Data Analytics and Machine Learning Innovation for Climate and Earth Surface Processes

Explore the intersection of data analytics, machine learning, and climate science in this 2-hour 27-minute conference session from AGU. Delve into the challenges and opportunities presented by the increasing volume of Earth observations and climate model outputs. Discover how tools from mathematics, statistics, and computer science are being adapted to advance physical understanding, improve Earth system modeling, and increase predictive ability for weather, climate, and Earth surface processes. Learn about identifying causal sources of predictability, quantifying climate variability, improving micro-scale parameterizations in climate models, and deciphering landscape responses to change. Gain insights from discussions between ocean and atmospheric scientists, hydrologists, geomorphologists, and data scientists on leveraging big data and machine learning for climate and Earth system modeling advancements. Topics covered include language translation, dynamic memory, model data machine learning, Google Earth Engine applications, pattern learning and matching, downscaling digital elevation models, and climate model analysis.

Intro
Welcome
Introduction
Language Translation
Dynamic Memory
Model Data Machine Learning
Matt Hunter Introduction
Google Earth Engine
Machine Learning
Training in Earth Engine
Lessons Learned
Example Workflows
Digital Elevation Model
Optical Spectrum
Typical Problems
Pattern Learning
Pattern Matching
Downscaling Digital Elevation Models
Dancing of Climate Models
Climate Model
Results