Reliability of in-band and broadband spectral index measurement: systematic study of the effect of signal to noise for uGMRT data

Md Rashid,Nirupam Roy,J. D. Pandian,Prasun Dutta,R. Dokara,S. Vig,K. M. Menten

2024-05-29

Abstract:Low radio frequency spectral index measurements are a powerful tool to distinguish between different emission mechanisms and, in turn, to understand the nature of the sources. Besides the standard method of estimating the ``broadband" spectral index of sources from observations in two different frequency ``bands", if the observations were made with large instantaneous bandwidth, the ``in-band" spectral index can be determined, either using images of emission at multiple frequency ranges within a band or using the novel Multi Term-Multi Frequency Synthesis (MT-MFS) imaging algorithm. Here, using simulated upgraded Giant Metrewave Radio Telescope (uGMRT) data, we have systematically studied the reliability of various methods of spectral index estimation for sources with a wide range of signal-to-noise ratio (SNR). It is found that, for synthetic uGMRT point source data, the MT-MFS imaging algorithm produces in-band spectral indices for SNR~$\lesssim100$ that have errors $\gtrsim 0.2$, making them unreliable. However, at a similar SNR, the sub-band splitting method produces errors $\lesssim 0.2$, which are more accurate and unbiased in-band spectral indices. The broadband spectral indices produce errors $\lesssim 0.2$ even for SNR $\gtrsim 15$, and hence, they are most reliable if there are no higher-order variations in the spectral index. These results may be used to improve the uGMRT observation and data analysis strategies depending on the brightness of the target source.

Instrumentation and Methods for Astrophysics

What problem does this paper attempt to address?

### What problems does this paper attempt to solve?

This paper aims to systematically study and evaluate the reliability of different methods in estimating the spectral index of low - frequency radio waves, especially for uGMRT (Upgraded Giant Metre - wave Radio Telescope) data. Specifically, the paper focuses on the following issues:

1. **The influence of signal - to - noise ratio (SNR) on spectral index measurement**:
- The reliability of estimating the "in - band" and "broad - band" spectral indices using methods such as the multi - frequency synthesis (MT - MFS) algorithm and the sub - band division method under different SNR conditions was studied.
- It was found that for the synthesized uGMRT point - source data, when the SNR is less than 100, the in - band spectral index error generated by the MT - MFS algorithm is large (≥0.2), making it unreliable; while the error generated by the sub - band division method under a similar SNR is small (≤0.2), which is more accurate and unbiased.

2. **Comparison of the advantages and disadvantages of different methods**:
- The performance of the MT - MFS algorithm, the sub - band division method, and the traditional broadband method under different SNR conditions was compared.
- The results show that the broadband spectral index has a small error (≤0.2) when the SNR is greater than 15, so it is the most reliable in the absence of higher - order spectral index changes.

3. **Spectral index estimation in the case of multiple point sources**:
- A region containing multiple point sources was simulated to check the in - band spectral index recovery of MT - MFS imaging in the presence of multiple point sources.
- The results show that as the SNR increases, the distribution range of the spectral index estimates gradually decreases, which is consistent with the single - point - source simulation results.

4. **Improvement strategies in practical applications**:
- Based on the research results, methods for improving uGMRT observation and data analysis strategies were proposed to improve the accuracy of spectral index estimation, especially when the target source has a low brightness.

### Formula summary

- **Definition of spectral index**:
\[
I_\nu \propto \nu^\alpha
\]
where \(I_\nu\) is the specific intensity, \(\nu\) is the frequency, and \(\alpha\) is the spectral index.

- **Broad - band spectral index formula**:
\[
\alpha = \frac{\log(F_1)-\log(F_2)}{\log(\nu_1)-\log(\nu_2)}
\]
where \(F_1\) and \(F_2\) are the flux densities at the reference frequencies \(\nu_1\) and \(\nu_2\), respectively.

- **Spectral index error formula**:
\[
\sigma_\alpha = \frac{1}{|\log(\nu_1/\nu_2)|}\sqrt{\left(\frac{\sigma_{F_1}}{F_1}\right)^2+\left(\frac{\sigma_{F_2}}{F_2}\right)^2}
\]
where \(\sigma_{F_1}\) and \(\sigma_{F_2}\) are the rms uncertainties of the two flux density maps, respectively.

Through these studies, the authors hope to provide more reliable spectral index estimation methods for future low - frequency radio astronomical observations and provide a solid foundation for interpreting the observation results.