Strange Range Values in GRACE KBR Data

Strange Range Values in GRACE KBR Data

Postby samuellow » Mon Mar 18, 2019 11:24 pm


Much thanks to Li Wen Hao who has helped me to retrieve the inter-satellite ranging data from GRACE A and B from the KBR1B data! I am working on precise orbit determination and also baseline determination algorithms using GNSS data processing. From Wikipedia, it seems that the nominal distance separation between GRACE A & B ~ 220km during normal mission operations.

To my dismay however, I found it a bit strange that the KBR data for 27-07-2010 shows that the range was ~ 330km instead. Even with the ionospheric, antenna-offset, and light time corrections, it was nowhere near the nominal 220km. I would like to ask if anyone has any idea as to whether this data is correct, or perhaps is it meant to be in some other form of units (although the L1B hand book says it is in meters).

Here is a snapshot of the KBR range values from the ASCII KBR file that Wen Hao has kindly sent to me: ... 3JTD0TwTas

Fortunately, or unfortunately, both my POD results (with header DSOLPD) and my double-differenced baseline measurement results (with header DSOLDD) show that the range between GRACE A & B was ~ 227km nominally, which seems close enough to the reported 220km formation distance for the mission. The simulations were done for the 27-07-2010 epoch, same as that for the date of KBR results. I used the RINEX observation files of the GRACE satellites A & B to do the POD and the baseline determination on Bernese GNSS software v5.2.

My simulation results, tabulated in Excel, are shown here: ... AYihGfxHNU

My question would be, why does the ASCII KBR data show 330km? Is this range really correct or am I missing something here? I am assuming it is in meters. If it is a separate unit, please kindly correct me! I don't think my algorithm is perfect, but I did found it a bit strange that two distinctly separate algorithms gave me such close values from the same RINEX file, but the KBR values were unexpected far off from ~220km. Thank you very much. I have attached the KBR ASCII file here too (thank you Wen Hao). ... -Bbc90v2kX

Once again, I appreciate the time taken by users and administrators in the PODAAC community for reading my queries. Thank you all :)
Posts: 6
Joined: Tue Feb 26, 2019 9:38 pm

Re: Strange Range Values in GRACE KBR Data

Postby podaac » Thu Mar 28, 2019 10:39 am

Hi Samuel,

Thanks for the nice note. To get to the heart of your question: we simply don’t measure absolute range with the K/Ka-Band ranging system. You can use this data to determine how the relative range of the two spacecraft changes over time, but keep in mind there is an absolute bias between them. Alternatively, you can use the GNV1B files, which have the inertial position of the two satellites determined with the onboard GPS receivers. You can use this to calculate the range – the expected accuracy in such a calculation is on the level of 1-2 cm, rather than the micrometer level of precision that the K/Ka-Band system provides. I should also mention that 220 km is a nominal separation distance. In practice, the satellites drift relative to each other. The separation distance throughout the mission is maintained to be between 170 km and 270 km. I hope this is helpful to you.



Dear Dr. David,

Thanks for getting back to me so quickly, and with a detailed answer to my queries too! May I then ask, does the 'micrometer precision' apply to the range rate then? Range rate would also be interesting to study. I would suppose the absolute range rate would be the same as the biased range rate (since the bias would have been omitted in the relative calculation I am assuming...).

I really appreciated the response considering that its getting late in Pasadena. Have a great evening~

Site Admin
Posts: 555
Joined: Mon Oct 22, 2012 4:00 pm

Return to Data Access