Table 2. Linear regression model for orang-utan abundance on an expedition, conditional on at least one detection.

	Estimate	Std. Error	z value	Pr(>|z|)
Intercept Model 2a	21.500714	5.689382	3.779	0.000527 ***
Intercept Model 2b	21.487783	5.685365	3.779	0.000513 ***
Year Model 2a	−0.011465	0.002917	−3.931	0.000337 ***
Year Model 2b	−0.011317	0.002911	−3.888	0.000372 ***
Person Model 2a	0.028371	0.029204	0.971	0.337299
log(Days) Model 2a	0.708796	0.108887	6.509	1.02e-07 ***
log(Days) Model 2b	0.677241	0.103856	6.521	8.76e-08 ***

Model 2a provides full model of non-zero abundance. Model 2b provides model of non-zero abundance, omitting Person. Year = year in which expedition was conducted. Person = number of people on an expedition. Log(Days) = natural logarithm of duration of expedition in days. Residual standard error in Model 2a: 0.97 on 39 degrees of freedom. Residual standard error in Model 2b: 0.9693 on 40 degrees of freedom. Significance code: ‘***’: p<0.001.