PyTracerLab.model.units module¶

Base classes for model units.

class PyTracerLab.model.units.DMUnit(mtt: float, DP: float)[source]¶

Bases: PyTracerLab.model.units.Unit

Dispersion Model (DM) unit.

Parameters:

mtt (float) – Mean travel time.
DP (float) – Dispersion parameter. Represents the inverse of the Peclet number. Also represents the ratio of the dispersion coefficient to the velocity and outlet / sampling position
PREFIX (str) – Prefix for local parameter names. Helper for GUI.
PARAMS (List[Dict[str, Any]]) – List of (default) parameter definitions. Helper for GUI.
Note (The parameter key for the mean travel time (mtt) is used in the)
and (GUI explicitly. The GUI assumes the parameter is given in years)
it. (internally converts)

DP: float¶

PARAMS = [{'bounds': (1.0, 10000.0), 'default': 10.0, 'key': 'mtt', 'label': 'Mean Travel Time'}, {'bounds': (0.0001, 10.0), 'default': 1.0, 'key': 'DP', 'label': 'Dispersion Param.'}]¶

PREFIX = 'dm'¶

get_impulse_response(tau: numpy.ndarray, dt: float, lambda_: float, prod: bool = False) → numpy.ndarray[source]¶

DM impulse response with decay.

The continuous-time DM response (without decay) is h(τ) = (1/mtt) * 1 / (sqrt(K)) * exp((1 - τ / mtt)^2 / K) with K = 4 pi DP (τ / mtt). We also apply an exponential decay term exp(-λ τ).

Parameters:

tau (ndarray) – Non-negative time axis (same spacing as simulation time grid).
dt (float) – Time step size of the discretization.
lambda (float) – Decay constant (1 / time units of tau).
prod (bool, optional) – If True, calculate production response (used to simulate 3He production from 3H decay).

Returns:

Impulse response evaluated at tau.

Return type:

ndarray

mtt: float¶

param_values() → Dict[str, float][source]¶

Get parameter values.

Returns:: Mapping from local parameter name to value.
Return type:: Dict[str, float]

set_param_values(values: Dict[str, float]) → None[source]¶

Set one or more local parameter values.

Parameters:: values (Dict[str, float]) – Mapping from local parameter name to new value. Keys not present are ignored.

class PyTracerLab.model.units.EMUnit(mtt: float)[source]¶

Bases: PyTracerLab.model.units.Unit

Exponential Model (EM) unit.

Parameters:

mtt (float) – Mean travel time.
PREFIX (str) – Prefix for local parameter names. Helper for GUI.
PARAMS (List[Dict[str, Any]]) – List of (default) parameter definitions. Helper for GUI.
Note (The parameter key for the mean travel time (mtt) is used in the)
and (GUI explicitly. The GUI assumes the parameter is given in years)
it. (internally converts)

PARAMS = [{'bounds': (0.0, 10000.0), 'default': 10.0, 'key': 'mtt', 'label': 'Mean Travel Time'}]¶

PREFIX = 'em'¶

get_impulse_response(tau: numpy.ndarray, dt: float, lambda_: float, prod: bool = False) → numpy.ndarray[source]¶

EM impulse response with decay.

The continuous-time EPM response (without decay) is h(τ) = (1/mtt) * exp(-τ / mtt). We also apply an exponential decay term exp(-λ τ).

Parameters:

tau (ndarray) – Non-negative time axis (same spacing as simulation time grid).
dt (float) – Time step size of the discretization.
lambda (float) – Decay constant (1 / time units of tau).
prod (bool, optional) – If True, calculate production response (used to simulate 3He production from 3H decay).

Returns:

Impulse response evaluated at tau.

Return type:

ndarray

mtt: float¶

param_values() → Dict[str, float][source]¶

Get parameter values.

Returns:: Mapping from local parameter name to value.
Return type:: Dict[str, float]

set_param_values(values: Dict[str, float]) → None[source]¶

Set one or more local parameter values.

Parameters:: values (Dict[str, float]) – Mapping from local parameter name to new value. Keys not present are ignored.

class PyTracerLab.model.units.EPMUnit(mtt: float, eta: float)[source]¶

Bases: PyTracerLab.model.units.Unit

Exponential Piston-Flow Model (EPM) unit.

Parameters:

mtt (float) – Mean travel time.
eta (float) – Ratio of total volume to the exponential reservoir (>= 1). eta=1 reduces to a pure exponential model; eta>1 adds a piston component.
PREFIX (str) – Prefix for local parameter names. Helper for GUI.
PARAMS (List[Dict[str, Any]]) – List of (default) parameter definitions. Helper for GUI.
Note (The parameter key for the mean travel time (mtt) is used in the)
and (GUI explicitly. The GUI assumes the parameter is given in years)
it. (internally converts)

PARAMS = [{'bounds': (0.0, 10000.0), 'default': 10.0, 'key': 'mtt', 'label': 'Mean Travel Time'}, {'bounds': (1.0, 2.0), 'default': 1.1, 'key': 'eta', 'label': 'Eta'}]¶

PREFIX = 'epm'¶

eta: float¶

get_impulse_response(tau: numpy.ndarray, dt: float, lambda_: float, prod: bool = False) → numpy.ndarray[source]¶

EPM impulse response with decay.

The continuous-time EPM response (without decay) is h(τ) = (η/mtt) * exp(-η τ / mtt + η - 1) for τ >= mtt*(1 - 1/η) and 0 otherwise. We also apply an exponential decay term exp(-λ τ).

Parameters:

tau (ndarray) – Non-negative time axis (same spacing as simulation time grid).
dt (float) – Time step size of the discretization.
lambda (float) – Decay constant (1 / time units of tau).
prod (bool, optional) – If True, calculate production response (used to simulate 3He production from 3H decay).

Returns:

Impulse response evaluated at tau.

Return type:

ndarray

mtt: float¶

param_values() → Dict[str, float][source]¶

Get parameter values.

Returns:: Mapping from local parameter name to value.
Return type:: Dict[str, float]

set_param_values(values: Dict[str, float]) → None[source]¶

Set one or more local parameter values.

Parameters:: values (Dict[str, float]) – Mapping from local parameter name to new value. Keys not present are ignored.

class PyTracerLab.model.units.ExEPMUnit(mtt: float, exp_part: float, piston_part: float)[source]¶

Bases: PyTracerLab.model.units.Unit

Explicit xponential Piston-Flow Model (EPM) unit. This model is essentially the same as the EPMUnit, but the EPM ratio (total volume / exponential volume or total area / area receiving recharge) is defined via two parameters instead of one aggregated parameter. Those two parameters are directly related and can never be estimated simultaneously.

Parameters:

mtt (float) – Mean travel time.
exp_part (float) – Area receiving recharge or exponential volume of the system.
piston_part (float) – Area not receiving recharge or piston-flow volume of the system.
PREFIX (str) – Prefix for local parameter names. Helper for GUI.
PARAMS (List[Dict[str, Any]]) – List of (default) parameter definitions. Helper for GUI.
Note (The parameter key for the mean travel time (mtt) is used in the)
and (GUI explicitly. The GUI assumes the parameter is given in years)
it. (internally converts)

PARAMS = [{'bounds': (0.0, 10000.0), 'default': 10.0, 'key': 'mtt', 'label': 'Mean Travel Time'}, {'bounds': (0.0, 100.0), 'default': 0.5, 'key': 'exp_part', 'label': 'Exponential Part'}, {'bounds': (0.0, 100.0), 'default': 1.0, 'key': 'piston_part', 'label': 'Piston Part'}]¶

PREFIX = 'epm'¶

exp_part: float¶

get_impulse_response(tau: numpy.ndarray, dt: float, lambda_: float, prod: bool = False) → numpy.ndarray[source]¶

ExEPM impulse response with decay.

Parameters:

tau (ndarray) – Non-negative time axis (same spacing as simulation time grid).
dt (float) – Time step size of the discretization.
lambda (float) – Decay constant (1 / time units of tau).
prod (bool, optional) – If True, calculate production response (used to simulate 3He production from 3H decay).

Returns:

Impulse response evaluated at tau.

Return type:

ndarray

mtt: float¶

param_values() → Dict[str, float][source]¶

Get parameter values.

Returns:: Mapping from local parameter name to value.
Return type:: Dict[str, float]

piston_part: float¶

set_param_values(values: Dict[str, float]) → None[source]¶

Set one or more local parameter values.

Parameters:: values (Dict[str, float]) – Mapping from local parameter name to new value. Keys not present are ignored.

class PyTracerLab.model.units.PMUnit(mtt: float)[source]¶

Bases: PyTracerLab.model.units.Unit

Piston-Flow Model (discrete delta at the mean travel time) with decay.

Parameters:

mtt (float) – Mean travel time where all mass is transported as a plug flow.
PREFIX (str) – Prefix for local parameter names. Helper for GUI.
PARAMS (List[Dict[str, Any]]) – List of (default) parameter definitions. Helper for GUI.
Note (The parameter key for the mean travel time (mtt) is used in the)
and (GUI explicitly. The GUI assumes the parameter is given in years)
it. (internally converts)

PARAMS = [{'bounds': (0.0, 10000.0), 'default': 10.0, 'key': 'mtt', 'label': 'Mean Travel Time'}]¶

PREFIX = 'pm'¶

get_impulse_response(tau: numpy.ndarray, dt: float, lambda_: float, prod: bool = False) → numpy.ndarray[source]¶

Discrete delta response on the grid with exponential decay.

The delta is represented by setting the bin at round(mtt/dt) to 1/dt to preserve unit mass in the discrete sum.

Parameters:

tau (ndarray) – Non-negative time axis (same spacing as simulation time grid).
dt (float) – Time step size of the discretization.
lambda (float) – Decay constant (1 / time units of tau).
prod (bool, optional) – If True, calculate production response (used to simulate 3He production from 3H decay).

Returns:

Impulse response evaluated at tau.

Return type:

ndarray

mtt: float¶

param_values() → Dict[str, float][source]¶

Get parameter values.

Returns:: Mapping from local parameter name to value.
Return type:: Dict[str, float]

set_param_values(values: Dict[str, float]) → None[source]¶

Set local parameter value.

Parameters:: values (Dict[str, float]) – Mapping from local parameter name to new value. Keys not present are ignored.

class PyTracerLab.model.units.Unit[source]¶

Bases: abc.ABC

Abstract base class for a model unit.

Concrete units represent hydrological transport schemata and must expose and accept their local parameter values via a mapping. Units are intentionally unaware of optimization bounds; those live in the Model’s parameter registry.

Notes

Implementations must keep local parameter names stable over time so that the Model’s registry stays consistent.
Names should be short (e.g., "mtt", "eta").

get_block(h: numpy.ndarray, tau: numpy.ndarray, dt: float, lambda_: float, prod: bool = False) → numpy.ndarray[source]¶: Get 1-dt block response.

abstractmethod get_impulse_response(tau: numpy.ndarray, dt: float, lambda_: float) → numpy.ndarray[source]¶

Evaluate the unit’s impulse response on a time grid.

Parameters:

tau (ndarray) – Non-negative time axis (same spacing as simulation time grid).
dt (float) – Time step size of the discretization.
lambda (float) – Decay constant (1 / time units of tau).

Returns:

Impulse response sampled at tau.

Return type:

ndarray

normalize_response(h: numpy.ndarray, dt: float) → numpy.ndarray[source]¶: Normalize impulse response to unit area for conservation of mass.

abstractmethod param_values() → Dict[str, float][source]¶

Return current local parameter values.

Returns:: Mapping from local parameter name to value.
Return type:: Dict[str, float]

abstractmethod set_param_values(values: Dict[str, float]) → None[source]¶

Set one or more local parameter values.

Parameters:: values (Dict[str, float]) – Mapping from local parameter name to new value. Keys not present are ignored.