. 2012 Jan 23;7(1):e30388. doi: 10.1371/journal.pone.0030388

Table 2. Results of the linear mixed-effects model fit by REML (model b) for predicting δD_h from latitude, longitude and elevation, with species as random intercept.

Model parameter	Estimate	SE	df	t-value	P
Intercept	−4.96	11.31	166	−0.44	0.66
Longitude	2.02	0.89	166	2.28	0.024
Latitude	−1.26	0.25	166	−5.00	<0.0001
Elevation²	0.0	0.0	166	1.14	0.26
Longitude:Latitude	−0.058	0.02	166	−3.13	0.002
Longitude:Elevation²	0.0	0.0	166	−1.121	0.26
Latitude:Elevation²	0.0	0.0	166	−1.41	0.25
Longitude:Latitude: Elevation²	0.0	0.0	166	1.09	0.28

Number of observations: 178. Number of groups (random effect species): 5. AIC = 1444.34, BIC = 1475.69, logLik −712.17. The random intercept was normally distributed (mean 0, SD 5.62), and so was its residual term (mean 0, SD 9.38). Model residuals were normally distributed (Lilliefors D = 0.0403, P = 0.1732).

: variable included as quadratic term.

Table 2. Results of the linear mixed-effects model fit by REML (model b) for predicting δDh from latitude, longitude and elevation, with species as random intercept.

Table 2. Results of the linear mixed-effects model fit by REML (model b) for predicting δD_h from latitude, longitude and elevation, with species as random intercept.