Potential Sources, Characteristics, and Controls over Emissions of Marine Ice Nucleating Particles

Marine ice nucleating particle (INP) emissions are an important source of uncertainty in modelling cloud dynamics and radiation balance over oceans. And yet we know little about which marine aerosols nucleate ice, their numbers or seasonality. INP concentrations from measured oceanic regions are typically low, but include some intriguing reports of elevated levels associated with phytoplankton blooms and above upwelling convergence zones. Mesocosm experiments conducted as part of the Center for Aerosol Impacts on Chemistry of the Environment confirmed previous observations that when phytoplankton blooms collapse the concentration of ice nucleating entities (INEs) in the seawater increases. These may be ejected within the sea spray aerosol (SSA). To test if phytoplankton type influences INE production, and to investigate specific sources of INEs, we separately added dead biomass of three phytoplankton species (a green alga, a diatom and a cyanobacterium) to a Marine Aerosol Reference Tank containing 7.5 L of filtered seawater. Throughout the ensuing decomposer successions, measurements of both the bulk water and SSA included: INP concentrations, measured using a Continuous Flow Diffusion Chamber and after collection on filters; cell counts of bacteria, viruses and grazers; Next-Generation sequencing of bacteria; catabolic enzyme activities; emission-excitation matrices; and aerosol size distributions. INEs in the bulk water and INPs in the SSA peaked 3-5 days after each detritus addition, coincident with a decline in heterotrophic bacteria, a peak in viruses and nanoflagellates, and a decrease in several enzyme activities. To directly probe INE composition, arrays of small droplets of fresh seawater were cooled, and frozen droplets mapped and then collected, to obtain sub-samples either enriched in INEs active at warm temperatures or devoid of them. nanoElectrospray-Orbitrap mass spectrometry and micro-Raman spectroscopy were then used to compare the composition of these samples to identify putative INE biomarkers. We also manipulated the sea surface microlayer (SML) to asses its control over emissions. For example, addition of SML increased INP abundance in the SSA by approximately 10% while addition of 1 mg/L protein rapidly reduced INP emission rates >100-fold.