WILKINSON, DAVID TODD

(b. Hillsdale, Michigan, 13 May 1935;

d. Princeton, New Jersey, 5 September 2002), experimental physics, observational astrophysics, observational cosmology, studies of the cosmic background radiation.

Wilkinson was one of the most respected and beloved physicists of the past half century. He was the acknowledged leader of research in the cosmic microwave background (CMB) and an exceptional teacher of professional physicists, as well as a successful expositor of science to the general public. He was a model for experimental physicists in both his style of research and his genuine lack of pretense and ability to focus colleagues on the job that needed to be done to get to the science. A testimony to his modesty lies in his curriculum vitae, which contains the minimum information required and, certainly, no reference to the honors and awards that he received during his professional life. We now know that he was awarded an honorary doctorate from the University of Chicago in 1996 and in 2001 received the James Craig Watson Medal for astronomy of the National Academy of Sciences (of which he was a member) for his and his student’s elegant experiments to measure the spectrum of the cosmic background radiation.

Beginnings with Atomic Physics . Wilkinson—known to all as “Dave”—was a product of his Midwest rural background in Michigan. He attended the University of Michigan at Ann Arbor, beginning in engineering physics as an undergraduate. As a graduate student he received a master’s degree in nuclear engineering and then moved into atomic physics to gain a PhD in 1962 with H. Richard Crane in a second iteration of the electron g-2 experiment. At the time precise experiments to measure the consequences of quantum electrodynamics (QED) were at the forefront of fundamental physics research. The key element of the g-2 experiment was that it provided a direct observation of the quantum corrections to the electron magnetic moment.

Moreover, the observation was made in a free-particle system with no need to estimate corrections derived from messy calculations of the many-body problem in atoms. Wilkinson, as he wrote in a review paper (1967), believed that a fundamental result would be best understood (and believed) in a system which was pristine and for which one could directly measure the systematic terms that influenced the result. He achieved a result precise to a part in 10⁸, good enough to show the quantum corrections at the level of the square of the fine structure constant to a part in ten thousand. The result confirmed the then-current theory. A prescient comment at the end of the g-2 paper (1963) states that the next-order corrections would be too difficult to measure by the technique used; a new idea would be needed.

The experience with the g-2 experiment carried forward to much of his research later. The experiments he would consider had to be fundamental, they had to be understood at the most basic level, and the systematic problems had to be measured and eliminated or at worst modeled. He recognized that developing technology to achieve a result was part of experimental physics. Finally, there was a time to quit and go on to imagine and invent a new idea or technique to get at the science.

Early Princeton Times. In 1963 Wilkinson was drawn to the brilliance and elegance of the experiments of Robert H. Dicke at Princeton University. Dicke was exploring the gravitational interaction as a branch of physics and not merely mathematics. He was attempting to establish the experimental basis for the general theory of relativity and knew enough of the rapid developments in technology to apply them to new measurements and observations. Princeton had become the world center for experimental relativity. The Dicke gravitation group was an intellectually exciting environment for a young physicist interested in fundamental experiments.

Precision measurement of the Moon’s orbit around Earth using laser ranging was one idea being considered. Wilkinson became part of a group that designed the array of corner reflectors placed on the Moon by the Apollo project. The corner reflectors have been one of the most enduring legacies of the Apollo program. With centimeter ranging precision using ground-based telescopes as laser transmitters and receivers, the ranging program established that the gravitational self-energy of Earth satisfies the weak principle of equivalence (Nordtvedt effect) and set stringent limits on the amplitude of the scalar part of the scalar-tensor theory.

As part of the investigations of the consequences of a scalar-tensor theory, Dicke began to think about the influence of the scalar component on cosmic evolution. At the time Dicke favored an oscillating universe and reasoned that, since the universe was not primarily made of iron (the nucleus with maximal binding energy per nucleon), there had to be heat to dissociate the nuclei remaining from prior cycles. In one of the evening meetings of the group, he suggested that one ought to think about making an observation of the radiation left from this hot epoch. Jim Peebles began to look into this theoretically and Dave Wilkinson and Peter Roll began to design an experiment. This was the beginning of cosmic microwave background research at Princeton, which in the hands of Wilkinson and his students and colleagues resulted in a revolution in our understanding of cosmology.

Initial CMB Spectrum Measurements. The story that led to the Nobel Prize–winning but accidental discovery of the 3 K background by Arno Penzias and Robert W. Wilson at Bell Labs is described in many cosmology books. The critical paper authored by Dicke, Peebles, Roll, and Wilkinson (1965) gave the rationale for the 3 K measurement at 7 centimeters wavelength made at Bell Labs. At the time of the discovery the Princeton group was well on the way to making their own measurements at X-band (3.2 cm).

By the end of 1968 Wilkinson, with his student Robert Stokes and colleagues Paul Boynton and R. Bruce Partridge, had made measurements of the cosmic background spectrum at 3.2, 1.58, 0.86, and 0.33 centimeters. The data, combined with the longer wavelength Bell Labs measurement, showed a spectrum that was closely thermal in the Rayleigh-Jeans part of the Planck blackbody curve of 2.7 K and also gave hints of a rollover at short wavelengths, the real telltale for a blackbody. The observations were all made from the ground using the type of differential radiometer Dicke had invented in his early work at the end of World War II.

The observations were not easy. They were absolute measurements in which one had to account for all the radiation entering the receiving system. The incoming radiation was ultimately compared with that from a source with known emission properties at a known calibration temperature. It was relatively easy to establish the properties of the calibration source using Kirchhoff’s laws of radiation but it was not easy to account for every scrap of radiation entering the observation port.

Any hot surface at 300 K (about room temperature) radiated more than the universe, it was very important not to “look” at any hot surface even with the wings of the antenna beam. The atmosphere itself emitted and this needed to be accounted for by measuring the amount of atmospheric radiation at different zenith angles. Finally, it was critical to account for the emission by foreground astrophysical processes in our own galaxy such as synchrotron, free-free, and dust emission. The relative importance of these foregrounds depends on the frequency of observation.

By the early 1970s Wilkinson realized that the point of diminishing returns of spectrum observations from the ground was approaching and that new ideas were needed. Some other groups had begun to use balloonborne observations to at least eliminate the atmospheric emission and to allow explorations of the spectrum in the region embracing the 3 K blackbody peak. Although he eventually also began to use balloon platforms for spectral measurements, Wilkinson’s attention turned to measurements of the angular distribution of the cosmic radiation.

Anisotropy of CMB Angular Distribution. Wilkinson and Peebles had the realization that the cosmic microwave background should not be the same in every direction. Instead it should exhibit a dipole anisotropy due to Earth’s motion relative to the average rest frame of the universe (or expressed another way, relative to the last scatterers of the radiation, which in effect form a plasma wall around us at a red shift of ∽1000). The major part of the velocity was expected to be due to the rotation of our galaxy (there is also a component from a galactic translation relative to the Hubble flow). The estimates for the dipole anisotropy were a fractional change of temperature of about a part in a thousand, a few millikelvin. In addition to this kinematic anisotropy, an intrinsic anisotropy due to gravitationally induced density fluctuations was expected. These fluctuations would in later epochs grow to become the large-scale structure of the universe. The amplitude and the angular scale of the fluctuations depend on the constituents of the primeval universe and were initially not well bounded by theoretical considerations.

The detection of the anisotropy became a new frontier in studies of the CMB and a new probe of the processes at work in the cosmic explosion. During this time it also was realized that the evolution of the universe was a giant physics laboratory sampling energies unattainable in particle accelerators and that the relics from the various epochs of the evolution of the universe may be still around for us to observe and measure. The synthesis of fundamental particle physics and cosmology had begun.

Wilkinson embarked on both large- and small-scale anisotropy measurements from the ground with the aim of measuring the dipole on large scales and intrinsic anisotropies on the smaller scales. The experiments carried different challenges than the absolute experiments to measure the spectrum. Now, sensitivity became important. The temperature differences expected were 100 microkelvin or smaller. There were many systematic effects which could produce false anisotropies: anisotropically distributed near radiators, lumpiness in the atmosphere, and structure in the galactic foregrounds. For large angular-scale measurements, it became important to reduce the atmospheric noise and to be able to operate at multiple wavelengths to understand and be able to remove the galactic foregrounds. Balloon experiments, even with their additional complexity and inconvenience, became the method of choice.

By the end of the 1970s both the Princeton group (Paul Henry, Paul Boynton, Edward Cheng, Peter Saulson, Stephen Boughn, and Brian Corey), using balloons, and a group at Lawrence Berkeley (Luis Alvarez, George Smoot, Mark Gorenstein, and Richard Muller), using the U2 airplane as an observing platform, had established the dipole anisotropy. The intrinsic anisotropy remained elusive. In the early 1980s Wilkinson made a major technical advance with Boughn, Dale Fixsen, and Cheng to apply a cryogenic low-noise MASER amplifier to a differential radiometer at 24.5 gigahertz to improve the sensitivity in a search for the intrinsic anisotropy.

COBE and WMAP: Satellite Measurements of theCMB. In 1972 NASA solicited proposals (Announcement of Opportunity AO6 and AO7) from a wide range of disciplines. John Mather, who had just received his PhD with Paul Richards at Berkeley, pulled together a team of scientists to propose a new satellite called the Cosmic Background Explorer (COBE). Mather conceived of a constellation of three mutually interdependent experiments to study the cosmic background spectrum and anisotropy with precision. A liquid helium dewar contained a Fourier Transform Infrared Absolute Spectrometer (FIRAS) to measure the spectrum between 3 to 20 cm^-1 and 20 to 50 cm^-1 as well as a radiometer to measure the Diffuse Infrared Background Radiation (DIRBE) in ten bands between 300 microns and 1 micron. Outside the dewar the satellite carried three Differential Microwave Radiometers (DMR) operating at 30, 50, and 90 gigahertz. The intent of the mission was to determine the spectrum and anisotropy of the cosmic background to a precision limited only by the ability to measure and model the foreground emission, and to search for an infrared background thought to arise from the formation of the first stars.

Wilkinson was part of the initial scientific group that proposed the COBE mission, which eventually grew to include twenty-one members who worked together for about twenty-five years. Although Wilkinson did not want an in-line position in this group, he played a key role in the project. His wisdom derived from prior experience and his continuous attention to the systematics and their potential effect on the science were critical to the mission success.

COBE measured the cosmic background spectrum with unprecedented precision, showing no deviations from a blackbody to a part in 10⁴ in the 3 to 20 cm^-1 band, the region embracing the peak. The universe was truly in thermal equilibrium at the time of decoupling. The blackbody temperature is 2.725 ± 0.002 K, (1994a) a value consistent with those measured by Wilkinson in the early days of the field.

The intrinsic anisotropies of the CMB were finally observed, extending from the largest angular scale (the quadrupole moment) down to the COBE DMR beam size of 7 degrees (1994b). The root-mean-square amplitudes were in the 100 microkelvin region and the spatial spectrum followed a Harrison-Zel’dovich power law with amplitude independent of scale. The observations supported inflation theories of the universe’s origin and strikingly demonstrated the need for cold dark matter as the source of the density perturbations that lead to the largescale anisotropy. The observation gave strong support to the idea that the structure seen in the surface of last scattering was due to quantum fluctuations of the much earlier universe preserved by the dark matter, which does not diffuse or respond to buffeting from the photons during the expansion.

Once the COBE mission was functioning, Wilkinson began planning the next step. He had experienced the significant advantages offered by space observations in these fussy experiments. In particular, he noted the ability to observe for years from a platform with a benign (unchanging) environment and, if the instrument was designed properly, the ability to measure and control the systematic errors that plagued both ground-based and balloonborne observations. The advantages and the elegance of the science one could do outweighed the painful prospects of the many meetings and conference calls as well as the politics of big science.

He jumped in with both feet and organized the collaboration between the NASA Goddard Space Flight Center and Princeton to design, build, and fly the Microwave Anisotropy Probe (MAP). The mission measures the CMB anisotropy in five bands between 23 and 96 gigahertz on angular scales extending from 0.2 (96 GHz) to 0.9 (23 GHz) degrees. Multifrequency observations were used to separate foregrounds from the CMB. Wilkinson insisted that the mission use differential radiometers, which had been found to exhibit low systematic errors, and, furthermore, that the satellite be placed at the L2 Lagrange point of the Earth-Sun system, where the solar illumination is invariant over the year (1999). At Goddard Charles Bennett, who had been the deputy principal investigator on the COBE DMR experiment, led the effort, while at Princeton Lyman Page and Norman Jarosik took the leads in the development.

The MAP mission was renamed the Wilkinson Microwave Anisotropy Probe (WMAP) in honor of his critical role. The publication of the first results from this mission (2003) was indeed a major step in precision cosmology and the best testimony to his unrelenting attention to performing a good and trustworthy experiment. WMAP results show structure in the last scattering surface due to acoustic oscillations in the plasma (one of these acoustic peaks had been observed from the ground earlier by Page and his students and confirmed by other groups). The angular scale of the acoustic peaks gives a standard length at a known distance and allows for the determination of a large set of cosmological parameters. The data show the universe to have a flat geometry; to be composed of specific proportions of baryons, dark matter, and dark energy; to have a well-determined age; and (besides the acoustic peaks) to have an approximately scale-invariant power spectrum of density fluctuations. In addition, WMAP, through its ability to measure the polarization of the radiation, has found when the universe was reionized at a later epoch when the first stars were formed.

On the larger angular scales, the WMAP and COBE results are consistent. More WMAP results are still to come at the time of this writing. The measurements of the CMB polarization (which Wilkinson began in the early 1970s with George Nanos, a student) is expected to start a new branch of cosmological investigations into processes during the very earliest moments of universal expansion, which may have generated gravitational waves that subsequently polarized the cosmic background through Thomson scattering. Suzanne Staggs, once Wilkinson’s student who worked on a spectrum measurement at 1.4 gigahertz, has concentrated her research on the measurement of the polarization patterns in the cosmic background.

Other Significant Astrophysical Research. Although studies of the primeval cosmic explosion are the major theme in Wilkinson’s scientific life, he had other interests in observational astrophysics. In 1968 it was discovered that the Crab Nebula pulsar (NP 0532) emitted optical pulses in addition to the radio pulses by which the pulsar was initially discovered. Sensing that there would be an interesting experiment using pulsars as clocks in different gravitational potentials, Wilkinson joined with Partridge, Boynton, and Edward Groth, then a graduate student, to measure the long-term frequency stability of the Crab pulsar. They developed a rapid transient photometer with high quantum efficiency to observe the pulses with high timing precision. The instrument was versatile and could be used effectively on a small telescope; hence it was portable and one did not have to compete for precious telescope time.

To their amazement and (possibly horror) they discovered that the Crab pulsar was an irregular clock. The clock experiment they had contemplated was a failure but they had discovered that the moment of inertia of the pulsar changed discontinuously; the pulsar seemed to have a solid surface which experienced star quakes. Much later in his life Wilkinson returned to this type of transient detector in his search (with Paul Horowitz, who by the way had also observed the optical pulses from the Crab) for extraterrestrial civilizations that might make themselves known by emitting optical pulses—OSETI.

Early on Jim Peebles and Dave Wilkinson began thinking about the other consequences and observable relics of the evolving cosmology that was being discovered. The various epochs in the cosmic explosion would have left fossils. An epoch of particular interest to the Princeton group was the time when the first stars and galaxies formed. How did the first luminous objects form in the large-scale structure set by the distribution of density fluctuations that originated in the prior plasma? An observational program was contemplated to look for the infrared light from redshifted ultraviolet light of the nuclear burning from the first aggregations—the so-called second ignition in the universe.

This program was part of the motivation for Wilkinson to investigate the use of charge-coupled devices (CCD) as array detectors in astrophysical measurements. The devices at the time had been developed for the military surveillance program from space. Wilkinson recognized that they also could have properties useful to astronomy and could even be a substitute for film if the arrays were made with sufficiently large formats. They could be used in the low light levels associated with astrophysical measurements and, in distinction to film, would be linear over a large dynamic range. Furthermore, one could accumulate charge in the pixels over time without loss and thereby integrate, especially if the CCD were operated at low temperatures. Finally, they should have much larger quantum efficiency than film and could be effective in the silicon band extending to wavelengths as long as 1.1 microns.

Wilkinson and his group learned how to convert CCDs to astronomical observations and devised methods to deal with their flaws. In the early 1980s a set of students including Stephan Meyer, Mark Bautz, and Edward Loh used the CCD detectors to measure the emission by lowbrightness red early galaxies and other low luminosity/mass ratio objects.

Teaching and Public Service. Wilkinson was the Cyrus Fogg Brackett Professor of Physics (a position that does not require teaching) and onetime head of the Physics Department at Princeton. He performed his departmental duties with care and excellence. He also performed a large number of public service functions for science by chairing numerous committees and studies for the National Science Foundation, NASA, and the National Academy of Sciences. His abiding interest was in teaching. Throughout his life as a scientist Wilkinson held the conviction that the most valuable thing he could do for others was to have them experience his joy in observing nature and understanding how it worked. Research was a medium for human interactions and a means of transmitting the pleasure of learning. There was never a dichotomy between research and teaching; they were part of the same process.

This approach applied not only to his graduate students but to his interaction with undergraduates and his own children. The interest in helping people understand science and the pleasure of knowing brought him into the Princeton Science and Technology Council, which was addressing the problems of bringing rational and quantitative reasoning to the average student, not just those interested in science and technology. As related by several of his colleagues, Wilkinson was one of the most influential and successful educational reformers at the university. In 1996, the year of Princeton’s 250th anniversary, he was given the Outstanding Teacher Award. His style of leading by example but from behind was not intimidating and drew people in as though they were the initiators.

In the process of organizing lectures and seminars for this program he invited Paul Horowitz to talk about the search for extraterrestrial intelligence (SETI). Horowitz explained that he was seeing a few optical pulses per day at the Harvard Observatory that could be candidates from an extraterrestrial intelligence. Wilkinson saw that it would be useful to make coincident measurements at Princeton to confirm that the pulses were coming from outside the atmosphere. This was the origin of OSETI at Princeton, which became a communal, university-wide science project to refurbish an existing telescope and to operate it with amateur observers. It turned out to be a marvelous way to get many people, not necessarily scientists, into the action.

Dave Wilkinson balanced a life in science with one rich in enjoying family and the pleasures of exploring nature in the laboratory and in the wild. He dealt with the cancer that afflicted him for the last fourteen years of his life with dignity and grace. He was active to the end.

BIBLIOGRAPHY

A list of Wilkinson’s publications can be found at http://www.physics.princeton.edu/www/jh/history/wilkinson_david.html.

WORKS BY WILKINSON

With H. R. Crane. “Precision Measurement of the g Factor of the Free Electron.” Physical Review 130 (1963): 852–863. One of Wilkinson’s earliest but also seminal papers was the subject of his PhD thesis.

With R. H. Dicke, P. J. E. Peebles, and P. G. Roll. “Cosmic Black Body Radiation.” Astrophysical Journal 142 (1965): 414–419. His most quoted paper, which interpreted the measurements by Penzias and Wilson.

“Properties of Free Electrons and Positrons.” In Methods of Experimental Physics 4. Atomic and Electron Physics, Part B:Free Atoms, edited by Vernon W. Hughes and Howard L. Schultz. New York: Academic Press, 1967.

With J. C. Mather, et al. “Measurement of the Cosmic Microwave Background Spectrum by the COBE FIRAS.” Astrophysical Journal 420 (1994a): 439–444. The spectrum measurement made by COBE.

With C. L. Bennett, et al. “Cosmic Temperature Fluctuations from Two Years of COBE DMR Observations.” AstrophysicalJournal 436 (1994b): 423–442. The paper announcing the discovery of intrinsic cosmic anisotropy.

With L. Page. “Cosmic Microwave Background Radiation.” Reviews of Modern Physics 71, no. 2 (1999): S173–S179. A summary of the observational state of the CMB.

With C. L. Bennett, et al. “First-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Preliminary Maps and Basic Results.” Astrophysical Journal Supplement Series 148 (2003): 1–27. The first WMAP results opening the epoch of precision cosmology.

OTHER SOURCES

Chang, Kenneth. “Dr. David T. Wilkinson, 67, a Physicist Who Searched for Big Bang’s Echoes, Is Dead.” New York Times, 8 September 2002.

Rainer Weiss