Cirrus clouds, ice supersaturation, and their dynamical background

Abstract

Although this conference is on cirrus clouds, we may not forget, that the condition sine-qua-non for cirrus is ice supersaturation. This lecture is therefore on ice supersaturation. Perhaps not surprisingly, ice supersaturation in the upper troposphere is quite difficult to forecast. As it is an extremal state in the humidity field, it reacts non-linearly to changes in ambient conditions. Water vapour takes place in many different chemical and physical processes and is thus highly variable with strong gradients. Reliable measurements of humidity in the tropopause region are largely lacking, so that there is not much for weather models to assimilate. In the lecture, we will talk about a related quantity termed upper-tropospheric humidity, a satellite-derived radiation quantity that is intended to describe the humidity field in the UT. We will introduce the contribution and weighting functions and show some applications. Second we will show how ice supersaturated regions are connected to the dynamic quantities in the synoptic fields and we show how one can try to exploit these dynamic quantities for an improved prediction of ice supersaturation.