Scientists Admit They Cannot Predict Drought

The threat of widespread and persistent drought, ruining crops and threatening water supplies, is constantly cited as an outcome of global warming. Media talking heads, climate scientists (who should know better) and even the American President have all made this assertion—and there is nothing to back up the claim. Results presented recently at the annual assembly of the European Geosciences Union in Vienna show that forecasting drought is still beyond the reach of current climate models. Models run against historical data have either predicted periods of drought at the wrong times or missed them all together. Yet climate alarmists continue to spread this pernicious lie, preaching damnation with the certitude of an Old Testament prophet.

As reported in the journal Nature, Columbia University’s Lamont-Doherty Earth Observatory ran tests to determine whether a state-of-the-art climate model could simulate the droughts known to have occurred in the southwest United States during the past thousand years. The model incorporated “realistic” numbers for factors that affect temperature and rainfall, such as atmospheric carbon dioxide levels, changes in solar radiation and ash from volcanic eruptions. It even included changes in the El Niño/Southern Oscillation (ENSO), a recurring temperature anomaly in the tropical Pacific Ocean that is known to affect weather in the western US and other parts of the world.

Sloan Coats and his colleagues then compared the results of its simulations with data from the North American Drought Atlas. The NADA is a detailed history of droughts based on the thickness of tree rings. According to the Nature article:

The results were puzzling. Although the simulation produced a number of pronounced droughts lasting several decades each, these did not match the timing of known megadroughts. In fact, drought occurrences were no more in agreement when the model was fed realistic values for variables that influence rainfall than when it ran control simulations in which the values were unrealistically held constant.

Sloan et al.'s first choice of climate model was not the only failure to perform. Other climate models tested by the team performed no better. Specifically, all the models failed to reproduce a series of multi-decadal droughts that occurred in the southwest during the Medieval Warm Period (900-1200AD), a period when global temperatures were as high or higher than they are today. Note that the MWP is often called the Medieval Climate Anomaly by disingenuous climate science types who just can't bring themselves to associate warm temperatures with anything but the past half century or so.

The Dust Bowl—computer models say this didn't happen.

“The model seems to miss some of the dynamics that drive large droughts,” says study participant Jason Smerdon, a Lamont-Doherty researcher who studies historical climate patterns. Bear in mind that these are documented climate events and the models were being fed the best “realistic” environmental data that the scientists could scrounge up (or make up). A major problem may be the models’ inability to reproduce the cycling between the ENSO’s El Niño and La Niña phases.

The ENSO “behaves much messier in the real world than in climate models”, says Jessica Tierney, a climate scientist at the Woods Hole Oceanographic Institution. “We’re not sure how it has varied in the past, and we don’t know how it might change in response to climate change. This is really one of the big uncertainties we’re facing.” This uncertainty is not limited to climate change in the American southwest. Tierney and colleagues investigated the connection between the El Niño/Southern Oscillation and droughts in the horn of Africa in a recent paper in Nature. Here is a quote from that paper:

The 2010–2011 drought in the Horn of Africa, by some measures the worst drought in 60 years, is a reminder that rainfall in this politically and socioeconomically vulnerable region can fluctuate dramatically. Prevailing La Niña conditions in the tropical Pacific were partly to blame; East African rainfall is teleconnected to the El Niño/Southern Oscillation (ENSO), and the Horn of Africa experiences droughts during La Niña events and pluvials during El Niño events. However, it is debated whether the failure of the ‘long rains’ (the rainy season of March, April and May) in 2011—which exacerbated the drought—is related to decadal variability in the Indo-Pacific region1 or anthropogenic forcing. It is critical for us to understand the character and mechanisms that drive decadal to centennial shifts in East African rainfall if we are to evaluate future regional projections of drought frequency and food security, but the short length of the instrumental record fundamentally limits our ability to understand variability on these timescales using observational data alone.

For data, Tierney et al. turned to proxy measurements from lake sediment, but they admit those data are not accurate. “If constrained by radiocarbon (14C) dating, can have a relatively large (~50–100-yr) temporal uncertainty due to compounded analytical and calibration errors,” they state. “This uncertainty can make the identification of shared trends between different site archives challenging, especially within the time frame of the past millennium.”

So naturally they turned to every climatologist's friend, the computer. They decided to synthesize hydroclimatic proxy records from East Africa using a Monte Carlo empirical orthogonal function (MCEOF) approach “to develop a spatiotemporal view of regional water balance during the past millennium that accounts for time uncertainty.” This is a variant of Monte Carlo simulations used to quantify the uncertainty of model predictions as a result of parameter uncertainty. In other words, if you don't have enough data make some up.

Comparison between East Africa MCEOF1 (blue) and an SST reconstruction from the Makassar strait in the western Pacific warm pool (orange).

In any case, the result of their study is shown in the figure, taken from the article. The right-hand axis is flipped such that cooler conditions plot upwards. Shading on MCEOF1 indicates the 68% (dark) and 95% (light) two-tailed uncertainty bounds. Shading on Makassar SSTs indicates the standard error on the binned proxy SST data. The implication is that the ENSO also affects drought conditions in East Africa, often in a way that climate models cannot replicate.

“Efforts to explain the recent decadal drought in the region and project future hydroclimatic change, whether forced or unforced, must acknowledge the existence of potent low-frequency hydroclimate variability related to the Indian Ocean that is not detectable from the instrumental record alone,” Tierney et al. conclude. “Present climate models predict that East Africa will get wetter as a consequence of increasing concentrations of greenhouse gases, but the region has in fact gotten drier in recent decades.”

Oops! The models got that exactly wrong. Even so, climate alarmists read the output of their models, like the auspices of ancient Rome reading entrails, and confidently predict a return to the Dust Bowl—a seven-year drought that devastated large swathes of US prairie land in the 1930s. In a 2011 Nature news article, titled “Mega-drought threat to US Southwest,” it is claimed that “global warming could tip region towards repeat of Pleistocene events.” Climate models suggest that the region will in future become even drier as atmospheric circulation patterns change and subtropical dry zones expand towards the poles.

The only wrinkle in this scenario is that the drying is predicted by computer climate models, and honest scientists know that the climate models cannot predict drought. “Models are often at odds over the very direction of regional changes,” the first Nature article concludes. “For example, different projections prepared for the Colorado Water Conservation Board disagree on whether mean precipitation in the state will increase or decrease by 2050.”

Why do they persist in making these predictions? Perhaps they just can't help themselves, perhaps they are all pathological liars. Or perhaps they just are not very good scientists. Lazy investigators who find it easier to believe their computer playthings than seek out hard evidence. Spellbound, they have committed the gravest sin in modeling—believing that their models are reality.

Be safe, enjoy the interglacial and stay skeptical.

So naturally they turned to

So naturally they turned to every climatologist's friend, the computer. They decided to synthesize hydroclimatic proxy records from East Africa using a Monte Carlo empirical orthogonal function (MCEOF) approach “to develop a spatiotemporal view of regional water balance during the past millennium that accounts for time uncertainty.

Models not right

In this case the old saying by Wolfgang Pauli comes to mind: "this isn't right, it's not even wrong". Truly, climate models are not even wrong.