Using Raspberry Pi microcomputers to remotely monitor birds and collect environmental data

Abstract

Technological innovations have expanded our ability to understand ecological systems and conserve biodiversity at a time when many species are at risk from climate change and habitat loss. One technology that is predicted to greatly benefit the field of avian ecology is the microcomputer. Raspberry Pi microcomputers are low-cost alternatives to personal computers and are able to house a variety of environmental sensors. Raspberry Pis have recently been used to monitor the behavior of birds in artificial nest cavities. Here, we deploy and assess the ability of five different Raspberry Pi-based devices to record observational and environmental data at avian feeding stations. From January to March 2018, each device generated digital photographs and recorded related temperature and wind speed measurements every 10 s, for three hours per day, at five locations in Stark County, OH, USA. Among photographs taken, 96% were classified as ‘usable’ and no differences were noted in percent of usable photographs among devices (P = .50). In addition, no differences were noted in the mean number of feeder visits of two common birds species detected by devices and detected by a field observer (P = .39 and 0.60). Collectively, temperature measurements recorded by devices were similar to temperatures reported at the Akron-Canton Airport (P = .88), but wind speeds were 1.78 ± 0.12 m/s lower (P < .001) at field sites. Our results demonstrate the ability of Raspberry Pis to remotely monitor birds and collect site-based environmental data in a cost-effective manner. We provide direction for making economical decisions when selecting cameras and temperature sensors without compromising data quality. While our study focuses on birds, our results are generalizable and could provide direction for remotely monitoring a variety of vertebrate and invertebrate taxa.