Also, history has shown us that when climate models make mistakes, they tend to underestimate the scale of the global warming problem. Take the Arctic sea ice: just a few years ago, GCMs were predicting it would completely melt around 2100. Now, the estimate has been revised to 2030.

Answering the big questions

At the end of the day, GCMs are the best prediction tools we have. If they converge on an outcome, it would be silly to bet against them. However, the big questions, like "Is human activity warming the planet?", don't even require a model. The only things you need to answer those questions are a few fundamental physics and chemistry equations that we've known about for over a century.

You could take climate models right out of the picture, and the answer wouldn't change. Scientists would still be telling us that the Earth is warming, humans are causing it, and the consequences will be severe - unless we take action to stop it.

2012-01-18 18:52:55
Rob Painting
Rob
paintingskeri@vodafone.co...
118.93.214.161

This is the best description of climate models we've had here, by quite some margin, but............... 

 

- Not that keen on the title - it creates doubt about the validity of climate models right from the get go.

-"You have to use multivariable calculus and parameters inferred from data" - plain english version? 

- Don't think you've adequately explained why the coupler doesn't share data between the ocean and land.

- Rather than saying the models underestimate global warming, would it be better to say that the climate models are inherently too stable? Paleoclimate data indicate abrupt change has happened in the past, whereas climate models don't seem to do abrupt, and therefore may underestimate some future changes. And would one of those graphs showing the mismatch between earlier model predictions of sea ice & observations be appropriate here? Couldn't hurt to ram home the point.  

2012-01-19 00:36:07
jyyh
Otto Lehikoinen
otanle@hotmail...
85.78.242.13

-"You have to use multivariable calculus and parameters inferred from data" - plain english version? 

maybe:"You have to use complex equations with many variables derived from physical laws, and use measured data to get some of their parameters.' ? Did I get that correctly??

- Don't think you've adequately explained why the coupler doesn't share data between the ocean and land.

yes, the IPSL model possibly has a variable for surface runoff incorporated in the atmospheric component (I don't know how it handles it), maybe it calculates fresh water processes in there. this is probably not the best image for this article maybe better use the other one (was it GISS-E?) which runs all variables through the coupler.

2012-01-19 03:08:32
Albatross
Julian Brimelow
stomatalaperture@gmail...
23.17.186.57

Hi Kate (i still owe you a bio, sorry, ugh),

Anyhow, nicely done.  I agree with Rob about the title-- it sounds a tad defensive and "trust" is such an emotive term.  I'm trying to think of a suggestion.  How about "What are climate models and why are they provide valuable information/guidence?"

But that is longer...  Or "What are climate models and why they are probably too conservative?  Which one you use should be consisytent with what you wish the focus of your post to be.

You might also want to explain the terms of that first equation.  That reminds me...I remember a prof that I was TA'ing for had his undergrad students code up that equation in Fortran :)

I also had a nice quote from a recent paper that states that the models are designed to be stable and that this means they are likely to underestimate dramatic change or miss tipping points.  I wish I knew where that was.  I'll try and track it down.

2012-01-19 05:03:37
Albatross
Julian Brimelow
stomatalaperture@gmail...
23.17.186.57

Hi Kate,

Upon relfection I do not think we should 'trust" the models at all-- some weather forecasters do that and it is referred to as "meteorological cancer", it is a problem nowadays with so much numerical guidance available.  So suggesting that one should  "trust" the models is not advisable.  

What we do is we accept that they are the best guidence available, whilst keeping in mind their limitations, strengths and weaknesses.  

2012-01-19 05:28:23How do climate models work?
climatesight
Kate
climatesight@live...
130.179.131.252

Thank you all very much for your comments! I made some more changes and version 3.0 is below. For the title, how about a snappy "How do climate models work?" I think that's probably better.

I previously had the terms of the equations explained in the title-text, but I moved them to the full text so they are more visible. Good call on the Arctic sea ice graph - the SkS image page had one all ready for me to use :)

Albatross - are you the guy I work with? No idea you were part of SkS!


This is a climate model:

T = [(1-α)S/(4εσ)]1/4

(T is temperature, α is the albedo, S is the incoming solar radiation, ε is the emissivity, and σ is the Stefan-Boltzmann constant)

An extremely simplified climate model, that is. It's one line long, and is at the heart of every computer model of global warming. Using basic thermodynamics, it calculates the temperature of the Earth based on incoming sunlight and the reflectivity of the surface. The model is zero-dimensional, treating the Earth as a point mass at a fixed time. It doesn't consider the greenhouse effect, ocean currents, nutrient cycles, volcanoes, or pollution.

If you fix these deficiencies, the model becomes more and more complex. You have to derive many variables from physical laws, and use empirical data to approximate certain values. You have to repeat the calculations over and over for different parts of the Earth. Eventually the model is too complex to solve using pencil, paper and a pocket calculator. It's necessary to program the equations into a computer, and that's what climate scientists have been doing ever since computers were invented.

A pixellated Earth

Today's most sophisticated climate models are called GCMs, which stands for General Circulation Model or Global Climate Model, depending on who you talk to. On average, they are about 500 000 lines of computer code long, and mainly written in Fortran, a scientific programming language. Despite the huge jump in complexity, GCMs have much in common with the one-line climate model above: they're just a lot of basic physics equations put together.

Computers are great for doing a lot of calculations very quickly, but they have a disadvantage: computers are discrete, while the real world is continuous. To understand the term "discrete", think about a digital photo. It's composed of a finite number of pixels, which you can see if you zoom in far enough. The existence of these indivisible pixels, with clear boundaries between them, makes digital photos discrete. But the real world doesn't work this way. If you look at the subject of your photo with your own eyes, it's not pixellated, no matter how close you get - even if you look at it through a microscope. The real world is continuous (unless you're working at the quantum level!)

Similarly, the surface of the world isn't actually split up into three-dimensional cells (you can think of them as cubes, even though they're usually wedge-shaped) where every climate variable - temperature, pressure, precipitation, clouds - is exactly the same everywhere in that cell. Unfortunately, that's how scientists have to represent the world in climate models, because that's the only way computers work. The same strategy is used for the fourth dimension, time, with discrete "timesteps" in the model, indicating how often calculations are repeated.

It would be fine if the cells could be really tiny - like a high-resolution digital photo that looks continuous even though it's discrete - but doing calculations on cells that small would take so much computer power that the model would run slower than real time. As it is, the cubes are on the order of 100 km wide in most GCMs, and timesteps are on the order of hours to minutes, depending on the calculation. That might seem huge, but it's about as good as you can get on today's supercomputers. Remember that doubling the resolution of the model won't just double the running time - in fact, the running time will increase by a factor of sixteen (one doubling for each dimension).

Cracking the code

A climate model is actually a collection of models - typically an atmosphere model, an ocean model, a land model, and a sea ice model. Some GCMs split up the sub-models (let's call them components) a bit differently, but that's the most common arrangement.

Each component represents a staggering amount of complex, specialized processes. Here are just a few examples from the Community Earth System Model, developed at the National Center for Atmospheric Research in Boulder, Colorado:

  • Atmosphere: sea salt suspended in the air, three-dimensional wind velocity, the wavelengths of incoming sunlight
  • Ocean: phytoplankton, the iron cycle, the movement of tides
  • Land: soil hydrology, forest fires, air conditioning in cities
  • Sea Ice: pollution trapped within the ice, melt ponds, the age of different parts of the ice

Each component is developed independently, and as a result, they are highly encapsulated (bundled separately in the source code). However, the real world is not encapsulated - the land and ocean and air are very interconnected. Some central code is necessary to tie everything together. This piece of code is called the coupler, and it has two main purposes:

  1. Pass data between the components. This can get complicated if the components don't all use the same grid (system of splitting the Earth up into cells).
  2. Control the main loop, or "time stepping loop", which tells the components to perform their calculations in a certain order, once per time step.

For example, take a look at the IPSL (Institut Pierre Simon Laplace) climate model architecture. In the diagram below, each bubble represents an encapsulated piece of code, and the number of lines in this code is roughly proportional to the bubble's area. Arrows represent data transfer, and the colour of each arrow shows where the data originated:

We can see that IPSL's major components are atmosphere, land, and ocean (which also contains sea ice). The atmosphere is the most complex model, and land is the least. While both the atmosphere and the ocean use the coupler for data transfer, the land model does not - it's simpler just to connect it directly to the atmosphere, since it uses the same grid, and doesn't have to share much data with any other component. Land-ocean interactions are limited to surface runoff and coastal erosion, which are passed through the atmosphere in this model.

You can see diagrams like this for seven different GCMs, as well as a comparison of their different approaches to software architecture, in this summary of my research.

Show time

When it's time to run the model, you might expect that scientists initialize the components with data collected from the real world. Actually, it's more convenient to "spin up" the model: start with a dark, stationary Earth, turn the Sun on, start the Earth spinning, and wait until the atmosphere and ocean settle down into equilibrium. The resulting data fits perfectly into the cells, and matches up really nicely with observations. It fits within the bounds of the real climate, and could easily pass for real weather.

Scientists feed input files into the model, which contain the values of certain parameters, particularly agents that can cause climate change. These include the concentration of greenhouse gases, the intensity of sunlight, the amount of deforestation, and volcanoes that should erupt during the simulation. It's also possible to give the model a different map to change the arrangement of continents. Through these input files, it's possible to recreate the climate from just about any period of the Earth's lifespan: the Jurassic Period, the last Ice Age, the present day...and even what the future might look like, depending on what we do (or don't do) about global warming.

The highest resolution GCMs, on the fastest supercomputers, can simulate about 1 year for every day of real time. If you're willing to sacrifice some complexity and go down to a lower resolution, you can speed things up considerably, and simulate millennia of climate change in a reasonable amount of time. For this reason, it's useful to have a hierarchy of climate models with varying degrees of complexity.

As the model runs, every cell outputs the values of different variables (such as atmospheric pressure, ocean salinity, or forest cover) into a file, once per time step. The model can average these variables based on space and time, and calculate changes in the data. When the model is finished running, visualization software converts the rows and columns of numbers into more digestible maps and graphs. For example, this model output shows temperature change over the next century, depending on how many greenhouse gases we emit:

Predicting the past

So how do we know the models are working? Should we trust the predictions they make for the future? It's not reasonable to wait for a hundred years to see if the predictions come true, so scientists have come up with a different test: tell the models to predict the past. For example, give the model the observed conditions of the year 1900, run it forward to 2000, and see if the climate it recreates matches up with observations from the real world.

This 20th-century run is one of many standard tests to verify that a GCM can accurately mimic the real world. It's also common to recreate the last ice age, and compare the output to data from ice cores. While GCMs can travel even further back in time - for example, to recreate the climate that dinosaurs experienced - proxy data is so sparse and uncertain that you can't really test these simulations. In fact, much of the scientific knowledge about pre-Ice Age climates actually comes from models!

Climate models aren't perfect, but they are doing remarkably well. They pass the tests of predicting the past, and go even further. For example, scientists don't know what causes El Niño, a phenomenon in the Pacific Ocean that affects weather worldwide. There are some hypotheses on what oceanic conditions can lead to an El Niño event, but nobody knows what the actual trigger is. Consequently, there's no way to program El Niños into a GCM. But they show up anyway - the models spontaneously generate their own El Niños, somehow using the basic principles of fluid dynamics to simulate a phenomenon that remains fundamentally mysterious to us.

In some areas, the models are having trouble. Certain wind currents are notoriously difficult to simulate, and calculating regional climates requires an unaffordably high resolution. Phenomena that scientists can't yet quantify, like the processes by which glaciers melt, or the self-reinforcing cycles of thawing permafrost, are also poorly represented. However, not knowing everything about the climate doesn't mean scientists know nothing. Incomplete knowledge does not imply nonexistent knowledge - you don't need to understand calculus to be able to say with confidence that 9 x 3 = 27.

Also, history has shown us that when climate models make mistakes, they tend to be too stable, and underestimate the potential for abrupt changes. Take the Arctic sea ice: just a few years ago, GCMs were predicting it would completely melt around 2100. Now, the estimate has been revised to 2030, as the ice melts faster than anyone anticipated:

Answering the big questions

At the end of the day, GCMs are the best prediction tools we have. If they all agree on an outcome, it would be silly to bet against them. However, the big questions, like "Is human activity warming the planet?", don't even require a model. The only things you need to answer those questions are a few fundamental physics and chemistry equations that we've known for over a century.

You could take climate models right out of the picture, and the answer wouldn't change. Scientists would still be telling us that the Earth is warming, humans are causing it, and the consequences will be severe - unless we take action to stop it.

2012-01-19 05:44:37
Albatross
Julian Brimelow
stomatalaperture@gmail...
23.17.186.57

Hi Kate,

Well you work in Winterpeg, I work for our employer remotely.  But yes you have emailed me a couple of times asking for my bio and I gave you a hint about Albatrosses ;)

To be fair, I understand of course that you are well aware of the models' limitations etc and that they are not 'truth" but guidance-- I should have prefaced my comments with that. 

I am still looking for that recent Nature paper about the models being too stable.

2012-01-19 06:09:42
John Hartz
John Hartz
john.hartz@hotmail...
98.122.98.161

Kate,

Given its source, I'm not sure whether the following statement is true. If it is, it would be an interesting factoid to incorporate into your article.

"Interestingly, the first supercomputers were developed in an effort to handle the millions of calculations required even for modest weather forecasts."

2012-01-19 06:16:47
John Hartz
John Hartz
john.hartz@hotmail...
98.122.98.161

Kate,

"Casting a critical eye on climate models" by Anil Ananthaswamy (posted on Jan 17, 2010 on the New Scientists magazine website) is a very informative and timely article. 

http://www.newscientist.com/article/mg20927951.400-casting-a-critical-eye-on-climate-models.html?full=true

Unfortunately, the online version of this article does not contain the graphic, "Model behaviour" that is in the print version. It is one of the best graphics about climate change that I have come across. 

2012-01-19 06:21:30
John Hartz
John Hartz
john.hartz@hotmail...
98.122.98.161

Kate,

Also check out: "What are computer models and can we trust them?" by Robin2. I'm not sure whether this article was posted on SkS for public cosnumption. 

 

2012-01-19 07:16:09
Rob Painting
Rob
paintingskeri@vodafone.co...
118.93.136.46

Looks good Kate. Do you want the first two lines in the last graphic deleted? I find it detracts from the point you are making. I've chopped that bit out, so you can use this instead (if you so wish).

Thumbs up from me.

2012-01-19 07:44:03
Albatross
Julian Brimelow
stomatalaperture@gmail...
23.17.186.57

Hi Rob,

Before I go even crazier, can you please remind me of that recent Nature paper that speaks to the fact that the models are probably too conservative and/or stable (using paleo data?).  I'm pretty sure that it is you who notified me about it, but I cannot for the life of me find it now. Thanks.

2012-01-19 07:59:33
Rob Painting
Rob
paintingskeri@vodafone.co...
118.93.136.46

Maybe you're referring to this commentary piece by Paul Valdes?.

2012-01-19 08:08:37
Albatross
Julian Brimelow
stomatalaperture@gmail...
23.17.186.57

God bless you, yes that is it!  Thanks Rob!

2012-01-19 08:18:27
John Hartz
John Hartz
john.hartz@hotmail...
98.122.98.161

By chance, I came across the following today. It is from "Climate Change: What Would Eisenhower Do?" by Steve Zwick posted on Forbes. Zwick's op-ed is a very condensed review of John Reisman's book, "Exposing the Climate Change Hoax: It's ALL About The Economy."

Exposing the Climate Hoax is an ambitious and intentionally provocative work that addresses everything from the structure of government to the nature of science, the psychology of tribal thinking, and the meaning of the word “conservative”.  It channels everyone from Karl Popper and Thomas Kuhn to Carl Sagan, William Safire and Mr. Wizard.

Early chapters focus on how science works and how models are employed.

“Climate models are not ‘sometimes’ wrong,” he writes.  “Climate models are ‘always’ wrong.”

But those same models, he points out, help us understand reality better and better – and to “build better aircraft, better buildings, better infrastructure.”

2012-01-19 09:02:58version 4.0
climatesight
Kate
climatesight@live...
142.161.233.206

Here is the next iteration of edits. Thanks for cropping that image - I think it works much better. John Hartz - yes, the quote about supercomputers is true. Paul Edwards' book "A Vast Machine" has some fascinating stories about early climate modelling, and the computing demand it created was a major theme.


This is a climate model:

T = [(1-α)S/(4εσ)]1/4

(T is temperature, α is the albedo, S is the incoming solar radiation, ε is the emissivity, and σ is the Stefan-Boltzmann constant)

An extremely simplified climate model, that is. It's one line long, and is at the heart of every computer model of global warming. Using basic thermodynamics, it calculates the temperature of the Earth based on incoming sunlight and the reflectivity of the surface. The model is zero-dimensional, treating the Earth as a point mass at a fixed time. It doesn't consider the greenhouse effect, ocean currents, nutrient cycles, volcanoes, or pollution.

If you fix these deficiencies, the model becomes more and more complex. You have to derive many variables from physical laws, and use empirical data to approximate certain values. You have to repeat the calculations over and over for different parts of the Earth. Eventually the model is too complex to solve using pencil, paper and a pocket calculator. It's necessary to program the equations into a computer, and that's what climate scientists have been doing ever since computers were invented.

A pixellated Earth

Today's most sophisticated climate models are called GCMs, which stands for General Circulation Model or Global Climate Model, depending on who you talk to. On average, they are about 500 000 lines of computer code long, and mainly written in Fortran, a scientific programming language. Despite the huge jump in complexity, GCMs have much in common with the one-line climate model above: they're just a lot of basic physics equations put together.

Computers are great for doing a lot of calculations very quickly, but they have a disadvantage: computers are discrete, while the real world is continuous. To understand the term "discrete", think about a digital photo. It's composed of a finite number of pixels, which you can see if you zoom in far enough. The existence of these indivisible pixels, with clear boundaries between them, makes digital photos discrete. But the real world doesn't work this way. If you look at the subject of your photo with your own eyes, it's not pixellated, no matter how close you get - even if you look at it through a microscope. The real world is continuous (unless you're working at the quantum level!)

Similarly, the surface of the world isn't actually split up into three-dimensional cells (you can think of them as cubes, even though they're usually wedge-shaped) where every climate variable - temperature, pressure, precipitation, clouds - is exactly the same everywhere in that cell. Unfortunately, that's how scientists have to represent the world in climate models, because that's the only way computers work. The same strategy is used for the fourth dimension, time, with discrete "timesteps" in the model, indicating how often calculations are repeated.

It would be fine if the cells could be really tiny - like a high-resolution digital photo that looks continuous even though it's discrete - but doing calculations on cells that small would take so much computer power that the model would run slower than real time. As it is, the cubes are on the order of 100 km wide in most GCMs, and timesteps are on the order of hours to minutes, depending on the calculation. That might seem huge, but it's about as good as you can get on today's supercomputers. Remember that doubling the resolution of the model won't just double the running time - instead, the running time will increase by a factor of sixteen (one doubling for each dimension).

Despite the seemingly enormous computer power available to us today, GCMs have always been limited by it. In fact, early computers were developed, in large part, to facilitate atmospheric models for weather and climate prediction.

Cracking the code

A climate model is actually a collection of models - typically an atmosphere model, an ocean model, a land model, and a sea ice model. Some GCMs split up the sub-models (let's call them components) a bit differently, but that's the most common arrangement.

Each component represents a staggering amount of complex, specialized processes. Here are just a few examples from the Community Earth System Model, developed at the National Center for Atmospheric Research in Boulder, Colorado:

  • Atmosphere: sea salt suspended in the air, three-dimensional wind velocity, the wavelengths of incoming sunlight
  • Ocean: phytoplankton, the iron cycle, the movement of tides
  • Land: soil hydrology, forest fires, air conditioning in cities
  • Sea Ice: pollution trapped within the ice, melt ponds, the age of different parts of the ice

Each component is developed independently, and as a result, they are highly encapsulated (bundled separately in the source code). However, the real world is not encapsulated - the land and ocean and air are very interconnected. Some central code is necessary to tie everything together. This piece of code is called the coupler, and it has two main purposes:

  1. Pass data between the components. This can get complicated if the components don't all use the same grid (system of splitting the Earth up into cells).
  2. Control the main loop, or "time stepping loop", which tells the components to perform their calculations in a certain order, once per time step.

For example, take a look at the IPSL (Institut Pierre Simon Laplace) climate model architecture. In the diagram below, each bubble represents an encapsulated piece of code, and the number of lines in this code is roughly proportional to the bubble's area. Arrows represent data transfer, and the colour of each arrow shows where the data originated:

We can see that IPSL's major components are atmosphere, land, and ocean (which also contains sea ice). The atmosphere is the most complex model, and land is the least. While both the atmosphere and the ocean use the coupler for data transfer, the land model does not - it's simpler just to connect it directly to the atmosphere, since it uses the same grid, and doesn't have to share much data with any other component. Land-ocean interactions are limited to surface runoff and coastal erosion, which are passed through the atmosphere in this model.

You can see diagrams like this for seven different GCMs, as well as a comparison of their different approaches to software architecture, in this summary of my research.

Show time

When it's time to run the model, you might expect that scientists initialize the components with data collected from the real world. Actually, it's more convenient to "spin up" the model: start with a dark, stationary Earth, turn the Sun on, start the Earth spinning, and wait until the atmosphere and ocean settle down into equilibrium. The resulting data fits perfectly into the cells, and matches up really nicely with observations. It fits within the bounds of the real climate, and could easily pass for real weather.

Scientists feed input files into the model, which contain the values of certain parameters, particularly agents that can cause climate change. These include the concentration of greenhouse gases, the intensity of sunlight, the amount of deforestation, and volcanoes that should erupt during the simulation. It's also possible to give the model a different map to change the arrangement of continents. Through these input files, it's possible to recreate the climate from just about any period of the Earth's lifespan: the Jurassic Period, the last Ice Age, the present day...and even what the future might look like, depending on what we do (or don't do) about global warming.

The highest resolution GCMs, on the fastest supercomputers, can simulate about 1 year for every day of real time. If you're willing to sacrifice some complexity and go down to a lower resolution, you can speed things up considerably, and simulate millennia of climate change in a reasonable amount of time. For this reason, it's useful to have a hierarchy of climate models with varying degrees of complexity.

As the model runs, every cell outputs the values of different variables (such as atmospheric pressure, ocean salinity, or forest cover) into a file, once per time step. The model can average these variables based on space and time, and calculate changes in the data. When the model is finished running, visualization software converts the rows and columns of numbers into more digestible maps and graphs. For example, this model output shows temperature change over the next century, depending on how many greenhouse gases we emit:

Predicting the past

So how do we know the models are working? Should we trust the predictions they make for the future? It's not reasonable to wait for a hundred years to see if the predictions come true, so scientists have come up with a different test: tell the models to predict the past. For example, give the model the observed conditions of the year 1900, run it forward to 2000, and see if the climate it recreates matches up with observations from the real world.

This 20th-century run is one of many standard tests to verify that a GCM can accurately mimic the real world. It's also common to recreate the last ice age, and compare the output to data from ice cores. While GCMs can travel even further back in time - for example, to recreate the climate that dinosaurs experienced - proxy data is so sparse and uncertain that you can't really test these simulations. In fact, much of the scientific knowledge about pre-Ice Age climates actually comes from models!

Climate models aren't perfect, but they are doing remarkably well. They pass the tests of predicting the past, and go even further. For example, scientists don't know what causes El Niño, a phenomenon in the Pacific Ocean that affects weather worldwide. There are some hypotheses on what oceanic conditions can lead to an El Niño event, but nobody knows what the actual trigger is. Consequently, there's no way to program El Niños into a GCM. But they show up anyway - the models spontaneously generate their own El Niños, somehow using the basic principles of fluid dynamics to simulate a phenomenon that remains fundamentally mysterious to us.

In some areas, the models are having trouble. Certain wind currents are notoriously difficult to simulate, and calculating regional climates requires an unaffordably high resolution. Phenomena that scientists can't yet quantify, like the processes by which glaciers melt, or the self-reinforcing cycles of thawing permafrost, are also poorly represented. However, not knowing everything about the climate doesn't mean scientists know nothing. Incomplete knowledge does not imply nonexistent knowledge - you don't need to understand calculus to be able to say with confidence that 9 x 3 = 27.

Also, history has shown us that when climate models make mistakes, they tend to be too stable, and underestimate the potential for abrupt changes. Take the Arctic sea ice: just a few years ago, GCMs were predicting it would completely melt around 2100. Now, the estimate has been revised to 2030, as the ice melts faster than anyone anticipated:

Answering the big questions

At the end of the day, GCMs are the best prediction tools we have. If they all agree on an outcome, it would be silly to bet against them. However, the big questions, like "Is human activity warming the planet?", don't even require a model. The only things you need to answer those questions are a few fundamental physics and chemistry equations that we've known for over a century.

You could take climate models right out of the picture, and the answer wouldn't change. Scientists would still be telling us that the Earth is warming, humans are causing it, and the consequences will be severe - unless we take action to stop it.

2012-01-19 14:15:40the thumb
jyyh
Otto Lehikoinen
otanle@hotmail...
85.76.43.238

I think I would have understood this when entering high school (at 15-16 years) or a bit earlier so a thumb.

2012-01-20 10:30:20
Riccardo

riccardoreitano@tiscali...
2.33.129.187

Good job, clear and informative. I think that you should explicitly say somewhere what you're going to do, i.e. climate models for beginners in plain english. It may even be in the title.

Two comments.
"treating the Earth as a point mass". It's not a point mass, it has a finite size and a surface. True that the surface area does not appear anywhere and that you'd get the same results whatever that surface is, but the model is called zero-dimensional because the physical quantities involved (albedo, emissivity and insolation) do not vary in space and time. In other words, the Earth is treated as a uniform mass with unchanging properties and illuminated uniformly.

"Also, history has shown us that when climate models make mistakes, they tend to be too stable, and underestimate the potential for abrupt changes."
Not happy with this sentence. It's not history, climate models are built in this way and we know it. History or, better, past climate or particular aspects of the climate have proved to be relatively unstable at times. Hence climate models "(tend to?) underestimate the potential for abrupt changes."

2012-01-20 16:30:04
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.102.70

FYI, John wants to publish this tomorrow (Friday, North America time).  Kate, give a thumbs-up when you feel it's ready.

2012-01-20 16:33:35
climatesight
Kate
climatesight@live...
74.216.64.2

Yes, John and I were emailing and I think publication tomorrow is a good plan. Kate

2012-01-20 16:37:58Following the thumbs up crowd
John Cook

john@skepticalscience...
124.186.107.58

Everyone else is doing it, felt like joining in :-)

2012-01-21 04:57:55Title?
climatesight
Kate
climatesight@live...
140.193.245.68

Hi John, I noticed that the published post has the old title "What are climate models and why should we trust them?" Can we change this to "How do climate models work?"

Thanks! Looks great.

Kate

2012-01-21 05:15:00
dana1981
Dana Nuccitelli
dana1981@yahoo...
64.129.227.4

Got it, Kate.  Very nice job on the post.

2012-01-22 04:31:16
John Mason

johntherock@btopenworld...
81.157.175.198

This is a superb post  - thanks, Kate! I'll make sure it is broadcast around the UK forums this next few days.

Cheers - John

2012-01-22 21:03:30
jyyh
Otto Lehikoinen
otanle@hotmail...
193.199.48.148

Looks like this article was succesful in attracting trollish behaviour. My guess is 'Jenpalex' is a troll and will spoil the thread if fed enough. DNFTT. Let's see. He seems to think Kate has a model and that Kate talks of this in the article, so he cannot have read the article. Sounds like Moron on a quest.