The observation of atmospheric far-infrared radiation is of significance for a deeper understanding of radiation exchange and balance in the earth-atmosphere system, especially in polar regions. More importantly, compared with other bands, far-infrared bands have inimitable advantages in remote sensing of ice clouds, upper layer water vapor in the troposphere, and atmosphere ingredients.

On the one hand, far-infrared radiation plays a crucial role in regulating climate and energy balance. Far-infrared radiation accounts for about 40% to 65% of the Earth's energy emitted to space and thus makes great contributions to the Earth's OLR (outgoing longwave radiation) and atmospheric cooling. However, there is still significant uncertainty in addressing the key issues related to heat flux regulation factors in cold and dry polar conditions due to the limited observations of far-infrared radiation, which has a negative influence on the accuracy of climate models. On the other hand, in atmospheric remote sensing, far-infrared spectra are highly sensitive to low-concentration water vapor in low temperature conditions, making it important for remote sensing of water vapor in polar regions, and in the upper troposphere and lower stratosphere. Additionally, the complex refractive indices of water and ice exhibit different spectral characteristics in mid-infrared and far-infrared bands, further enhancing the ability for cloud detection and phase recognition. Meanwhile, far-infrared hyperspectral radiation is considered to have the potential to improve the retrieval accuracy of microphysical and optical properties of thin ice clouds.

However, currently direct measurements of far-infrared radiation at hyperspectral resolution are still relatively limited due to technical issues related to precise spectroscopic and highly sensitive measurements. The most recent measurement of spaceborne far-infrared hyperspectral spectra can be traced back to the 1970s when the National Aeronautics and Space Administration of the United States (NASA) launched the Nimbus-III and Nimbus-IV using the IRIS (infrared interferometer sound) infrared interferometer, which measured far-infrared to mid-infrared radiation with a relatively rough spectral resolution (2.8 cm^-1) and a spatial resolution ranging from 400 cm^-1 to 1600 cm^-1. However, this is still the only satellite borne far-infrared radiation spectral observation data that can be obtained on a global scale. The main technical difficulties for spaceborne far-infrared radiation measurements lie in high-sensitivity detectors and hyperspectral optical systems (such as beam splitters). Due to the low photon energy in the infrared band, traditional infrared hyperspectral interferometers often require cooling by liquid helium (or liquid nitrogen) to improve measurement accuracy and signal-to-noise ratio, and this cannot be extended to satellite applications. Additionally, the moving mirror system of the Fourier spectrometer must also consider tilt and other errors when carried in space. These factors have become the main constraints on the development of high-precision and hyperspectral measurements of atmospheric far-infrared radiation for spaceborne payloads.

In recent years, with the development of high-sensitive uncooled detectors and beam splitters, a few comprehensive observation experiments of atmospheric far-infrared radiation at hyperspectral resolution have been conducted based on ground-based and airborne prototypes. Institutions such as the European Space Agency (ESA) and the NASA have also proposed a series of missions to observe far-infrared radiation by satellite instruments. Retrievals of ice cloud characteristics using hyperspectral far-infrared radiation have become an important frontier field and research hotspot. Thus, it is important and necessary to summarize the existing research to guide the future development of this field more rationally.

The main theoretical basis of far-infrared hyperspectral remote sensing is reviewed and summarized. We also introduce the advantages of far-infrared hyperspectral remote sensing of ice clouds from atmospheric absorption and ice crystal particle scattering sensitivities. Afterward, the development of far-infrared hyperspectral instruments for atmospheric remote sensing is sorted and summarized, with a focus on the technical parameters and key technical issues of the relevant instruments. From the perspective of technological breakthroughs in far-infrared radiation measurements, the key technologies associated with detectors, spectrometers, and beam splitters currently adopted have been classified and introduced (Tables 1-3). From the perspective of the platforms, the corresponding instruments and observation experiments of ground-based and airborne, and the main experimental results are introduced. Then, the main spaceborne missions to measure atmospheric far-infrared at hyperspectral resolution are summarized, including FORUM (ESA) and PREFIRE (NASA). Subsequently, the advantages and research progress of far-infrared hyperspectral technology for remote sensing of ice clouds are discussed. Since far-infrared spectra can provide complementary information on remote sensing of ice clouds, we compare studies about synergistic retrievals of ice cloud parameters and phase recognition by far-infrared and mid-infrared spectrum. In the end, the problems and the ongoing research trends in this field are discussed, including possible technological breakthroughs in the future and possible innovations in the future. The potential applications of far-infrared hyperspectral technology in ice cloud remote sensing in the future are also pointed out.

Far-infrared radiation measurements with hyperspectral resolution and highly sensitive measurements are gradually becoming a popular tool for atmospheric remote sensing. In summary, conducting global ice cloud remote sensing by hyperspectral far-infrared in the future still calls for in-depth and detailed explorations to promote the development of instrument technology, and also calls for a large number of observational experiments to develop accurate forward and retrieval algorithms.