Water Resources Research最新文献

{"title":"Two-Decadal Variability of Lacustrine Groundwater Discharge: Coupled Controls From Weather and Hydrologic Changes","authors":"Xiaoliang Sun, Yao Du, Jing Wu, Jiawen Xu, Hao Tian, Yamin Deng, Yiqun Gan, Yanxin Wang","doi":"10.1029/2024wr037173","DOIUrl":"https://doi.org/10.1029/2024wr037173","url":null,"abstract":"Lacustrine groundwater discharge (LGD) is a vital water and solute source for lakes. However, the understanding of the long-term temporal variability of LGD remains limited owing to insufficient insights into driving mechanisms, such as climatic and hydrologic changes. In this study, we examined the oxbow lake group in the central Yangtze River (YR) and assessed the LGD rates from 2000 to 2022 using ²²²Rn combined with meteorological and hydrological data. The findings revealed that groundwater was recharged during the wet season and discharged to the lakes during the dry season. We established a mathematical model to link the LGD rates to the meteorological and hydrological factors of the lakes, which accounted for 98.70% of the LGD rate variance. Using a predictive model combined with meteorological and hydrological data to assess the LGD rate over the past two decades, it was found that in wet years with higher precipitation and higher average YR water levels, the LGD rate was higher. The gradual increase in precipitation during the rising water and wet seasons, along with a slow rise in the YR water levels, will cause the LGD rate to exhibit a slightly increasing trend with fluctuations in the future. This study proposed an innovative approach to investigate the long-term temporal variation in LGD and identify the weather and hydrological influences on LGD.","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"217 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142325787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stream Nitrate Dynamics Driven Primarily by Discharge and Watershed Physical and Soil Characteristics at Intensively Monitored Sites: Insights From Deep Learning","authors":"G. Gorski, L. Larsen, J. Wingenroth, L. Zhang, D. Bellugi, A. P. Appling","doi":"10.1029/2023wr036591","DOIUrl":"https://doi.org/10.1029/2023wr036591","url":null,"abstract":"We developed a suite of models using deep learning to make hindcast predictions of the 7-day average backward-looking nitrate concentration at 46 predominantly agricultural sites across the midwestern and eastern United States. The models used daily observations of discharge and meteorological variables and watershed attributes describing anthropogenic modification to hydrology, nitrogen application, climate, groundwater, land use, watershed physiographic attributes, and soils. Across all sites, discharge and watershed soil and physiographic attributes showed a strong influence on model performance. Analysis of drivers across sites revealed considerable regional differences related to controlling processes such as groundwater contributions. We tested several ways to pool data across sites to develop accurate models and make the most effective use of available data. Single-site models, in which models are trained and tested at a single location, showed generally strong predictive performance (median Kling-Gupta Efficiency = 0.66), and accuracy at poorly performing sites could be improved by grouping sites with similar characteristics. Developing a single model for all sites reduced performance at several locations with distinct characteristics, suggesting that there is a threshold of dissimilarity beyond which more data does not improve the model. While many deep learning studies have shown that national or even global models can outperform local models, it is not clear that this is true for water quality constituents. This study demonstrates how data can be combined effectively, using deep learning to develop accurate and interpretable models of instream nitrate at sites where varying processes are responsible for changes in nitrate concentration.","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"38 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142321869","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring Subsurface Water Conditions in Dutch Canal Dikes During Drought Periods: Insights From Multiyear Monitoring","authors":"Bart Strijker, Timo J. Heimovaara, Sebastiaan N. Jonkman, Matthijs Kok","doi":"10.1029/2023wr036046","DOIUrl":"https://doi.org/10.1029/2023wr036046","url":null,"abstract":"Canal dikes in low-lying polders, as well as in other regions worldwide, are critical infrastructure for flood protection and water management. The subsurface water conditions can cause dike failures during excessive rainfall and prolonged periods of drought. There is a lack of multi-year monitoring of subsurface water conditions in canal dikes and an insufficient understanding of their geohydrological behavior. This study provides and analyses a novel multiyear data set of soil moisture and hydraulic heads (from February 2020 until March 2023) from a monitoring network covering various canal dikes with different characteristics in the western Netherlands. The data, including two extremely dry summers, highlight the impact of meteorological variations on the subsurface water conditions. Non-hydrostatic hydraulic head levels were observed during droughts that can be detrimental to dike stability and that are often not accounted for in safety assessments for drought situations. The effectiveness of various meteorological drought indicators applied to subsurface water conditions was evaluated: the precipitation deficit is the most reliable measure and outperforms the standardized drought indicators (SPEI and SPI). The drought recovery of dikes was analyzed to understand seasonal transitions and the sequence of different failure mechanisms, during dry and wet situations. This analysis also reveals differences between meteorological, soil moisture, and groundwater droughts, highlighting soil's storage capacity after drought and the limitations of meteorological drought indicators as proxies for soil moisture and groundwater. The insights from this study enhance assessments, inspection procedures and the identification of weak spots of dikes and other earthworks of infrastructure.","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"27 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142317410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Onset for Active Swimming of Microorganisms to Shape Their Transport in Turbulent Open Channel Flows","authors":"Zi Wu, Li Zeng, Guangmiao Li, Zheng Gong, Jie Zhan, Weiquan Jiang, Mengzhen Xu, Xudong Fu","doi":"10.1029/2024wr037586","DOIUrl":"https://doi.org/10.1029/2024wr037586","url":null,"abstract":"Research on active particles has primarily focused on transport in relatively weak flows, during which their active swimming plays a significant role. However, in natural or manmade waterways, the ambient flow velocity and water depth can be on the order of approximately 1 m/s and 1 m, respectively, generating turbulent diffusion that may be strong enough to potentially dominate the transport process, so that the active swimming might be negligible. In this paper, we propose a theoretical framework aiming at identifying the threshold at which the effects of active swimming become significant, under conditions of insufficient data for motion statistics of swimmers. While deriving the governing equation, we find that only the vertical component of the mean swimming has the potential to significantly influence the transport process. This manifests as the characteristic of inducing a non-uniform vertical concentration distribution, in competition with the mechanism of turbulent diffusion, which leads to a uniform distribution. We obtain the analytical solution for the vertical concentration distribution, with the key dimensionless parameter <i>α</i> representing the interplay between the active swimming and turbulent diffusion. The threshold is found to be approximately at the order of magnitude of <i>α</i> ∼ 0.1, below which active swimming is considered negligible. The theoretical predictions are validated by numerical simulations employing Direct Numerical Simulation and particle tracking methods. Applying the theory to two types of microorganisms transported under different flow conditions suggests that there are typical scenarios where the active swimming is negligible, and the swimmers can be treated as passive particles.","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"22 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142321871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring the Spatiotemporal Heterogeneity of Stream Nitrogen Concentrations in a Typical Human-Activity-Influenced Headwater Watershed in South China","authors":"Congsheng Fu, Haixia Zhang, Huawu Wu, Haohao Wu, Yang Cao, Ye Xia, Zichun Zhu","doi":"10.1029/2024wr038050","DOIUrl":"https://doi.org/10.1029/2024wr038050","url":null,"abstract":"Stream nitrogen concentrations significantly impact nitrogen loads and greenhouse gas emissions, but their spatiotemporal heterogeneity and human influences remain highly uncertain. This study thoroughly explored the spatiotemporal variations in stream nitrogen concentrations in a typical headwater watershed in South China. Spatially distributed measurements were conducted during 2020–2022, and mathematical modeling was implemented based on incorporating these data. More than 4,400 data points were collected for water temperature and concentrations of ammonium nitrogen (NH₄-N), nitrate nitrogen (NO_x-N), dissolved total nitrogen (DTN), total nitrogen (TN), and dissolved oxygen. Results showed that NO_x-N was the largest component of TN, with average concentrations of 1.20 and 1.66 mg L⁻¹, respectively. The stream N₂O concentration could be predicted using NH₄-N and NO_x-N concentrations via the Michaelis-Menten equation. Significant downstream decreases in NH₄-N, NO_x-N, DTN, and TN concentrations were identified in the largest river in the watershed, and clear spatial differences in these nitrogen concentrations existed among the three main rivers. Clear seasonal and annual variations in stream nitrogen concentrations were observed. NH₄-N, NO_x-N, DTN, and TN concentrations correlated with cumulative precipitation from the preceding 8–12 days, while stream N₂O concentrations correlated over 13–20 days. Stream N₂O concentrations and emissions averaged 12.77 nmol L⁻¹ and 1.12 nmol m⁻² s⁻¹, respectively, and were lower in summer than in other seasons. Upstream tea plantations, villages, and adjacent agricultural lands significantly affected nitrogen concentrations, while overflow dams did not. These findings highlight nitrogen cycle's complexity and the need for high-resolution data to guide effective watershed management.","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"214 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142317411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reconstruction of the Dynamics of a Catastrophic Crater Lake Outburst Flood, Changbaishan-Tianchi Volcano","authors":"Shengwu Qin, Jingyu Yao, Guangjie Li, Lingshuai Zhang, Xiaowei Liu, Chaobiao Zhang, Li Li","doi":"10.1029/2024wr037085","DOIUrl":"https://doi.org/10.1029/2024wr037085","url":null,"abstract":"Reconstruction of the catastrophic drainage following the Millennium Eruption (ME) of Changbaishan-Tianchi volcano in 946 ± 20 CE is of great significance, as it contributes to improving the regional maximum flood record and develop rare flood risk analysis. However, limited knowledge exists concerning the failure mode, magnitude, and transport processes of the outburst flooding. In this work, we present a whole system model that describes the paleohydrology of catastrophic drainage using geological records along the downstream valley. The model encompasses the crater lake dynamics, an approximation of the breach erosion process and flood propagation downstream. The boulder competence method was used to constrain by reasonable flow parameters, while mitigating the uncertainty caused by the ambiguous geological paleostage indicators. Paleohydrologic analysis indicates that at least 1 km³ of water was released from the caldera, with the vertical breach erosion rates as high as 34 m/hr. Volcanic activity during the ME II may have directly contributed to triggering of the flood event. The local hydrodynamic response of the downstream riverbed captures the dynamic migration patterns of sediments across spatio-temporal scales, offering a comprehensive interpretation of the specific scouring surfaces observed in the geological profile. The analysis of simulated inundation boundaries reveals that not all recorded inundations can be attributed to the crater lake outburst event. Reconstructions of megafloods based on downstream constraints on flood stage, velocity and discharge can help to infer and constrain the dynamics of dam failure mechanisms, and also contribute to our understanding of these complex paleohydrologic events.","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"28 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142306216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting Seasonal Deformation Using InSAR and Machine Learning in the Permafrost Regions of the Yangtze River Source Region","authors":"Jie Chen, Xingchen Lin, Tonghua Wu, Junming Hao, Xiaodong Wu, Defu Zou, Xiaofan Zhu, Guojie Hu, Yongping Qiao, Dong Wang, Sizhong Yang, Lina Zhang","doi":"10.1029/2023wr036700","DOIUrl":"https://doi.org/10.1029/2023wr036700","url":null,"abstract":"Quantifying seasonal deformation is essential for accurately determining the thickness of the active layer and the distribution of water content within it, providing insights into the freeze-thaw dynamics of permafrost environments and their sensitivity to climate change. Due to the limited hydraulic conductivity of the underlying permafrost, the freeze-thaw processes are largely confined to the active layer, allowing for predictable seasonal deformations. This study employed Independent Component Analysis to isolate large-scale seasonal deformation from Interferometric Synthetic Aperture Radar (InSAR) measurements taken from 2016 to 2020 in the Yangtze River Source Region (YRSR) of the Qinghai-Tibet Plateau (QTP), covering 18,500 km². We developed dedicated machine learning (ML) models that integrate these InSAR-derived measurements with various environmental proxies. By applying these models to the YRSR, we generated a comprehensive, full-coverage deformation map for permafrost terrains, achieving an <i>R</i>² value of 0.91 and an Root Mean Squared Error of approximately 0.5 cm, thus confirming the model's strong predictability of seasonal deformation in permafrost regions. Deformation magnitude varied from less than 1 cm to over 10 cm. Our analysis suggests that terrain attributes, influenced by climate and soil conditions, are the primary factors driving these deformations. This research provides valuable insights into quantifying permafrost-related seasonal deformation across expansive and rural landscapes. It also aids in assessing subsurface hydrological processes and the resilience and vulnerability of permafrost. The developed ML algorithm, with access to precise environmental data, is capable of forecasting seasonal deformations across the entire QTP and potentially throughout the Arctic.","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"32 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142306236","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Attributing Urban Evapotranspiration From Eddy-Covariance to Surface Cover: Bottom-Up Versus Top-Down","authors":"H. J. Jongen, S. Vulova, F. Meier, G. J. Steeneveld, F. A. Jansen, D. Tetzlaff, B. Kleinschmit, N. Haacke, A. J. Teuling","doi":"10.1029/2024wr037508","DOIUrl":"https://doi.org/10.1029/2024wr037508","url":null,"abstract":"Evapotranspiration &lt;span data-altimg=\"/cms/asset/5d18cc63-d71d-49d9-9c2a-db2a954fa1fb/wrcr27473-math-0001.png\"&gt;&lt;/span&gt;&lt;mjx-container ctxtmenu_counter=\"216\" ctxtmenu_oldtabindex=\"1\" jax=\"CHTML\" role=\"application\" sre-explorer- style=\"font-size: 103%; position: relative;\" tabindex=\"0\"&gt;&lt;mjx-math aria-hidden=\"true\" location=\"graphic/wrcr27473-math-0001.png\"&gt;&lt;mjx-semantics&gt;&lt;mjx-mrow data-semantic-children=\"4\" data-semantic-content=\"0,5\" data-semantic- data-semantic-role=\"leftright\" data-semantic-speech=\"left parenthesis upper E upper T right parenthesis\" data-semantic-type=\"fenced\"&gt;&lt;mjx-mo data-semantic- data-semantic-operator=\"fenced\" data-semantic-parent=\"6\" data-semantic-role=\"open\" data-semantic-type=\"fence\" style=\"margin-left: 0.056em; margin-right: 0.056em;\"&gt;&lt;mjx-c&gt;&lt;/mjx-c&gt;&lt;/mjx-mo&gt;&lt;mjx-mrow data-semantic-annotation=\"clearspeak:simple;clearspeak:unit\" data-semantic-children=\"1,2\" data-semantic-content=\"3\" data-semantic- data-semantic-parent=\"6\" data-semantic-role=\"implicit\" data-semantic-type=\"infixop\"&gt;&lt;mjx-mi data-semantic-annotation=\"clearspeak:simple\" data-semantic-font=\"italic\" data-semantic- data-semantic-parent=\"4\" data-semantic-role=\"latinletter\" data-semantic-type=\"identifier\"&gt;&lt;mjx-c&gt;&lt;/mjx-c&gt;&lt;/mjx-mi&gt;&lt;mjx-mo data-semantic-added=\"true\" data-semantic- data-semantic-operator=\"infixop,⁢\" data-semantic-parent=\"4\" data-semantic-role=\"multiplication\" data-semantic-type=\"operator\" style=\"margin-left: 0.056em; margin-right: 0.056em;\"&gt;&lt;mjx-c&gt;&lt;/mjx-c&gt;&lt;/mjx-mo&gt;&lt;mjx-mi data-semantic-annotation=\"clearspeak:simple\" data-semantic-font=\"italic\" data-semantic- data-semantic-parent=\"4\" data-semantic-role=\"latinletter\" data-semantic-type=\"identifier\"&gt;&lt;mjx-c&gt;&lt;/mjx-c&gt;&lt;/mjx-mi&gt;&lt;/mjx-mrow&gt;&lt;mjx-mo data-semantic- data-semantic-operator=\"fenced\" data-semantic-parent=\"6\" data-semantic-role=\"close\" data-semantic-type=\"fence\" style=\"margin-left: 0.056em; margin-right: 0.056em;\"&gt;&lt;mjx-c&gt;&lt;/mjx-c&gt;&lt;/mjx-mo&gt;&lt;/mjx-mrow&gt;&lt;/mjx-semantics&gt;&lt;/mjx-math&gt;&lt;mjx-assistive-mml display=\"inline\" unselectable=\"on\"&gt;&lt;math altimg=\"urn:x-wiley:00431397:media:wrcr27473:wrcr27473-math-0001\" display=\"inline\" location=\"graphic/wrcr27473-math-0001.png\" xmlns=\"http://www.w3.org/1998/Math/MathML\"&gt;&lt;semantics&gt;&lt;mrow data-semantic-=\"\" data-semantic-children=\"4\" data-semantic-content=\"0,5\" data-semantic-role=\"leftright\" data-semantic-speech=\"left parenthesis upper E upper T right parenthesis\" data-semantic-type=\"fenced\"&gt;&lt;mo data-semantic-=\"\" data-semantic-operator=\"fenced\" data-semantic-parent=\"6\" data-semantic-role=\"open\" data-semantic-type=\"fence\" stretchy=\"false\"&gt;(&lt;/mo&gt;&lt;mrow data-semantic-=\"\" data-semantic-annotation=\"clearspeak:simple;clearspeak:unit\" data-semantic-children=\"1,2\" data-semantic-content=\"3\" data-semantic-parent=\"6\" data-semantic-role=\"implicit\" data-semantic-type=\"infixop\"&gt;&lt;mi data-semantic-=\"\" data-semantic-annotation=\"clearspeak:simple\" data-semantic-font=\"italic\" data-semantic-parent=\"4\" data-semantic-role=\"latinletter\" data-semantic-type=\"identifier\"&gt;E&lt;/mi&gt;&lt;mo data-semantic-=\"\" data","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"21 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142273666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predictor Importance for Hydrological Fluxes of Global Hydrological and Land Surface Models","authors":"João Paulo L. F. Brêda, Lieke A. Melsen, Ioannis Athanasiadis, Albert Van Dijk, Vinícius A. Siqueira, Anne Verhoef, Yijian Zeng, Martine van der Ploeg","doi":"10.1029/2023wr036418","DOIUrl":"https://doi.org/10.1029/2023wr036418","url":null,"abstract":"Global Hydrological and Land Surface Models (GHM/LSMs) embody numerous interacting predictors and equations, complicating the understanding of primary hydrological relationships. We propose a model diagnostic approach based on Random Forest (RF) feature importance to detect the input variables that most influence simulated hydrological fluxes. We analyzed the JULES, ORCHIDEE, HTESSEL, SURFEX, and PCR-GLOBWB models for the relative importance of precipitation, climate, soil, land cover and topographic slope as predictors of simulated average evaporation, runoff, and surface and subsurface runoff. RF models functioned as a metamodel and could reproduce GHM/LSMs outputs with a coefficient of determination (<i>R</i>²) over 0.85 in all cases and often considerably better. The GHM/LSMs agreed that precipitation, climate and land cover share equal importance for evaporation prediction, and mean precipitation is the most important predictor of runoff, while topographic slope and soil texture have no influence on the total variance of the water balance. However, the GHM/LSMs disagreed on which features determine surface and subsurface runoff processes, especially with regard to the relative importance of soil texture and topographic slope. Finally, the selection of soil maps was only important for target variables of which soil is a relevant predictor. We conclude that estimating feature importance is a useful diagnostic approach for model intercomparison projects.","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"18 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142247158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving Discharge Predictions in Ungauged Basins: Harnessing the Power of Disaggregated Data Modeling and Machine Learning","authors":"Aggrey Muhebwa, Colin J. Gleason, Dongmei Feng, Jay Taneja","doi":"10.1029/2024wr037122","DOIUrl":"https://doi.org/10.1029/2024wr037122","url":null,"abstract":"Current machine learning methods for discharge prediction often employ aggregated basin-wide hydrometeorological data (lumped modeling) for parametric and non-parametric training. This approach may overlook the spatial heterogeneity of river systems and their impact on discharge patterns. We hypothesize that integrating spatiotemporal hydrologic knowledge into the data modeling process (distributed/disaggregated modeling) can improve the performance of discharge prediction models. To test this hypothesis, we designed experiments comparing the performance of identical Long Short-Term Memory Recurrent Neural Network (LSTM-RNN) models forced with either lumped or distributed features. We gather meteorological forcing and static attributes for the Mackenzie basin in Canada- a large and unique basin. Importantly, discharge performance is assessed out-of-sample with k-fold replication across gauges. Training LSTMs with disaggregated data significantly improved model accuracy. Specifically, there was a 9.6% increase in the mean Nash-Sutcliffe Efficiency and a 4.6% increase in the mean Kling-Gupta Efficiency, indicating a better agreement between predicted and actual observations in terms of mean, variability, and correlation. These experiments and results demonstrate the importance of integrating topologically guided geomorphologic and hydrologic information (distributed modeling) in data-driven discharge predictions.","PeriodicalId":23799,"journal":{"name":"Water Resources Research","volume":"332 1","pages":""},"PeriodicalIF":5.4,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142246266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}