Sentinel-2 Burned Area Analysis

../_images/tuto_burned-area_icon.png

Sentinel-2 Burned Area Analysis

This service takes as input a pair (pre-event and post-event) of Sentinel-2 MSI L2A products and generates a Burned Area Analysis map and NIR/SVWI RGB composites for the two input Sentinel-2 acquisitions.
This service gives the best results when the post fire image is very close to the fire date.
The masks for the water masks are applied by the service.

EO sources supported

  • Sentinel-2 MSI L2A

Output specifications

The service provides the following output product.

Output – Burned Area Analysis map
Correspondent file Burned Area Analysis map
Raster format GeoTIFF
resolution Native
Projection types EPSG:3857 - WGS84 – Pseudo Mercator
Output Filename example Burned area analysis (2020-10-05T07:37:49.0240000Z/2020-10-15T07:39:09.0240000Z)
Output – NIR/SVWI RGB composite of the pre-event input
Correspondent file NIR/SVWI RGB composite of the pre-event input
Raster format GeoTIFF
resolution Native
Projection types EPSG:3857 - WGS84 – Pseudo Mercator
Output Filename example NIR/SVWI RGB composite (2019-08-15T14:01:01.0240000Z/2019-08-15T14:01:01.0240000Z)
Output – NIR/SVWI RGB composite of the post-event input
Correspondent file NIR/SVWI RGB composite of the post-event input
Raster format GeoTIFF
resolution Native
Projection types EPSG:3857 - WGS84 – Pseudo Mercator
Output Filename example NIR/SVWI RGB composite (2019-08-30T14:00:59.0240000Z/2019-08-30T14:00:59.0240000Z)
Output – notebook used for generating dNBR and RBR
Correspondent file notebook used for generating dNBR and RBR
File format python file (ipynb)
Output Filename example Reproducibility notebook used for generating dNBR and RBR
Output – stage-in notebook for Sentinel-2 data for generating dNBR and RBR
Correspondent file notebook for Sentinel-2 data for generating dNBR and RBR
File format python file (ipynb)
Output Filename example Reproducibility stage-in notebook for Sentinel-2 data for generating dNBR and RBR

This tutorial will describe the processing of a pair of Sentinel-2 MSI images to generate a Burned Area Analysis map and NIR/SVWI RGB composites for the two input Sentinel-2 acquisitions on the GEP.
The Burned Area map is obtained through the interesection between AOI and the pre-fire, post-fire Sentinel-2 choosen as input products.
The formula used to estimate burn severity is the Normalized Band Ratio (NBR):

../_images/tuto_burned-area_0.png

A high NBR value indicates healthy vegetation while a low value indicates bare ground and recently burnt areas. Non-burnt areas are normally attributed to values close to zero. The Burned Area Analysis product output of the service provides the delta Normalized Burn Ratio (dNBR) and the Relativized Burn Ratio (RBR). The dNBR (the difference between pre-fire and post-fire) is useful to identify recently burned areas and differentiate them from bare soil and other non-vegetated areas. The RBR is advantageous when the absolute change between pre-fire and post-fire NBR is small.

Select the processing

  • Login to the platform (see user section)
  • Select the processing service “Sentinel-2 Burned Area Analysis”:
../_images/tuto_burned-area_1.png

The “Sentinel-2 Burned Area Analysis” panel is displayed with parameters values to be filled-in.

Fill the parameters

Pre-event product reference

  • Select the Sentinel-2 data collection in the EO Data button.
  • Select the area for which you want to do an anlysis, e.g Corumba in Brasil.
../_images/burned_area-2.png
  • Click on the lens icon and select S2MSI2A as Product Type in the Search Panel
  • Apply the date value 2019-08-15 in both time:start and time:end fields
../_images/burned_area-3.png
  • Drag and Drop the selected item in the first Input reference field:
../_images/burned_area-4.png

Note

pre-event input can be picked up directly by using the following text filter: S2A_MSIL2A_20190815T140101_N0213_R067_T21KUU_20190815T214633

Post-event product reference

Pick one of the results having the same track, then drag and drop one of the results in the Input reference field:

../_images/burned_area-5.png

Note

post-event input can be picked up directly by using the following text filter: S2B_MSIL2A_20190830T140059_N0213_R067_T21KUU_20190830T180923

Area Of Interest in WKT

  • Click on the Magic tool wizard and select AOI. The input parameter is automatically filled with the WKT representing the area selected.
../_images/burned_area-6.png

Note

you can also specify manually a different AOI in WKT format, or draw a new area on the map using the search tool and get its value from the Magic tool wizard.

Run the job

  • Click on the button Run Job and see the Running Job
../_images/burned_area-7.png
  • After about 20 minutes, see the Successful Job

Results: download and visualization

  • Click on the button Show results
  • See the result on map:
../_images/burned_area-8.png

  • The following files are produced:

    • Burned area analysis (2019-08-15T14:01:01.0240000Z/2019-08-30T14:00:59.0240000Z): Burned area analysis map
    • NIR/SVWI RGB composite (2019-08-15T14:01:01.0240000Z/2019-08-15T14:01:01.0240000Z): NIR/SVWI RGB composite of the pre-event input
    • NIR/SVWI RGB composite (2019-08-30T14:00:59.0240000Z/2019-08-30T14:00:59.0240000Z): NIR/SVWI RGB composite of the post-event input
    • Reproducibility notebook used for generating dNBR and RBR: notebook used for generating dNBR and RBR
    • Reproducibility stage-in notebook for Sentinel-2 data for generating dNBR and RBR: notebook for Sentinel-2 data for generating dNBR and RBR

Reference

  • Parks, S. A., Dillon, G. K. & Miller, C. A New Metric for Quantifying Burn Severity: The Relativized Burn Ratio. Remote Sens. 6, 1827–1844 (2014)
  • Keeley, J. E. Fire intensity, fire severity and burn severity: a brief review and suggested usage. Int. J. Wildland Fire 18, 116–126 (2009)

Further reading

  • Normalized Burn Ratio by Humbold State University - link.
  • UN-SPYDER Knowledge Portal – Normalized Burn Ratio - link.