Spatial-temporal flood inundation nowcasts by fusing machine learning methods and principal component analysis

Chang, Li-Chiu; Liou, Jia-Yi; Chang, Fi-John

doi:10.1016/j.jhydrol.2022.128086

Cited by 47 publications

(22 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, human-sensed crowdsourced data have become more available in different formats that can provide geo-located information regarding the flood status in a timely manner. For example, studies have analyzed anonymized social media content using ML and DL techniques and employed the extracted information for enhancing flood situational awareness 32,[66][67][68] . In another study example, Huang et al 69 integrated tweet data gathered by remote sensing and river water gauges to improve near real-time flood inundation maps.…”

Section: Related Workmentioning

confidence: 99%

“…Also, a growing number of researchers have used the predictive capability of various machine learning (ML) models for flood predictive monitoring [25][26][27][28][29][30] . These models can include more community features than tradition models to forecast flood status, which facilitates capturing the large number of heterogeneous community features needed for flood nowcasting 31,32 .…”

mentioning

confidence: 99%

See 1 more Smart Citation

A spatial–temporal graph deep learning model for urban flood nowcasting leveraging heterogeneous community features

Farahmand

Mostafavi

2023

Sci Rep

View full text Add to dashboard Cite

Flood nowcasting refers to near-future prediction of flood status as an extreme weather event unfolds to enhance situational awareness. The objective of this study was to adopt and test a novel structured deep-learning model for urban flood nowcasting by integrating physics-based and human-sensed features. We present a new computational modeling framework including an attention-based spatial–temporal graph convolution network (ASTGCN) model and different streams of data that are collected in real-time, preprocessed, and fed into the model to consider spatial and temporal information and dependencies that improve flood nowcasting. The novelty of the computational modeling framework is threefold: first, the model is capable of considering spatial and temporal dependencies in inundation propagation thanks to the spatial and temporal graph convolutional modules; second, it enables capturing the influence of heterogeneous temporal data streams that can signal flooding status, including physics-based features (e.g., rainfall intensity and water elevation) and human-sensed data (e.g., residents’ flood reports and fluctuations of human activity) on flood nowcasting. Third, its attention mechanism enables the model to direct its focus to the most influential features that vary dynamically and influence the flood nowcasting. We show the application of the modeling framework in the context of Harris County, Texas, as the study area and 2017 Hurricane Harvey as the flood event. Three categories of features are used for nowcasting the extent of flood inundation in different census tracts: (i) static features that capture spatial characteristics of various locations and influence their flood status similarity, (ii) physics-based dynamic features that capture changes in hydrodynamic variables, and (iii) heterogeneous human-sensed dynamic features that capture various aspects of residents’ activities that can provide information regarding flood status. Results indicate that the ASTGCN model provides superior performance for nowcasting of urban flood inundation at the census-tract level, with precision 0.808 and recall 0.891, which shows the model performs better compared with other state-of-the-art models. Moreover, ASTGCN model performance improves when heterogeneous dynamic features are added into the model that solely relies on physics-based features, which demonstrates the promise of using heterogenous human-sensed data for flood nowcasting. Given the results of the comparisons of the models, the proposed modeling framework has the potential to be more investigated when more data of historical events are available in order to develop a predictive tool to provide community responders with an enhanced prediction of the flood inundation during urban flood.

show abstract

Section: Related Workmentioning

confidence: 99%

mentioning

confidence: 99%

A spatial–temporal graph deep learning model for urban flood nowcasting leveraging heterogeneous community features

Farahmand

Mostafavi

2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Based upon time-series intercity mobility flow matrix, we use a rank reduction algorithm to identify the potential intercity mobility patterns. The regular rank reduction algorithms, such as PCA (principal component analysis) [33], ICA (independent component analysis) [34], and SVD (singular value decomposition) [12,20,21] have been widely used to extract a low number of latent components from high-dimensional data. However, traditional rank reduction algorithms can not guarantee the non-negativity of the results, even when the input initial matrix elements are all positive, leading to interpretability issues.…”

Section: Intercity Mobility Pattern Recognitionmentioning

confidence: 99%

Characterizing Intercity Mobility Patterns for the Greater Bay Area in China

Yin

2022

IJGI

View full text Add to dashboard Cite

Understanding intercity mobility patterns is important for future urban planning, in which the intensity of intercity mobility indicates the degree of urban integration development. This study investigates the intercity mobility patterns of the Greater Bay Area (GBA) in China. The proposed workflow starts by analyzing intercity mobility characteristics, proceeds to model the spatial-temporal heterogeneity of intercity mobility structures, and then identifies the intercity mobility patterns. We first conduct a complex network analysis, based on weighted degrees and the PageRank algorithm, to measure intercity mobility characteristics. Next, we calculate the Normalized Levenshtein Distance for Population Mobility Structure (NLPMS) to quantify the differences in intercity mobility structures, and we use the Non-negative Matrix Factorization (NMF) to identify intercity mobility patterns. Our results showed an evident ‘Core-Periphery’ differentiation characterized by intercity mobility, with Guangzhou and Shenzhen as the two core cities. An obvious daily intercity commuting pattern was found between Guangzhou and Foshan, and between Shenzhen and Dongguan cities at working time. This pattern, however, changes during the holidays. This is because people move from the core cities to peripheral cities at the beginning of holidays and return at the end of holidays. This study concludes that Guangzhou and Foshan have formed a relatively stable intercity mobility pattern, and the Shenzhen–Dongguan–Huizhou metropolitan area has been gradually formed.

show abstract

“…Expert judgments are usually used to determine the relative importance and weights of factors [22,23]. Consequently, the reliability of the results could be low and the model might not produce accurate results for different study areas [24,25].…”

Section: Introductionmentioning

confidence: 99%

Comparison of Machine Learning Algorithms for Flood Susceptibility Mapping

et al. 2022

View full text Add to dashboard Cite

Floods are one of the most destructive natural disasters, causing financial and human losses every year. As a result, reliable Flood Susceptibility Mapping (FSM) is required for effective flood management and reducing its harmful effects. In this study, a new machine learning model based on the Cascade Forest Model (CFM) was developed for FSM. Satellite imagery, historical reports, and field data were used to determine flood-inundated areas. The database included 21 flood-conditioning factors obtained from different sources. The performance of the proposed CFM was evaluated over two study areas, and the results were compared with those of other six machine learning methods, including Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), Deep Neural Network (DNN), Light Gradient Boosting Machine (LightGBM), Extreme Gradient Boosting (XGBoost), and Categorical Boosting (CatBoost). The result showed CFM produced the highest accuracy compared to other models over both study areas. The Overall Accuracy (AC), Kappa Coefficient (KC), and Area Under the Receiver Operating Characteristic Curve (AUC) of the proposed model were more than 95%, 0.8, 0.95, respectively. Most of these models recognized the southwestern part of the Karun basin, northern and northwestern regions of the Gorganrud basin as susceptible areas.

show abstract

Spatial-temporal flood inundation nowcasts by fusing machine learning methods and principal component analysis

Cited by 47 publications

References 51 publications

A spatial–temporal graph deep learning model for urban flood nowcasting leveraging heterogeneous community features

A spatial–temporal graph deep learning model for urban flood nowcasting leveraging heterogeneous community features

Characterizing Intercity Mobility Patterns for the Greater Bay Area in China

Comparison of Machine Learning Algorithms for Flood Susceptibility Mapping

Contact Info

Product

Resources

About