Items related to the evolution of particles Items related to Monte-Carlo radiation transport. More...

Namespaces
namespace	Tags
	Items related to the evolution of particles Items related to Monte-Carlo radiation transport Tags for MC.

Classes
struct	CellLightCrossingTimeCompute

struct	DiffusionMonteCarloParameters
	Precomputed quantities useful for the diffusion approximation in high-scattering opacity regions. We follow [81] Note that r_d in that manuscript should be (distance diffused)/(time elapsed) i.e. the paper is missing a normalization of r_d by delta_t. The upper bound of the integral in Eq (30) should just be sqrt(3*tau/4) All quantities in this struct are fixed; we could also just hard-code them to save us the on-the-fly calculation, or have a single instance of this struct in the global cache. More...

struct	FluidCouplingMutator
	Mutator adding the Monte-Carlo contribution to the evolution of the fluid. More...

struct	GhostZoneCouplingData
	Structure used to gather fluid coupling data for Monte-Carlo evolution. We need the energy, momentum, and composition coupling. More...

struct	InertialFrameEnergyDensityCompute

struct	InverseJacobianInertialToFluidCompute
	Inverse Jacobian of the map from inertial coordinate to an orthonormal frame comoving with the fluid. That frame uses the 4-velocity as its time axis, and constructs the other members of the tetrads using Gram-Schmidt's algorithm. More...

struct	McGhostZoneData
	The container for ghost zone data to be communicated in the Monte-Carlo algorithm. More...

struct	McGhostZoneDataInboxTag
	Inbox tag to be used before MC step. More...

class	MonteCarloOptions

struct	MortarData
	Structure used to gather ghost zone data for Monte-Carlo evolution. We need the rest mass density, electron fraction, temperature, and an estimate of the light-crossing time one cell deep within each neighboring element. More...

class	NeutrinoInteractionTable
	Class responsible for reading neutrino-matter interaction tables. More...

struct	Packet
	Struct representing a single Monte Carlo packet of neutrinos. More...

struct	SwapGrTags
	Swaps the inactive and active GR variables. More...

struct	System

struct	TemplatedLocalFunctions
	Structure containing Monte-Carlo function templated on EnergyBins and/or NeutrinoSpecies. More...

struct	TimeStepMutator
	Mutator advancing neutrinos by a single step. More...

Enumerations
enum class	CommunicationStep { PreStep , PostStep }
	This class is used to template communication actions in the Monte-Carlo code, to know which of the two communication steps we need to take care of (just before or just after the MC step). More...

Functions
void	cell_light_crossing_time (gsl::not_null< Scalar< DataVector > * > cell_light_crossing_time, const Mesh< 3 > &mesh, const tnsr::I< DataVector, 3, Frame::Inertial > &inertial_coordinates, const Scalar< DataVector > &lapse, const tnsr::I< DataVector, 3, Frame::Inertial > &shift, const tnsr::II< DataVector, 3, Frame::Inertial > &inv_spatial_metric)

void	cell_proper_four_volume_finite_difference (gsl::not_null< Scalar< DataVector > * > cell_proper_four_volume, const Scalar< DataVector > &lapse, const Scalar< DataVector > &sqrt_determinant_spatial_metric, double time_step, const Mesh< 3 > &mesh, const Scalar< DataVector > &det_jacobian_logical_to_inertial)
	Proper 4-volume of a cell for a time step time_step. We assume that determinant_spatial_metric is given in inertial coordinates, hence the need for det_jacobian_logical_to_inertial.

void	cell_inertial_coordinate_three_volume_finite_difference (gsl::not_null< Scalar< DataVector > * > cell_inertial_three_volume, const Mesh< 3 > &mesh, const Scalar< DataVector > &det_jacobian_logical_to_inertial)
	3-volume of a cell in inertial coordinate. Note that this is the coordinate volume, not the proper volume. This quantity is needed as a normalization factor for the terms coupling Monte-Carlo transport to the fluid evolution (as we evolved densitized fluid variables in inertial coordinates).

void	AddCouplingTermsForPropagation (gsl::not_null< Scalar< DataVector > * > coupling_tilde_tau, gsl::not_null< tnsr::i< DataVector, 3, Frame::Inertial > * > coupling_tilde_s, gsl::not_null< Scalar< DataVector > * > coupling_rho_ye, const Packet &packet, size_t extended_idx, double dt, double absorption_opacity, double scattering_opacity, double fluid_frame_energy, double lapse, double lorentz_factor, const std::array< double, 3 > &lower_spatial_four_velocity_packet)
	Contribution to the neutrino-matter coupling terms from evolving a packet by time dt, given absorption and scattering opacities. More...

void	evolve_single_packet_on_geodesic (gsl::not_null< Packet * > packet, double time_step, const Scalar< DataVector > &lapse, const tnsr::I< DataVector, 3, Frame::Inertial > &shift, const tnsr::i< DataVector, 3, Frame::Inertial > &d_lapse, const tnsr::iJ< DataVector, 3, Frame::Inertial > &d_shift, const tnsr::iJJ< DataVector, 3, Frame::Inertial > &d_inv_spatial_metric, const tnsr::II< DataVector, 3, Frame::Inertial > &inv_spatial_metric, const std::optional< tnsr::I< DataVector, 3, Frame::Inertial > > &mesh_velocity, const InverseJacobian< DataVector, 3, Frame::ElementLogical, Frame::Inertial > &inverse_jacobian_logical_to_inertial)

void	inertial_frame_energy_density (gsl::not_null< Scalar< DataVector > * > fluid_frame_energy_density, const std::vector< Packet > &packets, const Scalar< DataVector > &lapse, const Scalar< DataVector > &sqrt_determinant_spatial_metric, const Mesh< 3 > &mesh, const Scalar< DataVector > &det_jacobian_logical_to_inertial)

double	compute_fluid_frame_energy (const Packet &packet, const Scalar< DataVector > &lorentz_factor, const tnsr::i< DataVector, 3, Frame::Inertial > &lower_spatial_four_velocity, const Scalar< DataVector > &lapse, const tnsr::II< DataVector, 3, Frame::Inertial > &inv_spatial_metric)

void	DiffusionPrecomputeForElement (gsl::not_null< DataVector * > prefactor_diffusion_time_vector, gsl::not_null< DataVector * > prefactor_diffusion_four_velocity, gsl::not_null< DataVector * > prefactor_diffusion_time_step, const Scalar< DataVector > &lorentz_factor, const tnsr::i< DataVector, 3, Frame::Inertial > &lower_spatial_four_velocity, const Scalar< DataVector > &lapse, const tnsr::I< DataVector, 3, Frame::Inertial > &shift, const tnsr::ii< DataVector, 3, Frame::Inertial > &spatial_metric)
	Precompute quantities needed for evolving packets in the diffusion approximation, i.e. the transport velocity in logical coordinates and the coefficients defined by Eq (31-33)

std::array< double, 4 >	scatter_packet (gsl::not_null< Packet * > packet, gsl::not_null< std::mt19937 * > random_number_generator, const double &fluid_frame_energy, const Jacobian< DataVector, 4, Frame::Inertial, Frame::Fluid > &inertial_to_fluid_jacobian, const InverseJacobian< DataVector, 4, Frame::Inertial, Frame::Fluid > &inertial_to_fluid_inverse_jacobian)
	Elastic scattering, which is just redrawing the momentum from an isotropic distribution in the fluid frame, at constant fluid frame energy.

void	diffuse_packet (gsl::not_null< Packet * > packet, gsl::not_null< std::mt19937 * > random_number_generator, gsl::not_null< double * > neutrino_energy, gsl::not_null< Scalar< DataVector > * > coupling_tilde_tau, gsl::not_null< tnsr::i< DataVector, 3, Frame::Inertial > * > coupling_tilde_s, gsl::not_null< Scalar< DataVector > * > coupling_rho_ye, size_t extended_idx, double time_step, const DiffusionMonteCarloParameters &diffusion_params, double absorption_opacity, double scattering_opacity, const Scalar< DataVector > &lorentz_factor, const tnsr::i< DataVector, 3, Frame::Inertial > &lower_spatial_four_velocity, const Scalar< DataVector > &lapse, const tnsr::I< DataVector, 3, Frame::Inertial > &shift, const tnsr::i< DataVector, 3, Frame::Inertial > &d_lapse, const tnsr::iJ< DataVector, 3, Frame::Inertial > &d_shift, const tnsr::iJJ< DataVector, 3, Frame::Inertial > &d_inv_spatial_metric, const tnsr::ii< DataVector, 3, Frame::Inertial > &spatial_metric, const tnsr::II< DataVector, 3, Frame::Inertial > &inv_spatial_metric, const std::optional< tnsr::I< DataVector, 3, Frame::Inertial > > &mesh_velocity, const InverseJacobian< DataVector, 3, Frame::ElementLogical, Frame::Inertial > &inverse_jacobian_logical_to_inertial, const Jacobian< DataVector, 4, Frame::Inertial, Frame::Fluid > &inertial_to_fluid_jacobian, const InverseJacobian< DataVector, 4, Frame::Inertial, Frame::Fluid > &inertial_to_fluid_inverse_jacobian, const DataVector &prefactor_diffusion_time_step, const DataVector &prefactor_diffusion_four_velocity, const DataVector &prefactor_diffusion_time_vector)
	Evolve a packet for dt = time_step, assuming that we can use the diffusion approximation. More...

Detailed Description

Items related to the evolution of particles Items related to Monte-Carlo radiation transport.

Enumeration Type Documentation

◆ CommunicationStep

enum class Particles::MonteCarlo::CommunicationStep

strong

This class is used to template communication actions in the Monte-Carlo code, to know which of the two communication steps we need to take care of (just before or just after the MC step).

Enumerator
PreStep	PreStep should occur just before the MC step. It should send to the ghost zones of each element the fluid and metric data.
PostStep	PostStep should occur just after a MC step. It sends packets that have left their current element to the element they now belong to. NOT CODE YET: This is also when information about energy/momentum/lepton number exchance in the ghost zones should be communicated back to the corresponding live point for coupling to the fluid evolution.

Function Documentation

◆ AddCouplingTermsForPropagation()

void Particles::MonteCarlo::AddCouplingTermsForPropagation	(	gsl::not_null< Scalar< DataVector > * >	coupling_tilde_tau,
		gsl::not_null< tnsr::i< DataVector, 3, Frame::Inertial > * >	coupling_tilde_s,
		gsl::not_null< Scalar< DataVector > * >	coupling_rho_ye,
		const Packet &	packet,
		size_t	extended_idx,
		double	dt,
		double	absorption_opacity,
		double	scattering_opacity,
		double	fluid_frame_energy,
		double	lapse,
		double	lorentz_factor,
		const std::array< double, 3 > &	lower_spatial_four_velocity_packet
	)

Contribution to the neutrino-matter coupling terms from evolving a packet by time dt, given absorption and scattering opacities.

Details

We calculate total momentum exchanges, not densities; thus when coupling to the evolution of the energy/momentum density, division by the spatial volume is necessary. We calculate the source terms of Eqs (62-64) of [80], with integrals calculated as in Eqs (6-10) of that manuscript (except that we do not divide by the coordinate volume V).

◆ compute_fluid_frame_energy()

double Particles::MonteCarlo::compute_fluid_frame_energy	(	const Packet &	packet,
		const Scalar< DataVector > &	lorentz_factor,
		const tnsr::i< DataVector, 3, Frame::Inertial > &	lower_spatial_four_velocity,
		const Scalar< DataVector > &	lapse,
		const tnsr::II< DataVector, 3, Frame::Inertial > &	inv_spatial_metric
	)

Calculate energy of neutrinos in a frame comoving with the fluid

\begin{align} E = W \alpha p^t - \gamma^{ij} u_i p_j \end{align}

◆ diffuse_packet()

void Particles::MonteCarlo::diffuse_packet	(	gsl::not_null< Packet * >	packet,
		gsl::not_null< std::mt19937 * >	random_number_generator,
		gsl::not_null< double * >	neutrino_energy,
		gsl::not_null< Scalar< DataVector > * >	coupling_tilde_tau,
		gsl::not_null< tnsr::i< DataVector, 3, Frame::Inertial > * >	coupling_tilde_s,
		gsl::not_null< Scalar< DataVector > * >	coupling_rho_ye,
		size_t	extended_idx,
		double	time_step,
		const DiffusionMonteCarloParameters &	diffusion_params,
		double	absorption_opacity,
		double	scattering_opacity,
		const Scalar< DataVector > &	lorentz_factor,
		const tnsr::i< DataVector, 3, Frame::Inertial > &	lower_spatial_four_velocity,
		const Scalar< DataVector > &	lapse,
		const tnsr::I< DataVector, 3, Frame::Inertial > &	shift,
		const tnsr::i< DataVector, 3, Frame::Inertial > &	d_lapse,
		const tnsr::iJ< DataVector, 3, Frame::Inertial > &	d_shift,
		const tnsr::iJJ< DataVector, 3, Frame::Inertial > &	d_inv_spatial_metric,
		const tnsr::ii< DataVector, 3, Frame::Inertial > &	spatial_metric,
		const tnsr::II< DataVector, 3, Frame::Inertial > &	inv_spatial_metric,
		const std::optional< tnsr::I< DataVector, 3, Frame::Inertial > > &	mesh_velocity,
		const InverseJacobian< DataVector, 3, Frame::ElementLogical, Frame::Inertial > &	inverse_jacobian_logical_to_inertial,
		const Jacobian< DataVector, 4, Frame::Inertial, Frame::Fluid > &	inertial_to_fluid_jacobian,
		const InverseJacobian< DataVector, 4, Frame::Inertial, Frame::Fluid > &	inertial_to_fluid_inverse_jacobian,
		const DataVector &	prefactor_diffusion_time_step,
		const DataVector &	prefactor_diffusion_four_velocity,
		const DataVector &	prefactor_diffusion_time_vector
	)

Evolve a packet for dt = time_step, assuming that we can use the diffusion approximation.

The coupling terms include ghost zone information, while other tensors only include live points. extended_idx is the index of the packet in DataVectors including ghost zones.

Namespaces

Classes

Enumerations

Functions