Landsat-8 OLI image pixels for developing water mapping algorithms
Description
This dataset contains 153140 Landsat-8 OLI pixels. The pixels were collected from 180 subsite images (900 by 900 m) in 22 Landsat-8 OLI imageries that covering various water-land environments around the world. Each pixel is represented as a row of data with the attributes described here. DATE - The date of the Landsat 8 OLI image acquisition as YYYYMMDD. PATH/ROW - The Worldwide Reference System path and row of the Landsat image. SR - Spatial reference read from ArcGIS 10.5 CORX - x-coordinate of pixel center in the SR system CORY - y-coordinate of pixel center in the SR system WT.PCT - An integer value of the percentage area of the pixel that is covered by water. PIX.TYPE - Defined a pixel is water (1) or non-water (0) CLASS - The dominant land type contained in pixel. WT = water, VL = vegetated land, BL = built-up land, and SL = shadowed land BAND1, 2, 3, 4, 5, 6, 7 - Band surface reflectance. NDWI, MNDWI, AWEI0, AWEI1, WI2015 - The five water indices calculated with the algorithms listed below. NDWI = (ρb3-ρb5) / (ρb3+ρb5) MNDWI = (ρb3-ρb6) / (ρb3+ρb6) AWEI0 = 4(ρb3-ρb6)-0.25ρb5-2.75ρb7 AWEI1 = ρb2+2.5ρb3-1.5(ρb5+ρb6)-0.25ρb7 WI2015 = 1.7204+171ρb3+3ρb4-70ρb5-45ρb6-71ρb7 where ρ is surface reflectance and b1, b2, …, b7 are band numbers of Landsat-8 OLI images.
Files
Steps to reproduce
See the artical "Wen et al., 2021, Ensembles of multiple spectral water indices for improving surface water classification, International Journal of Applied Earth Observations and Geoinformation" for further information.