Title of article :
Evaluation of various satellite sensors for waterline extraction in a coral reef environment: Majuro Atoll, Marshall Islands
Author/Authors :
Yamano، نويسنده , , Hiroya and Shimazaki، نويسنده , , Hiroto and Matsunaga، نويسنده , , Tsuneo and Ishoda، نويسنده , , Albon and McClennen، نويسنده , , Caleb and Yokoki، نويسنده , , Hiromune and Fujita، نويسنده , , Kazuhiko and Osawa، نويسنده , , Yoko and Kayanne، نويسنده , , Hajime، نويسنده ,
Issue Information :
روزنامه با شماره پیاپی سال 2006
Pages :
14
From page :
398
To page :
411
Abstract :
The ability of five satellite sensor bands (IKONOS band 4, Terra ASTER bands 3 and 4, and Landsat ETM+ bands 4 and 5) was examined to extract the waterline at coral reef coasts (Majuro Atoll, Marshall Islands) using different wavelength regions (near infrared [NIR] and shortwave infrared [SWIR]) and different spatial resolutions (4, 15, and 30 m). After performing georeferencing and normalization of the images, density slicing was used to extract the waterline. Comparisons of extracted waterline positions with ground-level data for eight transects and global positioning system (GPS) tracks of the island shorelines showed that NIR bands were superior to SWIR bands because of the characteristics of the coral reef coasts, including a lack of foam and suspended sediments (which can affect the NIR wavelength region, if present) and the presence of remnant water on reef flats during low tide (which can affect the SWIR wavelength region). A linear relationship was found between the estimation errors of waterline positions and the spatial resolutions of the NIR sensors. Analysis on estimation errors and image costs showed that Terra ASTER band 3 was the most cost-effective sensor for extracting waterlines with reasonable accuracy. The results serve as general guidelines for using satellite-derived data to estimate intertidal topography and detect and monitor shorelines in coral reef environments.
Keywords :
Coral reef , Remote sensing , shoreline , IKONOS , Terra ASTER , Landsat ETM+
Journal title :
Geomorphology
Serial Year :
2006
Journal title :
Geomorphology
Record number :
2359087
Link To Document :
بازگشت