Transcript of the Interview with Dr. Louis Uccellini, Director of the National Weather Service, 27 February 2013


SSEC: Many influential names in meteorology came out of the UW-Madison in the 1970s. Talk a little bit about mentors during that time, including Don Johnson, your major professor, whose students along with yourself included Richard Anthes [former UCAR president], William Downey [AD, Australian Bureau of Meteorology], Ronald Townsend [President/CEO, Oakridge Associated Universities and Air Force Institute of Technology], Carlyle Wash [Head, Department of Meteorology, Naval Postgraduate School], John Zilman [former director, Australian Bureau of Meteorology and President, WMO]. It’s a pretty heady crowd — what it was like, the intellectual energy swirling around with your contemporaries?

Louis Uccellini (LU): I believe I was in the department at the right time. I was there as a student and post doc from 1968 to 1978. I received all three of my degrees from the University of Wisconsin, my bachelor’s in ’71, master’s in ’72 and PhD in ’77. And I did a post doc for one year under Vern Suomi and Don Johnson in 1978. The intellectual atmosphere at the university at the time was incredible, whether it was related to the meteorology department or what was going on outside of the department associated with the anti-war demonstrations during that period. Lyle Horn, who was one of the most important mentors I had, felt that we as students asked more questions during that period which made the classroom that much more exciting. And I had to agree with him. It was the exchange that occurred in the classroom that was really unbelievable. During that time period I think there were a lot more exchanges between students and professors and it certainly helped me in terms of my learning experience.

As for important people, the first one was Charlie Anderson who was my master’s advisor. I didn’t know I was going into graduate school until he tapped me on the shoulder one day and said you know, you do really well in asking questions and solving problems; you ought to go into research. You ought to go to graduate school and by the way, don’t bother applying any place else; you’re coming here to Wisconsin. That’s the way Charlie operated. So that’s all I did; I only applied to Wisconsin and he had it all set up for me. I switched over to Don Johnson who was my PhD advisor. He’s an incredible man. He is, I think, one of the top intellectual forces in the country when it comes to not only meteorology but the whole earth system. I think Hank [Revercomb] would agree with that. Working under him for a PhD was a major plus for me.

I had many opportunities to work with Vern Suomi even though I wasn’t his student. He wasn’t on any of my committees. I actually shared an office on the tenth floor and got to know him fairly well and he got to know me. He and Don set up a post doc for me a year after I received my PhD which allowed me to work on papers that were coming out of my PhD and that’s what I did.

When I decided to go to Goddard Space Flight Center from Wisconsin, I know Vern wanted me to stay at Wisconsin. He tried to entice me to stay but when I told him that I was going, and other professors had told me I needed to go, he called me into his office. Instead of getting annoyed at me for going, he spent a full hour explaining how to operate within a federal laboratory and make the best of it. I took his advice. Obviously I think I’ve done very well. There were incredible people there and I’m eternally grateful for what I gained from my experience at the University of Wisconsin.

SSEC: In addition to your connections with the meteorology department, you also worked with scientists associated with the Space Science and Engineering Center. You came up at a time when earth observation from a satellite platform was developing quite rapidly – with its great beginnings led by Verner Suomi. Tell us about that time and those connections, especially in the 1970s working on severe and convective storms with Johnson and Schlesinger; and then in the 1980s working on the VAS [VISSR Atmospheric Sounder] with Ralph Petersen, Dennis Chesters, Anthony Mostek, Bill Smith and others.

LU: I had interactions with Vern Suomi even as I was getting my PhD degree. He had two credit courses that he would bring outside speakers to the University of Wisconsin and I found that really fascinating. It was through one of those courses that I learned about the GARP experiment and SSEC’s role and all the tropical work they were doing, including Don Wylie’s and Fred Mosher’s work with the Boundary Layer Instrument System that they developed.

With that kind of ongoing interaction, Vern tapped me, (I guess he was working on a proposal for a geostationary sounder) and other students within one of these two credit seminar courses to help develop the proposal for specific satellite weather applications. I worked on the section that relayed how we would use more rapid soundings of temperature and moisture, for the analysis and prediction of severe convective storms. I don’t know if it was the first proposal for a geostationary sounder, but it was one of the proposals he submitted. That was my first real connection between meteorological applications and satellites.

One of the things that we talked about in the meeting I had with Suomi before I went to Goddard Space Flight Center was the importance of how to work within the management structure that the government can put in the way of getting things done — what to focus on in terms of lower earth orbiting satellites for numerical model applications — but what he was really interested in, of course, was the geostationary satellites. He was interested in everything, but at times he was really pushing for the geostationary sounders and how they would be used. I took that advice with the VISSR Atmospheric Sounder (VAS) Program. I became the VAS Demonstration Scientist at Goddard Space Flight Center and had to oversee what was going on at Wisconsin and Marshall Space Flight Center.

I continued my working relationships with people like Paul Menzel and Bill Smith and others at Wisconsin.

The work that Suomi had me involved in as he was starting at the University of Wisconsin certainly carried over into my career at Goddard Space Flight Center in a meaningful way. The VAS was considered a very successful project. Furthermore, the VAS sounder was actually launched as an operational sounder and has continued right up to the current GOES system — the geostationary satellites that are being flown today still have the sounding capability. It all came out of the VAS project that started at the University of Wisconsin and continued at Goddard Space Flight Center. I always felt very proud about that. And still do.



SSEC: Every decade, we tend to analyze the progress we’ve made in the atmospheric sciences. In the 1980s, under the auspices of the Board on Atmospheric Sciences and Climate, Charles Hosler led a review of “ major scientific problems, research objectives, experimental programs and policy issues in the atmospheric sciences for the remainder of the 20th century.“ You wrote a detailed response to Hosler’s call for comment, outlining scientific objectives and policy issues that should be addressed in the coming decades. Can you talk a little about your vision then and your vision now for the future of the discipline?

LU: That’s interesting. I had forgotten completely about this until I saw the letter this morning. There are two aspects of this that I think are still incredibly important today. One has to do with the interdisciplinary nature of the problems involving the atmosphere, ocean, land, hydrology, cryosphere, biology, and chemistry. We’re finding how important this is, not only in terms of increasing our understanding of the atmospheric circulation system and how it works. We also know that to make predictions, it’s not just the weather anymore. It’s predicting what happens with the rainfall, the runoff of water into a bay, to nutrient loading, to the chemical makeup being changed in a bay that impacts the sea life. All of this is connected and we now understand it. Furthermore, we have the ability to predict this synergistic relationship and the related feedback. It isn’t that you can pick out these individual components and think you’ve got the answer or the abilities to make a prediction; you have to put them together because they feed back on each other in ways that are incredibly important.

This was something that Francis Bretherton, Suomi’s successor, pointed to in his seminal document on the Earth System Science. He pointed to all these interactions. I think what I was trying to do with this letter in terms of atmospheric research, was to get people to understand that it’s not just the atmosphere anymore and it doesn’t just apply to climate. It relates to our understanding of and ability to predict the weather and the consequences for people who have to react to those weather regimes. That’s what I was trying to convey.

The other thing has to do with the experimental programs — whether I was pointing to a specific research issue or how to design experimental programs, I was trying to bring the research and operations together at the same time and deal with them in parallel rather than in series. People are still having trouble with this research to operations concept. It often takes too long to affect a transition from research results into operations or observation improvements into operations. The idea is that if the research and operational communities are working in parallel during the development phase or during these field programs, it should be easier to then make the results from that effort readily available within an operational environment. In a real sense, VAS was like that.

But going back to the question — in many areas, in other parts of the sciences, especially into the ‘80s and ‘90s — the research folks wanted to work by themselves and then deal with the operational community. What I was trying to get across whether it was forecasting or nowcasting what it really comes down to is that if you have an interest in affecting the transition from research to operations, these folks should work together right from the beginning.



SSEC: The National Weather Service has a record of increasing success in forecasting and warning. What do you consider to be your greatest challenge in your new role as head of the nation’s National Weather Service?

LU: I have lots of challenges. That’s what I’m finding out. And recall that I’ve only been sitting in this seat for three weeks.

One thing is that we don’t take enough time to recognize our successes. I think people who have been involved in the research and the operational communities over the same length of time that I’ve been involved, starting at the University of Wisconsin in 1968, have seen tremendous improvement in the ability to forecast extreme events further out in time.

When I was a student at the University of Wisconsin, I spent a lot of time on the fourteenth floor in the fax room, where everything was paper copy, everything was analog, nothing was digital. We were just starting to see the numerical models coming out operationally that actually predicted rainfall. Those models went out to twenty four hours. I, along with researchers who were coming up at the same time, thought we might be able to produce a thirty-six or forty-eight hour credible forecast of a weather producing event such as a rain storm, a snow storm, a severe weather outbreak. If we could get the forecast out to forty-eight hours we should declare it a success.

Well, today we’re making forecasts for extreme events out to seven, eight days in advance with enough credibility in some cases like Sandy. Some of the severe weather outbreaks in 2011, a major snow storm, like Snowmageddon, were all predicted seven days in advance. People are now starting to expect us to do this all the time.

One of the biggest challenges really is to manage expectations because there are still predictability issues — some systems tend to be more predictable with more consistency as you approach an event than others. That’s still a challenge — we want to capitalize on this ability but we also have to be able to deal with the uncertainty of the event so that decision makers are kept informed. They need to know when it’s time to make a decision to move forward with an evacuation, for example, and that they have enough confidence in our forecasts that it’s a worthwhile thing to do. Dealing with the uncertainty in the forecast for extreme events is still a major issue.

How we communicate the forecast and the warnings is still a very critical problem. We have as a strategic goal for the entire country of making the United States a Weather Ready Nation so that everyone can take the proper action with respect to an impending storm and not wait too late, or ignore it and then get themselves in trouble in a life threatening situation. This means that the way we communicate a forecast, the watches and warnings, really need to be looked at. We still have situations in which the message we give is not equal to the message received. There’s a lot of work that needs to go on if we want to take this to the point where people will actually take the action you expect.

From a purely science point of view, I always say there are three components to a successful forecast system. 1) There is the global observing system in which Wisconsin has played a huge role. 2) There is the data assimilation and modeling effort which involves a lot of science and there’s still a lot of science to be done — I mentioned the interdisciplinary nature of the work, the earth system approach where you’re dealing with the entire earth system and not just the atmosphere. 3) You need the computing capacity to make this all happen.

There are tremendous challenges in each one of these components. I think Wisconsin is certainly in a position to sustain its leadership role in earth observing issues. With respect to data assimilation and modeling, I know there’s work being done at Wisconsin but that’s a huge issue that involves scientists from all over the world. It’s the interaction of those scientists, by the way, that helps us make breakthroughs in the predictions we talked about earlier.

Then, of course, computers. The U.S. computer industry and several laboratories are making great strides. We’re looking at new approaches in computational sciences that will allow us to predict with much higher resolution and much faster with more accuracy so we can get very specific forecasts out in a timely way, down to maybe street level in ten years or so. We have to work with the computing industry to produce computers that will enable us to do that.



SSEC: You alluded to Hurricane Sandy earlier. Are there lessons to be learned from recent storms that have battered the east coast such as Hurricane Sandy or the recent New England snowstorm? Are these unusual events? Or was there anything, since you are a native of Long Island, anything that struck you about the recent snow in New England?

LU: Well, the thing that struck me about the recent snow storm in New England was the irony that the announcement for me becoming the Director of the National Weather Service, coincided exactly with the forecast and warning of this major New England snowstorm — a Northeast snowstorm which will make it into the third volume of the book that I wrote with Paul Kocin [Northeast Snowstorms, v.1 and v.2. Boston, MA, American Meteorological Society, 2004.] The irony and juxtaposition of events was certainly noteworthy to me.

The lessons learned — let’s go back to the severe weather outbreaks in 2011. Several times in 2011, we made a forecast for the potential for an extreme outbreak of severe weather for the southeast and south central part of the country seven, eight days in advance. As we approached those events we refined our outlooks. We provided the risk assessments two, three days in advance which pointed to a major outbreak. We issued watches hours in advance. We issued warnings twenty, thirty minutes in advance for a number of these cases. Still, hundreds of people died.

We had a seminal meeting in Norman, Oklahoma in December of 2011 with the social science community, the first responders, and the emergency management community. We realize that we have a lot of work to do in communicating warnings and getting the message out. People need to deal with decision support systems in a much more effective way because there are things that happen or messages that we provided which were not received the way we intended them to be received. What we take away is that we have to better communicate the potential impact of these storms. We actually used that communication strategy for Sandy and this impending snow storm, while it was an impending snowstorm, the major New England blizzard.

Sandy was a very complicated system and people were discussing whether it would be a hurricane or whether it would be a post tropical storm when it made landfall. What we focused on were not the characteristics of the storm — whether it was tropical or not when it made landfall — but the impacts. We had a very large and dangerous storm so we talked about the impacts of the very strong winds over a very large area from New England all the way to the Carolinas that would last for a long time, beyond twenty four hours.

Long Island, for example, had gale force to near hurricane force winds for up to thirty six hours that would compound the impact of the storm. You would have much more tree damage, many more downed trees. It hardly rained on eastern Long Island, yet they had near hurricane force winds for close to thirty six hours. People talked about this later. We emphasized the high winds right down the coast, the surge that would last through two tidal cycles. This would be a historic surge for New York City and northern New Jersey that would come in two events, not just one and the impact of the second one would be stronger than impact of the first.

Three days in advance, we emphasized the dangerous blizzard in West Virginia that would dump three feet of snow and the impact there was compounded because the area was still recovering from the severe storm they had on June 30, the derecho. Tarps were still on many roofs and Sandy eventually put three feet of very wet, heavy snow on top of these structures.

The impact on people in West Virginia and the recovery teams had to be emphasized. Furthermore, we stressed that this storm would extend all the way to Lake Michigan and that the damaging winds and wind driven surges in Lake Erie and Lake Michigan would yield life-threatening situations. Instead of just saying we had a hurricane that was going to strike, or a post-tropical storm that was going to strike the New Jersey coast which would have just gotten everybody focused on the Jersey coast, we dealt with the extent and the magnitude of the impact of this system. From the feedback we were getting, that strategy was successful. Those were lessons learned from this severe weather event.

With the snow storm we just had in New England, the key thing was the change over from rain to snow in southern New England, Long Island and New York City right before the evening rush hour. We conveyed to people, look we know it’s raining here at lunch hour but your evening rush hour is going to be impacted because the rain is going to switch over to very heavy snow. That’s exactly what happened. People who listened and left early didn’t get caught in traffic. The people who said, we’ll wait to see if it snows or not, got caught. There were people trapped on the expressway for many, many hours.

The idea that they didn’t believe what we were saying means we’ve still got work to do. But there were a lot of people who listened to us and got home before three o’clock that afternoon, even though it was raining. We’re really looking at this impact-based messaging as part of our weather ready nation vision for the country.



SSEC: In 2010 you were among a number of authors on a paper titled “An Earth-system prediction initiative for the twenty-first century” [Bulletin of the American Meteorological Society, v.91, no.10, 2010, pp1377-1388.] This paper involves the ideas of shared earth observation and information systems across disciplines. What do you see as the great challenge of satellite observations in the near term? In the longer term?

LU: One of the three major components of the whole forecast enterprise is the global observing system. And satellites, satellite observations, especially the lower earth orbiting satellites today provide the backbone of that observing system. One of Suomi’s visions for the GARP experiment I was talking about earlier and the FGGE experiment that followed, was that he was trying for weather forecasters to prove that we need satellite data and we can produce it in real time. The Space Science and Engineering Center proved that this necessary satellite data could be produced in real time so it should become a fundamental part of the real time observing system supporting numerical prediction models. Suomi was absolutely right. That is still true today. We’re still not optimally using all of the satellite data, whether it’s microwave or infrared. There is still a lot of science that’s needed, especially in the better use of infrared, the cloud cleared radiances and the like.

We now have hyperspectral resolution, from the infrared. We have many hundreds of channels that are being used today versus ten to twelve or fourteen or nineteen. We were successful in getting that data into our models but there is still a lot of work to be done, to ensure the optimal use of that data.

I think one of the biggest challenges that remains is advancing the use of geostationary data. Right now we will not have hyperspectral sounding data from geostationary orbit. Nevertheless, there will be more data coming from the GOES-R series. For example, GOES-R will have 16 imagery channels. This data is not hyperspectral, but we’re really going to receive the data over a large area over a short period of time, every fifteen minutes we’ll get a full view of the disk.

This is an incredible amount of data and if we’re updating that every fifteen minutes and we have a data assimilation system that can handle the increasing temporal resolution, I contend we’re going to get more information from that data for use in numerical models. That’s a challenge. That’s a challenge worth having as other countries launch geostationary sounders. The United States has to be prepared to use that data to show how important it really is. From an operational perspective I intend to do that.

The other opportunity, I believe, is the GPS RO [GPS Radio Occultation] the COSMIC-like data, a radio occultation data stream that allows us to extract temperature and humidity, moisture information with very high vertical resolution. If we combine that with data from the infrared and microwave instruments, I think we’re going to get even more accurate information to initialize our models. There’s a lot of work that’s going on in that area and we’re part of it. We have to continue this work because I think the rewards are going to be great over the next two to three years and beyond.

SSEC: The next part of that question — over the years there have been great successes in the collaboration among nations to share satellite data. People like Suomi believed that data should be available. Do you see this global collaboration continuing?

LU: Yes, the collaboration is continuing in the sharing of data among many countries. Sometimes it gets a little difficult if, there is fighting going on. The United States and Europe and parts of Asia and Canada are getting more active in launching satellite systems – these countries do exchange data. We rely on them, especially on the Europeans and their satellite systems. So, again, Suomi’s vision was right on the button. You need a free flow of information and data if you want to support this forecasting enterprise.

Remember I said it was the global observing system that’s the key part of this, the third leg. It’s not just the United States taking those observations. We depend upon the agreements with our World Meteorological Organization and the member states, to help provide all that data so that we can run our models and get the forecast out.



SSEC: The GAO, in a report just released this February, identified and added two new high risk areas to its annual list: 1) Limiting the federal government’s fiscal exposure by better managing climate change risks, and 2) mitigating gaps in weather satellite data. If you would talk a little bit about the relationship between climate and weather risks and the role of the National Weather Service in helping to reduce exposure to these risks.

LU: I think there is increasing evidence that the climate change aspect, which I know is controversial, but there are certain aspects of it that are becoming a very real that factor in to risk associated with extreme storms, extreme weather, major storm systems. I can cite two areas.

One area involves the warming atmosphere, which holds more water vapor. When latent heat is released as rain or snow, that heat release helps to fuel the storms. There is a hypothesis that there will be more intense storms related to that release of latent heat and it’s not just tropical storms, it’s extratropical storms, too. It’s the winter storms that we’re just going through. There is evidence to suggest that this increased warming and related increased water vapor is playing a role in more extreme storms.

We’re seeing more cases of extreme events and heavier precipitation associated with those events. But we don’t quite have enough cases yet to really nail that down and there’s a lot of research being done in that area. Some people are more sure about it than others, but the fact is, we need to pay attention because if we’re going to see more extreme events obviously we have to do a better job in forecasting those events so people can do a better job in responding to them and mitigating the potential impact and protecting their lives — whether it’s a major blizzard like we had in Texas and Oklahoma this past week or a hurricane or severe weather outbreaks.

The other aspect is rising sea levels. What we saw in Sandy and what we emphasized for New York City and northern New Jersey, were several unique aspects that were going to point to a very high surge for those areas, a dangerous surge, a life threatening surge. The track, intensity and large area of the storm was funneling all of that sea water towards the coast and towards New York Harbor.

But the other aspect that we emphasized was that there is a noted sea level rise that would compound the magnitude of the surge. Just because you haven’t seen water in these areas over the past several hundred years, we’re now dealing with a different background state, a sea level that’s higher. When you put this extreme event on top of a higher sea you’re going to get record surges, which is what happened in New York Harbor, Staten Island, Brooklyn, and the Queens area.

Sandy might not happen for another twenty, thirty, forty years, but it’s not a one-off kind of system either. We have to be concerned that these extreme events are going to have larger impacts because of this climate background state.

With respect to mitigating gaps in weather satellite data, I said several times now that a major component, a third component of the whole forecast enterprise is the global observing system and that satellites, the lower earth orbiting satellites, are a backbone to that entire system.

If we have a gap in coverage, if we can’t get the new satellites up before the existing satellites stop operating, that data gap will have a significant impact on our ability to predict these extreme events days in advance. That’s the key thing because for people to respond at day five you need a good forecast at days eight, seven and six leading to day five. And if then they’re responding at day three, you have to continue that consistent forecast skill. In order to get those kinds of forecasts in extended ranges you absolutely need those datasets to initialize the models over the entire globe. And, like I said, if we have gaps in satellite data, we’ll likely have a degradation in the forecast.

What are we doing to mitigate those gaps? There are budget initiatives that are being considered, supplemental funding that is being considered, not finally approved yet, that allows us to take action to improve our data assimilation, improve our models, and improve our use of existing data that could help mitigate the gap. It’s not going to replace the satellites by any measure. But we have to take steps to mitigate a potential gap, and that’s in fact what we’re planning to do.

SSEC: Can you address more specifically how you plan to move forward in this era of reduced resources?

LU: That’s the challenge for all of us. We have to prioritize what we need to do and map that against the budget that we actually have to work with. The bottom line is the bottom line and there’s absolutely no wiggle room around that fact.

What this means is that whether it’s from an observing system perspective, a science perspective, a modeling, model implementation, what IT you apply to your forecast process, what partnerships you form to share, to leverage and get more bang for the buck from a government perspective, it’s all of the above. I think what we’ll see is a lot more partnerships among the various agencies, leveraging, less duplication within an agency and among agencies to get focused on what needs to get done. I think that’s what you’ll see during these austere times.



SSEC: You’ve noted a commitment to transferring research into operations. How do you ensure the bridge between the two remain solid? Or how can it be improved? And then, what is your perspective or what is a good balance for the National Weather Service between conducting research and maintaining operations?

LU: Let me answer the second one first. The Weather Service focuses more on the transition of research to operations and that gets us back to the previous question about resources. If it’s so critical to actually be able to do research and there are other agencies that support research at a much higher level, financial level, then the key here is going to be how we properly leverage those research advances and how we work with the other agencies to effect a transition from research to operations. Those are the kinds of questions that I’ve been looking at for a long time. I’ve been responsible for setting up test beds, interagency test beds, test beds within the Weather Service, that really focus on this R2O [research to operations] issue and reaching back towards the research community that really wants to work with us in an operational world. That’s exactly where I think we need to be. We need to capitalize on that. Our focus on the research to operations transition through test beds and working with other agencies, to make that work better even in the face of these budget challenges, I believe we can improve what we do among the various agencies, relying on university research that really wants to be in this area.

Some researchers want to do basic research for the sake of doing basic research and that’s fine, I’m all for that. Other universities, and Wisconsin is one of them, want to influence the operational community and they work with us right from the beginning. They understand how we operate and they understand how to plug their advancements into the way our operational systems work. They’re usually up at the front of the line in terms of being able to affect transition from research into operations and Wisconsin’s examples are many. It’s not just the satellite arena but it’s how the satellites are used, for example, down at the Hurricane Center. That’s the kind of effort we need and that’s the kind of effort we need to build on with the larger community.



SSEC: What defines a “weather ready nation” and what is your vision?

LU: What defines a weather ready nation is a nation that’s ready to take action based on the forecast that we provide, an action that would save lives and mitigate property loss. Some parts of the country are better, more advanced in that area, they take the forecast seriously and they have action plans set up that actually respond to those forecasts, watches, and warnings, where other parts are not. We have weather ready communities. To do this on a larger domain we really do need to look to the nation as a whole.

We have to continue to work with the emergency management community from FEMA from the national perspective down to the regional down to the local down to the individual. To ensure that individuals have an action plan in one of these weather situations and that they’ll take the proper action.

Examples are numerous, but the ones that you hear about, like the meeting in Oklahoma after the severe weather outbreak. If you’re separated from your family, your instinct might be to drive across town to where your kids are in school, right? But, people were killed trying to reunite with their family during the event. People have to have a plan for these major events. As we saw in 2011, tornadoes are cutting right through a major metropolitan area, disrupting many, many lives.

What is your individual action plan in those events? We’re trying to encourage people to work not just from a weather forecast perspective but to work with our partners in the emergency management community, as an example, to deal with these types of scenarios and to deal with these events so that you’re ready before the event occurs.

Another example is our developing relationship with the Centers for Disease Control as they show that many of these [disease] outbreaks are related to either drought conditions or rain conditions, that there are linkages between outbreaks and weather and climate. And yet we’re just learning about the CDC’s requirements. They are learning about what we can provide not only here but globally. We need to work toward decision support services that will work for their mission. That would certainly make, even from a health perspective, make this country a weather ready nation. It is issues like that that really need to be connected to make us a weather ready nation.

In terms of a vision, I helped develop this vision so I’m not going to reinvent that coming into the Weather Service. I’m just going to try to focus our efforts in the way we advance throughout the entire forecast enterprise to be able to address these types of issues.



SSEC: When you reflect on your career and the development of the field, where do you see the greatest potentials?

LU: Well, you know, I interact with lots of students. Before I came into this position, I was the director of the National Centers for Environmental Prediction. I had the opportunity to visit up to five universities a year and talk with students. What I tell them is that there are a lot of challenges just being able to predict the weather. As great as the success has been we still have a lot of work to do — extending the prediction to one, two, three weeks in advance. There are a lot of applications for two week, three week, four week predictions whether it’s people taking vacations or energy companies operating their systems or where they’re going to transport natural gas, that kind of thing. So there are tremendous challenges and potential.

But the main thing is, and we hear this over and over again, it’s not just making the prediction, it’s not just the science aspect, it’s the societal impact and how do we connect to society in such a way that we literally can save lives and mitigate property loss and make this a worthwhile venture that goes beyond the science.

From a decision support system perspective, this is probably one of the greatest potential aspects of our science because we have built prediction into our science over the past one hundred years. We did this when other fields said that we were degrading science by attempting to predict. I actually sat in a seminar in 1970 in Wisconsin where a geologist said they should give up trying to predict earthquakes and oh, by the way, meteorologists should give up trying to predict the weather. That we were degrading the science by doing this.

We have taken basic physics, basic math, and turned it into a predictive science. Now we’ve got to connect with society. It’s a major challenge but it has huge potential. I encourage students to look at the problem holistically and work their careers accordingly.

[SSEC: Jean Phillips, Leanne Avila, Mark Hobson, Tom Achtor, Linda Hedges]