10:10 〜 10:30
[U02-04] Constraining climate sensitivity, cloud feedbacks using the satellite observations
★Invited Papers
Suki’s research work was not limited to numerical modelling. My work with Suki estimated radiative feedback in the annual variation from the satellite observations and climate models, which became a pioneering work on constraining climate feedback using observations.
The method of constraining feedback has evolved recently by combining constraints using multiple time scales of observations in the past, and also linking performance to climate sensitivity using machine-learning technology.
These ‘constraints’ have not been ‘popular’ with climate model development community. The reason is because they suggest ‘something is wrong’ but do not tell ‘what is wrong’, hence not so helpful as a clue to the key processes.
The recent constraint method was used in the IPCC AR6 and the estimated range of climate sensitivity is smaller than the inter-model spread. It is an interesting time for each modelling centre to decide the priority in the development of the model.
Cloud feedback has long been the largest uncertainty and the community has worked to understand the contributing cloud types and factors that lead to the model spread and recent studies have started to provide constraints on different cloud types.
The long-term satellite observations of clouds and radiative fluxes are the key to these achievements.
It has been twenty years since radiative fluxes started to be monitored from the satellites and the dataset is a treasure to the climate community.
However, maintaining the long-term observations is fragile, partly because of scientific challenges but fundamentally because of the current bureaucratic systems. In this talk, I take examples of the radiative flux observations from the satellite and the ground stations and describe the problem.
The method of constraining feedback has evolved recently by combining constraints using multiple time scales of observations in the past, and also linking performance to climate sensitivity using machine-learning technology.
These ‘constraints’ have not been ‘popular’ with climate model development community. The reason is because they suggest ‘something is wrong’ but do not tell ‘what is wrong’, hence not so helpful as a clue to the key processes.
The recent constraint method was used in the IPCC AR6 and the estimated range of climate sensitivity is smaller than the inter-model spread. It is an interesting time for each modelling centre to decide the priority in the development of the model.
Cloud feedback has long been the largest uncertainty and the community has worked to understand the contributing cloud types and factors that lead to the model spread and recent studies have started to provide constraints on different cloud types.
The long-term satellite observations of clouds and radiative fluxes are the key to these achievements.
It has been twenty years since radiative fluxes started to be monitored from the satellites and the dataset is a treasure to the climate community.
However, maintaining the long-term observations is fragile, partly because of scientific challenges but fundamentally because of the current bureaucratic systems. In this talk, I take examples of the radiative flux observations from the satellite and the ground stations and describe the problem.