‹ back to GIS Glossary

EVI (Enhanced Vegetation Index)

Definition

The Enhanced Vegetation Index (EVI) is an optimized vegetation index designed to improve upon the NDVI by reducing atmospheric influences (from aerosols and thin clouds) and minimizing the background signal from soil. It achieves this by incorporating the blue reflectance band to correct for aerosol scattering in the red band and includes a soil adjustment factor and a canopy background adjustment term. This makes EVI more sensitive to changes in high biomass regions and better suited for global vegetation monitoring.

Application

EVI is crucial for monitoring vegetation in areas with dense canopy cover, such as tropical rainforests, where NDVI tends to saturate. It is a standard product in global satellite missions like MODIS and VIIRS, used for tracking seasonal and interannual vegetation dynamics, studying carbon cycles, assessing forest productivity, and monitoring large-scale agricultural systems. Its reduced atmospheric sensitivity makes it more reliable in hazy conditions.

FAQ

What is the formula for EVI and how does it differ from NDVI?

EVI is calculated as: EVI = G * (NIR - Red) / (NIR + C1 * Red - C2 * Blue + L). G is a gain factor, C1 and C2 are coefficients for atmospheric resistance, Blue is the blue band reflectance, and L is a canopy background adjustment factor. Unlike NDVI, which uses only red and NIR bands, EVI adds the blue band to correct for aerosol effects and includes coefficients to reduce soil and atmospheric noise.

Why was EVI developed, and what problem does it solve?

EVI was developed to address two key limitations of NDVI:

1) Saturation in high-biomass areas, where NDVI loses sensitivity to further increases in leaf area index;

2) Susceptibility to atmospheric scattering and soil background effects. EVI provides a more linear response across a wider range of vegetation conditions and maintains better performance in atmospherically hazy regions.

When should I use EVI instead of NDVI?

Use EVI when working in regions with dense vegetation (e.g., tropical forests), in areas prone to atmospheric haze or aerosol pollution, or for time-series analysis where atmospheric conditions vary. Use NDVI for simpler applications, in areas with low to moderate vegetation, or when consistency with long historical datasets (which primarily use NDVI) is important.

What are the main challenges or drawbacks of using EVI?

Challenges include:

1) Requiring a blue band, which is not available on all sensors;

2) Being more computationally complex than NDVI;

3) Potential sensitivity to the specific coefficients (G, C1, C2, L) used, which may need tuning for different sensors or regions;

4) Possible introduction of noise in areas with very bright surfaces (e.g., snow, deserts) due to the blue band component.