Table 3 Results from the final GAM for identification success.

From: Assessing data bias in visual surveys from a cetacean monitoring programme

GAM n = 2496; Deviance explained = 6.06%; R-square = 0.078; UBRE = 0.31

Variables

Degrees of Freedom

Reference Degrees of Freedom

Chi-Square

P-value

s(Dist)

2.104

2.481

127.372

<0.001

s(Size)

1.238

1.421

63.710

<0.001

s(Vis)

2.340

2.699

9.264

0.048

s(Wind)

2.308

2.646

16.700

0.002

Variables

Estimate

Standard Error

Z-value

Pr(>|z|)

roup

0.71889

0.12366

5.813

<0.001

  1. Generalized Additive Model (GAM) to assess the bias on the identification success, fitted with variables related with the group size, the distance to the sighting, the meteorological conditions, and the experience of observers. Dist – distance of the sighting to the vessel (scale of 0 to 7); Group – cetacean group of species (Group A – Odontoceti sightings, excluding sperm whales; Group B – Mysticeti sightings, plus sperm whales); Size – group size (i.e., number of animals in the group) of the sighting; Vis – visibility state (scale of 0 to 10). Wind – wind state (Beaufort scale); For details on variables, see methods.
Back to article page