Imputing forest carbon stock estimates from inventory plots to a nationally continuous coverage

Table 1 Validation metrics by scale and forest C stock

Metric	Scale	Total	Above-ground	Below-ground	Understory	Standing dead	Down dead	Forest Floor	Soil organic
Agreement Coefficient^a	200 km	0.9911	0.9942	0.9944	0.9909	0.9399	0.9932	0.9947	0.9828
	100 km	0.9943	0.9926	0.9925	0.9866	0.9379	0.9924	0.9943	0.9933
	50 km	0.9866	0.9785	0.9785	0.9697	0.8635	0.9844	0.9850	0.9850
	25 km	0.9610	0.9295	0.9300	0.9220	0.6707	0.9534	0.9605	0.9639
KS Statistic^b	200 km	0.0364	0.0424	0.0485	0.0364	0.0485	0.0364	0.0364	0.0424
	100 km	0.0274	0.0316	0.0316	0.0247	0.0521	0.0288	0.0233	0.0288
	50 km	0.0866	0.0882	0.0882	0.0863	0.1135	0.0866	0.0863	0.0863
	25 km	0.1246	0.1260	0.1260	0.1244	0.1591	0.1244	0.1244	0.1244
RMA Slope^c	200 km	1.0073	1.0039	1.0050	0.9941	1.1515	1.0122	1.0003	1.0089
	100 km	1.0012	1.0025	1.0027	0.9995	1.0466	1.0050	1.0059	0.9987
	50 km	1.0102	1.0171	1.0174	1.0140	1.1405	1.0151	1.0114	1.0073
	25 km	1.0248	1.0405	1.0411	1.0403	1.2158	1.0319	1.0271	1.0206

^a Agreement Coefficient (larger values indicate better agreement, min = 0, max = 1).
^b Kolmogorov–Smirnov Statistic (smaller values indicate better agreement, min = 0, max = unbounded).
^c Reduced Major Axis Slope (values closer to 1 indicate better agreement, min, max = unbounded).

ISSN: 1750-0680