Using occupancy models to investigate the prevalence of ectoparasitic vectors on hosts: an example with fleas on prairie dogs

David A. Eads; Dean E. Biggins; Paul F. Doherty; Kenneth L. Gage; Kathryn P. Huyvaert; Dustin H. Long; Michael F. Antolin

doi:10.1016/j.ijppaw.2013.09.002

Home/Archive/USGS

USGS · 70066096

Using occupancy models to investigate the prevalence of ectoparasitic vectors on hosts: an example with fleas on prairie dogs

David A. Eads·Dean E. Biggins·Paul F. Doherty·Kenneth L. Gage·Kathryn P. Huyvaert·Dustin H. Long·Michael F. Antolin

View original source ↗Export metadata ↓

Abstract

Ectoparasites are often difficult to detect in the field. We developed a method that can be used with occupancy models to estimate the prevalence of ectoparasites on hosts, and to investigate factors that influence rates of ectoparasite occupancy while accounting for imperfect detection. We describe the approach using a study of fleas (Siphonaptera) on black-tailed prairie dogs ( Cynomys ludovicianus ). During each primary occasion (monthly trapping events), we combed a prairie dog three consecutive times to detect fleas (15 s/combing). We used robust design occupancy modeling to evaluate hypotheses for factors that might correlate with the occurrence of fleas on prairie dogs, and factors that might influence the rate at which prairie dogs are colonized by fleas. Our combing method was highly effective; dislodged fleas fell into a tub of water and could not escape, and there was an estimated 99.3% probability of detecting a flea on an occupied host when using three combings. While overall detection was high, the probability of detection was always <1.00 during each primary combing occasion, highlighting the importance of considering imperfect detection. The combing method (removal of fleas) caused a decline in detection during primary occasions, and we accounted for that decline to avoid inflated estimates of occupancy. Regarding prairie dogs, flea occupancy was heightened in old/natural colonies of prairie dogs, and on hosts that were in poor condition. Occupancy was initially low in plots with high densities of prairie dogs, but, as the study progressed, the rate of flea colonization increased in plots with high densities of prairie dogs in particular. Our methodology can be used to improve studies of ectoparasites, especially when the probability of detection is low. Moreover, the method can be modified to investigate the co-occurrence of ectoparasite species, and community level factors such as species richness and interspecific interactions.

Explore related subjects

International Journal for Parasitology: Parasites and Wildlife Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

David A. Eads, Dean E. Biggins, Paul F. Doherty, Kenneth L. Gage, Kathryn P. Huyvaert, Dustin H. Long, Michael F. Antolin. 2013. Using occupancy models to investigate the prevalence of ectoparasitic vectors on hosts: an example with fleas on prairie dogs. https://doi.org/10.1016/j.ijppaw.2013.09.002

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

Contrasting haemosporidian infections in two ecologically distinct wading birds from breeding colonies in the southeastern United States

Wading birds may serve as ideal hosts for avian hemoparasites, as they are long-lived, undertake extensive movements, form dense breeding colonies, and inhabit water-associated environments that support vectors. Although previous studies have reported parasite species and prevalence in various wading bird species, little is known about their associations with bird behavior and life stage. To address this gap, we examined haemosporidian infections in two ecologically distinct species, white ibis ( Eudocimus albus ) and tricolored heron ( Egretta tricolor ), to explore differences in life stage and movement that may explain prevalence differences. We combined blood screening for hemoparasites with satellite tracking data describing birds’ movement patterns. We screened 95 white ibis (67 juveniles and 28 adults or subadults) and 69 tricolored herons (45 juveniles and 24 adults). We detected a single Haemoproteus plataleae lineage in both species, with higher infection prevalence in white ibis (42.1 %) than in tricolored herons (14.5 %). Among white ibis, adults showed a higher prevalence (67.9 %) than juveniles (31.3 %), whereas in tricolored herons, adults had a prevalence of 8.3 % and juveniles 17.8 %. Non-breeding season movement data showed that white ibis used both freshwater and saline habitats across the southeastern United States, which may also serve as habitats for vectors. In contrast, tricolored herons remained mainly along coastal areas, using saline habitats in the southeastern United States and wintering sites in Central America, which may be less favorable for vectors. Overall, white ibis may serve as major reservoirs and sources of reinfection for H. plataleae , whereas tricolored herons may facilitate parasite dispersal between breeding colonies along the coasts of the southeastern United States and wintering areas in Central America. This study presents the first direct comparison of hemoparasite infections in two ecologically distinct wading birds and highlights movement data as key to explaining infection differences, providing a baseline for future studies.

Alabama↗

USGS2026

Effects of fipronil bait pellets on two cricetid species: Potential implications for plague mitigation and wildlife conservation

We evaluated the effects of fipronil bait pellets on two cricetids that commonly occupy colonies of black-tailed prairie dogs ( Cynomys ludovicianus ; BTPDs): western deer mice ( Peromyscus sonoriensis ) and northern grasshopper mice ( Onychomys leucogaster ). In one experiment, bait pellets (0.96 mg fipronil/bait) were applied at 75 baits/ha to three 1.44-ha plots on a BTPD colony. Mouse abundance declined by 70% from before to 6-10 d after treatment. In a second experiment, bait pellets (0.46 or 1.52 mg fipronil/bait) were applied at 125 baits/ha to four plots (0.85-1.86 ha) on two BTPD colonies; two non-treated plots were baselines (1.09 and 2.06 ha). From before to 11-15 d after treatment, mouse abundance declined by 51%- 67% on the treated plots vs. a decline of 9% on the non-treated plots. Mouse survival from before to 11-15 d after treatment was 51% lower on the treated plots. In a third experiment, bait pellets (0.84 mg fipronil/bait) were applied at 125 baits/acre on two 1.44-ha plots on a BTPD colony; two 1.44-ha non-treated plots were baselines. Mouse survival from before to 30-44 d after treatment was 45% lower on the treated plots; the abundance of deer mice on the treated plots remained similar from before to 30-44 d after treatment, perhaps due to juvenile recruitment and/or immigration. In a laboratory experiment, 33 deer mice offered one bait pellet (0.86 mg fipronil/bait) consumed 27% of their bait, on average (range = 0-100%). Over 3 d, deer mouse mortality was estimated at 53%; mortality increased with fipronil dose, which averaged 11 mg fipronil/kg body mass (range = 3-46 mg/kg). Brain samples were available from 31 deer mice; all tested positive for fipronil sulfone, the primary mammalian metabolite of fipronil, at 19 to 61,205 ng fipronil sulfone/g. Additional experiments could determine if these findings scale up to larger landscapes.

South Dakota↗

USGS2024

Diversity, distribution, and methodological considerations of haemosporidian infections among Galliformes in Alaska

Using samples spanning 10-degrees of latitude in Alaska, we provide the first comparative assessment of avian haemosporidia distribution of Arctic Alaska with subarctic host populations for four species of grouse and three species of ptarmigan (Galliformes). We found a high overall prevalence for at least one haemospordian genus (88%; N = 351/400), with spruce grouse ( Canachites canadensis ) showing the highest prevalence (100%; N = 54/54). Haemoproteus and Plasmodium lineages were only observed within grouse, while Leucocytozoon species were found within both grouse and ptarmigan. Further, different Leucocytozoon lineages were obtained from blood and tissue samples from the same individual, potentially due to the differential timing and duration of blood and tissue stages. Using different primer sets, we were able to identify different Leucocytozoon lineages within 55% (N = 44/80) of sequenced individuals, thereby detecting coinfections that may have otherwise gone undetected. The commonly used Haemoproteus / Plasmodium primers amplified Leucocytozoon for 90% (N = 103/115) of the products sequenced, highlighting the potential value of alternate primers to identify intra-genus coinfections and the importance of obtaining sequence information rather than relying solely on PCR amplification to assess parasite diversity. Overall, this dataset provides baseline information on parasite lineage distributions to assess the range expansion associated with climate change into Arctic regions and underscores methodological considerations for future studies.

Alaska↗