Predicting the provisioning potential of forest ecosystem services using airborne laser scanning data and forest resource maps

Forest Ecosystems

Table 2 The features and performance of ALS-based models for predicting ratio-scaled ES proxy values. W – Shapiro-Wilk test statistic

ES	Predictor	W^a	RMSE	RMSE_LOOCV
BIOD	cc_shrub × h40_first	0.965^***	0.259	0.266
	+ prop_{first/FP_ground} × d50_first	0.981	0.235	NA
	+ diff_only–LP × hstd_LP	0.986	0.217	0.226
	+ h95_first × h10_LP	0.977^*	0.203	NA
TIMB	cc_{only_ground} × hmean_FP	0.977^*	0.220	0.228
	+ h40_last × cc_{LP_ground}	0.970^**	0.202	0.213
	+ h05_last × d05_first	0.989	0.182	NA
	+ cc_{only_ground} × h10_FP	0.988	0.174	NA
CARB	cc_shrub × h60_first	0.980	0.158	0.163
	+ h20_last × cc_{only_ground}	0.988	0.144	0.152
	+ d60_first × d30_LP	0.987	0.138	0.148
	+ prop_{first/FP_ground} × d05_first	0.984	0.132	NA
BILB	h60_first × h70_LP	0.987	0.279	0.286
	+ d50_only × cc_{FP_ground}	0.984	0.255	0.274
	+ r_understory × d05_FP	0.982	0.238	NA
	+ d70_first × d50_only	0.984	0.222	0.267
COWB	d20_LP × h40_FP	0.966^***	0.281	0.295
	+ diff_only–FP ²	0.981	0.250	NA
	+ prop_{first/FP_ground} × d05_first	0.983	0.233	NA
	+ d60_first × h05_only	0.971^**	0.217	NA
AMEN	h10_first × hmean_FP	0.993	0.239	0.246
	+ d30_last × d70_first	0.991	0.219	0.229
	+ h20_last × cc_{FP_ground}	0.994	0.204	NA
	+ hstd_first × hmean_LP	0.991	0.187	NA

^aThe asterisks refer to the significance of the test statistic at the 90% (*), 95% (**), and 99% (***) confidence level