Research Article

Assessment of Sentinel-2 and Landsat-8 OLI for Small-Scale Inland Water Quality Modeling and Monitoring Based on Handheld Hyperspectral Ground Truthing

Table 4

Evaluation of the performance of various atmospheric correction methods for Sentinel-2 image processing.

Atmospheric correctionStatistical indexB1B2B3B4B5B6B7B8

DOSRMSE0.0290.0330.0290.0470.0450.0520.0560.055
NRMSE0.2310.1740.1870.1140.1980.1470.1830.177
MAE0.0230.0250.0210.0370.0340.0410.0450.043
Bias-0.022-0.023-0.016-0.036-0.034-0.04-0.044-0.04
ATCORRMSE0.0430.0450.0520.0610.0580.0610.0630.065
NRMSE0.3430.2370.3350.1480.2550.1720.2060.209
MAE0.0380.0330.0420.0490.0460.0440.0450.046
Bias-0.037-0.027-0.041-0.049-0.044-0.038-0.038-0.04
DSFRMSE0.0240.0380.0480.0590.060.0630.0650.065
NRMSE0.1910.2000.3100.1430.2640.1780.2120.209
MAE0.0190.0290.0410.0520.0530.0570.060.059
Bias-0.013-0.028-0.041-0.051-0.053-0.057-0.059-0.06
EXPRMSE0.0330.0410.0420.0510.0520.0590.0660.061
NRMSE0.2630.2160.2710.1240.2290.1670.2160.196
MAE0.0480.0360.0320.0370.0410.0380.0480.049
Bias-0.047-0.027-0.031-0.037-0.041-0.038-0.038-0.04
L2ARMSE0.0510.0550.0480.0680.610.0590.0760.072
NRMSE0.4060.2900.3100.1652.6840.1670.2480.232
Bias0.0490.0490.0530.0590.0570.0690.0710.66
MAE-0.44-0.057-0.049-0.056-0.04-0.063-0.067-0.61