Version History

From EPICwiki

Jump to: navigation, search

Contents

Version 4.3

(04/2009)
  1. Horizontal velocities, U and V, have been moved from the layers to the layer interfaces, as has the turbulence variable NU_TURB. Already on the interfaces were THETA and the mixing rations Qsp (s is species index, p is phase index). The goal is a reduction of computational modes caused by vertical averaging. The hybrid density, H, remains the only prognostic variable carried in the layers.
  2. Hydrostatic integration has been improved. The model now defaults to the finite-volume formulation for horizontal pressure gradient force. These changes reduce the spurious-wind problem in steep terrain.
  3. For arrays, the model is now writing and reading from epic.nc layers K = 0 to grid.nk (instead of 1 to grid.nk), which includes the top and bottom interfaces of the model. The epic.nc file format now includes P3, full pressure on the interfaces, instead of HDRY, the dry hybrid densith in the layers, since pressure is more intuitive and the density does not contain information about the pressure at the top of the model.
  4. Time is now computed as UTC calendar time, instead of arbitrary time, to facilitate models with seasonal forcing. The model's starting time is user specified. The macro TIME behaves as before, giving the elapsed time since this start in seconds as a floating point number.
  5. Keplerian orbital elements are now included for calculation of orbit position, plus season with respect to time, for Epoch J2000.
  6. Gravity has been upgraded from a constant to a function of latitude, based on the planet's Omega, J2 and GM. Saturn in particular has a large variation in g with latitude.
  7. We are now distinguishing between the sidereal and synodic angular velocities for planetary rotation (planet->omega_sidereal and planet->omega_synodic).
  8. The model's path environment variable has been changed from "EPIC_PATH" to "EPIC4_PATH". The version without the "4" now refers to the older EPIC Version 3 (the pure-isentropic coordinate model). Making this distinction facilitates compiling different versions of the model on the same computer.
  9. The executable epic (and mpi_epic) now gets its t_vs_p and t_cool_vs_p profiles from the input epic.nc file, rather than reading them in at runtime from epic/data. The executable initial now looks for these profiles, as well as u_vs_lat, in the local directory first, and then in epic/data.
  10. The EPIC user community now has a wiki site: http://epicwiki.atmos.louisville.edu
  11. Fixed known bugs.

Version 4.2

(10/2007)
  1. Advection of theta in the sigma region improved from the 1st-order upwind scheme, which is highly dissipative and therefore undesirable, to the Hsu predictor corrector or Akima monotonized, centered scheme. The Hsu scheme is the default and is recommended.
  2. Improved hyperviscosity operators. The meridional hyperviscosity now uses finite-difference weights generated by fitting the first three Legendre polynomials to the 3-pt stencil. This improves the model's behavior at the poles. It alse runs faster than the previous algorithm. The zonal hyperviscosity is now based on the Crank-Nicolson scheme, which provides both stability and accuracy.
  3. Added the Earth case. This is being developed primarily at New Mexico Tech, by Dave Raymond's group. Their integrated "toy physics" cumulus parameterization, precipitation and radiation package is called nmt_physics.
  4. Added the option of Fred Fritsch's PCHIP monotonic spline. The issue is that a regular cubic spline is smooth but not monotonic, whereas the PCHIP (piecewise cubic Hermite interpolating polynomial) spline is both smooth and monotonic.
  5. Added the Akima advection scheme as an option. As with other centered, fourth-order schemes, this one is noisy and typically requires biharmonic hyperviscosity (nu4) to keep under control. However, it provides a useful alternative to the Hsu predictor corrector.
  6. Upgraded the netcdf output files (epic.nc, extract.nc) to meet the Climate and Forecast (CF) and Common Data Model (CDM) metadata conventions. This means that plotting packages like Panoply and IDV will automatically recognize the variables, units, and axes in EPIC output files.
  7. Fixed known bugs.

Version 4.1

(09/2006)
  1. Included Jupiter cloud microphysics model. (Csaba Palotai)
  2. Added new initialization scheme based on inversion of an elliptic problem (Aaron Herrnstein). This yields improved initial balance between the mass and momentum fields.
  3. Latest Matlab tools. (Aaron Herrnstein)
  4. Fixed known bugs.

Version 4.0

(08/2005):
  1. Converted to hybrid isentropic-sigma vertical coordinate.
  2. Added terrestrial planets.
  3. Added phase changes and latent heating of multiple condensables.
  4. Converted remaining Fortran subroutines into C.
  5. Fixed known bugs.

Version 3.9

12/2002):

This version number was reserved as a fork for Adam Showman for independent development.

Version 3.8

(07/1998)
  1. Incorporated P. Gierasch's saturation vapor-pressure fortran functions.
  2. Switched to the NetCDF standard (Network Common Data Format).
  3. Added chemical_elements.h, which contains the atomic numbers, names, symbols, molar masses (atomic weights), and solar abundances of the elements.
  4. Added specific humidities (water, ammonia, methane, etc.) as optional prognostic variables.
  5. Fixed known bugs.

Version 3.7

(04/1998)
  1. Added virial equation of state. The choices are now "ideal" and "virial."
  2. 1st and 2nd time steps now use 1st and 2nd order Adams-Bashforth steps, instead of a cold start (zero tendencies) for the model's 3rd order Adams-Bashforth time step.
  3. Switched from rsh to ssh (secure remote shell) for EPIC_Receive AVS module.
  4. Voyager velocity data files for Jupiter's Great Red Spot and White Oval BC now included in $EPIC4_PATH/data/jupiter.
  5. Fixed known bugs.

Version 3.6

(04/1998)
  1. First version posted on the NASA Planetary Data System (PDS) Atmospheres Node.
  2. The EPIC Model is described in Dowling et al (1998, Icarus 132, 221-238) "The Explicit Planetary Isentropic Coordinate (EPIC) Atmospheric Model."
  3. (C. Santori, 05/30/1997) u_vs_lat.saturn: u at 89S latitude changed from 16.20 to 16.30
  4. mqspec structure added to epic.h.
  5. Definitions of initialization types (search for SORPTOP) added to epic.h.
  6. Function prototypes for mont_from_q and low_mode added to epic.h.
  7. epic_init_funcs.c replaced entirely.
  8. Section inside "else if (grid.init_type == USE_Q_VS_M) {" replaced.
  9. epic_init_eigen.c added to epic/src/shared.
  10. Makefile.epic: epic_init_eigen.c added, mimicking epic_init_funcs.c.
  11. Spinup questions removed for case Q(M).
  12. New defaults added.
Retrieved from "http://epicwiki.atmos.louisville.edu/index.php/Version_History"
Personal tools