Effects of avian hematozoa on behavior and energetic condition of migratory landbirds
Luke W. DeGroote, MS
Advisor: Paul Rodewald
Diseases may have a profound effect on the persistence and evolution of populations by altering energetic condition or reproductive success of individuals. For example, it has been hypothesized that hematozoan infections may drive the evolution of sexual dichromatism in birds (Hamilton & Zuk 1982). Ecological studies have provided evidence in support of parasite-mediated sexual selection by demonstrating that hematozoan infection results in reduced reproductive success. Few studies, however, have investigated the effects of hematozoa on migrating landbirds. This research contributes to our knowledge of hematozoan parasites and their effects on migratory birds by: (1) developing a semi-automated method to quantify hematozoan intensity, (2) determining interspecific differences in prevalence and intensity of infection at a northerly migratory stopover site, and (3) elucidating pathogenic effects of hematozoans on migrating landbirds.
Intensity of hematozoan infection is infrequently quantified because accurate calculations require visual counts of parasites relative to the number of erythrocytes counted. Manual quantification of erythrocytes can be circumvented using ImageJ software (Natl. Inst. of Health) to count erythrocyte nuclei from digital images. Intensity of infection calculated from traditional visual erythrocyte counts was compared to that calculated from digital erythrocyte counts. I found that the ratio of microscope erythrocyte counts to image erythrocyte counts did not differ between two randomly selected groups (mean difference=0.14, SE=0.07, F1, 34=2.15, P=0.15), indicating that image erythrocyte counts can accurately estimate the number of erythrocytes within larger microscopic fields of view. Intensity of hematozoan infection calculated from manual quantification of 2,000 erythrocytes was 0.46 times lower (bootstrap P=0.02) than intensity calculated from digital quantification of 50,000 erythrocytes. I contend that digital quantification of hematozoa infection offers a rapid and more accurate method to quantify infections of varying intensity.
Using digital quantification techniques, I examined whether hematozoan infection was associated with extent of prealternate molt, refueling performance, energetic condition, and arrival timing during migratory stopover. Blood samples were collected from Yellow-rumped Warblers (Dendroica coronata, n = 519), Magnolia Warblers (Dendroica magnolia; n = 385), and Yellow Warblers (Dendroica petechia; n = 216) from mid-April through May of 2004 and 2005 in northwestern Ohio. I found that 36.3% of Magnolia Warblers, 12.7% of Yellow Warblers, and 58.3% of Yellow-rumped Warblers were infected with one or more genera of hematozoa. Fewer Yellow-rumped Warblers (2.1%) were infected by Plasmodium spp. than Magnolia (8.9%) or Yellow Warblers (4.4%) (Fisher’s Exact P<0.01), while Leucocytozoon spp. infected fewer Yellow Warblers (0.0%) than Magnolia (8.9%) and Yellow-rumped Warblers (5.8%) (P<0.01). Interspecific differences in prevalence and increased prevalence at a northerly stopover site suggests that relapse occurs during spring migration.
Akaike Information Criterion and multi-model averaging provided evidence that late arrival (prevalence _=0.13, SE=0.01, RVI=0.98; intensity _=0.13, SE=0.03, RVI=1.00), and reduced energetic condition (intensity _=_0.27, SE=0.12, RVI=1.00) at stopover sites were associated with higher probability and intensity of infection in Yellow-rumped Warblers. In contrast, probability and intensity of infection in migrant Magnolia and Yellow Warblers were not significantly associated with either energetic condition or capture date. Extent of prealternate molt and refueling performance were not associated with hematozoan infection in migrant Magnolia and Yellow-rumped Warblers. However, intensity of Haemoproteus or Plasmodium infection was higher in second year Magnolia and Yellow-rumped Warblers (_=_0.98, SE=0.52, RVI=0.98; _=_1.23, SE=0.34, RVI=1.00 respectively) than after second year individuals. This study provides evidence that for some migrating landbird species, events during one stage of the annual cycle could influence parasitization, energetic condition, migratory timing, and reproductive performance in subsequent stages.