Use of Airborne Hyperspectral Imagery to Map Soil Properties in Tilled Agricultural Fields
Table 2
Partial least squares (PLS) prediction model goodness of fit () associated with each of 15 math treatments, for the 13 analytes that predicted with , calculated using data from all 315 soil sampling locations1. The first derivative gap two (1Dg2, depicted in bold) was selected as the overall most successful model although it was occasionally outperformed (italic). See Section 2 for analyte descriptions.
Derivative
Gap
C
Sand
Silt
Clay
pH
OM
K
Ca
Mg
Mn
Zn
Fe
Al
average
NON
0
0.578
0.762
0.763
0.585
0.442
0.685
0.555
0.669
0.692
0.642
0.670
0.754
0.777
0.659
1ST
1
0.555
0.770
0.761
0.565
0.517
0.706
0.522
0.654
0.719
0.624
0.606
0.707
0.799
0.654
1ST
2
0.591
0.763
0.763
0.617
0.549
0.717
0.514
0.676
0.708
0.638
0.647
0.740
0.782
0.670
1ST
4
0.595
0.754
0.740
0.596
0.451
0.692
0.513
0.630
0.616
0.565
0.675
0.703
0.773
0.639
1ST
8
0.584
0.748
0.739
0.600
0.413
0.668
0.578
0.635
0.642
0.580
0.636
0.637
0.727
0.630
1ST
16
0.588
0.740
0.744
0.550
0.442
0.672
0.547
0.610
0.619
0.567
0.591
0.666
0.725
0.620
1ST
32
0.525
0.730
0.722
0.554
0.381
0.640
0.520
0.648
0.619
0.542
0.580
0.669
0.765
0.607
1ST
64
0.338
0.623
0.619
0.470
0.317
0.537
0.413
0.519
0.580
0.488
0.568
0.636
0.669
0.521
2ND
1
0.542
0.585
0.597
0.424
0.361
0.526
0.484
0.708
0.705
0.529
0.599
0.665
0.729
0.573
2ND
2
0.507
0.681
0.682
0.372
0.499
0.566
0.493
0.665
0.664
0.566
0.591
0.673
0.769
0.595
2ND
4
0.576
0.714
0.724
0.409
0.449
0.660
0.484
0.672
0.669
0.606
0.613
0.749
0.752
0.621
2ND
8
0.536
0.698
0.689
0.533
0.472
0.602
0.528
0.595
0.631
0.542
0.638
0.743
0.753
0.612
2ND
16
0.580
0.729
0.725
0.573
0.487
0.661
0.498
0.590
0.640
0.598
0.617
0.680
0.712
0.622
2ND
32
0.564
0.715
0.698
0.609
0.432
0.631
0.548
0.622
0.592
0.548
0.606
0.639
0.688
0.607
2ND
64
0.525
0.596
0.589
0.421
0.156
0.599
0.511
0.556
0.464
0.501
0.538
0.602
0.614
0.513
1Analytes that predicted poorly (<0.5) included: (<0.303), (<.355), Cu (<0.358), B (<0.169), S (<0.282), and P saturation (<0.127).