Predictive distribution modeling of Swertia bimaculata in Darjeeling-Sikkim Eastern Himalaya using MaxEnt: current and future scenarios

Ecological Processes

Table 5 Prediction accuracy of Swertia bimaculata distribution models

		Present		Year 2050			Year 2070
		Comp.	Uncor.	RCP 2.6	RCP 4.5	RCP 8.5	RCP 2.6	RCP 4.5	RCP 8.5
AUC value		0.982	0.980	0.985	0.982	0.985	0.986	0.985	0.982
TSS value		0.783	0.645	0.714	0.728	0.638	0.689	0.733	0.796
Percentage of contribution	Variable code	BIO13	BIO18	BIO18	BIO18	BIO18	BIO18	BIO18	BIO18
Percentage of contribution	Value	39.1%	40.2%	40.5%	41.2%	40.9%	40.8%	41.2%	43.3%
Permutation of importance	Variable code	BIO6	BIO15	ELEV	ELEV	ELEV	ELEV	ELEV	ELEV
Permutation of importance	Value	53.3	26.4	50.8	42.3	49.5	39.6	48.9	45.3
Area in km² (10^th percentile training presence threshold)		451.89	869.48	84.03	238.38	48.88	3.43	147.49	0
Percentage of area (10^th percentile training presence threshold)		3.92%	7.55%	0.73%	2.07%	0.42%	0.03%	1.28%	0%
Highest probability of species occurrence		0.843	0.902	0.658	0.640	0.605	0.493	0.539	0.301