Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

A practical guide for developing your SDS320 spatial data science project

Why this handbook exists

SDS320 is organised around an independent spatial data science project. The project is the main structure through which you apply SDS concepts, Python workflows, GeoAI methods, evaluation, documentation and communication.

This handbook helps you move from a broad idea to a feasible project. It supports the decisions you need to make during the semester: What is my question? Which data can answer it? Which method fits the data and question? What should I implement first? How can I document my workflow so that another student can understand and rerun it?

The handbook is practical rather than encyclopaedic. It helps you to apply assessment details to your own project.


How to use this handbook

You do not need to read every page at once. Use the handbook when a specific project decision becomes relevant.

At the beginning of the semester, focus on the pages about the project lifecycle, milestones and research questions. During implementation, return to workflow design, data-method fit and reproducibility. Toward the end, use the repository, figures, report and presentation pages.

SDS320 projects are iterative. You may return to an earlier page when your data are harder to access than expected, when a method does not work, or when feedback shows that your question needs to be narrower.


Project pathway

Use the pages in this section as a project pathway:

  1. Project lifecycle — understand the full process from first idea to final project.

  2. Milestones — translate the semester schedule into project checkpoints.

  3. Research question — turn a broad topic into a feasible spatial question.

  4. Workflow design — plan the steps from raw data to interpretable results.

  5. Data, methods and scope — check whether your question, data and method fit together.

  6. Reproducibility — make your project understandable and rerunnable.

  7. Repository — organise the assessed public project repository.

  8. Figures and maps — create visuals that support your project argument.

  9. Report writing — write a concise report that explains what you did and what you found.

  10. Presentation guide — prepare the concept and final presentations.


What makes a good SDS320 project?

A good SDS320 project starts with a feasible spatial question. It uses suitable data, applies a method that fits the question, produces a meaningful output and explains the result honestly.

A strong project does not need to use the most complex method. It should show that you can make informed decisions, implement a clear workflow, evaluate outputs, document limitations and communicate what the result means.

Good projects usually have:


Key message

Treat your SDS320 project as something you develop continuously. You should test ideas early, document decisions while you work, and revise your project when evidence shows that something is too broad, too fragile or not well aligned. The project should not be treated as something that can be completed shortly before the deadline.