Discovering Spatial Correlations between Earth Observations in Global Atmospheric State Estimation by using Adaptive Graph Structure Learning

This study aims to discover spatial correlations between Earth observations and atmospheric states to improve the forecasting accuracy of global atmospheric state estimation, which are usually conducted using conventional numerical weather prediction (NWP) systems and is the beginning of weather forecasting. NWP systems predict future atmospheric states at fixed locations, which are called NWP grid points, by analyzing previous atmospheric states and newly acquired Earth observations without fixed locations. Thus, surrounding meteorological context and the changing locations of the observations make spatial correlations between atmospheric states and observations over time. To handle complicated spatial correlations, which change dynamically, we employ spatiotemporal graph neural networks (STGNNs) with structure learning. However, structure learning has an inherent limitation that this can cause structural information loss and over-smoothing problem by generating excessive edges. To solve this problem, we regulate edge sampling by adaptively determining node degrees and considering the spatial distances between NWP grid points and observations. We validated the effectiveness of the proposed method by using real-world atmospheric state and observation data from East Asia. Even in areas with high atmospheric variability, the proposed method outperformed existing STGNN models with and without structure learning.

Explainable Graph Neural Networks for Observation Impact Analysis in Atmospheric State Estimation

This paper investigates the impact of observations on atmospheric state estimation in weather forecasting systems using graph neural networks (GNNs) and explainability methods. We integrate observation and Numerical Weather Prediction (NWP) points into a meteorological graph, extracting $k$-hop subgraphs centered on NWP points. Self-supervised GNNs are employed to estimate the atmospheric state by aggregating data within these $k$-hop radii. The study applies gradient-based explainability methods to quantify the significance of different observations in the estimation process. Evaluated with data from 11 satellite and land-based observations, the results highlight the effectiveness of visualizing the importance of observation types, enhancing the understanding and optimization of observational data in weather forecasting.

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

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.