Evaluating Climate Models
By Bruce Murray
AnalysisOnline Editor
Knowns and Unknowns
 Global climate change forecasts are based on a pantheon of numerous different climate models – complex scientific and mathematical constructs that factor together the earth’s multiple physical geographic systems to predict future outcomes.
A major source of confusion in the climate change story is the variety of climate models currently in use and the differing results they produce. Presently there are 23 major different Atmosphere-Ocean General Circulation Models from 11 different countries, including the United States, the United Kingdom, Canada, Australia, Germany, Norway, France, Russia, Japan, Korea, and China. (See Climate Change 2007: The Physical Science Basis, published by the Intergovernmental Panel on Climate Change, Chapter 8, “Climate Models and Their Evaluation,” pages 597-599, 646.)
Sorting through the multitude of climate models makes it difficult to provide a single numerical answer to the climate change question, noted Caltech Professor Nathan Lewis, speaking at the National Energy Symposium at USC. (Click here for video of Nate Lewis discussing the viability of current energy technologies.) The climate models all differ in detail in terms of how they compute interactions of the earth’s climate systems. Differing results among the models raises the possibility that one or all of them might very well be wrong, as critics of global warming are quick to point out. (See Lewis on 'The Axes of Energy' - how the world can meet its energy needs and confront environmental challenges.)
“Observational evidence does not support today’s computer climate models, so there is little reason to trust model predictions of the future,” wrote a group of 60 scientists to Canadian Prime Minister Stephen Harper in 2006. “Even if the climate models were realistic, the environmental impact of Canada delaying implementation of the Kyoto Protocol or other greenhouse-gas reduction schemes, pending completion of consultations, would be insignificant.”
The Intergovernmental Panel on Climate Change, with more than 600 experts from 30 countries, asserts the validity of climate models in predicting global climate change. “Confidence in models comes from their physical basis, and their skill in representing observed climate and past climate changes,” according to the IPCC report, Climate Change 2007 (pg. 601). “Models have proven to be extremely important tools for simulating and understanding climate, and there is considerable confidence that they are able to provide credible quantitative estimates of future climate change, particularly at larger scales.
“Nevertheless, models still show significant errors. Although these are generally greater at smaller scales, important large-scale problems also remain,” according to the report.
The report, at 940 pages, provides excruciating detail of climate models’ capabilities and limitations – particularly problems in models’ representations of cloud systems, tropical precipitation and El Niño-Southern Oscillation in the Pacific Ocean.
Consensus/ No Consensus
Climate models are mathematical representations of the climate system based on established physical laws such as the conservation of mass, energy and momentum. These mathematical representations are generated on massive computers.
At the broadest level, there are three different types of climate models: Simple Climate Models, Earth System Models of Intermediate Complexity, and Atmosphere-Ocean General Circulation Models – the latter being most comprehensive type of climate model that is most often cited in climate change predictions. In addition to the 23 Atmosphere-Ocean General Circulation Models, there are also eight major different Earth System Models of Intermediate Complexity, and many more Simple Climate Models. (See Climate Change 2007, pages 597-599, 646.)
After sorting through the variety of climate models, a major difficulty in confronting the climate change story is the sheer complexity of climate modeling. Models take into account the five major components of the earth’s climate system: the atmosphere, the hydrosphere (oceans, rivers and lakes), the cryosphere (polar regions), the land surface and the biosphere (plants and animals) – and all of the interactions between them.
In the context of global climate change, the most important use of climate models is their ability to predict climate sensitivity, or the change in global surface temperature in response to a doubling of carbon dioxide in the atmosphere. (pg. 66.) If world energy consumption continues along projected paths, the carbon dioxide concentration levels could more than double by 2050. (See slide 12 of the Power Point Presentation, “Global Energy Perspective,” by Nathan Lewis.)
The 23 different Atmosphere-Ocean General Circulation Models produce varying results with regard to climate sensitivity. When atmospheric carbon dioxide concentrations are doubled, the models calculate an increase in global temperature ranging from 2.1 to 4.4 degrees Celsius, with a mean value of 3.2 degrees Celsius. (See Climate Change 2007, pgs. 66, 631.)
“Models continue to display a substantial range of global temperature change in response to specified greenhouse gas forcing,” according to the IPCC report (pg. 601). “Despite uncertainties, models are unanimous in their prediction of substantial climate warming under greenhouse gas increases.”
Certainty and Uncertainty
In assessing the validity of climate models, scientists first test a model’s ability to simulate the present climate – in itself no easy task. “For models to predict future climatic conditions reliably, they must simulate the current climatic state with some as yet unknown degree of fidelity,” according to the IPCC report (pg. 608).
After accurately assessing the present climate, climate models must leap into the unknown – predicting the future. Testing future climate cannot be done directly, since there are no observed periods with forcing changes exactly analogous to those expected over the 21st century.
What does the accuracy of a climate model’s simulation of past or contemporary climate say about the accuracy of its projections of climate change? “This question is just beginning to be addressed, exploiting the newly available ensembles of models,” according to the report. (pg. 594)
Climate researchers are making significant advances in the representation of terrestrial processes, oceanic systems and cryospheric components. But the multitude of variables involved in climate modeling presents an ongoing challenge.
“There is currently no consensus on the optimal way to divide computer resources among: finer numerical grids, which allow for better simulations; greater numbers of ensemble members, which allow for better statistical estimates of uncertainty; and the inclusion of a more complete set of processes – for example, carbon feedbacks, atmospheric chemistry interactions,” according to the report. (pg. 592)
Living with Uncertainty
All climate models factor in scientific error, bias and uncertainty – as is part of the scientific method. “The models should always be viewed critically,” according to the IPCC report (pg. 594).
“A specific prediction based on a model can often be demonstrated to be right or wrong,” the report continues. “This is true for both weather prediction and climate prediction. Weather forecasts are produced on a regular basis, and can be quickly tested against what actually happened. Over time, statistics can be accumulated that give information on the performance of a particular model or forecast system. In climate change simulations, on the other hand, models are used to make projections of possible future changes over time scales of many decades and for which there are no precise past analogues. Confidence in a model can be gained through simulations of the historical record.”
The discipline of science always factors in error, bias and simple chance in any study – climate modeling or otherwise. “The perfect study has never been done,” wrote Michael A. Kamrin, Delores J. Katz and Martha L. Walter in the book, Reporting on Risk. “Each study is performed at a distinct time and place with a unique group of subjects by investigators who utilize a particular study design. A diversity of methods are used to execute the study and to analyze its results. Chance always plays a part as well.”
According to the scientific method, “error” is not necessary a bad word, according to Dr. David L. Goodstein, Vice Provost and Professor of Physics and Applied Physics at Caltech. “In common usage, error is undistinguished from the word ‘mistake.’ But errors are intrinsic to all scientific experiments. Error is intrinsic in the interaction between people and nature,” Goodstein said. “The responsibility of scientists is to report how big the errors are. Every paper on science always has a section on errors. Every experiment has an error budget.”
The 23 major different Atmosphere-Ocean General Circulation Models, while dizzying in their complexity and differing results, provide the diversity of methods necessary to draw general conclusions.
“The expanded evaluation of models, encompassing a diversity of perspectives, makes it less likely that significant model errors are being overlooked,” according to the IPCC report. (pg. 591) “In order to identify errors that are systematic across models, the statistical mean is often shown. There is some evidence that the multi-modal mean field is often in better agreement with observations than any of the fields simulated by the individual models, which supports continued reliance on a diversity of modeling approaches in projecting future climate change and provides some further interest in evaluating the multi-model mean results.” (pg. 608)
In summary, when climate models and global warming are viewed in the context of the scientific method and risk assessment, one must learn to live with uncertainty.
“The uncertainty inherent in risk assessment means that risk assessors cannot precisely describe the risk. Rather, they should state the range of probabilities which they found.” Kamrin, Katz and Walter wrote. “In the end, the risk assessor provides an estimate of risk, along with a description of the uncertainties that cause his/her report to be a ‘best guess,’ not an irrefutable statement of fact.”
Posted September 27, 2007
Be the first to comment on Evaluating Climate Models
|
