UFalcon 0.2.0

Ultra Fast Lightcone

Package for constructing full-sky weak lensing maps from lightcones based on N-Body simulation output within a minimal runtime. Written in Python 3.

Introduced in Sgier et al. 2019 and extended in Sgier et al. 2020 (in prep.).

Why use UFalcon?

In order to accurately infer cosmological constraints from current and future weak lensing data, a large ensemble of survey simulations are required to model cosmological observables and a well-converged covariance matrix. UFalcon applied to PKDGRAV3-output combines accuracy and minimal computational runtime: The simulation of the density field guarantees to satisfy a certain force accuracy and is therefore not an approximate N-Body code. Furthermore, the PKDGRAV3 code is highly efficient and can be run with graphics processing units (GPU) support. The subsequent post-processing with UFalcon can be parallelized on a computer cluster and has a runtime of less than 30 min walltime per convergence mass map. The package offers a high flexibility for the lightcone construction, such as user-specific redshift ranges, redshift distributions and single-source redshifts. Furthermore, UFalcon offers the possibility to compute the galaxy intrinsic alignment signal, which can be treated as an additive component to the cosmological signal.

Features

• Fast computation of fullsky Healpix maps (Gorski et al. 2005) containing particle counts (shells).

• Fast construction of weak lensing maps (convergence, shear) for user-specific redshift distributions and single-source redshifts.

• Computation of galaxy intrinsic alignment (IA) signal (additive to the cosmological signal) based on the nonlinear intrinsic alignment model (NLA) (Bridle et al. 2007 , Hirata et al. 2004 and Joachimi et al. 2011) and applied in Zürcher et al. 2020.

N-Body Simulations

Currently supported N-Body simulation codes are:

• PKDGRAV3 (Potter et al. 2016) available here.

• L-PICOLA (Howlett et al. 2015) available here.

Note that UFalcon currently only supports post-processing of simulation output generated in lightcone mode.

Getting Started

The following sections provide an overview of the UFalcon pipeline and some example-scripts, which you can use for your analysis.

Credits

• Copyright (C) 2020 ETH Zurich, Institute for Particle Physics and Astrophysics, Cosmology Group

• Authors: Raphael Sgier and Jörg Herbel

• Contact: Raphael Sgier [e-mail].

Documentation

The full documentation can be generated with Sphinx

:

History

0.1.0 (2019-04-04)

• Development version

0.1.0 (2020-04-06)

• First version on TestPyPI

0.2.0 (2020-05-14)

• Completely moved all cosmological computation to astropy

• support for continuous (not tabulated) redshift distribution

• Run external test with established pipeline

• Extended tests for lensing weights