CAFRI Labs: Landsat Ensemble AGB 1.1.0: More Data = More Better

Lucas Johnson

Evaluation Results

Using version of FIA models 1.1.1 More Data x LiDAR Zeroes

Using version of LiDAR models 1.1.0 More Data

Notes

Ensembling five component models: three Landsat:FIA based, two Landsat:LiDAR based
All models were used to generate predictions against data clipped to FIA plots:
- FIA models were run through 10-fold CV, and predictions were made against the hold-out fold
- LiDAR models, which have never “seen” FIA plots before except indirectly through the LiDAR-AGB surface, predicted the FIA training data directly without any CV
RMSE weighting and LINMOD fitting were performed as normal against these predictions
This test set maps directly to the Landsat-FIA model test set

Change Summary

This version adds 65 indices and their respective 1 year deltas (130 predictors) computed through awesome spectral indices library.

	FIA-based			LiDAR-based		Ensemble
	RF	GBM	SVM	RF	GBM	RMSE	LINMOD
RMSE	52.121	53.165	52.132	55.214	54.561	51.057	50.652
RMSE %	41.243	42.069	41.252	43.690	43.173	40.401	40.080
MBE	1.243	1.992	-1.460	7.782	6.006	3.039	1.667
R2	0.522	0.503	0.521	0.472	0.481	0.545	0.547

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

Corrections

If you see mistakes or want to suggest changes, please create an issue on the source repository.

Citation

For attribution, please cite this work as

Johnson (2022, Sept. 15). CAFRI Labs: Landsat Ensemble AGB 1.1.0: More Data = More Better. Retrieved from https://cafri-labs.github.io/acceptable-growing-stock/posts/landsat-ensemble-agb-110-more-data-more-better/

BibTeX citation

@misc{johnson2022landsat,
  author = {Johnson, Lucas},
  title = {CAFRI Labs: Landsat Ensemble AGB 1.1.0: More Data = More Better},
  url = {https://cafri-labs.github.io/acceptable-growing-stock/posts/landsat-ensemble-agb-110-more-data-more-better/},
  year = {2022}
}