Abstract. The halo ratio (HR) is a quantitative measure characterizing the occurrence of the 22∘ halo peak associated with cirrus. We propose to obtain it from an approximation to the scattering phase function (SPF) derived from all-sky imaging. Ground-based fisheye cameras are used to retrieve the SPF by implementing the necessary image transformations and corrections. These consist of geometric camera characterization by utilizing positions of known stars in a camera image, transforming the images from the zenith-centred to the light-source-centred system of coordinates and correcting for the air mass and for vignetting, the latter using independent measurements from a sun photometer. The SPF is then determined by averaging the image brightness over the azimuth angle and the HR by calculating the ratio of the SPF at two scattering angles in the vicinity of the 22∘ halo peak. In variance from previous suggestions we select these angles to be 20 and 23∘, on the basis of our observations. HR time series have been obtained under various cloud conditions, including halo cirrus, non-halo cirrus and scattered cumuli. While the HR measured in this way is found to be sensitive to the halo status of cirrus, showing values typically >1 under halo-producing clouds, similar HR values, mostly artefacts associated with bright cloud edges, can also be occasionally observed under scattered cumuli. Given that the HR is an ice cloud characteristic, a separate cirrus detection algorithm is necessary to screen out non-ice clouds before deriving reliable HR statistics. Here we propose utilizing sky brightness temperature from infrared radiometry: both its absolute value and the magnitude of fluctuations obtained through detrended fluctuation analysis. The brightness temperature data permit the detection of cirrus in most but not all instances.