Climate Science Challenges and Opportunities for Machine Learning

Climate Science Challenges and Opportunities for Machine Learning

Scott Hosking

Abstract:

In this lecture we will focus on three areas: 1) monitoring environmental change; 2) modelling our climate; and 3) quantifying climate risks. The first section will highlight some of the key observational datasets available for assessing how our atmosphere and oceans have changed over the recent past (from 1900 to the present day), and some of the challenges surrounding the non-uniform distribution of in-situ measurements. The second section will briefly explain what a climate model simulator is, what they are good at doing and the challenges associated with comparing their output with real-world climate change. Then in the third section we will discuss how the climate research community increases spatial granularity of our climate models and zoom-in on specific regions of interest, such as densely populated regions or vulnerable environments. At the end we will go through a Google Colab notebook using some gridded climate model simulation output to setup a Multifidelity Climate Modelling data challenge!

repo with Google Colab Notebook: https://github.com/scotthosking/mf_modelling

Relevant links and publications: