Sigma0, change detection confusion

Microwave Toolbox
#1

I have two sigma0 radarsat-2 images only with one month diff. Water bodies on the one in Aug does not appear as dark as that of Sept.

In summer time, no frozen water. The image also does Not look noisy! Other water bodies in Aug image look good (in black). only for these few sites it s different!

This causes confusion in change detection. Does anyone know by any chance why it’s like that?

Thanks.

comparisonSAR.jpg546×555 159 KB

#2

There can be a lot of reasons for this. Are the polarizations the same? Different polarizations wll have a different noise floor. The darker sea might have been on a calmer day while the brighter sea may have had higher winds. This is the nature of SAR imagery and why it’s good to work with many images over the same scene.

1 Like
#3

Thanks for the reply. The odd thing is that the Sept. one is the windy one while water is in dark. the Aug one is in calmer condition. Both HH. other water bodies in Aug look fine!

#4

Wind direction also affects sea backscattering. For a given wind speed, wind along range has higher returns than across range.
Do you have any in-situ measurement or picture of the lakes at the SAR acquistion times that could indicate some physical phenomenon affecting the water surface? Or valid images from optical satellites (Sentinel-2, -3, Landsat, MODIS)?

1 Like
#5

Thanks for your reply. For that exact date I do not have any optical data for now.

All other water bodies for that particular image (Aug) look fine in absolute dark color. The wind speed is 8km/hr and wind direction is 160. Only these two water bodies do not appear in dark! This shows that wind should not be an issue for that. Conversely, the other image (Sept) is a windy one!

I have checked all 20 images ascending and descending , these two water bodies look fine over there! No clue!

Thanks again