Weather Forecasters Balance Experience with Technology
When people talk about a meteorologist cooking up a weather forecast, they may be more right than they realize, said one of the forecasters NASA counts on to predict conditions ahead of a launch.
"I compare forecasting a lot to cooking, to be honest," said Joel Tumbiolo, a meteorologist with the Air Force's 45th Weather Squadron, the unit that handles forecasting for rockets launched at the Eastern Range on the Atlantic Coast of the United States. "In cooking, you have recipes that you follow, but to be a good cook you have to have a certain taste and feel for it, and I feel there's a lot of that in weather forecasting."
The weather team monitors conditions from the ground level to a few thousand feet in the air, a region the rocket will fly through in a minute or two at most. But even a low-hanging cloud can be enough to call off a launch.
"If those couple minutes don't go right, bad things happen," Tumbiolo said. "You always wonder, 'How can a rocket going at that velocity be affected by a cloud?' But we've learned through trial and error that it does affect it."
The launch teams quickly learn the impact of weather on a countdown, said Omar Baez, launch director for NASA's Launch Services Program, or LSP.
"Weather is one of those things you never think about coming into the rocket business and you quickly learn how it affects our business," Baez said. "And it's not just during the launch phase."
Weather conditions dictate many of the activities around the launch site, not only the launches themselves. For instance, high winds can prevent crews from hoisting a spacecraft onto the top of a rocket. Thunderstorms can stop all activities on the launch pad. So getting a prediction wrong for even minor preparation work can result in a launch delay down the road.
Florida weather doesn't make it any easier on forecasters. From the thunderstorm that appears almost out of nowhere on a sunny afternoon to invisible winds thousands of feet up, the state's weather patterns offer plenty of seeming contradictions.
"In a recipe, if you have A, B, C and D, you get a certain result," Tumbiolo said. "In weather, you can have all the data that tells you something's going to happen and at the end of the day having something totally different happen. Not only does that challenge me, it interests me."
Learning to expect and predict frequent changes is perhaps the most important lesson. That is a significant departure from the conditions he saw growing up in the Midwest, where whatever conditions were to the west would reliably become the conditions to the east in a short time.
"Here, a lot of weather comes in off the ocean, of course," Tumbiolo said. "That was my biggest transition, getting my hands around the fact that weather comes in from all different directions depending on what kind of day we're having."
The key to deciphering changes is experience, Tumbiolo said. Still, the weather holds a few surprises.
"Sometimes things happen, and to be honest, you just don't know, 'Why did it happen?' But that's part of being a meteorologist."
Tumbiolo, who has been performing the job for 21 years, forecasts for about a dozen launches a year, including missions for LSP.
And, yes, weather forecasters keep score on how many predictions they get right.
"You always want to know that you're doing well or what you can improve, so, yeah, I keep a batting average. Over the past 21 years, I'd have to say my batting average is in the 80 to 85 percentile. If I can get over 80, I'm pretty pleased."
For Tumbiolo and the group of five weather officers, the payoff for a correct forecast is a spectacular rocket launching into the sky to begin a multimillion-dollar mission. The penalty for an inaccurate prediction can be dire.
"We have to forecast for a very specific time, a specific location," Tumbiolo said. "So we can't give a general, broad-brush (forecast), like, 'There's a 30 percent chance of showers today.' "
The meteorologists work from a set of rules that everyone must agree are "go" before a launch is allowed. Each rule covers a specific condition, such as the likelihood of lightning occurring during launch.
"We are evaluating rules, not just making subjective judgments," Tumbiolo said.
The good news is that the forecasters have a lot of technological help to show them everything from clouds, rain and humidity levels to wind high above the surface. From weather balloons to Doppler radar and sophisticated computer models, the forecasters aren't working alone to decipher the future.
"We probably have the densest network of weather instrumentation than any other place that I know of," Tumbiolo said.
Sometimes, though, forecasters want their own perspective. As a countdown moves toward zero, Tumbiolo makes his way to the roof of the Morrell Operations Center at Cape Canaveral Air Force Station. The view covers most of the sprawling base and the sky.
"To me, your best instrument is your eyeballs," Tumbiolo said.
There have been a few times when instruments were overruled by the forecasters. For example, radar picked up a small cloud ahead of an Atlas launch. The cloud was predicted to dissipate quickly. When it started growing, Tumbiolo went outside for a firsthand look.
"We have a rule called the 'good sense rule' where it's just that," Tumbiolo said, "If all the other rules are not in violation but it just doesn't look right to you, things are happening fast, or clouds are forming fast or it just doesn't feel good, we can invoke the rule and in all the time I've been here there's been maybe once or twice when we invoked that rule.
"The rules are a safety net. They're conservative, they're restrictive, and that's a good thing. You want that safety net. It gives you confidence."
That confidence can be especially welcome on those occasions when the rocket and spacecraft are ready, but the weather is not cooperating. It's up to the launch director to give a final "go" to liftoff, but Tumbiolo said he's never felt pressure from them to green-light a forecast just to get the mission started.
"Most of the launch directors are very weather-knowledgeable," Tumbiolo said. "They don't get on your case."
Baez said he trusts the forecasters to know their field and relies on them heavily.
"I have learned more than I had ever thought about weather, and I keep the literature handy," Baez said. "It's the one section in the launch console notebook that I dog ear and put tabs in to be able to reference it quickly."
There is one weather criteria the forecasters don't determine: upper-level winds. Instead, the data from the weather balloons and other instruments is sent to launch vehicle engineers who have specific computer models at hand that quickly simulate the launch of a specific rocket through specific conditions. If the computer says it is not safe, the engineers could call off the launch.
"It could be a picture-perfect, chamber of commerce day and we stand down due to upper level winds which we can't visibly see but have the potential of disturbing or tumbling a launch vehicle," Baez said.
Although he's been predicting the weather in the same place for more than two decades, Tumbiolo said he has no trouble getting motivated each day to do it some more.
"The weather involved in every launch is always different," Tumbiolo said. "There's always a different weather scenario involved, so to me that's always challenging and motivating."
NASA's John F. Kennedy Space Center