A Multi-Method Approach to Flood Mapping: Reconstructing Inundation Changes in the Cambodian Upper Mekong Delta

Datasets for the assessment of changes in the incidence, extents, and spatial patterns of inundations in the Cambodian Mekong Delta, based on a water level – flood link calculated from in-situ water levels, and Sentinel-derived inundation maps

“… [7] ) was employed, which merges statistical methods and kinematic elements to pinpoint optimal correlation factors between two time series in a set number of sub-sections of these series (also cf. [1] ). In this way, the lowest water level at which each individual pixel in the case study area becomes flooded could be identified, as well as the typical temporal delay depending on the water level.…”

“…These dates carry particular significance for the agricultural harvesting and planting cycles in the region (cf. [1] ).…”

“…2008 is known to be a hydrological break point in the region (cf. [ 1 , 9 , 10 ]). For each period, the average flood frequency for each date and the total flood duration were calculated for each pixel and a raster file generated on this basis.…”

“…Generally, the terminology of the files is self-explanatory, with the A suffix referring to the case study sub-zone A and the B suffix referring to the case study sub-zone B. Shapefiles showing their geographical delineation are provided in the sub-folder Study_Area_Delineation. Extensive descriptions of their terrain and land cover characteristics are given in [1] . These shape files were delineated manually and encompass a total area of approximately 500 km2.…”

“…This is the result of linking water levels measured at the hydrological station at Koh Khel with inundation extents extracted via k-means clustering from Sentinel-1 and −2 imagery using the cloud computing platform Google Earth Engine and a correlation model. Details on the methodology are provided in [1] .…”