Using SPICE kernels with hermpy

In this example, we show how hermpy.net.ClientSPICE can be used to download, access, and use SPICE kernels from multiple sources.

import datetime as dt

import spiceypy as spice

hermpy.net introduces a client to handle the caching and fetching of SPICE kernels.

There are some default kernels included by default, but more can be added on the fly. If you feel your additions would make sense as permanent inclusion, please open a PR.

from hermpy.net import ClientSPICE

spice_client = ClientSPICE()

Adding kernels

Kernels can be added to the SPICE client through updating the KERNEL_LOCATIONS dictionary. KERNEL_LOCATIONS expects the format seen below, with a base url (BASE), a subdirectory (DIRECTORY), and a list of filepatterns to search for (PATTERNS). The outermost key is purely to help describe the addition and is not used internally. Character wildcards are represented as ‘?’.

Here we show an example of adding MESSENGER mission kernels, and a coordinate system kernel defined for the BepiColombo mission.

spice_client.KERNEL_LOCATIONS.update(
    {
        "MESSENGER Frames (tf)": {
            "BASE": "https://naif.jpl.nasa.gov/pub/naif/",
            "DIRECTORY": "pds/data/mess-e_v_h-spice-6-v1.0/messsp_1000/data/fk/",
            "PATTERNS": ["msgr_dyn_v600.tf"],
        },
        "MESSENGER": {
            "BASE": "https://naif.jpl.nasa.gov/pub/naif/",
            "DIRECTORY": "pds/data/mess-e_v_h-spice-6-v1.0/messsp_1000/data/spk/",
            "PATTERNS": ["msgr_??????_??????_??????_od431sc_2.bsp"],
        },
        "BepiColombo Frames": {
            "BASE": "https://naif.jpl.nasa.gov/pub/naif/",
            "DIRECTORY": "pds/pds4/bc/bc_spice/spice_kernels/fk/",
            "PATTERNS": ["bc_sci_v12.tf"],
        },
    }
)

Loading kernels with spiceypy

We open a context in which we load kernels from ClientSPICE. For more details see the ``spiceypy documentation.

with spice.KernelPool(spice_client.fetch()):
    et = spice.datetime2et(dt.datetime(2012, 6, 1))
    position, _ = spice.spkpos("MESSENGER", et, "BC_MSO_AB", "NONE", "Mercury")

    print(position)

[-1111.54790425 -4790.21635147 -2979.94219262]

Loading kernels directly wit ClientSPICE

Equivalently (and often preferably), we define a context manager within ClientSPICE which extends the above. All that happens under the hood here is that the spice client performs the fetch, passes it to spice.KernelPool(), and yields the contextmanager.

with spice_client.KernelPool():
    et = spice.datetime2et(dt.datetime(2012, 6, 1))
    position, _ = spice.spkpos("MESSENGER", et, "BC_MSO_AB", "NONE", "Mercury")

    print(position)

[-1111.54790425 -4790.21635147 -2979.94219262]