Abstract. Sea spray aerosol contains ice nucleating particles (INPs), which affect the formation and properties of clouds. Here, we show that aerosols emitted from fast growing marine phytoplankton produce effective immersion INPs, which nucleate at temperatures significantly warmer than the atmospheric homogeneous freezing (−38.0 ∘C) of pure water. Aerosol sampled over phytoplankton cultures grown in a marine aerosol reference tank (MART) induced nucleation and freezing at temperatures as high as −15.0 ∘C during exponential phytoplankton growth. This was observed in monospecific cultures representative of two major groups of phytoplankton: a cyanobacterium (Synechococcus elongatus) and a diatom (Thalassiosira weissflogii). Ice nucleation occurred at colder temperatures (−28.5 ∘C and below) when the cultures were in the stationary or death phases of growth. Ice nucleation at warmer temperatures was associated with relatively high values of the maximum quantum yield of photosystem II (ΦPSII), an indicator of the physiological status of phytoplankton. High values of ΦPSII indicate the presence of cells with efficient photochemistry and greater potential for photosynthesis. In the North Atlantic Ocean, high net growth rates of natural phytoplankton assemblages were associated with marine aerosol that acted as effective immersion INPs at relatively warm temperatures. Data were collected over 4 days at a sampling station maintained in the same water mass as the water column stabilized after deep mixing by a storm. Phytoplankton biomass and net phytoplankton growth rate (0.56 day-1) were greatest over the 24 hours preceding the warmest mean ice nucleation temperature (−25.5 ∘C). Collectively, our laboratory and field observations indicate that phytoplankton physiological status is a useful predictor of effective INPs, and more reliable than biomass or taxonomic affiliation. Ocean regions associated with fast phytoplankton growth, such as the North Atlantic during the annual spring bloom, may be significant sources of atmospheric INPs.

No abstract