CloudNine: Analyzing Meteorological Observation Impact on Weather Prediction Using Explainable Graph Neural Networks

TL;DR

CloudNine introduces an explainable graph neural network system that enables detailed, multi-scale analysis of how individual meteorological observations influence weather predictions, addressing limitations of previous impact analysis methods.

Contribution

The paper presents a novel XGNN-based system that provides localized, spatio-temporal impact analysis of meteorological observations on weather forecasting.

Findings

Enables visualization of observation impacts at specific locations and times.

Integrates atmospheric state estimation with weather prediction models.

Offers a web application for interactive impact analysis.

Abstract

The impact of meteorological observations on weather forecasting varies with sensor type, location, time, and other environmental factors. Thus, quantitative analysis of observation impacts is crucial for effective and efficient development of weather forecasting systems. However, the existing impact analysis methods are difficult to be widely applied due to their high dependencies on specific forecasting systems. Also, they cannot provide observation impacts at multiple spatio-temporal scales, only global impacts of observation types. To address these issues, we present a novel system called ``CloudNine,'' which allows analysis of individual observations' impacts on specific predictions based on explainable graph neural networks (XGNNs). Combining an XGNN-based atmospheric state estimation model with a numerical weather prediction model, we provide a web application to search for observations in the 3D space of the Earth system and to visualize the impact of individual observations on predictions in specific spatial regions and time periods.