“Environmetrics is still in the formation stage”: An interview with Walter W. Piegorsch and Abdel El-Shaarawi

The Encyclopaedia of Environmetrics is now available online in a new edition with brand new content included. StatisticsViews.com talks to Walter W. Piegorsch and Abdel El-Shaarawi about the development of this project, what readers can expect in this new edition and the influence of statistics on their careers.

1. Encyclopaedia of Environmetrics is now available online in a new edition with new content included. How does one go about updating such a wealth of informative literature and how long did the process take you?

A: We were contacted by our colleagues at Wiley prior to the JSM 2010 meeting about updating the 2002 Encyclopaedia and we decided then to discuss the process seriously at JSM 2010 in Vancouver. At JSM we held a dinner meeting where two section editors (Eric Smith and Mac Hallin joined us in a serious discussion about the details of updating such a massive work.

W: Environmental Science was and continues to be a growing area but our timing with the first encyclopaedia was a little off, unfortunately. The topic was gaining attention in the late 1990s and we started working on the first edition ’round about 1998. But, it took some time to put together and came out in early 2002. That was right after 9/11 and the consequent economic and political changes in the US led to reduced spending on environmental sciences for quite some time. Coming into the 2010s, tides changed. The economic situation is still difficult, but the folks at Wiley looked ahead and thought: if the interest in environmental sciences continues to grow, we will be ahead of the game. The timing this time was right.

A: At this stage we planned to contact the editors of the first issue to see who would be willing to be involved in the project. Some agreed to be involved because they were still working in the area while others were no longer interested in getting involved and some had retired. So we needed to put a new team together and we wanted this team to reflect the current scope of environmetrics research and application Environmetrics is an area still in growing and formation phase, not just because of science catching up with environmental issues but also because it is connected with issues of greater concern to human life such as climate change, and water shortage and water quality. The environment is a rapidly changing and science has to catch up with the society needs. This was the beauty of the project in the first place: linking environmental applications with statistical and quantitative science. Environmetrics was never thought of in terms of statistics only but with other aspects of quantitative sciences such as physics and applied mathematics which are relevant to environmental problems.

A: In updating the Enclyopaedia we decided to start with an entire look at we had in the first issue and how to define a formula for the new phase so revised the editorial board and specified well defined topic areas so the section editors can identify contributors for covering these topics. Only one or two editors from the original Board joined the board and we aimed at adding to the board younger and active individuals who are willing to put the efforts into the encyclopaedia and they actually did a tremendous job. They were familiar with current issues and that helped give the encyclopaedia the edge that was needed.

W: ‘Rapidly evolving’ is often used just as a cliché, but honestly, the field of environmetrics really is constantly changing. The new board of section editors looked at all the entries from the first edition and decided what could stay and what needed updating. The beauty of our vision of environmetrics is that it’s about developing new technology and new methodology, which try to address real problems, which then lead to real solutions.

2. What were your main objectives during the process? What did you set out to achieve in reaching your readers?

W: Our first question was, what has happened in last 10 years that we need to focus attention on, what information has changed during this time? I am thrilled with the new edition, as we have managed to cover most of the topics we wanted. We wanted to modernise certain topic areas and so we marked those that needed updates. I was concerned about entries we didn’t get to in the first edition, so we bounced ideas off of the section editors using massively detailed spread-sheets. Those topics — some new, some old — determined which editor would work with what material, so when we started the current edition, we already had strong foundations on which to build from the first edition.

A: We asked ourselves whom are we going to be serving – graduates, researchers, decision-makers, the range of individuals who are concerned about the environment. We hope that the Encyclopaedia will provide the tools to keep them informed with up-to-date with precise and accurate knowledge that will help them address all issues related to research and application. This new updated issue covers areas such as the huge massive data sets that became common as a result of the advances in computing technology.

We asked ourselves whom are we going to be serving – graduates, researchers, decision-makers…about the environment. We hope that the Encyclopaedia will provide the tools to keep them up-to-date with precise and accurate knowledge that will help them address all issues related to research and application.

– Abdel El-Shaarawi

3. Were there areas that you found more challenging to write and if so, why?

A: The environment engineering and technology section was a challenge to include in the updated issue. There were a few items that were difficult to include in a particular section these were placed in the general section. In first edition, we had 12 sections and in this new edition we do have now 14 sections. I think some of the topics that went into the general section may be developing into new sections in the future as they grow with the accumulation of knowledge.

W: I would agree, it was challenging to write the new material. Once you have a basis to work on, it’s not too hard to rewrite an entry, but the brand new sections of course had to come from scratch. There’s a lot of new, multidisciplinary material in the second edition, some of it so current and standalone that it almost seems adhoc, and maybe these newer topics will find homes, as Abdel says, in new titled sections in future editions. Big Data comes up a lot in the new entries, as it is so integrated in what we do now in the quantitative sciences.

4. With the release of this new edition, what can the readers can expect in this new version?

W: A lot of solid updates, and overall it’s 50% bigger. It was a great opportunity to work with John Wiley & Sons in producing it — they gave us a lot more space to work with. There are about 150 totally new topic entries.

5. Why did you decide to add the new section, Environmental Engineering and Technology, to the new edition?

W: Abdel was the original Editor-in-Chief of the journal Environmetrics, and I took over a few years ago; from that post we could see what was happening in the field when we looked at the submissions authors were sending in. Environmental Engineering and Technology was one area that has now been growing for maybe ten years, but in the late 90s, it was less established. There wasn’t a place for it in the first edition and it was an obvious step to take up for the new one. It was an important goal for us not to simply produce the same material with a pretty new cover.

6. How does the Encyclopaedia help practitioners as well as researchers in the field?

A: It helps the practitioners on many sides – it defines for them the scope and extent  of important environmental problems, e.g. climate change. The current knowledge is there in this encyclopaedia and provides them with the tools to study and understand these problems. I organised a full day workshop in ISI Congress in 2012 just dealing with the issue of water and many people were trying to figure out to deal with water shortage and water quality problems. The environmetrics are the techniques for dealing with environmental issues. You can determine the limit if something is harmful to the environment. Now it has moved from human health to dealing with the broader aspects of ecosystem health. It provides reasonable guidance on how to apply quantitative methods to address real applications.

W: I couldn’t agree more. We did follow the model from previous Wiley MRWs [Major Reference Works] and focussed on topics that would be of interest to researchers and practitioners. Most of these are not very long entries but rather are short, quick introductions with references interested readers can follow up on. With the online version, embedded links will give users even more flexibility; they’ll give readers convenient signposts for where to learn more

7. What is environmetrics and how does it apply to our everyday lives?

A: Environmetrics is still in the formation stage. For some people, it means environmental statistics but to me, it means quantitative environmental analysis. It applies to all the quantitative subjects that tries to address environmental issues – electrical engineers are dealing with massive data sets, as are applied mathematicians – all these go under the analysis of environmental change as the result of the impact of pollution on the environment. This is the objective of science in my opinion and environmetrics is part of the structure.

W: Abdel helped define environmetrics.

With the online edition, embedded links will give users even more flexibility; they’ll give readers convenient signposts for where to learn more.

– Walter W. Piegorsch

A: Walt did a lot too!

W: Who did coin the term?

A: David Cox submitted a project to the National Research Council asking to fund a project under the title, ‘Environmetrics.’ Of the first issue of the Journal Environmetrics, Cox was identified as the founder of that term and not Stu Hunter. George Barnard emailed Stu Hunter congratulating him on founding the name but Stu replied that David Cox came up with the name first and not Stu. I still have all the correspondence! It probably should be archived. I have the details of the first conference in 1989 under ‘Environmetrics’.

W: In terms of our every-day lives, environmetrics applies to water quality, environmental regulations, environmental risk, big data applications, satellite imagery and pattern recognition. Almost always, there will be a quantitative question that has an environmetric solution. Abdel has always said that environmetrics is more than environmental statistics and I think we’ve worked hard to ensure this distinction. At the same time, I do notice that when you have an environmental issue with a quantitative aspect to it, there is almost always a statistical aspect buried in there somewhere. Maybe I’m biased, but more and more whenever I see a general societal question with a quantitative issue, a lot of it seems to be made up of statistical components.

A: Environmental statistics is a subset of environmetrics. Statistics is the scientific method as it is connected with understanding and the collection of measurements. The basis of scientific methods is statistics in my opinion. Environmetrics is designed to deal with a specific subset of areas, which are connected with environmental issues. These issues became very important during the end of the last century and the beginning of this century. Everyone is concerned about life on Earth and I think the way to learn is to have quantitative measures and this where environmetrics is important.

8. How did you both begin to pursue a career in statistics and what was it that brought you to recognise statistics as a discipline in the first place?

A: I can answer on behalf of both of us – we know how to count!

W: Well, I still make small mistakes. Can that be allowed? As an undergraduate, I took a probability and statistics sequence and I noticed that you could do something with mathematics that solved the problem right then and there. One can show that a pharmaceutical works better than a placebo, say, and that got me going. I really get a buzz now from doing interdisciplinary statistics — we’re always evolving and incorporating information across disciplines that teaches you things you never thought you’d learn.

A: The beauty of statistics is that once you know it and work with people from other areas of science – chemists, biologists, etc., you tend to learn their language without going into the lab and doing the work. You build on what you have learnt. When I worked at the National Water research Institute, I did not know about water related problems my concern was with its use for maintaining our lives. The successful statistician understands what the medical doctor is trying to communicate. Statistics is not a dull subject that people should shy away from as it really opens up opportunities to learn about other areas. If you go into linguistics, you can distinguish between Bacon and Shakespeare. It is a field that is of service to others as well as to those working in it. My grandfather wanted me to become a medical doctor but at school, I did not like it then I moved instead to the faculty of economics and political science. I specialised in statistics in the second year after I decided not study political science or economics. Statistics had equations and techniques that I could use and enjoy the breadth of its area of applications.

9. As university professors, what do you think the future of teaching statistics will be? What do you think will be the upcoming challenges in engaging students?

W: Science needs to be more interactive and we need to keep talking with each other, as statisticians are trained to do. I say this as someone who has just stepped down as head of an interdisciplinary training programme. Students often think statistics is a just a bunch of equations and numbers, but then they are not being taught by a statistician who can show them how it can all fit together (like that first instructor of mine did for me). That’s one of the problems we face. Of course, even the statistician-as-instructor has to bring the excitement through – show that we’re here to learn and understand science, rather than being a cubby-holed person doing the same thing over and over again. Insist on learning as you go. For instance, what is limnology? It’s the study of lakes and rivers [technically, inland waters], and I know that because I worked with limnologists one time. All of this comes to you as a practicing statistician. It is tough for instructors in the classroom as we have to bring in really good, motivating, technical material. Statistics is a way of integrating yourself into the scientific method and it will be successful if we can keep that philosophy into the future. I can spend weeks preparing pretty data visualizations, but I also have to make sure the students can go away and do complicated analysis. Science is getting more complicated, and so is statistics: You cannot just take one course and become an expert.

A: We need to motivate the students and show them the value of statistics as a science on its own right with strong impact in almost all fields of sciences.

10. Over the years, how has your teaching, consulting, and research motivated and influenced each other? Do you get research ideas from statistics and incorporate your ideas into your teaching?

A: Of course; this semester I am teaching time series, which is an area that has impact on making future decisions. For example you can use intervention analysis to study the current instability in Egypt on the Egyptian economy. You need to train the students and show them how research and applications are linked and how new thinking arise from day to day applications.If you want to know a subject, you need to teach it and then you can see new angles for extending its range of applications. Student questions can help you to think about limitations of existing methods and thus help in defining new research questions.

W: Yes, what Abdel just said! Once I finished grad school I worked for the US Federal government as a practicing statistician. I saw that it was about how you take the scientific method and work out the questions, and then how statistics can help you achieve the answers. I have been doing this for many years and my teaching and research are based on real problems that need solutions right now.

I worked for the US Federal government as a practicing statistician. I saw that it was about how you take the scientific method and work out the questions, and then how statistics can help you achieve the answers.

– Walter W. Piegorsch

11. What do you think the most important recent developments in the field have been? What do you think will be the most exciting and productive areas of research in statistics or environmetrics during the next few years?

A: I think the integration of environmental research will be a challenge as environmental problems are really just a sequence of layers upon layers of connected issues, e.g. climate change is connected to ocean pollution. , data from different disciplines need to be integrated, e.g. the water of the Nile is to be shared among nine countries. How do you determine share of each country in this limited water source? When you are dealing with massive data sets, it is important how the data is quantified at both the regional level and the global level. Analysis has been done on a smaller scale but I think it’s a much bigger issue and that needs further development.

W: I would second that. I think that ‘layers’ is a good way of putting it. Environmental sciences and statistics are not easy. Due to the advent of Big Data and the attention it is receiving, there has been massive development in statistics and there has been a coming-together of statisticians to figure out what these Big Data issues are. The ideas are also applicable in environmetrics and there are areas that continue to be challenging, such as stochastic modelling or how you analyse spatial and temporal data.

12. What do you see as the greatest challenges facing the profession of statistics in the coming years?

W: Pulling off our answer to question 9: answering all of these big questions on Big Data and data mining, and how to drive ourselves forward but being open to change and new things. It’s now the IYOS and I have a colleague who is not a statistician but who knew about the IYOS! Maybe the overarching goal is to increase the visibility of the field; while I don’t think we’re invisible, the general public isn’t always aware of all that statistics can be and, for that matter, how they can be misled by statistics. All this Big Data stuff is opening that up and we need to be able to answer the questions it’s making us ask. A change in perception is vital. Statistics is a way to do science better and it doesn’t take you long to find that out once you start to learn about the methods and practices. But, achieving this recognition takes time.

A: We are visible enough. The ASA is the second oldest surviving institution in the US and the RSS has been going since 1800s. One of the oldest records is the Nile Meter – these are statistics and actually, they are environmetrics. How statistics is intertwined with the lives of people, learning about issues in the economy statistics has been playing a great role in designing data collections, analysing data and interpreting results and actually in communicating the results to decision makers.

13. Are there people or events that have been influential in your careers?

A: Walt.

W: Why thanks, Abdel! I’d return the compliment and say Abdel has been an important influence: he really did help define what we now think of as environmetrics and laid the foundations for the rest of us. I’m not sure I have had that kind of influence…

A: We met in 1992 or 1993 in Australia – same people influenced us who are the forefathers of statistics and we owe it all to them – Newton, Fisher, etc.

W: The shoulders of giants, we sit on them (paraphrasing Sir Isaac Newton…). One such was George Casella, a Professor at Cornell who advised me as his interests were gravitating towards environmental sciences; he helped me get excited by environmental questions and it’s stuck ever since. George was a leader and published in the area tremendously. He was so broad that he had moved on and his latest work was on computation and genetics. (Sadly, he passed away last July, well before his time.)

A: He died in his prime and was very active publishing – the sheer output of his work over the years. I visited Cornell and gave a talk and met him. I met him at a conference I organised in Burlington in 1994 and he was very supportive of people working in this area. He was sending the graduate students to the seminars and asking them to summarise the statistical components, which I found really interesting that he did that.

W: I had an email with George a week before he died, on a new project we might do together but which now is uncertain. I know he had a few more books he still wanted to write — almost by definition they would’ve been very influential. As mentioned earlier, Abdel was the founding Editor-in-Chief of the journal Environmetrics, a job I’m now honoured to hold with Peter Guttorp; when we heard the sad news, Peter and I moved quickly to commission a special issue of the journal in George’s honour, which will appear sometime this year.