Table 6: Detailed partial least squares regression model results for soil total carbon ( ) prediction from the subsets of visible/near-infrared diffuse reflectance spectra based on spectral classification with k-means cluster analysis. The range of values reflects the results of 10 random iterations of the models, and the number in parentheses is the mean. Detailed results are also given for full sample set models with no subsetting for comparison. For models with full cross validation (i.e., leave-one-out cross validation), the same samples used to calibrate the model were used to validate the model.


Cluster 0780.934.52Full cross validation0.885.872.865.40

Cluster 1870.68–0.881.92–3.26 370.60–0.911.74–3.471.54–3.331.94–5.50
(0.77)(2.86) (0.75) (2.89) (2.16) (3.14)

Cluster 2730.54–0.960.65–2.22 320.62–0.910.98–1.721.67–3.340.79–2.56
(0.81)(1.29) (0.80) (1.33) (2.39)

Full sample set2150.83–0.962.82–5.84 920.74–0.953.10–5.831.89–4.541.80–3.92
(0.90)(4.30) (0.88) (4.30) (3.28) (3.06)

Full sample set3070.953.09Full cross validation0.943.394.093.80

aNumber of samples.
bCoefficient of determination.
cRoot mean squared error.
dResidual prediction deviation.
eRatio of performance to interquartile distance.