GUIbrush®: Characterizing the atmospheres of the new worlds with Python

Video: https://video.linux.it/w/igKJp9HopEWjsA69zsQTkd?start=57m12s&stop=2h9m45s

Vediamo un progetto basato su Python per il recupero delle proprietà atmosferiche dei pianeti extrasolari, concentrandoci sulle sfide astrofisiche e computazionali.

Dott. Paolo Giacobbe — Ricercatore presso l'Osservatorio Astrofisico di Torino

Transcript

1. GUIbrush®
   i.e.
   Characterizing the atmospheres of new
   worlds
   with Python
   Paolo Giacobbe1 & Francesco Amadori1
   (1INAF-Osservatorio Astrofisico di Torino, Pino Torinese, Italy.)
   

   View Slide

2. When it all began (?)
   On 6 October 1995, Michel
   Mayor and Didier Queloz
   announced the discovery of a
   planetary mass object (0.5
   times Jupiter) orbiting the
   solar-type star 51 Peg.
   

   View Slide

3. Indirect detection methods:
   Radial Velocities
   

   View Slide

4. Indirect detection methods:
   transits
   Credits: NASA
   

   View Slide

5. Why study exoplanetary
   atmospheres?
   The Holy Grall of an exoplanetologist
   Biosignatures
   “..object, substance, and/or pattern whose origin specifically requires
   a biological agent”
   (Des Marais and Walter, 1999; Des Marais et al., 2008; Schwieterman
   et al. 2017)
   

   View Slide

6. Why study exoplanetary
   atmospheres?
   ▪We are in the era of comparative
   exoplanetology
   ▪Already now we reveal a rich diversity of
   chemical compositions and atmospheric
   processes hitherto unseen in the Solar
   System.
   ▪The spectrum of an exoplanet reveals the
   physical, chemical, and biological processes
   that have shaped its history and govern its
   future.
   

   View Slide

7. View Slide

8. View Slide

9. A graphical representation of the Trappist-1 system
   Credit: NASA/JPL-Caltech/Robert Hurt (IPAC)
   A C/O > 1 suggest that the planet formed
   beyond the water snowline
   and
   later migrated towards its star at the we
   observe it today
   

   View Slide

10. Variation of the C/O ratio of the gas in a disc due to
    freeze-out (Madhusudan 2019, Booth+2017).
    

    View Slide

11. Transmission Spectroscopy for Exoplanet Atmospheres
    

    View Slide

12. Credits: NASA
    

    View Slide

13. Transmission Spectroscopy for Exoplanet Atmospheres
    Sedaghati et al 2017
    Low resolution spectroscopy R < 1000
    

    View Slide

14. Low resolution spectroscopy and hot jupiters
    Figure from Sing+2016
    

    View Slide

15. View Slide

16. View Slide

17. View Slide

18. View Slide

19. Telescopio Nazionale Galileo @ La Palma
    

    View Slide

20. GIANO-B @ TNG - La Palma (Spain)
    NIR spectrograph mounted at the 3.6-metre
    Telescopio Nazionale Galileo (TNG).
    Simultaneous coverage in the 0.92-2.45 µm
    range (fifty orders )
    Spectral resolving power of R = 50,000.
    Oliva et al. 2006
    

    View Slide

21. View Slide

22. At high spectral resolution, molecular features are resolved into a dense forest of of tens of thousands of
    individual lines in a pattern that it is unique for a given molecule -> a kind of fingerprint
    High Resolution ( R = 25,000 - 100,000)
    Transmission Spectroscopy
    

    View Slide

23. View Slide

24. The high resolution spectroscopy helps to disentangle and isolate the exoplanet’s spectrum.
    Disentangle moving planet lines from stationary telluric & stellar lines
    High Resolution ( R = 25,000 - 100,000)
    Transmission Spectroscopy
    Snellen et al. (2010) C
    - HD209458b
    

    View Slide

25. View Slide

26. View Slide

27. GUIbrush®
    Graphic User Interface
    for Bayesian Retrieval
    Using Spectroscopy at
    High Resolution
    

    View Slide

28. VS
    

    View Slide

29. Guibrush® is coded in Python > 3.8
    THE DEMCMC is
    parallelized with the
    Multiprocessing
    Python library
    DEMCMC
    PetitRadTrans:
    Radiative Transfer Code
    ~10 sec for one model in
    the 0.9-2.5 micron range
    GPU?
    The goal is a 100x faster
    code for ANDES/JWST
    range, 3D models, etc
    etc
    

    View Slide

30. Bottleneck #1
    The Bayesian estimator
    

    View Slide

31. Bottleneck #1
    The Bayesian estimator
    

    View Slide

32. Bottleneck #2
    The radiative transfer code
    

    View Slide

33. Bottleneck #2
    The radiative transfer code
    

    View Slide

34. Radiative equilibrium calculations for HD209458b
    

    View Slide

35. C/O =1
    

    View Slide

36. Are there clouds?
    

    View Slide

37. The “final” matrix for Giano is 102’400 x 60 = 6’144’000
    Bottleneck #3
    The model reprocessing
    

    View Slide

38. View Slide

39. View Slide

40. When will the first 'good' news
    about biosignatures be
    released?
    

    View Slide