The effect of calibration data length on the performance of a conceptual hydrological model versus LSTM and GRU
- We systematically explore the effect of calibration data length on the performance of a conceptual hydrological model, GR4H, in comparison to two Artificial Neural Network (ANN) architectures: Long Short-Term Memory Networks (LSTM) and Gated Recurrent Units (GRU), which have just recently been introduced to the field of hydrology. We implemented a case study for six river basins across the contiguous United States, with 25 years of meteorological and discharge data. Nine years were reserved for independent validation; two years were used as a warm-up period, one year for each of the calibration and validation periods, respectively; from the remaining 14 years, we sampled increasing amounts of data for model calibration, and found pronounced differences in model performance. While GR4H required less data to converge, LSTM and GRU caught up at a remarkable rate, considering their number of parameters. Also, LSTM and GRU exhibited the higher calibration instability in comparison to GR4H. These findings confirm the potential of modernWe systematically explore the effect of calibration data length on the performance of a conceptual hydrological model, GR4H, in comparison to two Artificial Neural Network (ANN) architectures: Long Short-Term Memory Networks (LSTM) and Gated Recurrent Units (GRU), which have just recently been introduced to the field of hydrology. We implemented a case study for six river basins across the contiguous United States, with 25 years of meteorological and discharge data. Nine years were reserved for independent validation; two years were used as a warm-up period, one year for each of the calibration and validation periods, respectively; from the remaining 14 years, we sampled increasing amounts of data for model calibration, and found pronounced differences in model performance. While GR4H required less data to converge, LSTM and GRU caught up at a remarkable rate, considering their number of parameters. Also, LSTM and GRU exhibited the higher calibration instability in comparison to GR4H. These findings confirm the potential of modern deep-learning architectures in rainfall runoff modelling, but also highlight the noticeable differences between them in regard to the effect of calibration data length.…
Author details: | Georgy AyzelORCiDGND, Maik HeistermannORCiDGND |
---|---|
DOI: | https://doi.org/10.1016/j.cageo.2021.104708 |
ISSN: | 0098-3004 |
ISSN: | 1873-7803 |
Title of parent work (English): | Computers & geosciences : an international journal devoted to the publication of papers on all aspects of geocomputation and to the distribution of computer programs and test data sets ; an official journal of the International Association for Mathematical Geology |
Subtitle (English): | a case study for six basins from the CAMELS dataset |
Publisher: | Elsevier |
Place of publishing: | Amsterdam |
Publication type: | Article |
Language: | English |
Date of first publication: | 2021/04/01 |
Publication year: | 2021 |
Release date: | 2023/02/02 |
Tag: | Artificial neural networks; Calibration; Deep learning; Rainfall-runoff; modelling |
Volume: | 149 |
Article number: | 104708 |
Number of pages: | 12 |
Funding institution: | Geo.X, the Research Network for Geosciences in Berlin and Potsdam; ClimXtreme - the Research Network on Climate Change and Extreme Events |
Organizational units: | Mathematisch-Naturwissenschaftliche Fakultät / Institut für Umweltwissenschaften und Geographie |
DDC classification: | 0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 004 Datenverarbeitung; Informatik |
5 Naturwissenschaften und Mathematik / 55 Geowissenschaften, Geologie / 550 Geowissenschaften | |
Peer review: | Referiert |