LiDAR pulses are absorbed by water if you could get aerial coverage during flood. Cloud coverage may limit. High density aerial infrared photos may be helpful if enough canopy openings. DEMs created using LIDAR can also better define ground surface and Vegetation detail, both helpful in mapping soils and flood features. Small streams and rivers may have cloud cover during severe storms, while larger rivers may have flood peaks multiple hours to days later, offering greater opportunity for remote sensing. To some degree, soil maps developed by soil scientists can be also helpful in identifying extent of flood prone soils as they use soil, vegetation and hydrologic indicators in mapping soils using both remote sensing with field verification sampling. Knowing stream type also can be helpful, as gullies and other entrenched streams have no or very limited floodplains, some have terraces (former floodplains). Braided streams can have floodplains covering much of valley as excessive sedimentation limits channel capacity necessitating development of excessive flooding and eventually added flood channels and instability.

Hello Dinuke Munasinghe ,

Not sure what type of optical imagery you intend to use, but if you're talking about multi-spectral satellite imagery such as Sentinel-2 or Landsat-8, you could have a look at the approach presented here:

Article Introducing WIW for Detecting the Presence of Water in Wetla...

(https://sentinel.esa.int/web/sentinel/news/-/article/copernicus-sentinel-2-helps-track-changes-in-seasonal-water-of-wetlands)

Using optical satellite imagery obviously has its limitations when it comes to 'detecting something under closed canopy' and the quality of the results will largely depend on the type of vegetation, the duration of flooding and the persistence of clouds...

Best,

Sebastian

William F. Hansen Thanks for the ideas. I didn't think of the idea of the flood-prone soils! Thanks for that! I'll check some of the global soil databases.

@Sebastian: Thank you for the read! I'll check it out. I work with multispectral imagery.