Table 2.
Suite of models considered in Stage 1 to estimate probability of hematozoa infection (Ψ) and probability of detecting hematozoa mtDNA (p)
| Model | K | Model description |
|---|---|---|
| (.) | 1 | pr (infection) does not vary |
| month | 3 | pr (infection) varies by month (May, June, August) |
| May models (17 May – 27 May) | ||
| sex | 2 | pr (infection) varies during May by sex |
| species | 4 | pr (infection) varies during May by species (MALL, NOPI, AMWI, LESC) |
| species + sex | 5 | pr (infection) varies during May by species and by sex to the same degree for all species |
| June models (10 July – 19 June) | ||
| sex | 2 | pr (infection) varies during June by sex |
| species | 3 | pr (infection) varies during June by species (MALL, NOPI, AMWI) |
| species + sex | 4 | pr (infection) varies during June by species and by sex to the same degree for all species |
| August models (28 July – 20 August)a | ||
| age | 2 | pr (infection) varies during August by age (adult, juvenile) |
| sex | 2 | pr (infection) varies during August by sex |
| age + sex | 3 | pr (infection) varies during August by age and by sex to the same degree for both age classes |
| age * sex | 4 | pr (infection) varies during August by each age and sex class |
| (ad * sex) + (juv) | 3 | pr (infection) varies during August by age and by sex for adult birds |
| species | 3 | pr (infection) varies during August by species (MALL, NOPI, AGWT) |
| species * age | 6 | pr (infection) varies during August by species and age |
| species + age | 4 | pr (infection) varies during August by species and by age to the same degree for all species |
| species * sex | 6 | pr (infection) varies during August by species and sex |
| species + sex | 4 | pr (infection) varies during August by species and by sex to the same degree for all species |
| species * age + sex | 7 | pr (infection) varies during August by species and age, and by sex to the same degree for all species |
| species * sex + age | 7 | pr (infection) varies during August by species and sex, and by age to the same degree for all species |
| species * age * sex | 12 | pr (infection) varies during August by species, age, and sex |
| (ad * species * sex) + (juv * species) | 9 | pr (infection) varies during August by species and by sex for adult birds |
| August temporal trend models | ||
| + day | 1 | pr (infection) during August varies in a linear trend with a single slope for all sources of variation |
| + day * ageb | 2 | pr (infection) during August varies in a linear trend with a unique slope for each age class |
| * day | 2+ | pr (infection) during August varies in a linear trend with a unique slope for each source of variation |
This suite of models was considered independently for each of three hematozoa genera (Leucocytozoon, Haemoproteus, Plasmodium). K = number of model parameters applicable only to modeling of Ψ or p in a given month and is not representative of the total number of model parameters. Age of birds was either after-hatch-year (adult) or hatch-year (juvenile). Species abbreviations: AGWT American green-winged teal, AMWI American wigeon, LESC lesser scaup, MALL mallard, NOPI northern pintail
aAugust models were considered alone, with the additive temporal trend (+ day), with the temporal trend varying by age (+ day * age), and with the multiplicative temporal trend (* day)
bModel structure was applied only to August models containing variation in age