- Fixed sign for exponential decay of magn. field strength with Galactic height in LogarithmicSpiralField
- Fixed r term in source distribution for SNR and Pulsar
- Fixed wrong mass inheritance for secondaries other than nuclei or electron/positron
- Added new backwards-compatible function particleMass that returns particle mass also for non-nuclei
- Added the new Galactic magnetic field models from Unger&Farrar arXiv:2311.12120
- Added EBL model from Finke et al. 2022
- AMRMagneticField - underlying library (saga) is no longer supported.
- ObserverPoint: Use Observer1D instead.
- Re-added ToroidalHaloField and LogarithmicSpiralField models. Note, that the class name was also corrected in spelling: TorroidalHaloField --> ToroidalHaloField
- Synchronized signature of ParticleSplitting constructor
- New candidate property tagOrigin to trace back which source or which interaction created the candidate
- New interface for massdistributions given on a Grid1f
- Grids can be restricted to the volume without repetition (clipVolume parameter)
- SourceFeature to sample the source position from a given massdistribution
- EBL model from Saldana-Lopez et al. 2021
- New module CandidateSplitting for better statistics at high energies for e.g. diffusive shock acceleration
- New advection fields for modeling diffusive shock acceleration at 1D planar, oblique and spherical shocks
- Weight column in hdf-Output is now called "W", which is the same as for TextOutput.
- ObserverPoint will be renamed into Observer1D.
- AMRMagneticField - underlying library (saga) is no longer supported.
- External extensions DINT and Eleca, which can be replaced with the EM*-modules combined with the thinning option for reasonable computation times.
- FieldlineIntegrator
- grplinst
- monopole
- ROOTOutputPlugin
- Fix of reflective boundary condition for scalar- and vectorgrids that showed asymmetry and discontinuities (See issue [#361]).
- Fix in EMTripletPairProduction
- Fix of the data files of the Hackstein EGMF models as well as the corresponding example notebook.
- Fix of axis normalization of getRotated in Vector3.h.
- Fix of secondary spectra in electromagnetic interactions (EM*-modules), issue [#334] and pull request [#15] in crpropa-data.
- Fix weight inheritance for secondary particles; they are created with their parents weights as intial weights now.
- Add modules for first and second order Fermi acceleration
- Make custom photon fields available in the PhotoPionProduction module.
- Add tricubic- and nearest neighbour interpolation routines for scalar- and vectorgrids.
- Add the new PolarizedSingleModeMagneticField class for polarized/ helical single mode magnetic field models.
- Add a source feature for targeted emission, following the von-Mises-Fisher distribution
- Updates in SNR and pulsar source distributions
- Plane wave and grid turbulence models use same parameter convention now
- External extensions DINT and Eleca, which can be replaced with the EM*-modules combined with the thinning option for reasonable computation times.
- Updated version of the CLUES EGMF models
- Re-add URB that was temporarily not available in the (unreleased) master branch.
- Turbulent fields generated on a grid were limited up to 2048 grid-size due to an integer overflow (i.e. 2048^3 index > signed int); solved by replacing int with size_t
- New model for the radio background added: Nitu et al. 2021 measurements.
- New model for the radio background added: Fixsen et al. 2011 (ARCADE-2) measurements.
- Weighted sampling thinning of electromagnetic processes (EMPairProduction, EMInverseComptonScattering, EMDoublePairProduction, EMTripletPairProduction).
- Planck JF12b variant of the JF12Field. See arXiv:1601.00546. Thanks to Mikhail Zotov for contributing.
- ParticleCollector can provide Candidates directly to ModuleList::run
- Basic file versioning of the data archive in CMakeLists.txt
- Python docstrings are generated automatically from doxygen documentation
- Photon field classes replaced the photon field enumerator, consequently, the new photon fields implementation follows the same logic as of the other modules making it is easier to introduce custom ones
- New class-based interface for turbulent fields introduced
- New turbulence modules implemented:
- GridTurbulence (with the bendover scale) which should in general be used instead of initTurbulence (before, it was implemented as initTurbulenceWithBendover);
- SimpleGridTurbulence which provides the exact field as initTurbulence;
- HelicalGridTurbulence which provides the exact field as initHelicalTurbulence
- PlaneWaveTurbulence a new algorithm based on Giacalone & Jokipii, 1999 and Tautz & Dosch, 2013.
- New CMake option:
BUILD_DOC
for building Doxygen & Sphinx docs
- ParticleCollector::getAll() -> ParticleCollector::getContainer()
- Photon fields are no longer items of the PhotonField enumerator but independent
classes that share the same interface, so instead of
CMB
one should useCMB()
, instead ofIRB_Kneiske04
-IRB_Kneiske04()
, etc. - initTurbulenceWithBendover() removed (as it was just briefly present in the code) and replaced with GridTurbulence
- Turbulence-related functions: initTurbulence, turbulentCorrelationLength, initHelicalTurbulence
- Fix of (#254): Redshift evolution in PhotoPionProduction The reshift evolution was always handled with simple scaling and never with the more accurate 2 dimensional interpolation.
- New source feature SourceLambertDistributionOnSphere Simplifies simulations of scenarios with an initially isotropic and homogeneous distribution of cosmic rays on a sphere, e.g. for investigation of propagation of extagalactic cosmic rays in the Milky Way (see issue #246 and pull request #247)
- For PhotoPionProduction now a two dimensional interpolation of the redshift evolution is available (see pull request #255)
- A new break condition MinimumChargeNumber is added (see pull request #256)
- Vector3 now has index based access to its components. This improves interoperability with third party libraries, in particular with numpy arrays. (see pull request #262)
- Random seeds can be accessed from python (see pull request #263)
- New galactic magnetic field model by Terral & Ferriere (2017) (see pull request #258)
- Reimplementation of SOPHIA's photon field sampling used in PhotoPionProduction in c++, leading to a factor 2-3 speed up of the module (see pull request #260).
- Introducing a method, sophiaEvent(onProton, Eprimary, Ephoton), to directly call SOPHIA's event generator from python (see pull request #260).
- Fixed issue with secondaries in the PhotoPionProduction potentially relevant for primary energies above approx. 10**21 eV (See issue [#225]).
- Fix of azimuthal component of ArchimedeanSpiral Field (See commit 1f79e2c). Thanks to Leander Schlegel for reporting and providing a patch.
- The PropagatorBP implements the Boris-Push algorithm as as alternative to the Cash-Karp integrator PropagatorCK. Thanks to Patrick Reichherzer for this contribution.
- A basic geometry system was added to e.g. restrict modules to a spatial region.
- Candidates now have weights.
- User can define custom candidate properties identified with a string (slow but sometimes handy).
- Random seeds can be stored in output files. Please note that perfect reproducibility of simulations is still not guaranteed when using multi cores.
- The 'ParticleCollector' allows in-memory storage of Candidates for collective-processing during a simulation.
- The SDE solver supports advection and adiabatic cooling.
- A selection of models for mass distributions in our Galaxy as first step of the development of a hadron-hadron interaction module was added. Thanks to Julien Dörner for implementing these models.
- Source distributions following the Galactic pulsar distribution and the Galactic SNR distributions are available.
- Turbulent magnetic fields can now be optionally helical (initHelicalTurbulence) or obey the homogeneous turbulence theory requirement (initTurbulenceWithBendover).
- Modified JF12 Galactic magnetic field model with a truly solenoidal disk spiral field and a smooth X field with parabolic field lines according to Kleimann et al. (arXiv:1809.07528). Thanks to Timo Schorlepp for sharing this code.
- PhotonOutput1D is obsolete as the functionality is provided by the regular output modules (See #211).
- Future releases will use C++11 features. Outdated compilers will not be supported anymore.
- ROOTOutput. This code has been moved from CRPropa to a plugin and is no longer maintained by us. If you want to maintain this code, please contact us.
- Data for the constrained 'Hackstein' models of the local Universe using initial conditions from the CLUES project. Thanks to Stefan Hackstein for sharing.