Skip to main content

Table 4 Goodness-of-fit and validation statistics for two binomial generalized additive models to estimate percent canopy cover based on crown radius-DBH models (\( {\hat{CC}}_{\mathrm{VERT}} \)) and measurements using a spherical densiometer (\( {\hat{CC}}_{\mathrm{HEMI}} \))

From: Model-based estimation of above-ground biomass in the miombo ecoregion of Zambia

  Fit statistics Validation statistics
Model RMSE (6) MD (8) RMSPE (4) MPD (10)
\( {\hat{CC}}_{\mathrm{VERT}} = -1.8747+f\left(\mathrm{NDVI}\right)+f\left(\mathrm{A}\mathrm{RVI}.\mathrm{texture}\right) + f\left(\mathrm{SWIR}1.\mathrm{texture}\right) + f\left(\mathrm{D}2\mathrm{cap}\right)+f\left(\mathrm{A}\mathrm{W}\mathrm{C}\right)+f\left(\mathrm{slope}\right) \) 8.5 % 0.0 % 8.4 % 1.5 %
\( {\hat{CC}}_{\mathrm{HEMI}}=0.6788+\mathrm{Fire}.\mathrm{late}+\mathrm{D}2\mathrm{cap}+\mathrm{carbon}+f\left(\mathrm{bulk}\right)+f\left(\mathrm{c}\mathrm{r}\mathrm{s}.\mathrm{f}\mathrm{r}\mathrm{g}\right) \) 13.4 % −0.7 % 17.2 % −0.5 %
  1. RMSE root mean square error, MD mean difference, RMSPE root mean square prediction error, MPD mean prediction difference
  2. See Tables 2 and 3 for descriptions of predictor variables