5 August 2019 in Water Resources Research
Spatiotemporally continuous global river discharge estimates across the full spectrum of stream orders are vital to a range of hydrologic applications, yet they remain poorly constrained. Here we present a carefully designed modeling effort (Variable Infiltration Capacity land surface model and Routing Application for Parallel computatIon of Discharge river routing model) to estimate global river discharge at very high resolutions. The precipitation forcing is from a recently published 0.1° global product that optimally merged gauge‐, reanalysis‐, and satellite‐based data. To constrain runoff simulations, we use a set of machine learning‐derived, global runoff characteristics maps (i.e., runoff at various exceedance probability percentiles) for grid‐by‐grid model calibration and bias correction. To support spaceborne discharge studies, the river flowlines are defined at their true geometry and location as much as possible—approximately 2.94 million vector flowlines (median length 6.8 km) and unit catchments are derived from a high‐accuracy global digital elevation model at 3‐arcsec resolution (~90 m), which serves as the underlying hydrography for river routing. Our 35‐year daily and monthly model simulations are evaluated against over 14,000 gauges globally. Among them, 35% (64%) have a percentage bias within ±20% (±50%), and 29% (62%) have a monthly Kling‐Gupta Efficiency ≥0.6 (0.2), showing data robustness at the scale the model is assessed. This reconstructed discharge record can be used as a priori information for the Surface Water and Ocean Topography satellite mission's discharge product, thus named “Global Reach‐level A priori Discharge Estimates for Surface Water and Ocean Topography”. It can also be used in other hydrologic applications requiring spatially explicit estimates of global river flows.
13 October 2021 in Nature Communications
The behaviors and skills of models in many geosciences (e.g., hydrology and ecosystem sciences) strongly depend on spatially-varying parameters that need calibration. A well-calibrated model can reasonably propagate information from observations to unobserved variables via model physics, but traditional calibration is highly inefficient and results in non-unique solutions. Here we propose a novel differentiable parameter learning (dPL) framework that efficiently learns a global mapping between inputs (and optionally responses) and parameters. Crucially, dPL exhibits beneficial scaling curves not previously demonstrated to geoscientists: as training data increases, dPL achieves better performance, more physical coherence, and better generalizability (across space and uncalibrated variables), all with orders-of-magnitude lower computational cost. We demonstrate examples that learned from soil moisture and streamflow, where dPL drastically outperformed existing evolutionary and regionalization methods, or required only ~12.5% of the training data to achieve similar performance. The generic scheme promotes the integration of deep learning and process-based models, without mandating reimplementation.
23 January 2018 in Journal of Geophysical Research: Atmospheres
9 September 2019 in Water Resources Research
Conventional basin‐by‐basin approaches to calibrate hydrologic models are limited to gauged basins and typically result in spatially discontinuous parameter fields. Moreover, the consequent low calibration density in space falls seriously behind the need from present‐day applications like high resolution river hydrodynamic modeling. In this study we calibrated three key parameters of the Variable Infiltration Capacity (VIC) model at every 1/8° grid‐cell using machine learning‐based maps of four streamflow characteristics for the conterminous United States (CONUS), with a total of 52,663 grid‐cells. This new calibration approach, as an alternative to parameter regionalization, applied to ungauged regions too. A key difference made here is that we tried to regionalize physical variables (streamflow characteristics) instead of model parameters whose behavior may often be less well understood. The resulting parameter fields no longer presented any spatial discontinuities and the patterns corresponded well with climate characteristics, such as aridity and runoff ratio. The calibrated parameters were evaluated against observed streamflow from 704/648 (calibration/validation period) small‐to‐medium‐sized catchments used to derive the streamflow characteristics, 3941/3809 (calibration/validation period) small‐to‐medium‐sized catchments not used to derive the streamflow characteristics as well as five large basins. Comparisons indicated marked improvements in bias and Nash‐Sutcliffe efficiency. Model performance was still poor in arid and semiarid regions, which is mostly due to both model structural and forcing deficiencies. Although the performance gain was limited by the relative small number of parameters to calibrate, the study and results here served as a proof‐of‐concept for a new promising approach for fine‐scale hydrologic model calibrations.
16 June 2025 in Water Resources Research
To expand the spatial coverage of the conventional Basin Long Short‐Term Memory (LSTM) model for river discharge estimation beyond pre‐selected individual locations, we developed a discharge modeling scheme, Grid LSTM‐RAPID, to estimate discharge for every river reach worldwide. Grid LSTM‐RAPID applies LSTM runoff estimation to the grids (0.25°), small rectangular hydrological response units (HRUs) rather than basins (irregularly shaped HRUs of any size), and then routes the grid runoff over all reaches on a global river network using the RAPID routing model. It largely maintains the strong performance of Basin LSTM over gauged basins and achieves a median Kling‐Gupta Efficiency (KGE) of 0.653 for small basins out‐of‐sample both temporally and spatially (0.688 for out‐of‐sample temporally), and a median KGE of 0.592 for other basins with larger areas and less data quality. Compared to Basin LSTM, Grid LSTM‐RAPID loses about 0.03 in median KGE for basins out‐of‐sample in both time and space in exchange for global all‐reach coverage without heavy cost. Despite this tradeoff, it significantly outperforms a well‐calibrated process‐based benchmark model. Using the new scheme, we create an improved global reach‐level daily discharge data set from 1980 to near present named GRADES‐hydroDL, which is openly shared at https://www.reachhydro.org/home/records/grades‐hydrodl.
10 April 2025 in Water Resources Research
The National Water Model (NWM) is a key tool for flood forecasting, planning, and water management. Key challenges facing the NWM include calibration and parameter regionalization when confronted with big data. We present two novel versions of high‐resolution (∼37 km2) differentiable models (a type of hybrid model): one with implicit, unit‐hydrograph‐style routing and another with explicit Muskingum‐Cunge routing in the river network. The former predicts streamflow at basin outlets whereas the latter presents a discretized product that seamlessly covers rivers in the conterminous United States (CONUS). Both versions use neural networks to provide a multiscale parameterization and process‐based equations to provide a structural backbone, which were trained simultaneously (“end‐to‐end”) on 2,807 basins across the CONUS and evaluated on 4,997 basins. Both versions show great potential to elevate future NWM performance for extensively calibrated as well as ungauged sites: the median daily Nash‐Sutcliffe efficiency of all 4,997 basins is improved to around 0.68 from 0.48 of NWM3.0. As they resolve spatial heterogeneity, both versions greatly improved simulations in the western CONUS and also in the Prairie Pothole Region, a long‐standing modeling challenge. The Muskingum‐Cunge version further improved performance for basins >10,000 km2. Overall, our results show how neural‐network‐based parameterizations can improve NWM performance for providing operational flood predictions while maintaining interpretability and multivariate outputs. The modeling system supports the Basic Model Interface (BMI), which allows seamless integration with the next‐generation NWM. We also provide a CONUS‐scale hydrologic data set for further evaluation and use.