A note on leveraging synergy in multiple meteorological datasets with deep learning for rainfall-runoff modeling

In this paper we show the benefits of using multiple meteorological forcing products at the same time in a single LSTM-based rainfall-runoff model over just using a single product.

Abstract

A deep learning rainfall–runoff model can take multiple meteorological forcing products as input and learn to combine them in spatially and temporally dynamic ways. This is demonstrated with Long Short-Term Memory networks (LSTMs) trained over basins in the continental US, using the Catchment Attributes and Meteorological data set for Large Sample Studies (CAMELS). Using meteorological input from different data products (North American Land Data Assimilation System, NLDAS, Maurer, and Daymet) in a single LSTM significantly improved simulation accuracy relative to using only individual meteorological products. A sensitivity analysis showed that the LSTM combines precipitation products in different ways, depending on location, and also in different ways for the simulation of different parts of the hydrograph.

Paper

Kratzert, F., Klotz, D., Hochreiter, S., and Nearing, G. S.: A note on leveraging synergy in multiple meteorological data sets with deep learning for rainfall–runoff modeling, Hydrol. Earth Syst. Sci., 25, 2685–2703, https://doi.org/10.5194/hess-25-2685-2021, 2021.

Code

Code, data and pre-trained models to reproduce every detail of this paper can be found in this GitHub repository.

Citation

@Article{kratzert2021multi,
author = {Kratzert, F. and Klotz, D. and Hochreiter, S. and Nearing, G. S.},
title = {A note on leveraging synergy in multiple meteorological data sets with deep learning for rainfall--runoff modeling},
journal = {Hydrology and Earth System Sciences},
volume = {25},
year = {2021},
number = {5},
pages = {2685--2703},
doi = {10.5194/hess-25-2685-2021}
}