Conductivity from Sentinel-1 Image

Hello,
I want to find the conductivity values of each pixel from the Sentinel-1 image. I have already completed the processing steps and i want to know how to find the conductivity values of each pixel?
Thank you

1 Like

Can you please explain a bit more what you mean with conductivity?
SAR backscatter is related to the conductivity of a medium, but also to the roughness of a surface, its orientation, shape, ect. Conductivity is not stored as a separate information, it is rather a part of the overall signal.

SAR polarimetry deals with the decomposition of signals into their physical components, e.g. the contribution of volume scattering, but it requires fully-polarimetric data (HH HV VH VV) while Sentinel 1 is dual-pol only (VV VH).

2 Likes

@ABraun helloo, thank you for the quick reply.
Firstly i want to tell Im new to SAR imagery works. In my research, I want to find the electrical conductivity value of the ground directly from sentinel-1 image for a selected area. Meanwhile, I will do the field survey as well to compare the conductivity values obtained between Sentinel-1 image and field survey.

Now i want to know is it possible to extract the electric conductivity values from Sentinel-1 image after done the image processing work?

Thank you so much.

1 Like

This is a pretty challenging task for a beginner :slight_smile:

Although there is a linear relationship between conductivity and and SAR backscatter, you would have to remove all other contributions to the signal.

If you had two absolute identical areas (same soil type, same roughness, same vegetation cover,…) and one had a higher conductivity than the others you would see a difference in backscatter intensity between them. But as no area is like the other and SAR data additionally contains speckle, you can never know if the backscatter variation results from moisture, roughness, vegetation by looking at the image.

It is good to collect field data on conductivity, so you can relate your measurements to the pixel values, but this will require a well-designed systematic data collection over multiple dates to reduce the impact variations of the other contributing factors (roughness). If there is vegetation in your study area, this is almost impossible, because the backscatter is not coming from the ground but from the leaves and branches.

I see many similarities to the retrieval of soil moisture which has the same challenge: You might want to have a look at these topics, especially johngan’s comments:

1 Like

@ABraun thank you sir I will check them. Yea im planing to do the field survey and image processing in the same location. Also the area will not cover with vegetation as well. Its a plain ground area. If you found anything about finding conductivity of the sentinel-1 image please let me know.
Thank you.

I find nothing in the scientific literature on this specific topic. Ground conductivity is usually mentioned together with ground penetrating radar. Of course, soil moisture changes conductivity as Andreas pointed out. Soil moisture retrieval is only now being solved for SAR, after decades of work. So definitely not something a beginner could do alone.

https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q="ground+conductivity"+radar+SAR&btnG=

2 Likes

Thank you for the reply. Yeah thats true. Also I have found this thread in ESA website as well.

The dielectric constant of the materials on the ground also play a factor in the measurements. In the microwave region of the spectrum, most natural materials have a dielectric constant in the range of 3 to 8 when dry, whereas water has a dielectric constant of approximately 80. The dielectric constant indicates the reflectivity and conductivity of materials. Therefore, the presence of moisture in soil and vegetation results in significant reflectivity.

But I don’t know how to find dielectric constant as well but it may have relationship with conductivity as it mention. If you found something please add it here and Im really appreciate the help you giving.
Thank you so much.

1 Like