Ice production in northern hemisphere cold air-outbreak clouds: two contrasting aircraft campaigns

Cold-air outbreaks (CAOs) are common high-impact weather events that produce extensive boundary layer clouds that have a substantial influence on our planet’s climate. These clouds are often supercooled and therefore their properties are susceptible to the formation of ice.  The amount of ice in these clouds has been identified as being particularly important for defining the magnitude of the cloud-climate feedback and climate sensitivity. To address ice production in northern hemisphere CAOs we conducted two contrasting aircraft campaigns in 2022.  One campaign (ACAO, 11 flights) was in March in the Norwegian and Barents Sea where cold air flowed from the ice-covered Arctic Ocean. The other (M-Phase, 12 flights) was in October-November and focused on the Labrador Sea with air coming from the Arctic Archipelago. In both campaigns, we used similar instruments on the FAAM BAe-146 research aircraft designed to probe the aerosol properties, cloud microphysics and atmospheric thermodynamics of the CAO events.  Flight sorties were designed to study aerosol-cloud interactions as the CAO developed through the stratus and into the cumulus regime. We found that INP concentrations in these Northern Hemisphere CAOs were orders of magnitude greater than CAO events over the Southern Ocean.  The springtime ACAO cases had systematically greater INP (and aerosol) concentrations than the autumnal Labrador Sea M-Phase cases. The presence of substantial amounts of mineral dust in the springtime Arctic, despite all local sources being covered in ice and snow, implies a reservoir of old INPs and aerosol in the springtime Arctic that originated from the low latitudes. This is supported by our global aerosol model. Primary ice production by INPs is shown to define the ice concentrations in the stratus regime in many cases, but in the cumulus regime there are pockets of very high ice concentrations that are indicative of secondary ice production. Our modelling work has demonstrated that INPs are key to defining the stratus to cumulus transition and the cases are providing an excellent test for the high-resolution regional modelling with the Met Office Unified Model.  We are also using ACAO cases to study how INPs interact with clouds in CAOs, where warm temperature INPs are preferentially lost through nucleation scavenging. Furthermore, we envisage that the data from these campaigns will provide a valuable resource for model development, hypothesis testing and contrasting with other CAO campaigns in other places and times.  Given the stark contrast of primary ice production in CAO clouds in different locations and times around the globe, we conclude that the primary production of ice in model CAO clouds should be linked to the aerosol properties and knowledge of the local INP population to reduce uncertainty in cloud feedback and climate sensitivity.