Satellite Azimuth Direction/Heading

Go to the layer manager and add the mapping tools layer. Add look direction. You will get a cyan arrow representing the look direction overlaid on the image. It should be easy enough to add another for satellite heading. I’ll add it.
The overlays are useful after your interferogram has been projected.

1 Like

Thanks for the information.

Although I can now see the look direction graphically, what the geodetic modelling community really needs is a quantitative value of the look vector orientation, which can normally be generated with a knowledge of the satellite azimuth, incidence angle and the assumption the look vector is within the vertical plane parallel to the range direction, which I’m not sure is valid for the TOPSAR mode (what we really want is the unit length look vector in North-East-Vertical co-ordinates relative to the ground).

Clearly this information is stored somewhere in the metadata in order to generate the look vector through the mapping tools, but I’m not sure how to access it. Is this also a feature that could be easily built into SNAP?


Sam W.

Ok we’ll add it to the overlay as well as to the metadata so it’s accessible from command line processing. Thanks

1 Like

Hi Iveci,

I wonder that this update has been applied or not?
Other than the yellow north arrow, It would be such a big help if SNAP could give quantitative measurement of this heading direction, especially with any post-processing work with other software. For now, i still calculate it manually, without any “official” azimuth direction comparison.

Thank you, so much.

Is there any news on this? I too would like to be able to azimuth direction or look direction from the metadata. I am after the quantitative result, not an image layer, for further processing.

Is it reasonable to use 90 degrees to the right of the azimuth direction for the look direction? I am using orbit corrected, calibrated, ellipsoid corrected L1 GRD IW data, so I think the azimuth look direction movements used to make up the SLC image are averaged out in the further processing.


Heading, azimuth direction, of the Sentinel-1 product is provided in metadata as “platformHeading”.

It is explained in Sentinel-1 Product Specification as follows

Table 6-31 Data Type – productInformationType
platformHeading: Platform heading relative to North [degrees].

Example from SNAP

Is this information what you are looking for?


How do we calculate the heading angle for ALOS-2 Level 1.1 data?
The metadata shows blank values in its place.
Any help is appreciated.

Thanks !!

did you open the ALOS-2 product with File > Import > SAR sensors > ALOS-2?

You can also insert the flight direction in the layer manager.

Yes, I have exactly done that.
Still there is no option in layer manager to add flight direction.

well, look direction (range) is perpendicular to flight direction (azimuth) :slight_smile:

Thanks !!! :slight_smile:

This is related to my previous question on calculating the satellite heading angle.
Visually we can see the true north and flight direction.
But can we get a value for the heading angle from the software? since in ALOS-2 Level 1.1 metadata there is no value

Thanks in advance.

I can imagine that it is somewhere hidden in the metadata but I don’t know exactly either, sorry.
Have you tried the search for metadata?

Yes, I have searched the metadata completely.
In the ALOS-2 SAR Product Description Manual it says ‘blank’ for platform heading angle. Also while reading the metadata in SNAP, there are no values.
So I wish to inquire if there are other means of finding the metadata from the other satellite parameters.


Hi, I am trying to find the value of the heading angle for ERS2 data. I can’t find the platformheading parameter you are talking about. Furthermore, for the modelling I am doing I would be better to obtain a file similar as the incidence angle. Is that possible ?


I’m not sure how the platform heading angle is defined but it should be equal to (or derivable from) the orbit inclination.

Hello everyone,
I am really confused about the heading angle values. Let me present an example.
I am using the following product S1A_IW_SLC__1SDV_20170221T043903_20170221T043931_015377_0193B0_C530.
The relative orbit of this product is 80 and the orbit direction is descending.
As you can see in from the scihub platform the heading value is denoted with the blue arc.

When i open the product in SNAP v6 i can see the yellow arrow that denotes the North. The direction of the yellow arrow make sense because of the way that the Sentinel-1 sensor “writes” the data. After a mirroring in the E-W direction the angle between the azimuth and the yellow arrow looks like the blue arc in the scihub printscreen. Looking at the image, I expect that the heading angle is greater than 180 degrees (assuming that clockwise direction is positive).

Let me now ask these questions.

  1. What is the difference between center_heading and center_heading2? (Is the center_heading calculated from North and the center_heading2 from South). if yes, why the difference between them is not 180 degrees? (Is it due to the curvature of the orbit of the sensor, because these values are calculated in different azimuth positions/times?)
  2. I suppose that platformHeading expresses the heading angle of the sensor. So the value of the heading value is about -166.5 degrees or 193.5 degrees. How the difference from the center_heading can be explained?
  3. Finally, lets say i want to process only a part of an image which let say its located in the second burst of the second swath. Which value should i use (In the above image each swath has a different platformHeading value) ? Is it makes sense to calculate a heading angle focused on my AOI?

Thanks in advance!

Dear all,

I realize that this is already an older topic, but since I am basically struggling with exactly the same questions as @kleok and there isn’t a definite answer yet I didn’t want to open a new thread.

Specifically, I am trying to convert LOS displacement data from a PSI-analysis into vertical and horizontal displacement. In order to do so, I need the difference in satellite heading between ascending and descending orbits. While this shouldn’t be too complicated, I got somewhat confused when having a look at the S-1 metadata.

To summarize the problem, S-1 Metadata contain centre_heading and centre_heading2 in the Abstracted_Metadata item, as well as platformHeading under Original_Product_Metadata/annotation/s1a-iw...xml/product/generalAnnotation/productInformation.
Based on educated guesses by @kleok, centre_heading and centre_heading2 seem to indicate the direction from North and South, respectively, as the difference is pretty close to 180°.

The product specification only states that platformHeading indicates the platform heading relative to North [degrees] (in azimuth direction). It can be observed that the value for platformHeading changes slightly across the three sub-swaths.

However, I am still unsure whether it is correct to use the differences between center_heading or rather the platformHeading of the respective sub-swaths for both orbits.

On a sidenote, I am slightly baffled by the fact that there doesn’t seem to be an easily accessible metadata description of S-1 data. The above mentioned Sentinel-1 Product Specification, is helpful to some extent, but does not seem to contain all information, though.


1 Like

Abstracted Metadata is constructed by SNAP, I assume to harmonise the metadata parameters of the various satellites. I don’t think centre_heading and centre_heading2 are Sentinel-1 parameters - at least they don’t show in the Product Specification doc - so they are purely SNAP parameters and you have to look at SNAP documentation/ask SNAP team to find what their meaning is.

platformHeading is a Sentinel-1 parameter, and I’d take it to be the most accurate value for the satellite azimuth heading.

It can be observed that the value for platformHeading changes slightly across the three sub-swaths.

Given that the satellite heading (relative to North) changes with azimuth time and that the IW sub-swaths have slightly different acquisition times (start/stop/mid acquisition times), my guess is that the sub-swaths have small differences in platformHeading values because they are computed at different azimuth times.
Judging by the description on how to compute platformHeading in assembled products (table 3-11 in Product Specification doc), I’d say that platformHeading is computed at mid-acquisition azimuth times, but this is just my guess.

On a sidenote, I am slightly baffled by the fact that there doesn’t seem to be an easily accessible metadata description of S-1 data. The above mentioned Sentinel-1 Product Specification , is helpful to some extent, but does not seem to contain all information, though.

Are you refering to platformHeading, or to other cases?


Thanks for the clarification, platformHeading seems to be what I was looking for.

As for the other two parameters: I didn’t realize that these were purely SNAP parameters, assuming they had to be constructed somehow from the product metadata. However, neither the Java code in the snap-engine GitHub nor the documentation from the SNAP Developer Guide provided any conclusive information on the abstract metadata section in SNAP.

Consequently, my comment regarding metadata descriptions is rather related to SNAP / Sentinel Toolboxes rather than Sentinel-1 itself. I think that a documentation of the SNAP metadata parameters would be helpful, as this could save both time in looking for plausible answers and making assumptions on what they could possibly mean.

1 Like