Table 1. Ranking of the best model fits.

The selection is based on the second-order Akaike’s information criterion (AIC_c) corrected for small sample sizes (see Materials and Methods). The Generalized Linear Mixed Models were fitted with binomial errors for predicting seabird strandings as functions of environmental variables and oil pollution along the coast of Brazil.

Models	AICc	ΔAICc	wi
Brown booby
wh + dbi + ds	311.7	0.0	0.23
wh + dbi	311.7	0.0	0.22
wh + wd + dbi + ds	312.1	0.4	0.18
sst + wh + wd + dbi + ds	313.3	1.6	0.10
wh + dbi + sst	313.4	1.7	0.09
Magnificent frigatebird
owf + os + dbi	126.2	0.0	0.28
owf + ro + dbi	127.5	1.3	0.15
owf + ro + os +dbi + ds	127.5	1.3	0.14
owf + os + dbi + ds	128.1	1.9	0.11
Kelp gull
sst + cl + dbi + ds	120.7	0.0	0.44
sst + cl + wp + dbi + ds	121.6	0.9	0.28
Cabot's tern
sst + ro + wh + os + dbi * bp	227.6	0.0	0.18
sst + cl + ro + wh + os + dbi * bp	227.8	0.2	0.16
mc + sst + ro + wh + ws + os + dbi * bp	228.2	0.6	0.14
mc + sst + cl + ro + wh + ws + os + dbi * bp	228.7	1.1	0.11
sst + ro + wh + ws + os + dbi * bp	228.8	1.2	0.10
sst + cl + ro + wh + os + dbi * bp	229.4	1.7	0.08

Abbreviations indicate: chlorophyll-a concentrations (cl), distance from breeding islands (dbi), breeding period (bp), surveyed distance (ds), meridional currents (mc), oil spills (os), onshore wind frequency (owf), river outflow (ro), sea surface temperature (sst), wave height (wh), wave period (wp), wind direction (wd) and wind speed (ws). AIC_c = Second-order Akaike’s information criterion corrected for small sample sizes, ΔAIC_c = difference in AIC_c score between ranked models; w_i = AIC weights.