Abstract:
Insect foraging behaviour is influenced by various environmental factors; however,
continuous monitoring under natural conditions remains technically and financially
challenging. In this study, we present a cost-effective, autonomous multi-sensor
monitoring system developed to investigate the foraging behaviour of a ground
dwelling ant species in relation to temperature, humidity, and light intensity.
Leveraging a low-cost ESP32 microcontroller and a MATLAB-based optical flow
computer vision algorithm, the system enables continuous monitoring and analysis of
ant activity (ingress and egress). A comprehensive dataset was collected every 15
minutes from 23 to 29 August 2024, including environmental parameters
(temperature, relative humidity, and light intensity) and corresponding ant ingress
and egress counts. During the observation period, a 5-minute video feed for each time
interval of 15 minutes was captured. This, paired with the environmental parameter
logging system, captured data over a total duration of 168 hours (7 days), resulting in
672 discrete 15-minute intervals. Nearly 709 Carpenter ants (Family Formicidae)
were observed moving either in or out of the monitored mound per 5 minutes of each
interval. Distinct nocturnal patterns were observed, with peak foraging activity
between 0000 and 0600 hours, and reduced activity during the daytime. The
multivariate regression analysis revealed that ant activity was significantly associated
with the tested parameters (p = 0.009; R² = 0.216), where temperature (p = 0.07) and
humidity (p < 0.001) were found to be significantly influencing the ant activity;
however, the adopted model only moderately supports these variables due to the
limited number of data points. This case study reports baseline information on
affordable automated systems, which can be used in behavioural ecology. Further,
the findings enhance the understanding of the environmental factors influencing ant
foraging and suggest a methodological framework applicable to other small-bodied
terrestrial invertebrates.
