I'm a professor at U Michigan and lead a course on climate change problem solving. These articles often come from and contribute to the course.
By: Dr. Ricky Rood , 1:41 AM GMT on October 05, 2012
Modeling Summary and A Change in the Weather: Models, Water, and Temperature (9)
In this entry, I am doing a first summary of my modeling series, and exercising my habit of discussing a paper of special interest or importance. For those who came in late, here is the Introduction to the series, and the previous entry. All of entries in the series are linked at the end.
Doing Science with Models 1.6: I have tried to demystify the use of models in science and climate science in several ways. Here is the series of ideas that I have tried to line up.
1) Models are everywhere, and we use them all of the time. I introduced examples of commonly used models such as ledger sheets and building plans. In fact, whenever we are faced with a new problem, we naturally look to models for possible solutions. Most commonly that model is – do I have experience in a previous situation that helps me in this situation? That might be followed with - do I have friends who have relevant experience? Can I hire expertise? When we are faced with no experience of a situation; that is, we have no model, then we are thrown into a situation where we might have difficulty understanding impacts, risks, and what to do. Whether or not we explicitly recognize it, models are part of human thinking and problem solving. (Models are Everywhere, Ledgers, Graphics, and Carvings)
2) The arithmetic that we use to figure out how much money we have, the budget equation, is a model.
Today’s Money = Yesterday’s Money + Money Gained – Money Spent
Some of the models that we use are mathematical and provide us with a way to quantify things that are important to us. (Balancing the Budget)
3) We have become comfortable with coding models on computers. With the spread of computers in the past 20 years, we, for example, use computers to balance our checkbooks and plan our budgets. We enter numbers and words into forms and press some buttons, and seconds later, we have categorized accounting of our income and expenses. (Ledgers, Graphics, and Carvings)
4) The same form of the budget equation that we use to balance our checkbook can be used to make an accounting of the energy of the Earth.
Today’s Energy = Yesterday’s Energy + Energy Gained – Energy Lost.
Therefore, if we can measure energy, sources of energy, and loses of energy, we can make a quantitative accounting. (Balancing the Budget)
5) Point of view is important. The accounting of the Earth’s energy, hence a description of the climate, depends on your point of view. If you were sitting on Mars, then you might only be interested in the energy that comes to and leaves the Earth. If you are a person on the surface of the Earth, you need to know the energy in the atmosphere, the land, the ocean, and the ice. Therefore you need budget equations for each of these components of the Earth’s climate. This is like having several energy accounts. The transfers between accounts appear as exchanges: loses to one account and gains to others. (Point of View)
6) Complexity arises because there are many energy accounts and many ways to transfer energy from one account to another. Even though every energy exchange might be simple, when we put all of the exchanges together the total system is complex. Energy might collect in one place, for example evaporated water in the tropical atmosphere, and it might be lost and deposited some place else, for example ice sheets in Greenland. There is the possibility of transfer of large amounts of energy between these collections of energy. (Looking Under the Cloak of Complexity)
7) The Earth’s climate is constrained by the processes that govern the transfer of energy from one account to another. Well-known rules, or laws, govern the way that energy is transferred. They are strictly and precisely defined. The Earth’s climate can be quantified by accounting for the energy. Because of the laws that govern energy transfer, we can in principle make credible estimates of the Earth’s climate in the future. (The Free Market and the Climate Model)
8) It takes time for energy to move around to the different energy accounts. For example, a lot of energy can be stored in the ocean for long periods of time. Long? Compared to what? Long compared to the atmosphere and perhaps compared to the life times of humans. Ice sheets have had life times of hundreds of thousands of years, and they represent the accumulation of many years of energy transport. (Looking Under the Cloak of Complexity)
If we use this framework to think about climate, climate models, and climate change, then when we add carbon dioxide to the atmosphere what are we changing? From the point of view of the human on the surface of the Earth, we are changing the amount of time that the energy from the Sun is stored near the surface of the Earth. Some of this energy shows up as an increase in temperature at the surface. Some warms the ocean. Some changes the water budget and the weather. Ultimately, some makes it way back to outer space, but not until after it causes a set of changes important to the person on the surface of the Earth.
Going forward, I will explore more deeply complexity and how we can manage this complexity to make and interpret predictions.
Interesting Research: A Change in the Weather ? - This past summer saw a record low in Arctic sea ice. (nice blog in Washington Post) The previous record low was in 2007. There are those who dismiss this as a record low of sea ice because it is from “satellite data,” which are only about 30 years of observations. But I would argue that we can make a pretty convincing argument that these are record lows for, well, thousands of years.
The paper I want to write about is “The Recent Shift in Early Summer Arctic Circulation,” by James Overland and co-authors. This paper documents a “shift” in the Arctic climate that has persisted through the past 6 years (2007 – 2012). This shift is in the atmospheric circulation, and it is described as an increase in atmospheric surface pressure on the North American side of the Arctic and a decrease in pressure on the Siberian side. (To get this perspective, look down on a map of the Earth from above.) This circulation pattern has been especially strong in June.
A consequence of this circulation pattern is that there is flow of air from the south along the date line in the Pacific Ocean, essentially through the Bering Strait into the Arctic. This pushes sea ice northward, and brings warm air towards the North Pole. This contributes to rapid melting of sea ice. To the point of complexity, this movement of warm air into the Arctic is not the only contributor to the melting ice. During years of extreme melting, there has been reduction in cloudiness, allowing more Sun to get to the surface. There has also been more heat transport by the ocean. Finally, ice melting has been accelerated by mixing of warm(ish) water from the MacKenzie River farther into the ocean. Rather than each of these processes being viewed as perhaps “the cause” of enhanced sea ice melting, all of these processes should be viewed as a system, where they all add up to more melting.
James Overland and co-authors label this a “shift.” It is unarguably a persistent pattern, and the authors present statistical evidence that such a pattern has not been present in more than 60 years of observations. The question of whether this is a shift to a new pattern that will persist going forward remains open. One way to study this is to study whether or not known changes to the surface might cause these patterns in the atmosphere. Of special interest, of course, have the changes in sea ice initiated a change in circulation that has accelerated the loss of sea ice? Overland and co-authors also point to the possibility that the large decrease of snow cover in late spring and early summer could potentially enhance the circulation pattern. Such persistent circulation patterns are one of the most difficult phenomena for climate and weather models to represent.
Figure 1: Rutgers University Global Snow Lab. Departure of Snow Cover in June 2012 from a 30 year, 1971-2000, average. The legend is percent difference, with oranges less than zero and blues greater than zero. Make your own maps here.
As a final remark I want to return to the idea of whether or not the melting of the sea ice might force an atmospheric circulation that then contributes to the melting of more sea ice. In my series on modeling and my discussion of complexity, I talked about how billions of simple transactions in millions of accounts can come together to represent a very complex system. One of the characteristics of complex systems is the presence of processes that once they occur, they amplify themselves. In this case, could a change in sea ice cause a change in the atmosphere that amplifies the change in sea ice? This type of reinforcing behavior, or positive feedback, contributes to complexity in a fundamentally different way than a damping or negative feedback. In a damped system, a change that decreased sea ice would cause a change that would increase sea ice to maintain a balance. It is important to recognize the difference between a damped system and a system in a state of balance. A balanced state, when perturbed, might find a new balance. Or, it might dance around all over the place until a new balance is found. Most of the evidence is that the Earth’s climate is not a damped system, but a balanced system. As we change it by increasing the temperature at the surface, we should expect it to bounce around looking for a new balance.
Models, Water, and Temperature
Models are Not All Wet: Series Introduction
Models are Everywhere
Ledgers, Graphics, and Carvings
Balancing the Budget
Point of View
Looking Under the Cloak of Complexity
The Free Market and the Climate Model
The views of the author are his/her own and do not necessarily represent the position of The Weather Company or its parent, IBM.