Numerical Weather Prediction

Numerical weather prediction (NWP) aims to forecast atmospheric conditions using mathematical models and observational data. Current research emphasizes improving forecast accuracy, particularly for extreme weather events, through the development and application of advanced machine learning models, including transformers, convolutional neural networks, and graph neural networks, often integrated with or replacing aspects of traditional physics-based NWP systems. This involves exploring novel data assimilation techniques, efficient model architectures, and improved post-processing methods to enhance forecast reliability and reduce uncertainties. The advancements in NWP have significant implications for various sectors, including disaster preparedness, resource management, and energy production.

Papers

January 17, 2022

SunCast: Solar Irradiance Nowcasting from Geosynchronous Satellite Data
Dhileeban Kumaresan, Richard Wang, Ernesto Martinez, Richard Cziva, Alberto Todeschini, Colorado J Reed, Hossein Vahabi
Numerical Weather Prediction Solar Irradiance Satellite Data Solar Nowcasting

January 10, 2022

An application of the splitting-up method for the computation of a neural network representation for the solution for the filtering equations
Dan Crisan, Alexander Lobbe, Salvador Ortiz-Latorre
Solution Path Computation Method Numerical Weather Prediction Splitting Algorithm Neural Network Representation Approximation Theory Numerical Approximation Unnormalized Statistical Model

December 13, 2021

Efficient spatio-temporal weather forecasting using U-Net
Akshay Punjabi, Pablo Izquierdo Ayala
Supervised Autoencoder U Net Numerical Weather Prediction Spatiotemporal Forecasting Weather Prediction Task

December 9, 2021

Model-Agnostic Hybrid Numerical Weather Prediction and Machine Learning Paradigm for Solar Forecasting in the Tropics
Nigel Yuan Yun Ng, Harish Gopalan, Venugopalan S. G. Raghavan, Chin Chun Ooi
Numerical Weather Prediction Machine Learning Paradigm Solar Forecasting

Numerical Weather Prediction

Papers

SunCast: Solar Irradiance Nowcasting from Geosynchronous Satellite Data

An application of the splitting-up method for the computation of a neural network representation for the solution for the filtering equations

Efficient spatio-temporal weather forecasting using U-Net

Model-Agnostic Hybrid Numerical Weather Prediction and Machine Learning Paradigm for Solar Forecasting in the Tropics