1. The questions being asked fall into two broad categories listed below and each category has multiple workstrand components. Solutions can be viewed generally as the development of mathematical and statistical tools and methodology that were motivated by specific natural resource and agricultural issues faced by Scotland including the effects of climate change on plant and animal populations, biodiversity, and ecosystem health. 

   I. Modelling natural population & community dynamics 

• WS 1.1. Mathematical models for natural pest management (A1, RQ 2& RQ3; Katharine Preedy). One aim is the development of a mechanistic, overarching ecological theory for crop-pest-natural enemy dynamics which includes cereal crops, aphids and parasitoids. A manuscript in internal review is titled “A modelling framework to estimate vector dispersal and disease spread in an important agricultural pathosystem”.
• WS 1.2. Modelling tools for spatio-temporal population and community dynamics (D4: RQ4 & RQ15; Stephen Catterall and Glenn Marion). One area of focus has been the spatio-temporal modelling of the spread of the great spruce bark beetle in Scotland and incorporating spatially explicit population genetics modelling approaches to better quantify the ability of tree populations to adapt to environmental change.
• WS 1.3. Process-based life cycle modelling (A2: RQ5; Dave Ewing and Helen Kettle). One area of focus has been on modelling the cabbage stem flea beetle life cycle.  Two papers have been published: Temperature and time of season are the predominant drivers of cabbage stem flea beetle, “Temperature and time of season are the predominant drivers of cabbage stem flea beetle” (2024) and “A process-based life cycle of the cabbage stem flea beetle on winter oilseed rape: effects of temperature” (2025).
• WS 1.4. Hierarchical spatio-temporal modelling of aphid dynamics (A1, RQ3; C5, RQ8; Ken Newman and Zhou Fang). A primary aim is develop improved aphid arrival forecasting tools for farmers and agronomist including a user-friendly app.  The models underlying these tools use the environmental forcing factors, particularly daily temperatures, without coarse temporal aggregation and borrow strength by including a hierarchical spatial framework that uses data across the UK. A manuscript under internal review is titled “Adding structure to generalized additive models, with applications in ecology”.

        II. Statistical tools for biodiversity and environment.

• WS 2.1. Modern approaches to time series data analysis: nonlinear Hidden Markov models and change points. (D1, RQ5&6; D2, RQ4; Luigi Spezia and Katherine Whyte). A primary aim is the development of data augmentation techniques for fitting Markov switching autoregressive models (MSARMs) that use high frequency data, with a second aim being the inclusion of changepoints, relevant to climate change effects. Two papers from this work are “Bayesian analysis of high-frequency water temperature time series through Markov switching autoregressive models” (2023) and “Bayesian joint longitudinal models for assessing the exploitation rates of sardine stock in the Mediterranean Sea” (2024).  
• WS 2.2. Statistical methods for biodiversity modelling (D4: RQs 1, 4, 8, 15, 17, 19, 21, 22; Nick Schurch and Anastasia Frantsuzova). Focus has been on multi-species joint distribution modelling with application to biodiversity assessment in Scotland and includes development of the multi-JSDM framework codebase and multispecies abundance simulation.