spectral curvature in the synchrotron component is almost impossible. We expect to see spectral steepening in the synchrotron emission due to energy losses in the cosmic-ray electron (CRE) population (e.g. Strong et al. 2011) or potentially even flattening due to reenergization of the CRE population from active star formation regions (Bennett et al. 2003). The ARCADE2 results suggest that there is evidence for synchrotron spectral curvature (Kogut et al. 2011); however, the ARCADE2 survey was limited to a relatively small portion of the sky.
There are also still many unknown quantities regarding dust emission at radio frequencies, commonly referred to as either anomalous microwave emission (AME) (Kogut et al. 1996; Leitch et al. 1997) or, when linked with polycyclic aromatic hydrocarbons (PAHs) or very small grains, spinning dust emission (Draine & Lazarian 1998b; Ali-Haı̈moud, Hirata & Dickinson 2009). The Planck COMMANDER analysis has provided the current best all-sky map of AME (Planck Collaboration XII 2014; Planck Collaboration IV 2020). However, it was found that the COMMANDER analysis derived AME amplitude can differ by 30–50 per cent when compared to other analyses due to correlations with free–free and synchrotron emission components even in mid-to-high latitude (b > 10°) regions (Planck Collaboration XXV 2016). Similarly, Cepeda-Arroita et al. (2021) found significant differences in the fitted AME amplitude when using both the C-BASS and 408-MHz data to constrain low frequencies rather than just the 408-MHz data alone in the λ-Orionis region.
The study of AME, free–free and synchrotron emission at mid-to-high latitudes can be greatly improved by using a cleaner estimate of the synchrotron emission at WMAP and Planck frequencies. CBASS offers just such as estimate as it is nearer in frequency, meaning that it will be less impacted by spectral curvature. Further advantages of the C-BASS map over the 408-MHz map are a well-understood beam response, allowing the map to be deconvolved to a known Gaussian beam pattern, and well-understood noise and systematics; all of which are not available for the 408-MHz map. At present, only the northern part of the C-BASS survey is complete, but this will be extended to the full-sky in the future.
In this paper, we use a preliminary version of the C-BASS intensity map of the northern sky to perform a pixel-space template fitting correlation analysis, a method that has a long heritage for studying Galactic foregrounds (e.g. de Oliveira-Costa et al. 1997; Banday et al. 2003; Bennett et al. 2003; Davies et al. 2006; Ghosh et al. 2012; Peel et al. 2012). We use the C-BASS map in combination with the 408MHz map to estimate the degree of spectral curvature across the sky in regions of approximately 100 deg2 . We also investigate if there is any change in the fitted dust coefficients at 22.8 GHz when using C-BASS in place of the 408-MHz map, and look for any evidence of a hard synchrotron component that may explain some or all of the observed AME at mid-to-high latitudes (Peel et al. 2012). Finally, we look to see if there is any preference for any particular dust tracer when measuring AME at mid-to-high latitudes.
The paper is organized as follows: In Section 2, we provide an overview of the C-BASS experiment, data-reduction methods, and map-making. In Section 3, we describe all the ancillary data sets used in this analysis. In Section 4, we outline the template fitting method as well as the masks and regions that we use to divide up the sky. In Section 5, we outline the results of the template fitting analysis. In Section 6, we fit model spectra of spinning dust and thermal dust to the dust template fitting coefficients. Finally, in Section 7, we discuss the results of the template fitting analysis in a wider context before giving concluding remarks and a summary in Section 8.
2 C-BASS NORTH
In this work, we will use a preliminary version of the C-BASS intensity data. The C-BASS map presented here is identical to that used in Dickinson et al. (2019) and Cepeda-Arroita et al. (2021) except for a global calibration correction resulting in a 2.1 per cent decrease in brightness; there is also a decrease in calibration uncertainty from 5 to 3 per cent. This map will be almost identical to the final C-BASS intensity map except for small changes in data selection and processing aimed at eliminating subtle artefacts in the polarization data, which have negligible effect on the intensity maps. In the following sections, we give an overview of the C-BASS data-reduction pipeline. More details will be given in the upcoming C-BASS survey papers (Taylor et al., in preparation and Pearson et al., in preparation).
2.1 Northern survey and instrument
The C-BASS project aims to map the entire sky in intensity and polarization at 4.76 GHz (Jones et al. 2018). The northern survey used a 6.1-m dish based at the Owens Valley Radio Observatory, with a nominal full-width half-maximum (FWHM) resolution of 0°.73. The antenna uses a Gregorian optical configuration that was designed to minimize sidelobe power by both underilluminating the primary reflector and surrounding it in a radio absorbing baffle. Furthermore, to ensure a circularly symmetric beam pattern, the support struts for the secondary reflector were removed and it was instead supported by a low-loss dielectric foam cone (Holler et al. 2013).
The C-BASS instrument is a dual circularly polarized correlation radiometer that can obtain instantaneous measurements of Stokes parameters I, Q, and U. The receiver’s nominal bandpass is 4.5– 5.5 GHz but notch filters were used to suppress local radio frequency interference (RFI), reducing the effective bandwidth to 0.5 GHz. Receiver stability is maintained by a continuous comparison between the sky and a resistive load that minimizes 1/f noise fluctuations. For details of the C-BASS receiver system, see King et al. (2014).
The C-BASS northern survey observations were taken between 2012 and 2015. The survey scanning strategy slewed the telescope at a set of fixed elevations, between 37°.2 and 77°.2, over full 360° sweeps in azimuth. The slewing rate was ≈4° s−1 but was varied slightly between observations to average out any possible scan-synchronous instrumental systematics from the final C-BASS map. This observing strategy maps all declinations above −15°, and covers approximately a 26 000 deg2 area of sky. A full description of the northern survey maps will be given in Taylor et al. (in preparation).
2.2 C-BASS data reduction
The C-BASS data are processed through a standard pipeline procedure, the details of which will be outlined in forthcoming papers (Pearson et al., preparation and Taylor et al., in preparation). However, we provide a brief summary here. The key tasks of the pipeline are to flag sources of RFI and Solar system objects, calibrate the data to brightness temperature units, and remove a number of systematic effects from the data. The two main systematics that are modelled and tracked are: a microphonics signal caused by the cryocooler system, which induces a 1.2-Hz signal into the timeordered data (TOD); and emission from the ground detected in the far sidelobes.
Relative calibration is performed by injecting a regular noise diode signal into the front-end of the receiver. The noise diode is found to be stable to the level of 1 per cent or better over many months,
MNRAS 513, 5900–5919 (2022)
Downloaded from https://academic.oup.com/mnras/article/513/4/5900/6581346 by guest on 30 April 2023
