Monitoring vegetation health is one of the most common applications of remote sensing. One of the most widely used vegetation indices is the Normalized Difference Vegetation Index (NDVI).

In this tutorial, we demonstrate:

• How to compute NDVI manually in QGIS
• How to automate the process using a QGIS Model3 file
• Common mistakes and how to fix them

This workflow is based on Sentinel-2 Level-2A (Surface Reflectance) imagery and QGIS 3.28.

1. What is NDVI?

NDVI is calculated using two spectral bands:

• Red band (B04)
• Near Infrared band (B08)

The formula is:$$NDVI = \frac{(NIR – Red)}{(NIR + Red)}$$NDVI=(NIR+Red)(NIR−Red)​

NDVI values range between -1 and +1.

2. Preparing Sentinel-2 Data

Download Sentinel-2 Level-2A (L2A) imagery.

Inside the SAFE directory, navigate to:

GRANULE → IMG_DATA → R10m

You will find:

• B04_10m.jp2 → Red
• B08_10m.jp2 → NIR

These are the only bands required for NDVI.

3. Method 1 — Manual NDVI Calculation in QGIS

Step 1: Load Data

Drag into QGIS:

• B04_10m.jp2
• B08_10m.jp2
• (Optional) Study area polygon

Step 2: Clip to Study Area (Optional)

Go to:

Raster → Extraction → Clip raster by mask layer

Clip both bands using your AOI.

Step 3: Raster Calculator

Open:

Raster → Raster Calculator

Use this formula:

(A - B) / (A + B + 0.0001)

Where:

• A = B08 (NIR)
• B = B04 (Red)

Important settings:

• Output data type → Float32
• NoData value → -9999
• Save as NDVI.tif

Step 4: Fix Symbology

Right click NDVI layer → Properties → Symbology

• Render type → Singleband Pseudocolor
• Min = -1
• Max = 1
• Color ramp: Red → Yellow → Green

Now vegetation will appear green and water will appear negative (blue/red).

4. Method 2 — Automating NDVI Using a Model3 File

For repeated analysis (multiple dates, tiles, or study areas), automation is essential. QGIS Model Designer allows you to build reusable workflows saved as .model3 files.

Why Use a Model3 File?

• Eliminates repetitive steps
• Reduces user error
• Ensures consistent processing
• Useful for teaching and research
• Easy batch processing

Download a Ready for Use Model3 File for NDVI Estimation using Sentinel images

Creating an NDVI Model in QGIS

Step 1: Open Model Designer

Processing → Graphical Modeler

Step 2: Add Inputs

Add three inputs:

• Raster layer → B04
• Raster layer → B08
• Vector layer (Polygon) → AOI

Step 3: Add Algorithms

A. Clip B04

Algorithm:

GDAL → Clip raster by mask layer

• INPUT = B04
• MASK = AOI

B. Clip B08

Repeat same step for B08.

Step 4: Add Raster Calculator

Algorithm:

GDAL → Raster Calculator

Parameters:

• A = Clipped B08
• B = Clipped B04
• Formula: (A - B) / (A + B + 0.0001)
• Output type → Float32

Step 5: Save Model

Click:

Save Model

This creates a .model3 file.

You can now:

• Reuse it anytime
• Share it with colleagues
• Integrate into larger workflows
• Run batch NDVI calculations

5. Using an Existing Model3 File

If you already have a .model3 file:

1. Open QGIS
2. Go to: Processing Toolbox → Models → Open Existing Model
3. Select your NDVI model file
4. Run it
5. Provide:
   • B04
   • B08
   • AOI
   • Output location

The entire NDVI process runs automatically.

6. Common NDVI Problems

Problem: NDVI values are greater than 1

Cause:

• Output data type not set to Float32

Solution:

• Recalculate with Float32 output

Problem: Only water appears

Cause:

• Winter/dry season image
• Incorrect symbology stretch

Fix:

• Manually set NDVI range from -1 to 1

Problem: Strange extreme values

Solution:
Always add small constant:

+ 0.0001

7. Extracting Vegetation or Water

Vegetation mask:

NDVI > 0.2

Water mask:

NDVI < 0

These can be added as additional steps inside your Model3 file.

8. Final Thoughts

Combining Sentinel-2 imagery with QGIS provides a powerful, open-source solution for vegetation monitoring.

Manual NDVI calculation is useful for learning.
Model3 automation is essential for professional workflows.

For environmental monitoring, agriculture, hydrology, or climate studies, mastering NDVI in QGIS is a foundational ski