Skip to main content

Table 1 Distribution models for ten amphibians and two reptiles by logistic regression analysis of presence-absence data for Portugal. Model conditions are with 13 variables (condition 1) or with nine variables (condition 2; details see text). Environmental data are standardized, except for LITH. Paramers with large effect (values >1) are shown in boldface type. The fit of the descriptive models is expressed by Cohen's kappa and the 'Area Under the Curve in Receiver Operating Characteristic' plots (AUC) and asymptotic standard error (SE). Distribution models derived from the equations are shown as 'probability of occurrence' maps for Portugal in figures 1 and 2 and as range maps for the Iberian Peninsula in figures 3 and 4.

From: From descriptive to predictive distribution models: a working example with Iberian amphibians and reptiles

  Observed Modelling Model equation † Model fit
Species presences condition ACID ALTI FROD FROM HARD HUMI INSO LITH ‡ NDVI PRET RELI TEMP TJUL Constant kappa AUC ± SE
Amphibians                    
a) Chioglossa lusitanica 202 1 -1.120 -1.635          3.405 1.919   -1.226 -2.511 0.8847 0.989 ± 0.002
   2   -1.864          3.900 1.743 -1.170   -1.857 0.8547 0.987 ± 0.003
b) Pleurodeles waltl 144 1, 2             -0.803 0.848   -0.559 0.3885 0.745 ± 0.018
c) Triturus marmoratus 226 1      -1.406    -1.057 -0.316    0.671    -0.368 0.6305 0.853 ± 0.013
   2       -0.522 -0.611    -0.780 0.607   -0.565   -0.649 0.596 0.854 ± 0.013
d) T. pygmaeus 130 1, 2    -0.512     -0.652    0.253 -0.943   0.984 0.444 -0.796 0.4072 0.751 ± 0.018
e) Alytes cisternasii 223 1    -0.623    -0.411 1.066 -1.376 0.759   -1.210     -0.915 0.5967 0.871 ± 0.012
f) A. obstetricans 169 1   0.872    -0.617   -0.416     0.535     -0.836 0.5737 0.879 ± 0.013
   2   1.028      -0.554     0.610     -0.674 0.5581 0.879 ± 0.013
g) Hyla arborea 132 1 -0.458   -0.523 0.336 -0.404 -0.367   0.789 -0.538 0.486 -0.849    -0.631 -0.337 0.3533 0.742 ± 0.021
   2    -0.463    -0.335      -0.526 -0.480 -0.615 -0.386 -0.167 0.2575 0.682 ± 0.023
h) H. meridionalis 95 1, 2   0.972 -0.577     0.408      -0.707 1.115   -0.709 0.4301 0.777 ± 0.02
i) Pelodytes ibericus 91 1    -0.371 -0.604 0.785         1.753 -0.717 -1.451 0.5909 0.863 ± 0.016
   2    -0.738           2.533 -0.639 -1.446 0.5052 0.843 ± 0.018
j) Rana iberica 229 1   0.635    -0.902       1.988     -0.934 0.7448 0.936 ± 0.008
   2   0.753      -0.605     1.890     -0.746 0.7494 0.935 ± 0.008
Reptiles                    
k) Anguis fragilis 83 1, 2       -0.360 -0.605     0.512    -1.046 -1.141 0.5866 0.876 ± 0.016
l) Lacerta schreiberi 272 1     -0.273    -1.080 -1.239 -0.501      -1.327 -0.103 0.6441 0.899 ± 0.010
   2        -0.858     0.479    -1.104 -0.594 0.6353 0.897 ± 0.010
  1. †) Model equation for e.g. Chioglossa lusitanica under model conditions of type 1 is: probability of occurrence = (1/(1+exp(1.120*ACID+1.35*ALTI-3.405*PRET-1.919*RELI+1.226*TJUL+2.511))).
  2. ‡) The categorical variable 'LITH' Is represented by two binary variables.