Abstract:
Over the past years, Earth system modelling has substantially changed with the advent of machine learning in this field. Those new models are either trained on existing physical models (mostly reanalyses) or satellite observations to produce direct observation predictions (DOP). However, both are limited by the skill of the model or the amount of satellite observations available to train a data-driven model. Neither are available for the land surfaces to the extent that physically meaningful predictions can be learned, be it due to simplifications of the physics within land surface models or the fact the satellites cannot observe the subsurface (with one exemption).
Physical model development for the operational forecasting system at ECMWF therefore continues. In this presentation, the recent developments towards a more explicit integration of hydrology into ecLand (including work on cold-region hydrology), new ways to spatially optimise model parameters, and the pathway to a data-driven model (aiLand) will be presented. The presentation will close with a suggested strategy to leverage the strength of both physics- and data-driven modelling to advance them jointly, rather than in competition.
What is the Joint Oslo Seminar (JOS):
- Atmospheric and climate sciences have a stronghold in Oslo among the four institutions University of Oslo, the Meteorological Institute, CICERO and NILU.
- This joint seminar invites renowned international experts to contribute to an informal series of lectures, meant to create interaction with the Oslo atmospheric and climate science community on recent highlights and analysis in the field.
- Normally seminars will be held on Thursdays (12:15 pm -1pm).