Application of infrared remote sensing to constrain in-situ estimates of ice crystal particle size during SPartICus

Abstract. In a prior paper (Cooper and Garrett, 2010), an infrared remote sensing technique was developed that quantifies the effective radius re of ice crystals in cirrus clouds. By accounting for a broad range of expected inversion uncertainties, this retrieval scheme isolates those radiometric signatures that can only occur if the cirrus has nominally "small" values of re below 20 μm. The method is applicable only for specific cloud and atmospheric conditions. However, it can be particularly useful in constraining in-situ estimates of cirrus cloud re obtained from aircraft. Recent studies suggest that airborne measurements may be compromised by the shattering of ice crystals on airborne instrument inlets, so robust, independent confirmation of these measurements is needed. Here, we expand the Cooper and Garrett (2010) retrieval scheme to identify ice clouds that are likely to have "large" values of re greater than 20 μm. Using MODIS observations, we then compare assessments of cirrus cloud re with in-situ measurements obtained during three test cases from the 2010 SpartICus campaign. In general, there is good agreement between retrievals and in-situ measurements for a "small" and "large" crystal case. For a more ambiguously "small" re case, the 2D-S cloud probe indicates values of re that are slightly larger than expected from infrared retrievals, possibly indicating a slight bias in the 2D-S results towards large particles. There is no evidence to support that an FSSP-100 with unmodified inlets produces measurements of re in cirrus that are strongly biased low, as has been claimed.