GADM — Global Administrative Areas

The most widely used academic boundary dataset — clean, consistent, and freely downloadable for research.

---

Provider

University of California, Davis

Website

gadm.org

Access Level

🟢 No Login Needed

Formats Available

Shapefile, GeoJSON, GeoPackage, R SpatialPolygons

Coverage

National → District (Level 0–3)

Version

GADM v4.1 (current)

---

What Is GADM?

GADM (Global Administrative Areas) is a database of the world's administrative boundaries. For India, it provides:

Level	Admin Level	Example
Level 0	Country	India
Level 1	State / Union Territory	Maharashtra, Goa, Delhi
Level 2	District	Pune, Nashik, Nagpur
Level 3	Sub-district / Taluka	Haveli, Mulshi

Why researchers love GADM: - Consistent geometry across all countries - Works perfectly for cross-country comparisons - Each boundary comes with an attribute table of names and codes - Peer-reviewed and widely cited in academic papers

---

How to Download

Direct Download from gadm.org 🟢

1. Go to https://gadm.org/download_country.html
2. Select Country: India
3. Choose format:
4. Shapefile — for QGIS or ArcGIS
5. GeoJSON — for web development or Python
6. GeoPackage — modern single-file format (recommended)
7. Click the level you want (Level 0, 1, 2, or 3)
8. Download immediately — no account needed

Using R (for R users) 🔵

# Install the geodata package
install.packages("geodata")
library(geodata)

# Download India districts (Level 2)
india_districts <- gadm("IND", level=2, path=tempdir())
plot(india_districts)

# Access attribute data
head(india_districts@data)

Using Python + geopandas 🔵

import geopandas as gpd

# If you've downloaded the GeoJSON:
india = gpd.read_file("gadm41_IND_2.json")

# View first few rows
print(india.head())
print(f"Number of districts: {len(india)}")

# Plot
india.plot(figsize=(12,10), edgecolor='black', facecolor='lightyellow')

---

Key Attribute Fields

Field	Description	Example
GID_0	Country ISO code	IND
NAME_0	Country name	India
GID_1	State code	IND.16_1
NAME_1	State name	Maharashtra
GID_2	District code	IND.16.20_1
NAME_2	District name	Pune
GID_3	Sub-district code	IND.16.20.1_1
NAME_3	Sub-district name	Haveli

---

Licence

Academic Use Only

GADM data is free for non-commercial academic use. For commercial use (building products, apps), you need permission or must use an alternative like OSM or official government data.

---

GADM vs OSM vs Official Boundaries

	GADM	OSM	Official (Census/SOI)
Best for	Research papers	Web maps, routing	Legal/official purposes
Village level	❌	✅	✅ (hard to get)
Always up-to-date	Sometimes outdated	✅	Sometimes outdated
Free	Academic use	✅	Varies
Easy to download	✅	✅	Varies

---

✏️ Practice Exercise

Exercise 1.3 — Count Districts per State

Goal: Use GADM data to find which Indian state has the most districts.

import geopandas as gpd
import matplotlib.pyplot as plt

# Load GADM Level 2 (Districts)
india = gpd.read_file("gadm41_IND_2.json")

# Count districts per state
district_count = india.groupby('NAME_1')['NAME_2'].count().sort_values(ascending=False)

print("Top 10 states by number of districts:")
print(district_count.head(10))

# Plot bar chart
district_count.head(15).plot(kind='bar', figsize=(12,6), color='saddlebrown')
plt.title("Districts per State (Top 15)")
plt.xlabel("State")
plt.ylabel("Number of Districts")
plt.xticks(rotation=45, ha='right')
plt.tight_layout()
plt.savefig("districts_per_state.png", dpi=150)
plt.show()

Expected answer: Uttar Pradesh has the most districts (~75), followed by Madhya Pradesh (~55).

---

Next Dataset: Bhuvan Basemaps (ISRO) →