Martian Summer

by Andrew Kessler

What could possibly go wrong?

  “Please look at yestersol’s scraping activities carefully to be sure it’s a well-prepared surface. This might be our one go,” Vicky says, imploring the team to look closely at the new data when it comes in so we can optimize today’s plan. She too wants us to put this episode behind us and move on.

  We start to work through the day’s activities, examining each for possible conflicts or problems. They call this “scrubbing” the plan. These plans can be messy. This plan is no exception. There are conflicts, inconsistencies, and activities that still need to go through the validation and verification process. This collective editing process helps the SPI I catch errors before the plan gets to the approval stage when engineers can cut out anything they don’t like. Some notice that the RAC and SSI heaters are scheduled to run at the same time.

  “Can we do that?” an engineer asks.

  Then as downlink starts, there are more delays. Neither TEGA nor the RA unsafed. RA timed out yesterday after hitting hard material. And we won’t have the scrape test from yesterday to help us today.

  “Isn’t there a time-out block?” one of the systems engineers asks.

  It’s not validated yet. We can’t use it.

  For the engineering fans out there wondering why TEGA and the RA didn’t unsafe, here’s what happened: no one told Phoenix to tell the RA and TEGA to unsafe. We solved their underlying problems and it was time to turn back on; they just didn’t know. That’s the simple part. The reason for not telling them is more complicated.

  The “science master” is a file that tells the spacecraft when to wake up and go to sleep each day. Systems engineers carefully prepare these schedules in advance to make sure that this document is in sync with another important engineering file called the background master. It’s the background master’s job to tell Phoenix when the relay orbiters are in range. The background master then nags Phoenix to phone home. It’s mothering software. They uploaded an updated version at the last minute, but no one included the unsafe commands attached to it. Since Phoenix looked to this file to see what to do first, it didn’t know to unsafe the RA and TEGA before it tried to use them. Robots are so literal.

  The modicum of excitement present when we started this meeting bleeds from the room. It doesn’t seem like we will get our ice. New problems with the hastily constructed plan come to light. With a flight rule violation here, a lien against an instrument there, and a key activity forgotten, it seems like we need to temper our excitement.

  Things feel a little chaotic. Midpoint II is delayed. Twice. The tactical timeline starts to slip away.

  When we finally get started, Vicky hands things over to the SPI I. Things should move quickly now. Today’s SPI I is Mike McCurdy. He built the planning software, Phoenix Science Interface (PSI), that the SPI I uses to schedule Phoenix’s day. Yes, the SPI’s use PSI. NASA acronym humor. Good thing Mike is on duty. He’s our best hope to get back on track. We need someone to work quickly; it’s gonna be down to the wire. But before Mike can say a thing, PSI crashes, then the computer goes down. This is awkward.

  It becomes clear that there’s too much work and too few tired engineers to rush this TEGA delivery through today.

  “Perhaps we should consider eliminating the test?” Miles Smith asks the shift II lead and the shift II mission manager.

  “You’re going risk pushing the TEGA delivery back by days if you don’t do it now,” Heather Enos says with exasperation.

  Joel Krajewski steps in.

  A few weeks ago, when I was about to leave the SOC after a 16-hour day, I passed Joel Krajewski in the hall. The SOC was still full of scientists and engineers. I said I didn’t know how they do it so many nights in a row.

  “Crushing guilt,” he said. People stay all night to make sure things are perfect because the guilt that someone would feel from a mistake that ruins the next day destroys them from the inside.

  I guess fifty days is a maximum for lander guilt. There are still many late nights, but with so many new problems it’s just not possible to stay a few extra hours and triple-check your work. The guilt-transport-system slows. Guilt, excitement, or anything but a deep passion for sleep are muted by this point in the mission.

  It’s becoming clear that at least from a psychological perspective, Nilton Renno and Carol Stoker were right. We should have scooped ice from Dodo-Goldilocks weeks ago and moved on.

  Joel tries to clarify what’s at stake to head off a fight. A good mission manager lays out the consequences for taking each path and then lets the stakeholders decide.

  “There are two critical paths running in parallel, RA and TEGA,” Joel says. Both need attention in order to move ahead in a timely fashion. He explains that if rasping works well, we need to be ready with TEGA, which requires a one-day gap between opening the door and delivery. Dropping the monster activity will cause a delay. Not getting it done properly will too.

  Everyone agrees: let’s push ahead with the plan. Team members vote with their 80% confidence in the plan.

  THE 50S ARE NOT A GOOD TIME FOR THE MISSION. EACH DAY BEGINS with a new hope that’s quickly dashed. The test on sol 53 is a moderate success. They manage to open the TEGA doors, only to discover a problem with a frozen valve. The valve is fouling up the internal pressure in TEGA. Then a problem with the load plate at the back of the rasp pushes back the next drilling exercise. Preheating and validating code goes slower than expected. Waiting is not interesting.

  CHAPTER TWENTY

  MARTIAN COLDS

  SOL 57

  ON THE FRONT PAGE OF TODAY’S TUCSON CITIZEN APPEARS A picture of George W. Bush waving to adoring fans in Tucson, Arizona. The President of the United States comes to the desert; he probably wants to see firsthand the amazing discovery made by Phoenix. Nope, he doesn’t stop by Mission Control. Doesn’t he know there’s a Mars mission happening here? Maybe he wasn’t invited. Maybe he doesn’t care.

  RA team stayed up late looking at yesterday’s rasp test. They seem frustrated. I even think I hear one of them curse. No, he said “shoot.” Sorry for the misunderstanding. Something has changed. Where is the happy-go-lucky, get-the-sample-at-all-costs team I know and love? It seems that the RA team has a visitor: contempt.

  “They still don’t get just how hard this arm is to operate,” angry RA engineer #1 says. They’re worried I’m going to write about their anger, so I’ll protect the identities of the cranky.

  “This is not like the arm on MER! How many times can you say it!? This arm is two times longer and doesn’t have contact sensors. And the DEMS are so noisy. What are we supposed to do with those? You try to measure distances in the images and run them through the compliance models to understand the flex,” overwhelmed RA engineer #2 says.

  Having never used a compliance model, not once in my entire life, it’s hard to empathize. Still, it’s good to vent their frustration. They’re frustrated because they have to precisely position the load plate at the back of the scoop. They spend a lot of time on this placement. There’s a lot of pressure to make these precise movements, quickly and without all the information they need. Every day they don’t deliver ice just makes the feelings of frustration grow.

  And I’m not the only one who noticed the tension. The mission managers mandated a pre-delivery down day for the RA to make their final calculations. The day before delivery, the RA team will spend their day calculating and coding only. And by official order are excused from any other duties.

  JOEL KRAJEWSKI CALLED IN SICK YESTERDAY. PETER HAS A COLD. And there are ten or so other sneezers blasting germs into the SOC atmosphere. Mars lag is slowly killing us all. It’s time to consult Edna Fiedler from the counter-fatigue group.

  “I’m glad to see you,” Edna says.

  Before I can start complaining, she says there’s something she’d like to speak to me about.

  Me?

  “Yes.”

  Then she waits for me to talk. This is an awkward exchange. It seems my basic so
cial skills are breaking down too. We both just stand there.

  “Don’t you have some questions about how you’re feeling?” she asks.

  It’s like she can read my mind. I have questions about how everyone feels. She says she can’t tell me about the others because that’s confidential, but I can ask her about myself. And then, we agree, it’s all about me anyway.

  “There are probably changes in your T-cell count happening,” she says. Oh dear. My immune system is not working efficiently. I knew it.

  “There are possible pulmonary problems and the liver too,” she adds. “Your endocrine system is slugging along. The function hasn’t stopped, but it’s sluggish. You’re not moving hormones.” All this is due to the slow but constantly debilitating sol schedule.

  “So what’s the good news?” I ask.

  They’re really learning a lot for future missions. So Edna is pleased. Now that I know what’s wrong with me and why I feel this way, it’s her turn.

  “Have you ever heard of astrosociology?” Edna asks. I haven’t. Great, one more thing I’m supposed to know but I don’t.

  “‘It’s the sociology of space travel and other planets,” she says. “We’re having a conference and we’d like you to speak about your time here. It takes place in Huntsville, Alabama.” I play coy.

  “Oh, that sounds interesting,” I say. But really it’s the most awesome thing ever. Suddenly my symptoms don’t feel so overwhelming. Fiedler says she’ll email me the details right away. How can I resist? She goes to lunch. I Google astrosociology and start checking airfare to Alabama. I wonder what I’d talk about. Maybe the bags of pee everyone always carries around Mission Control would make for a nice opener.

  IT’S MORNING IN TUCSON, EARLY EVENING ON MARS. THE START OF SHIFT I nearly syncs with a regular Earth day. For a few glorious days we’ll come into the SOC at seven, eight, nine, and then ten o’clock in the morning. Even if our bodies are getting worse, there’s a huge psychological benefit to waking with the sun. The last week of coming in at three, four, and five in the morning was especially demotivating. And that’s on top of the continued hiccups with TEGA and the other elements of the dig.

  It’s nearly time for midpoint. Scientists mill about before taking their seats. Bill Boynton grabs the microphone.

  “Everyone needs a balloon for the meeting,” Bill says.

  Heather comes out of the TEGA office hugging an armload of balloons.

  “Please, you must have a balloon for midpoint,” Bill repeats. And he goes to help Heather pass them out.

  Three minutes before midpoint begins, everyone has a balloon. The balloons come with smiles and a little confusion. We look at each other and wonder what wondrous discovery warrants this great balloon giveaway. Anything to get us through these doldrums is appreciated. There has not been a lot of good news in the last few days, but balloons are a universal indicator something great is going to happen.

  “We’re having a prickly-ass contest,” Heather says. “Everyone will sit on their balloon and whoever pops the balloon first, wins. Get ready … Go!”

  There’s no time for questions. An overzealous engineer goes splat when his balloon rejects his overture. We race to pop our balloons. Popping and laughing ensues.

  But that’s it. There’s no news. Just a random balloon-popping contest to keep us from bursting ourselves. Well timed. We all win.

  The fun is over. Doug Ming starts midpoint.

  It’s been a slow and monotonous few days. The most exciting event was a giant monsoon that collapsed the roof of the small conference room. This is where we hold the end-of-sol science and planning meetings. Not any more. Did NASA send the storm to put a little more pressure on us? I’m not at liberty to say.

  TODAY, TEGA IS FINALLY UNSAFED. THE PROBLEM WITH THE VALVE IS under control. The pre-bake checkout looks good, and so does everything else. That’s good news. All the big hitters from NASA and JPL are here. Ramon, Barry, and Dara are back. The President is somewhere in Tucson. And the instruments are nearly ready. Let’s do this. We could get past this NASA imposition any day now.

  “No data gaps or errors,” reports a TEGA engineer. TEGA is ready to accept a sample. As soon as NASA says “Go,” the TEGAns will open their doors to make way for glory.

  Apart from the curmudgeonly RA engineers, things are going great at chez RA. They just tried their autonomous fault recovery. It’s a scheme where the RA times out but does not go into safe mode. Fault recovery is a bit of code that lets the RA stare a problem in the face and then instead of backing down, it confronts the problem and gets back to work. This only works for a few situations, but it should save a lot of time now and potentially prevent some future slowdowns.

  THE END-OF-SOL SCIENCE MEETING IS IN DOWNLINK. BEFORE WE GET started, Pat Woida comes in with a big smile.

  “Someone made lunch for the whole team,” he says. A local woman, impressed that a real space mission was happening in her Tucson neighborhood, decided to cook authentic Sonoran food for one hundred people. There are tamales, stewed meats, vegetables, and salads. A small woman with an enormous smile stands near the feast and smiles at the convoy of hungry engineers. I make myself a towering plate of corn-husk-encased goodness and stewed meats. I get so lost in my plate that I fail to ask her name or why she fed us. I’m a good eater but a terrible reporter. Thank you, mystery woman; it was delicious.

  CHAPTER TWENTY-ONE

  THERE IS NO TRY

  SOL 59

  “TODAY IS A BIG DAY,” DOUG MING SAYS, AGAIN, TO START kickoff. “We will do our 16 rasping holes and acquire the sample. Then deliver and TEGA low bake. There will be little else.”

  The plan will go something like this: 5:00 a.m., wake up. Preheat RA. Rasping. Scraping. Preheat TEGA. Deliver sample.

  The downlink priorities are all rearranged. The table that controls them gets a special update—tested carefully—to ensure that the TEGA data comes down with the highest priority tag, its APID. The engineers moved it above some of the most critical engineering data.

  The weather report doesn’t bode well for us. There are storms on Mars! And for the first time, the team sees ice clouds brewing overhead.

  “Wow, the weather will be changing!” Doug says. It’s autumn on Mars.

  Everything is double-checked. The storms pose no threat, but we need to be extra-careful today.

  “Measure twice. Cut once,” an engineer says, passing some colleagues rechecking the plan.

  Matt Robinson feels good about the load plate test results.

  “We’re set to give it our best shot. But a lot of things have to go right. sixteen rasps give you two cc’s of material,” Matt says. “I sure hope that’s enough.” The TEGA team has reset light/dark parameters on the LED for the oven-full signal. They’re gaming the system a bit to give them an oven-full with a smaller-than-usual quantity. Don’t tell Ramon.

  Midpoint arrives and it’s time to decide “go” or “no go.” The teams report on their general status. RAC is healthy and ready.

  “TEGA is healthy and clean, but ready to get dirty. There is no evidence of an electrical short,” a TEGA engineer says.

  “I looked at the RAC images,” Doug says. There are no problems. He’s ready to declare it a “go.”

  All that is left for the SPI I to explain the plan to the shift II team. They’re relatively happy. But there are still lingering concerns. Questions and answers come in short staccato.

  “What happens if MRO fails?”

  “The data is protected.”

  “Memory concerns?”

  “Critical data will be written to flash … with plenty of margin.”

  There are opportunities to re-transmit. Everyone gets their questions answered. The core plan gets the APID “0”: highest priority to be transmitted from MRO. What about everything else? There is nothing else.

  “Please cross your fingers and toes,” Doug says.

  “And pray to the ice gods,” Bill adds.

  Meeting
dismissed.

  “THIS IS IT FOR ME,” DARA SABAHI SAYS. HE CAME BACK TO HELP BRING TEGA into the proper delivery position and offer some guidance to the team. “I’ve done everything I can.” Now he’s leaving for good. He doesn’t know when and if he’ll return. Dara says he’s not worried about the team getting ice.

  “If it doesn’t work and it’s an engineering issue, then it can be addressed. If it doesn’t work because of Mars uncertainty, we can learn something and try to work around it. Either way, we can address it. The only time we don’t succeed is when we don’t try,” he says. It’s tough to argue with his life-affirming view of the world.

  “Remember, this is about managing many points of risk. You must balance them but you can’t run from them,” Dara says as he passes the baton to his team.

  “I will,” I think to myself. Maybe I’m internalizing.

  “Our clients can’t go to the mechanic if something goes wrong.” Dara is partial to automotive engineering metaphors.

  “We have a prototype that has to work perfectly, the first on the road,” Dara says. “You can do that, but it involves managing a lot of risk. And you might imagine that it’s a technical challenge. And that technical challenge is complicated. But on these big projects, it’s not as complicated as cooperation.”

  It’s rare that engineers play down technical aspects for those of basic human interactions which are vastly more difficult to plan and almost impossible to predict. I guess that what makes Dara a universally praised member of this team. He sees space problems in terms of human interactions.

  “The little things matter. Mutual respect matters. It goes a long way if you want to be successful.” With that, Dara says good-bye.

  He put his heart and everything else he had into getting Phoenix on the ground and scooping ice. And now he’s leaving. It’s time for him to get away. Maybe for good.

  “I’m going to get in my car and drive. No computers. No email. No radio. No people. No politics,” he says. “You can try to contact me. Just don’t count on a response,” and then smiles sweetly.