Strange data artifacts on Juno data

I have been using SpiceyPy (to great effect), to process trajectory information from the Juno Mission. I have been using the subpnt function to gather latitude, longitude and the skpos function to gather the position vectors for Juno during its orbits. For each orbit I use the meta kernel provided by the JADE team. My trajectory results appear to be generally accurate however several orbits have strange artifacts in them. These artifacts appear as large leaps in the the position vector and corresponding jumps in the latitude/longitude data.

Expected behavior The web based SPICE interface (with the same meta kernel) produces data which is similar to the SpiceyPy output but without the artifacts. (I have attached plots for comparison)

To Reproduce This is the code I’ve been using to gather the position information:

    spice.furnsh('./Kernels/PJ{}_metakernel.txt'.format(pj))

    time_str = np.array([t.iso for t in data_Table['UTC']])
    spiceyTimes = spice.str2et(time_str)

    positions, lightTimes = spice.spkpos('JUNO', spiceyTimes, 'IAU_JUPITER', 'NONE', 'JUPITER_BARYCENTER')
    lon = []
    lat = []
    for spiceyTime in tqdm(spiceyTimes):
        spoint, trgepc, srfvec = spice.subpnt('NEAR POINT: ELLIPSOID', 'JUPITER', spiceyTime, 'IAU_JUPITER', 'NONE', 'JUNO')
        srad, slon, slat = spice.reclat(spoint)
        lon.append(slon * spice.dpr())
        lat.append(slat * spice.dpr())

data_Table[‘UTC’] is a list of times with a dt of 60 seconds spanning from 2017-297/16:12:31 to 2017-297/19:12:31

And the meta kernel I have been using is:

PJ09_metakernel.txt The specific kernels referenced can be downloaded through NAIF with the urls contained In the following file

Kernel_url.txt

Desktop (please complete the following information):

• SpiceyPy Version 3.1.1
• OS: Mac OsX 10.15.5
• OS Architecture 64bit
• Python Version 3.9.0
• Python Architecture 64bit

[gabrielb09]

The web based interface is WebGeocalc its provided and maintained by NAIF. While the plots above have different aspect ratios they cover the exact same span of time (2017-297/16:12:31 to 2017-297/19:12:31) and show the same data (longitude for the two top plots and latitude for the two bottom plots).

[gabrielb09]

Thank you so much for your help, after further investigation you were right! My time array had some missing values, that will teach me to blindly trust my data!