An interactive dashboard that models the multi-layered drivers behind Delhi's rising temperatures through the year 2100. Rather than utilising a generic flat warming projection, this engine dynamically integrates global climate data with localised land-use master plans and UN demographic forecasts to construct a hyper-local climate simulation.
⚠️ Deployment Note: This application is deployed on a free resource tier via Streamlit Community Cloud. If the application has been inactive for a few days, it may go to "sleep" to conserve server energy. If you see a sleeping state screen, simply click the "Wake up app" button, and the live interface will spin back up in roughly 30 seconds.
The architecture synthesises data from three foundational local data structures:
| Data Asset | Type | Strategic Purpose |
|---|---|---|
DelhiHistoricData19702024.csv |
Weather | Establishes the 50-year base seasonal climatology to retain monthly variations (e.g., maintaining sharp pre-monsoon heat peaks). |
DelhiLULC.xlsx - Sheet1.csv |
Land Use | Correlates chronological "Built-up" area growth (in Lakh Hectares) with rising minimum baseline temperatures to isolate local UHI sensitivity. |
Indiapop.csv |
Demographics | Houses UN India projections through 2100. Applies a dynamic, capped migration share ( |
- Language: Python 3.x
- Core Interface: Streamlit (UI Engine)
- Statistical Modeling: Scikit-Learn (Linear Regression Core Engine)
- Data Processing: Pandas & NumPy
- Data Visualization: Plotly Graph Objects (High-Contrast Presentation Layer)