I approach my research on ecological systems through the lens of monitoring and modeling, with the objective of managing them for a desired outcome. I design and develop custom automated and AI-powered devices to monitor and understand the factors that influence the dynamics of the ecological system. I then use the understanding I acquire from the monitoring effort to develop models to represent the ecological system. The models allow me to simulate and explore different scenarios and pathways to achieve a desired outcome in the ecological system. Finally, I link my research directly to management decision-making to drive the desired outcome in the real world. My dissertation project exemplifies this approach.
The objectives of my dissertation project are;
(1) Develop a device to monitor and understand the influence of temperature,
precipitation and landscape on Osmia cornifrons population dynamics.
(2) Develop a climate-sensitive population model to simulate the influence of climate and landscape on Osmia cornifrons population dynamics across Pennsylvania between 2008 and 2023.
(3) Create a decision-support tool to help apple farmers consider and incorporate Osmia cornifrons for their pollination needs.
I've developed an automated and AI-powered monitoring system to monitor the influence of environmental factors on Osmia cornifrons. This device can also study other solitary bees in different landscapes. I adopted a modular systems design to ensure this system can be replicated and customized for different contexts. The hardware system can be broken into two modules (1) the recording and storage module and the (2) energy module. Each module works independently of each other. My study's recording and storage module consists of the following components: Raspberry Pi 4, witty Pi 4, Raspberry Pi HD camera, 3D printed weatherproof case, 256 SD cards, waterproof USB connectors, etc. The components of the energy system were 100W Renology solar panel, 10A Renology charge controller, 12V 30Ah Lithium battery, DC-DC converter, weatherproof battery case, etc.
One major bottleneck that my dissertation project has overcome and contributed to the field is that I've developed an AI-powered algorithm to analyze the video data gathered in the field. Previous monitoring studies had to rely on humans to process the video data, which limited the quantity of video data that could be analyzed to a few days [1]. The algorithm I've developed makes it possible to analyze amounts of video data like never before. With this system, we can now observe solitary bees and make new insightful discoveries about the population dynamics and how climate and landscape influence them.
Using the SolBeePop framework [2], I've put together a model to simulate the influence of climate and landscape on Osmia cornifrons' population dynamics. The model consists of two components: (1) emergence date predictions and (2) reproductive success estimation. The emergence date prediction is based on degree day models previously investigated and developed by other studies [3-5. The reproductive success estimation component is the step function, which is parameterized with the results from the monitoring effort.
Finally, I am working with my collaborators to assemble a decision-support tool that will enable farmers across PA to know when and how much pollination they can expect from the natural Osmia cornifrons population and how that aligns with the apple flowering period. The goal is that this tool will help farmers consider and incorporate solitary bees for pollination needs.
Image source (https://iconiccollection.com/solitary-bees-are-having-their-moment-in-the-sun/)
[1] McKinney, M.I. and Park, Y.L., 2012. Nesting activity and behavior of Osmia cornifrons (Hymenoptera: Megachilidae) elucidated using videography. Psyche: A Journal of Entomology, 2012(1), p.814097.
[2] Schmolke, A., Galic, N. and Hinarejos, S., 2023. SolBeePop: A model of solitary bee populations in agricultural landscapes. Journal of Applied Ecology, 60(12), pp.2573-2585.
[3] Adams, L.R., 2001. IN PENNS YLVANIA.
[4] White, J., Son, Y. and Park, Y.L., 2009. Temperature-dependent emergence of Osmia cornifrons (Hymenoptera: Megachilidae) adults. Journal of economic entomology, 102(6), pp.2026-2032.
[5] Ahn, J.J., Park, Y.L. and Jung, C., 2014. Modeling spring emergence of Osmia cornifrons Radoszkowski (Hymenoptera: Megachilidae) females in Korea. Journal of Asia-Pacific Entomology, 17(4), pp.901-905.