wiki:NewC07.02

New in C07.02

This page summarizes all major changes and improvements to Cloudy that were included in this release. You may also view the HotFixes and KnownProblems pages, or return to the main RevisionHistory page.


Changes in the C07.02.01 service release

Version C07.02.01 was posted on the web site on 2007 July 8. This version contains all bug fixes through this date, and three other substantial changes.

  • Part 3 of Hazy has been updated. Parts 1 and 3 are now correct for this version.
  • The treatment of dielectronic recombination for ions with no published rates has been changed. A command has been added to allow these guesses to be further changed.
  • The absorption lines in the fine continuum are now deferrenced onto the coarse continuum. This has no impact on the calculations but absorption lines will now appear in the punch continuum output.

Further details are in AddendaC07_02_01.pdf


The C07.02.00 original release

Infrastructure

The code is now C++.

It runs a model in about 2/3 the time needed by C06.02. This is a significant speedup compared with C06.02, while doing more physics with greater stability.

Robin Williams has rewritten much of the logic of the molecular solvers. As a result the code is now substantially shorter and more stable.

Peter van Hoof has totally revised the table stars command so that the same code base is used for many options. This will make it easier to include new grids of stellar atmospheres in the future. He also has included support for several more grids.

The state command is being developed. This allows the code to save state at the end of a calculation (using the command state put filename) and later restart using the saved optical depth scale (using the command state get filename). This is intended for debugging dynamics simulations which can easily require a hundred iterations. Further discussion on this page

Ryan Porter added the grid command, that makes it possible to compute grids of simulations in which one or more parameters are varied. Many commands include a vary keyword and is followed by a grid range command to set the range over which it will be varied. The following will compute models with three ionization parameters, -2, -1, and 0, and three densities, 3, 4, and 5. A total of nine models will be computed.

ionization parameter -1 vary
grid range from -2 to 0 in 10. size steps
hden 3 vary
grid range from 3 to 5 in 1.0 steps

Commands and output

The punch physical conditions output now includes the filling factor and is tab delimited.

The files created by punch commands will no longer overwrite one another when a series of models are computed. There is a clobber option to use the old behavior, in which the output from each model would overwrite the file. This logic is the opposite of the old

The set H2 Solomon command has been renamed to set H2 small model. the new name better describes the full effects of the command.

The assert temperature command has a grid option to assert the temperatures in a series of grid calculations.

A new punch averages command will report the average temperature, ionization fraction, or column density, for a model. This can be combined with the grid command to save information from large grids of calculations.

The punch radius command now has the outer option, to only produce this information for the outer radius.

The punch column densities command now has the some option. If the keyword some appears then the code will read in a list of species and write the column densities as one long out.

A new stop nTotalIoniz command was introduced. This will cause the code to abort after the ionization solver has been called n times, where n is the number that appears on the command line. This is a debugging aid.

A new print continuum indices command was introduced. This prints all lines that lie with a given cell in the coarse continuum. This, together with the print line sort wavelength command, provides a way to determine what lines contribute to various features in the continuum.

The print arrays command now accepts the keyword only followed by the name of an element. If these appear then only the arrays for that element are printed. By default all elements are printed.

The no buffering command works differently. Output will go to stderr which is unbuffered according to the standard.

The first columns of the punch heating and punch cooling commands have been changed to be the same. They will give the depth, temperature, heating, and cooling. Thanks to Yumihiko Tsuzuki for finding this inconsistency.

Routine cdEDEN_last to return the electron density of the last zone has been added.

The set grain heat command multiplies the grain heating rate by the factor that appears on the line. This allows arbitrary changes in the grain heating efficiency. Allers et al. (2005; ApJ, 630, 368) argue that the grain heating should be increased by a factor of three.

The new punch linelist command reads in a list of emission lines from a file and gives the predicted intensities of those lines in a punch file. It calls routine cdGetLineList which is described in part 3 of Hazy. Several useful line lists are in the files LineList*.dat in the data/ directory.

The cosmic ray command now requires the keyword background to do galactic background cosmic rays. Previously background was assumed if no other keywords were recognized. That is unsafe since an accidental misspelling of one of the other keywords would be interpreted as background.

Ryan Porter added the punch FITS command to output continua in FITS format.

Ryan Porter added the punch XSPEC command to create output for the XSPEC x-ray data analysis program.

The no free free command now turns off free-free emission in addition to heating and cooling.

The magnetic field command now has vary option. Only the field itself is varied, not its gamma law.

The main emission line printout has always listed the intrinsic intensity of lines. Previously, in a dusty open geometry, the code also predicted the emergent intensity as observed from the illuminated face of the H II region. This is discussed further in the section "Line intensities in a dusty open geometry" in the "Observed Quantities" chapter of Part 3 of Hazy. The code now predicts and prints intrinsic and emergent intensities for both closed and open geometries. This requires careful thought - the emergent spectrum is very geometry dependent. The new print line emergent command will tell the code to use the emergent rather than intrinsic intensity for most quantities. 2006 Mar 11

The drive helium command no longer worked and has been removed. Thanks to Kevin Blagrave for discovering this. 2006 March 11.

Atomic & molecular data

The Badnell recombination data set is now the default.

The Gu et al. (2006; ApJ 641, 1227) UTA date are now the default for Fe0 - Fe+15.

A new set UTA command allows either the Gu et al. (2006) or the older Behar et al. (2001; ApJ, 563, 497) to be used. The Behar et al. data were the default in c06.02 and the Gu et al. data are now the default. The Kisielius et al. (MNRAS 344, 696) data continue to be used for Fe+13 - Fe+15. The no Kisielius UTA command will cause the code to use only the Behar or Gu UTA data.

The set recombination Badnell commands now accept both ON and OFF keywords. This makes it possible to turn off the Badnell rates once they are the default.

Nick Abel updated the collision strength for [O IV] 26 micron to Tayal 2006 ApJS, 166, 634

The collision strengths for the [O I] optical lines have been updated to Barklem,P.S.,2006,A&A (astroph 0609684) by Humeshkar Nemala. The new collision strengths for the 6300, 6363 lines can be several times larger than the old vales at 5000 - 10,000K. As a result the [OI] optical lines are now often 20% to 30% stronger.

The collision rates for excitation of [OI] by H0 were updated to Krems, R.V., Jameson, M.J. & Dalgarno, A. 2006, ApJ, 647, 1531

Dielectronic rate coefficients for Fe updated to Badnell astro-ph 0608339.

The wavelength of the 2 3P - 2 3S multiplet of the He-like iso-electronic sequence is now the mean of the three lines rather than the wavelength of the lowest J line. This multiplet has the label TOTL. The three lines within the multiplet are unchanged.

The collision strengths for FUV lines of Fe 12 have been changed from the g-bar approximations given by Mewe et al. 1985, A&AS, 62, 197, to Storey, P.J., Del Zanna, G., Mason, H.E., #* refercon & Zeippen, C.J., 2005, A&A, 433, 717. The new values are roughly a factor of 2.5 smaller.

The wavelengths of some continuum bands have changed. The code integrates all lines and continua over a set of wavelength bands that are given in the file bands_continuum.dat, contained in the data directory. These bands are part of the standard output and make it possible to obtain the total luminosity that would be detected across particular filters. The following bands changed: FIR 8.3m, TIR 1800m (was TIR 550m), NIRa 2.85m, and NMIR 21.75m. Further details are in the bands_continuum.dat file contained in the data directory.

The formalism used to evaluate collision rates for hydrogenic levels of all ions treated with the unified iso-electronic sequences has been updated to Porter, Bauman, Ferland & MacAdam (2005, ApJ 622L, 73) by Ryan Porter. These rates are considerably smaller than these used previously, which were based on Ferguson & Ferland (1997, ApJ, 479, 363). H and He emission lines from gas with high densities, especially BLR clouds, are now stronger. In some cases the gas temperature is higher due to the smaller collision rates and consequent smaller cooling. The consequences can be seen by examining the test suite predictions - these changes were introduced in the time 2006 July 20-25.

Charge exchange rates coefficients for O+2 and N+2 with H0 updated to Barragan et al. ApJ 636, 544. These are substantially smaller than the previous rates.

Collision strengths for neon ions updated by Humeshkar Nemala. These include [Ne II], [Ne III], and [Ne V]. The largest changes were in the collision strengths for infrared fine structure lines. All changed by some extent, the [Ne III] 15.55 mm line the most - it is now roughly 2/3 its previous intensity.

Collision strength for F iso sequence updated to Berrington, Saraph, & Tully A&A 129, 161 by Humeshkar Nemala.

Dielectronic recombination rate coefficient for Fe+13 -> Fe+12 updated to Schmidt et al. 2006, ApJ, 641, 157L

Dielectronic recombination for S ions updated to Badnell ApJ, 379, 356

Nick Abel added N2H+ to the chemistry.

The set atomic data command has been introduced to allow different choices of atomic data sets. This is introduced to select between two sets of transition probabilities for [O II]. With this change the default is now Zeippen (1982) rather than the Zeippen (1987) values recommended by NIST. Observations suggest that the older values are better.

The Badnell dielectronic recombination data file has been updated to his 2006 March 11 version.

The model of N0 has been upgraded to the atomic data presented in Tayal (2006, ApJS, 163, 207). The model is now a five-level atom.

Nick Abel has continued to improve the chemistry network. Recent additions include charge transfer of Mg0 and Fe0 with C+, Si+, and S+, and an added Boltzmann factor for certain reactions which did not have the factor in some references, but did in others, and which clearly did need one. The CO column density and amount of freeze-out onto grain surfaces changed for some low-density PDRs. These changes will be part of the next minor release, C06.02a. 2006 March 06.

Peter van Hoof has improved the grain-molecular interactions. Most importantly, the projected grain surface area was missing a factor GrnVryDpth(), leading to an overestimation of the PAH surface area in molecular regions. Also, the expressions for the average velocity of the molecules were missing a factor sqrt(8/pi), leading to an underestimation of the velocity. These changes will be part of the next minor release, C06.02a. 2006 March 06.

Other physics

Turbulence is now included as a pressure turn in the gas equation of state. There is a "no pressure" option on the turbulence command that makes it possible to not include this term. The factor F relating turbulent velocity and the energy density (Heiles & Troland 2005, ApJ, 624, 773) is now the second number on the command line, with a default of 3. F must be specified if a turbulence dissipation length is specified.

A time-varying continuum source can once again be treated. The first paper on this was Ferland & Truran (1981, ApJ, 244, 1022, available here on ADS) but that physics was not developed along with the rest of the code. There is now a time command that describes how the continuum source varies. This physics builds upon that described in Henney et al (2005, ApJ, 621, 328, available here on ADS).

The following shows two cases. The first is the instantaneous turn-off of a O star near an H II region. The plot shows the atomic hydrogen fraction (the y-axis) as a function of depth into the cloud (cm - the x-axis). Each curve is for a successive time step. The cloud starts highly ionized, the continuum source is switched off, and the H0 fraction increases with time. It eventually becomes atomic. The simulation stopped before molecular hydrogen had time to form.


Return to the RevisionHistory page.

Return to the StepByStep instructions.

Return to main wiki page

Return to nublado.org


Last modified 10 years ago Last modified on 2008-03-04T00:40:38Z

Attachments (1)

Download all attachments as: .zip