Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets

In this manuscript we show for the first time how to train a single LSTM-based neural network as general hydrology model for hundreds of basins. Furthermore, we proposed the Entity-Aware LSTM (EA-LSTM) in which static features are used explicitly to subset the model for a specific entity (here a catchment).

Abstract

Regional rainfall-runoff modeling is an old but still mostly out-standing problem in Hydrological Sciences. The problem currently is that traditional hydrological models degrade significantly in performance when calibrated for multiple basins together instead of for a single basin alone. In this paper, we propose a novel, data-driven approach using Long Short-Term Memory networks (LSTMs), and demonstrate that under a “big data” paradigm, this is not necessarily the case. By training a single LSTM model on 531 basins from the CAMELS data set using meteorological time series data and static catchment attributes, we were able to significantly improve performance compared to a set of several different hydrological benchmark models. Our proposed approach not only significantly outperforms hydrological models that were calibrated regionally but also achieves better performance than hydrological models that were calibrated for each basin individually. Furthermore, we propose an adaption to the standard LSTM architecture, which we call an Entity-Aware-LSTM (EA-LSTM), that allows for learning, and embedding as a feature layer in a deep learning model, catchment similarities. We show that this learned catchment similarity corresponds well with what we would expect from prior hydrological understanding.

Paper

Kratzert, F., Klotz, D., Shalev, G., Klambauer, G., Hochreiter, S., and Nearing, G.: Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets, Hydrol. Earth Syst. Sci., 23, 5089–5110, https://doi.org/10.5194/hess-23-5089-2019, 2019.

Code

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

Citation

@article{kratzert2019ealstm,
author = {Kratzert, F. and Klotz, D. and Shalev, G. and Klambauer, G. and Hochreiter, S. and Nearing, G.},
title = {Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets},
journal = {Hydrology and Earth System Sciences},
volume = {23},
year = {2019},
number = {12},
pages = {5089--5110},
url = {https://www.hydrol-earth-syst-sci.net/23/5089/2019/},
doi = {10.5194/hess-23-5089-2019}
}