The all-sky assimilation of AMSU-A radiances has been investigated and compared to the clear-sky approach currently used operationally. This work is motivated by the aim to assimilate all microwave instruments through the all-sky system in the future. With this in mind, a detailed investigation of differences in the assimilation configurations between the separate clear-sky and all-sky assimilation systems has been performed. The majority of these differences have been addressed with the all-sky configuration now sharing many common settings with the clear-sky configuration. This is motivated by the fact that the AMSU-A clear-sky configuration has been fine-tuned and optimised over many years so improving upon it with different settings would be difficult.
The experimentation shows that the all-sky assimilation of AMSU-A now replicates the large forecast impact of the clear-sky assimilation in the extra-tropics, while allowing up to 20% more observations to be assimilated for the lowest AMSU-A sounding channels. There are small but significant improvements to short-range forecasts of temperature, humidity and wind in the extra-tropics, while medium-range forecasts are generally neutral. However, short-range temperature, humidity and wind forecasts in the tropics are slightly degraded.
A number of possible future enhancements to the AMSU-A all-sky configuration have been identifiedwhich will hopefully lead to improved results and, along with the progress made so far, should enable the all-sky assimilation of AMSU-A and other microwave temperature sounding instruments to become operational in the future.
The all-sky assimilation of AMSU-A radiances has been investigated and compared to the clear-sky approach currently used operationally. This work is motivated by the aim to assimilate all microwave instruments through the all-sky system in the future. With this in mind, a detailed investigation of differences in the assimilation configurations between the separate clear-sky and all-sky assimilation systems has been performed. The majority of these differences have been addressed with the all-sky configuration now sharing many common settings with the clear-sky configuration. This is motivated by the fact that the AMSU-A clear-sky configuration has been fine-tuned and optimised over many years so improving upon it with different settings would be difficult.
The experimentation shows that the all-sky assimilation of AMSU-A now replicates the large forecast impact of the clear-sky assimilation in the extra-tropics, while allowing up to 20% more observations to be assimilated for the lowest AMSU-A sounding channels. There are small but significant improvements to short-range forecasts of temperature, humidity and wind in the extra-tropics, while medium-range forecasts are generally neutral. However, short-range temperature, humidity and wind forecasts in the tropics are slightly degraded.
A number of possible future enhancements to the AMSU-A all-sky configuration have been identifiedwhich will hopefully lead to improved results and, along with the progress made so far, should enable the all-sky assimilation of AMSU-A and other microwave temperature sounding instruments to become operational in the future.