Landsat:LiDAR-AGB v0.0.8 Map Accuracy

Map accuracy/agreement assessment following the Riemann et. al. framework.

Lucas Johnson
2022-04-06

Description

Map accuracy assessment for the v0.0.8 model-weighted-ensemble Landsat model trained on AOA-masked LiDAR-AGB surfaces. FIA data inventoried between 2013 and 2019 were considered for this assessment.

Model Document

Riemann et. al.

Branch used to produce this document: lj_landsat-lidar-0.0.8

FIA plot inclusion criteria

Masking

LCMAP Collection 1.2 primary classification layers were used to mask out water, developed, and barren areas i.e. “!class %in% c(1, 5, 8)” pixels from the modeled AGB surfaces. Annual AGB surfaces were matched to the LMCAP layer from the same year.

NOTE: Masked AGB pixels are set to 0, and included in the agreement assessment just as any other AGB pixel.

Masked classes:

Agreement Statistics (SI corrected)

Table 1: Plot:Pixel Pooled x LCPRI
LCPRI Group n PPH Mean FIA MBE RMSE MAE R2 KS AC ACs ACu
1 pooled 426 NA 7.15 -7.15 26.03 7.15 0.00 0.15 0.00 NA NA
2 pooled 1139 NA 9.62 22.27 32.03 27.62 0.29 0.81 0.27 0.57 0.70
3 pooled 94 NA 30.76 24.09 39.89 34.13 0.42 0.57 0.44 0.72 0.71
4 pooled 2374 NA 130.61 3.92 57.55 45.45 0.32 0.17 0.00 0.55 0.02
5 pooled 122 NA 9.50 -9.50 28.54 9.50 0.00 0.15 0.00 NA NA
6 pooled 254 NA 79.95 1.33 58.43 47.24 0.41 0.30 0.00 0.27 0.12
8 pooled 15 NA 7.17 -7.17 18.32 7.17 0.00 0.27 0.00 NA NA
Table 2: Plot:Pixel Comparison
Group n PPH Mean FIA MBE RMSE MAE R2 KS AC ACs ACu
target_2013 680 NA 76.71 10.57 48.11 35.98 0.63 0.22 0.57 0.94 0.64
target_2014 661 NA 79.69 8.42 50.39 36.90 0.63 0.23 0.54 0.92 0.62
target_2015 651 NA 75.06 9.99 47.74 35.56 0.66 0.26 0.60 0.93 0.67
target_2016 627 NA 80.54 6.05 47.52 35.28 0.66 0.22 0.56 0.93 0.63
target_2017 608 NA 81.51 4.96 51.77 37.86 0.65 0.25 0.50 0.89 0.61
target_2018 601 NA 77.51 7.93 47.66 35.18 0.65 0.25 0.57 0.93 0.64
target_2019 596 NA 80.83 3.60 46.61 34.55 0.70 0.26 0.56 0.90 0.66
pooled 4424 NA 78.78 7.45 48.58 35.91 0.65 0.24 0.56 0.92 0.63
Table 3: 8660 Ha Hex
Group n PPH Mean FIA MBE RMSE MAE R2 KS AC ACs ACu
target_2013 645 1.05 76.51 10.68 48.35 36.05 0.62 0.21 0.56 0.94 0.63
target_2014 617 1.07 78.24 9.00 48.72 35.96 0.64 0.23 0.56 0.93 0.63
target_2015 574 1.13 74.87 10.74 46.30 34.38 0.66 0.25 0.61 0.94 0.68
target_2016 564 1.11 79.54 6.39 46.82 34.61 0.66 0.22 0.56 0.93 0.63
target_2017 568 1.07 82.36 4.64 51.26 37.19 0.65 0.24 0.50 0.89 0.61
target_2018 551 1.09 77.10 8.28 46.90 34.75 0.65 0.25 0.58 0.94 0.65
target_2019 521 1.14 81.11 2.94 45.56 33.65 0.70 0.25 0.55 0.90 0.65
pooled 1502 2.95 77.25 8.00 33.64 25.44 0.71 0.14 0.67 0.95 0.72
Table 4: 78100 Ha Hex
Group n PPH Mean FIA MBE RMSE MAE R2 KS AC ACs ACu
target_2013 190 3.58 71.43 10.80 30.07 22.66 0.71 0.15 0.72 0.96 0.76
target_2014 184 3.59 75.98 10.17 30.67 23.75 0.74 0.15 0.72 0.95 0.77
target_2015 184 3.54 70.38 11.66 30.16 23.05 0.71 0.17 0.70 0.93 0.77
target_2016 180 3.48 79.06 4.67 33.23 24.16 0.71 0.12 0.62 0.93 0.69
target_2017 182 3.34 79.84 3.37 34.80 26.19 0.71 0.14 0.61 0.93 0.68
target_2018 186 3.23 76.60 7.61 34.83 25.79 0.64 0.15 0.58 0.93 0.64
target_2019 179 3.33 80.22 5.26 33.19 23.32 0.70 0.14 0.63 0.94 0.69
pooled 205 21.58 72.36 9.52 21.21 14.89 0.80 0.13 0.80 0.95 0.85
Table 5: 216500 Ha Hex
Group n PPH Mean FIA MBE RMSE MAE R2 KS AC ACs ACu
target_2013 80 8.50 67.69 9.70 23.18 15.71 0.78 0.15 0.79 0.96 0.82
target_2014 75 8.81 74.61 8.24 24.24 18.58 0.75 0.15 0.73 0.94 0.79
target_2015 77 8.45 69.17 9.79 25.00 17.54 0.67 0.17 0.69 0.95 0.74
target_2016 77 8.14 75.22 6.93 20.01 14.73 0.85 0.12 0.84 0.97 0.87
target_2017 74 8.22 78.20 5.08 20.53 16.15 0.83 0.11 0.81 0.97 0.84
target_2018 77 7.81 70.36 10.95 24.84 18.17 0.73 0.17 0.74 0.94 0.79
target_2019 76 7.84 77.96 6.18 28.28 19.92 0.68 0.12 0.64 0.95 0.68
pooled 85 52.05 70.32 9.56 19.15 13.50 0.81 0.20 0.81 0.95 0.86

ECDF Comparisons Across Scales - FIA Plots vs Mapped (SI corrected)

1:1 and GMFR Lines Across Scales - FIA Plots vs Mapped (SI corrected)

Spatial/Distribution Patterns of Local Differences (SI Corrected)

Target 2013 Distribution

Target 2016 Distribution

Target 2019 Distribution

Pooled Distribution

Spatial Differences

Patterns of Local Variability (SI Corrected)

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, April 6). CAFRI Labs: Landsat:LiDAR-AGB v0.0.8 Map Accuracy. Retrieved from https://cafri-labs.github.io/acceptable-growing-stock/posts/landsatlidar-agb-v008-map-accuracy/

BibTeX citation

@misc{johnson2022landsat:lidar-agb,
  author = {Johnson, Lucas},
  title = {CAFRI Labs: Landsat:LiDAR-AGB v0.0.8 Map Accuracy},
  url = {https://cafri-labs.github.io/acceptable-growing-stock/posts/landsatlidar-agb-v008-map-accuracy/},
  year = {2022}
}