Expected impact of ionospheric range delay on CSLC-S1

[hfattahi]

I'm starting this discussion to better understand if we need to correct for ionospheric range delay for CSLC or not.

Let's start with extracting range delay for a stack of Sentinel-1 observations over Rosamond CRs. @seongsujeong probably already has the list of date-times for S1 over ascending and descending orbits or can be queried from ASF.

The task is to :
1- download ionosphere data from IGS website. Here is some more info:

The TEC data are publicly available here: https://igs.org/products/#ionosphere_2237
The link where the data are located are here:
https://cddis.nasa.gov/archive/gnss/products/ionex/
Note that earthdata login is required to access. (if you can already download data from ASF DAAC, the same user/pass should work here as well).
Once you are in then you will notice there are folders for each year and then for each month of the year. Within each month there are multiple different files. They are different Ionosphere TEC files or from different processing centers. I usually use those from JPL. Their filename start with “jpl”. Here is an example name: “jplg0860.15i.Z”

So if the date is "20150327", then the day of the year is:
(datetime.datetime(2015,3,27)-datetime.datetime(2015,1,1)).days+1

which gives 86

so you have to download the data from the folder named 86 for that specific date.

Once the data are downloaded, you need to uncompress them:

uncompress jplg0860.15i.Z

which gives the file "jplg0860.15i"

2- Read the TEC data for each acquisition date and the lat, lon of inetrest:

Use the Iono reader in compsass:

tec_val = iono.get_ionex_value("jplg0860.15i",
                                       utc_sec
                                       , lat, lon,
                                        interp_method='linear3d',
                                        rotate_tec_map='False',
                                         )

Take a look here for more info on how to use the reader.

3- when you read the TEC data for all dates and over the lat/lon of interest, then I expect that we will have a vector (time-series) of the TEC values. Now we need to convert them to range delay.

TECU=1e16
K=40.31
f: S1 center frequency
slant_range_delay = KtecTECU/f**2/cos(incidence angle)

4- once we have the first data for slant_range_delay, then we can discuss the ground range and impact on North and East.

Note: the analysis should be done separately for ascending and descending.

@taliboliver, @vbrancat , @seongsujeong @scottstanie