Uncovering Flooding Mechanisms Across the Contiguous United States Through Interpretive Deep Learning on Representative Catchments

Abstract

The past decade has witnessed astonishing growth in the volume and diversity of data for water resource research, owing to the increasing number of installed observation sensor systems (McCabe et al., 2017) and the latest crowdsourcing and opportunistic sensing technologies (Jiang et al., 2019). A new challenge arising in this context is extracting information and knowledge from the data deluge (Reichstein et al., 2019). The concept of data mining, which aims to elicit implicit relationships and structures hidden within variables through modern data-driven science (Mount et al., 2016), has been used for over two decades in the field of hydrology (Abbott, 1999). Machine learning (ML), a major branch of artificial intelligence (AI), offers practical tools for data mining in hydrologic analysis (Babovic, 2005). Numerous studies have examined the roles of various ML models, such as genetic programming (Babovic & Keijzer, 2000), classification and regression trees (Fang & Shen, 2017), and self-organizing maps (Ley et al., 2011), in providing insights into hydrological processes. However, a significant tradeoff between model accuracy and interpretability persists for early generation ML models (Murdoch et al., 2019).Deep learning (DL) is a cutting-edge AI paradigm whose superiority in automatically capturing abstract information hidden in data has been acknowledged across different fields (LeCun et al., 2015;Reichstein et al., 2019). To date, one of the most successful DL architectures for modeling dynamic hydrological variables is long short-term memory (LSTM) networks (Feng et al., 2020;Kratzert et al., 2018), whose recurrent structure and unique gating mechanism can accurately capture nonlinear and temporal dependencies between variables (Hochreiter & Schmidhuber, 1997). While substantial efforts have been made to demonstrate and enhance the predictive power of LSTM networks in hydrology, few studies have illustrated how to gain scientific insights from the model (Shen, 2018). This is largely due to the accuracy-interpretability dilemma (Florez-Lopez & Ramon-Jeronimo, 2015), which is

Stream Temperature Prediction in a Shifting Environment: Explaining the Influence of Deep Learning Architecture

“…However, the DRB is a rain‐dominated system where near‐term soil saturation, precipitation, and air temperature drive key hydrological factors like stream temperature and discharge. Although work done in more snow dominated catchments has shown that LSTMs can learn long‐term snow‐water‐storage dynamics (Jiang et al., 2022), other antecedent meteorological variables generally lose their autocorrelation with hydrological variables beyond 30 days prior to the target conditions (Khanal et al., 2019). It is possible that the LSTM within RGCN is learning relationships between shallow‐groundwater and meteorological drivers, which could influence stream temperature on the scale of ∼100 days.…”

Exploring the Potential of Long Short‐Term Memory Networks for Improving Understanding of Continental‐ and Regional‐Scale Snowpack Dynamics

“…Notice however that pursuing this direction may be subject to an “accuracy‐interpretability” dilemma that has been suggested to arise from a perhaps irreconcilable conflict between a model's predictive accuracy and the possibility of understanding its behaviors (Florez‐Lopez & Ramon‐Jeronimo, 2015). It would be valuable to devote more attention to investigating the use of DL interpretation methods to facilitate AI‐assisted scientific discovery (Jiang et al., 2022). By doing so, the advances in AI might more properly be able to leverage existing theories, helping to revolutionize the next generation of earth and environmental sciences (ESS) models in the near future (Fleming, Watson, et al., 2021).…”

Trends and variability of rainfall characteristics influencing annual streamflow: A case study of southeast Australia

“…Jiang et al . (2022) uncovered the flooding mechanisms across the contiguous United States through interpretive deep learning on representative catchments.…”