Nicholas Kaiser

Ph.D. 1983, Cambridge University
  • Theoretical astrophysics and cosmology
IfA Mānoa, C-212
(808) 956-8580(808) 956-8580


I work mostly in the general area of “observational cosmology”. I helped develop the theory of how to use observations of the microwave background, galaxy clustering, large-scale `cosmic-flows’, evolution of massive clusters of galaxies and gravitational lensing to probe dark matter and dark energy and test theories for the origin and evolution of the Universe. I also led the development of the Pan-STARRS wide-field survey telescope system.

Awards and honours:

  • Ontario Fellow of the CIAR Cosmology Program (1988)
  • Helen Warner Prize of the American Astronomical Society (1989)
  • Canadian NSERC Steacie Fellowship (1991-92)
  • Herzberg Medal of the Canadian Association of Physicists (1993)
  • Rutherford Medal of the Royal Society of Canada (1997)
  • Fellow of the Royal Society (2008)
  • U. Hawaii Regents Medal for Excellence in Research (2014)
  • Royal Astronomical Society Gold Medal (2017)


Nick Kaiser’s Research Achievements

Kaiser has made many important theoretical contributions to modern observational cosmology.  His PhD thesis was on Anisotropies of the Cosmic Microwave Background (CMB) and he has subsequently diversified into many other areas, mostly related to analysis of observational probes of large-scale structure and dark matter.

Cosmic Microwave Background Anisotropy.  Kaiser started his career working with Martin Rees on a variety of problems relating to anisotropies of the CMB, which was then emerging as a practically useful probe of cosmology.  This was also the time that models such as the Cold Dark Matter (CDM) model began to emerge and there was much activity in generating quantitative predictions for observations. His thesis resulted in two major papers; one analyzing the CMB anisotropy in open models via analytic modeling and the second making quantitative calculations of small-angle anisotropies including, for the first time, polarization in the radiative transfer [1].  An important result from this paper was that the damping rate for adiabatic fluctuations was significantly modified by the inclusion of polarization; a factor that is now of some importance in the era of precision cosmology.

An important source of uncertainty in these predictions was the neglect of reionization, which can erase primary anisotropies.  On taking up a post-doctoral fellowship at Berkeley, Kaiser performed the first detailed analysis of secondary anisotropies that would be generated by Doppler scattering off inhomogeneous ionized gas [2].

Subsequently, Kaiser and graduate student Albert Stebbins were the first to predict the characteristic line-like discontinuities in the microwave sky generated by cosmic strings [4].

Biased Clustering of Clusters and Galaxies.  Kaiser then turned his attention to a long-standing mystery in large-scale clustering studies: while both cluster and galaxies display power-law correlations, the amplitude for clusters is much higher and increases with richness.  In a seminal paper [3], Kaiser showed that this could naturally be accounted for in a simple model where clusters of galaxies are associated with peaks of the primordial mass fluctuations when smoothed on an appropriate scale.  While developed to account for the anomalous clustering of clusters, this result was seized upon by advocates of the CDM theory as a possible way to reconcile the low dynamical mass estimates with an Einstein-de Sitter cosmology, and this was dubbed biased galaxy formation. This idea was developed in some detail in the famous `BBKS’ paper [5], and biased galaxy clustering is a central concept in modern studies of galaxy clustering.

Cluster Evolution.  In the late ’80s and early ’90s Kaiser developed powerful analytic models for the evolution and clustering of massive clusters.  In 1986 he used scaling laws (based on the locally power-law like behaviour of the initial fluctuation spectrum in CDM models) to predict the evolution of clustering of massive clusters [6], and this has formed the basis for much subsequent work.  In 1991 [10], he showed that X-ray observations showed deviations from pure scaling expected from gravitational clustering, but that these deviations could be understood if there had been entropy injection into the intra-cluster gas from early galaxy formation, and this concept has held up well in the light of more recent studies.

Redshift Space Distortions.  In 1987, Kaiser published a detailed analysis of the effects of peculiar motions on the apparent structure as measured from from redshift surveys [7].  Of particular importance was the prediction that, on large scales, the clustering pattern would appear to be squashed along the line of sight — in contrast to the then well-known `finger of god’ effect that causes elongation along the line of sight on small scales.  This effect has now been dramatically verified in the 2dF survey and provides a direct way to measure the rate of growth of structure.  This in turn provides crucial constraints on the matter and dark energy densities.

Bulk Flows and Galaxy Clustering.  Over his career, Kaiser has been instrumental in developing quantitative analysis of various probes of large scale structure.  Following the discovery of the `Great-Attractor’ by Burnstein et al, he devoted considerable effort to how these `bulk-flows’ could be used to constrain the initial spectrum of fluctuations and also the density of the universe [8] and applying this analysis to the IRAS redshift survey [9], a survey that successfully exploited the `sparse-sampling’ strategy that Kaiser had proposed as a way to make more efficient use of telescope time.

Another exciting observational development at the time was the suggestion of large-scale periodicity in the distribution of galaxies in pencil beam redshift surveys.  In his paper with Peacock [11], Kaiser showed that the spikes like those seen in the power spectrum would arise quite naturally.  This analysis was extended to show how the spectrum of mass fluctuations could be optimally measured from 3-dimensional galaxy redshift surveys [15], and this technique has been widely adopted by observers.

Weak Gravitational Lensing.  In the early ’90s, a new observational probe of structure appeared on the scene.  Pioneered by Tony Tyson, this was the measurement of distortion of shapes of faint galaxies caused by gravitational lensing by clusters and by galaxies.  While clearly providing a powerful new window on the dark matter, the detailed relation between the mass and the observed shear was initially only poorly understood.  In a series of papers, Kaiser first showed how to link the power spectrum of mass fluctuations to that of the shear [12]; with graduate student Gordon Squires he showed how one could reconstruct the projected mass density in clusters in  the weak-lensing limit [13]; and he then extended this into the strong-lensing limit [16].  With Broadhurst and Squires, Kaiser (KBS) developed a practical technique for quantitative shear estimation [17], which is still widely used and remains competitive with all other methods for WL measurement.  In [18] Kaiser generalized WL spectral theory to treat realistic cosmologies and in [20] proposed a new technique that expands on KBS in using the convergence, or magnification, together with the polarization to improve shear measurement.

Wide Field Imaging Surveys.  With Tonry and Luppino [19], Kaiser proposed a new type of wide-field survey telescope exploiting low-order adaptive optics. This led to the development of the Pan-STARRS project that Kaiser and colleagues at Hawaii devoted approximately one decade to developing. The PS1 telescope commenced operations around 2011 and carried out a highly productive 3.5 year survey and the  PS1+2 two telescope system is currently largely devoted to near earth object searches, but is also now becoming active in follow up of optical counterparts to LIGO gravitational wave bursts. The data from the 3.5 year survey have now been made publicly available through the STScI MAST archives.

Gravitational Redshifts in Galaxy Clusters.  In 2011, Wojtak et al. made the first measurement of cluster gravitational redshifts.  Zhao et al. pointed out that one needed to also consider the transverse Doppler effect.  In [21] Kaiser showed that there were two further, previously ignored, effects that appeared at the same order of magnitude as the naive gravitational redshift: One of these is the light-cone effect; images of clusters will tend to contain an excess of objects that are moving away from the observer. The other is that the transverse Doppler effect, when averaged over photons, is actually a blue-shift, and that this counteracts the simple transverse Doppler red-shift when one is dealing with magnitude limited samples.

The Nature of Astronomical Redshifts.  The tiny gravitational redshifts observed in clusters of galaxies raise some interesting questions about the nature of redshifts more generally.  In both flat space-time, and in non-empty FRW cosmologies, where space-time is curved, the wavelength of photons changes in lock-step with the proper separation of observer and source.  This led to the interesting argument of Bunn & Hogg that all redshifts can, and should, be considered to be `kinematical’ in nature. But applied to a non-expanding cluster, this would suggest that there would be no observable effect.  In [22]  Kaiser showed that the redshift can be decomposed into a kinematic  part plus an intrinsically gravitational redshift, where the latter is an integral of the gradient of the tidal field along the line-of-sight.  This shows that the kinematic behaviour seen in FRW models is a happenstance of the symmetry of these models in which the tidal field is spatially constant.

Perturbations to the Luminosity Distance – Redshift Relation.   Cosmic flow measurements probe large-scale structure using its effect on the cosmological distance-redshift relation D(z).  Initially analysed in a quasi-Newtonian manner, and first measured in the 70’s, this technique was reconsidered (or rediscovered) by relativists and cosmologists in the new millennium. It was shown that this is important for analysis of SN1a measurements and claimed that this provided a new probe of cosmic structure dubbed `Doppler lensing’. Interestingly, these studies gave results that appear to violate the equivalence principle. Kaiser & Hudson [23] showed how the unphysical results resulted from neglecting the gravitational redshift and thus provided a physically consistent framework for analysing supernova observations. 

Non-linear relativistic cosmological perturbation studies have found biases in the mean luminosity distance and distance modulus at low redshift.  Kaiser & Hudson [24] showed that these effects may be understood as a non-relativistic,  and purely kinematic, Malmquist-like bias.  This effect is essentially identical to the distance bias from small-scale random velocities  that has previously been considered by astronomers, though we find that the standard formula overestimates the homogeneous bias by a factor 2.

Biases in D(z) from Gravitational LensingA long-standing question in cosmology is whether gravitational lensing changes the distance-redshift relation D(z). Interest in this has been rekindled by recent studies in non-linear relativistic perturbation theory  that claim to find large biases in both the area of a surface of constant redshift and in the mean distance to this surface. If correct this would significantly impact both CMB and the SN1a supernova Hubble diagram, two of the major under-pinnings of modern cosmology.

Kaiser & Peacock [25] have shown show that the perturbation to the area of a surface of constant redshift is in reality much smaller than claimed, being of the order of the cumulative bending angle squared, or roughly a part-in-a-million effect. This validates the arguments of Weinberg that the mean magnification of sources is unity. It also validates the conventional treatment of CMB lensing. But the existence of a scatter in magnification will cause any non-linear function of these conserved  quantities to be statistically biased. The fractional bias in such quantities is generally of the same order of magnitude as the mean squared convergence,  which is orders of magnitude larger than the area perturbation. Claims for large bias in area or  flux density of sources appear to have resulted from misinterpretation of such effects: they do not represent a new non-Newtonian effect, nor do they invalidate standard cosmological analyses.


I developed an intensive graduate level course in theoretical astrophysics.  The notes for this can be found here.

I also teach the “stars, galaxies and cosmology” undergraduate astrophysics course A242.