Hydrological Concept Formation inside Long Short-Term Memory (LSTM) networks
Preprint
- 23 November 2021
- preprint
- Published by Copernicus GmbH in EGUsphere
- Vol. 2021, 1-37
Abstract
Neural networks have been shown to be extremely effective rainfall-runoff models, where the river discharge is predicted from meteorological inputs. However, the question remains, what have these models learned? Is it possible to extract information about the learned relationships that map inputs to outputs? And do these mappings represent known hydrological concepts? Small-scale experiments have demonstrated that the internal states of Long Short-Term Memory Networks (LSTMs), a particular neural network architecture predisposed to hydrological modelling, can be interpreted. By extracting the tensors which represent the learned translation from inputs (precipitation, temperature) to outputs (discharge), this research seeks to understand what information the LSTM captures about the hydrological system. We assess the hypothesis that the LSTM replicates real-world processes and that we can extract information about these processes from the internal states of the LSTM. We examine the cell-state vector, which represents the memory of the LSTM, and explore the ways in which the LSTM learns to reproduce stores of water, such as soil moisture and snow cover. We use a simple regression approach to map the LSTM state-vector to our target stores (soil moisture and snow). Good correlations (R2 > 0.8) between the probe outputs and the target variables of interest provide evidence that the LSTM contains information that reflects known hydrological processes comparable with the concept of variable-capacity soil moisture stores. The implications of this study are threefold: 1) LSTMs reproduce known hydrological processes. 2) While conceptual models have theoretical assumptions embedded in the model a priori, the LSTM derives these from the data. These learned representations are interpretable by scientists. 3) LSTMs can be used to gain an estimate of intermediate stores of water such as soil moisture. While machine learning interpretability is still a nascent field, and our approach reflects a simple technique for exploring what the model has learned, the results are robust to different initial conditions and to a variety of benchmarking experiments. We therefore argue that deep learning approaches can be used to advance our scientific goals as well as our predictive goals.Keywords
All Related Versions
- Published version: Hydrology and Earth System Sciences, 26 (12), 3079.
Funding Information
- Natural Environment Research Council (NE/L002612/1)
This publication has 21 references indexed in Scilit:
- Enhancing Streamflow Forecast and Extracting Insights Using Long‐Short Term Memory Networks With Data Integration at Continental ScalesWater Resources Research, 2020
- Deep learning, hydrological processes and the uniqueness of placeHydrological Processes, 2020
- The ERA5 global reanalysisQuarterly Journal of the Royal Meteorological Society, 2020
- Toward Improved Predictions in Ungauged Basins: Exploiting the Power of Machine LearningWater Resources Research, 2019
- A Transdisciplinary Review of Deep Learning Research and Its Relevance for Water Resources ScientistsWater Resources Research, 2018
- Anthropogenic warming exacerbates European soil moisture droughtsNature Climate Change, 2018
- ESA CCI Soil Moisture for improved Earth system understanding: State-of-the art and future directionsRemote Sensing of Environment, 2017
- Getting the right answers for the right reasons: Linking measurements, analyses, and models to advance the science of hydrologyWater Resources Research, 2006
- Data-based mechanistic modelling of environmental, ecological, economic and engineering systemsEnvironmental Modelling & Software, 1998
- The role of groundwater in storm runoffJournal of Hydrology, 1979