Probing atmospheric ice nucleation using field-collected and laboratory-generated aerosol particles and global model evaluation

Poster PDF

Description

Atmospheric ice nucleation from a diverse aerosol population is investigated making use of field-collected and laboratory-generated particles and application of a global model. While experiments typically examine the ice nucleation efficiency of individual components, the focus here is to study ice nucleation when competing ice nucleating particles (INPs) are present, which closer reflects actual ambient conditions. This is achieved by using particles sampled at the SGP site and in the Amazonian region that include soil-derived and mineral dust particles, organic aerosol, and bioaerosol. The particle composition and mixing state of these field-collected particles is evaluated by scanning transmission X-ray microscopy with near-edge X-ray absorption spectroscopy (STXM/NEXAFS). The particles’ propensity to act as INP is investigated as a function of temperature as low as 210 K and RH as high as water saturation. In cases, we were able to identify and micro-spectroscopically characterize the INP. INPs of the SGP site are dominated by inorganic material, whereas INPs from the Amazonian region can be mostly organic in nature. Furthermore, laboratory immersion freezing experiments are employed to investigate the freezing of droplets that contain two different mineral dust species such as illite and kaolinite acting competitively as INPs. This experimental work is complemented by a modeling study that aims to analyze how the INP concentration depends on the emitted size distribution and mixing state of mineral dust particles. The model uses a significantly improved module of dust emission. The focus here is on K-feldspar since it has been demonstrated from immersion freezing experiments that this mineral is the most proficient INP at temperatures higher than 248 K, and it allows a comparison to a previous global model study that applied a simpler dust emission description. The current analysis will work toward an answer to the question of whether re-aggregation and partial fragmentation need to be considered to calculate INP concentration from K-feldspar, or whether neglecting those is a sufficient approach because errors from other sources of uncertainty can have a much larger effect on the calculated INP numbers than errors from differences in the size distribution and mixing state between the two methods. These efforts will improve our predictive understanding of atmospheric ice nucleation.