Climate Models & Accuracy

A hypothesis is an informal idea that has not been thoroughly tested by the scientific community. Most are discarded. A hypothesis becomes a theory when it can explain and predict observations and it also has been thoroughly tested by the scientific community. Even theories, over time, can be disproved and discarded. If a theory stands the test of time (years and decades) it may be called a law or unifying theory and is the closest approximation to "the truth" possible. One must keep in mind that it is impossible to prove that a theory is true, only that is is untrue. Because one cannot study world climate in a laboratory setting, the best method to test climate change hypotheses is to use climate models. Climate models are extremely sophisticated computer programs that simulate climate.

Climate Modeling 101 from National Academy of Sciences

Are Climate Models Accurate?

There is considerable confidence that climate models provide credible quantitative estimates of future climate change, particularly at continental scales and above. This confidence comes from the foundation of the models in accepted physical principles and from their ability to reproduce observed features of current climate and past climate changes. Confidence in model estimates is higher for some climate variables (e.g., temperature) than for others (e.g., precipitation). Over several decades of development, models have consistently provided a robust and unambiguous picture of significant climate warming in response to increasing greenhouse gases. (IPCC, 2007)

1. Climate models are based upon well-established laws of physics and use a wealth of actual observations.

2. These models are able to simulate the current climate. According to the IPCC (2007): "Models show significant and increasing skill in representing many important mean climate features, such as the large-scale distributions of atmospheric temperature, precipitation, radiation and wind, and of oceanic temperatures, currents and sea ice cover. Models can also simulate essential aspects of many of the patterns of climate variability observed across a range of time scales. Examples include the advance and retreat of the major monsoon systems, the seasonal shifts of temperatures, storm tracks and rain belts, and the hemispheric-scale seesawing of extratropical surface pressures (the Northern and Southern ‘annular modes’). Some climate models, or closely related variants, have also been tested by using them to predict weather and make seasonal forecasts. These models demonstrate skill in such forecasts, showing they can represent important features of the general circulation across shorter time scales, as well as aspects of seasonal and interannual variability. Models’ ability to represent these and other important climate features increases our confidence that they represent the essential physical processes important for the simulation of future climate change."

3. These models are able to simulate past climate. According to the IPCC (2007): "Models have been used to simulate ancient climates, such as the warm mid-Holocene of 6,000 years ago or the last glacial maximum of 21,000 years ago. They can reproduce many features (allowing for uncertainties in reconstructing past climates) such as the magnitude and broad-scale pattern of oceanic cooling during the last ice age. Models can also simulate many observed aspects of climate change over the instrumental record. One example is that the global temperature trend over the past century (shown in Figure 6.2) can be modelled with high skill when both human and natural factors that influence climate are included. Models also reproduce other observed changes, such as the faster increase in nighttime than in daytime temperatures, the larger degree of warming in the Arctic and the small, short-term global cooling (and subsequent recovery) which has followed major volcanic eruptions, such as that of Mt. Pinatubo in 1991. Model global temperature projections made over the last two decades have also been in overall agreement with subsequent observations over that period."
   
   Figure 6.1: Global temperature trend over the past century modeled quite well

   Figure 6.2 (ibid) belows shows how climate model temperature predictions compare to reconstructed temperatures. Thick lines represent model predictions with human and natural forcing (All) and thin lines represent model predictions with just natural forcing (Nat). Models do a good job of simulating past climate using just natural forcing but they can only reproduce the modern temperature record by including human emissions of greenhouse gases. The thick and thin lines begin to diverge around 1850 around the time that the Industrial Revolution ramped up. Futhermore, the models predict that the modern climate should be COOLING due to natural forcing which means that the human forcing dominates climate in the recent record.

   
   Figure 6.2: Simulated temperatures during the last 1,000 yr with and without anthropogenic forcing,

In a January 2010 analysis, Tamino at Open Mind shows that there is very good agreement between model projections and actual observations. Take a few minutes to view his analysis.

According to the IPCC 2007 WGI, Chapter 8 report by Randall, et al. (2007):

1. There is considerable confidence that Atmosphere-Ocean General Circulation Models (AOGCMs) provide credible quantitative estimates of future climate change, particularly at continental and larger scales.
2. Models now being used in applications by major climate modeling groups better simulate seasonally varying patterns of precipitation, mean sea level pressure and surface air temperature than the models relied on by these same groups at the time of the IPCC Third Assessment Repport (TAR).
3. Model global temperature projections made over the last two decades have also been in overall agreement with subsequent observations over that period.
4. Some AOGCMs can now simulate important aspects of the El Niño-Southern Oscillation (ENSO).
5. The ability of AOGCMs to simulate extreme events, especially hot and cold spells, has improved.
6. Atmosphere-Ocean General Circulation Models are able to simulate extreme warm temperatures, cold air outbreaks and frost days reasonably well.
7. Models also reproduce other observed changes, such as the faster increase in nighttime than in daytime temperatures and the larger degree of warming in the Arctic known as polar amplification.
8. Models account for a very large fraction of the global temperature pattern: the correlation coefficient between the simulated and observed spatial patterns of annual mean temperature is typically about 0.98 for individual models. This supports the view that major processes governing surface temperature climatology are represented with a reasonable degree of fidelity by the models.
9. The models, as a group, clearly capture the differences between marine and continental environments and the larger magnitude of the annual cycle found at higher latitudes, but there is a general tendency to underestimate the annual temperature range over eastern Siberia. In general, the largest fractional errors are found over the oceans (e.g., over much of tropical South America and off the east coasts of North America and Asia). These exceptions to the overall good agreement illustrate a general characteristic of current climate models: the largest-scale features of climate are simulated more accurately than regional- and smaller-scale features.
10. Models predict the small, short-term global cooling (and subsequent recovery) which has followed major volcanic eruptions, such as that of Mt. Pinatubo in 1991
11. Simulation of extratropical cyclones has improved. Some models used for projections of tropical cyclone changes can simulate successfully the observed frequency and distribution of tropical cyclones.
12. The models capture the dominant extratropical patterns of variability including the Northern and Southern Annular Modes, the Pacific Decadal Oscillation, the Pacific-North American and Cold Ocean-Warm Land Patterns.
13. With a few exceptions, the models can simulate the observed zonal mean of the annual mean outgoing LW within 10 W/m² (an error of around 5%) The models reproduce the relative minimum in this field near the equator where the relatively high humidity and extensive cloud cover in the tropics raises the effective height (and lowers the effective temperature) at which LW radiation emanates to space.
14. The seasonal cycle of the outgoing LW radiation pattern is also reasonably well simulated by models.
15. The models capture the large-scale zonal mean precipitation differences, suggesting that they can adequately represent these features of atmospheric circulation. Moreover, there is some evidence that models have improved over the last several years in simulating the annual cycle of the precipitation patterns.
16. Models also simulate some of the major regional characteristics of the precipitation field, including the major convergence zones and the maxima over tropical rain forests, although there is a tendency to underestimate rainfall over the Amazon.
17. Confidence has also increased in the ability of GCMs to represent upper-tropospheric humidity and its variations, both free and forced. Together, upper-tropospheric observational and modeling evidence provide strong support for a combined water vapor/lapse rate feedback of around the strength found in GCMs (approximately 1 W/m^{2 o}C^-1, corresponding to around a 50% amplification of global mean warming).

Climate models do have their limitations and modelers are constantly improving their models with newer data as the understanding of climate processes improves with research. According to the IPCC (ibid): "Nevertheless, models still show significant errors. Although these are generally greater at smaller scales, important large-scale problems also remain. For example, deficiencies remain in the simulation of tropical precipitation, the El Niño-Southern Oscillation and the Madden-Julian Oscillation (an observed variation in tropical winds and rainfall with a time scale of 30 to 90 days). The ultimate source of most such errors is that many important small-scale processes cannot be represented explicitly in models, and so must be included in approximate form as they interact with larger-scale features. This is partly due to limitations in computing power, but also results from limitations in scientific understanding or in the availability of detailed observations of some physical processes. Significant uncertainties, in particular, are associated with the representation of clouds, and in the resulting cloud responses to climate change. (A good paper on this subject is Trends in Observed Cloudiness and Earth’s Radiation Budget: What Do We Not Know and What Do We Need to Know? by Norris and Slingo (2009).) Consequently, models continue to display a substantial range of global temperature change in response to specified greenhouse gas forcing. Despite such uncertainties, however, models are unanimous in their prediction of substantial climate warming under greenhouse gas increases, and this warming is of a magnitude consistent with independent estimates derived from other sources, such as from observed climate changes and past climate reconstructions."

Tamino (2009) has an excellent blog post that shows how even much simpler mathematical models show that the trend in global temperatures cannot occur without greenhouse gas forcing. Including greenhouse gas forcing results in a very good fit.

Figure 6.3: Global temperature trend over the past century without greenhouse gas forcing = poor fit (ibid)

Figure 6.4: Global temperature trend over the past century with greenhouse gas forcing = very good fit (ibid)

Three more sites with an excellent discussion include: SkepticalScience: How reliable are climate models?, Open Mind: Models, and Realclimate: On Attribution & Updates to model-data comparisons.

Peter Sinclair's Climate Crock of the Week: This Year's Model Climate science is not completely dependent on climate models. There are many threads of supporting evidence. Still, it is clear that climate models are telling us something important that we cannot afford to ignore.

Predicting Climate is Easier than Predicting Weather

Predicting weather is essentially trying to predict individual events within a system. Predicting climate is essentially trying to predict the general trend or the statistical probability of changes in the entire system. Just as it is easy to predict the current (flow of all electrons) through an electronic circuit (climate), it is essentially impossible to predict the path of a single electron through this circuit (weather).

Another example: mathematicians cannot say with much certainty which side will land on a single coin flip. That is like predicting weather. However, a mathematician can say with great certainty how many heads and tails will be flipped after one million flips. That is like climate prediction. It is not 100% but it much closer to 100% than 50%. Of course, this is a gross oversimplification because weather and climate are more complicated than a choice between heads or tails on each iteration. However, the valid point being made is that long term trends are easier to predict than individual events.

A slightly more technical explanation is offered by the Atmospheric Science Assessment and Integration Section Science and Technology Branch Environment Canada (2008) below:

Climate is average weather, which is more predictable than day-to-day and hour-to-hour weather changes. Weather behaviour is chaotic and often difficult to predict beyond a week or so into the future. By comparison, climate is largely determined by global and regional geophysical processes that change slowly. Hence, if these factors are properly understood and predictable, then the climate can be forecast far into the future with a significant degree of confidence.

Day-to-day local weather is largely determined by atmospheric circulation, the formation of large-scale weather systems, and local convective processes. Because of the chaotic nature of the atmosphere, predictability decreases with time, and is quite poor beyond a week or so. Climate, on the other hand, represents average weather and its expected variability. These are determined by factors such as incoming solar radiation (which varies with latitude and time of year), the influence of prevailing characteristics of cloud cover, aerosols and other components of the atmosphere on the flow of the Sun’s energy into the atmosphere and of heat energy out again, prevailing winds and other atmospheric conditions, and local geophysical conditions that, in general, change slowly and in a more predictable manner. Thus, while forecasters would be unable to predict day-to-day weather 6 months into the future, they can provide good approximations of the changes in seasonal climates because of known physical processes that cause conditions to change from winter to summer and back again. They can also provide estimates of the changes in probability of different kinds of weather events, such as sub-zero minimum temperatures, maximum temperatures in excess of 30°C, snow blizzards or thunderstorms. Likewise, climate models, when looking much farther into the future, project how the climate characteristics, averaged over several decades, might change in response to projected changes in the factors that determine the climate.

A person who claims "how can we predict climate change over decades when we can't even predict tomorrow's weather?" has a fundamental misunderstanding of modeling.

Help Climate Modelers Using Your Computer

Climateprediction.net is a distributed computing project to produce predictions of the Earth's climate up to 2080 and to test the accuracy of climate models. To do this, they need people around the world to give them time on their computers - time when they have their computers switched on, but are not using them to their full capacity.

The model will run automatically as your screensaver. As the model runs, you can watch the weather patterns of the world evolve. The results are sent back to them via the internet, and you will be able to see a summary of your results on their web site. Climateprediction.net uses the same underlying software, BOINC, as many other distributed computing projects and, if you like, you can participate in more than one project at a time.

Participating in the project is very easy. All you need to do is download and install BOINC, then attach to the Climateprediction.net project — you can select the project from a list of all available projects when you run the client.

Download BOINC from the BOINC home page. A sample of the screen you will see when running this screensaver appears to the right and you can click for a larger image.

Next: Modern Day Climate Change

• Introduction
• The Scientific Consensus
• Determining the Climate Record
• Impact of Greenhouse Gases
• Climate Change: The Smoking Gun for Humans
• Natural Causes of Climate Change
• Global Cooling?
• Modern Day Climate Change
• Impacts of Climate Change
• The Global Warming Denial Machine
• Global Warming Misinformation
• Reasons for the Misinformation
• How to Talk to a Conservative about Climate Change
• Summary of Key Points
• Suggested Reading
• Works Cited

Last updated: 09/11/12