skip to main content

Explore Research Products in the NSF-PAR

Title: Data-Driven Flood Alert System (FAS) Using Extreme Gradient Boosting (XGBoost) to Forecast Flood Stages
Heavy rainfall leads to severe flooding problems with catastrophic socio-economic impacts worldwide. Hydrologic forecasting models have been applied to provide alerts of extreme flood events and reduce damage, yet they are still subject to many uncertainties due to the complexity of hydrologic processes and errors in forecasted timing and intensity of the floods. This study demonstrates the efficacy of using eXtreme Gradient Boosting (XGBoost) as a state-of-the-art machine learning (ML) model to forecast gauge stage levels at a 5-min interval with various look-out time windows. A flood alert system (FAS) built upon the XGBoost models is evaluated by two historical flooding events for a flood-prone watershed in Houston, Texas. The predicted stage values from the FAS are compared with observed values with demonstrating good performance by statistical metrics (RMSE and KGE). This study further compares the performance from two scenarios with different input data settings of the FAS: (1) using the data from the gauges within the study area only and (2) including the data from additional gauges outside of the study area. The results suggest that models that use the gauge information within the study area only (Scenario 1) are sufficient and advantageous in terms of their accuracy in predicting the arrival times of the floods. One of the benefits of the FAS outlined in this study is that the XGBoost-based FAS can run in a continuous mode to automatically detect floods without requiring an external starting trigger to switch on as usually required by the conventional event-based FAS systems. This paper illustrates a data-driven FAS framework as a prototype that stakeholders can utilize solely based on their gauging information for local flood warning and mitigation practices.  more » « less
Award ID(s):
1832065 1940163
NSF-PAR ID:
10347490
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
Water
Volume:
14
Issue:
5
ISSN:
2073-4441
Page Range / eLocation ID:
747
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ( , Hydrology)
    null (Ed.)
    Flooding during extreme weather events damages critical infrastructure, property, and threatens lives. Hurricane María devastated Puerto Rico (PR) on 20 September 2017. Sixty-four deaths were directly attributable to the flooding. This paper describes the development of a hydrologic model using the Gridded Surface Subsurface Hydrologic Analysis (GSSHA), capable of simulating flood depth and extent for the Añasco coastal flood plain in Western PR. The purpose of the study was to develop a numerical model to simulate flooding from extreme weather events and to evaluate the impacts on critical infrastructure and communities; Hurricane María is used as a case study. GSSHA was calibrated for Irma, a Category 3 hurricane, which struck the northeastern corner of the island on 7 September 2017, two weeks before Hurricane María. The upper Añasco watershed was calibrated using United States Geological Survey (USGS) stream discharge data. The model was validated using a storm of similar magnitude on 11–13 December 2007. Owing to the damage sustained by PR’s WSR-88D weather radar during Hurricane María, rainfall was estimated in this study using the Weather Research Forecast (WRF) model. Flooding in the coastal floodplain during Hurricane María was simulated using three methods: (1) Use of observed discharge hydrograph from the upper watershed as an inflow boundary condition for the coastal floodplain area, along with the WRF rainfall in the coastal flood plain; (2) Use of WRF rainfall to simulate runoff in the upper watershed and coastal flood plain; and (3) Similar to approach (2), except the use of bias-corrected WRF rainfall. Flooding results were compared with forty-two values of flood depth obtained during face-to-face interviews with residents of the affected communities. Impacts on critical infrastructure (water, electric, and public schools) were evaluated, assuming any structure exposed to 20 cm or more of flooding would sustain damage. Calibration equations were also used to improve flood depth estimates. Our model included the influence of storm surge, which we found to have a minimal effect on flood depths within the study area. Water infrastructure was more severely impacted by flooding than electrical infrastructure. From these findings, we conclude that the model developed in this study can be used with sufficient accuracy to identify infrastructure affected by future flooding events. 
    more » « less
  2. ; ; ; ( , Journal of Hydrometeorology)
    Abstract

    Flash flooding in the arid/semiarid southwestern United States is frequently associated with convective rainfall during the North American monsoon. In this study, we examine flood-producing storms in central Arizona based on analyses of dense rain gauge observations and stream gauging records as well as North American Regional Reanalysis fields. Our storm catalog consists of 102 storm events during the period of 1988–2014. Synoptic conditions for flood-producing storms are characterized based on principal component analyses. Four dominant synoptic modes are identified, with the first two modes explaining approximately 50% of the variance of the 500-hPa geopotential height. The transitional synoptic pattern from the North American monsoon regime to midlatitude systems is a critical large-scale feature for extreme rainfall and flooding in central Arizona. Contrasting spatial rainfall organizations and storm environment under the four synoptic modes highlights the role of interactions among synoptic conditions, mesoscale processes, and complex terrains in determining space–time variability of convective activities and flash flood hazards in central Arizona. We characterize structure and evolution properties of flood-producing storms based on storm tracking algorithms and 3D radar reflectivity. Fast-moving storm elements can be important ingredients for flash floods in the arid/semiarid southwestern United States. Contrasting storm properties for cloudburst storms highlight the wide spectrum of convective intensities for extreme rain rates in the arid/semiarid southwestern United States and exhibit comparable vertical structures to their counterparts in the eastern United States.

     
    more » « less
  3. ; ; ( , Earth Interactions)
    Abstract

    The Mississippi River basin drains nearly one-half of the contiguous United States, and its rivers serve as economic corridors that facilitate trade and transportation. Flooding remains a perennial hazard on the major tributaries of the Mississippi River basin, and reducing the economic and humanitarian consequences of these events depends on improving their seasonal predictability. Here, we use climate reanalysis and river gauge data to document the evolution of floods on the Missouri and Ohio Rivers—the two largest tributaries of the Mississippi River—and how they are influenced by major modes of climate variability centered in the Pacific and Atlantic Oceans. We show that the largest floods on these tributaries are preceded by the advection and convergence of moisture from the Gulf of Mexico following distinct atmospheric mechanisms, where Missouri River floods are associated with heavy spring and summer precipitation events delivered by the Great Plains low-level jet, whereas Ohio River floods are associated with frontal precipitation events in winter when the North Atlantic subtropical high is anomalously strong. Further, we demonstrate that the El Niño–Southern Oscillation can serve as a precursor for floods on these rivers by mediating antecedent soil moisture, with Missouri River floods often preceded by a warm eastern tropical Pacific (El Niño) and Ohio River floods often preceded by a cool eastern tropical Pacific (La Niña) in the months leading up peak discharge. We also use recent floods in 2019 and 2021 to demonstrate how linking flood hazard to sea surface temperature anomalies holds potential to improve seasonal predictability of hydrologic extremes on these rivers.

     
    more » « less
  4. ; ; ; ; ; ( , Earth Science Informatics)
    Abstract

    Unprecedented floods from extreme rainfall events worldwide emphasize the need for flood inundation mapping for floodplain management and risk reduction. Access to flood inundation maps and risk evaluation tools remains challenging in most parts of the world, particularly in rural regions, leading to decreased flood resilience. The use of hydraulic and hydrodynamic models in rural areas has been hindered by excessive data and computational requirements. In this study, we mapped the flood inundation in Huron Creek watershed, Michigan, USA for an extreme rainfall event (1000-year return period) that occurred in 2018 (Father’s Day Flood) using the Height Above Nearest Drainage (HAND) model and a synthetic rating curve developed from LIDAR DEM. We compared the flood inundation extent and depth modeled by the HAND with flood inundation characteristics predicted by two hydrodynamic models, viz., HEC-RAS 2D and SMS-SRH 2D. The flood discharge of the event was simulated using the HEC-HMS hydrologic model. Results suggest that, in different channel segments, the HAND model produces different degrees of concurrence in both flood inundation extent and depth when compared to the hydrodynamic models. The differences in flood inundation characteristics produced by the HAND model are primarily due to the uncertainties associated with optimal parameter estimation of the synthetic rating curve. Analyzing the differences between the HAND and hydrodynamic models also highlights the significance of terrain characteristics in model predictions. Based on the comparable predictive capability of the HAND model to map flood inundation areas during extreme rainfall events, we demonstrate the suitability of the HAND-based approach for mitigating flood risk in data-scarce, rural regions.

     
    more » « less
  5. ; ; ( , Scientific Reports)
    Abstract

    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.

     
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email