Polysaccharides─Important Constituents of Ice-Nucleating Particles of Marine Origin

Abstract

Remote marine regions are characterized by a high degree of cloud cover that greatly impacts Earth's radiative budget. It is highly relevant for climate projections to represent the ice formation in these clouds. Therefore, it is crucial to understand the sources of ice-nucleating particles (INPs) that enable primary ice formation. Here, we report polysaccharides produced by four different aquatic eukaryotic microorganisms (Thraustochytrium striatum, Tausonia pullulans, Naganishia diffluens, Penicillium chrysogenum) as responsible ice-nucleating macromolecules (INMs) in these samples originating from the marine biosphere. By deriving a classical nucleation theory-based parametrization of these polysaccharidic INMs and applying it to global model simulations, a comparison to currently available marine atmospheric INP observations demonstrates a 44% contribution of polysaccharides to the total INPs of marine origin within -15 to -20 °C. The results highlight the relevance of biological INMs as part of the INP population in remote marine regions.

Keywords: ice-nucleating macromolecules; ice-nucleating particles; polysaccharides; remote marine regions.

Figure 1

Ice nucleation site density per mass n_m is given as a function of temperature for known mineral dust, biogenic INMs, and in this study examined marine microbial INMs. ‡This data was converted from n_s or n_v to n_m. Details of the conversion are given in the SI. * The desert dust compilation includes Asian dust, Canary Island dust, Israeli dust, and Saharan dust. Refs^,−,,,,,−

Figure 2

Physical properties of analyzed eukaryotic microorganisms indicate a non proteinaceous origin of INMs mainly attached to the microbial cells. The INM number concentration per sample volume is given for unmodified eukaryotic microorganisms together with CNT-based fits (A) and after physical treatments: heating at 95 °C for 1 h (B), filtration (<0.2 μm; C) and the combination of both (D). Heating does not change INM number concentrations, whereas filtration significantly reduces INM number concentration, and subsequent heating does not alter the INM concentration. This points toward the existence of heat-stable most likely non proteinaceous INMs freely suspended and to a larger extent connected to microbial cells.

Figure 3

High agreement of the ice nucleation activity of tested aquatic eukaryotic microorganisms with that of marine polysaccharides. The temperature-dependent ice nucleation site densities per carbohydrate carbon mass n_m, C-TCCHO is presented with CNT-based parametrizations (parameters are presented in Table S4 in the SI). The derived CNT-parametrization of T. striatum is used for HSZ25.

Figure 4

Modeled against observed INP concentration between −15 and −20 °C. Measurement data from 14 different campaigns (n = 5364; see Tab S5 and Figure S6 for references and regional coverage) in predominantly marine air masses was used. These were compared to annual mean INP concentrations derived from modeled mineral dust and sea salt concentrations simulated with a global model. For mineral dust INPs, the parametrization by Niedermeier et al., and for marine polysaccharide INPs, the parametrization derived in this work was applied. Gray dots show the comparison between modeled mineral dust INPs and observation, whereas colored dots (color-code by observation temperature) present the sum of modeled mineral dust + marine polysaccharide INPs. For orientation, the figure shows the 1:1 line (solid) and the 1:10 and 10:1 lines (dashed).

Figure 5

Percentage of modeled polysaccharide-based marine INPs in the sum of modeled INPs (mineral dust + marine polysaccharides) in the lowermost model layer for different temperatures. Annual mean INP concentrations were derived from mineral dust concentrations and from polysaccharides estimated based on sea salt concentrations simulated by a global model.