Table 2. Sample sizes and accuracy metrics for all native models and projections.

Training Landscape	United States			Italy			Kazakhstan†
Projection	Native	IT	KZ	Native	KZ	US	Native	KZ	IT	US
Training	48	-	-	28	-	-	24	-	-	-
Testing	12	35	39	7	39	60	8	15	35	60
AUC	0.93	0.51	0.48	0.84	0.56	0.43	0.90	0.71	0.45	0.48
SE	0.05	0.05	0.05	0.09	0.05	0.04	0.09	0.08	0.05	0.04
Z	5.84‡	7.72‡	5.99‡	3.86‡	97.91‡	16.42‡	3.13‡	4.52‡	15.93‡	7.71‡
Total Omission	0	62.1	13.2	0	86.8	96.4	0	0	0	0
Average Omission	6.7	71	51	1.4	74.8	82.2	0	19.2	10	17.7
Total Commission	8.38	0	0.84	27.86	0	0	19.77	13.17	9.04	28.67
Average Commission	22.88	7.71	43.54	50.21	0.05	3.17	46.3	54.28	90.9	83.35
SAO	16.67	-	89.47	14.29	-	100	0	46.15	100	62.50

Native projection are under the curve (AUC) scores are shown in bold for comparison. SE = standard error, Z = z-score, SAO = summed area omission.

^† Native training and testing are for southern training area only

^‡ statistically significant value