A lot of my work over the years has lived quietly in private repositories, institutional projects, and old research folders. That is often the nature of applied environmental and data science work: the analysis is real, substantial, and useful, but not always visible. I have wanted for some time to change that by sharing more of the thinking and methodology behind the work in a way that is public, practical, and reusable.
I recently published a public GitHub repository based on the core analytical themes of my PhD research in climate variability, vegetation dynamics, and remote sensing. You can view the repository here: https://github.com/clairedavis-reddy/phd-vegetation-climate-r
Rather than uploading older scripts exactly as they were, I rebuilt the work into a set of clean, accessible R notebooks that reflect the analytical logic behind the research in a way that is easier to follow, run, and adapt.
The repository focuses on three connected workflows: climate trend analysis, vegetation phenology, and long-term spatial vegetation change. Together, they reflect the broader question that shaped much of my PhD: how do we detect meaningful ecological change over time, and how do we interpret vegetation signals in relation to climate?
A key part of this process was deciding not to publish raw legacy code unchanged. Older research workflows often contain hard-coded paths, project-specific file structures, and package dependencies that do not translate well into a public repository. Instead, I built lightweight, reproducible examples using small sample datasets. The aim was to preserve the analytical thinking while making the work more transparent and useful for others.
The notebooks are designed to be practical as well as readable. They show how environmental time series can be explored, how seasonal vegetation signals can be interpreted, and how spatial patterns of change can be summarised in a reproducible format. While the repository is rooted in my PhD, it is also intended to be useful more broadly: as a teaching resource, as a starting point for similar analyses, and as a public example of how environmental data workflows can be communicated clearly.
For me, this repository is about more than sharing code. It is about making the analytical approach visible. It shows how I think about linking environmental questions to data, how I structure workflows, and how I translate technical analysis into something scientifically grounded and practically useful. That same thread runs through much of my wider work in geospatial analysis, climate information systems, and applied data science.
In a field where valuable analysis often remains buried in reports, project folders, or private repositories, there is real value in making methods more open and reusable. For me, publishing this repository is a way of making that work visible again. It is a clear and honest reflection of how I work, what I value in environmental data science, and the kind of analysis I hope others can build on.