Describing the Uncertainties in VLBI Images

TL;DR

This paper introduces a new method for accurately modeling pixel-based uncertainties in VLBI intensity images, improving the reliability of uncertainty estimates in astrophysical imaging.

Contribution

The paper presents a novel approach to quantify uncertainties in VLBI images, validated against Monte Carlo simulations, with ongoing refinements for compact sources.

Findings

Method accurately reproduces uncertainty patterns for non-compact sources.

Comparison with Monte Carlo simulations confirms the method's effectiveness.

Refinements are underway to improve accuracy for compact sources.

Abstract

We are investigating a new approach to modelling uncertainties in individual pixels of Very Long Baseline Interferometry (VLBI) intensity (Stokes I, Q, U) images. Comparison of distributions of our calculated uncertainties for model sources with the results of Monte Carlo simulations shows that our method correctly reproduces the overall level and pattern of uncertainties in intensity images for model sources that are not too compact. Refinement of the approach to better reproduce uncertainties in compact sources is being studied. This new approach should ultimately provide a means to correctly estimate pixel-based uncertainties in the Stokes parameters I, Q and U, including the effect of correlations between values in different pixels.