Testing source confusion and identification capability in Cherenkov Telescope Array data

TL;DR

This study assesses CTA's ability to resolve and identify overlapping gamma-ray sources, especially PWNe, through simulations based on existing data, highlighting its potential to disentangle source confusion in the Galactic Plane survey.

Contribution

The paper presents the first detailed simulation-based analysis of source confusion and identification capabilities of CTA for TeV PWNe, using real H.E.S.S. data as a basis.

Findings

CTA can resolve up to 60% of confused sources above 500 GeV.

Source confusion occurs in about 30% of simulated scenarios.

Over 95% accuracy in matching simulated sources to morphological templates.

Abstract

The Cherenkov Telescope Array will provide the deepest survey of the Galactic Plane performed at very-high-energy gamma-rays. Consequently, this survey will unavoidably face the challenge of source confusion, i.e., the non-unique attribution of signal to a source due to multiple overlapping sources. Among the known populations of Galactic gamma-ray sources and given their extension and number, pulsar wind nebulae (PWNe, and PWN TeV halos) will be the most affected. We aim to probe source confusion of TeV PWNe in forthcoming CTA data. For this purpose, we performed and analyzed simulations of artificially confused PWNe with CTA. As a basis for our simulations, we applied our study to TeV data collected from the H.E.S.S. Galactic Plane Survey for ten extended and two point-like firmly identified PWNe, probing various configurations of source confusion involving different projected separations, relative orientations, flux levels, and extensions among sources. Source confusion, defined here to appear when the sum of the Gaussian width of two sources is larger than the separation between their centroids, occurred in $\sim$30% of the simulations. For this sample and 0.5$\deg$ of average separation between sources, we found that CTA can likely resolve up to 60% of those confused sources above 500 GeV. Finally, we also considered simulations of isolated extended sources to see how well they could be matched to a library of morphological templates. The outcome of the simulations indicates a remarkable capability (more than 95% of the cases studied) to match a simulation with the correct input template in its proper orientation.