Theory, Empiricism, Forests, and Global Warming Models
A popular statement, usually attributed to George Box, is that “all models are wrong, but some are useful.” The usefulness of models fall into two broad classes: theory and prediction. Theoretical models attempt to map known physical, chemical, and biological relationships. Predictive models (sometimes called “black box”) attempt to make accurate predictions.
There is a strong tendency to confuse or combine these utilities, and that is true in any modeling (my specialty is forest growth and yield models). Proponents of theoretical models are often adamant that their models are best (a value judgment) and insist that they be used in predictive situations. Predictive modelers, in contrast, may use crude rules of thumb that are unattractive to theoreticians, but predictive modelers emphasize that their goal is accurate prediction.
Hence Box’s assertion that models are wrong must also be bifurcated. Theoretical models are wrong if the theories behind them are invalid. Predictive models are wrong if they make poor predictions. It is easy (but not useful) to confuse these wrong-itudes.
Predictive models are generally empirical, that is, data-driven. Predictions are validated (or invalidated) by the data on actual outcomes. Theoretical models are validated (or invalidated) by tests of theory, which may or may not be empirical. Experiments (empiricism) are used to test theories, but theoretical models do not rest on predicted outcomes because theoretical models are not predictive by design.
The best weather prediction models are more empirical than theoretical. They look at current conditions (fronts, pressure gradients, jet streams, etc.) as they are cadastrally arrayed across the globe, and compare those to past dates when the same or very similar arrays occurred. Then the weather outcomes of the similar past conformations are examined, and used to predict the immediate future weather. Not much theory to that, more of a data mining of the past; hence the descriptor “empirical.”
For instance, the Fleet Numerical Meteorology and Oceanography Center (or FNMOC), known prior to 1995 as the Fleet Numerical Oceanography Center (FNOC), is a meteorological and oceanographic center located in Monterey, California. A United States Navy facility, it prepares worldwide weather and oceanographic forecasts every six hours, which are made available to the public by the National Oceanic and Atmospheric Administration. Meteorological observations use an EMPIRICAL atmospheric data base which is queried for every weather prediction. Current methodologies have evolved from the global, primitive-equation model (GPEM) which used a staggered, spherical, sigma-coordinate system with real input data interpolated to the sigma surfaces, to the current constant feedback loop system using REAL DATA crunched in state-of-the-art silicon graphics super computers, enabling even higher-resolution meteorological and oceanographic products that are the best weather predictions in the world.
Climate models are much more theoretical because we basically lack empirical data about past climate. Some attempts are made to use proxies such as ice cores and tree rings, sunspot data, Milankovitch cycles, etc. but the data are sparse and time frames vary widely. In general we can predict a decline in temperatures and a return to Ice Age conditions based on fairly good evidence at a long time scales, but when and how that slide will occur is unpredictable at short time scales. When theoretical GHG “forcings” are included in climate models, empiricism is almost completely absent.
So we are in a situation where theoretical climate models are being used to make short-term climate predictions without empirical (data-driven) checks. There has been minimal empirical ground-truthing to evaluate the validity of climate models, including (but certainly not limited to) carbon sequestration and low cloud albedo effects.
Since they were first formed in 1988, the UN’s Intergovernmental Panel on Climate Change (IPCC) has predicted a linear rise in global temperature. The IPCC has issued four climate assessment reports (1990, 1995, 2001, and 2007). Every assessment has predicted a linear (straight line) increase in global temperature. Yet global temperature has not risen in a straight line; indeed, global temps have varied nonlinearly and not risen significantly since 1998. In the first three months of this year, global temperatures dropped to their lowest levels in at least 20 years, though some say 50 years. The empirical data invalidate the predictions of the IPCC models, which are theoretical and not data-driven.
So we are in a situation where theoretical climate models are being improperly used to make short-term predictions. Further, those predictions have been invalidated by real data. The IPCC predictions have generated some fairly Draconian suggested measures that are extremely distasteful, at least to many people. More taxes, less freedom, “sacrifices,” economic disruptions, authoritarianism, rising food prices, etc. are being recommended (imposed) based on the predictions of theoretical models. Political “solutions” to fuzzy predictions from “wrong” and improperly classed models are greatly feared, and I think properly so.
The discourse cannot help but become impolite in this situation. Neither “side” is immune or has demurred from casting imprecations and insults.
How much better it would be if we realized that we cannot predict the climate (in the short term) since climate models are theoretical and not empirical (and have been invalidated by real data), and instead we prepared to be adaptable to whatever happens, while preserving (enhancing) as much freedom, justice, and prosperity as we possibly can.
Some climate alarmists have claimed that increasing forest fires are driven by climate change. Yet forest fires occur from the tropics to boreal forests and always have. No climatic zone lacks forest fires. Empiricists (such as myself) point out that the amount and characteristics of fuels are the critical elements; fuels are what burn in forest fires. Managing fuels forestalls and diminishes forest fires, especially catastrophic fires. Regardless of theory, the practical management of fuels will prevent catastrophe. This is the logic behind “defensible space” and forest restoration. Regardless of climate change, fuels management via restoration forestry can prevent catastrophic forest fires.
Being adaptable means preparing forests to receive fire without whole stand incineration. Prepared forests are resilient to fire, regardless of climatic zone. Resilient forests are valuable for wildlife, watershed, and other resource qualities and characteristics, whereas incinerated forests lack those values.
Empirical logic applied in the real world can save forests. Non-empirical theories offer no such solutions, but instead recommend no action, no tending, and no stewardship, and the inevitable destruction of forests.
We cannot predict climate change with any degree of accuracy. We have only theoretical models for climate, and those are by nature and design poor predictors, as has been demonstrated by real, empirical data. We must not rely on such theoretical models because the hopelessness they project will only destroy forests. Instead we must get real, get empirical, and tend our forests so that they are not destroyed, regardless of what the climate may or may not do.
These truisms about modeling apply to forest carbon sequestration models as well. Those are theoretical models and lack predictive power.
John M. Reynolds stated on the wonderful William M. Briggs Statistician Blog [here]
A theory-based forest carbon sequestration model may predict that a forest will absorb (via net fixation) X amount of atmospheric CO2 each year, but unless those models account for the release of CO2 by forest fires, they are very inaccurate over any lengthy time period (and hence are not very useful).
Dr. Thomas Bonnicksen’s Forest Carbon And Emissions Model [here] is also theory-based and has not been calibrated to actual measured emissions (very little of that kind of data has been collected to use in calibration). However, the FCEM at least attempts to account for the de-sequestration of carbon by forest fires.
To the extent that forest fires are predictable (and the empirical evidence indicates that in the aggregate they certainly are), then forest carbon sequestration models MUST include forest fire emissions to have any utility at all.
To put this in plain language, if forests sequestered carbon indefinitely, then massive amounts of carbon would build up on the forest floor. That doesn’t happen. Instead the fixed carbon is unfixed by eventual fire as well as constant decay. In the most extreme cases, down woody debris can become a few feet thick, but that condition is ephemeral because a fire is sure to occur sooner or later if ample fuels are present.
Short-term predictive models are useful if the model assumptions remain valid (i.e. forest growth and yield models are fairly accurate for 20 to 50 year outlooks, if beetles, disease, and/or fires do not occur). Longer term predictive models and theoretical models that do not account for all possible phenomena are not accurate at all, and not much use because of that.