Bibliographie complète

Abstract Several cloud retrieval algorithms based on satellite observations in the infrared have been developed in the last decades. However, these observations only cover the midinfrared (MIR, λ < 15 μm) part of the spectrum, and none are available in the far‐infrared (FIR, λ ≥ 15 μm). Using the optimal estimation method, we show that adding a few FIR channels to existing spaceborne radiometers would significantly improve their ability to retrieve ice cloud radiative properties. For clouds encountered in the polar regions and the upper troposphere, where the atmosphere is sufficiently transparent in the FIR, using FIR channels would reduce by more than 50% the uncertainties on retrieved values of optical thickness, effective particle diameter, and cloud top altitude. Notably, this would extend the range of applicability of current retrieval methods to the polar regions and to clouds with large optical thickness, where MIR algorithms perform poorly. The high performance of solar reflection‐based algorithms would thus be reached in nighttime conditions. Since the sensitivity of ice cloud thermal emission to effective particle diameter is approximately 5 times larger in the FIR than in the MIR, using FIR observations is a promising venue for studying ice cloud microphysics and precipitation processes. This is highly relevant for cirrus clouds and convective towers. This is also essential to study precipitation in the driest regions of the atmosphere, where strong feedbacks are at play between clouds and water vapor. The deployment in the near future of a FIR spaceborne radiometer is technologically feasible and should be strongly supported. , Plain Language Summary The size of ice cloud particles can be estimated from space by measuring the infrared emission of ice clouds. However, this method no longer works when clouds are too thick or when particles are too large, although such clouds are encountered in many regions of the Earth and are critical for the Earth's climate. Using new sensors that measure cloud emission at longer wavelengths, in the so‐called far‐infrared region, would extend the potential of satellite observations to thick clouds, to clouds made of larger particles, and to clouds found in the polar regions which are very poorly known. These new sensors have become available in the last years, but none is deployed in space so far. We show that a satellite equipped with a few channels in the far‐infrared would greatly increase our capacity to observe ice clouds. This is promising for our understanding of cloud microphysics and precipitation in general, and for studying the water cycle in the polar regions in particular. For these reasons, it is highly recommended to deploy a far‐infrared radiometer in space. , Key Points Ice cloud remote sensing would be greatly improved by adding far‐infrared channels to existing midinfrared spaceborne radiometers Far‐infrared radiometry is well suited to study ice clouds in the polar regions and upper troposphere Using far‐infrared radiometry extends current infrared capabilities to large optical thickness and large cloud particles