Fig 1.
Location of the two bays selected for this study within Chiloé Island (see text for details).
Caulín (disturbed bay); Pullao (non-disturbed bay).
Fig 2.
Variation (means ± SE) in Hudsonian godwit abundance within a fixed sector length (a proxy of density) throughout the low-tide period (i.e. 4 hours) in the disturbed (grey bars) and non-disturbed bay (black bars) (see text for details) over the 2-year study.
Fig 3.
Expected godwit abundance within a fixed sector length (a proxy of density) during the low-tide period in the disturbed (grey line) and non-disturbed bay (black line) (see text for details) in relation to (A) Days after October 20th; (B) Tide height. Note that values close to 0 are referred to spring tides and close to 60 are referred to neap tides.
Table 1.
Generalized linear model showing the effects of site (disturbed and ‘non-disturbed bay’), tide period (hours with respect to low tide), ‘daysafter’ (number of days after October 20th) and tide height (tidal amplitude), and their interactions on Hudsonian godwit abundance in two bays within Chiloé Island (see text for details).
Fig 4.
Variation (means ± SE) in Hudsonian godwit foraging activity (%) throughout the low-tide period (i.e. 4 hours) in the disturbed (open triangles and grey line) and non-disturbed bay (full squares and black line) (see text for details) over the 2-year study.
Fig 5.
Expected abundance of Hudsonian godwits as a function of the number of people present in the disturbed bay.
Note observed data (grey dots) superimposed (jitter plot), whereas the line represents the predicted values from the ZINB model (see text for details).