Towards a global seasonal and permanent reference water product from Sentinel-1/2 data for improved flood mapping

Martinis, Sandro; Groth, Sandro; Wieland, Marc; Knopp, Lisa; Rättich, Michaela

doi:10.1016/j.rse.2022.113077

Cited by 37 publications

(15 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep neural networks, specifically CNNs, are the most widely used [36]. In this sense, the more training data the CNNs have, the better results they will obtain [37].…”

Section: Related Workmentioning

confidence: 99%

Segmentation and Visualization of Flooded Areas Through Sentinel-1 Images and U-Net

Pech-May,

Aquino-Santos,

Álvarez-Cárdenas

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Floods are the most common phenomenon and cause the most significant economic and social damage to the population. They are becoming more frequent and dangerous. Consequently, it is necessary to create strategies to intervene effectively in the mitigation and resilience of the affected areas. Different methods and techniques have been developed to mitigate the damage caused by this phenomenon. Satellite programs provide a large amount of data on the Earth's surface, and geospatial information processing tools help manage different natural disasters. Likewise, deep learning is an approach capable of forecasting time series that can be applied to satellite images for flood prediction and mapping. This paper presents an approach for flood segmentation and visualization using the U-Net architecture and Sentinel-1 SAR satellite imagery. The U-Net architecture can capture relevant features in SAR images. The approach comprises various phases, from data loading and preprocessing to flood inference and visualization. For the study, the georeferenced dataset Sen1Floods11 is used to train and validate the model through different epochs and training. A study area in southeastern Mexico that presents frequent floods was chosen. The results demonstrate that the segmentation model achieves high accuracy in detecting flooded areas, with promising metrics regarding loss, precision, and F1-score.

show abstract

“…Deep neural networks, specifically CNNs, are the most widely used [36]. In this sense, the more training data the CNNs have, the better results they will obtain [37].…”

Section: Related Workmentioning

confidence: 99%

Segmentation and Visualization of Flooded Areas Through Sentinel-1 Images and U-Net

Pech-May,

Aquino-Santos,

Álvarez-Cárdenas

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Copernicus Sentinel-2 data are systematically processed to L1C products and made available online between 2 and 12 h from sensing (on average, seven-hour after sensing) in the Copernicus Open Access Hub and Copernicus Services Data Hub. The effectiveness of the Sentinel-2 missions, operated by the European Space Agency (ESA) in the frame of the European Union's Copernicus program, in drought and flood monitoring have been shown within the last years in many studies [28,[31][32][33][34].…”

Section: Monitoring Flood Areas and Nature-based Solutions Using Spac...mentioning

confidence: 99%

Evaluating Nature-Based Solution for Flood Reduction in Spercheios River Basin Part 2: Early Experimental Evidence

et al. 2022

View full text Add to dashboard Cite

A number of Nature Based Solutions (NBS) are being used around the world in order to address various hydrometeorological hazards as a more environmentally friendly alternative to hard structures. Such a solution has been created in the Spercheios river basin in Central Greece, which is susceptible to heavy rainfall and river bank overflow due to flood water from upstream, in order to mitigate flood and drought impacts under current and future climate conditions. Here a first attempt is made to use actual measurements taken from various sources in the river, including in-situ and satellite data, in order to establish early experimental evidence of the NBS efficiency in the area. The measurements include data from automated hydrological stations from the OpenHi network, satellite remote sensing data and field measurements performed along the Spercheios River basin. For each measurement used, different analysis has been performed based on data availability and pertinence to the NBS efficiency. Preliminary results presented here show that the NBS functions as designed and provides protection against flooding in the area, and can potentially diminish the risk of drought. The results are in agreement with the numerical outputs already presented in our previous work.

show abstract

“…They found that 16-23 % of Fortunately, geospatial data for identifying aquatic systems, including wetlands, are burgeoning (Khare et al, in review). Global land cover and land use geospatial datasets that include a wetland cover class continue to propagate (Hu et al, 2017a), taking advantage of both lengthy time-series Landsat data (Homer et al, 2020) as well as recently launched advanced high-resolution and/or synthetic aperture radar (SAR) equipped satellites (e.g., Sentinel-1, Sentinel-2, plus many commercially available platforms; Martinis et al, 2022) and topographic data sources and analyses (e.g., Wu et al, 2019b). Examples include the GlobeLand30 (Chen et al, 2015), the European Space Agency (ESA) WorldCover 2020 (ESA, 2020), the Dynamic World (Brown et al, 2022), as well as consortiums focusing on annual land cover change mapping (e.g., Tsendbazar et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

“…performance when used in combination with vegetation-based assessments or spectral analyses identifying water (Devries et al, 2017;Evenson et al, 2018b). Similarly, emerging synthetic aperture radar-based landscape classifications (e.g., Huang et al, 2018;Martinis et al, 2022;Brown et al, 2022) and both airborne and satellite-borne hyperspectral and advanced analyses, including LiDAR, as well as analytical capabilities (e.g., machine-learning approaches, object-oriented classifications, Berhane et al, 2018); topographically based models, Xi et al, 2022) hold great promise for improved resolution and performance in identifying non-floodplain wetlands (Christensen et al, 2022).…”

mentioning

confidence: 99%

Mapping Global Non-Floodplain Wetlands

Lane

D’Amico

Christensen

et al. 2023

Preprint

View full text Add to dashboard Cite

Abstract. Non-floodplain wetlands – those located outside the floodplains – have emerged as integral components to watershed resilience, contributing hydrologic and biogeochemical functions affecting watershed-scale flooding extent, drought magnitude, and water-quality maintenance. However, the absence of a global dataset of non-floodplain wetlands limits their necessary incorporation into water quality and quantity management decisions and affects wetland-focused wildlife habitat conservation outcomes. We addressed this critical need by developing a publicly available Global NFW (non-floodplain wetland) dataset, comprised of a global river-floodplain map at 90 m resolution coupled with a global ensemble wetland map incorporating multiple wetland-focused data layers. The floodplain, wetland, and non-floodplain wetland spatial data developed here were successfully validated within 21 large and heterogenous basins across the conterminous United States. We identified nearly 33 million potential non-floodplain wetlands with an estimated global extent of over 16 million km2. Non-floodplain wetland pixels comprised 53 % of globally identified wetland pixels, meaning the majority of the globe’s wetlands likely occur external to river floodplains and coastal habitats. The identified Global NFWs were typically small (median 0.039 km2), with a global median size ranging from 0.018–0.138 km2. This novel geospatial Global NFW dataset advances wetland conservation and resource-management goals while providing a foundation for global non-floodplain wetland functional assessments, facilitating non-floodplain wetland inclusion in hydrological, biogeochemical, and biological model development. The data are freely available through the United States Environmental Protection Agency’s Environmental Dataset Gateway (https://gaftp.epa.gov/EPADataCommons/ORD/Global_NonFloodplain_Wetlands/) and through https://doi.org/10.23719/1528331 (Lane et al., 2023).

show abstract

Towards a global seasonal and permanent reference water product from Sentinel-1/2 data for improved flood mapping

Cited by 37 publications

References 33 publications

Segmentation and Visualization of Flooded Areas Through Sentinel-1 Images and U-Net

Segmentation and Visualization of Flooded Areas Through Sentinel-1 Images and U-Net

Evaluating Nature-Based Solution for Flood Reduction in Spercheios River Basin Part 2: Early Experimental Evidence

Mapping Global Non-Floodplain Wetlands

Contact Info

Product

Resources

About