Regional Weather Variable Predictions by Machine Learning with Near-Surface Observational and Atmospheric Numerical Data

2024-12-11

arXiv.org

AI Summary

To address the limitations of conventional regional weather forecasting—namely, low spatiotemporal resolution and poor accuracy in ungauged areas—this paper proposes a Micro-Macro dual-layer machine learning framework. The MiMa model integrates 5-minute surface observations with hourly numerical weather predictions to enable high-resolution nowcasting. The Re-MiMa model extends this capability via geography-aware feature embedding–driven regional transfer learning and station-specific modelet design, enabling, for the first time, generalizable nowcasting in ungauged regions. Both models adopt a multivariate time-series Transformer encoder-decoder architecture, eliminating reliance on dense observational networks. Experiments demonstrate that MiMa achieves high accuracy for single-station nowcasting on the Kentucky Mesonet. Re-MiMa reduces nowcasting errors in ungauged areas by over 30%, significantly enhancing fine-grained,全域 meteorological service capabilities.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Computer Vision: Multi-modal Vision
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Accurate and timely regional weather prediction is vital for sectors dependent on weather-related decisions. Traditional prediction methods, based on atmospheric equations, often struggle with coarse temporal resolutions and inaccuracies. This paper presents a novel machine learning (ML) model, called MiMa (short for Micro-Macro), that integrates both near-surface observational data from Kentucky Mesonet stations (collected every five minutes, known as Micro data) and hourly atmospheric numerical outputs (termed as Macro data) for fine-resolution weather forecasting. The MiMa model employs an encoder-decoder transformer structure, with two encoders for processing multivariate data from both datasets and a decoder for forecasting weather variables over short time horizons. Each instance of the MiMa model, called a modelet, predicts the values of a specific weather parameter at an individual Mesonet station. The approach is extended with Re-MiMa modelets, which are designed to predict weather variables at ungauged locations by training on multivariate data from a few representative stations in a region, tagged with their elevations. Re-MiMa (short for Regional-MiMa) can provide highly accurate predictions across an entire region, even in areas without observational stations. Experimental results show that MiMa significantly outperforms current models, with Re-MiMa offering precise short-term forecasts for ungauged locations, marking a significant advancement in weather forecasting accuracy and applicability.

Problem

• Improves regional weather prediction accuracy
• Integrates micro and macro data sources
• Forecasts weather at ungauged locations

Innovation

• Machine learning model MiMa
• Encoder-decoder transformer structure
• Regional-MiMa for ungauged locations