Abstract
A water body detection technique is an essential part of digital elevation model (DEM) generation to delineate land–water boundaries and to set flattened elevations. The initial tile-based water body data that are created during production of the Advanced Spaceborne Thermal Emission and Reflection radiometer (ASTER) GDEM, as a by-product, are incorporated into ASTER GDEM V3 to improve the quality. At the same time as ASTER GDEM V3, the Global Water Body Data Base (ASTWBD) Version 1 is also released to the public. The ASTWBD generation consists of two parts: separation from land area, and classification into three categories: sea, lake, and river. Sea water bodies have zero elevation. Lake water bodies have flattened elevations. River water bodies have a gradual step-down from upstream to downstream with a step of one meter. The separation process from land area is carried out automatically using an algorithm, except for sea-ice removal, to delineate the real seashore lines in the high latitude areas; almost all of the water bodies are created through this process. The classification process into three categories, i.e., sea, river, and lake, is carried out, and incorporated into ASTER GDEM V3. For inland water bodies, it is not possible to perfectly detect all water bodies using reflectance and spectral index, which are the only available parameters for optical sensors. The only way available to identify the undetected inland water bodies is to manually copy them with visual inspection from the earth’s surface images, like Landsat images. GeoCover2000 images are the main part of the object images. Color–Land ASTER MosaicS (CLAMS) images are used to cover the deficiency of the GeoCover2000 images. This kind of time-consuming, unsophisticated way is inevitable as it is a manual-based method to improve the quality of the ASTWBD. This paper describes the manual-based improvement method; specifically, how deficient water body images are efficiently copied as rasterized images from the earth’s surface images to obtain a more complete global water body data set.
Highlights
The Advanced Spaceborne Thermal Emission and Reflection radiometer (ASTER) is an advanced multispectral imaging sensor that was launched on board the Terra spacecraft in December, 1999 [1,2]
Water body detection is an essential part of digital elevation model (DEM) generation, because image matching is not possible for water bodies
Color–Land ASTER MosaicS (CLAMS) [8] images are used to cover the deficiency of the GeoCover images
Summary
The Advanced Spaceborne Thermal Emission and Reflection radiometer (ASTER) is an advanced multispectral imaging sensor that was launched on board the Terra spacecraft in December, 1999 [1,2]. For inland water bodies like lakes and rivers, it is not possible to perfectly detect all water bodies using reflectance and spectral index, which are the only available parameters for optical sensors like ASTER. The only way available to identify the missed inland water bodies is to manually copy them with time-consuming visual inspection from the earth’s surface images, like Landsat images. The original GeoCover data set covers the earth with a 14.25 m spatial resolution and UTM coordinate. The original CLAMS data set covers the earth every 4◦ latitude by 4◦ longitude with 0.5 arcsecond posting Both data sets were converted to the same spatial resolution and coordinates as the ASTER GWBD, i.e., geographic latitude/longitude coordinates with 1 arcsecond posting, and a 1◦ latitude by 1◦ longitude tile size. The support tool “ROI” was used to copy the missed inland water bodies on either object image. The support tool “Masking” was used to import the copied images to the water body image tile, and the improved image tile was saved as a GeoTIFF file
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
