Scholarlink

<p>This study compares daily and monthly precipitation data from the Brazilian Daily Weather Gridded Data (BR-DWGD), Climate Hazards Center InfraRed Precipitation With Station Data (CHIRPS) and Integrated Multi-satellitE Retrievals for GPM (IMERG) products with observations from four rainfall stations in the Ipojuca River Basin, in Pernambuco, Brazil, from 2001 to 2022. Daily data from four rain gauge stations in the municipalities of Arcoverde, Belo Jardim, Caruaru and Primavera were used as reference. The analysis revealed high variability in precipitation, with distinct trends for each season and product. In areas further from the coast, the products presented smaller errors. BR-DWGD had greater adherence to rain gauge data based on historical monthly averages, while CHIRPS and GPM-IMERG overestimated in the rainy months. BR-DWGD presented the lowest daily Root Mean Absolute Square Error (RMSE) (5.96&#xa0;mm), compared to CHIRPS (8.85&#xa0;mm) and GPM-IMERG (12.17&#xa0;mm). On a monthly scale, all products showed correlations above 0.8, with emphasis on BR-DWGD (0.88) which also presented the highest POD (Probability of Detection) for the different thresholds. For extreme events, GPM-IMERG was the best at detecting daily rainfall above 50&#xa0;mm. Overall, BR-DWGD was the most effective for most indicators, while GPM-IMERG was superior at capturing extreme events. CHIRPS was most effective at detecting dry days, but struggled with heavy precipitation events.</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

Evaluation of BR-DWGD, CHIRPS and GPM-IMERG precipitation products: case study in the Ipojuca River Basin

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月6日

查看全文
  • Juarez A. da S. Júnior,
  • Estephania S. Jovino,
  • Alfredo R. Neto,
  • Leidjane M. M. de Oliveira

摘要

This study compares daily and monthly precipitation data from the Brazilian Daily Weather Gridded Data (BR-DWGD), Climate Hazards Center InfraRed Precipitation With Station Data (CHIRPS) and Integrated Multi-satellitE Retrievals for GPM (IMERG) products with observations from four rainfall stations in the Ipojuca River Basin, in Pernambuco, Brazil, from 2001 to 2022. Daily data from four rain gauge stations in the municipalities of Arcoverde, Belo Jardim, Caruaru and Primavera were used as reference. The analysis revealed high variability in precipitation, with distinct trends for each season and product. In areas further from the coast, the products presented smaller errors. BR-DWGD had greater adherence to rain gauge data based on historical monthly averages, while CHIRPS and GPM-IMERG overestimated in the rainy months. BR-DWGD presented the lowest daily Root Mean Absolute Square Error (RMSE) (5.96 mm), compared to CHIRPS (8.85 mm) and GPM-IMERG (12.17 mm). On a monthly scale, all products showed correlations above 0.8, with emphasis on BR-DWGD (0.88) which also presented the highest POD (Probability of Detection) for the different thresholds. For extreme events, GPM-IMERG was the best at detecting daily rainfall above 50 mm. Overall, BR-DWGD was the most effective for most indicators, while GPM-IMERG was superior at capturing extreme events. CHIRPS was most effective at detecting dry days, but struggled with heavy precipitation events.