Table 3. Significant parameters in the regression model using Batrachochytrium dendrobatidis (Bd) occurrence and data from wild-occurring amphibians having precise locations in the global Bd database (N = 3,733 locations).

Parameter	Coefficient	SE	P value
Intercept	1.186	0.262	––
Enigmatic Declines	0.141	0.028	<0.001
Temperature Range	−0.089	0.017	<0.001
Average Annual Precipitation	−0.0005	0.000	<0.001
Biome	––	––	<0.001
Tundra (included in intercept)	––	––
Montane Grasslands & Shrublands	1.929	0.366
Mediterranean Forest, Woodlands & Scrub	−0.598	0.126
Tropical & Subtropical Dry Broadleaf Forests	−0.346	0.406
Boreal Forests/Taiga	−0.649	0.340
Mangroves	0.251	0.773
Deserts & Xeric Shrublands	0.471	0.186
Flooded Grasslands & Savannas	0.323	0.576
Temperate Broadleaf & Mixed Forests	0.397	0.122
Temperate Grasslands, Savannas & Shrublands	−0.346	0.172
Temperate Coniferous Forests	0.229	0.124
Tropical & Subtropical Grasslands	−0.582	0.198
Tropical & Subtropical Moist Broadleaf Forests	0.224	0.172
Tropical & Subtropical Coniferous Forests	0.503	0.278

Individual biome coefficients represent additive shifts in the Bd-occurrence odds ratio, whereas other coefficients represent multiplicative changes in odds of detecting Bd. Logistic regression with likelihood ratio test statistic (Chi²) was used.