Sentinel-2 Swath identifiers?

Optical Toolbox
#1

Is there a standard way of referring to a repeat-swath of a Sentinel-2 MSI acquisition?

For Landsat-8 there is an implicit repeat-swath identifier in the WRS-2 orbital path number. I can’t see a mention of such an identifier in the Sentinel-2 Handbook

Get relative orbits for a given UTM tile?
#3

@gbrelstaff

Sentinel-2 has Relative Orbits. These Relative Orbits (there are 143 of them) provide the 10-day repeat cycle of the mission acquisition. There is a KML on the Mission Guide S2 Relative Orbit KML available on Sentinel Online that identifies the postion of each Cycle on the surface of the Earth. IN each Tile, the Relaive Orbitr can be deduced from the component RXXX in the filename, where XXX is a number between 001 and 143. Information on the naming convention used, and the components within it, can be found here:

https://sentinel.esa.int/web/sentinel/user-guides/sentinel-2-msi/naming-convention

Cheers

Jan

S2 MPC/CC Operations Manager

#4

Thanks Jan

I will adopt the XXX notation.

#5

Just checking Jan

Can you confirm that the XXX numbers are not ordinally mapped in the same way as Landsat’s WRS-2 paths?

For my example Sardinia - the island is covered by two swaths in either case, but where for:

  • WRS-2 the two paths are labelled 192 and 193

  • XXX the two relative orbits are labelled 022 and 065

with the XXX thus not mapping ordinal/incremental around the globe.

#6

@gbrelstaff

Hi,

No. The number of the Relative Orbit is sequential within the 143 Orbit Cycle of the SENTINEL-2 mission. Thus, Relative Orbits 021 to 024 process westwards:

image
image.jpg841×346 173 KB

Cheers

Jan

#7

Thanks Jan

Since clearly there are going to be other orbits (at other dates) covering the gaps between R022 and R023 on your map then I think you’ve clarified things for me:

  • Sentinel-2 relative orbits numbers are ordinal in time (of acquisition)

  • Landsat’s are ordinal in space (e.g longitude at equator) rather than time

1 Like