The utility of fused airborne laser scanning and multispectral data for improved wind damage risk assessment over a managed forest landscape in Finland

Table 7 Effect sizes of alternate LR models, obtained by successive elimination of least significant variables

	Estimate	Std. error	z value	Pr(> \|z\|)
Four variables:
Intercept	-4.970	0.244	-20.370	< 0.0001
isSpruce	0.321	0.328	0.978	0.328
domHtStand	0.388	0.147	2.638	0.008
distToEdge_NW	-0.309	0.199	-1.552	0.121
gapize_NW	-0.294	0.213	-1.381	0.167
Three:
Intercept	-4.822	0.181	-26.667	< 0.0001
domHtStand	0.429	0.142	3.012	0.003
distToEdge_NW	-0.323	0.199	-1.622	0.105
gapize_NW	-0.292	0.213	-1.375	0.169
Two:
Intercept	-4.790	0.176	-27.193	< 0.0001
domHtStand	0.411	0.142	2.892	0.004
distToEdge_NW	-0.311	0.201	-1.548	0.122
One:
Intercept	-4.750	0.169	-28.055	< 0.0001
domHtStand	0.440	0.142	3.099	0.002

Here, the northwest wind direction model from Table 6 is taken, and further analysed/simplified
Values that are significant at the p = 0.05 level are highlighted (boldface)

ISSN: 1297-966X