Rapid environmental change is outpacing the capacity of many species to respond. Understanding and predicting whether populations will adapt, migrate or decline is one of the central challenges of contemporary evolutionary biology. We address this question across two complementary systems: the model plant Arabidopsis thaliana, in which we develop and validate genomic prediction frameworks for evolutionary forecasting, and invasive common ragweed (Ambrosia artemisiifolia), where adaptation to novel environments can be directly observed across continents.

A central challenge in evolutionary ecology is translating current genomic variation into predictions under future conditions whcih is critical to assess how populations will respond to climate change or novel biotic and anthropic challenges. This long-term research program, building on Prof. Johanna Schmitt's FIBR project using Arabidopsis thaliana, leverages multi-environment field experiments across a wide climatic range and seasonal conditions. By capturing the statistical relationship between genome-wide variation and fitness across contrasting environments, we model the evolution of genetic architectures to forecast how trait values and population genetic composition are expected to shift under novel conditions, without requiring organisms to be directly exposed to those conditions.

Foundational work demonstrated that natural variation in fitness across European climates is underlain by genomic regions with strong environment-specific effects, and that the adaptive significance of flowering time plasticity, as a highly heritable trait driving fitness, strongly responds to these local climates (Fournier-Level et al. 2011, Science; Fournier-Level et al. 2022, New Phytologist). Building on this framework, we generated genome-informed forecasts of trait responses for diverse plant genetic makeups sourced from across the native range, providing a practical tool for seed provenancing decisions in restoration and conservation contexts (Putra et al. 2023, Molecular Ecology Resources).

The AraCast application below lets you explore these forecasts interactively, comparing predicted trait values and evolutionary trajectories across populations and climate scenarios.

Invasive species are natural experiments in rapid evolution: founding populations are identifiable, the timescale of spread is documented, and novel selection pressures can be inferred from the environments colonised. Common ragweed (Ambrosia artemisiifolia) is a globally invasive annual plant that has spread across Europe and into Australia from a North American origin, rapidly adapting to a wide range of climates and becoming a major agricultural weed and allergen. In collaboration with Kathryn (Kay) Hodgins (Monash University) and John Stinchcombe (University of Toronto), we use ragweed's global invasion as a tractable system to ask how genomic diversity shapes invasion outcomes and whether patterns of adaptation in the native range can be used to forecast trait expression and establishment potential in invaded environments.

This programme is being developped by first integrating population genomics structure into ecological niche modelling, showing that the genomic composition of founding populations has predictable consequences for invasion potential: some source genotypes are better matched to invaded environments than others, and this can be quantified from genome-wide variation before introduction events occur (Putra et al. 2024, Evolutionary Applications). Second, by examining trait architecture across the native and invasive range, we found that the genetic basis of key functional traits is variable across populations, which places fundamental limits on how well cross-range genomic predictions can perform — a caution for any programme that seeks to extrapolate trait-based forecasts beyond the populations used for model training (Putra et al. 2026, Evolutionary Applications). Together, this work illustrates both the promise and the constraints of genome-based approaches for forecasting evolutionary dynamics in novel environments.