Weak Lensing with the SKAO: Radio Shear Measurement
Abstract
Cosmic shear measurements have traditionally been dominated by optical surveys, which offer higher resolution, better sensitivity, higher galaxy number density, and wider area coverage than their radio counterparts. With the advent of upcoming radio surveys, particularly those planned with the SKA-Mid, we would reach the high sensitivity and resolution required for weak lensing studies, allowing radio observations to begin competing with optical surveys in this domain. However, radio observations are affected by fundamentally different instrumental and astrophysical systematics, meaning that shape and shear measurement techniques developed for optical surveys cannot be straightforwardly applied. Fully realizing the weak lensing potential of these next generation radio surveys therefore requires the development and validation of methods tailored specifically for radio datasets. In this chapter, we present a simulation pipeline to generate realistic observations of isolated radio galaxies based on the SKA-Mid AA4 array configuration. We apply three recent radio shape measurement techniques: SuperCALS, RadioLensfit, and DeepShape, to assess their performance on these simulations. Our results show that RadioLensfit and DeepShape yield the most accurate shape estimates, although RadioLensfit is significantly more computationally intensive. We also perform shear recovery on simulated data using RadioLensfit and DeepShape, finding multiplicative and additive shear biases of order a few 10−2 and 10−4, respectively. Finally, we highlight the challenge of source separation, which will play a critical role in the success of future radio weak lensing analyses.
Related
- Weak lensing shear and aperture mass from linear to non-linear scales
- Analytical weak-lensing shear response of galaxy model fitting
- Two-dimensional weak lensing shear for cluster mass, concentration, ellipticity and miscentering estimation
- Low-variance renoising for weak gravitational lensing shear estimation
- Weak-Lensing Shear Response for Photometric Redshift-Based Tomographic Binning
BibTeX
@misc{tripathi2026weak,
title = {{Weak Lensing with the SKAO: Radio Shear Measurement}},
author = {Priyamvad Tripathi and Marzia Rivi and Simon Prunet and Ian Harrison and André Ferrari},
howpublished = {HAL},
year = {2026},
month = oct,
eprint = {2606.24638},
archivePrefix = {arXiv},
url = {https://cnrs.hal.science/hal-05829996v1},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - Weak Lensing with the SKAO: Radio Shear Measurement
AU - Priyamvad Tripathi
AU - Marzia Rivi
AU - Simon Prunet
AU - Ian Harrison
AU - André Ferrari
PY - 2026
DA - 2026/10/21
PB - HAL
UR - https://cnrs.hal.science/hal-05829996v1
AB - Cosmic shear measurements have traditionally been dominated by optical surveys, which offer higher resolution, better sensitivity, higher galaxy number density, and wider area coverage than their radio counterparts. With the advent of upcoming radio surveys, particularly those planned with the SKA-Mid, we would reach the high sensitivity and resolution required for weak lensing studies, allowing radio observations to begin competing with optical surveys in this domain. However, radio observations are affected by fundamentally different instrumental and astrophysical systematics, meaning that shape and shear measurement techniques developed for optical surveys cannot be straightforwardly applied. Fully realizing the weak lensing potential of these next generation radio surveys therefore requires the development and validation of methods tailored specifically for radio datasets. In this chapter, we present a simulation pipeline to generate realistic observations of isolated radio galaxies based on the SKA-Mid AA4 array configuration. We apply three recent radio shape measurement techniques: SuperCALS, RadioLensfit, and DeepShape, to assess their performance on these simulations. Our results show that RadioLensfit and DeepShape yield the most accurate shape estimates, although RadioLensfit is significantly more computationally intensive. We also perform shear recovery on simulated data using RadioLensfit and DeepShape, finding multiplicative and additive shear biases of order a few 10^−2 and 10^−4, respectively. Finally, we highlight the challenge of source separation, which will play a critical role in the success of future radio weak lensing analyses.
ER -