Skip to content

Add HARP2 (PACE) polarimetric cloudbow simulator - #167

Closed
zzbatmos wants to merge 1 commit into
CFMIP:masterfrom
zzbatmos:claude/great-lovelace-wsm2hk
Closed

zzbatmos wants to merge 1 commit into
CFMIP:masterfrom
zzbatmos:claude/great-lovelace-wsm2hk

Conversation

@zzbatmos

Copy link
Copy Markdown

Add a simulator of the HARP2 polarimetric retrieval of liquid-cloud droplet effective radius (re) and effective variance (ve) from the multi-angle polarized reflectance of the 670 nm band.

Forward model: single-scattering polarized reflectance of each subcolumn (liquid, ice and Rayleigh layers over a black surface), using a Mie look-up table of -P12(Theta; re, ve) for gamma size distributions. Retrieval: Breon and Goloub (1998) parametric fit A*(-P12) + B cos^2 + C over the 135-165 degree cloudbow window, with the minimum refined by parabolic interpolation. Grid-box outputs: cloudbow cloud fraction, mean re, mean ve and an re-ve joint histogram.

  • src/simulator/HARP2_simulator: core module, miepython LUT generator, 670 nm LUT and documentation
  • src/simulator/cosp_harp2_interface.F90: parameters, LUT reader, viewing geometry, daylight and swath masks (swath index 7)
  • src/cosp.F90: outputs, switches, error checks, calls; COSP_INIT takes optional Lharp2 and harp2_lut_file arguments (backward compatible)
  • New optical inputs cospIN%reffLiq and cospIN%veffLiq
  • Offline driver: example optics, netCDF output, namelist entries (HARP2 outputs off in the regression namelists because the KGOs do not contain them yet)
  • unit_testing/harp2: standalone tests, run in the gfortran CI

With HARP2 off or on, all existing outputs of the UKMO test case are bit-identical to master (also with chunking, model levels and swathing).

Claude-Session: https://claude.ai/code/session_019JV1kcA6ySXWErFBHBBZ6T

Add a simulator of the HARP2 polarimetric retrieval of liquid-cloud droplet
effective radius (re) and effective variance (ve) from the multi-angle
polarized reflectance of the 670 nm band.

Forward model: single-scattering polarized reflectance of each subcolumn
(liquid, ice and Rayleigh layers over a black surface), using a Mie
look-up table of -P12(Theta; re, ve) for gamma size distributions.
Retrieval: Breon and Goloub (1998) parametric fit A*(-P12) + B cos^2 + C
over the 135-165 degree cloudbow window, with the minimum refined by
parabolic interpolation. Grid-box outputs: cloudbow cloud fraction, mean
re, mean ve and an re-ve joint histogram.

- src/simulator/HARP2_simulator: core module, miepython LUT generator,
  670 nm LUT and documentation
- src/simulator/cosp_harp2_interface.F90: parameters, LUT reader,
  viewing geometry, daylight and swath masks (swath index 7)
- src/cosp.F90: outputs, switches, error checks, calls; COSP_INIT takes
  optional Lharp2 and harp2_lut_file arguments (backward compatible)
- New optical inputs cospIN%reffLiq and cospIN%veffLiq
- Offline driver: example optics, netCDF output, namelist entries
  (HARP2 outputs off in the regression namelists because the KGOs do
  not contain them yet)
- unit_testing/harp2: standalone tests, run in the gfortran CI

With HARP2 off or on, all existing outputs of the UKMO test case are
bit-identical to master (also with chunking, model levels and swathing).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019JV1kcA6ySXWErFBHBBZ6T
@zzbatmos zzbatmos closed this Sep 24, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants