Landsat:FIA-AGB v0.0.99 Map Accuracy

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

Lucas Johnson
2022-05-05

Description

Map accuracy assessment for the v0.0.99 model-weighted-ensemble Landsat model trained directly on forested FIA plots. 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-fia-0.0.99

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.

AOA masking was also implemented, and results are shown both with and without AOA masking.

NOTE: Masked AGB pixels are set to NA, and excluded from the analysis.

Masked LCMAP classes:

Agreement Statistics

Table 0: Plot:Pixel Pooled x LCPRI
LCPRI n Mean FIA %RMSE RMSE % MAE MAE MBE R\(^2\)
Cropland 32 25.38 147.52 37.44 128.52 32.62 24.77 -0.23
Grass/Shrub 34 32.35 111.56 36.09 88.50 28.63 24.54 0.31
Tree cover 1823 142.47 35.85 51.07 27.74 39.52 -2.54 0.39
Wetland 126 112.11 44.30 49.66 36.19 40.57 -2.16 0.46
Table 1: Plot:Pixel Pooled x LCPRI - AOA masked
LCPRI n Mean FIA %RMSE RMSE % MAE MAE MBE R\(^2\)
Cropland 27 24.04 144.68 34.78 125.93 30.27 21.09 0.09
Grass/Shrub 24 39.66 91.81 36.41 75.86 30.08 25.91 0.46
Tree cover 1827 141.99 36.04 51.17 28.02 39.79 -0.94 0.40
Wetland 125 113.08 44.80 50.66 36.54 41.32 -0.61 0.46
Table 2: Pooled
Dist n PPH Mean FIA %RMSE RMSE %MAE MAE MBE R\(^2\)
Plot:pixel 2015 136.85 36.96 50.58 28.71 39.29 -1.63 0.45
10 1053 1.91 137.41 31.87 43.80 24.24 33.31 -2.51 0.49
25 260 7.75 134.17 23.04 30.92 16.59 22.26 -0.48 0.58
50 79 25.51 135.29 16.29 22.04 11.52 15.59 -1.18 0.68
100 27 74.63 125.98 14.40 18.14 8.17 10.30 2.74 0.71
Table 3: Pooled - AOA Masked
Dist n PPH Mean FIA %RMSE RMSE %MAE MAE MBE R\(^2\)
Plot:pixel 2003 137.37 36.98 50.80 28.86 39.64 -0.30 0.44
10 1051 1.91 137.75 31.96 44.02 24.43 33.65 -1.17 0.49
25 260 7.7 134.54 23.34 31.40 16.83 22.64 0.68 0.56
50 79 25.35 135.60 16.91 22.92 11.73 15.91 -0.01 0.66
100 27 74.19 126.47 16.63 21.04 8.68 10.97 4.48 0.61

Continuous Metrics (Pooled, AOA Masked)

Scatter Plots (Pooled, AOA Masked)

Menlove and Healey (2019, AOA Masked)

61.44% of 236 estimates within FIA 95% CI. 7 estimates filtered (\(\geq\) 425 Mg ha\(^{-1}\)).

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, May 6). CAFRI Labs: Landsat:FIA-AGB v0.0.99 Map Accuracy. Retrieved from https://cafri-labs.github.io/acceptable-growing-stock/posts/landsatfia-agb-v0099-map-accuracy/

BibTeX citation

@misc{johnson2022landsat:fia-agb,
  author = {Johnson, Lucas},
  title = {CAFRI Labs: Landsat:FIA-AGB v0.0.99 Map Accuracy},
  url = {https://cafri-labs.github.io/acceptable-growing-stock/posts/landsatfia-agb-v0099-map-accuracy/},
  year = {2022}
}