Machine learning for accelerating 2D flood models: Potential and challenges     

Behzad Jamali  Ehsan Haghighat  Aleksandar Ignjatovic  João P. Leitão  Ana Deletic 《水文研究》2021,35(4):e14064

Two-dimensional hydrodynamic models numerically solve full Shallow Water Equations (SWEs). Despite their high accuracy, these models have long simulation run times and therefore are of limited use for exploratory or real-time flood predictions. We investigated the possibility of improving flood modelling speed using Machine Learning (ML). We propose a new method that replaces the computationally expensive parts of the hydrodynamic models with simple and efficient data-driven approximations. Our hypothesis is that by integrating ML with physics-based numerical methods, we can achieve improved generalization performance: that is, the trained model for one case study can be used in other studies without the need for new training. We tested two ML approaches: for the first, we integrated curve fitting, and, for the second, artificial neural networks (ANN) with a finite volume scheme to solve the local inertial approximation of the SWEs. The data-driven models approximated the Momentum Equation, which explicitly solved the time derivative of flow rates. Water depths were then updated by applying a water balance equation. We also tested two different training datasets: the simulated dataset, generated from the results of hydrodynamic model, and the random dataset, generated by directly solving the momentum equation on randomly sampled input data. Various combinations of input features, for example, water slope and depth, were explored. The proposed models were trained in a small hypothetical case and tested in a different hypothetical and in two real case studies. Results showed that the curve-fitting method can be implemented successfully, given sufficient training and input data. The ANN model trained with a random dataset was substantially more accurate than that of the model trained with the simulated dataset. However, it was not successful in the real case studies. The curve-fitting method resulted in better generalization performance and increased the simulation speed of the local inertial model by 23%. Future research should test the performance of ML in terms of an increase in stable time step size and approximation of the full SWEs.  相似文献   

Environmental characterization of a karst polje: an example from Izeh polje,southwest Iran     

Hamid Reza Nassery  Farshad Alijani  Leila Mirzaei 《Environmental Earth Sciences》2009,59(1):99-108

Groundwater contamination is a well-known phenomenon, which occurs on local and regional scales in Izeh polje. The aims of this paper are investigation of the impact of human activities on the polje ecosystem, determination of the vulnerability of ground water, and to solve environmental problems. Nitrate contamination of groundwater in the Izeh polje was predicted using a solute transport model. The nitrate concentration in groundwater in most parts of Izeh polje is greater than maximum concentration permissible for drinking water, i.e., 45 mg/l. The main source of nitrate in the eastern underground areas of Izeh city is the domestic sewage. Bacterial pollution of shallow ground water in Izeh polje is severe and widespread. About 45% of ground water samples in May and September 2001 have positive MPN coliforms. Infiltration of polluted surface waters and decrease of water table depth, have lead to bacterial pollution of 80% of ground water samples in January 2002. The northeast, south and southwest areas of Izeh polje have higher pollution potential rather than its middle parts. The aquifer vulnerability indices in the middle, eastern, and northern parts of the polje are moderately lower as a result of decreased sediment size of the aquifer. The pollution in the polje depends on the amount and presence of pollutants. If they do exist, the possibility of pollution is considerable due to the coarseness of materials and shallow depth of groundwater table.  相似文献   

Locating suitable sites for the construction of subsurface dams using GIS   总被引：1，自引：1，他引：0  

Imran Ali Jamali  Bo Olofsson  Ulla Mörtberg 《Environmental Earth Sciences》2013,70(6):2511-2525

Subsurface dams constitute an affordable and effective method for the sustainable development and management of groundwater resources when constructed on suitable sites. Such dams have rarely been constructed in crystalline rock areas and to best of our knowledge, geographic information system (GIS) has never been used in any methodology for locating suitable sites for constructing these dams. This paper presents a new methodology to locate suitable sites for the construction of subsurface dams using GIS software supported by groundwater balance modelling in a study area Boda-Kalvsvik, Sweden. Groundwater resources were calculated based on digitized geological data and assumptions regarding stratigraphic layering taken from well archive data and geological maps. These estimates were then compared with future extractions for domestic water supply using a temporally dynamic water balance model. Suitability analyses for subsurface dams were based on calculated topographic wetness index (TWI) values and geological data, including stratigraphic information. Groundwater balance calculations indicated that many of the most populated areas were susceptible to frequent water supply shortages. Of the 34 sub-catchments within the study area: ten were over-extracted, nine did not have any water supply demand at all, one was self-sufficient and the remaining 14 were able to meet the water supply demand with surplus storage capacity. Six suitable sites for the construction of subsurface dams were suggested in the vicinity of the over-extracted sites based on suitability analysis and groundwater balance estimates. The new methodology shows encouraging results for regions with humid climate but having limited natural water storage capacities. The developed methodology can be used as a preliminary planning step for subsurface dam construction, establishing a base for more detailed field investigations.  相似文献   

Experimental analysis for the dynamic initiation mechanism of debris flows     

LI Chi  ZHU Wen-hui  LI Lin  LU Xiao-bing  YAO De  Farshad AMINI 《山地科学学报》2016,13(4):581-592

Debris flow is one of the major secondary mountain hazards following the earthquake. This study explores the dynamic initiation mechanism of debris flows based on the strength reduction of soils through static and dynamic triaxial tests. A series of static and dynamic triaxial tests were conducted on samples in the lab. The samples were prepared according to different grain size distribution, degree of saturation and earthquake magnitudes. The relations of dynamic shear strength, degree of saturation, and number of cycles are summarized through analyzing experimental results. The findings show that the gravelly soil with a wide and continuous gradation has a critical degree of saturation of approximately 87%, above which debris flows will be triggered by rainfall, while the debris flow will be triggered at a critical degree of saturation of about 73% under the effect of rainfall and earthquake(M>6.5). Debris flow initiation is developed in the humidification process, and the earthquake provides energy for triggering debris flows. Debris flows are more likely to be triggered at the relatively low saturation under dynamic loading than under static loading. The resistance of debris flow triggering relies more on internal frication angle than soil cohesion under the effect of rainfall and earthquake. The conclusions provide an experimental analysis method for dynamic initiation mechanism of debris flows.  相似文献   

Spatiotemporal analysis of precipitation and temperature concentration using PCI and TCI: a case study of Khuzestan Province,Iran     

Ahmadi  Farshad  Nazeri Tahroudi  Mohammad  Mirabbasi  Rasoul  Kumar  Rohitashw 《Theoretical and Applied Climatology》2022,149(1-2):743-760

Theoretical and Applied Climatology - Temperature and precipitation are the basic elements of climate, and their variation can change the water demands of different uses. In this study, the trend...  相似文献   

Evaluation of drought events in various climatic conditions using data-driven models and a reliability-based probabilistic model     

Barzkar  Ali  Najafzadeh  Mohammad  Homaei  Farshad 《Natural Hazards》2022,110(3):1931-1952

Natural Hazards - Due to a wide range of socio-economic losses caused by drought over the past decades, having a reliable insight of drought properties plays a key role in monitoring and...  相似文献   

Trends in hydrological and climatic variables affected by four variations of the Mann-Kendall approach in Urmia Lake basin,Iran     

Farshad Fathian  Zohreh Dehghan  Mohammad Hadi Bazrkar  Saeid Eslamian 《水文科学杂志》2013,58(5):892-904

ABSTRACT

The trends in hydrological and climatic time series data of Urmia Lake basin in Iran were examined using the four different versions of the Mann-Kendall (MK) approach: (i) the original MK test; (ii) the MK test considering the effect of lag-1 autocorrelation; (iii) the MK test considering the effect of all autocorrelation or sample size; and (iv) the MK test considering the Hurst coefficient. Identification of hydrological and climatic data trends was carried out at monthly and annual time scales for 25 temperature, 35 precipitation and 35 streamflow gauging stations selected from the Urmia Lake basin. Mann-Kendall and Pearson tests were also applied to explore the relationships between temperature, precipitation and streamflow trends. The results show statistically significant upward and downward trends in the annual and monthly hydrological and climatic variables. The upward trends in temperature, unlike streamflow, are much more pronounced than the downward trends, but for precipitation the behaviour of trend is different on monthly and annual time scales. Furthermore, the trend results were affected by the different approaches. Specifically, the number of stations showing trends in hydrological and climatic variables decreased significantly (up to 50%) when the fourth test was considered instead of the first and the absolute value of the Z statistic for most of the time series was reduced. The results of correlations between streamflow and climatic variables showed that the streamflow in Urmia Lake basin is more sensitive to changes in temperature than those of precipitation. The observed decreases in streamflow and increases in temperature in the Urmia Lake basin in recent decades may thus have serious implications for water resources management under the warming climate with the expected population growth and increased freshwater consumption in this region.