I am a postdoc at UC-Berkeley and an affiliate of LBNL.
Currently, I am working primarily with Uros Seljak on finding methods to improve
constraints on cosmological models using observations of the
large-scale structure of the Universe.
My research interests span a variety of topics in theoretical cosmology, including
tests of dark energy models and signatures of reionization and
inflation in the cosmic microwave background. Previously,
I was a CCAPP postdoctoral fellow
at Ohio State.
I completed my Ph.D. in physics at the University of
Chicago, and before that was an undergraduate in Course 8 at MIT.
A few recent results from my research are summarized below. For more details,
click on the links to my publications and talks in the sidebar on the left.
Observational Probes of Cosmic Acceleration
Collaborators: David Weinberg, Daniel Eisenstein, Chris Hirata, Adam Riess, Eduardo Rozo
Our review of observational methods for studying cosmic acceleration and
improving constraints on theories of dark energy and modified gravity is
now available on arXiv.
We hope this will be a useful resource for graduate students and other
researchers who are interested in learning about the basic observational
techniques. In the review, we summarize the basic principles,
the current status of observations, and the main challenges that future
experiments are likely to face for
Type Ia supernovae, baryon acoustic oscillations, weak gravitational
lensing, galaxy cluster abundances, and other probes of acceleration.
We also present a new, comprehensive set of forecasts for the constraints
on cosmological parameters that these observational methods will provide
within the next 5 to 10 years.
Consistency tests of dark energy theories
Collaborators: Wayne Hu, Dragan Huterer, Ali Vanderveld, Tim Eifler
Two of the main methods we have to try to understand cosmic acceleration
are measurements of cosmological distances as a function of redshift and
measurements of the growth of large-scale structure in the universe.
Both distance and growth are affected by theories of dark energy or
modified gravity that attempt to explain cosmic acceleration, but
some theories that make the same predictions for distances make different
predictions for growth, and vice versa. So, by measuring distances and
growth, we can distinguish among the different theories.
Typically, distances are easier to measure accurately than the growth of
structure. Distance indicators such as Type Ia supernovae (SNeIa),
baryon acoustic oscillations (BAO), and the cosmic microwave background
(CMB) currently measure the distance-redshift relation with an accuracy
of a few percent or better over much of the history of the universe.
For the simplest theory of cosmic acceleration, a spatially flat universe with
a cosmological constant and cold dark matter (flat ΛCDM),
we found that these measurements of distances can be used to
make predictions for the growth of structure with percent-level
The figure on the right shows the flat ΛCDM predictions from
current SNeIa, BAO, and CMB data, plus a measurement of the Hubble
constant (H0). Shading shows the regions with 68%
confidence level (CL), and solid curves mark the boundaries of the
95% CL regions. The top panel shows predictions for the growth function
G and for the same function with an alternate normalization
G0, the middle panel shows predictions
for the differential growth rate fG and the growth index γ
(often used as
a diagnostic for modified gravity), and the bottom panel shows predictions
for the extrapolation of distance D to high redshifts
and for the Hubble expansion rate H.
Signatures of reionization in the CMB
Collaborators: Wayne Hu, Cora Dvorkin, Hiranya Peiris
During the epoch of reionization, hydrogen atoms in the intergalactic
medium are ionized by radiation from the first stars and quasars.
The resulting free electrons rescatter photons
from the CMB, reducing the amplitude of CMB temperature anisotropies and
generating additional CMB polarization anisotropies on large scales.
The effect on the angular power spectrum of the CMB temperature
(CℓTT) is difficult to distinguish
from changes to the amplitude of primordial fluctuations from inflation,
but in the E-mode polarization power spectrum
(CℓEE) reionization produces a
detectable feature on large scales (low ℓ). From WMAP measurements
of the amplitude of this feature, we know that the total optical depth
to reionization is approximately τ=0.09, which implies that
reionization occurred at a redshift of about 10.
While measurements of CℓEE can
determine the total optical depth τ fairly well, it is more difficult to
tell whether the transition from a neutral medium to an ionized medium
happened quickly or slowly. The figure on the right shows two very
different ionization histories with similar values of τ (see inset)
and their corresponding polarization power spectra. Although there are
clear differences in the spectra at ℓ<30, data from WMAP are
not precise enough to distinguish between these models
(see points and error bars,
showing 3-year WMAP data). Until the large-scale polarization measurements
improve with data from upcoming experiments like Planck, the detailed
evolution of the ionized fraction during reionization will remain
largely unknown. Given this uncertainty, most analyses of CMB data
assume a simple form for the ionization history (e.g., an instant
In a series of papers, we studied the impact of this uncertainty in
the ionization history on the estimated values of cosmological parameters
using a general parametrization of the ionization history based on
principal components. With current WMAP data, the impact is fairly
small: relative to an analysis that assumes instant
reionization, accounting for uncertainty in the ionization history
slightly increases both the central value of the estimated optical depth
(to about 0.10) and the uncertainty in τ. Even this small change
can significantly affect parameters correlated with τ.
For example, evidence for deviations from a scale-invariant primordial
power spectrum weakens when reionization uncertainty is considered.
Future CMB polarization measurements from Planck and other experiments
will improve constraints on reionization but will also
be more susceptible to assumptions about the form of the ionization history;
for Planck, accounting for uncertainty in the evolution
of the ionized fraction is expected to increase the error on τ from 0.005