Regional models improve fast atmospheric gas transport estimates worldwide
From Regional to Global: Transfer Learning for Atmospheric Transport Emulators
Machine Learning
Summary
Estimating greenhouse gas emissions often uses complex models that simulate how gases move through the atmosphere. These models can be very slow, especially for global data from satellites. The paper shows how a faster emulator, previously made for South America, can be adapted to work across different world regions by training on multiple areas and testing transfer between them. This helps make quick, accurate estimates of gas transport possible over the whole globe without running expensive simulations everywhere.
What this means in practice
- •For climate monitoring teams: Quickly produce gas transport footprints globally by using cross-region trained emulators for faster emissions estimation from satellite data.
- •For environmental consulting firms: Reduce time and cost in greenhouse gas tracking by implementing multi-region emulator models instead of running costly physical simulations everywhere.
Authors
Jeff Clark, Elena Fillola, Nawid Keshtmand, Raul Santos-Rodriguez, Matthew Rigby
Abstract
Greenhouse gas emissions estimates can be derived using inverse methods by combining atmospheric concentration observations with chemical transport models. The latter traditionally use physics-driven simulators such as Lagrangian Particle Dispersion Models (LPDMs), which are expensive to run and do not scale well to modern satellites' high resolution data. Previously we developed a performant atmospheric transport emulator that approximates LPDM outputs ("footprints") over South America ~1,000X faster than the UK Met Office's LPDM. Expanding towards global emulation is not straightforward, as atmospheric transport is regionally heterogeneous. This paper evaluates spatial transferability capabilities of models across four world regions: South America, East Asia, South Asia, North Africa using both region-specific and multi-region models, and leave-one-region-out experiments. Regional differences are characterised in the context of input variable and output footprint distributions. This work builds intuition in cross-region generalisation and transfer learning, aiding regional performance towards efficient global emissions estimates.