Visualization of Normalized Satellite indexes using Sentinel-2 data

The aim of the project is to familiarize students with calculating Normalized satellite indexes, electromagnetic spectrum bands, visualization of map data. Besides the development of software, students should also work on a literature review of the topic and prepare a demonstration of Normalized Satelite Indexes (NSIs) for their final presentation.

The result of the project should be an interface that could demonstrate differences between NSIs. The interface could be graphical (GUI) or at least command line-based (CLI) capable of generating adequate visualizations.

Input data of the project should be data downloaded from any of the services that provide the Sentinel-2 dataset. Usage of other satellite missions is allowed if students present sufficient arguments for the change in the assignment. The toy dataset created in this project might contain a combination of data from several missions – for instance, Sentinel-2 primarily, and for situations when Sentinel-2 data are not available, MODIS.

Some steps of the work

The project should have roughly the following parts:

  1. Review of existing normalized satellite indexes/spectral indices
    • Mapping the satellite indexes to Sentinel-2 bands or bands of the products used to construct the toy dataset
    • Table of used normalized satellite indexes/spectral indexes with references to literature, notes, examples, etc.
    • Approximately 5-page review.
  2. Construction of a toy dataset that includes samples for different times of the year at the same location, different geographical locations, known fire areas, etc. Students should develop software, which handles the downloading of the data, for instance from the Copernicus Open Access Hub.
  3. Creating or adapting a user interface capable of displaying the maps and raster overlays that supports quick “switching” and comparison of the various satellite indexes. The user interface should use adequate colormap for each of the used satellite indexes. Ideally, all data in the toy dataset should be shown – in a similar way to the Google Maps’ satellite view.
  4. Preparation of a presentation of the developed software, including demonstrations of the satellite normalized indexes on the data from the toy dataset constructed in this project. There should be a suitable example for each of the indices calculated in this project.

Computational resources

A remote computer for long-running code can be provided to students after discussion.

Semester 2020/2021

This project is assigned to Gabriel Gorta and Michal Marcinčin.

Links/Resources

NDWI (Normalized Difference Water Index)

  • Water value is positive
  • McFeeters S.K. (1996)
    • Monitor changes related to water content in water bodies
    • NDWI = (G – NIR) / (G + NIR)
    • In Sentinel-2:
      • NDWI = (B03 – B08) / (B03 + B08)
    • S. K. McFEETERS (1996) The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features, International Journal of Remote Sensing, 17:7, 1425-1432, DOI: 10.1080/01431169608948714
  • B.C. Gao. (1996)
  • Xu H. (2006) – “Modified Normalized Difference Water Index”
    • MNDWI = (Green − MIR)/(Green + MIR)
    • MIR – Mid-Infrared, 300–5000 nm
    • Sentinel-2: B11 and B12 are in MIR
    • Hanqiu Xu (2006) Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery, International Journal of Remote Sensing, 27:14, 3025-3033, DOI: 10.1080/01431160600589179
    • MIR: LANDSAT TM band 5 ≈ Sentinel-2 band 11

Landsat Surface Reflectance-Derived Spectral Indices

6 Spectral Indexes To Make Vegetation Analysis Complete

https://eos.com/blog/6-spectral-indexes-on-top-of-ndvi-to-make-your-vegetation-analysis-complete 

  •  The Soil Adjusted Vegetation Index (SAVI)
  • soil adjustment factor L to the equation of NDVI in order to correct for soil noise effects
  • https://doi.org/10.1016/0034-4257(88)90106-X 
  • Atmospherically Resistant Vegetation Index (ARVI)
  • notrelatively prone to atmospheric factors (such as aerosol)
  • NDVI corrected for atmospheric scattering effects in the red reflectance spectrum by using the measurements in blue wavelengths
  • more robust to topographic effects
    • A highly effective monitoring tool for tropical mountainous regions
  • ARVI = (NIR – (2 * Red) + Blue) / (NIR + (2 * Red) + Blue)
  • Enhanced Vegetation Index (EVI)
    • EVI = 2.5 * ((NIR – Red) / ((NIR) + (C1 * Red) – (C2 * Blue) + L))
    • Aim to correct NDVI results for atmospheric influences and soil background signals
    • coefficients 
      • C1 and C2 to correct for aerosol scattering present in the atmosphere, 
      • L to adjust for soil and canopy background
      • NASA’s MODIS:  C1=6, C2=7.5, and L=1
  • Green Chlorophyll Index (GCI)
    • GCI = (NIR) / (Green) – 1
    • content of leaf chlorophyll in various species of plants
      • reflects the physiological state of vegetation
      • decreases in stressed plants
  • Structure Insensitive Pigment Index (SIPI)
    • SIPI = (NIR – Blue) / (NIR – Red)
    • analysis of vegetation with the variable canopy structure
    • Normalized Burn Ratio (NBR)
    • Highlight burned areas following the fire
    • Healthy vegetation shows high reflectance in NIR spectrum
    • recently burned areas of vegetation reflect highly in the SWIR spectrum
    • A common practice to assess burn extent and severity with the relative differenced NBR

References