BioSense: An automated sensing node for organismal and environmental biology

Abstract

Automated remote sensing has revolutionized the fields of wildlife ecology and environmental science. Yet, a cost-effective and flexible approach for large scale monitoring has not been fully developed, resulting in a limited collection of high-resolution data. Here, we describe BioSense, a low-cost and fully programmable automated sensing platform for applications in bioacoustics and environmental studies. Our design offers customization and flexibility to address a broad array of research goals and field conditions. Each BioSense is programmed through an integrated Raspberry Pi computer board and designed to collect and analyze avian vocalizations while simultaneously collecting temperature, humidity, and soil moisture data. We illustrate the different steps involved in manufacturing this sensor including hardware and software design and present the results of our laboratory and field testing in southwestern United States.

Keywords: Avian ecology; Bioacoustics; Microclimate; Raspberry Pi; Remote sensing.

Graphical abstract

Fig. 1

The main components of a BioSense node. The basic hardware configuration of each unit includes an integrated a Raspberry Pi computer board, a BME280 weather station for air temperature, humidity and pressure data collection protected by a radiation shield, a Polyvinyl Chloride (PVC) waterproof box fitted caps clamped down with stainless steel bolts and nuts, a stereo microphone, two soil moisture probes, and an optional solar-charged battery pack if no electricity plugs are available. A) protective enclosure, B) Raspberry Pi 4, C) radiation shield, D) Wi-Fi antenna, E) two soil moisture sensors, F) microphone, G) BME280.

Fig. 2

Real Time Clock (RTC) port and color table.

Fig. 3

Raspberry Pi cable and power supply configuration before insertion into the waterproof case.

Fig. 4

1) Set of 40 pre-soldered general purpose input/ouput (GPIO) connectors for additional sensors. 2) Quad-core Cortex-A72 (ARM v8) 64-bit and 1.5 GHz. 3) RAM up to 8 GB LPDDR4-2400 SDRAM. 4) Gigabit Ethernet. 5) Set of 2 USB 3.0 ports. 6) Set of 2 USB 2.0 ports. 7) Stereo audio port (4-pole). 8) MIPI-CSI camera port (2-lane). 9) Set of 2 Micro HDMI ports supported up to 4kp60. 10) Bluetooth 5.0, 2.4 GHz and 5.0 GHz IEEE 802.11b/g/n/ac wireless LAN. 11) MIPI-CSI display port (2-lane). 12) Micro SD card is placed under the display port. 13) USB C power port (5V/3A).

Fig. 5

BioSense configuration workflow.

Fig. 6

Examples of BioSense prototypes deployed in the field at the Brackenridge Field Laboratory, Texas (panel A), and at the Balcones Canyonlands Preserve, Texas (panel B). Note that the radiation shield in panel A can be modified to accommodate more than three protective discs, if needed.

Fig. 7

Blue Jay vocalization spectrogram extracted from the BioSense recording with the highest confidence level (0.31) as provided by BirdNET analysis (panel A). Blue Jay vocalization spectrogram obtained with a separate recording unit while visually confirming the species sighting nearby the microphone for comparison purposes (panel B). The spectrogram of the calls inside the white rectangles in panel A are nearly identical to the Blue Jay signature calls plotted inside the white rectangles in panel B. This result represents a simple validation of the BirdNET analysis implemented through a Raspberry Pi platform which we embedded into the BioSense node configuration. While a thorough investigation of avian species occurrence at the test site is beyond the scope of this manuscript, this illustration provides evidence that at least for common species such as the Blue Jay, BioSense can perform a neural network classification analysis and detect the correct species. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 8

Environmental data profiles for air humidity, soil moisture, and ambient temperature during laboratory testing. The capacitive measurement through the ATSAMD10 chip in one soil moisture probe (SM1) showed a sharp decrease over two consecutive timestamps corresponding to the measurements taken on September 25 at 09:20 pm and 09:30 pm. We do not have a clear explanation for these observations, but it is possible that we may have recorded a couple of faulty capacitive readings and we recommend caution with the interpretation of the results. Potential faulty readings and data outliers could be mitigated by averaging and/or filtering multiple data points. Overall, air temperature and humidity recorded by the BME280 probe ranged between 24.3 °C and 26.2 °C and 48.8 % and 55 %, respectively. These measurements are consistent to the controlled temperature and humidity laboratory condition in which we tested our BioSense prototypes.