Prediction of Effective Angular Momentum (EAM) functions

Introduction to the methods

The specifications of the methods that are currently operational are mentioned in the following table.

Each component of EAM is analyzed separately. Predictions are made for mass terms (χ1,p, χ2,p, χ3,p) and motion terms (χ1,w, χ2,w, χ3,w). The EAM components that are predicted are as follows

EAM type Description Predicted terms
AAM Atmospheric Angular Momentum mass and motion
HAM Hydrological Angular Momentum mass and motion
OAM Oceanic Angular Momentum mass and motion
SLAM Sea-Level Angular Momentum mass


Acknowledgements
The data sources used in the current framework are taken GFZ German Research Centre for Geosciences, section Earth System Modelling (ESMGFZ). The original data can be accessed here.

.

Methods description

Method Input EAM Algorithm Forecasting horizon (days) Comments
fonode-GFZ-14d-AAM GFZ AAM mass terms (χ1,p, χ2,p, χ3,p) and motion terms (χ1,w, χ2,w, χ3,w) First order neural oridinary differential equations 14 6 models in total. One model for each term (mass and motion)
fonode-GFZ-14d-HAM GFZ HAM mass terms (χ1,p, χ2,p, χ3,p) and motion terms (χ1,w, χ2,w, χ3,w) First order neural oridinary differential equations 14 6 models in total. One model for each term (mass and motion)
fonode-GFZ-14d-OAM GFZ OAM mass terms (χ1,p, χ2,p, χ3,p) and motion terms (χ1,w, χ2,w, χ3,w) First order neural oridinary differential equations 14 6 models in total. One model for each term (mass and motion)
fonode-GFZ-14d-SLAM GFZ SLAM mass terms (χ1,p, χ2,p, χ3,p) First order neural oridinary differential equations 14 3 models in total. One model for each term (mass)
GFZ-6d-AAM GFZ AAM mass terms (χ1,p, χ2,p, χ3,p) and motion terms (χ1,w, χ2,w, χ3,w) 6
GFZ-6d-HAM GFZ HAM mass terms (χ1,p, χ2,p, χ3,p) and motion terms (χ1,w, χ2,w, χ3,w) 6
GFZ-6d-OAM GFZ OAM mass terms (χ1,p, χ2,p, χ3,p) and motion terms (χ1,w, χ2,w, χ3,w) 6
GFZ-6d-SLAM GFZ SLAM mass terms (χ1,p, χ2,p, χ3,p) 6