The overall major challenge facing the development of satellite remote sensing is to solve new problems with affordable innovative technologies and measurement concepts, and to address old problems accumulated in the past half a century of satellite remote sensing experiments. Specifically, the following key aspects need to be addressed and supplemented:
One of the main advantages of satellite remote sensing is the ability to quickly observe large regions on Earth. At the same time, the limitations of the coverage range of available satellite data are also obvious. For example, polar orbiting imagers in low Earth orbit (LEO) usually achieve global coverage within at least one day (but mostly two days or more), so many natural phenomena with high time and spatial variability have not been fully captured.
In this regard, the geostationary orbit (GEO) for remote sensing, which provides frequent daily observations of the same celestial body, solves this limitation. However, there is a trade-off between the spatial coverage and resolution of satellite images (usually, higher coverage leads to lower spatial resolution). Achieving extensive observation of both time and space coverage and high spatial resolution is valuable for many industry applications, but also challenging. Therefore, the design of satellite observations may require new innovations, auxiliary data, and complementary observations to address specific objects and related issues.
Although the observation capacity of satellites has been proven, the information content provided by current satellite data is still limited for many industrial applications. Therefore, deploying satellite sensors with enhanced capabilities is feasible and urgently needs to be put on the agenda.
For example, it has been clearly realized that multi-angle polarimeters (MAPs) provide the most accurate data suitable for describing the detailed vertical characteristics of atmospheric aerosols and clouds, so polarized data in aerosol and cloud characterization is expected to significantly increase in the coming decade.
The quality of remote sensing inversion algorithms is another key aspect that affects the final quality of data products. Because once the instrument is put into use, the quality of the obtained observation data cannot be fundamentally improved, while the inversion algorithm can be continuously improved, which may lead to significant differences in the final satellite remote sensing data products. This is not only due to data from different instruments, but also due to improvements in inversion concepts.
In this regard, there has been significant progress in the new generation of remote sensing inversion algorithms in the past decade. For example, new algorithms often rely on fast and accurate atmospheric modeling (rather than using pre-calculated look-up tables or LUTs) and can invert a large number of parameters. In addition, synchronous inversions of aerosol properties and land surface and/or cloud properties have been achieved. Finally, in the CO2M EU / Copernicus framework mentioned above, joint inversion of CO2 and aerosol properties is a promising method to reduce the impact of aerosol pollution on derived CO2 products.
In addition, the idea of developing instrument-independent algorithms that can be applied to different observations or their combinations is becoming increasingly popular. Generalized Retrieval of Aerosol and Surface Properties (GRASP) is an example, which can be used for various passive and active measurements based on satellites and ground-based instruments, and has also been successfully applied to the synchronous and coordinated inversion of lidar profiles and columnar radiation observations.