Scholarlink

<p>This contribution presents the first multivariate deep learning frameworks for jointly forecasting four tropospheric parameters using nine long-sequence architectures representing Transformer, Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Multi-Layer Perceptron (MLP) model families. Models were trained on seven years of hourly observations from 501 globally distributed stations with a 96-hour input and a 24-hour forecast horizon. The results demonstrate that the multivariate formulation consistently outperforms univariate forecasting, reducing 24-hour Root Mean Square Error (RMSE) by approximately 7% for both Zenith Tropospheric Delay (ZTD) and Zenith Wet Delay (ZWD) and providing typical full-horizon errors of 19.3 to 20.9&#xa0;mm for ZTD and ZWD, 4.1 to 5.6&#xa0;mm for Zenith Hydrostatic Delay (ZHD), and 3.1 to 3.3&#xa0;mm for Precipitable Water Vapor (PWV). Moreover, the outcomes highlight that ZTD and ZWD can be forecasted at the sub-cm level in the 1–3&#xa0;h time range. A central contribution of the study is the evaluation of physical consistency, which shows that the forecasted parameters preserve core atmospheric relationships, including the PWV/ZWD ratio and the short-term coupling between ZTD and ZWD, with violation rates below 0.01%. Although a few Transformer-based models show minor inconsistencies in the ZTD, ZHD, and ZWD closure, some architectures sustain high forecast accuracy with closure deviations constrained to 0.2&#xa0;mm. These findings demonstrate the substantial benefit of multivariate deep learning for forecasting tropospheric parameters and highlight the need for future approaches that integrate explicit physical constraints to further enhance numerical stability and physical realism.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Long-sequence deep learning frameworks for multivariate forecasting of tropospheric parameters

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月12日

查看全文
  • Mert Bezcioglu

摘要

This contribution presents the first multivariate deep learning frameworks for jointly forecasting four tropospheric parameters using nine long-sequence architectures representing Transformer, Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Multi-Layer Perceptron (MLP) model families. Models were trained on seven years of hourly observations from 501 globally distributed stations with a 96-hour input and a 24-hour forecast horizon. The results demonstrate that the multivariate formulation consistently outperforms univariate forecasting, reducing 24-hour Root Mean Square Error (RMSE) by approximately 7% for both Zenith Tropospheric Delay (ZTD) and Zenith Wet Delay (ZWD) and providing typical full-horizon errors of 19.3 to 20.9 mm for ZTD and ZWD, 4.1 to 5.6 mm for Zenith Hydrostatic Delay (ZHD), and 3.1 to 3.3 mm for Precipitable Water Vapor (PWV). Moreover, the outcomes highlight that ZTD and ZWD can be forecasted at the sub-cm level in the 1–3 h time range. A central contribution of the study is the evaluation of physical consistency, which shows that the forecasted parameters preserve core atmospheric relationships, including the PWV/ZWD ratio and the short-term coupling between ZTD and ZWD, with violation rates below 0.01%. Although a few Transformer-based models show minor inconsistencies in the ZTD, ZHD, and ZWD closure, some architectures sustain high forecast accuracy with closure deviations constrained to 0.2 mm. These findings demonstrate the substantial benefit of multivariate deep learning for forecasting tropospheric parameters and highlight the need for future approaches that integrate explicit physical constraints to further enhance numerical stability and physical realism.