Table 1 Summary statistics for the three GAM models: the two spatial models tested the effect of the environmental predictors on little bustard density in each survey period (2003–2006 and 2016) and the population variation model tested the effect of the environmental predictors on the delta in little bustard density across surveys (2016–2003–2006).

From: Changes in grassland management and linear infrastructures associated to the decline of an endangered bird population

 

Model coefficients

Estimate

SE

t

edf

F

p value

Deviance explained (%)

Spatial models

Survey 2003–2006 Density

Intercept

0.87

0.16

5.49

  

0.000

24.6

Habitat

   

1.00

14.82

0.000

 

Survey 2016 Density

Intercept

− 0.09

0.18

− 0.47

  

0.643

71.2

Habitat

   

1.00

55.94

0.000

 

Cattle proportion

   

1.88

5.42

0.007

 

Population variation model

Density variation

Intercept

− 1.24

0.14

− 8.73

  

0.000

80.6

Density survey 2003–2006

   

1.76

85.81

0.000

 

Habitat_mean

   

1.84

21.41

0.000

 

Cattle proportion_mean

   

1.82

9.47

0.001

 

Power lines_mean

   

1.00

5.17

0.028

 
  1. SE Standard error, t T statistics, edf estimated degrees of freedom, F F statistics.
Back to article page