Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios
This study compared the performance of Long Short-Term Memory networks (LSTM) and Soil Water Assessment Tool (SWAT) in simulating observed runoff and projecting future runoff using 11 CMIP6 GCMs. The projected runoff was estimated for two Shared Socioeconomic Pathways (SSPs), 2–4.5 and 5–8.5 for nea...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
Elsevier B.V.
2022
|
| Subjects: |
| _version_ | 1796867564168544256 |
|---|---|
| author | Song, Young Hoon Chung, Eun Sung Shahid, Shamsuddin |
| author_facet | Song, Young Hoon Chung, Eun Sung Shahid, Shamsuddin |
| author_sort | Song, Young Hoon |
| collection | ePrints |
| description | This study compared the performance of Long Short-Term Memory networks (LSTM) and Soil Water Assessment Tool (SWAT) in simulating observed runoff and projecting future runoff using 11 CMIP6 GCMs. The projected runoff was estimated for two Shared Socioeconomic Pathways (SSPs), 2–4.5 and 5–8.5 for near (2021–2060) and far (2061–2100) futures, respectively. The biases in GCM simulated climate variables were corrected using quantile mapping considering observations at 6 weather stations as reference data over the historical period (1985–2014). Five evaluation metrics were used to quantify the GCM's and hydrological models' capability to reconstruct climate variables and runoff in the Yeongsan Basin of South Korea. Uncertainties in LSTM and SWAT simulated runoff for the historical and projected periods were quantified using Bayesian Model Averaging (BMA) and reliability ensemble averaging (REA), respectively. The results showed significant improvement in bias-corrected GCMs in replicating observations in terms of all evaluation metrics. The extreme runoff estimated using general extreme value (GEV) distribution revealed the better capability of LSTM than SWAT in reproducing observed runoff at all gauging locations. The SWAT projected an increase (17.7%) while LSTM projected a decrease (−13.6%) in the future runoff for both SSPs at most locations. The uncertainty in LSTM simulated runoff was lower than in SWAT runoff at all stations for the historical period. However, the uncertainty in SWAT projected runoff was lower than LSTM projected runoff for both SSPs. This study helps assessing the ability of deep-learning versus physically-based models in hydrological modeling and therefore opens new perspectives for hydrological modeling applications. |
| first_indexed | 2024-03-05T21:29:09Z |
| format | Article |
| id | utm.eprints-103964 |
| institution | Universiti Teknologi Malaysia - ePrints |
| last_indexed | 2024-03-05T21:29:09Z |
| publishDate | 2022 |
| publisher | Elsevier B.V. |
| record_format | dspace |
| spelling | utm.eprints-1039642023-12-10T04:46:08Z http://eprints.utm.my/103964/ Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios Song, Young Hoon Chung, Eun Sung Shahid, Shamsuddin TA Engineering (General). Civil engineering (General) This study compared the performance of Long Short-Term Memory networks (LSTM) and Soil Water Assessment Tool (SWAT) in simulating observed runoff and projecting future runoff using 11 CMIP6 GCMs. The projected runoff was estimated for two Shared Socioeconomic Pathways (SSPs), 2–4.5 and 5–8.5 for near (2021–2060) and far (2061–2100) futures, respectively. The biases in GCM simulated climate variables were corrected using quantile mapping considering observations at 6 weather stations as reference data over the historical period (1985–2014). Five evaluation metrics were used to quantify the GCM's and hydrological models' capability to reconstruct climate variables and runoff in the Yeongsan Basin of South Korea. Uncertainties in LSTM and SWAT simulated runoff for the historical and projected periods were quantified using Bayesian Model Averaging (BMA) and reliability ensemble averaging (REA), respectively. The results showed significant improvement in bias-corrected GCMs in replicating observations in terms of all evaluation metrics. The extreme runoff estimated using general extreme value (GEV) distribution revealed the better capability of LSTM than SWAT in reproducing observed runoff at all gauging locations. The SWAT projected an increase (17.7%) while LSTM projected a decrease (−13.6%) in the future runoff for both SSPs at most locations. The uncertainty in LSTM simulated runoff was lower than in SWAT runoff at all stations for the historical period. However, the uncertainty in SWAT projected runoff was lower than LSTM projected runoff for both SSPs. This study helps assessing the ability of deep-learning versus physically-based models in hydrological modeling and therefore opens new perspectives for hydrological modeling applications. Elsevier B.V. 2022-09-10 Article PeerReviewed Song, Young Hoon and Chung, Eun Sung and Shahid, Shamsuddin (2022) Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios. Science of the Total Environment, 838 (NA). pp. 1-19. ISSN 0048-9697 http://dx.doi.org/10.1016/j.scitotenv.2022.156162 DOI:10.1016/j.scitotenv.2022.156162 |
| spellingShingle | TA Engineering (General). Civil engineering (General) Song, Young Hoon Chung, Eun Sung Shahid, Shamsuddin Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios |
| title | Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios |
| title_full | Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios |
| title_fullStr | Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios |
| title_full_unstemmed | Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios |
| title_short | Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios |
| title_sort | differences in extremes and uncertainties in future runoff simulations using swat and lstm for ssp scenarios |
| topic | TA Engineering (General). Civil engineering (General) |
| work_keys_str_mv | AT songyounghoon differencesinextremesanduncertaintiesinfuturerunoffsimulationsusingswatandlstmforsspscenarios AT chungeunsung differencesinextremesanduncertaintiesinfuturerunoffsimulationsusingswatandlstmforsspscenarios AT shahidshamsuddin differencesinextremesanduncertaintiesinfuturerunoffsimulationsusingswatandlstmforsspscenarios |