Satellite Flood Inundation Assessment and Forecast Using SMAP and Landsat

Abstract

The capability and synergistic use of multi-source satellite observations for flood monitoring and forecasts is crucial for improving disaster preparedness and mitigation. Here, surface fractional water cover (FW) retrievals derived from Soil Moisture Active Passive (SMAP) L-band (1.4 GHz) brightness temperatures were used for flood assessment over southeast Africa during the Cyclone Idai event. We then focused on five sub-catchments of the Pungwe basin and developed a machine learning-based approach with the support of Google Earth Engine (GEE) for daily (24-hour) forecasting of FW and 30-m inundation downscaling and mapping. The Classification and Regression Trees (CART) model was selected and trained using retrievals derived from SMAP and Landsat coupled with rainfall forecasts from the NOAA Global Forecast System (GFS). Independent validation showed that FW predictions over randomly selected dates are highly correlated (R=0.87) with the Landsat observations. The forecast results captured the flood temporal dynamics from the Idai event; and the associated 30-m downscaling results showed inundation spatial patterns consistent with independent satellite Synthetic Aperture Radar (SAR) observations. The data-driven approach provides new capacity for flood monitoring and forecasts leveraging synergistic satellite observations and big data analysis, which is particularly valuable for data sparse regions.

Remote Sensing Methods for Flood Prediction: A Review

“…This pie graph shows that the use of machine learning and hydrological models has been equally and most frequently observed in the literature for flood forecasting. A significant issue in remote sensing technologies is the constraint regarding orbital cycles and spaces between trajectories of satellites, which makes the continuous monitoring of fixed objects a difficult task [31,71].…”

“…Floods, being natural disasters, are uncertain and unpredictable, which makes flood modelling a complex task having numerous uncertainties. Hence, hydrological and nu- A significant issue in remote sensing technologies is the constraint regarding orbital cycles and spaces between trajectories of satellites, which makes the continuous monitoring of fixed objects a difficult task [31,71].…”

“…What are the current gaps in remote sensing-based flood prediction technologies? A significant issue in remote sensing technologies is the constraint regarding orbital cycles and spaces between trajectories of satellites, which makes the continuous monitoring of fixed objects a difficult task[31,71].Floods, being natural disasters, are uncertain and unpredictable, which makes flood modelling a complex task having numerous uncertainties. Hence, hydrological and numerical models seldom provide imprecise results and fail to give reliable predictions regarding floods.…”

Spatial modeling of flood hazard using machine learning and GIS in Ha Tinh province, Vietnam

“…Several different techniques highlighted in previous studies have been used to detect flood locations. For example, we can derive indices like Enhanced Vegetation Index (EVI), NDWI, and Normalized Difference Surface Water Index (NDSWI) to detect flood locations using optical remote sensing data such as Landsat imagery (Tong et al 2018;Du et al 2021). However, these data may be influenced by cloud cover; they also suffer from limited spatial and temporal resolutions.…”

Increased flooded area and exposure in the White Volta river basin in Western Africa, identified from multi-source remote sensing data

“…Global-scale flood datasets derived from satellite data are available, such as the MODIS NRT Global Flood Product 17 , the MODIS Global Flood Database 18 , and the Global Flood Detection System 19 . These datasets again have coarse spatial resolution and validation of them is highly challenging, particularly in data-sparse regions such as SSA 14 where accurate data on flood extent at the local scale is greatly needed for risk management 20 , 21 .…”