About
PhD candidate in Statistical Ecology at the University of St Andrews, focusing on developing and applying hierarchical and simulation-based models to large, heterogeneous citizen science datasets. Currently modelling sources of observer bias in the eBird citizen science dataset as part of my PhD.
Research interests: Citizen Science, Occupancy Modelling, eBird, Statistical Ecology, R.
Research
PhD focus: Species distribution models for citizen science data.
Citizen science data is a great source of information about the world. However, in order to properly use this data source we have to account for the lack of formal survey structure. My PhD focuses on how we quantify and account for biases in this data to further understand the natural world.
Supervisors: Dr Alison Johnston, Professor David Borchers.
Projects
Presence of the Merlin Bird Identification App in the eBird Database
The use of automated bird identification apps is changing how citizen scientists collect data, with implications for ecological modelling. This project uses a customised zero-inflated beta family linear mixed model to assess how reporting rates vary across species, observers, and levels of identification experience, revealing heterogeneous effects of app use on detection data.
Quantifying and Modelling Spatiotemporal Trends in Bird Detection Modality
Using archival data from the Macaulay Library (bird photos and audio recordings), this project quantifies how observers detect species and how detection modality changes across space and time. Patterns are modelled as functions of avian phylogeny and environmental interactions, offering insights into how observer behaviour shapes biodiversity datasets.
Publications
- Backstrom, L. J., Drake, R. L., Worthington, H., & Johnston, A. (2025). Detections of rare species lead citizen scientists to initiate data recording. Diversity and Distributions, 31(10), e70103.
Skills
Technical Skills
- Programming & Data Analysis: Proficient in R (tidyverse, ggplot2, terra, luna, ubms, Stan, spOccupancy, spAbundance, ggtree); experience with Python for statistical and mathematical workflows.
- Statistical Modelling: Skilled in custom-built GLMMs, occupancy models, and simulation studies designed from real ecological data. Practical experience implementing Bayesian models using Stan.
- Spatial & Ecological Data: Experienced with large and complex datasets, including eBird (>4 billion records), the Macaulay Library, British Trust for Ornithology, and landcover databases.
- Data Management & Quality Assurance: Managing restricted-access and sensitive datasets, ensuring compliance with ethical and security requirements. Strong command of data cleaning, integration, and quality control workflows.
- Statistical Communication: Advanced data visualisation and reporting using R (ggplot2 and custom graphics).
- Software & Tools: R, Python, GitHub, Microsoft Office Suite.
Core Skills
- Interdisciplinary Collaboration
- Communication & Stakeholder Engagement
- Scientific Writing & Reporting
- Project & Time Management
- Teamwork & Independence
- Analytical Agility
- Knowledge Integration