RICHARD J. REED

A Mini-Autobiography

              A professorship in a modern research university, such as the one I held at the University of Washington, carries responsibilities in three areas—teaching, research, and professional service. The following account focuses mainly on the research aspect of my career and on the meteorological training and education that laid the foundation for my future work.

I graduated from Braintree (Massachusetts) High School in June 1940 in the period between the Great Depression—the effects of which were still being felt and US entry in World War II. Though I was a good student with well-rounded abilities and no strong leaning toward any particular subject or career choice, it was assumed that I would join two mathematically and scientifically talented classmates in attending MIT. Certainly math and science had been two of my favorite subjects, and I would have welcomed the opportunity to go to a school like MIT or another elite school such as Harvard. However, the tuition at these institutions was out of reach for my family, even when supplemented by a scholarship awarded at my graduation. Consequently it was decided that I would enroll in the fall at a less expensive school, Boston College, where I would satisfy my love for numbers by majoring in accounting. Actually it was the depression mentality, not the fondness for numbers, which shaped my choice of major. Decent jobs were scarce in those days, and I was reminded by family and friends that certified public accountants (CPAs) were well paid and reliably employed.

My scholarship having run out at the end of my freshman year and having no other source of funds for the next year’s tuition, I took a job in a factory with the purpose of earning enough money to return to college the following year. Then fate intervened. On December 7, 1941 the Japanese attacked Pearl Harbor, and war was declared. Imbued with the near universal love of country that marked the public mood in those genuinely patriotic times, I enlisted in the Navy at the bottom rank—Apprentice Seaman—and in January 1942 reported to Newport, Rhode Island, for boot training. At the finish of the training I was invited, as were all recruits, to select from a long list of specialties, such as boatswain, yeoman, etc., the three on the list that I found most attractive. Among the three I selected was aerology (the naval term for meteorology), probably motivated by a long, but ill-defined, interest in weather. I had never read on the subject or been exposed to it in high school or college, and I had never maintained a weather station as was commonly done by young weather enthusiasts, though I did keep a daily record of the local wind direction taken from the wind vane atop a nearby church. In any event the authorities decided that aerology was to be my specialty and assigned me to an on-the-job training program at the nearby naval air station at Quonset Point, Rhode Island, rather than to the customary training school at Lakehurst, New Jersey, which was already full to capacity.

On the practical side I was instructed in map plotting and weather observation by enlisted men stationed there, the foremost of whom was Albert Lewis, a colorful character from Arkansas with limited academic background but high native intelligence. Lewis rose to the rank of Chief Petty Officer while I was still billeted at Quonset Point and subsequently became a commissioned officer, attaining the rank of Lieutenant Commander by war's end. Such were the unusual opportunities that existed for advancement during the war years when merit rather than formal credentials and time served in rank was the basis for promotion. Lewis was an excellent weather analyst and forecaster who either by direct instruction or by example helped me acquire skills in weather analysis and forecasting. Although analysis and forecasting were mainly the duties of the officers, enlisted men shared these duties increasingly as they rose in rank. Lewis’s map analyses were a thing of beauty that had a profound effect on my aesthetic appreciation of meteorology. The emotional charge I have always felt in analyzing frontal cyclones stemmed from his early example.

 Freshly minted young ensigns Max Edelstein and Alvin Morris, the latter to become a longtime friend after the war, were assigned the job of teaching the trainees the elements of meteorology. To aid their instruction they suggested that we read a popular— and deservedly so—elementary textbook by Blair. I have never forgotten this experience. Once started on the book I could not put it down, staying up that night until I had finished reading it and feeling at the conclusion that I had thoroughly absorbed the material despite my relatively weak scientific background. If there ever was a case of love at first sight for a scientific subject, I experienced it that day (and night). There are those who view unusual ability in math and physics as the key to scientific success and its manifestation in a particular subject as largely a matter of accident. I have never subscribed to this view. The aesthetic feelings aroused in me by weather patterns and the fascination I felt for weather phenomena as physically evolving entities have always seemed to me inborn facets of my being. I cannot picture any other field of study having had the same emotional effect.

Given the chance to progress at a pace determined by merit alone, I advanced three grades in a little over a year to the rank of Aerographer Mate First Class. At that stage—undoubtedly at Al Lewis’s instigation—I was nominated to take the examination for a fleet appointment to the US Naval Academy. After a brief period of preparation at a navy facility in Norfolk, Virginia, I took the exam and passed with flying colors whereupon I went to Annapolis in June 1943 with the expectation of being admitted. But then Dame Fortune smiled on me as never before, as anyone who has trouble visualizing me as a career naval officer can appreciate. Despite having better than 20-20 vision, I was found to have astigmatism, which at that time—but assuredly not now—was grounds for rejection. As a consolation prize, I was sent under the Navy V-12 program to Dartmouth College for one year (three semesters) to acquire further credits in standard academic subjects (math, physics, English, history, etc.) at the conclusion of which I was slated to go to midshipman school and be commissioned an officer in the Navy Reserve. But again Lady Luck intervened.  Shortly before the year ended an announcement was posted inviting trainees to become candidates for a lone position that was available for an additional year (three semesters) of study in the meteorology program at Caltech. I jumped at the chance, and in view of my background was not surprised when I was selected. Thus my formal academic training in meteorology was begun in July 1944.

From my earliest days as an enlisted man, I had heard of the controversial Irving P. Krick who headed the meteorological unit at Caltech, a special division of the Aeronautical Engineering Department. Now I was to be trained in his preserve—a somewhat disconcerting thought! Fortunately Krick was on duty as a colonel in the Air Force so I was not directly exposed to his influence, and his cohorts in the department taught, with acceptable competence, the less demanding analysis and forecasting courses, subjects in which I was already well versed. The more demanding meteorology courses, physical and dynamical, were taught by regular members of the Caltech faculty—the eminent geophysicist Beno Gutenberg and the brilliant aeronautical engineer Homer Joe Stewart—as were supporting math courses. So despite the Krick influence I received a sound undergraduate education in meteorology.

Having completed eight semesters at the college level, two at Boston College and three each at Dartmouth and Caltech, I was awarded a Bachelor of Science Degree in Meteorology from Caltech in June 1945. From there I was sent to midshipman school at Notre Dame and commissioned an ensign in November of that year. Meanwhile the war with Japan had ended (that in Europe had concluded while I was still at Caltech) so that after commissioning, rather than being sent to the Pacific theater as originally planned, I was assigned to air station duty in Rhode Island, within hailing distance of my home near Boston, Mass. After being released from active duty in June 1946, I enrolled at MIT, entering the Meteorology Department in September 1946 and graduating with an ScD (Doctor of Science) degree in June 1949. The relatively short period for attaining the ScD was the norm for the Department in those days, when demonstration of the ability to perform original research, rather than the accomplishment of noteworthy research, was the goal.

The department at MIT had been founded by Carl-Gustav Rossby, the foremost meteorologist of his day, and subsequently was headed by Sverre Petterssen, the leading synoptic meteorologist of the era. By the time I arrived, Henry Houghton, a man valued for his wisdom and leadership ability as well as his scientific acumen, was in charge. Houghton taught physical meteorology, Bernard Haurwitz and Victor Starr dynamic meteorology, James Austin synoptic meteorology, and Hurd Willett long-range forecasting and climate change. Their courses were always well prepared and were uniformly excellent in content. I was indeed fortunate to have the benefit of instruction by such knowledgeable and dedicated teachers. Though synoptic meteorology was always my primary interest, I wrote my ScD thesis under Willett on the seemingly exotic subject "The Effect of Atmospheric Motions on Ozone Distribution and Variations," motivated largely by what I understood as an unwritten rule that mainline synoptic meteorology was not a suitable area for an MIT doctoral dissertation. The rule, if ever such existed, was broken a few years later by my early collaborator Fred Sanders. On the other hand, I may have been drawn to Willett by his interest in solar-weather relationships and the notion, posed at one time by Haurwitz (then no longer at MIT), that the source of such a relationship, if any, might reside in the ozone layer. Certainly neither at that time nor at any later time was I sympathetic to the idea of a direct solar effect on weather. I merely wished to understand the already well-documented ozone-weather relationship and for this purpose sought out as thesis advisor the faculty member most closely associated with the problem.

After completing the ScD in June 1949, I was offered a position at MIT on a project, sponsored by the Office Naval Research (ONR), known as the Pressure Change Project, and held the position for the next five years. The project was led primarily by Austin and had as its goal studying the mechanism of atmospheric pressure change in cyclones and anticyclones. Austin subscribed to a thermal mechanism that had little appeal to me; I preferred instead a more dynamic approach based on, admittedly, ill-defined ideas related to vorticity. This divergence of approach was not a source of conflict, since earlier Austin had made extensive use of vorticity in his ScD thesis, written under Petterssen. The truth is that Austin, to his credit, made little effort to influence my thinking or control my activities, allowing me to follow my interests in whatever direction they carried me. Besides preparing project reports of minor importance, I published three articles in the Journal of Meteorology on atmospheric ozone (the initial one in 1950 representing my first publication in a refereed journal) and a jointly authored paper on atmospheric cooling by melting snow, which was inspired by a late season, surprise snowstorm in Boston. During this period I read critically the works of Bjerknes and Palmén, became familiar with, and intrigued by, the seemingly odd ideas of E. Kleinschmidt, Jr., and was drawn to Charney’s work on baroclinic instability, which seemed to me to provide a more realistic explanation of cyclone development than the prevailing Norwegian model. I also taught my first class, basically a synoptic laboratory.

By far the most important papers that I published during the postdoctoral period were a pair on upper-tropospheric frontogenesis and the related topic of tropopause folding or stratospheric intrusions. The first of these was co-authored with Fred Sanders, then working on his dissertation. Fred and I were kindred spirits who fed each other's discontent with what we regarded as the nearly blind acceptance by many meteorologists of the model of cyclone development advocated by the Norwegian School. It seemed obvious to us—and we were not alone in this view—that fronts often strengthened during cyclogenesis rather than providing sharp preexisting thermal discontinuities on which cyclones formed. Also we objected to the view, suggested by cross section analyses of the Norwegian School, that fronts extended with more or less equal strength throughout the troposphere rather than, at least for truly strong and unambiguous fronts, being mainly confined to restricted regions of the atmosphere. These regions were the lower troposphere, particularly the near-surface layers, and in some instances the layer near the tropopause. Fred deserves the main credit for proposing this dichotomy, which we felt not only better described nature but equally importantly encouraged more open thinking on atmospheric structure. Following the completion of our joint paper Fred decided to concentrate on low-level fronts, as he did in his thesis (and in the subsequently published version of the thesis), and suggested that I continue treating upper-tropospheric fronts, as I did in my paper on tropopause folding. The papers on upper-level fronts made extensive use of potential vorticity in distinguishing between tropospheric and stratospheric air and in tracing air trajectories. I would be remiss if I failed to acknowledge the part played by Victor Starr in stimulating my interest in potential vorticity and in calling my attention to its role as a tracer. Victor also encouraged skepticism regarding accepted ideas, a necessary attitude for the young scientist intent upon breaking fresh ground.

During the five-year postdoctoral stint I stayed alert for a teaching opportunity at a reputable university. Some offers were received that seemed worthy of consideration, but in each case Houghton advised me against accepting them. Thus when an offer came for a position at the University of Washington, at that time a little-known player in the meteorological firmament, he surprised me by urging that I give the offer serious consideration. Trusting his judgment, I looked further into the proposition and decided to make the move to Seattle. I owe Houghton much credit for his sage advice on important issues and am deeply indebted to Austin for both his direct and behind-the-scenes efforts to advance my career. Like my father and Al Lewis before him, I count him as one of my main boosters—those who served not so much as mentors but as confidence builders and promoters of my interests. At Washington, where I began my duties in September 1954, I was to meet another great booster, Phil Church, the Chairman of the Department of Meteorology and Climatology.

In accepting the position at the University of Washington, I had agreed to work on an arctic project being negotiated by Phil with the Geophysical Research Directorate (GRD) of the. Air Force Cambridge Research Center (AFCRC). In the next several years a number of reports, MS theses, and published papers on arctic weather analysis and forecasting resulted from this effort, only one of which represented a substantial contribution. An "Arctic Forecast Guide" for naval forecasters, prepared while I was a consultant to the Navy Weather Research Facility at Norfolk, Virginia, was a byproduct of this work. I pursued a number of other interests beside arctic meteorology during those early years at UW. With Ed Danielsen, like Fred Sanders, a soul mate who shared my reservations about the Norwegian model, I published a further paper on upper-tropospheric fronts. Nominally Ed was my first doctoral student, but he was so independent in nature that I always felt that he viewed his dissertation—much of it conceived before my arrival at Washington and later carried out separately from our joint research—as falling in a special unsupervised category.

I also published a couple of papers on graphical prediction, which at that time was competitive with numerical prediction as a method of preparing prognostic surface charts. The latter soon surpassed the graphical method, not to my surprise since I always regarded graphical prediction more as a teaching tool for synoptic labs than as a long-term approach to forecasting. Nonetheless, during my forthcoming sabbatical in 1961-62 at the National Meteorological Center (NMC) in Suitland, Maryland, the equations and procedures involved in the graphical method were programmed and the output of the numerical version used for several years as background for manually prepared surface prognoses. Eventually, in 1966, the more sophisticated mainline numerical surface prognoses reached the stage where they could be used directly and the graphically based prognoses slowly faded into oblivion.

 One contribution published during those early years was written purely for fun. This was a paper entitled "Flying Saucers over Mt. Rainier," inspired by a picture in a Seattle newspaper of a saucer shaped cloud formation downwind of that majestic mountain. Informed by my colleague Bob Fleagle that the original sighting of flying saucers was made by the pilot of a light plane flying in the vicinity of Mt. Rainier, I did an analysis of the later situation and proposed on the basis of the investigation that the flyer in the original sighting likely mistook lenticular clouds for vehicles from outer space. Needless to say, this proposal had no impact on the true believers. In fact it outraged some. I've always been proud of this offbeat paper and disappointed that it has not received greater notice in the literature on the subject. I freely admit to having a passionate dislike of UFOs, sasquatches, Loch Ness monsters, and other products of credulous minds.

My second major paper of this early period—the paper in which I first described the phenomenon that came to be known as the Quasi Biennial Oscillation (QBO)—had its origin in an improbable series of events, which I will now relate. Some will say that I was only the co-discoverer of this amazing phenomenon, and I will not contest them. As with all discoveries bits and pieces of the final picture emerged over a period of time, and two British meteorologists, Veryard and Ebdon, published a paper similar to mine in the same month. But when I made the discovery, I felt—and justifiably so—that I had found and described in its essential form a bizarre phenomenon not recognized as such in previous work. Indeed much of the preceding literature was unknown to me at the time. For reasons that are now obscure, I was invited to present a review paper on the stratospheric circulation at the annual meeting of the AMS held in Boston in January 1960. Perhaps the invitation stemmed in part from my early work on atmospheric ozone and in part from my participation in summer seminars held by McGill University at Stanstead College in Quebec. In the beginning, the seminars were devoted to arctic meteorology, which was then my main field of research. But subsequently they took on an increasingly stratospheric complexion that, according to my recollection, stemmed from an interest on the part of some of the participants, including myself, in the stratospheric sudden warming phenomenon, a rare event that was seen most strikingly in the arctic stratosphere. Whatever the reason for my invitation to speak at Boston, I felt a need to be brought up to date on the latest work being done on stratospheric meteorology and decided, accordingly, to attend an AMS-sponsored meeting being held on that subject in Minneapolis, Minnesota, in September 1959. Conveniently, as it turned out, the organizer of the meeting, Arthur Belmont, requested all speakers to bring multiple preprints of their contributions for distribution to the participants—a procedure that I encountered there for the first time—and I made a point of collecting a copy of each. My own participation was confined to chairing one of the sessions.

The following November my colleague, Joost Businger, who had accompanied me on the trip to Minneapolis, visited my office one morning and asked what I was doing. I replied that I was working on the Boston paper. "What do plan to say about the strange winds at Christmas Island reported by Colonel Frank McCreary?" he inquired. I responded that I was so busy watching the time clock, as chair of the session at which McCreary spoke, that I had not paid much attention to what he said; but, no matter, I had a copy of the preprint of his paper on the shelf nearby. After Businger left, I perused the paper and was intrigued by figures showing the alternation between easterly and westerly winds at different levels in the equatorial stratosphere during the two years of nuclear testing at Christmas Island, and in the seeming downward progression of the changes. This behavior was contrary to the accepted picture at that time of persistent Krakatoa Easterlies between 25 km and 30 km, as first seen from the drift of the dust ejected in the volcanic eruption of Krakatoa in 1883, and somewhat fitful Berson westerlies underneath, as first observed in wind soundings taken in East Africa by the German meteorologist von Berson. At that time my office shelves were laden with booklets containing mean-monthly sounding data for US stations, including those operated in the Pacific by US personnel. I feverishly pulled these from the shelves and roughly tabulated the stratospheric wind data for nine stations in the equatorial Pacific. It turned out that at one of these stations, Canton Island, there were enough high reaching soundings to establish that a pattern of downward propagating easterly and westerly winds occurred for at least two cycles and that all stations in the sample shared the pattern. Intuitively (or wishfully) I assumed that the cycle was a regular feature of the equatorial stratosphere and so reported it in Boston.

The discovery came too late to announce in the abstract of my paper published in the Bulletin of Meteorology, but is on record in a preprint of the paper, faded copies of which still exist. Formal publication of the results in the Journal of Geophysical Research was delayed until March 1961, the same month in which Veryard and Ebdon published similar findings in the Meteorological Magazine. The delay was a consequence of my decision to hold off submitting the paper for publication until confirming data were received from Nairobi, Kenya, a station on the far side of the world from the Pacific stations. I thought the oscillation must be of global extent but felt the need to establish the fact before submitting the paper for formal publication. Even so, one of the reviewers rejected this seminal paper, perhaps because it conflicted with the prevailing view on equatorial stratospheric winds. Fortunately the editor had decided to proceed with publication before the negative review was received.

Eventually AFCRC decided to end its support of arctic research and, in the person of Ed Kessler, persuaded me to turn my efforts instead to looking into synoptic applications of radar data. For this purpose they supplied the Department with a vertically pointing 1.87-cm radar, and I began studying, in collaboration with Carl Kreitzberg, a PhD candidate, time strips of the radar echoes observed in the Pacific storms that passed overhead. Subsequently AFCRC loaned us a GMD-1 sounding system for taking serial ascents in the storms, thereby allowing better definition of their thermal, moisture, and wind structures. In the winter and spring of 1961, extensive data were gathered in five storms that traversed the area. Carl and I began analyzing the data before I left on sabbatical, and during my absence he completed the analyses and used the findings as the basis of his doctoral dissertation. Carl was my second PhD student (if Ed Danielsen, is counted as my first).

Upon returning in July 1962 from the sabbatical at NMC, I turned my attention increasingly to stratospheric meteorology, obtaining support for the effort from the National Science Foundation (NSF). The NSF administrative procedures were simpler than for the Air Force sponsored research and, at least at that time, allowed principal investigators more freedom in modifying grant objectives when unexpected opportunities arose. The change to NSF support was indeed a welcome one; I am deeply grateful for the 35 years of unbroken support that followed. Topics studied during the period between my first and second sabbaticals  (1962-1968) were the QBO, the stratospheric sudden warming phenomenon, the transport of trace substances in the stratosphere, tidal motions in the stratosphere and mesosphere and a variety of other features of the circulation at high levels. Some of these studies employed wind data acquired from rocket launches at a sparse network of firing ranges. The rocket data opened up a new region of the atmosphere to exploration, and I have to thank Willis Webb, the head of the meteorological unit at the White Sands Missile Range in New Mexico, for inviting me to become a consultant to the unit and providing me with the incentive to enter this new realm. An unexpected bonus of the work on tidal winds took place when Richard Lindzen joined me as a postdoc in 1964-5 and tackled the problem of explaining the tidal winds theoretically. He made some progress with the theory before leaving the University of Washington for a second postdoc at the University of Oslo. While in Oslo, and subsequently at NCAR, he succeeded brilliantly in achieving his objective. It was the source of much satisfaction when later observational work that I carried out with a number of my students confirmed Lindzen's predictions. It should be mentioned also that Dick was the de facto supervisor of my fifth PhD student Donald McKenzie, who wrote his dissertation on an aspect of tidal theory. My third and fourth doctoral students, Stuart Muench and John Perry, had earlier written dissertations on the energetics of stratospheric warming events that occurred on two different years, and later Taroh Matsuno, a visiting postdoc supported by my NSF grant, wrote an important paper relevant to this phenomenon.

As the time for my second sabbatical neared, I was approached, probably on Bob Fleagle's recommendation, to spend the year in Washington, D.C., as the Executive Scientist of the US Committee for the Global Atmospheric Research Program (GARP). Thus I spent the period from September 1968 to August 1969 assisting Jule Charney, the committee chairman, and Joe Smagorinsky and Vern Suomi, the vice-chairmen, in drawing up the conceptual plan for US participation in this large international undertaking. My motivation in accepting an administrative post was two fold: I was a firm backer of Jule Charney’s idea of conducting an experiment to gather a global data set, and I welcomed the opportunity to participate in a program that had as one of its objectives improving US-Soviet relationships by bringing US and Soviet scientists together in cooperative ventures. It is often forgotten in this time of more harmonious relationships between our two nations that many scientists, on both sides of the Iron Curtain, shared this idealistic motivation for participating in GARP.

Although the busy year in Washington, DC, afforded little time for research, the GARP experience had a profound effect on my future research. As part of my duties I was involved with working groups that were convened to recommend subprograms in areas that were felt to contribute importantly to the overall objectives. Two of the subjects treated by these groups particularly caught my fancy—tropical meteorology and clear air turbulence—and upon returning home, I decided to leave stratospheric meteorology behind and strike out in new directions. No doubt the occasional urge felt by scientists, as well as by those in other walks of life, to try something new and different played a part in the decision. The venture into clear air turbulence—actually begun shortly before leaving on the GARP stint—was relatively short-lived, resulting after my return in two papers coauthored with Ken Hardy of the Geophysical Research Directorate of the Air Force Cambridge Research Laboratories who was a visiting scientist in the Department during part of the time of our collaboration. The venture into tropical meteorology lasted for a much longer time. A first paper, published in 1971, on the structure and properties of synoptic-scale wave disturbances in the equatorial Western Pacific drew much favorable attention, and this together with my GARP experience no doubt led to my appointment as the chairman of the panel charged with drawing up the US plan for participation in the GARP Atlantic Tropical Experiment (GATE). The actual writing of the plan fell to my colleague Mike Wallace whose spectral studies of the Pacific waves predated my synoptic work and aroused my interest in them. My interest was also stimulated by theoretical work done by Jim Holton during the same period. While waiting for the experiment to begin, I did further work on the waves, aided by my student Richard Johnson. Dick subsequently became my sixth PhD recipient, writing his dissertation on the role of precipitation downdrafts in cumulus-synoptic scale interaction.

During the experiment, held in 1974 off the West coast of Africa, I was stationed for nearly three months at the headquarters in Dakar, Senegal, as an internationally appointed advisor to Joachim Kuettner, the leader of the experiment. Watching the weather unfold each day on the maps prepared on site and on the satellite imagery, I became fascinated with the easterly or African waves that crossed the region with great regularity and upon returning home determined to make them the main object of my research. I regard the output of this research to be among my most important contributions to meteorology, though more for its definitive nature than for its originality. Toby Carlson and particularly Bob Burpee must be credited with the seminal work on the subject. My research on the waves spanned more than a decade, ending with a pair of papers written in collaboration with Tony Hollingsworth while on sabbatical leave at the European Centre for Medium-range Weather Forecasting (ECMWF) in 1985-86. In studying convective activity in the waves I was struck, as were others, by the strong diurnal cycle that was present in the convection whether the waves were located over land or sea. This observation sparked a general interest on my part in the nature of the diurnal variation of tropical convection. Subsequently several papers were written in the period 1981-85 on the diurnal cycle in the GATE area and, based on data from the First GARP Global Experiment (FGGE), in the tropical Pacific as well. Some of these papers turned up interesting regional differences in the satellite signatures of the deep convection that to my knowledge have never been explained or for that matter widely acknowledged. In view of the current emphasis on simulating cloud effects in the general circulation models employed in global warming studies, I can only hope that these important differences will eventually be recognized and used to test the fidelity of the models. Incidentally, I must credit Bill Gray’s work on tropical convection for greatly stimulating my own interest.

In the late 1970s while still working on African waves I became interested in small synoptic-scale cyclones that, risking controversy, I will refer to here as polar lows. This interest stemmed from the connection in my mind between the convectively enhanced cyclonic systems that formed off Africa and the observed small cyclones that occasionally developed during winter in cold, unstable Pacific air masses poleward of the jet stream and that were seen by satellite imagery to involve the organization of deep convection. Following the initial paper on the subject "Cyclogenesis in Polar Air Streams," published in 1979, several years elapsed before I resumed research on polar lows—whether of the Pacific variety (sometimes referred to as comma clouds) or of the so-called "true" type (a term coined by Erik Rasmussen whose work I found particularly informative and stimulating) that form over high latitude ocean bodies far removed from middle latitude influences. The research included observational studies done with Warren Blier prior to my ECMWF sojourn, a combined observational and theoretical study carried out with Charles Duncan while I was at ECMWF and, after my return home, studies that employed the Penn State/NCAR mesoscale model as a diagnostic tool. Blier earned his PhD in 1989 on one such study, becoming number seven on my list of doctoral students. Mark Albright and Jim Bresch also did yeoman work in application of the model to specific cases of polar lows.

Shortly after starting the work on cold air cyclogenesis, I undertook studies of two events that produced damaging winds in Western Washington—one connected with a vortex formation in the lee of Mt. Olympus during the passage of an intense synoptic-scale cyclone in February 1979 and the other associated with an episode of particularly destructive downslope winds on the west slope of the Cascades in November 1979. The first of the studies was undertaken in response to a request from the Washington State Department of Transportation that I join a team of experts set up to investigate the cause of the collapse of the Hood Canal Bridge during the windstorm. The second was started of my own volition, simply out of a desire to document the conditions underlying a destructive event of a type known to occur occasionally during winter in and near the western foothills of the Cascades. This was only one of several instances, dating back to the April snowstorm in Boston, in which my research was prompted by curiosity regarding a directly experienced event.

Before embarking on the sabbatical to ECMWF in September 1985 I began research on a subject– explosive cyclogenesis – that was attracting much attention at the time. Sanders and Gyakum had written the seminal paper on the subject, and follow-up papers by Gyakum, Anthes, Bosart, and Uccellini had looked in depth at two remarkable cases that had been observed in the Western Atlantic. Aware of an equally remarkable case in the Eastern Pacific—a region hardly noted for extreme events—I undertook, with Mark Albright’s help, a synoptic study of the case. During the late stages of the study, Rick Anthes paid a visit to the Department of Atmospheric Sciences and upon seeing the results of the research carried out thus far suggested that I collaborate with his junior colleague at NCAR—Bill Kuo—in an effort to extend the investigation by modeling the storm with use of the Penn State/NCAR mesoscale model. The synoptic study was completed and the results submitted for publication prior to my departure for England, but sufficient time did not remain to begin the collaboration with Bill. This had to await my return. During the year at ECMWF I became greatly impressed with the ability of state-of-the-art numerical models to perform realistic simulations of atmospheric developments so that I returned home in July 1986 not only eager to apply model diagnostics to explosive cyclogenesis, but also to polar lows and other synoptic-scale and mesoscale entities.

I have already alluded to the modeling effort on polar lows. Even more productive was the work done with Bill Kuo on explosive cyclogenesis, nearly ten papers resulting from our joint efforts (with valuable assistance from Simon Low-Nam and Georg Grell) in the period 1988-1996. The first of these treated the Eastern Pacific case; the others were based on cases observed in the Atlantic. Among specific items examined in the latter was the evolution of the storm structures, the airflow within the storms, the sensitivity of the developments to various physical processes—such as latent heat release in clouds and heat flux from the ocean surface—and the storm developments from the perspective of the "potential vorticity thinking" advocated by Hoskins, McIntyre, and Robertson.  Mark Stoelinga collaborated with Bill and me on the most ambitious of the potential vorticity papers and subsequently, working with the same case, Mark used a highly sophisticated scheme of potential vorticity inversion to assess the role of frictional and diabatic processes in the development. For this achievement he was awarded a doctorate in 1993, becoming the eighth PhD on my list of nine. In addition to the sensitivity studies carried out with Bill, I took advantage of my ECMWF connection to join with Adrian Simmons in looking into the role of latent heat release and surface fluxes in the development of a number of other major Atlantic storms.

My baseball team was completed in 1994 with Jordan Powers' successful attainment of his PhD. In his work he employed the Penn State/NCAR model to investigate three cases of mesoscale gravity wave development, evaluating the effects on the developments of such factors as background atmospheric structure, physical processes, and grid resolution. Before closing I would like to remark on two papers that I coauthored, one that appeared early in the post-ECMWF period and the other published this year (2001) with Bill Kuo and associates, which, barring a miraculous recovery of my health, is likely to be my final research contribution. In the first, coauthored with Dan Keyser, we presented a generalization of the Petterssen frontogenesis function and examined its role in the forcing of vertical motion in idealized cyclones. I had developed the basic idea while teaching my graduate synoptic course and showed Dan my class notes during a visit he made to the University of Washington. He informed me that he had recently obtained a similar result and suggested that we collaborate on a paper. I agreed.  But at the time I was in the middle of preparations for my sabbatical leave, precluding any immediate effort on my part. Dan took charge and expanded the topic in his inimitable style, producing a paper that, despite the fact that I can take little credit for its final form, engenders a special feeling of pride. Perhaps this is because the underlying idea was a theoretical one—at least from the standpoint of a lowly synoptic meteorologist.

The second paper dealt with a small or mesoscale cyclone in the Mediterranean Sea that bore a close resemblance to a tropical cyclone. The paper, which appeared in a recent issue of Meteorology and Atmospheric Physics, documented in detail the initial development and subsequent behavior of the storm and tested the ability of the NCAR model to simulate the observed sequence of events. A unique feature of the study, in the sense that I had not employed it previously in my work, was the use, thanks to my collaborators, of adjoint sensitivity to alter the initial conditions at one point in the simulation and from the altered state to obtain a revised forecast. This procedure, in combination with a vortex implant method, resulted in a considerable improvement of the "forecast."

In summary, during my early education I demonstrated the capacity for a career in science but felt no strong urge to pursue a scientific career, much less a career in meteorology. It was an accident of World War II that I was shunted into this field as an enlisted man in the Navy and a matter of rare good fortune that I was given the opportunity to obtain a BS degree in Meteorology while still in uniform. However, it was no accident that once introduced to the subject, my future course was set. From the start I loved meteorology with a deep-seated passion and experienced aesthetic feelings toward meteorological phenomena that motivated and sustained my efforts. These emotional attachments lasted throughout my career. Without them it is unthinkable that I could have achieved what I did. My choice of specific research topics was determined or inspired by a number of circumstances, ranging from healthy curiosity, to obligations imposed by others, to opportunities proffered by outside parties, to the occasional desire to strike out in new directions, to contact with directly experienced weather events, and, in my most notable work, to a series of improbable events. In a word there was no common thread except for an insatiable desire to observe and understand atmospheric phenomena.