NDVI Index

The NDVI index is probably the most well-known of environmental indicators.

NDVI (Normalized Difference Vegetation Index) is a vegetation index based on satellite data that makes it possible to assess the condition, density, and intensity of photosynthesis in plants.

It uses differences in the reflection of electromagnetic radiation by vegetation in the red (RED) and near-infrared (NIR) ranges.

NDVI values range from –1 to 1. Low values (below 0) correspond to water, clouds, or snow; values close to zero (0–0.2) characterize non-vegetated surfaces, soils, and artificial objects; medium values (0.2–0.5) indicate sparse or stressed vegetation; and high values (0.5–1) correspond to dense and healthy vegetation.

Both free and commercial satellite data sources can be used to calculate NDVI. Sentinel-2 (Copernicus, ESA)provides imagery with a resolution of 10 meters, where the red band corresponds to channel B4 and the near-infrared band to channel B8. These data are available, among others, through the Copernicus Open Access Hub or Copernicus Browser. Landsat (NASA/USGS) provides imagery with a resolution of 30 meters, where for Landsat 8 and 9 missions the red band is B4 and the near-infrared band is B5; data can be downloaded from the USGS Earth Explorer or accessed via Google Earth Engine. MODIS (NASA) enables analyses on a global scale, although with a resolution of 250 to 1000 meters it is more suitable for climate studies than detailed analyses. On the commercial side, there are systems such as PlanetScope (Planet Labs), offering daily imagery with a resolution of 3–5 meters, WorldView (Maxar), providing images with resolutions down to below 1 meter, and SPOT (Airbus), with a resolution of 1.5 to 10 meters.

To calculate NDVI in QGIS, one must first download the appropriate images and identify the bands corresponding to the red and near-infrared ranges. In the program, the raster calculator is opened, and then the formula (NIR – RED) / (NIR + RED) is entered, replacing the names with the appropriate channels, for example (“B8” – “B4”) / (“B8” + “B4”) for Sentinel-2. The result is saved as a new raster in which the values correspond to the NDVI index. To present the map in a clear way, in the raster symbology settings one selects the “pseudocolor” rendering and assigns a color scale – from red for low values, through yellow for medium values, up to green for high values. This produces a map of the spatial distribution of vegetation, showing where vegetation is lush and healthy, and where it is sparse or absent altogether.