Landsat FIA AGB 1.1.1: LiDAR Zeros

Evaluation Results

Last version of FIA models: 2022-04-11 More Data = More Better

Change Summary

• This version adds 176 completely non-forested FIA plots, with maximum LiDAR heights < 1m, and an inferred AGB value of 0 Mg/ha

	RF (ranger)	GBM (LightGBM)	SVM (kernlab)	Ensemble (model weighted)	Ensemble (RMSE weighted)
%RMSE	41.231	41.825	41.252	40.523	40.835
RMSE	52.106	52.857	52.132	51.211	51.606
MAE	39.401	40.277	39.373	38.884	39.194
MBE	1.252	0.854	-1.460	0.671	0.205
R2	0.523	0.506	0.521	0.537	0.532

Against the previous (more data) test set (without LiDAR zeroes)

	RF (ranger)	GBM (LightGBM)	SVM (kernlab)	Ensemble (model weighted)	Ensemble (RMSE weighted)
%RMSE	39.516	40.085	39.517	38.860	39.140
RMSE	54.285	55.066	54.286	53.383	53.769
MAE	42.497	43.422	42.387	42.070	42.235
MBE	1.028	0.567	-2.000	0.728	-0.146
R2	0.370	0.346	0.366	0.386	0.381

AGB Distribution

summary(bind_rows(training, testing)$agb_mgha)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
   0.00   75.17  126.45  125.90  174.95  425.00 

Metadata

Ensembles

• RMSE-weighted model weights:

• Linear model weights:

Call:
lm(formula = agb_mgha ~ rf_pred * lgb_pred * svm_pred, data = k_fold_preds)

Residuals:
    Min      1Q  Median      3Q     Max 
-143.41  -33.96   -1.57   27.86  211.21 

Coefficients:
                            Estimate Std. Error t value Pr(>|t|)   
(Intercept)               -1.951e+00  5.355e+00  -0.364  0.71566   
rf_pred                    3.187e-01  2.903e-01   1.098  0.27245   
lgb_pred                  -1.012e-01  3.105e-01  -0.326  0.74447   
svm_pred                   1.897e-01  2.238e-01   0.848  0.39678   
rf_pred:lgb_pred           1.080e-03  1.692e-03   0.638  0.52340   
rf_pred:svm_pred           1.906e-03  2.305e-03   0.827  0.40837   
lgb_pred:svm_pred          4.986e-03  2.576e-03   1.936  0.05307 . 
rf_pred:lgb_pred:svm_pred -2.463e-05  8.718e-06  -2.825  0.00479 **
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 50.89 on 1532 degrees of freedom
Multiple R-squared:  0.5438,    Adjusted R-squared:  0.5417 
F-statistic: 260.9 on 7 and 1532 DF,  p-value: < 2.2e-16

Coverages

• 0 coverages:

\(n\) and \(p\)

• 2204 observations
  • 1542 training
  • 662 testing
• 162 predictors
  • aweinsh, delta_aweinsh, aweish, delta_aweish, bcc, delta_bcc, dvi, delta_dvi, embi, delta_embi, exgr, delta_exgr, exr, delta_exr, fcvi, delta_fcvi, gari, delta_gari, gbndvi, delta_gbndvi, gemi, delta_gemi, gli, delta_gli, gndvi, delta_gndvi, gosavi, delta_gosavi, grndvi, delta_grndvi, gvmi, delta_gvmi, ikaw, delta_ikaw, ipvi, delta_ipvi, lswi, delta_lswi, mag, mbi, delta_mbi, mcari2, delta_mcari2, mlswi26, delta_mlswi26, mlswi27, delta_mlswi27, mndwi, delta_mndwi, mrbvi, delta_mrbvi, msavi, delta_msavi, msi, delta_msi, msr, delta_msr, mtvi2, delta_mtvi2, muwir, delta_muwir, nbr, delta_nbr, nbr2, delta_nbr2, nbsims, delta_nbsims, nddi, delta_nddi, ndsi, delta_ndsi, ndvi, delta_ndvi, ndvimndwi, delta_ndvimndwi, ndwi, delta_ndwi, ngrdi, delta_ngrdi, nirv, delta_nirv, nli, delta_nli, nmdi, delta_nmdi, normg, delta_normg, normnir, delta_normnir, normr, delta_normr, nrfig, delta_nrfig, nsds, delta_nsds, nwi, delta_nwi, pisi, delta_pisi, rcc, delta_rcc, rdvi, delta_rdvi, rgbvi, delta_rgbvi, rgri, delta_rgri, s3, delta_s3, si, delta_si, sr, delta_sr, sr2, delta_sr2, swi, delta_swi, tcb, delta_tcb, tcg, delta_tcg, tcw, delta_tcw, tdvi, delta_tdvi, tgi, delta_tgi, trivi, delta_trivi, tvi, delta_tvi, ui, delta_ui, vgnirbi, delta_vgnirbi, vig, delta_vig, vrnirbi, delta_vrnirbi, wi1, delta_wi1, wi2, delta_wi2, yod, precip, tmin, tmax, dem, twi, slope, aspect, chm, lcpri4, lcpri6, lcpri3, lcpri2, lcpri5, lcsec3, lcsec2, lcsec1, lcsec4, lcsec8

Component Models

• Tuning used 5-fold CV
• Final hyperparameters:

Random forest:

$num.trees
[1] 1500

$mtry
[1] 45

$min.node.size
[1] 4

$sample.fraction
[1] 0.75

$replace
[1] FALSE

$formula
agb_mgha ~ .

LGB:

$nrounds
[1] 50

$params
$params$learning_rate
[1] 0.1

$params$num_leaves
[1] 12

$params$max_depth
[1] 3

$params$extra_trees
[1] TRUE

$params$min_data_in_leaf
[1] 6

$params$bagging_fraction
[1] 0.9

$params$bagging_freq
[1] 5

$params$feature_fraction
[1] 1

$params$min_data_in_bin
[1] 14

$params$lambda_l1
[1] 0.5

$params$lambda_l2
[1] 1

$params$force_col_wise
[1] TRUE

SVM:

$x
agb_mgha ~ .

$kernel
[1] "laplacedot"

$type
[1] "eps-svr"

$kpar
$kpar$sigma
[1] 0.0078125


$C
[1] 12

$epsilon
[1] 0.125

$nu
[1] 0.2