MSAVI
Modified Soil Adjusted Vegetation Index. An improvement over SAVI that automatically adjusts the soil brightness correction factor, reducing the need to manually estimate the L parameter for different soil conditions.
Formula
MSAVI = (2 × NIR + 1 - √((2 × NIR + 1)² - 8 × (NIR - Red))) / 2Overview
The Modified Soil Adjusted Vegetation Index (MSAVI) eliminates the need to manually specify SAVI's L factor by automatically adjusting soil correction per pixel. Developed by Qi et al. in 1994, it solves SAVI's circular logic problem and is most effective in areas with very sparse vegetation.
How It Works
MSAVI = (2 × NIR + 1 − √((2 × NIR + 1)² − 8 × (NIR − Red))) / 2. The formula makes L increase as vegetation decreases — exactly the behavior needed. When vegetation is dense, it converges toward NDVI. When sparse, maximum soil suppression is applied. The nonlinear formula increases dynamic range at low cover densities.
Key Facts
- No user-specified L factor needed — soil correction adapts automatically per pixel.
- Outperforms SAVI and NDVI in areas with less than 15% vegetation cover.
- Developed by Qi et al. in 1994 as a self-adjusting alternative to SAVI.
- Sentinel-2 bands: B8 (NIR) and B4 (Red).
Applications
Arid Vegetation Monitoring
Detecting low-density shrubs, grasses, and biological soil crusts in desert environments.
Post-Disturbance Recovery
Monitoring vegetation regrowth after fire, mining, or construction.
Early-Season Crop Emergence
Assessing seedling establishment when most of the field is visible soil.
Limitations & Considerations
Saturates at high LAI values like other red-NIR indices. Performance over water, snow, or highly reflective surfaces may be unpredictable. Atmospheric correction still required for temporal comparisons.
History & Background
Published in 1994 in Remote Sensing of Environment. The MSAVI2 formulation with the inductive L function became the standard version. Supported in USGS documentation, ESA SNAP, and Google Earth Engine.
Analyze MSAVI data with LYRASENSE
Use our agentic notebook environment to work with satellite data and compute indices like MSAVI — no setup required.