Table 5: Calibration and validation accuracies for approach B—partial least squares regression. The models further to be applied for a large-scale prediction of soil constituents based on HyMap imagery are highlighted.

Spectral datasetPretreatmentNo. of factors Calibration (941/1232 samples) Validation (311/402 samples)
RMSCalRPDCal RMSValRPDVal

ASD spectral resolution(1) In situ field1st derivative60.790.452.160.530.521.30
(2) Bare soil fieldMSC + 1st derivative60.820.412.380.650.421.61
(3) LaboratoryMean center + 1st derivative60.820.402.360.690.45 1.53
(1) In situ fieldMean center70.660.761.740.390.841.14
Iron oxides(2) Bare soil fieldMean center80.690.731.800.430.801.20
(3) LaboratoryLog(1/R) + mean center90.820.642.390.430.98 1.14
(1) In situ fieldMean center70.234.341.150.024.630.97
Clay(2) Bare soil fieldLog(1/R)50.114.671.060.064.421.02
(3) LaboratoryMean center80.334.311.220.084.431.02

HyMap spectral resolution(1) In situ fieldLog(1/R) + mean center50.770.462.110.590.471.43
(2) Bare soil fieldLog(1/R) + 1st derivative50.820.412.380.620.451.51
(3) LaboratoryLog(1/R) 90.790.432.200.790.32 2.11
(1) In situ fieldMean center70.670.751.750.430.761.26
Iron oxides(2) Bare soil fieldMean center70.660.761.730.450.771.25
(3) LaboratoryLog(1/R) + mean center90.810.672.280.490.88 1.27
(1) In situ fieldMean center40.144.581.090.015.060.89
Clay(2) Bare soil fieldMean center50.154.561.090.034.471.01
(3) Laboratory1st derivative80.374.161.270.104.431.02

1Number of samples in training and test sets for field datasets. 2Number of samples in training and test sets for laboratory datasets.