the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 07 Feb 2025
An explainable deep learning model based on hydrological principles for flood simulation and forecasting
Abstract. Deep learning (DL) models always perform well in hydrological simulation but lack physical-based principles. To address this gap, we integrate the runoff generation and flow routing principals of Xinanjiang (XAJ) model into the architecture of recurrent neural network (RNN) units and establish a physical-based XAJRNN neural network layer. Subsequently, this layer is fused with LSTM layers to construct an explainable deep learning (EDL) model, which underwent testing at the Lushui River and Qingjiang River basins in China. Compared to benchmark models, the proposed EDL model performs very well, the average Nash-Sutcliffe efficiency (NSE)values for these two basins are 0.98 and 0.94, respectively. The small flood peak relative errors (PRE) and peak timing difference (∆T) close to zero demonstrate that the EDL model can accuracy simulate flood events. Notably, the EDL model not only enhances simulation accuracy over ordinary DL models but also enhances interpretability by incorporating physical principles, thereby offering fresh insights for the fusion of DL and hydrological models for flood simulation and forecasting.
- Preprint
(3309 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 07 Apr 2025)
-
RC1: 'Comment on egusphere-2025-279', Anonymous Referee #1, 11 Mar 2025
reply
The paper is well-structured and provides a solid foundation. However, there are a few suggestions for improvement regarding this study:
- Lines 30–31 mention the limitations of traditional hydrological models. More details on these limitations should be provided.
- Line 51 contains a typo: “As an RNN subset in DL.”
- Figure 1 shows that the rainfall gauge color is too similar to the elevation color. Please use a different color for better distinction. Besides that, the elevations of the two basins can be unified in color scale to make it easier to compare the terrain differences.
- In section 2.2, mention that the timestep of data are different in two basins. For example, the rainfall data of Lushui River basin is 3h, but Qingjiang River basin is 6h.
- The Xinanjiang (XAJ) hydrological model should be explicitly mentioned in Line 182.
- Equation 2 and Figure 3 present the proposed XAJRANN layer. How are its parameters optimized?
- Lines 247-248 mention that “The choice of LSTM is based on the numerous studies demonstrating its ability to improve the performance of hydrological model simulations.”, the author should add some references for these statements.
- Figure 4 illustrates the structure of the EDL model. In the XAJRNN layer, the model outputs actual evapotranspiration (Et), areal mean free water storage (S0), areal mean tension water storage (W), and basin outflow discharge (Q). Is this output generated through supervised learning? Subsequently, in the LSTM layer, the model produces runoff, which may also be trained using supervised learning. Does it make sense to train the model twice?
- Is there a typo in Lines 296 “and vis vasa.”? (vice versa?)
- The RMSE values of the two basins in Table 1 are quite different. Please explain the results based on data statistics.
- Table 1 presents the model performance. During the testing phase, the LSTM model achieved better NSE, RE, and RMSE for the Qingjiang River. What could be the reason for this?
- What do the colors in the scatter plots of Figure 5 and Figure 6 represent? Please add a legend.
- Figure 5 shows that there are flood events exceeding 3000 m3/s during the test period, while there are fewer flood events exceeding 3000 m3/s during the training period. Please explain whether this is the reason why the scatter points are below the 1:1 ideal line in the high flow range.
- Table 2 indicates that the ∆𝑇 of XAJ and LSTM model are more than one day in 20170702 event. Is there a calculation error since the NSE is 0.93 for LSTM?
- There is a problem with the statements for Lines 394-396. The discrepancies in the rising speed during the flood rising phase compared to the observations may be due to the slow response of the model to rainfall rather than to the models' insufficient ability to simulate low flow conditions.
- Lines 401-403 mention that all three models underestimated the peak flow, and the simulated peak was significantly delayed compared to the observed peak, especially under complex terrain conditions. Please select stations with complex terrain and simple terrain for result comparisons to illustrate the impact of terrain on model simulations.
- The author should add some statements about the simulated time horizon (e.g. T+1, T+2, …).
Citation: https://doi.org/10.5194/egusphere-2025-279-RC1 -
RC2: 'Comment on egusphere-2025-279', Anonymous Referee #2, 12 Mar 2025
reply
This paper integrates the runoff generation and flow routing principles of the Xinanjiang model into a recurrent neural network framework, proposing the XAJRNN layer and constructing an EDL model. This approach enhances the physical interpretability of deep learning-based flood forecasting. Using the Lushui River and Qingjiang River basins as case studies, the EDL model is compared with benchmark models, demonstrating superior performance in flood simulation. The study is well-structured, data-driven, and methodologically rigorous, offering a novel perspective and valuable tool for explainable deep learning in hydrology. However, improvements in clarity, graphical details, and language are needed.
(1) Line 128: Please provide additional explanation on why the runoff generation and flow routing principles of the Xinanjiang model were chosen to construct an explainable deep learning model, specifically elaborating on its advantages and applicability.
(2) Line 131: Please further explain the rationale for using LSTM neural network layers to construct the model, highlighting its superiority.
(3) Lines 154-166: To enhance the completeness of the research background, it is recommended that information on the magnitude and frequency of historical floods in the study area be supplemented.
(4) Line 168: Please adjust the scale of the river curves in Figure 1 to improve the aesthetic quality and clarity of the illustration.
(5) Line 224: The author mentions "a similar structure"; please specify in which aspects this similarity is reflected to improve clarity.
(6) Equations (1) and (2) and Figure 3 (b): The parameter symbols in the equations do not match those used in Figure 3 (b). Please carefully verify and ensure consistency.
(7) Lines 258-260: The XAJRNN layer outputs four physical variables of interest. Please explain why these four variables were selected as outputs instead of others.
(8) Lines 274-280: The paper mentions that a genetic algorithm was used to optimize the parameters of the Xinanjiang model. Please provide the obtained optimal parameter values and include them in the relevant section.
(9) Figure 8 (d): The simulation performance of the EDL and the benchmark models appears to be poor. Please analyze the potential reasons for this issue.
(10) Language expression: Some parts of the paper contain repetitive phrasing. It is recommended to refine the text to improve fluency and conciseness.
(11) Reference formatting: Please carefully check the reference formatting to ensure compliance with the journal’s requirements, including the correct spelling of author names, publication year format, DOI, and page ranges.
Citation: https://doi.org/10.5194/egusphere-2025-279-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 146 | 42 | 5 | 193 | 4 | 5 |
- HTML: 146
- PDF: 42
- XML: 5
- Total: 193
- BibTeX: 4
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1