TAEM- This past month our publisher, Joseph J. O’Donnell, has offered a challenge to science professionals and enthusiasts to offer advice and guidance to NASA. This came about when we discovered a statement in another media source that stated that the agency seemed to be ‘Lost in Space’. Apparently NASA stated that there were no set programs for future space exploration. George Mason University in Northern Virginia was the first college to rise to the occasion to offer those needed boosts to the space agency.
On the staff of the university we found Dr. Bob Weigel of the Department of Computational and Data Sciences. Dr. Weigel, please tell our readers about your educational training and how this led to the courses that you teach.
NB: I am with the School of Physics, Astronomy, and Computational Sciences (SPACS; [http://spacs.gmu.edu/]). The Department of Computational and Data Sciences combined with the Department of Physics and Astronomy to form SPACS.
BW-My undergraduate degree is in Mechanical Engineering. I did a MS in Physics in which I studied complex dynamical systems. Space Weather is both complex and and a dynamic system, so I did a PhD in Physics with an emphasis on plasma physics and space weather.
The courses that I regularly teach at the graduate level are Plasma Physics, Space Weather, and Statistical Methods in the Space Sciences. These are all topics that I deal with daily in my research.
Initially, I taught many standard physics and astronomy courses at the undergraduate level. Over the past five years, we have developed a new undergraduate degree called “Computational and Data Sciences”. We observed that many scientists graduated with a strong scientific background, but they lagged in scientific computing and data analysis skills. However, to be most successful in the current job market, one must demonstrate competency in all three fields.
BW-One of the courses that I am teaching this semester is “Computing for Scientists” [http://cds130.org/]. We developed this course to address the fact that students in upper-division science courses need to know scientific computing. This course covers everything from how a computer works internally to writing actual programs in the MATLAB programming language. A typical final project involves the students being given a set of black and white images, some of which show a malignant tumor. Their job is to write a computer program that detects which images have a malignant versus benign tumor in a set of 10,000 images.
In recent years, I have begun to offer a “distance education hybrid” version of the course. Instead of meeting 150 minutes in class per week, the students only meet for 75 minutes on campus. The rest of the previously in-class time, students read lecture notes and watch video lectures of what I would typically present in class. Instead of coming to class to hear me lecture, they do it on their own time. When they are in class, students work in groups on problems that extend what they read or watched before class. I like this format. Instead of fighting traffic to watch a non-interactive presentation, they watch the presentation on their own time as many times as they feel necessary. I have not tried a fully online version. My feeling is that there is much value in working with other students on problems in class and that it will be difficult to deliver an equivalent learning experience online. I’ll probably try it sometime in the future after I have a better understanding of the limitations.
The graduate-level courses that I teach are typical in format. The first half of the semester, students do homework problems and turn in hand-written solutions. In the second half of the semester, I begin to assign projects and the students give presentations on their progress during class while I and the other students give suggestions and ask questions.
TAEM- You have many interests and among them is Magnetospheric physics and geomagnetism. Please explain these and the connections between them.
BW-Earth’s magnetosphere extends from the upper-most regions of our atmosphere and encompasses a huge region of space around Earth. The magnetosphere is created by the interaction of plasma emitted by the solar wind with the Earth’s magnetic field. Most of us are familiar with the sun’s interaction with Earth’s atmosphere – photons from the sun pass through and interact with atmospheric gasses. In a similar way, charged particles emitted from the sun, which form the solar wind, interact with Earth’s magnetic field.
The connection between geomagnetism and the magnetosphere is that although Earth’s magnetic field is generally quite steady, electric currents in the magnetosphere cause perturbations in the magnetic field measured on the Earth’s surface. If you had a very, very sensitive compass, you would see it fluctuate more at times when the solar wind is more active.
The motivation for studying geomagnetism is that the magnetosphere-induced fluctuations indirectly tell us something about what is happening in the magnetosphere and the solar wind. This is similar to the reason that climate scientists study measurements of the wind near the surface of the Earth – these measurements tell us a bit about the winds that are happening in places higher up and surface wind measurements are cheaper than those made by instruments sent into the upper atmosphere.
BW-Space Weather involves the study of processes driven by the solar wind. Previously I mentioned one of these processes – fluctuations of the magnetic field observed near the surface of Earth. But there are many, many other processes, including ones that take place in the magnetosphere and ionosphere. Research into solar wind/magnetosphere/ionosphere coupling asks a basic question: given a measurement in the solar wind, what will happen in the magnetosphere and ionosphere? Both the magnetosphere and ionosphere are complex systems. To answer this question, we need to understand the basic physical processes that govern each system and how they interact.
The more we know about solar wind/magnetosphere/ionosphere coupling, the better we can make predictions, which is one of the primary goals of space weather scientists.
TAEM- What effects does space weather have on space travel ?
BW-Actually space weather has impacts on airplane travel. At times the solar wind is such that there is an enhanced amount of radiation at airplane altitudes near the poles. It is often least expensive for certain flights to fly over the poles, but if there is enhanced radiation, the flights are re-routed to protect the passengers. In space, you don’t have Earth’s atmosphere, which absorbs much of this space weather induced radiation.
Another space weather impact is high energy charged particles interacting with computers on satellites. If a charged particle collides with a computer component on the satellite, it can either damage the computer or cause it to send a “phantom” signal – if a charged particle hits an electrical component in a certain way, it can mimic a real signal. In the worst case, this signal could be “shut down” or “change orbit”.
To protect against space weather, satellites must have additional protections. These protections can be costly and increase the weight of the spacecraft. This is why forecasting is important. If we can predict when something is going to happen, the satellite operator can sometimes take protective action.
TAEM- Your interests also apply to inverse methods for Magnetospheric modeling and nonlinear dynamics. Please inform our readers about these and their importance to the space program.
BW-There are broadly two ways of gaining a better understanding of a complex system like the magnetosphere. In the forward modeling approach, one writes down an approximate set of equations that describe the system. The solutions to these equations are then compared with real data.
In the inverse approach, one uses data to infer the characteristics the forward model must have by looking at the data. We say that we “invert the data to get a model of the system”.
Both the forward and inverse modeling methods are used to gain a better understanding of systems with complex and nonlinear dynamics. The forward modeling approach works well for simple and linear systems. Both approaches are needed otherwise.
TAEM- Tell us about the decision theory as applied to rare event forecasting, scientific visualization, and collaborative software development.
BW-Space weather forecasts typically make a prediction about how big a space weather event will be. For example, one prediction could be about how much radiation will there be near the poles at 10pm. Given this information, an air traffic controller needs to make a decision. Should a flight be re-routed? If space weather predictions were 100% accurate, the decision would be easy. In reality, space weather predictions are not perfect – sometimes there will be a prediction for a high amount of radiation and a low amount of radiation will actually occur. Decision theory involves answering the question of what action to take in the face of imperfect forecasts. The calculation involves consideration of (1) the forecasts past performance and (2) the costs and benefits for taking action based on the forecast.
As an example, suppose an operator for a communications satellite learns that a large space weather event has been forecasted. If he takes protective action, for example, by shutting down the satellite for a few hours, his company will lose revenue because their customers are not being served. If he does not take protective action, he could lose the satellite. Decision theory can be used to give guidance to the operator interested in minimizing financial loss.
TAEM- How would you and your students collaborate on a plan to offer advice and support to NASA, and where do you suggest space exploration head in both the short and long terms, and why?
BW-NASA does two types of explorations in space – manned and unmanned.
I think the “Lost in Space” perspective arises because NASA must face two realities:
1) Manned missions generally have a higher public relations value; and
2) Un-manned missions generally have a higher scientific value.
However, when they do too many un-manned missions, the public asks for more public relations value missions. NASA reacts and then the public says, “If your objective is science, why are you doing things with little science value? Are you losing focus?”
Of course, scientists could make NASA’s life easier if the science we did with data from unmanned missions had more PR value. The reality is that sometimes it is quite difficult. There are times that I wish my data could be made as beautiful as the data from the Hubble Telescope. At one point I put some thought into a Space Weather Picture of the Day web site – something to popularize what we do in the same way that Astronomers popularize their science with the Astronomy Picture of the Day web site. I came up with very few pictures that non-specialists found interesting, so I dropped the idea. I still do outreach and given lectures about space weather, but still seek a way of communicating our science in a way that can capture the interest of a broad audience in the same way that astronomy does.
The time for an unmanned mission is constrained by the instrument lifetimes, fuel, the budget for receiving the data, and the budget for monitoring the spacecraft.
In addition, the most interesting part of space for science often don’t require humans and would be quite boring for the public.
Given these two realities that NASA must juggle, it is not surprising that they are often viewed as “Lost in Space”.
TAEM- Dr. Weigel, It has been both a pleasure and an honor to be able to do this interview with you for The Arts and Entertainment Magazine and I assure you that the many students who follow our publication have learned much from you. We also look forward to further input from you and your students, and fellow faculty members, in the near future.