Landsat:FIA v0.0.5 Map Accuracy - Holdout Plots Only

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

Lucas Johnson
2021-07-20

Description

Map accuracy assessment for Landsat models trained directly on FIA plot data for the years 2013-2019.

Model Document

Riemann et. al.

Branch used to produce this document: lj_landsat-fia-0.0.5-holdout

FIA plot inclusion criteria

Masking

No masking applied.

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 137 NA 6.42 4.59 20.83 11.70 0.32 0.88 0.00 0.76 0.09
2 pooled 334 NA 9.34 3.39 22.65 12.52 0.29 0.78 0.00 0.58 -0.05
3 pooled 17 NA 33.89 6.72 40.69 29.67 0.34 0.35 0.00 0.66 0.21
4 pooled 715 NA 130.42 0.84 56.24 44.42 0.35 0.19 0.00 0.38 -0.05
5 pooled 25 NA 9.99 3.07 11.18 7.30 0.83 0.84 0.75 0.91 0.85
6 pooled 85 NA 72.52 4.34 53.85 42.91 0.47 0.29 0.02 0.70 0.32
8 pooled 3 NA 0.00 9.17 11.88 9.17 NA 1.00 0.05 NA NA
Table 2: Plot:Pixel Comparison
Group n PPH Mean FIA MBE RMSE MAE R2 KS AC ACs ACu
target_2014 179 NA 87.65 -0.31 51.15 35.42 0.63 0.31 0.42 0.91 0.51
target_2015 195 NA 76.56 5.30 49.27 35.31 0.63 0.33 0.50 0.92 0.58
target_2016 201 NA 81.73 2.72 41.39 28.49 0.72 0.32 0.64 0.97 0.68
target_2017 166 NA 75.28 3.32 47.91 32.15 0.71 0.37 0.58 0.91 0.67
target_2018 190 NA 71.00 3.87 43.36 30.19 0.68 0.39 0.60 0.96 0.64
target_2019 168 NA 87.65 -4.56 42.51 29.42 0.78 0.35 0.69 0.92 0.77
pooled 1316 NA 79.21 2.24 45.88 31.85 0.68 0.34 0.56 0.94 0.63
Table 3: 8660 Ha Hex
Group n PPH Mean FIA MBE RMSE MAE R2 KS AC ACs ACu
target_2014 178 1.01 87.09 0.05 51.00 35.26 0.63 0.31 0.42 0.91 0.51
target_2015 187 1.04 75.73 6.39 49.28 35.28 0.63 0.33 0.52 0.92 0.60
target_2016 192 1.05 81.51 2.50 41.25 28.25 0.71 0.32 0.64 0.97 0.67
target_2017 163 1.02 75.14 3.36 47.70 31.79 0.71 0.37 0.58 0.91 0.67
target_2018 185 1.03 71.01 3.13 42.77 29.99 0.69 0.39 0.61 0.96 0.65
target_2019 163 1.03 87.55 -4.97 42.99 29.69 0.78 0.34 0.68 0.91 0.77
pooled 820 1.34 79.25 2.51 41.94 29.46 0.70 0.29 0.60 0.94 0.65
Table 4: 78100 Ha Hex
Group n PPH Mean FIA MBE RMSE MAE R2 KS AC ACs ACu
target_2014 120 1.49 81.70 1.24 45.83 29.87 0.65 0.28 0.49 0.94 0.55
target_2015 121 1.61 73.38 8.96 43.12 32.45 0.60 0.26 0.56 0.96 0.60
target_2016 126 1.60 76.59 3.06 35.72 23.74 0.73 0.27 0.69 0.98 0.71
target_2017 112 1.48 70.99 5.27 40.35 28.37 0.71 0.29 0.64 0.94 0.69
target_2018 134 1.42 71.38 4.21 41.45 28.97 0.67 0.33 0.59 0.96 0.63
target_2019 118 1.42 85.86 -3.90 38.50 26.78 0.79 0.28 0.68 0.91 0.77
pooled 196 5.61 75.94 2.29 23.84 17.96 0.80 0.10 0.77 0.97 0.79
Table 5: 216500 Ha Hex
Group n PPH Mean FIA MBE RMSE MAE R2 KS AC ACs ACu
target_2014 67 2.67 79.90 1.08 28.11 21.43 0.81 0.16 0.76 0.97 0.79
target_2015 63 3.10 69.75 9.49 33.11 26.34 0.63 0.24 0.59 0.93 0.66
target_2016 68 2.96 74.36 5.27 23.94 17.75 0.81 0.16 0.80 0.98 0.82
target_2017 65 2.55 73.37 5.00 34.67 24.47 0.74 0.18 0.66 0.94 0.73
target_2018 66 2.88 70.03 4.34 28.01 22.12 0.77 0.18 0.73 0.96 0.77
target_2019 65 2.58 87.11 -3.87 30.38 21.46 0.83 0.14 0.74 0.91 0.83
pooled 79 13.91 74.68 4.84 19.66 13.56 0.80 0.11 0.78 0.97 0.82

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 2014 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 (2021, July 29). CAFRI Labs: Landsat:FIA v0.0.5 Map Accuracy - Holdout Plots Only. Retrieved from https://cafri-labs.github.io/acceptable-growing-stock/posts/landsatfia-v005-map-accuracy-holdout-plots-only/

BibTeX citation

@misc{johnson2021landsat:fia,
  author = {Johnson, Lucas},
  title = {CAFRI Labs: Landsat:FIA v0.0.5 Map Accuracy - Holdout Plots Only},
  url = {https://cafri-labs.github.io/acceptable-growing-stock/posts/landsatfia-v005-map-accuracy-holdout-plots-only/},
  year = {2021}
}