Abstract:Regionalization techniques have been comprehensively discussed as the solution for runoff predictions in ungauged basins (PUB). Several types of regionalization approach have been proposed during the years. Among these, the physical similarity one was demonstrated to be one of the most robust. However, this method cannot be applied in large regions characterized by highly variable climatic conditions, such as sub-tropical areas. Therefore, this study aims to develop a new regionalization approach based on an e… Show more
“…For the scenarios without PCA denoising operation, the rainfall and runoff data is normalized with Equation (1), called Min-Max normalization. The normalized data has a value range of [1]:…”
Section: Normalizationmentioning
confidence: 99%
“…Runoff forecasting plays an essential role in flood mitigation management, agricultural water management, water transportations, and other socio-economical activities closely related to water resources. Many approaches, including physics-based models [1][2][3] and data-driven models [4][5][6], have been developed during recent decades to obtain a precise prediction result of the runoff.…”
A gated recurrent unit (GRU) network, which is a kind of artificial neural network (ANN), has been increasingly applied to runoff forecasting. However, knowledge about the impact of different input data filtering strategies and the implications of different architectures on the GRU runoff forecasting model’s performance is still insufficient. This study has selected the daily rainfall and runoff data from 2007 to 2014 in the Wei River basin in Shaanxi, China, and assessed six different scenarios to explore the patterns of that impact. In the scenarios, four manually-selected rainfall or runoff data combinations and principal component analysis (PCA) denoised input have been considered along with single directional and bi-directional GRU network architectures. The performance has been evaluated from the aspect of robustness to 48 various hypermeter combinations, also, optimized accuracy in one-day-ahead (T + 1) and two-day-ahead (T + 2) forecasting for the overall forecasting process and the flood peak forecasts. The results suggest that the rainfall data can enhance the robustness of the model, especially in T + 2 forecasting. Additionally, it slightly introduces noise and affects the optimized prediction accuracy in T + 1 forecasting, but significantly improves the accuracy in T + 2 forecasting. Though with relevance (R = 0.409~0.763, Grey correlation grade >0.99), the runoff data at the adjacent tributary has an adverse effect on the robustness, but can enhance the accuracy of the flood peak forecasts with a short lead time. The models with PCA denoised input has an equivalent, even better performance on the robustness and accuracy compared with the models with the well manually filtered data; though slightly reduces the time-step robustness, the bi-directional architecture can enhance the prediction accuracy. All the scenarios provide acceptable forecasting results (NSE of 0.927~0.951 for T + 1 forecasting and 0.745~0.836 for T + 2 forecasting) when the hyperparameters have already been optimized. Based on the results, recommendations have been provided for the construction of the GRU runoff forecasting model.
“…For the scenarios without PCA denoising operation, the rainfall and runoff data is normalized with Equation (1), called Min-Max normalization. The normalized data has a value range of [1]:…”
Section: Normalizationmentioning
confidence: 99%
“…Runoff forecasting plays an essential role in flood mitigation management, agricultural water management, water transportations, and other socio-economical activities closely related to water resources. Many approaches, including physics-based models [1][2][3] and data-driven models [4][5][6], have been developed during recent decades to obtain a precise prediction result of the runoff.…”
A gated recurrent unit (GRU) network, which is a kind of artificial neural network (ANN), has been increasingly applied to runoff forecasting. However, knowledge about the impact of different input data filtering strategies and the implications of different architectures on the GRU runoff forecasting model’s performance is still insufficient. This study has selected the daily rainfall and runoff data from 2007 to 2014 in the Wei River basin in Shaanxi, China, and assessed six different scenarios to explore the patterns of that impact. In the scenarios, four manually-selected rainfall or runoff data combinations and principal component analysis (PCA) denoised input have been considered along with single directional and bi-directional GRU network architectures. The performance has been evaluated from the aspect of robustness to 48 various hypermeter combinations, also, optimized accuracy in one-day-ahead (T + 1) and two-day-ahead (T + 2) forecasting for the overall forecasting process and the flood peak forecasts. The results suggest that the rainfall data can enhance the robustness of the model, especially in T + 2 forecasting. Additionally, it slightly introduces noise and affects the optimized prediction accuracy in T + 1 forecasting, but significantly improves the accuracy in T + 2 forecasting. Though with relevance (R = 0.409~0.763, Grey correlation grade >0.99), the runoff data at the adjacent tributary has an adverse effect on the robustness, but can enhance the accuracy of the flood peak forecasts with a short lead time. The models with PCA denoised input has an equivalent, even better performance on the robustness and accuracy compared with the models with the well manually filtered data; though slightly reduces the time-step robustness, the bi-directional architecture can enhance the prediction accuracy. All the scenarios provide acceptable forecasting results (NSE of 0.927~0.951 for T + 1 forecasting and 0.745~0.836 for T + 2 forecasting) when the hyperparameters have already been optimized. Based on the results, recommendations have been provided for the construction of the GRU runoff forecasting model.
“…To compare the accuracy of the implemented techniques in reconstructing missing water-quality data, Kling-Gupta efficiency (KGE), percent bias (PBIAS) and the Nash-Sutcliffe efficiency (NSE) were used. The latter was employed as the objective function since it is the most restrictive [35], while KGE and PBIAS were both used for validation. Equations ( 1)-( 3) present these metrics, where x i o is the ith observed value, x i c is the ith computed (or imputed) value, x o is the mean of observed values and n is the testingdataset size.…”
The monitoring of surface-water quality followed by water-quality modeling and analysis are essential for generating effective strategies in surface-water-resource management. However, worldwide, particularly in developing countries, water-quality studies are limited due to the lack of a complete and reliable dataset of surface-water-quality variables. In this context, several statistical and machine-learning models were assessed for imputing water-quality data at six monitoring stations located in the Santa Lucía Chico river (Uruguay), a mixed lotic and lentic river system. The challenge of this study is represented by the high percentage of missing data (between 50% and 70%) and the high temporal and spatial variability that characterizes the water-quality variables. The competing algorithms implement univariate and multivariate imputation methods (inverse distance weighting (IDW), Random Forest Regressor (RFR), Ridge (R), Bayesian Ridge (BR), AdaBoost (AB), Hubber Regressor (HR), Support Vector Regressor (SVR) and K-nearest neighbors Regressor (KNNR)). According to the results, more than 76% of the imputation outcomes are considered “satisfactory” (NSE > 0.45). The imputation performance shows better results at the monitoring stations located inside the reservoir than those positioned along the mainstream. IDW was the model with the best imputation results, followed by RFR, HR and SVR. The approach proposed in this study is expected to aid water-resource researchers and managers in augmenting water-quality datasets and overcoming the missing data issue to increase the number of future studies related to the water-quality matter.
“…In the case of completely ungauged basins, it is common to build the IDF/IDAF curves using the parameters determined with previous regional studies, if any exist for the case study area, or to consider the IDF/IDAF curves defined for gauged watersheds belonging to the same homogeneous region or with similar climatic and topographic characteristics of the case study, according to the spatial proximity and physical similarity principles, respectively [15].…”
The estimation of the design peak discharge is crucial for the hydrological design of hydraulic structures. A commonly used approach is to estimate the design storm through the intensity–duration–area–frequency (IDAF) curves and then use it to generate the design discharge through a hydrological model. In ungauged areas, IDAF curves and design discharges are derived throughout regionalization studies, if any exist for the area of interest, or from using the hydrological information of the closest and most similar gauged place. However, many regions around the globe remain ungauged or are very poorly gauged. In this regard, a unique opportunity is provided by satellite precipitation products developed and improved in the last decades. In this paper, we show weaknesses and potentials of satellite data and, for the first time, we evaluate their applicability for design purposes. We employ CMORPH—Climate Prediction Center MORPHing technique satellite precipitation estimates to build IDAF curves and derive the design peak discharges for the Pietrarossa dam catchment in southern Italy. Results are compared with the corresponding one provided by a regionalization study, i.e., VAPI—VAlutazione delle Piene in Italia project, usually used in Italy in ungauged areas. Results show that CMORPH performed well for the estimation of low duration and small return periods storm events, while for high return period storms, further research is still needed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.