How it works

Detection is a hypothesis test on the visibilities (Vio & Andreani 2021). For a trial position \(\boldsymbol{\ell}\), rotate the visibilities there and keep the real part, \(x_{jk} = \mathrm{Re}[V_{jk}\,e^{-2\pi i\,\mathbf{u}_{jk}\cdot\boldsymbol{\ell}}]\) for row \(j\) and channel \(k\), and test

\[H_0:\ \mathbf{x} = \mathbf{n}, \qquad H_1:\ \mathbf{x} = a\,\mathbf{s} + \mathbf{n}, \qquad s_{jk} = E(\mathbf{u}_{jk})\,g(\nu_k),\]

with \(\mathbf{n}\) Gaussian noise of covariance \(\mathbf{C} = \mathrm{diag}(w^{-1})\) from the weights, \(E\) the visibility amplitude of an elliptical Gaussian source, and \(g\) a Gaussian line. The matched filter

\[T(\mathbf{x}) = \frac{\mathbf{s}^T\mathbf{C}^{-1}\mathbf{x}}{\sqrt{\mathbf{s}^T\mathbf{C}^{-1}\mathbf{s}}}\]

is \(\mathcal{N}(0, 1)\) under \(H_0\) and has mean equal to the line’s S/N under \(H_1\). It does not depend on the template’s amplitude, so the response is in S/N units and the flux is not searched.

Spatial collapse. The template is separable, so each channel is first summed over rows, weighting every visibility by the trial source’s own amplitude:

\[y_k = \mathrm{Re}\sum_j E_{jk}\,X_{jk}\,e^{-2\pi i\,\mathbf{u}_{jk}\cdot\boldsymbol{\ell}}, \qquad W_k = \sum_j w_{jk}\,E_{jk}^2, \qquad X = wV.\]

For a point source \(y_k / W_k\) is the naturally weighted dirty image, computed exactly. On a regular position grid the phase factorises, so every channel is one complex matrix product (luv_finder.kernel).

Line and continuum. At each position the spectrum is fitted as a line template \(\mathbf{g}\) on a polynomial continuum \(\mathbf{P}\). The template is projected off the polynomials, \(\mathbf{h} = \mathbf{g} - \mathbf{P}(\mathbf{P}^T\mathbf{W}\mathbf{P})^{-1}\mathbf{P}^T\mathbf{W}\mathbf{g}\), and normalised by its exact variance under the correlation \(\rho\) between channels:

\[T = \frac{\mathbf{h}^T\mathbf{y}}{\sqrt{\sum_{kk'} h_k h_{k'}\,\rho_{|k-k'|}\sqrt{W_k W_{k'}}}}.\]

ALMA’s Hanning response gives \(\rho = (1, 2/3, 1/6)\); it is measured on the jackknife per spectral window.

Null hypothesis. The jackknife differences consecutive integrations of each baseline, which cancels the sky and keeps the noise. The same grid run on it gives the correlation between channels, the shape of a source’s response, and the false-alarm rate of the peaks: a blind search selects peaks among many correlated tests, whose \(P_\mathrm{FA}\) is not that of a single value. Candidates need a likelihood ratio \(N_\mathrm{data}(\ge\gamma)/N_\mathrm{jackknife}(\ge\gamma) \ge 3\) and a fidelity \(\ge 0.6\), with \(N_\mathrm{jackknife}\) a model of the jackknife’s peak counts, \(n_k (u/2.5)^m e^{-(u^2 - 2.5^2)/2}\) (Catalogue and line fit).

Mosaics. Each pointing is divided by its primary beam and the pointings are combined by inverse variance on one sky grid (Mosaics and the primary beam).