Stochastic Parameterizations: Better Modelling of Temporal Correlations using ML

Add to your list(s) Download to your calendar using vCal

• Raghul Parthipan, University of Cambridge
• Tuesday 05 April 2022, 11:00-12:30
• https://zoom.us/j/6708259482?pwd=Qk03U3hxZWNJZUZpT2pVZnFtU2RRUT09.

If you have a question about this talk, please contact Herbie Bradley.

The modelling of small-scale processes is a major source of error in climate models, hindering the accuracy of low-cost models which must approximate such processes through parameterization. Using stochasticity and machine learning have led to better models but there is a lack of work on combining the benefits from both. We show that by using a physically-informed recurrent neural network within a probabilistic framework, our resulting model for the Lorenz 96 atmospheric simulation is competitive and often superior to both a bespoke baseline and an existing probabilistic machine-learning (GAN) one. This is due to a superior ability to model temporal correlations compared to standard first-order autoregressive schemes. The model also generalises to unseen regimes. We evaluate across a number of metrics from the literature, but also discuss how the probabilistic metric of likelihood may be a unifying choice for future probabilistic climate models.

This talk is part of the AI4ER Seminar Series series.

This talk is included in these lists:

• AI4ER Seminar Series
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge talks
• Chris Davis' list
• Interested Talks
• Trust & Technology Initiative - interesting events
• bld31
• https://zoom.us/j/6708259482?pwd=Qk03U3hxZWNJZUZpT2pVZnFtU2RRUT09
• ndk22's list
• ob366-ai4er
• rp587

Note that ex-directory lists are not shown.