Sebastian T. Brinkmann 2022-10-13
This is the github repository for my bachelor thesis “Investigating associations between multiple greenspace exposures and symptoms of depression. Results from the Prospective Urban and Rural Epidemiology (PURE) Study”. Here I provide reproducible workflows for the analyses. Data supporting this publication is accessible via the following DOI XXX.
All code is written in R (v. 4.2) and packages used in this analysis can be installed from the 00_setup.R file.
The elevation data required downloading and pre-processing. In this workflow I have provided R scripts that describe the complete process. Processed input has also been uploaded on Zenodo for easy access.
This study used three different greenspace metrics that are used in the existing literature, (i) Normalised Difference Index (NDVI), (ii) Leaf Area Index (LAI), and (iii) Land Use Cover (LULC). Processed data has also been uploaded on Zenodo for easy access.
The Greenspace Availability Exposure Index (GAVI) measures the availability of greenspace. The final GAVI has been uploaded in this GitHub repository (GAVI.tif).
The Greenspace Visibility Exposure Index (VGVI) measures the visibility of greenspace. The final VGVI has been uploaded in this GitHub repository (VGVI.tif).
The Greenspace Accessibility Exposure Index (GACI) measures the acces of public greenspace such as parks. The final GACIhas been uploaded in this GitHub repository (GACI.tif).
Due to reassons of confidentiality the original data from the PURE study can not be made publicly available. In order to follow the statistical modelling, I have generated synthetical data. This workflow covers (i) data pre-processing, (ii) statistical modelling with GLMs, and (iii) metric specific weight calculation using BART.