Scholarlink

<p>This paper studies the impact on model precipitation forecast of the combined assimilation of GNSS-ZTD (global navigation satellite system–zenith total delay) data and gradients compared to the GNSS-ZTD assimilation alone. Two case studies are considered over Lombardy, northern Italy, where data from 28 GNSS receivers were available, with three different model configurations: one without GNSS data assimilation (CTRL), one with GNSS-ZTD data assimilation (ZTD) and one with both ZTD and gradients data assimilation (ZTDGRAD). Simulations are performed by Weather Research and Forecasting (WRF) model and last 12&#xa0;h for each case study. Assimilation is performed by 3DVar through WRFDA software. Two forecast phases lasting 3&#xa0;h are analyzed for each case study: hours from 1 to 4 (1–4&#xa0;h) and hours from 3 to 6 (3–6&#xa0;h) after the end of the assimilation phase. Precipitation is verified against data coming from about 1300 rain gauges. GNSS data assimilation improves CTRL forecast for both cases, better catching events location and intensities and reducing false alarms. For the first case, GNSS data assimilation has a positive impact at both 1–4&#xa0;h and 3–6&#xa0;h phases, while for the second case the impact given by GNSS data assimilation is positive for the 1–4&#xa0;h phase and neutral for the 3–6&#xa0;h phase. Among the two configurations with GNSS data assimilation, ZTDGRAD shows the best performances for both cases.</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

Assimilation of anisotropic GNSS observations in the WRF model: preliminary results of the NEW-ARGENT project

  • 期刊论文

  • 发表日期：

    2026年2月14日

查看全文
  • Rosa Claudia Torcasio,
  • Eugenio Realini,
  • Mattia Crespi,
  • Stefano Federico

摘要

This paper studies the impact on model precipitation forecast of the combined assimilation of GNSS-ZTD (global navigation satellite system–zenith total delay) data and gradients compared to the GNSS-ZTD assimilation alone. Two case studies are considered over Lombardy, northern Italy, where data from 28 GNSS receivers were available, with three different model configurations: one without GNSS data assimilation (CTRL), one with GNSS-ZTD data assimilation (ZTD) and one with both ZTD and gradients data assimilation (ZTDGRAD). Simulations are performed by Weather Research and Forecasting (WRF) model and last 12 h for each case study. Assimilation is performed by 3DVar through WRFDA software. Two forecast phases lasting 3 h are analyzed for each case study: hours from 1 to 4 (1–4 h) and hours from 3 to 6 (3–6 h) after the end of the assimilation phase. Precipitation is verified against data coming from about 1300 rain gauges. GNSS data assimilation improves CTRL forecast for both cases, better catching events location and intensities and reducing false alarms. For the first case, GNSS data assimilation has a positive impact at both 1–4 h and 3–6 h phases, while for the second case the impact given by GNSS data assimilation is positive for the 1–4 h phase and neutral for the 3–6 h phase. Among the two configurations with GNSS data assimilation, ZTDGRAD shows the best performances for both cases.