Reliably capturing transient animal behavior in the field and laboratory remains a logistical and financial challenge, especially for small ectotherms. Here, we present a camera system that is affordable, accessible, and suitable to monitor small, cold-blooded animals historically overlooked by commercial camera traps, such as small amphibians. The system is weather-resistant, can operate offline or online, and allows collection of time-sensitive behavioral data in laboratory and field conditions with continuous data storage for up to four weeks. The lightweight camera can also utilize phone notifications over Wi-Fi so that observers can be alerted when animals enter a space of interest, enabling sample collection at proper time periods. We present our findings, both technological and scientific, in an effort to elevate tools that enable researchers to maximize use of their research budgets. We discuss the relative affordability of our system for researchers in South America, which is home to the largest population of ectotherm diversity.

Detailed instructions on camera setup and usage can be found in the Supplementary materials on GitHub (https://github.com/lobsu/WyzeSetUps/wiki).

In the laboratory, Wyze v3 cameras were adhered by velcro onto the side of the Exo Terra tanks and suspended above the tadpole canister, with the face of the camera approximately 15.5–17.5 cm above the bottom of the canister. For egg care observations, Wyze v3 cameras are oriented horizontally, with the face of the camera facing approximately 10 cm from the canister entry site. Cameras were charged using their prepackaged USB cords and connected via a Nexwell USB charging station (Amazon, Bellevue, WA, USA) to the building outlets. Cameras were given 128 Gb SD cards for continuous recording. 128 Gb SD cards fill after approximately 10 days and are switched by researchers and replaced with an empty SD card.

In the field, Wyze v3 cameras were equipped with a 128 GB SD card and set to the desired operating mode (motion triggered or continuous) using the WiFi of the field station before being transported to the field. We recommend setting the phone to the time zone in which the camera will be used during setup to avoid incorrect timestamps after changing time zone. Cameras were set to motion detection using the lowest sensibility, sound detection and notifications were switched off. All SD cards were formatted in the camera prior to use. Cameras were disconnected from power sources and WiFi during the transport to the study site.  We used Anker PowerCore Fusion 10000 Model A1623 (9700mAh/35.21Wh) power banks for short-term (1 day) monitoring and Anker PowerCore26800 Model A1277 as well as 2nd Gen Astro E7 ModelA1210 (both 26800mAh/96.48Wh) for longer-term (10 days) surveil of tadpoles in their natural habitat. For long-term operation, we replaced the power bank every 2 days and recharged the devices in the camp. Prior to field implementation, we tested the efficiency of these power banks and found the operating mode of the camera (continuous vs motion-triggered) did not affect the monitoring time.

Cameras were labeled and secured to bromeliad leaves 20–40 cm above the water reservoir housing a tadpole using commercial rubber bands. Power banks were wrapped with an absorbent diaper and placed in 2 plastic zip lock bags with their opening facing into different directions to minimize the amount of rainwater seeping in. We purposefully chose inexpensive zip-lock plastic bags to cover the power supply from rain to test the operation of our system on a budget. Alternatively, plastic bags can be replaced by a variety of outdoor waterproof boxes designed for electrical equipment. We tried to shelter Power banks from the sun and rain by placing them under bromeliad leaves. To assure an accurate timestamp, we recorded the screen of our mobile phone showing the accurate time as a first action with the camera after reconnecting the camera to the power source. We replaced the SD card every 2 days when changing the power source to back up contained footage and moved the camera to a different tadpole nursery after 10 days.

Data Processing

With continuous recording, every minute of every day is recorded. Videos were first processed and kept only when frogs entered the camera. The scorer marked the times the frog entered the focal canister, defining entrance and exit as when the entire frog’s body was within the canister. Data processing was performed in RStudio (version 4.2.3, PBC, Boston, MA). 

Excel, RStudio.