Dadu River Runoff Forecasting via Seq2Seq

2019 ◽  
Author(s):  
Jian Xu ◽  
Wei Luo ◽  
Ying Huang
Keyword(s):  
River Runoff ◽  
Runoff Forecasting ◽  
Dadu River

scholarly journals A system-theory-based model for monthly river runoff forecasting: model calibration and optimization

2014 ◽  
Vol 62 (1) ◽  
pp. 82-88 ◽  
Author(s):  
Jianhua Wu ◽  
Hui Qian ◽  
Peiyue Li ◽  
Yanxun Song
Keyword(s):  
System Theory ◽  
Frequency Analysis ◽  
River Runoff ◽  
Optimization Technique ◽  
Essential Element ◽  
Forecasting Model ◽  
Observation Data ◽  
Prediction Errors ◽  
Groundwater Exploitation ◽  
Runoff Forecasting

Abstract River runoff is not only a crucial part of the global water cycle, but it is also an important source for hydropower and an essential element of water balance. This study presents a system-theory-based model for river runoff forecasting taking the Hailiutu River as a case study. The forecasting model, designed for the Hailiutu watershed, was calibrated and verified by long-term precipitation observation data and groundwater exploitation data from the study area. Additionally, frequency analysis, taken as an optimization technique, was applied to improve prediction accuracy. Following model optimization, the overall relative prediction errors are below 10%. The system-theory-based prediction model is applicable to river runoff forecasting, and following optimization by frequency analysis, the prediction error is acceptable.

DISTRIBUTION OF DISSOLVED CHEMICAL ELEMENTS IN THE YENISEI RIVER ESTUARY AND ADJACENT WATER AREA OF THE KARA SEA

2017 ◽  
Author(s):  
Alla Savenko ◽  
Alla Savenko ◽  
Oleg Pokrovsky ◽  
Oleg Pokrovsky ◽  
Irina Streletskaya ◽  
...  
Keyword(s):  
Trace Elements ◽  
River Runoff ◽  
Major Ions ◽  
Chemical Elements ◽  
River Estuary ◽  
Water Area ◽  
Kara Sea ◽  
Yenisei River ◽  
Adjacent Water ◽  
The Yenisei River

The distribution of dissolved chemical elements (major ions, nutrients, and trace elements) in the Yenisei River estuary and adjacent water area in 2009 and 2010 are presented. These results were compared to the data obtained during previous hydrochemical studies of this region. The transport of major cations (Na, K, Mg, Ca) and some trace elements (Rb, Cs, Sr, B, F, As, Mo, U) in the estuary follows conservative mixing. Alkalinity also belongs to conservative components, however this parameter exhibits substantial spatial heterogeneity caused by complex hydrological structure of the Yenisei Bay and adjoining part of the Kara Sea formed under the influence of several sources of desalination and salty waters inflow. Concentrations of Pmin, Si, and V in the desalinized waters of photic layer decrease seaward owing to uptake by phytoplankton. The losses of these elements reach 30–57, 30, and 9% of their supply by river runoff, respectively. The content of dissolved phosphates and vanadium in the intermediate and near-bottom layers of the Yenisei River estuary strongly increases with salinity due to regeneration of precipitated organic matter, whereas silica remineralization is much less pronounced. Barium is characterized by additional input of dissolved forms in the mixing zone in the quantity comparable to that carried out by river runoff. This may be caused by its desorption from river suspended matter due to ion exchange. The transport of dissolved Al and Mn in the estuarine zone is probably controlled by the coagulation and flocculation of organic and organomineral colloids, which is indicated by a decrease in the concentration of these elements at the beginning of the estuary (31 and 56%, respectively) followed by a stable concentration further seaward.

Improving Runoff Simulations using Satellite-observed Time-series of Snow Covered Area

2003 ◽  
Vol 34 (4) ◽  
pp. 281-294 ◽  
Author(s):  
R.V. Engeset ◽  
H-C. Udnæs ◽  
T. Guneriussen ◽  
H. Koren ◽  
E. Malnes ◽  
...  
Keyword(s):  
Time Series ◽  
Radar Sensors ◽  
Noaa Avhrr ◽  
Runoff Forecasting ◽  
Snow Covered Area ◽  
Covered Area ◽  
Satellite Sensors ◽  
Major Floods ◽  
Using Data ◽  
Southern Norway

Snowmelt can be a significant contributor to major floods, and hence updated snow information is very important to flood forecasting services. This study assesses whether operational runoff simulations could be improved by applying satellite-derived snow covered area (SCA) from both optical and radar sensors. Currently the HBV model is used for runoff forecasting in Norway, and satellite-observed SCA is used qualitatively but not directly in the model. Three catchments in southern Norway are studied using data from 1995 to 2002. The results show that satellite-observed SCA can be used to detect when the models do not simulate the snow reservoir correctly. Detecting errors early in the snowmelt season will help the forecasting services to update and correct the models before possible damaging floods. The method requires model calibration against SCA as well as runoff. Time-series from the satellite sensors NOAA AVHRR and ERS SAR are used. Of these, AVHRR shows good correlation with the simulated SCA, and SAR less so. Comparison of simultaneous data from AVHRR, SAR and Landsat ETM+ for May 2000 shows good inter-correlation. Of a total satellite-observed area of 1,088 km2, AVHRR observed a SCA of 823 km2 and SAR 720 km2, as compared to 889 km2 using ETM+.

Contemporary and Scenario Changes in River Runoff in the Don Basin

2020 ◽  
Vol 47 (6) ◽  
pp. 913-923
Author(s):  
A. G. Geordiadi ◽  
I. P. Milyukova ◽  
E. A. Kashutina
Keyword(s):  
River Runoff

scholarly journals Zonal aspects of the influence of forest cover change on runoff in northern river basins of Central Siberia

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
A. Onuchin ◽  
Т. Burenina ◽  
А. Shvidenko ◽  
D. Prysov ◽  
A. Musokhranova
Keyword(s):  
River Runoff ◽  
Catchment Area ◽  
Forest Cover ◽  
River Basins ◽  
Forest Vegetation ◽  
Middle Taiga ◽  
Northern Eurasia ◽  
Runoff Formation ◽  
Forest Tundra

Abstract Background Assessment of the reasons for the ambiguous influence of forests on the structure of the water balance is the subject of heated debate among forest hydrologists. Influencing the components of total evaporation, forest vegetation makes a significant contribution to the process of runoff formation, but this process has specific features in different geographical zones. The issues of the influence of forest vegetation on river runoff in the zonal aspect have not been sufficiently studied. Results Based on the analysis of the dependence of river runoff on forest cover, using the example of nine catchments located in the forest-tundra, northern and middle taiga of Northern Eurasia, it is shown that the share of forest cover in the total catchment area (percentage of forest cover, FCP) has different effects on runoff formation. Numerical experiments with the developed empirical models have shown that an increase in forest cover in the catchment area in northern latitudes contributes to an increase in runoff, while in the southern direction (in the middle taiga) extensive woody cover of catchments “works” to reduce runoff. The effectiveness of geographical zonality in regards to the influence of forests on runoff is more pronounced in the forest-tundra zone than in the zones of northern and middle taiga. Conclusion The study of this problem allowed us to analyze various aspects of the hydrological role of forests, and to show that forest ecosystems, depending on environmental conditions and the spatial distribution of forest cover, can transform water regimes in different ways. Despite the fact that the process of river runoff formation is controlled by many factors, such as temperature conditions, precipitation regime, geomorphology and the presence of permafrost, the models obtained allow us to reveal general trends in the dependence of the annual river runoff on the percentage of forest cover, at the level of catchments. The results obtained are consistent with the concept of geographic determinism, which explains the contradictions that exist in assessing the hydrological role of forests in various geographical and climatic conditions. The results of the study may serve as the basis for regulation of the forest cover of northern Eurasian river basins in order to obtain the desired hydrological effect depending on environmental and economic conditions.

scholarly journals Influence of Random Forest Hyperparameterization on Short-Term Runoff Forecasting in an Andean Mountain Catchment

Atmosphere ◽  
2021 ◽  
Vol 12 (2) ◽  
pp. 238
Author(s):  
Pablo Contreras ◽  
Johanna Orellana-Alvear ◽  
Paul Muñoz ◽  
Jörg Bendix ◽  
Rolando Célleri
Keyword(s):  
Random Forest ◽  
Model Performance ◽  
Early Warning Systems ◽  
Point Of View ◽  
Lead Times ◽  
Physical Parameters ◽  
Runoff Forecasting ◽  
Spatio Temporal ◽  
Improved Model ◽  
Search Approach

The Random Forest (RF) algorithm, a decision-tree-based technique, has become a promising approach for applications addressing runoff forecasting in remote areas. This machine learning approach can overcome the limitations of scarce spatio-temporal data and physical parameters needed for process-based hydrological models. However, the influence of RF hyperparameters is still uncertain and needs to be explored. Therefore, the aim of this study is to analyze the sensitivity of RF runoff forecasting models of varying lead time to the hyperparameters of the algorithm. For this, models were trained by using (a) default and (b) extensive hyperparameter combinations through a grid-search approach that allow reaching the optimal set. Model performances were assessed based on the R2, %Bias, and RMSE metrics. We found that: (i) The most influencing hyperparameter is the number of trees in the forest, however the combination of the depth of the tree and the number of features hyperparameters produced the highest variability-instability on the models. (ii) Hyperparameter optimization significantly improved model performance for higher lead times (12- and 24-h). For instance, the performance of the 12-h forecasting model under default RF hyperparameters improved to R2 = 0.41 after optimization (gain of 0.17). However, for short lead times (4-h) there was no significant model improvement (0.69 < R2 < 0.70). (iii) There is a range of values for each hyperparameter in which the performance of the model is not significantly affected but remains close to the optimal. Thus, a compromise between hyperparameter interactions (i.e., their values) can produce similar high model performances. Model improvements after optimization can be explained from a hydrological point of view, the generalization ability for lead times larger than the concentration time of the catchment tend to rely more on hyperparameterization than in what they can learn from the input data. This insight can help in the development of operational early warning systems.

scholarly journals River runoff modeling in the European territory of Russia

CATENA ◽  
2021 ◽  
Vol 203 ◽  
pp. 105327
Author(s):  
O. Yermolaev ◽  
S. Mukharamova ◽  
E. Vedeneeva
Keyword(s):  
River Runoff ◽  
Runoff Modeling

scholarly journals The mineralogic and isotopic fingerprint of equatorial carbonates: Kepulauan Seribu, Indonesia

2021 ◽  
Author(s):  
Dwi Amanda Utami ◽  
Lars Reuning ◽  
Maximillian Hallenberger ◽  
Sri Yudawati Cahyarini
Keyword(s):  
River Runoff ◽  
Patch Reef ◽  
High Water ◽  
Sem Analysis ◽  
Shallow Waters ◽  
Carbonate System ◽  
Riverine Water ◽  
High Turbidity ◽  
Java Sea ◽  
River Catchments

AbstractKepulauan Seribu is an isolated patch reef complex situated in the Java Sea (Indonesia) and is a typical example for a humid, equatorial carbonate system. We investigate the mineralogical and isotopic fingerprint of Panggang, one of the reef platforms of Kepulauan Seribu, to evaluate differences to other carbonate systems, using isotope in combination with XRD and SEM analysis. A characteristic property of shallow water (< 20 m) sediments from Kepulauan Seribu is their increased LMC content (~ 10%) derived from some genera of rotaliid foraminifers and bivalves. The relative abundance of these faunal elements in shallow waters might be related to at least temporary turbid conditions caused by sediment-laden river runoff. This influence is also evidenced by the presence of low amounts of siliciclastic minerals below the regional wave base. Kepulauan Seribu carbonates are characterized by very low δ13C and δ18O values. This is related to the isotopically depleted riverine input. The δ13CDIC in riverine water is reduced by the contribution of 12C from riverside mangroves. Deep atmospheric convection and intensive rains contribute 18O-depleted freshwater in the river catchments, finally reducing salinity in the Java Sea. The depleted δ13C signature in carbonates is further enhanced by the lack of green algae and inorganic carbonates and abundance of coral debris. Low δ18O values in carbonates are favored by the high water temperatures in the equatorial setting. Since equatorial carbonates in SE Asia, including the Java Sea, are typically influenced by high turbidity and/or river runoff, the observed distinctively low isotope values likely are characteristic for equatorial carbonate systems in the region.

Sign in / Sign up

Export Citation Format

Share Document