Preflight Interview: Janice Voss

Before we get into the specifics of this particular flight, I want you to tell me a little bit about yourself. Why did you want to become an astronaut?

I first got interested in the space program in sixth grade. I was going on vacation with my family for the summer. We stopped by the library as we always do to pick up reading books for the trip. And I picked up a book called, "A Wrinkle in Time" by Madeleine L'Engel, which is a science fiction story about some kids that get involved in space stuff and save the world, and I just thought it was a great story. It really opened my mind to a lot of possibilities that I thought were fascinating. So I started reading more science fiction. That led me into more science nonfiction and math and [the] space program, and I just found the whole thing so fascinating that I've never been interested in doing anything else.

Give me an overview of your career that led to this point. Tell me how you got here.

In the course of going through my schooling, I continued to pursue math and science with a space interest. So in high school I did some special programs at the University of Illinois to further my knowledge of computers and how to program. I ended up going to Purdue University, which, at the time, had more civilian astronauts and more military astronauts than any other school in the country except the military academies. And they also had a very strong program in engineering as they continue to do. Had some great opportunities there. Joined the co-op program at NASA which is a work study program that allows you to work, as you go to school, in the area that you plan to study as a career. And that was a great opportunity to see what the real space program is like and to know that I would indeed enjoy working at it in any capacity, not just as an astronaut, but as an engineer or a flight controller or many different opportunities in space at NASA. Once I finished my undergraduate curriculum, I worked here at NASA for a little while as a crew trainer. And I went back to school and got a masters and a Ph.D., came back here working for Orbital Sciences Corporation in a mission integration role, getting their rocket that's used to take satellites from low Earth orbit to higher orbits ready to fly. And then in 1990, I was selected as an astronaut for the American space program.

Now briefly summarize what you'll be doing during this flight. What are your primary responsibilities?

My role in this flight is Payload Commander, so preflight and continuing into the flight, I have overall responsibility for the crew interaction with the payload. Any issues that have a crew impact would be one that I would be coordinating the crew comments and issues with. For example, right now we're working some issues. We need to see some flight hardware to review it for stowage, and they need it for some testing, and what are some opportunities for the crew to see it that won't interfere with the testing. Those are some of the small detailed issues. Bigger issues are the crew procedures. What do we need to have in them? What kinds of situations do we need to cover? What level of detail needs to be in the procedures? The whole crew, of course, works these issues, but I'm the point of contact and the single collector of all the information and the interface to the rest of the payload team. Once we get into mission, I become more of a crewmember and less a Payload Commander although I still deal with overall issues. Then, mostly, I'm just a member of my shift. On this flight, the crew is divided in half. Half of the crew is sleeping when the other half is working, so we can do operations around the clock. So my shift has three people on it, and I am the payload lead on that shift. So any payload problems that we have, I'm supposed to be the authority onboard to know how to deal with [them]. Nominal operations, hopefully, are going to be the way most of the flight goes. And for nominal operations, once we get the payload set up and checked out, which takes about 12-15 hours, all we do is change tapes because all the data's being recorded onboard, and there's so much data that we need to change tapes roughly every hour. And so that's going to keep us pretty busy. And then monitoring the systems, the health. We're the eyes on the spot, and we have the first insight as to what's going on. So if we had to have a quick response, we could be there right on the spot to take care of it. And that's what all the payload people do. Of course the orbiter people watch the orbiter systems while all that is going on. We do have a little bit of interaction between orbiter and payload because we have some orbiter maneuvering that's being done to help keep the payload radiating in the right places on the Earth. And in those situations I'm working with the Pilot to make sure that all of the payload stuff gets done in support of those orbital maneuvers.

This mission will look back at Earth using radar interferometry to acquire topographic data. So tell me - what is radar interferometry?

It's a general concept. It's the same thing that you would know about interferometry here on the ground. It's two different beams looking at an object, and they interfere with each other, and they give you these fringes and patterns and things. And you can get information from the interaction of the two beams that you can't get from either beam individually. And we're all familiar with holograms. I have them on my credit card these days, and it's the same kind of thing. You get three-dimensional information from this kind of an image, whereas, with one beam, you get flat pictures. So it takes a flat picture, and it makes it three-dimensional. That's the same thing we're doing, conceptually, and I have a little model to show you how we're doing it. Here's how we get the two beams. If the Earth is in my lap ([the shuttle] actually, of course, flies upside down so that we can view the Earth), you see this big antenna in the payload bay. That's the main radiating antenna. It creates a single beam, but it goes down to Earth at an angle, and so it goes down and reflects back. And we have a mast that's 60 meters or 200 feet long [and] has a receiver on the end. So the main antenna in the bay is a receiver and a transmitter. The outboard antenna is a receiver only. So when it's facing the Earth, it radiates down and then both these antennas receive. That's how you get the two-beam pattern that allows you to get three-dimensional information out of the images as we go by.

You're doing all this work; tell me, why is this flight important? Is there more we can still learn about the surface of the Earth?

This flight is a mapping mission, and, as is true of mapping through the history of time, a better map allows you to understand and do things that you couldn't do before because you have better knowledge of what's there. This map will be a three dimensional map which is what makes it so different from previous flights of this hardware. It will allow us to map the surface of the Earth depending on what criteria you use for one sigma, three sigma which are all statistical things. It's something like 10 meters vertical and 30 meters horizontal resolution. That is the best resolution we have from any map at this point. And those maps that have that resolution only cover a small percentage of the surface of the Earth because it's very difficult to get that kind of data at that resolution. So we hope to map the entire landmass of the Earth between roughly 57 degrees south latitude and 60 degrees north latitude which is the tip of South America to part way up Canada. All of that landmass, which is 80 percent of the Earth's landmass, in [an] eleven-day mission at the highest resolution that any map we have has been made. So what we're getting is a much better three dimensional map than we've ever had of all that land before, and it will allow us to do a number of things. Ones that you can immediately envision are things like firefighters jumping into forests have a better understanding [of] where the mountains are and the valleys and the terrain. And [for] airplanes flying into airports, we have a better map that will allow us (a) to build better simulators that train pilots better and (b) to develop the ground systems that are used to guide an airplane during landing so it has a better knowledge of the mountains around various airports to help the airplanes make safer landings. And then there's a whole bunch of less obvious applications in the research area like water flow. If you want to know how rain falls down and how it gets into rivers and streams and how it affects our environment, you need to know where the mountains are. All kinds of land issues like that. How the crops are likely to grow because the elevation's changed. You can look at things like volcanic matter propagating after a volcano. [You can see] where it went, so there [are] all kinds of issues. To a certain extent you can do some things having to do with earthquakes, how things have moved and where they're going. The resolution isn't high enough to track individual movements but you can look at the bigger picture, over time, of larger scale movements. There [are] a lot of applications that will help us understand better how the Earth is changing, evolving because we have a map that's more accurate than we've ever had before.

You mentioned a little bit about resolution a while ago, but what kind of resolution do you expect and how would you compare this to other forms of spaceborne imaging?

There are several different kinds of spaceborne imaging, and they have different goals and different purposes. Right now there is no radar in space-based applications that can give you interferometric data on a single pass, which means two beams simultaneously. They have been able to do some of this work by having a satellite that goes around the Earth and then on a later pass gets a second look at the same landmass and then you can combine those two images in a similar way. But because they weren't taken at the same time, there's a lot more uncertainty about exactly what you're looking at, and it increases the errors. So you can't get as accurate a signal doing it that way as you can doing it with a single pass interferometer. Also different shuttle altitudes give you closer perspective and allow you to map the swath in a different way than a satellite that's up at, say, geosynchronous orbit. The shuttle on this flight will be at about 125 nautical miles, I don't know what that is in statute from the top of my head. The geosynchronous satellites are much farther up. They're at about 22,000 miles statute above the surface of the Earth. So they're much, much farther away and the image looks quite different. There [are] tradeoffs on power and resolution for those very different altitudes. It's also often true that the satellites up at those, because they're up in that position and there for a long time, [are] typically not just mapping satellites. They have different frequencies that look at infrared and visible to get crop usage data. So the primary purpose of those satellites is for other reasons, at least the civilian satellites that I know about.

You mentioned the mast here a couple of times. Could you maybe explain to me the process of deploying that mast? What happens? How long does it take?

Here, let me use my model here to show about the mast deploy process. This design of this mast is an amazingly clever mechanical design. This whole mast, which is 60 meters or 200 feet long, fits into this little silver canister. This is the setup in the payload bay. The main antenna and this little canister on the end [were] added to hold this mast. This whole mast accordions down inside this little canister and you're left in launch configuration. If you can imagine that this whole thing is squished down so you don't see it, it looks basically like this. In fact it's actually flipped over. In launch configuration, it looks like this with this whole thing smashed down. We have some of the motors that have the mast come out like this all the way out the side, and then we flip the antenna over this way. Of course, with my model I can only flip it this way, but on orbit, it would flip the other direction. And that's the deploy process. It takes about 17 minutes to get this whole thing out the side, and we have cameras that are mounted on the four corners of the payload bay that can look at this process. We plan to take this camera and focus it right down here where all the unfolding is happening. [There is] another camera here that can look at the bigger length of the mast and make sure the whole thing is coming out properly, and then we have backup cameras on these other sides. This camera is mostly blocked by the antenna, but, as you can see, this comes out at an angle, so once it gets a little bit clear of the payload bay, this camera back here can still see what's going on. And this one has a clear view because there's nothing in front of it. And this camera up here is actually a low light level camera so that we can watch this process in either daylight or darkness should we happen to have a transition as the mast is coming out. Hopefully that will all go very smoothly.

If it doesn't go smoothly, does the mast need to be fully deployed to be able to get acceptable science out of this? Can you get any sort of information at all if it only partially deploys?

The mast needs to be pretty much fully deployed to get good science because there are lots of processing bits that go into reconstructing this three-dimensional map of the Earth. One of those very important bits is where these antennas are. We talked earlier about the interferometric nature. We have a radiating beam coming off this antenna, and it reflects off the Earth. You get two reflective signals, and you get the height information [from] one signal going into this receiver and one signal in this receiver. In order to reconstruct that image, you have to very critically know what this angle is. So you have to know very, very accurately the position of the outboard antenna with respect to the inboard antenna. We have a set of sensors that are indicated by this little black and white box here on the center that include a Star Tracker which has been recalibrated to look at this outboard antenna, [which] has some lights on it that the Star Tracker [will] track. It uses that information to figure out the rotational position of this antenna, and we also have some lights that reflect off a little corner cube that's on the corner of this antenna that tells you exactly how long this distance is. So with that information from the Star Tracker and the electronic distance meters you can very accurately reconstruct this. However, in order to get high accuracy, these things are finely tuned. If the mast is only partly deployed, they are tuned to a range that the mast is not in, so they will not be able to measure the position of this outboard antenna in respect to the inboard antenna, which means you can't reconstruct the three- dimensional map. The radar data will be fine but you can't do anything with it because you don't have enough information to convert it back to a three-dimensional map. So we have to get the mast pretty much fully deployed in order to do science. That's the big picture issue. There's also the lesser issue that if it's not fully deployed, there's a reason, which means it possibly isn't mechanically sound. With this large mast, you can see it makes for a very large system altogether and there's enough weight out here that it actually doubles the rotary moment of inertia of this whole system. And we have to carefully design our attitude control system to handle that, and if this is in some intermediate position and it's loose in some sense, we wouldn't be able to safely maneuver either the shuttle or the whole antenna system. So there's two issues with not getting it fully deployed. So we just plan to get it fully deployed.

What is the process of mast retraction? Is it just the reverse of the deploy?

The mast retraction is basically a reversal of the deploy. The issues are a little different because the science has been completed, hopefully, so you're not worried about the end state. It just has to come back and unlatch. There's no tuning to be done when you finish. On the deploy, you get it out there, and then there's a whole bunch of tuning you do to get it set up. On the retraction, once you get it in, there's not tuning to be done so that is a much simpler and shorter process because there's less going on. On the other hand, it has to fit. Going out you're taking a small thing and making it big, and it's just expanding and that's easy. Going in you're taking a big thing and trying to make it small, and as we have all tried to do packing suitcases, that's a much more difficult problem. And so trying to get all that stuff back into the volume is going be potentially a challenge. They have worked lots of systems to make sure that that's going to happen smoothly. They have a heater so if these things get cold and don't want to fold properly, we can warm them up and make them fold more easily. They have a back up motor that has more torque than the primary motors, so, if it gets almost closed and it's just a little stiff, then they can put a little extra power on it and help get it in. They've tested all this on the ground, and they think all those things help. We have cameras that are watching the process, and, of course, we have mechanical latches and things that will help us tell if it's all safely latched in. But it is conceptually the reverse, the details are a little bit different.

You mentioned a couple of the potential problems and how you would deal with them there. If it came right down to it, could you jettison the mast, just get rid of it if you had to?

We do have a jettison capability for the mast. We want to always be safe. It's the top priority [of] the American space program, as well as all the other space programs in the world, to be safe in our operations. And so we have a jettison capability to cover any situation that we otherwise get stuck in. It's all operated from inboard in the crew compartment so we don't have to do a spacewalk in order to make that happen. We have pyrotechnic devices, explosive devices that'll cut the lines and the interfaces and just make the payload loose. Back to my model again, once this all folds up, the mast is in. The antenna is folded onto the canister so we have this little silver canister that holds all the mast, and the antenna will be in here. You can see that's pretty tight, the spacing there, so the way the jettison system works, it just cuts all interface and it's just loose. It doesn't go anywhere. It just sits there and then the orbiter will fly away from it, and you can jettison it in any position. Of course, if it gets stuck and you jettison it here, you don't have as much of a clearance problem. But now you have the problem that you've got an off-center mast that's going off, potentially, a little bit sideways, so you just have to be careful. But the same process applies. You cut all the interfaces here, and the orbiter flies away from it, and that's how we would jettison it if required.

Does this flight have any specific mapping targets on the surface of the Earth like the SRL missions did?

The previous SRL missions had these specific targets because they were doing the flat map and they had some particular science objectives. Like looking at the lava structure in Hawaii, where the flat map can tell you a lot about the texture of the Earth. Because this is a mapping mission, the goal of this flight is to map the surface. The entire surface is our target, not just specific areas that have high science interest. So while there are areas that some people are more interested in than others, we're going to map them all. So, we don't have any individual scientists working individual sites. They will get that data back when the data's all analyzed, but we don't need any help from them real time during the mission because we're going to map everything. So, unlike other missions, we only have the one customer, which is JPL, and we don't have individual scientists that we're working with for the procedures development. They'll get their data post-flight.

Is the radar on for the entire flight? Are you collecting data as you go over the oceans?

We do collect data over the water for two reasons. One, there [are] little islands all over the oceans, and the islands are important, too. So we map what might look like over the ocean, but we're actually mapping little islands in the ocean. The second thing is that they're using the oceans to calibrate the data. Since we're going to map anytime over land, basically [when] we map, [we're] going to go over an entire continent, and they use the water level at the beginning of the continent to reference the pass. On the previous missions, they had science sites of particular interest, and they put special corner cubes and things at the reference site. So, you get a good reflection from this baseline data system, and you know exactly where the corner cubes are so you can use that to reference the data on that site. In this mission, since we're mapping the entire continent, or whatever it is that you're mapping, we use the water on either end as a reference. So, we map a little bit of ocean on either side of the land to give us a zero baseline for doing the relative heights, too, and this radar's so sensitive that it's not quite that simple. You have to worry about tides because the water goes up and down enough with tides that it would change your baseline. So they measure the water, but then they adjust that by using a tidal model of the Earth and the exact shape of the Earth to get a really good baseline for the height measurements of the land.

Why is the data being recorded instead of downlinked live?

We are downlinking some data live, but we don't have the downlink bandwidth to downlink all of the data that we're recording because we're recording so much. We're recording a total of 270 megabits per second, which is a lot. We're doing 90 per second on one tape and 180 per second on another tape. The bandwidth we have to downlink data is only 45 megabits per second, so we can't downlink any of that entirely real time. We can slow the recorders down and play them back at half speed and quarter speed. So, when we're between land areas, we do some playbacks, [and] we get the data down somewhat delayed. And at that time we get all the data down [that] we recorded real time to check the antenna health. We do send down 45 megabits per second real time, which is half of the X-band data or a quarter of the C-band data. The X-band is the one at 90 and the C-band's at 180. And the way we get the quarters, on the X-band both wavelengths have an inboard and outboard antenna. So, the X-band has an inboard beam and an outboard beam, each at 45 [megabits]. So, you can look at either the inboard or the outboard real time, and then in playback you can look at both. For the C-band they have two polarizations, horizontal and vertical, and then they also have inboard and outboard. So, they have two, a horizontal and a vertical inboard, and they have the two that are horizontal and vertical outboard. So that's the four beams of C-band. You can look at any one of those four real time, but to see all four of them you look at the playback data some time later. And you have to see both an inboard and an outboard simultaneously to reconstruct a three dimensional image. You can check radar health, and you can get a flat map with one beam. But to get a three-dimensional map, you have to have at least two beams.

What kind of tapes are you using to collect the data?

We're using special high rate tapes. They look like the really old cassette tapes I [knew] when I was a child. [They are] really big clunky tapes because they record a lot of data, and it's very special tape that records at various speeds, depending on what the recorder's running. But the fastest we're using [is] the 180 megabits per second which requires tapes that don't stretch and that have a special way of writing on the tape so that they will record a lot of data on a small amount of tape.

Once you get all this information gathered, how long will it take to process it all?

They will process some of the data real time at relatively low resolution just to see if things seem to be working properly. They use that information to adjust the gains on the antenna [to] make sure it's tuned properly. They won't be able to start the high resolution processing until they get data back from the Global Positioning Satellites. The GPS satellites [are the ones] that a lot of people use on boats, and you can have them in your car these days. We're using that same data to get us a good reference for where the radar actually is. You have to know where it really is to figure out where it's measuring to. So, once you know where the radar is in inertial space or the space around the Earth, then, when you get the radar signal back which gives you a delta distance to the ground, you can figure the height of the ground. So they can't do the really high-resolution stuff until they get a high-resolution global positioning satellite set of data back post-flight. Then they start the processing. They expect it to take roughly a year to process the whole map. Of course, they'll start processing, and they'll have pieces of the map earlier. But to get the whole map will take about a year for the JPL side. Part of the people involved in this work is the National Imagery and Mapping Agency. They will be doing their own processing independently because they have somewhat different requirements on the data set, and we're not part of that effort. So, I don't know how long they'll be taking on their processing.

Are the X-band data and the C-band data processed in a similar manner?

The processing for X-band and C-band are similar. They're not identical because they're different wavelengths, and they're two different groups. The X-band data is being processed by Germany, and the C-band data is being processed by the Jet Propulsion Lab in California. But, of course, they have somewhat different software. The concepts are the same, and the basic processing is the same. The C-band folks at JPL [that] will be working with that data have four beams, and the X-band has two beams. So, again, that processing is somewhat different, but they will be conceptually the same.

If a volcano erupts or there's some other type of natural disaster while you are up there, will you be able to pay any special attention to it?

Not in any detail. If we have something like a volcano erupting, the problem is that our goal is to get a complete map, and in order to do that, we have to use every minute of data that we have to do the mapping. We have an hour and a half or two hours of extra time in the flight. So, we can't spend any time refocusing and changing directions, or we'll lose a piece of the map that we're planning on taking. So, it would obviously be of interest. They've asked us to take hand held camera pictures to document anything unusual because it may affect their data, and the investigators may be particularly interested in it post flight. But we won't be able to refocus or retune or repoint the radar to spend more time on those sites.

Once you know the payload is functioning, the crew slips into a regular routine changing out tapes. Do you think the work will ever get boring or tiring for you?

The repetitiveness of the flight certainly is something that we are aware of and trying to actively work to make sure that it doesn't become an issue. Not because it would become boring, but just because it makes you somewhat complacent. [When] you get in a rhythm and you think you know what you're doing, it's easy to miss things. Just like on the ground. You're driving along, on a highway stretch for hours and hours and not much traffic. You have to be very careful to keep your attention focused. So you do things like you turn on the radio, and you have something to drink. We do similar things in orbit. We're going to rotate responsibilities a little bit so you're doing something a little different. We have the tape change to be done. We have monitoring the system that's measuring the outboard with respect to the inboard antennas. We have orbiter general cleanliness and upkeep that we have to do. We have these flycast maneuvers that happen about once a day. We have zero doppler steering maneuvers, which are used to keep the antenna pointed along the velocity path, the speed path, and that happens every 45 minutes. There [are] enough different activities that you normally have different people doing that you can rotate some during the flight in order to keep everybody fresh. It is beyond my comprehension that a spaceflight could ever be boring. So boring is not going to be a problem. If you ever have a little bit of extra time and [a] little bit of need to change your perspective, you would just look out the window and freshen your mind up for a few seconds and come back to what you're doing. You can always do that, and the view is gorgeous. Because of what we're doing with mapping, the windows are pointed at the Earth. So we'll have views of the Earth all the time, and you can always use that to perk you up a little bit. Not, of course, ever distracting you from the primary mission of keeping track of the radar, but you can use that just as a few seconds of change between data tapes when the radar's not running to keep you fresh and remind yourself of what a truly spectacular space program we have.

Tell me about the relationship with the international partners on this flight.

In addition to working with JPL for the C-band radar, which is one of the wavelengths they're working with, we have Germany and Italy working on the X-band radar. So, it's been a challenge for them because of the time zone changes doing simulations and stuff. The two parts of the Earth are several hours apart in time zone, so we're trying to work to keep both teams, [the] JPL team and, of course, the folks here in Houston and [the] Germans and Italians about equally [focused] on time management. They will have a separate payload operations control center in Germany during the mission that will help monitor the data, and they have one at JPL in California. Those two teams are working very closely together, and I think it's been a great benefit to both teams. And, of course, it's wonderful to have this kind of information spread globally with several different countries taking interest in getting the data distributed and used.