Anomalous ice crystal production in the evaporation zones of supercooled clouds

It is widely acknowledged that currently identified mechanisms of ice formation and multiplication in clouds with tops warmer than –35°C have often proven grossly insufficient when compared with field measurements of high ice crystal number concentration. In the absence of any obvious solutions, there has remained the hope that such discrepancies would be resolved simply by improved measurements of cloud particle size distributions as a function of phase, on the one hand, and ice nuclei activity as a function of nucleation mode and environmental conditions, on the other. However, here we argue that several recent field campaigns have only deepened the mystery. Namely, there is increasing confidence that state-of-the-art measurements in a variety of cloud types with improved cloud particle imaging and spectrometer probes have revealed large numbers of small ice crystals (up to at least 1000 per liter) where simultaneous measurements of ice nuclei are routinely at least two orders of magnitude too low and ice multiplication processes appear far too weak to explain the difference. In a particularly clear instance, Cotton and Field (2002) recently demonstrated that large numbers of ice crystals first appeared in the evaporation zones of supercooled wave clouds during the 1999 Interactions of Aerosol and Cold Clouds (INTACC) campaign. Here we present additional evidence for evaporation-related ice nucleation, based on comparison of observations from two recent field campaigns with large-eddy simulations using size-resolved multi-phase microphysics. We focus first on the anomalously high ice concentrations that were recently observed at the base of supercooled arctic stratocumulus during the 2004 Multi-Phase Arctic Cloud Experiment (M-PACE). We then focus on the evidence for anomalous ice formation in the updraft entrainment zones of continental cumulonimbus clouds during the 2002 Cirrus Regional Study of Tropical Anvils and Cirrus Layers—Florida Area Cloud Experiment (CRYSTAL-FACE). We explore explanations for evaporation-related ice nucleation, including proposed physical processes, as well as possible errors and omissions of field data sets, models, and laboratory measurements.