Models(3) Predictable Arguments: In the comments to the previous blog there was this quote attributed to John Christy in a Wall Street Journal Commentary.
“It is my turn to cringe when I hear overstated-confidence from those who describe the projected evolution of global weather patterns over the next 100 years, especially when I consider how difficult it is to accurately predict that system's behavior over the next five days.”
(John Christy WSJ Nov 1, 2007)This statement implies that the ability or the inability to predict weather at five days has a direct relationship to the ability or the inability to predict that the planet will warm because of the increasing burden of greenhouse gases. There is no knowledge-based foundation to support this assertion.
Start: Consider the role of weather in the climate system. Atmospheric motion comes in response to temperature variability. It is straightforward to show that if there is temperature variability (a temperature gradient), then there will be pressure variability (a pressure gradient), and pressure gradients are the initiator of motion in the atmosphere. Because of the tilt of the Earth’s axis, the planet gets more heating in the tropics and less at poles. From the point of view of climate, weather in the atmosphere and motion in the ocean, exists to smooth out this persistently forced equator to pole temperature difference.
From that global climate perspective weather is like turbulence. If you were to place a smoky oil lamp in the corner of a closed room, it would be difficult for you to say exactly where you could smell (detect) smoke at any moment, but with some certainty you could say that the room would get smoky. We can say with absolute certainty that the atmosphere will maintain its role of moving heat from the equator to the pole. We can confidently conclude that if the temperature difference between the equator and the pole changes, there will be an impact on the weather.
Forced Behavior: Consider another example, the surface of a lake or the ocean. There are generally waves on the surface. It is difficult for you to stand on the shore and to predict the behavior of each particular wave and the how those waves will interact with each other. You could, if you identified a specific wave, predict with some skill how that wave would propagate for a while. You could predict, for instance, when it gets to shore. If you set out to predict all of the waves it would drive you crazy. That’s the weather problem, predicting all of those individual waves.
What if the wind increased? You could predict with great certainty that the waves on the surface of the ocean would get larger. You could predict how far the waves would impede on the beach. There are many attributes about the waves that you could predict that would provide useful, actionable information for someone with a house on the beach. Yet, you would still not be able to very accurately describe each of the individual waves.
In the above example the increased wind forces the wave field to a new basic state.
Weather and Climate: The weather problem is largely a problem of predicting waves that are sloshing around trying to smooth out local variations of thermal energy. We can pick out the weather systems with our observations, and make useful predictions. But we have to keep observing and picking out those systems to make useful predictions. The climate problem is one of increased forcing. If we change the atmosphere to hold more energy near the surface of the Earth, it will warm. This climate prediction is independent of the ability to predict the weather.
-----
In the Wall Street Journal commentary Christy speaks of being humbled by the complexity of the climate system, and then states a belief:
“Mother Nature simply operates at a level of complexity that is, at this point, beyond the mastery of mere mortals (such as scientists) and the tools available to us.”
(John Christy WSJ Nov 1, 2007)This is a belief. It is not a belief that I share.
Consider complexity.
Watch the movie
Apollo 13 . The complexity of the rocket launch and a successful lunar mission is staggering. Apollo 13 was the one where there was a major failure in the space vehicle. This magnified the complexity. Rocket science is about complexity, but like the weather and climate problem, the physics of rocket are astoundingly simple and well understood. We push around objects in the field of gravity. It would have been easy to look at that accident and say that the complexity is staggering; it can’t be understood; there is no actionable information. That is not what people did.
Rockets though are the product of humans. Consider your body. It is complex. It is difficult to predict with precision what a drug will do to an individual. It is difficult to predict how an individual will respond to a particular environmental exposure. Does this inability to predict the response of a particular individual, mean that warnings about excess heat and dehydration have no value? Does this inability to understand and represent the entire complexity of even a single human, tell us that there is no sense in pursuing medical solutions?
I stand humbled by the complexity of the Earth. That does not mean that I look at the Earth and conclude that its complexity is beyond the ability of people to understand what is happening and what is going to happen.
I am reminded of a friend of mine who did not believe in getting medical procedures. He had a heart attack and survived. He knew, then, that he had heart disease. He knew he could get it treated, but he chose not to get it treated. He knew he would die earlier if he did not get it treated. It was a belief of his, and it was a belief that, in the end, was his alone to make.
There is a profound difference from my friend’s individual decision and when we have knowledge about what is going to happen, and that knowledge is consequential to all. Be clear - it is a belief that the complexity of Mother Nature is beyond mere mortals to understand – John Christy’s personal belief. There is tremendous evidence that we can approach Mother Nature’s complexity, extract information, and provide knowledge about the future. It is responsible to take action on this knowledge because the impact is not individual.
Here are the previous blogs on models.Uncertainty and Types of ModelsModels (1) AssumptionsModels (2) Forgotten LayersChapter 16: Fundamentals of Modeling ....