A full-particle Martian upper thermosphere-exosphere model using the DSMC method

Kaori Terada, Naoki Terada, Hiroyuki Shinagawa, Hitoshi Fujiwara, Yasumasa Kasaba, Kanako Seki, François Leblanc, Jean Yves Chaufray, Ronan Modolo

Research output: Contribution to journalArticlepeer-review

5 Citations (Scopus)


A one-dimensional full-particle model of the Martian upper thermosphere-exosphere has been developed, where the Direct Simulation Monte Carlo (DSMC) method is applied to both thermal and nonthermal components. Our full-particle model can self-consistently solve the transition from collisional to collisionless domains in the upper thermosphere, so that the energy deposition from nonthermal energetic components to thermal components in the transition region is properly described. For the solar EUV condition during the Viking 1 measurement (1 EUV case), computed density profiles are in good agreement with those observed by Viking 1 and with the conventional model. For a solar EUV flux 6 times the Viking 1 condition (6 EUV case), the computed heating efficiency is essentially the same as the 1 EUV case but slightly increases by about 10% below the exobase, and temperature deviates from the conventional model in and above the transition region. This result suggests that the conventional heating efficiency of 0.18 is a good approximation for low (1 EUV case) to moderately strong (6 EUV case) solar EUV conditions but would be inappropriate for an extremely strong solar EUV (up to ~100 times stronger flux) environment. We also find that applying different models of the CO2-O collisional energy transfer rate results in a difference in the calculated exobase temperature by 150 K for the 6 EUV case.

Original languageEnglish
Pages (from-to)1429-1444
Number of pages16
JournalJournal of Geophysical Research: Planets
Issue number8
Publication statusPublished - 2016 Aug 1


  • DSMC simulation
  • exosphere
  • Mars
  • upper thermosphere


Dive into the research topics of 'A full-particle Martian upper thermosphere-exosphere model using the DSMC method'. Together they form a unique fingerprint.

Cite this