Machine Learning in Environmental Science and Prediction: An Overview

Time: Thursday 22-Oct-2020 16:00 (This is a past event.)


Artifacts
slides: please to see content

Motivation / Abstract
This presentation is the debut of a new stream on "ML for Environmental Science", and outline the vision for this stream, as well as a brief overview of current applications of machine learning in climate and earth system science.

As an introduction to the topic, I will first highlight some work of researchers I would like to invite and that would be representative of the stream topics in the near future. Some of them will be related to improving existing physical models or gaining insight from observational and model data (i.e. science), while others will be concerned with direct prediction (forecasting). 

Then I will present my own take on where I think machine learning and AI are at in environmental science right now, and what the major challenges are. It will be the perspective of a scientist in the field, but I will focus on areas that I think are of interest to machine learning researchers. With this I also hope to provide a starting point for machine learning experts interested in applying their skills in environmental or climate science, and thus foster collaboration between the scientific and machine learning communities

About the Speaker:
I am an atmospheric scientist by training and my primary background is in climate modelling. I have long had an interest in the use of machine learning in atmospheric science and have been involved with the climate informatics workshop since 2015. For a few years I have also been working in hydrological modeling and since then have become very interested in ML in hydrology and earth system science.

Due to my background the stream will still have a heavy focus on ML in climate science and modeling, but in the immediate future I will also include hydrology, as well as applications in weather prediction. In the long-term I hope to expand the scope to include a wider range of environmental and earth system sciences, including, e.g. remote sensing and ecology.
Stream Categories:
 ML in Climate ScienceML in Physics