Network Models: An Underutilized Tool in Wildlife Epidemiology?

Table 1

Direct and indirect techiques that could be used to collect contact network data on wildlife populations and selected examples using these techniques.

Technique	Useful for which type of species?	Comments	Selected references
Direct
Behavioral observations of known individuals	Diurnal habituated animals that can be easily observed (not cryptic species)	Potentially a “gold standard” for contact networks (multiple types of social interactions can be recorded); labor intensive	[16–19]
Viewpoint scanning	Visible animals active during the day; open habitat (not cryptic species)	Allows between-species observations at replicable sites; labor intensive yet incomplete observations	[20]
Indirect
Biologging	Easily captured and handled individuals	Population needs to be saturated with detectors; excellent resolution of proximity data although proximity does not mean contact; continuous time record; cannot distinguish between types of close contacts (e.g., fighting versus mating)	[21]
Biologging:  animal-borne  acoustic proximity  receiver	Marine mammals	Need to handle animal to retrieve device; good between-animal resolution	[22]
Biologging: PIT (Passive  Integrated Transponder)  tags	Useful for small mammals	Good data on duration of presence/absence of marked individuals at specific places (e.g., supplemented foraging sites) equipped with PIT loggers; approximation of contacts	[23]
Biologging: proximity  data loggers/collars	Medium to large animals	measure frequency and duration of contact; complete temporal data; need to recover loggers	[24–26]
Capture-mark-recapture	Easily captured and handled individuals	A contact is defined as occupying same area during same period of time; good for capturing movement/dispersal data, not good at capturing within-group contacts	[27–31]
Direct manipulation	Captive populations of common animals	Great for repeatable experiments on experimentally infected individuals to measure transmission, but does this reflect contact patterns in wild?	[32, 33]
GPS recorders	Easily captured and handled medium to large individuals	Need recorders on all individuals in select area; if recorders are synced well, excellect contact data for the time the GPS takes point (with spotty coverage in between). Maybe local avoidance happens but would be undetected?	[15]
Powder marking	Easily trapped and handled individuals	Gives good contact data if contacts involve direct phyical contact; can only monitor a few indivuals at a time due to contstraints on the number of powder colors	[23]
Radio telemetry	Handled individuals, not good for very small individuals	Contact defined as occupying same area during same period of time. Good indicator of (i) scale of interaction but gives coarse resolution of a “contact”, (ii) mixing between groups of animals, but not within groups and (iii) den-sharing contacts. Presence of fieldworkers may alter behavior.	[27, 34–36]
Trapping and bait  marking	Easily trapped and handled individuals who use latrines to mark territories	Good data on home range overlap and intergroup movement rates	[37]
Video tracking from  animal’s perspective	Animal must be able to be caught and wear something like a video backpack	Great contact data from individual perspective	[38]
Video trapping from  fixed perspective  (automated)	Social insects that can be individually tagged and the group monitored	Great resolution of contact data; software records duration and frequency of contacts	[39]