[Gsoc-orga] GSoC 2025 Project Proposal: Analyzing Urban Expansion vs. Agricultural Land Loss using Historical OSM Data

[Ansuman Mahapatro]

Dear OpenStreetMap GSoC Organizers,

My name is Ansuman and I am currently pursuing the MITx MicroMasters
program in Data Science on edX, building upon my Bachelor's degree in
Mechanical Engineering. This background has instilled in me a strong
analytical mindset and a passion for data-driven solutions.

While I am in the process of developing expertise in areas such as
geospatial analysis and working with OpenStreetMap data, the potential of
this project – "Analyzing Urban Expansion vs. Agricultural Land Loss using
Historical OSM Data" – deeply resonates with my interests in sustainable
development and leveraging data for societal good. I am particularly
inspired by the possibility of using OSM's rich historical information to
understand and visualize critical land-use changes.

Although my direct experience with the specific technical skills required
for this project (such as in-depth use of the Overpass API, Osmium, and
geospatial libraries like GeoPandas/Shapely) is currently limited, I am
actively learning these technologies as part of my data science journey. I
am a dedicated and quick learner, and I am confident in my ability to
acquire the necessary skills with the guidance of an experienced mentor. I
am eager to contribute to this project and believe that with mentorship, I
can make meaningful progress in developing the proposed tool and analysis.
My enthusiasm for this topic and my commitment to learning make me a
motivated candidate for this project.

Project Idea:

Urban growth often leads to the loss of valuable agricultural land,
impacting food security and environmental sustainability. This project aims
to develop a tool to quantify and visualize this trend using
OpenStreetMap's historical data.

The proposed solution involves:


   - Data Acquisition: Developing methods to extract historical landuse
   data (e.g., using the Overpass API with date filters or processing
   historical OSM data extracts) for a specified geographic region and time
   range. This will involve querying and filtering OSM data based on tags like
   landuse=residential, landuse=farmland, etc.
   - Spatial Analysis: Implementing functions to calculate the area covered
   by urban and agricultural land uses at different time points, identify
   areas of land-use change (agricultural to urban), and compute summary
   statistics (e.g., total agricultural land lost, rate of urban expansion).
   Libraries like GeoPandas and Shapely in Python would be crucial here.
   - Visualization: Creating visualizations to effectively communicate the
   results, such as maps showing land use at different times, change maps
   highlighting areas of transition, and charts illustrating trends in land
   cover.
   - Tool Development: Packaging the analysis into a reusable tool (e.g., a
   Python library with a command-line interface or potentially a web-based
   application) with clear documentation.

OSM Relevance:

This project directly leverages OpenStreetMap's core strength: its rich
historical data. By analyzing landuse changes over time, we can demonstrate
the value of OSM data for understanding and addressing critical real-world
challenges related to land management and sustainable development. The tool
developed would be a valuable asset for researchers, planners, and the OSM
community.

Technical Skills:

I have a foundational understanding of:


   - Python programming
   - Data analysis concepts and libraries like Pandas and NumPy (acquired
   through the MITx MicroMasters program)
   - Problem-solving and logical thinking (developed through my engineering
   background)

I am actively learning and eager to gain practical experience in:


   - Accessing and processing OpenStreetMap data (Overpass API, Osmium)
   - Geospatial data analysis (GeoPandas, Shapely)
   - Data visualization techniques
   - I am highly motivated to learn these skills under the guidance of a
   mentor and contribute effectively to this project.


Project Scope:

I believe this project is well-suited for a 350-hour GSoC project,
especially with the support of a mentor. I am committed to dedicating the
necessary time and effort to learn and contribute meaningfully. The
deliverables would include:


   - Progress towards a working tool (Python library/application) for
   querying and processing historical OSM landuse data.
   - Development of functions for calculating land-use statistics and
   generating visualizations (with mentor guidance).
   - Documentation of the learning process and the developed components.
   - I am excited about the possibility of contributing to the
   OpenStreetMap project and would be particularly grateful for the
   opportunity to work with a mentor who can guide me through the technical
   aspects of this project. I am eager to learn and contribute to the OSM
   community.


Thank you for your time and consideration.

Sincerely,

Ansuman Mahapatro

*P.S.* I would also like to acknowledge the helpful discussions with a
language AI assistant that aided me in formulating and refining this
project idea and proposal.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/gsoc-orga/attachments/20250406/323954cd/attachment.htm>