From the single-scattering properties of ice crystals to climate prediction: A way forward

Abstract

Cirrus is composed of non-spherical ice crystals, and against the blue background of the sky, they appear as tenuous wispy clouds, usually located at altitudes greater than about 6km. Their spatial and temporal distribution about the Earth's atmosphere is significant. With such distributions, their contributions to the Earth's natural greenhouse effect and hydrological cycle are important. Therefore, it is important that climate models are able to predict the radiative effect of cirrus, as well as their contribution to the total amount of ice mass that occurs in the Earth's atmosphere. However, cirrus is composed of ice crystals that can take on a variety of geometrical shapes, from pristine habits such as hexagonal ice columns, hexagonal ice plates and bullet-rosettes, to highly randomized habits, which may have roughened surfaces and/or air cavities. These habits also aggregate together, to form chains of aggregates and compact aggregates. The sizes of these habits may also vary, from about less than 10μm, to several cm, with the smaller ice crystals usually existing toward cloud-top and the larger ice crystals existing toward the cloud-bottom. Due to this variability of geometrical complexity, size, and ice mass, predicting the magnitude of the cirrus greenhouse effect has proven problematic. To try to constrain these radiative and hydrological uncertainties, since about 2006 there is now available the A-train constellation of satellites, which attempt to quantify the radiative and hydrological contributions of cirrus to the Earth's atmosphere. The A-train obtains nearly simultaneous measurements of cirrus from across the electromagnetic spectrum. Such simultaneous measurements pose challenges for theoretical scattering models of cirrus, as these models must conserve ice mass and be physically consistent across the electromagnetic spectrum.In this review paper, the microphysical properties of cirrus are summarized. The current idealized habit mixture models that have been proposed to represent the observed variability in ice crystal shape, size and mass are discussed. The theoretical light scattering methods that are currently applied to the idealized habit mixture models to solve for their scattering and absorption properties are discussed. The physical inconsistency of the current approach to parameterize the bulk scattering and absorption properties of cirrus in climate models is highlighted. An alternative parameterization, which couples cloud physics more directly with radiation, is proposed. Such a coupling is required, if climate models are to be physically consistent and radiatively interactive.