دورية أكاديمية
Intercomparison of Deep Learning Architectures for the Prediction of Precipitation Fields With a Focus on Extremes.
| العنوان: | Intercomparison of Deep Learning Architectures for the Prediction of Precipitation Fields With a Focus on Extremes. |
|---|---|
| المؤلفون: | Otero, Noelia, Horton, Pascal |
| المصدر: | Water Resources Research; Nov2023, Vol. 59 Issue 11, p1-18, 18p |
| مصطلحات موضوعية: | DEEP learning, EXTREME weather, ATMOSPHERIC models, MACHINE learning, HUMIDITY, WEATHER forecasting, EARTH sciences |
| مصطلحات جغرافية: | EUROPE |
| مستخلص: | In recent years, the use of deep learning methods has rapidly increased in many research fields. Similarly, they have become a powerful tool within the climate scientific community. Deep learning methods have been successfully applied for different tasks, such as the identification of atmospheric patterns, weather extreme classification, or weather forecasting. However, due to the inherent complexity of atmospheric processes, the ability of deep learning models to simulate natural processes, particularly in the case of weather extremes, is still challenging. Therefore, a thorough evaluation of their performance and robustness in predicting precipitation fields is still needed, especially for extreme precipitation events, which can have devastating consequences in terms of infrastructure damage, economic losses, and even loss of life. In this study, we present a comprehensive evaluation of a set of deep learning architectures to simulate precipitation, including heavy precipitation events (>95th percentile) and extreme events (>99th percentile) over the European domain. Among the architectures analyzed here, the U‐Net network was found to be superior and outperformed the other networks in simulating precipitation events. In particular, we found that a simplified version of the original U‐Net with two encoder‐decoder levels generally achieved similar skill scores than deeper versions for predicting precipitation extremes, while significantly reducing the overall complexity and computing resources. We further assess how the model predicts through the attribution heatmaps from a layer‐wise relevance propagation explainability method. Plain Language Summary: With the increasing success of machine learning methods in Earth Sciences, deep learning is becoming a promising tool for building data‐driven models for meteorological applications. Yet, predicting extreme events, such as heavy rainfall, is still challenging. Here, we present an intercomparison of deep learning models to assess the ability of different architectures to predict precipitation events. While most of the models perform relatively well, we show that U‐Net‐based architectures outperformed the rest of the models. We additionally applied explainability methods to quantify which input features are the most important to predict precipitation events. Overall, the relative humidity showed the highest relevance values, followed by both wind components, particularly in western and southern Europe. Key Points: We present an intercomparison of deep learning models to assess the ability of different architectures to predict precipitation eventsU‐Net‐based architectures outperformed the rest of the modelsA layer‐wise relevance propagation explainability method quantify the most important feature to predict precipitation extreme [ABSTRACT FROM AUTHOR] |
| Copyright of Water Resources Research is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| قاعدة البيانات: | Complementary Index |
Full Text Finder كن أول من يترك تعليقا!