Open AccessEvaluation of Quantitative Precipitation Estimations through Hydrological Modeling in IFloodS River Basins
Author(s) -
Huan Wu,
Robert F. Adler,
Yudong Tian,
Guojun Gu,
George J. Huffman
Publication year - 2017
Publication title -
journal of hydrometeorology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.733
H-Index - 123
eISSN - 1525-755X
pISSN - 1525-7541
DOI - 10.1175/jhm-d-15-0149.1
Subject(s) - streamflow , environmental science , quantitative precipitation estimation , precipitation , flood forecasting , drainage basin , flood myth , climatology , rain gauge , structural basin , data assimilation , meteorology , hydrology (agriculture) , geology , geography , paleontology , cartography , geotechnical engineering , archaeology
A multiple-product-driven hydrologic modeling framework (MMF) is utilized for evaluation of quantitative precipitation estimation (QPE) products, motivated by improving the utility of satellite QPE in global flood modeling. This work addresses the challenge of objectively determining the relative value of various QPEs at river basin/subbasin scales. A reference precipitation dataset is created using a long-term water-balance approach with independent data sources. The intercomparison of nine QPEs and corresponding hydrologic simulations indicates that all products with long-term (2002–13) records have similar merits as over the short-term (April–June 2013) Iowa Flood Studies period. The model performance in calculated streamflow varies approximately linearly with precipitation bias, demonstrating that the model successfully translated the level of precipitation quality to streamflow quality with better streamflow simulations from QPEs with less bias. Phase 2 of the North American Land Data Assimilation System (NLDAS-2) has the best streamflow results for the Iowa–Cedar River basin, with daily and monthly Nash–Sutcliffe coefficients and mean annual bias of 0.81, 0.88, and −2.1%, respectively, for the long-term period. The evaluation also indicates that a further adjustment of NLDAS-2 to form the best precipitation estimation should consider spatial–temporal distribution of bias. The satellite-only products have lower performance (peak and timing) than other products, while simple bias adjustment can intermediately improve the quality of simulated streamflow. The TMPA research product (TMPA-RP; research-quality data) can generate results approaching those of the ground-based products with only monthly gauge-based adjustment to the TMPA real-time product (TMPA-RT; near-real-time data). It is further noted that the streamflow bias is strongly correlated to precipitation bias at various time scales, though other factors may play a role as well, especially on the daily time scale.