Research

My research examines how environmental datasets and modelling choices influence the signals we detect in the terrestrial water cycle. The unifying aim is to make hydroclimate evidence more transparent, comparable, and reproducible.

Hydroclimate dataset evaluation

Global and regional studies often rely on gridded products that differ in their observations, models, assumptions, and spatial coverage. I compare these products to identify where they agree, where they diverge, and how those patterns vary among climatic and environmental settings.

This work includes:

• multi-product comparison of precipitation and evapotranspiration;
• spatial analysis of dataset agreement and disagreement;
• evaluation of trends and long-term hydroclimatic change; and
• interpretation of differences in relation to dataset construction and environmental context.

Evapotranspiration

Evapotranspiration links the water, energy, and carbon cycles, yet it cannot be observed directly at the scales required for global assessment. My work investigates estimates from multiple sources and the methodological choices behind them.

Current themes include terrestrial evapotranspiration trends, topology-based comparison of trend behaviour, exploratory analysis of evapotranspiration products, and the effects of potential evapotranspiration method selection on hydrological conclusions. Recent work introduces a topology framework for classifying agreement, opposition, and dataset influence across global ET trend products.

Uncertainty and method sensitivity

Dataset choice is part of the scientific result. I assess uncertainty arising from alternative input products, estimation methods, and analytical decisions rather than treating a single dataset as a neutral reference.

The goal is not only to measure spread, but to understand its structure: where uncertainty is concentrated, which conclusions are robust, and which depend on a particular methodological pathway.

Environmental modelling and soil physics

My broader research background includes coupled water and heat transport, bare-soil evaporation, preferential flow, and numerical modelling of processes in the soil–atmosphere system. This process-based perspective informs my interpretation of large-scale data products.

Reproducible environmental research

I use reproducible, code-based workflows to make analyses auditable and reusable. This includes R packages and applications, version-controlled analysis, documented data products, and publication-linked repositories.

Typical methods include spatial and temporal data analysis, statistical validation, sensitivity analysis, uncertainty visualization, and workflow design for large environmental datasets.