Multi-criteria land suitability mapping for Kallakurichi district, Tamil Nadu — identifying optimal land use opportunities across forestation, agriculture, water harvesting, housing, industry, and solar energy.
Unused and fallow lands represent significant untapped potential for sustainable development — but their optimal use varies by terrain, ecology, infrastructure, and climate. This project maps and scores unused land across Kallakurichi district for six competing land use categories using satellite-derived LULC classification, change detection, and multi-criteria spatial analysis.
The study was commissioned by the Tamil Nadu State Planning Commission (under Auroville Consulting) and outputs were delivered to state and district government departments for evidence-based spatial planning.
| Metric | Value |
|---|---|
| Study area | Kallakurichi district, Tamil Nadu |
| Analysis period | 2019 – 2023 (5-year LULC change detection) |
| Satellite imagery | Sentinel-2 + Landsat time series |
| Land use categories assessed | 6 (forestation, agriculture, water harvesting, housing, industry, solar) |
| Ground truthing | Field validation across district |
| Deliverables | Interactive online map + detailed district report |
| Client | Tamil Nadu State Planning Commission (Government of Tamil Nadu) |
Policy impact: Results were presented to relevant state and local departments and informed district-level spatial planning decisions. An interactive online map was created for ongoing use by government stakeholders.
- Identify unused and fallow lands across Kallakurichi district using satellite-based LULC classification
- Detect land use change (2019–2023) to understand trends in land availability
- Assess multi-criteria suitability of identified lands for 6 development and conservation categories
- Ground truth findings through field validation
- Disseminate results to government departments via interactive maps and detailed reports
| Category | Key Criteria |
|---|---|
| Forestation | Ecological connectivity, slope, soil type, proximity to existing forest |
| Agriculture | Soil quality, water availability, accessibility, existing agricultural patterns |
| Water Harvesting | Watershed position, DEM (drainage accumulation), soil permeability |
| Housing | Proximity to roads and settlements, slope, flood risk, infrastructure |
| Industrial Development | Road/rail access, flat terrain, distance from sensitive ecological zones |
| Ground-mounted Solar | Solar irradiance, slope (<3%), land availability, grid proximity |
1. Sentinel-2 + Landsat imagery acquisition (2019 & 2023)
↓
2. LULC classification (supervised, multi-class)
↓
3. LULC change detection (2019 → 2023)
↓
4. Unused / fallow land identification
↓
5. Multi-criteria suitability scoring per land use category
(weighted overlay using spatial layers: DEM, soil, hydrology,
road networks, ecological constraints)
↓
6. Competing use analysis (overlapping suitability zones)
↓
7. Ground truthing and field validation
↓
8. Interactive map + district-level report generation
Kallakurichi/
├── kallakurichi/
│ ├── Kallakuruchi_analysis.ipynb # Master LULC analysis notebook
│ ├── Kallakuruchi_analysis.py
│ ├── Kallakurichi_competinguse.ipynb # Competing land use overlay analysis
│ ├── Kallakurichi_competinguse.py
│ ├── Kallakuruchi_forest.ipynb # Forestation suitability
│ ├── kallakurichi_agri.ipynb # Agriculture suitability
│ ├── kallakurichi_housing.ipynb # Housing suitability
│ ├── kallakurichi_industry.ipynb # Industrial suitability
│ ├── kallakurichi_solar.ipynb # Solar energy suitability
│ ├── kallakurichi_water.ipynb # Water harvesting suitability
│ ├── kallakuruchi_input.ipynb # Input data preparation
│ └── Input_template.ipynb # Reusable analysis template
├── requirements.txt
└── README.md
Each suitability category has a dedicated notebook with its own criteria, weights, and spatial outputs — enabling modular updates when policy priorities change.
# Create environment
conda create --name lila python=3.10
conda activate lila
# Install dependencies
conda install jupyter nbconvert
conda install --file requirements.txt -c conda-forge
pip install dask-geopandas geopandas datashaderOpen the relevant notebook for the land use category you want to analyse:
jupyter notebook kallakurichi/Kallakuruchi_analysis.ipynb # Master analysis
jupyter notebook kallakurichi/kallakurichi_solar.ipynb # Solar suitability
jupyter notebook kallakurichi/Kallakurichi_competinguse.ipynb # Competing use overlayInputs required per notebook:
- District boundary shapefile (EPSG:32644)
- Sentinel-2 / Landsat imagery stack (2019, 2023)
- Ancillary spatial layers: DEM, road network, soil map, hydrological data
| Output | Description |
|---|---|
| LULC classification maps | Land cover for 2019 and 2023 |
| Change detection map | Transitions between LULC classes (2019→2023) |
| Unused land inventory | Mapped fallow/unused parcels with area statistics |
| Suitability maps (x6) | Scored land parcels per development category |
| Competing use map | Spatial overlay showing conflicting suitability zones |
| Interactive online map | Web-deliverable for government stakeholders |
| District report | Detailed PDF report with maps, charts, and policy recommendations |
- District-level spatial planning and land governance
- Solar energy potential assessment for distributed generation
- Forestation and ecological restoration planning
- Agricultural land identification and rural development
- Evidence-based policy for climate adaptation at district scale
Athithiyan M R — Geospatial Data Scientist | Remote Sensing | Climate Analytics
This work was carried out at Auroville Consulting in collaboration with the Tamil Nadu State Planning Commission.
MIT License © 2026 Athithiyan M R