The role
We invite applications from data scientists/quantitative geographers/demographers for an 36-month postdoctoral research position (with a possibility to extend) to support the FuturePop project. This project aims to generate high resolution gridded population estimates globally for future scenarios until 2100, with estimates further disaggregated by age and sex. In addition, uncertainty estimates will also be provided. The data will support a spectrum of fields, primarily health and natural hazard applications. The applicant will work closely with the WorldPop group at the University of Southampton, and will work within a wider team to develop methods and generate the new data. We encourage applications from those with and without geospatial experience, but are willing to learn.
What will you be doing?
The primary purpose of the post-holder is to provide advanced quantitative and geospatial analysis skills to:
a) support the development of new high resolution gridded population projections until 2100 for the Shared Socio-economic (SSP) scenarios,
b) produce disaggregated population projects by age and sex,
c) contribute to uncertainty estimates of population disaggregation,
d) produce opensource code to facilitate the dissemination of these methods,
e) lead and contribute to drafting key scholarly publications,
f) co-develop research and applications,
g) disseminate FuturePop outputs at meetings and conferences.
You will join the world-leading Quantitative Spatial Science (formerly Spatial Modelling) research group at the University of Bristol, the Jean Golding Institute for data science, and benefit from our institutional partnership with the Alan Turing Institute.
You should apply if
The candidate will hold a PhD (or be near completion) in a relevant field and should have extensive experience working, or are working towards furthering themselves in the majority of the following areas:
1. An interest in and passion for issues of demography and socio-economic scenarios
2. Experience in machine learning and deep learning techniques, particularly random forest and convolutional neural networks
3. Expertise in R and/or Python programming language
4. Experience of working with UNIX
5. Sound data management skills, and experience working with ‘big data’
6. Experience, or an interest in co-developing research
7. Self-motivation, initiative and organizational skills in planning and carrying out research
8. Conduct advanced research to a high standard both independently and as part of an interdisciplinary team.
9. Regularly disseminate project findings by, for example, authoring peer-reviewed journal articles in leading academic journals and presenting papers at key conferences in the field.
10. Ability to use initiative, and apply creativity, to solve problems that are encountered in the teaching and/or research context