Infrastructure and the ethnographic-cartographic production of urban bird species richness

The qualitative bio-geographies of human geographers and the quantitative mappings of biogeographers share a goal: how to understand living with non-human life. Yet they rarely bridge the conceptual and methodological gap between them. This paper theorizes how the concept of infrastructure can bridge this ethnographic-cartographic divide. Infrastructure is not just an inanimate shell. It is also a system of relation, a dynamic patterning of socionatural form emerging out of the experiences and affective moments of its constituents. As a proof of concept, we quantified and compared urban bird species richness and frequency for Tallahassee, Florida over a 17-year period (2000–2017) for two co-occurring observational infrastructures, eBird and a wildlife rehabilitation center that serves the city. Species common to both infrastructures comprised 94% of all eBird observations and 99% of all rehab records. Their differences reflected contrasts in how the motivations for experiencing birds intersected with bird habitat preferences, behavior, and contingencies of urban history and development. eBird observations had high species richness (295 spp) and reflected the growing popularity among birds and a small number of active birders for visiting stormwater retention lakes recently modified to improve bird habitat. Rehabilitation records had a lower richness (194 spp) and exhibited a much more even distribution of bird encounters among individual residents as well as community institutions like schools, universities, law enforcement, and other government organizations. Infrastructural perspectives convey how affective and individualistic encounters with non-humans can link to emergent biogeographic mappings and how urban biodiversity is relationally and heterogeneously produced rather than simply contained in cities. 

Methods
The raw wildlife rehabilitation data set was obtained from St Francis Wildlife Association and Rehabilitation Center in Quincy, Florida. These data consisted of the records of birds brought in for rehabilitation over the period 2000–2017. These data were cleaned up to include only complete records. Addresses for each record are associated with the individual or entity that brought the birds in for rehab. These were geocoded to produce a distribution of point locations for these rehab events. eBird data was obtained from the online website. These were filtered to include only those observations that were part of completed checklists.