Table 2 Binomial logistic regression models (reduced from the full models, Additional file 1: Table S1) testing the effects of biometric and environmental predictor variables on the overall presence of Hematodinium in the population. Models separated by location: Model 1, total population; Model 2, Dock; Model 3, Pier

Model 1

Hemat ~ Season + Sex

Season (spring)

0.8137

0.2628

0.00196**

+ HemoCol

Season (summer)

0.7437

0.2670

0.00535**

df = 1188

Season (winter)

0.4678

0.2740

0.08776

AIC: 914.72

Sex (male)

0.7940

0.1828

1.4e−05***

HemoCol (milky)

1.1187

0.2716

3.8e−05***

Model 2

HematDock ~ Sex

Sex (male)

1.4790

0.2918

7.4e−07***

+ Colour + HemoCol

Colour (orange)

− 0.5830

0.3827

0.1277

df = 595

Colour (yellow)

0.3667

0.2808

0.1915

AIC: 450.92

HemoCol (milky)

1.1323

0.4070

0.0054**

Model 3

HematPier ~ Season

Season (spring)

0.99428

0.38751

0.010293*

+ CW + HemoCol

Season (summer)

0.75876

0.40444

0.060645

df = 586

Season (winter)

0.35552

0.42248

0.400067

AIC: 430.7

CW

− 0.06889

0.02065

0.000848***

HemoCol (milky)

1.17749

0.38636

0.002306**