Skip to main content

Table 3 Model selection results based on lowest AIC value from GLMM models examining the environmental cues for Common Snook migration during the spawning season

From: Primed and cued: long-term acoustic telemetry links interannual and seasonal variations in freshwater flows to the spawning migrations of Common Snook in the Florida Everglades

Hypothesized driver

Model variables

df†

AIC

ΔAIC‡

AIC Weight

Conditional R2

Marginal R2

Flow dynamics

Water level

3

2082

144.2

< 0.001

0.265

0.119

 

Daily water level change

3

2051

112.8

< 0.001

0.119

0.086

 

Water level + daily water level change

4

1980

41.5

< 0.001

0.292

0.175

Other hydrologic/abiotic conditions

Temperature

3

2174

235.3

< 0.001

0.042

0.006

 

Salinity

3

2171

233.1

< 0.001

0.042

0.006

 

Temperature + salinity

4

2169

230.6

< 0.001

0.039

0.012

Phenology

Day length

3

2176

237.6

< 0.001

0.053

0.003

 

Lunar cycle

3

2175

236.3

< 0.001

0.051

0.005

 

Day length + Lunar cycle

4

2174

235.6

< 0.001

0.058

0.008

Interannual variation

Year

9

2174

235.9

< 0.001

0.061

0.024

Global model

Water level + daily water level change + temperature + salinity + lunar cycle + year

14

1938

0

0.680

0.319

0.247

Best model

Water level + daily water level change + temperature + salinity + year

13

1940

1.6

0.320

0.316

0.245

  1. The response variable is a binary indicator of whether or not each fish was detected migrating for each day detected. For variable descriptions see Additional file 1: Table S2
  2. †Model degrees of freedom; ‡ Difference in AIC score between each model and lowest AIC model