Search

In this study, a new hybridized machine learning algorithm for urban flood susceptibility mapping, named MultiB-MLPNN, was developed using a multi-boosting technique and MLPNN. The model was tested in Amol City, Iran, a data-scarce city in an ungauged area which is prone to severe flood inundation events and currently lacks flood prevention infrastructure. Performance of the hybridized model was compared with that of a standalone MLPNN model, random forest and boosted regression trees. Area under the curve, efficiency, true skill statistic, Matthews correlation coefficient, misclassification rate, sensitivity and specificity were used to evaluate model performance. In validation, the MultiB-MLPNN model showed the best predictive performance. The hybridized MultiB-MLPNN model is thus useful for generating realistic flood susceptibility maps for data-scarce urban areas. The maps can be used to develop risk-reduction measures to protect urban areas from devastating floods, particularly where available data are insufficient to support physically based hydrological or hydraulic models.

A total maximum daily load (TMDL) is required for water bodies in the U.S. that do not meet applicable water quality standards. Computational watershed models are often used to develop TMDL pollutant reduction scenarios. Uncertainty is inherent in the modeling process. An explicit uncertainty analysis would improve model performance and result in more robust decision making when comparing alternative pollutant reduction scenarios. This paper presents a risk-based framework for evaluating alternative pollutant allocation scenarios considering reliability in achieving water quality goals. We demonstrate a generic routine for the application of Generalized Likelihood Uncertainty Estimation (GLUE) to Hydrological Simulation Program-FORTRAN (HSPF) using existing softwares to evaluate two bacteria reduction scenarios from a recently developed TMDL that addressed a bacterial impairment in a mixed land use watershed in Virginia, U.S. Our probabilistic analysis showed that for reliability levels <25%, the recommended TMDL bacterial load reduction scenario had the same exceedance rate as the full reduction scenario (fully reducing all bacterial loads except wildlife), while for reliability levels between 25% and 50%, the exceedance rates for the two pollutant reduction scenarios were similar, with the TMDL recommended scenario violating the water quality criteria only slightly more often. The full reduction scenario performed better in higher reliability levels, although it could not meet the water quality criteria. Our results indicated that, in this case, achieving water quality goals with very high reliability was not possible, even with extreme levels of pollutant reduction. The risk-based framework presented here illustrates a method to propagate watershed model uncertainty and assess performance of alternative pollutant reduction scenarios using existing tools, thereby enabling decision makers to understand the reliability of a given scenario in achieving water quality

A total maximum daily load (TMDL) is required for water bodies in the U.S. that do not meet applicable water quality standards. Computational watershed models are often used to develop TMDL pollutant reduction scenarios. Uncertainty is inherent in the modeling process. An explicit uncertainty analysis would improve model performance and result in more robust decision making when comparing alternative pollutant reduction scenarios. This paper presents a risk-based framework for evaluating alternative pollutant allocation scenarios considering reliability in achieving water quality goals. We demonstrate a generic routine for the application of Generalized Likelihood Uncertainty Estimation (GLUE) to Hydrological Simulation Program-FORTRAN (HSPF) using existing softwares to evaluate two bacteria reduction scenarios from a recently developed TMDL that addressed a bacterial impairment in a mixed land use watershed in Virginia, U.S. Our probabilistic analysis showed that for reliability levels <25%, the recommended TMDL bacterial load reduction scenario had the same exceedance rate as the full reduction scenario (fully reducing all bacterial loads except wildlife), while for reliability levels between 25% and 50%, the exceedance rates for the two pollutant reduction scenarios were similar, with the TMDL recommended scenario violating the water quality criteria only slightly more often. The full reduction scenario performed better in higher reliability levels, although it could not meet the water quality criteria. Our results indicated that, in this case, achieving water quality goals with very high reliability was not possible, even with extreme levels of pollutant reduction. The risk-based framework presented here illustrates a method to propagate watershed model uncertainty and assess performance of alternative pollutant reduction scenarios using existing tools, thereby enabling decision makers to understand the reliability of a given scenario in achieving water quality

Floods are some of the most destructive and catastrophic disasters worldwide. Development of management plans needs a deep understanding of the likelihood and magnitude of future flood events. The purpose of this research was to estimate flash flood susceptibility in the Tafresh watershed, Iran, using five machine learning methods, i.e., alternating decision tree (ADT), functional tree (FT), kernel logistic regression (KLR), multilayer perceptron (MLP), and quadratic discriminant analysis (QDA). A geospatial database including 320 historical flood events was constructed and eight geo-environmental variables—elevation, slope, slope aspect, distance from rivers, average annual rainfall, land use, soil type, and lithology—were used as flood influencing factors. Based on a variety of performance metrics, it is revealed that the ADT method was dominant over the other methods. The FT method was ranked as the second-best method, followed by the KLR, MLP, and QDA. Given a few differences between the goodness-of-fit and prediction success of the methods, we concluded that all these five machine-learning-based models are applicable for flood susceptibility mapping in other areas to protect societies from devastating floods.

Floods are some of the most destructive and catastrophic disasters worldwide. Development of management plans needs a deep understanding of the likelihood and magnitude of future flood events. The purpose of this research was to estimate flash flood susceptibility in the Tafresh watershed, Iran, using five machine learning methods, i.e., alternating decision tree (ADT), functional tree (FT), kernel logistic regression (KLR), multilayer perceptron (MLP), and quadratic discriminant analysis (QDA). A geospatial database including 320 historical flood events was constructed and eight geo-environmental variables—elevation, slope, slope aspect, distance from rivers, average annual rainfall, land use, soil type, and lithology—were used as flood influencing factors. Based on a variety of performance metrics, it is revealed that the ADT method was dominant over the other methods. The FT method was ranked as the second-best method, followed by the KLR, MLP, and QDA. Given a few differences between the goodness-of-fit and prediction success of the methods, we concluded that all these five machine-learning-based models are applicable for flood susceptibility mapping in other areas to protect societies from devastating floods.

A two-phase Monte Carlo simulation (TPMCS) uncertainty analysis framework is used to analyze epistemic and aleatory uncertainty associated with simulated exceedances of an in-stream fecal coliform (FC) water quality criterion when using the Hydrological Simulation Program-FORTRAN (HSPF). The TPMCS framework is compared with a single-phase or standard Monte Carlo simulation (SPMCS) analysis. Both techniques are used to assess two total maximum daily load (TMDL) pollutant allocation scenarios. The application of TPMCS illustrates that cattle directly depositing FC in the stream is a greater source of epistemic uncertainty than FC loading from cropland overland runoff, the two sources specifically targeted for reduction in the allocation scenario. This distinction is not possible using SPMCS. Although applying the TPMCS framework involves subjective decisions about how selected model parameters are considered within the framework, this uncertainty analysis approach is transparent and the results provide information that can be used by decision makers when considering pollution control measure implementation alternatives, including quantifying the level of confidence in achieving applicable water quality standards. © American Society of Civil Engineers (ASCE).

This study presents a probabilistic framework that considers both the water quality improvement capability and reliability of alternative total maximum daily load (TMDL) pollutant allocations. Generalized likelihood uncertainty estimation and Markov chain Monte Carlo techniques were used to assess the relative uncertainty and reliability of two alternative TMDL pollutant allocations that were developed to address a fecal coliform (FC) bacteria impairment in a rural watershed in western Virginia. The allocation alternatives, developed using the Hydrological Simulation Program-FORTRAN, specified differing levels of FC bacteria reduction from different sources. While both allocations met the applicable water-quality criteria, the approved TMDL allocation called for less reduction in the FC source that produced the greatest uncertainty (cattle directly depositing feces in the stream), suggesting that it would be less reliable than the alternative, which called for a greater reduction from that same source. The approach presented in this paper illustrates a method to incorporate uncertainty assessment into TMDL development, thereby enabling stakeholders to engage in more informed decision making.

A two-phase Monte Carlo simulation (TPMCS) uncertainty analysis framework is used to analyze epistemic and aleatory uncertainty associated with simulated exceedances of an in-stream fecal coliform (FC) water quality criterion when using the Hydrological Simulation Program-FORTRAN (HSPF). The TPMCS framework is compared with a single-phase or standard Monte Carlo simulation (SPMCS) analysis. Both techniques are used to assess two total maximum daily load (TMDL) pollutant allocation scenarios. The application of TPMCS illustrates that cattle directly depositing FC in the stream is a greater source of epistemic uncertainty than FC loading from cropland overland runoff, the two sources specifically targeted for reduction in the allocation scenario. This distinction is not possible using SPMCS. Although applying the TPMCS framework involves subjective decisions about how selected model parameters are considered within the framework, this uncertainty analysis approach is transparent and the results provide information that can be used by decision makers when considering pollution control measure implementation alternatives, including quantifying the level of confidence in achieving applicable water quality standards. © American Society of Civil Engineers (ASCE).

This study presents a probabilistic framework that considers both the water quality improvement capability and reliability of alternative total maximum daily load (TMDL) pollutant allocations. Generalized likelihood uncertainty estimation and Markov chain Monte Carlo techniques were used to assess the relative uncertainty and reliability of two alternative TMDL pollutant allocations that were developed to address a fecal coliform (FC) bacteria impairment in a rural watershed in western Virginia. The allocation alternatives, developed using the Hydrological Simulation Program-FORTRAN, specified differing levels of FC bacteria reduction from different sources. While both allocations met the applicable water-quality criteria, the approved TMDL allocation called for less reduction in the FC source that produced the greatest uncertainty (cattle directly depositing feces in the stream), suggesting that it would be less reliable than the alternative, which called for a greater reduction from that same source. The approach presented in this paper illustrates a method to incorporate uncertainty assessment into TMDL development, thereby enabling stakeholders to engage in more informed decision making.