Computer Models and Atmospheric Circulation: A Major Failure

by Dr. Tim Ball on May 5, 2011

in Atmosphere,Data,Government,Theory,Tropics

The computer is down. I hope it is something serious. ~Stanton Delaplane

We still don’t understand how the atmosphere circulation works, so it is incredible that the computer modelers imagined it could be simulated. Of course, this is beyond the question of the basis for the models. There is simply insufficient surface weather data, and it is worse above the surface. As Denis Rancourt commented,

Although these models are among the most elaborate predictive models of complex non-linear phenomena, they are nonetheless sweeping oversimplifications of the global climate system and its mechanistic intricacies.

Three-cell climate model Figure 1: The three-cell model.

When I started studying weather and climate it was believed there were three cells: the Polar, Ferrel, and Hadley (Figure 1).

In that theory, the complexity was between the Ferrel and Polar Cells. Then it was argued that the Ferrel Cell did not exist, and air movement between the Polar and Hadley Cells was extremely complicated (Figure 2).

The Greenhouse Effect compares the atmosphere to a greenhouse, but they don’t compare in a multitude of ways.

The atmosphere is mostly heated by circulation that doesn’t occur in a greenhouse. More important, transfer of most of the heat is by evaporation and wind.

The more complicated climate model Figure 2: The more complicated model.

Critical Role of Moisture

When solar energy strikes a surface, it sets the molecules in motion; that raises the temperature of the surface. If the surface is water, the molecules can reach escape velocity and move from the water to the air. In doing so, they go through a phase change because they change from a liquid (water) to a gas (water vapor). The heat is used, but not lost. It’s retained within the gas molecule and known as latent heat. Movement of the air, known as convection if vertical, and advection if horizontal, then transports it. Advection is more commonly known as wind. When the air is cooled the process is reversed and known as condensation. The latent heat is released to warm the atmosphere, which is why the temperature rises when precipitation occurs.

Maximum heating occurs at what is called the Heat Equator. It doesn’t coincide with the geographical Equator, but with the latitude where the sun’s rays are vertical and therefore imparting maximum energy. Therefore, the Heat Equator migrates with the seasonal movement of the vertical rays.

George Hadley identified the equatorial cells named after him in 1735 based on wind direction records from British ships. It was a remarkable intuition to extrapolate surface winds into a three dimensional model (Figure 3).

Simple diagram of a Hadley cell Figure 3: A simple diagram of the Hadley Cell

Called convective cells, they’re driven by the heated air rising where the two cells connect. This creates a distinctive zone called the Inter Tropical Convergence Zone (ITCZ). Where the air is rising, winds are essentially vertical, which is of no help to a sailing ship. It was a dead zone they named the doldrums. Nowadays, the primary use of the word is to define a feeling of boredom or depression. Similar calm occurs in the subtropics, but here the wind is downward. For sailing ships the problem is the same, a lack of horizontal wind, so they were becalmed. These regions are called the horse latitudes, reputedly because the Spanish were unable to feed or water the horses so the crew was forced to throw them overboard.

Clouds depicted in Figure 3 are a vertical development form called cumulonimbus (Cb). They build to the top of the troposphere and occasionally punch through into the stratosphere. Recorded internal vertical wind speeds reach 500 kph and carry enormous amount of heat and moisture to high altitudes. They are the major transport mechanism of heat to the higher latitudes as the atmosphere tries to offset the imbalance of surplus energy in the Tropics with deficit energy in the Polar Regions.

Big Holes in the Computer Models

Computer climate models are called General Circulation Models (GCM). They are mathematical models that divide the atmosphere into boxes. Their surface footprint is a rectangle defined by latitude and longitude. As Essex and McKitrick (E&M) in their book Taken By Storm (revised edition) explain,

Grid boxes are to GCMs what pixels are to computer or TV screens. A pixel is the smallest dot that makes up the picture that you watch on the screen. … Extracting climate information from an individual grid box is like watching a TV program on mute from one pixel.

What they are talking about is known as the sub-grid problem. We know thunderstorms are ”like an enormous, natural, self-powered, heavy-duty air conditioner.” This means it is taking the surplus heat and moving it up in the atmosphere to create cooling and move it to deficit areas. As E&M note, we know the role of the thunderstorm, but are unable to model how it functions. They wrote,

Even with a mathematical theory, no supercomputer in the world can capture the imponderable complexity of a thunderstorm, nor is one expected to emerge that can.

So what’s the problem? There are thousands of thunderstorms every day mostly along the ITCZ that are a critical part of the earth’s atmospheric system. E&M explain,

However, the mighty creature overhead, along with all its cousins, is too small to show up in even the biggest and grandest global climate models. They are, in the jargon of the field, sub-grid scale. That is a computerese way of saying that they fall between the cracks. Not only can we not handle today’s thunderstorms, no such storm ever shows up, even in our very best climate computer models. Thunderstorms are not dealt with in climate models, from first principles or any other way.

They can’t ignore their effect, because the models wouldn’t work at all; so they create empirical rules to mimic the overall effect, called parameterizations.

E&M identify the values of the models. I am not so sure, however, I agree with their assessment that, “somewhere along the line, this straightforward common sense about the uncertainty of models and their relationship to global climate change was pushed aside.” They say they don’t know when it started, “but it is definitely not scientific in nature”. It was followed by what they call The Doctrine of Certainty. I can help them: it arrived when computer modelers with no weather or climate knowledge took over the discipline in the 1980s. I watched them dominate climate conferences and meetings. They then moved to centre stage with the Intergovernmental Panel on Climate Change (IPCC). Maurice Strong said he went to the UN because he could get all the money he wanted and not be accountable to anyone. Computer modeling is very expensive, and as we now know, members of the IPCC have not been held accountable.

Computers can figure out all kinds of problems, except the things in the world that just don’t add up. ~James Magary