Abstract. This is the first paper to investigate the relationship between the shape of the scattering phase function of cirrus and the relative humidity with respect to ice (RHi, using space-based solar radiometric angle-dependent measurements. The relationship between RHi and the complexity of ice crystals has been previously studied using data from aircraft field campaigns and laboratory cloud chambers. However, to the best of our knowledge, there have been no studies to date that explore this relationship through the use of remotely sensed space-based angle-dependent solar radiometric measurements. In this paper, one case study of semi-transparent cirrus, which occurred on 25 January 2010 off the north-east coast of Scotland, is used to explore the possibility of such a relationship. Moreover, for the first time, RHi fields predicted by a high-resolution numerical weather prediction (NWP) model are combined with satellite retrievals of ice crystal complexity. The NWP model was initialised at midnight, on 25 January 2010, and the mid-latitude RHi field was extracted from the NWP model at 13:00 UTC. At about the same time, there was a PARASOL (Polarization and Anisotropy of Reflectance for Atmospheric science coupled with Observations from a Lidar) overpass, and the PARASOL swath covered the NWP-model-predicted RHi field. The cirrus case was located over Scotland and the North Sea. From the satellite channel based at 0.865 μm, the directionally averaged and directional spherical albedos were retrieved between the scattering angles of about 80 and 130°. An ensemble model of cirrus ice crystals is used to predict phase functions that vary between phase functions that exhibit optical features (referred to as pristine) and featureless phase functions. For each of the PARASOL pixels, the phase function that best minimised differences between the spherical albedos was selected. This paper reports, for this one case study, an association between the most featureless phase function model and the highest values of NWP-predicted RHi (i.e. when RHi > 1.0). For pixels associated with NWP-model-predicted RHi < 1, it was impossible to generally discriminate between phase function models at the 5% significance level. It is also shown that the NWP model prediction of the vertical profile of RHi is in good agreement with dropsonde, in situ measurements and independent aircraft-based physical retrievals of RHi. Furthermore, the NWP model prediction of the cirrus cloud-top height and its vertical extent is also found to be in good agreement with aircraft-based lidar measurements.