Simulating the Hadley Cell using Energy2D

Download the models

Although it is mostly used as an engineering tool, our Energy2D software can also be used to create simple Earth science simulations. This blog post shows some interesting results about the Hadley Cell.

The Hadley Cell is an atmospheric circulation that transports energy and moisture from the equator to higher latitudes in the northern and southern hemispheres. This circulation is intimately related to the trade winds, hurricanes, and the jet streams.

As a simple way to simulate zones of ocean that have different temperatures due to differences in solar heating, I added an array of constant-temperature objects at the bottom of the simulation window. The temperature gradually decreases from 30 °C in the middle to 15 °C at the edges. A rectangle, set to be at a constant temperature of -20 °C, is used to mimic the high, chilly part of the atmosphere. The viscosity of air is deliberately set to much higher than reality to suppress the wild fluctuations for a somehow averaged effect. The results show a stable flow pattern that looks like a cross section of the Hadley Cell, as is shown in the first image of this post.

When I increased the buoyant force of the air, an oscillatory pattern was produced. The system swings between two states shown in the second and third images, indicating a periodic reinforcement of hot rising air from the adjacent areas to the center (which is supposed to represent the equator).

Of course, I can't guarantee that the results produced by Energy2D are what happen in nature. Geophysical modeling is an extremely complicated business with numerous factors that are not considered in this simple model. Yet, Energy2D shows something interesting: the fluctuations of wind speeds seem to suggest that, even without considering the seasonal changes, this nonlinear model already exhibits some kind of periodicity. We know that it is all kinds of periodicity in Mother Nature that help to sustain life on the Earth.

Simulating geometric thermal bridges using Energy2D

Fig. 1: IR image of a wall junction (inside) by Stefan Mayer

One of the mysterious things that causes people to scratch their heads when they see an infrared picture of a room is that the junctions such as edges and corners formed by two exterior walls (or floors and roofs) often appear to be colder in the winter than other parts of the walls, as is shown in Figure 1. This is, I hear you saying, caused by an air gap between two walls. But not that simple! While a leaking gap can certainly do it, the effect is there even without a gap. Better insulation only makes the junctions less cold.

Fig. 2: An Energy2D simulation of thermal bridge corners.

A typical explanation of this phenomenon is that, because the exterior surface of a junction (where the heat is lost to the outside) is greater than its interior surface (where the heat is gained from the inside), the junction ends up losing thermal energy in the winter more quickly than a straight part of the walls, causing it to be colder. The temperature difference is immediately revealed by a very sensitive IR camera. Such a junction is commonly called a geometric thermal bridge, which is different from material thermal bridge that is caused by the presence of a more conductive piece in a building assembly such as a steel stud in a wall or a concrete floor of a balcony.

Fig. 3: IR image of a wall junction (outside) by Stefan Mayer

But the actual heat transfer process is much more complicated and confusing. While a wall junction does create a difference in the surface areas of the interior and exterior of the wall, it also forms a thicker area through which the heat must flow through (the area is thicker because it is in a diagonal direction). The increased thickness should impede the heat flow, right?

Fig. 4: An Energy2D simulation of a L-shaped wall.

Unclear about the outcome of these competing factors, I made some Energy2D simulations to see if they can help me. Figure 2 shows the first one that uses a block of object remaining at 20 °C to mimic a warm room and the surrounding environment of 0 °C, with a four-side wall in-between. Temperature sensors are placed at corners, as well as the middle point of a wall. The results show that the corners are indeed colder than other parts of the walls in a stable state. (Note that this simulation only involves heat diffusion, but adding radiation heat transfer should yield similar results.)

What about more complex shapes like an L-shaped wall that has both convex and concave junctions? Figure 3 shows the IR image of such a wall junction, taken from the outside of a house. In this image, interestingly enough, the convex edge appears to be colder, but the concave edge appears to be warmer!

The Energy2D simulation (Figure 4) shows a similar pattern like the IR image (Figure 3). The simulation results show that the temperature sensor placed near the concave edge outside the L-shape room does register a higher temperature than other sensors.

Now, the interesting question is, does the room lose more energy through a concave junction or a convex one? If we look at the IR image of the interior taken inside the house (Figure 1), we would probably say that the convex junction loses more energy. But if we look at the IR image of the exterior taken outside the house (Figure 3), we would probably say that the concave junction loses more energy.

Which statement is correct? I will leave that to you. You can download the Energy2D simulations from this link, play with them, and see if they help you figure out the answer. These simulations also include simulations of the reverse cases in which heat flows from the outside into the room (the summer condition).

Time series analysis tools in Visual Process Analytics: Cross correlation

Two time series and their cross-correlation functions

In a previous post, I showed you what autocorrelation function (ACF) is and how it can be used to detect temporal patterns in student data. The ACF is the correlation of a signal with itself. We are certainly interested in exploring the correlations among different signals.

The cross-correlation function (CCF) is a measure of similarity of two time series as a function of the lag of one relative to the other. The CCF can be imagined as a procedure of overlaying two series printed on transparency films and sliding them horizontally to find possible correlations. For this reason, it is also known as a "sliding dot product."

The upper graph in the figure to the right shows two time series from a student's engineering design process, representing about 45 minutes of her construction (white line) and analysis (green line) activities while trying to design an energy-efficient house with the goal to cut down the net energy consumption to zero. At first glance, you probably have no clue about what these lines represent and how they may be related.

But their CCFs reveal something that appears to be more outstanding. The lower graph shows two curves that peak at some points. I know you have a lot of questions at this point. Let me try to see if I can provide more explanations below.

Why are there two curves for depicting the correlation of two time series, say, A and B? This is because there is a difference between "A relative to B" and "B relative to A." Imagine that you print the series on two transparency films and slide one on top of the other. Which one is on the top matters. If you are looking for cause-effect relationships using the CCF, you can treat the antecedent time series as the cause and the subsequent time series as the effect.

What does a peak in the CCF mean, anyways? It guides you to where more interesting things may lie. In the figure of this post, the construction activities of this particular student were significantly followed by analysis activities about four times (two of them are within 10 minutes), but the analysis activities were significantly followed by construction activities only once (after 10 minutes).

Time series analysis tools in Visual Process Analytics: Autocorrelation

Autocorrelation reveals a three-minute periodicity

Digital learning tools such as computer games and CAD software emit a lot of temporal data about what students do when they are deeply engaged in the learning tools. Analyzing these data may shed light on whether students learned, what they learned, and how they learned. In many cases, however, these data look so messy that many people are skeptical about their meaning. As optimists, we believe that there are likely learning signals buried in these noisy data. We just need to use or invent some mathematical tricks to figure them out.

In Version 0.2 of our Visual Process Analytics (VPA), I added a few techniques that can be used to do time series analysis so that researchers can find ways to characterize a learning process from different perspectives. Before I show you these visual analysis tools, be aware that the purpose of these tools is to reveal the temporal trends of a given process so that we can better describe the behavior of the student at that time. Whether these traits are "good" or "bad" for learning likely depends on the context, which often necessitates the analysis of other co-variables.

Correlograms reveal similarity of two time series.

The first tool for time series analysis added to VPA is the autocorrelation function (ACF), a mathematical tool for finding repeating patterns obscured by noise in the data. The shape of the ACF graph, called the correlogram, is often more revealing than just looking at the shape of the raw time series graph. In the extreme case when the process is completely random (i.e., white noise), the ACF will be a Dirac delta function that peaks at zero time lag. In the extreme case when the process is completely sinusoidal, the ACF will be similar to a damped oscillatory cosine wave with a vanishing tail.

An interesting question relevant to learning science is whether the process is autoregressive (or under what conditions the process can be autoregressive). The quality of being autoregressive means that the current value of a variable is influenced by its previous values. This could be used to evaluate whether the student learned from the past experience -- in the case of engineering design, whether the student's design action was informed by previous actions. Learning becomes more predictable if the process is autoregressive (just to be careful, note that I am not saying that more predictable learning is necessarily better learning). Different autoregression models, denoted as AR(n) with n indicating the memory length, may be characterized by their ACFs. For example, the ACF of AR(2) decays more slowly than that of AR(1), as AR(2) depends on more previous points. (In practice, partial autocorrelation function, or PACF, is often used to detect the order of an AR model.)

The two figures in this post show that the ACF in action within VPA, revealing temporal periodicity and similarity in students' action data that are otherwise obscure. The upper graphs of the figures plot the original time series for comparison.