Effects of environmental temperature and humidity in future climates on ice nucleation and resulting precipitation in idealized supercell simulations

Parameterizations for heterogeneous ice nucleation based solely on temperature or ice supersaturation often under- or over-estimate observed ice crystal number concentrations over the range of the parameter space. Such errors may propagate into errors in model predictions of precipitation and, by extension, the life cycle of a convective storm. These errors also confound efforts to predict how the precipitation efficiency in convective storms differs in different environments, important for understanding regional climate change. As more observations of ice nuclei become available, in order to reflect the current state of knowledge, these parameterizations must be refined and tested. In the current study, new observations of ice nuclei from the INSPECT I and INSPECT II field campaigns are applied to several single- and double-moment microphysical parameterization schemes currently available in the Weather Research and Forecasting (WRF) model, including Purdue Lin, Morrison, Thompson, and the WRF single- and double-moment schemes. Data from the INSPECT campaigns include new estimates of ice nuclei representative of dust, soot, and organic particle concentrations in the free troposphere over North America, each chemical species having different nucleation efficiencies, not currently represented in WRF. New parameterizations representing these data are used to examine the sensitivity of ice crystal production to different temperature and relative humidity profiles in the storm environment, and the resulting effects on storm evolution, precipitation, and precipitation efficiency.