Predicting Standardized Streamflow index for hydrological drought using machine learning models

Shamshirband, Shahabbodin; Hashemi, Sajjad; Salimi, Hana; Samadianfard, Saeed; Asadi, Ebrahim; Shadkani, Sadra; Kargar, Katayoun; Mosavi, Amir; Nabipour, Narjes; Chau, Kwok Wing

doi:10.1080/19942060.2020.1715844

Cited by 189 publications

(80 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Standardized precipitation index (SPI) is the most usable index in drought monitoring for its simplicity [21], and its accuracy compared to other indicators [31,32]. To follow the process, cumulative distribution function was first computed by gamma distribution function, fitted into time series.…”

Section: Methodology and Study Areamentioning

confidence: 99%

Copula Based Spatial Analysis of Drought Return Period in Southwest of Iran

Kavianpour¹,

Seyedabadi²,

Moazami³

et al. 2020

Period. Polytech. Civil Eng.

View full text Add to dashboard Cite

In the past years, Khuzestan province which is located in the southwest of Iran has experienced severe droughts. Drought can be explained by its characteristics known as duration or severity. However, combination of the two features by probabilistic approach is appeared to be a well improved method to describe the phenomena. The aim of this study is to provide a more accurate statistical method of determining drought based on simultaneous analysis of two drought characteristics. Here, precipitation data from twenty stations were used to determine drought characteristics, by Standardized Precipitation Index (SPI). Joint probability function of two variables were built via copula functions. The drought return period was calculated in the form of two scenarios. The first scenario is, based on an assumption that drought is recognized by at least one of the two specific characteristics. Drought in the second scenario is distinguished by the two characteristics in a joint probabilistic form. According to research results, there was no significant difference between the north and south of Khuzestan in the study of single characteristics of drought. While analyzing two characteristics of the drought, the return period in the north was shorter than the south. The return period of droughts in the east was always shorter than in the west. The drought return period varies from 30 to 52 months and 50 to 87 months for the first and second scenarios, respectively.

show abstract

Section: Methodology and Study Areamentioning

confidence: 99%

Copula Based Spatial Analysis of Drought Return Period in Southwest of Iran

Kavianpour¹,

Seyedabadi²,

Moazami³

et al. 2020

Period. Polytech. Civil Eng.

View full text Add to dashboard Cite

show abstract

“…The machine learning is also a statistical method with data-driven and self-adaptive features. Last decades, machine learning techniques have been employed in solving prediction problems in many domains [20]including hydrological research [21], [22]. These methods, such as Support Vector Machine (SVM) [23]- [25] and Artificial Neural Network (ANN) [19], [27], [28], have been widely utilized for river flow prediction.…”

Section: Introductionmentioning

confidence: 99%

Streamflow Prediction Using Deep Learning Neural Network: Case Study of Yangtze River

Liu

Jiang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

The most important motivation for streamflow forecasts is flood prediction and longtime continuous prediction in hydrological research. As for many traditional statistical models, forecasting flood peak discharge is nearly impossible. They can only get acceptable results in normal year. On the other hand, the numerical methods including physics mechanisms and rainfall-atmospherics could provide a better performance when floods coming, but the minima prediction period of them is about one month ahead, which is too short to be used in hydrological application. In this study, a deep neural network was employed to predict the streamflow of the Hankou Hydrological Station on the Yangtze River. This method combined the Empirical Mode Decomposition (EMD) algorithm and Encoder Decoder Long Short-Term Memory (En-De-LSTM) architecture. Owing to the hydrological series prediction problem usually contains several different frequency components, which will affect the precision of the longtime prediction. The EMD technique could read and decomposes the original data into several different frequency components. It will help the model to make longtime predictions more efficiently. The LSTM based En-De-LSTM neural network could make the forecasting closer to the observed in peak flow value through reading, training, remembering the valuable information and forgetting the useless data. Monthly streamflow data (from January 1952 to December 2008) from Hankou Hydrological Station on the Yangtze River was selected to train the model, and predictions were made in two years with catastrophic flood events and ten years rolling forecast. Furthermore, the Root Mean Square Error (RMSE), Coefficient of Determination (R 2), Willmott's Index of agreement (WI) and the Legates-McCabe's Index (LMI) were used to evaluate the goodness-of-fit and performance of this model. The results showed the reliability of this method in catastrophic flood years and longtime continuous rolling forecasting. INDEX TERMS Yangtze River, hydrological time series forecasting, streamflow prediction, empirical mode decomposition, deep learning.

show abstract

“…Coastal city case studies should employ longer timeframes to find risk areas. In longer timeframes, properties such as hourly, daily, and monthly precipitation can be employed to predict flood events with accuracy through data-driven and statistical methods [55][56][57][58][59]. A case study with longer timeframes can also demonstrate how risky developments increase, and it can also predict such developments in port hinterlands.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

The Effect of Coastal City Development on Flood Damage in South Korea

Park

Paterson²,

Zigmund³

et al. 2020

Sustainability

View full text Add to dashboard Cite

Flood damage continues to be an issue in coastal cities. Impervious areas that contribute to flood damage are increasing due to the continuous development of ports in cities. However, previous research has not explored development in port hinterlands and in the coastal flood risk areas of coastal cities. Therefore, this study analyzed the impact of coastal city development on flood damage in Korea. A panel Tobit analysis was conducted on 58 coastal cities between 2002 and 2018. The results revealed that a 1% increase in impervious surfaces and one coastal development permit would increase damage costs by 1.29% and 2%, respectively. The analysis revealed that the increasing development of coastal cities had a significant impact on flood damage. The findings suggest that land-use plans highlight a conflict between port development and safety. This article provides insight that can be used by policy makers to manage risk areas near ports.

show abstract

Predicting Standardized Streamflow index for hydrological drought using machine learning models

Cited by 189 publications

References 38 publications

Copula Based Spatial Analysis of Drought Return Period in Southwest of Iran

Copula Based Spatial Analysis of Drought Return Period in Southwest of Iran

Streamflow Prediction Using Deep Learning Neural Network: Case Study of Yangtze River

The Effect of Coastal City Development on Flood Damage in South Korea

Contact Info

Product

Resources

About