Skip to content

Latest commit

 

History

History
111 lines (78 loc) · 4 KB

Horizons_manual.md

File metadata and controls

111 lines (78 loc) · 4 KB
Raw

How to get satellite ephemeris using JPL Horizons service?

The JPL Horizons service allows retrieving highly accurate positions of satellites. We need these positions to calculate how microlensing parallax affects event magnification seen by the satellite. This file explains how to access Horizons via e-mail.

The MulensModel package contains short template e-mail:

documents/horizons_short_template.txt

This file contains all the basic settings that Horizons requires. There are only a few settings that one will modify on normal basis. These are:

Satellite choice

This is set by keyword COMMAND and has a numerical value. There are many natural and many artificial Solar System objects, but we are interested in just a few. Let's have a look at this list Some particular choices:

  • Hubble Space Telescope -48
  • Spitzer Space Telescope -79
  • New Horizons -98
  • EPOXI -140
  • Rosetta -226
  • Kepler/K2 -227

Oops... Gaia is not there! And yes, Gaia is observing the microlensing events. In that case let's go to another website. By querying it, we can find:

  • Gaia -139479
  • Swift -128485
  • Euclid -680

Note that all the values are negative and as long as we don't start observing microlensing events from other planets or Moon, we will only use negative values for keyword COMMAND

Epochs selection

You can select epochs for which you want calculations to be performed in two ways:

  1. by selecting START_TIME, STOP_TIME, and STEP_SIZE, or
  2. by providing a list of epochs via TLIST keyword.

The first one should be obvious. If you want to use the second one, then see the manual linked below. These epochs don't have to be exactly the same as the epochs in which you have observations. If they don't match, then they'll be interpolated. However, keep in mind that the time step used for interpolation cannot be too short -- if you're using data from satellite on a low Earth orbit, then the step size cannot be 1 day! It is a good practice to extend epochs range to be wider than your data. This allows one to make nice plots that contain both ground-based and space-based data.

Expected output

The QUANTITIES keyword is set to '1,20,21' and this translates to, respectively:

  • 'Astrometric RA & DEC'
  • 'Obsrv range & rng rate'
  • 'One-Way Light-Time'

These are the quantities required by MulensModel. If you need more quantities for whatever reason, then add the additional ones at the end of the list. There are currently 39 other quantities you may ask for. If you wish, there are even more options available if you change TABLE_TYPE keyword to something other than 'OBS', but we're not discussing it here as MulensModel will not read those.

If you need more detailed description, see:

ftp://ssd.jpl.nasa.gov/pub/ssd/horizons_batch_example.brief

or

ftp://ssd.jpl.nasa.gov/pub/ssd/horizons_batch_example.long

Once you are done with all the settings, you're ready to send an e-mail. Just open your e-mail program and:

  • type [email protected] in To field of the e-mail,
  • type JOB in Subject field,
  • copy-paste the settings prepared above into the main body of the e-mail,
  • make sure that your e-mail is in plain text mode (if you send html message, then the Horizons won't be able to process it!),
  • send an e-mail,
  • wait for reply and save it once it arrives,
  • if you're happy with what you got, switch your e-mail program back to html mode, rich text mode etc.

Congratulations! You now know how to access Horizons via e-mail.

If you send default or only slightly modified settings, then you can compare the results with mine, which are here:

data/ephemeris_files/Spitzer_ephemeris_01.dat

(C) R. Poleski, Mar 2017, last changes: Aug 2023