Larry Vardiman’s Weather Report

Larry Vardiman retired from the ICR back in June, but has continued some of his writing work. Our final January Acts & Facts article of interest – A Mountain of Snow after the Genesis Flood – is largely cribbed from one of his (and Wesley Brewer’s) more recent Answers Research Journal papers, “Numerical Simulations of Three Nor’easters with a Warm Atlantic Ocean.” You can find a brief analysis of the paper at RationalWiki, but let’s take a look for ourselves.

The Acts & Facts article is basically an edited down version of the ARJ paper, with many of the details of the actual simulations cut out for space. I’ll quote from both sources, and differentiate between them via the quote backgrounds. Both the paper and the article begin by talking about snowstorms and nor’easters:

A nor’easter is a type of synoptic scale storm that occurs along the east coast of the United States and the Atlantic coast of Canada. It is so named because the storm travels up the coast, and the winds spiral around the storm from the northeast in coastal areas. The storms sometimes have characteristics similar to a hurricane. They feature a low-pressure area with the center of the rotation just off the east coast and with leading winds in the left front quadrant rotating onto land.

You get the idea – the nature of the storm itself is largely tangential to the issue. More important are the premises that the research is based on. Obviously there’s the biblical flood, but there are also a few more:

Oard (1990) and Austin et al. (1994) have argued that heat released from catastrophic processes of the Genesis Flood would have heated the oceans. Ocean drilling of sediments has found that the oceans were at least 20°C (36°F) hotter in the past than today (Kennett et al. 1975; Shackleton and Kennett 1975). Also, Heezen and Tharp (1977) published images of the ocean floor obtained in the 1960s which show mid-ocean ridges and undersea volcanoes that released large quantities of heat during past earth upheavals. Higher sea-surface temperatures would have evaporated larger quantities of water vapor from the oceans, energized mid-latitude storms and hurricanes, and precipitated large quantities of rain and snow on the continents. Previous numerical simulation studies by Vardiman and Brewer (2010a, 2010b, 2010c, 2011, 2012) have verified many of these predictions.

(See the ARJ page to track down the citations.) In translation, he believes that the flood would have increased the temperature of the oceans. Vardiman reckons that this would change weather patterns such that they would dump enough snow onto North America to produce an ice sheet and explain the ice age. Beyond the creationist speculation he actually offers some evidence to back up the 20°C claim, in the form of the 1975 ocean drilling papers. The primary problem with him doing so, however, is that the time when the high temperatures are found is much earlier than the ice age. Nevertheless, Vardiman is taking it as read that approximately 50 million years of time can be compressed down to fit in order for these temperatures to have caused the later ice age.

Indeed, this is the (or at least ‘a’) problem with all of Vardiman’s research. If young Earth creationism were the dominant scientific paradigm and these premises had been verified he would actually be doing good science. But because we do not live in such a world Vardiman’s work is very much out of place and a waste of resources to boot. A much better use of his time would have been trying to model the flood itself, or some other laying-the-groundwork research.

And there are more problematic premises:

We compared the three actual nor’easters and the simulated storms to ensure that the numerical model used faithfully replicated them. The sea-surface temperatures in the Atlantic were then artificially warmed by 18°F, and the storms were re-analyzed for any changes.

For his analysis Vardiman has simply taken a handful of recent storms and cranked up the temperature of the ocean to watch what would happen. He posits no other changes to the system than this. That’s a little odd given that the storms are supposed to have begun right after a catastrophic flood that destroyed most life on Earth, and other changes to the relevant parts of the climate. Was the Gulf stream running? Would the lack of trees do anything to the conditions, by any chance? There are many considerations that are not accounted for by Vardiman, which would be problematic if the event he was trying to simulate had really happened.

There are some things that he does recognise, however:

It is possible that the boundaries of the model simulation may have artificially influenced the circulation patterns, particularly near the eastern boundary. Air can flow through the boundaries of the model, but other effects such as pressure, temperature, and divergence interacting with the boundary may have caused some of the irregularities. The most likely cause for such an artificial influence may be that the warmer sea-surface temperature was specified only in the domain shown. One way to mitigate these effects, if they exist, would be to enlarge the domain.

If only the area that he is directly modelling is set to the increased temperature that would make one hell of a butterfly…

The conclusion that Vardiman draws from his modelling is that the increase in ocean temperature would lead to an increase in precipitation (specifically, snow) from storms like these. He thinks that this would be sufficient to explain the size of the North American ice sheet during the ice age:

Even with the heaviest precipitation falling over the Atlantic Ocean in these simulations, the widespread lighter precipitation of about 100 mm (4 in.) water equivalent per 24 hours would have contributed to a significant accumulation of snow during the ice age. This rate would produce about 500 mm (20 in.) of snow in 24 hours. If Vardiman and Brewer (2010c) are right that during the ice age following the Genesis Flood a storm formed and moved across the United States every three days all year round, this would produce an uncompressed snow pack of about 60 m (~200 ft.) per year. Upon compression to solid ice this would produce an ice layer of about 1,200 m (4,000 ft.) in 100 years. If bands of additional precipitation were swept around the centers of circulation of enhanced nor’easters off the coast of New England, the accumulation would have been even greater. Just the accumulation of snow and ice from the widespread lighter precipitation would have been more than adequate to explain the evidence for a thick Laurentide ice sheet in eastern Canada during a recent ice age.

I haven’t looked at the storm every three days claim, but it certainly sounds fishy. There is another aspect to consider here, however: could this snowfall be sustained? Even if we supposed that an icepack really was being built up, its very accumulation would change the climate and weather patterns that is producing it. And how long would this warmed ocean take to cool? Vardiman is a long way from having results that are “more than adequate” to explain the observations.

So there are a lot of potential criticisms that could be levelled against this research. To recap, they can be separated into three groups. The first relates to the premises on which the research is based: that the flood happened, that the seas warmed because of this, that all geographical features were nevertheless the same. These premises are a very long way from being sound, but we will have to take them as read to even be able to look at the model itself.

The second relates to the model itself. I could be wrong here, but it seems to me that the modelling doesn’t really simulate a world with seas 18°F (10°C) higher than normal. Instead, Vardiman has taken the starting conditions for a number of known storms and placed a hot plate beneath them to see what would happen. If I’m right, this has provoked his “lighter precipitation” across most of the area of study without regard for the storm, meanwhile the temperature imbalance between the area inside the simulation and the normal temperatures outside may be responsible for the apparent movement of the larger quantities of precipitation out into the Atlantic (suspiciously close to the edge in places). If even some of that is true then it casts serious doubts over the whole enterprise.

The third group relates to the conclusions drawn in the final quote above. Even if there was no problem with the models and premises, it may well be that through the accumulation of snow and the cooling of the oceans the conditions that Vardiman credits with producing his ice age would not have been able to sustain themselves for long enough.

Anyway, that’s the cutting-edge creation science that Vardiman used to produce before he retired. I’m sure he will be missed.