{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Implementation of a Devito self adjoint variable density visco- acoustic isotropic modeling operator -- Linearized Ops --\n",
"\n",
"## This operator is contributed by Chevron Energy Technology Company (2020)\n",
"\n",
"This operator is based on simplfications of the systems presented in:\n",
" **Self-adjoint, energy-conserving second-order pseudoacoustic systems for VTI and TTI media for reverse time migration and full-waveform inversion** (2016)\n",
" Kenneth Bube, John Washbourne, Raymond Ergas, and Tamas Nemeth\n",
" SEG Technical Program Expanded Abstracts\n",
" https://library.seg.org/doi/10.1190/segam2016-13878451.1"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Introduction \n",
"\n",
"The goal of this tutorial set is to generate and prove correctness of modeling and inversion capability in Devito for variable density visco- acoustics using an energy conserving form of the wave equation. We describe how the linearization of the energy conserving *self adjoint* system with respect to modeling parameters allows using the same modeling system for all nonlinear and linearized forward and adjoint finite difference evolutions. There are three notebooks in this series:\n",
"\n",
"##### 1. Implementation of a Devito self adjoint variable density visco- acoustic isotropic modeling operator -- Nonlinear Ops\n",
"- Implement the nonlinear modeling operations. \n",
"- [sa_01_iso_implementation1.ipynb](sa_01_iso_implementation1.ipynb)\n",
"\n",
"##### 2. Implementation of a Devito self adjoint variable density visco- acoustic isotropic modeling operator -- Linearized Ops\n",
"- Implement the linearized (Jacobian) ```forward``` and ```adjoint``` modeling operations.\n",
"- [sa_02_iso_implementation2.ipynb](sa_02_iso_implementation2.ipynb)\n",
"\n",
"##### 3. Implementation of a Devito self adjoint variable density visco- acoustic isotropic modeling operator -- Correctness Testing\n",
"- Tests the correctness of the implemented operators.\n",
"- [sa_03_iso_correctness.ipynb](sa_03_iso_correctness.ipynb)\n",
"\n",
"There are similar series of notebooks implementing and testing operators for VTI and TTI anisotropy ([README.md](README.md)).\n",
"\n",
"Below we continue the implementation of our *self adjoint* wave equation with dissipation only Q attenuation, and linearize the modeling operator with respect to the model parameter velocity. We show how to implement finite difference evolutions to compute the action of the ```forward``` and ```adjoint``` Jacobian. \n",
"\n",
"## Outline \n",
"1. Define symbols \n",
"2. The nonlinear operator \n",
"3. The Jacobian opeator \n",
"4. Create the Devito grid and model fields \n",
"5. The simulation time range and acquistion geometry \n",
"6. Implement and run the nonlinear forward operator \n",
"7. Implement and run the Jacobian forward operator \n",
"8. Implement and run the Jacobian adjoint operator \n",
"9. References \n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Table of symbols\n",
"\n",
"There are many more symbols for the Jacobian than we had in the previous notebook. We need to introduce terminology for the nonlinear operator, including total, background and perturbation fields for several variables. \n",
"\n",
"| Symbol | Description | Dimensionality | \n",
"|:---|:---|:---|\n",
"| $\\overleftarrow{\\partial_t}$ | shifted first derivative wrt $t$ | shifted 1/2 sample backward in time |\n",
"| $\\partial_{tt}$ | centered second derivative wrt $t$ | centered in time |\n",
"| $\\overrightarrow{\\partial_x},\\ \\overrightarrow{\\partial_y},\\ \\overrightarrow{\\partial_z}$ | + shifted first derivative wrt $x,y,z$ | shifted 1/2 sample forward in space |\n",
"| $\\overleftarrow{\\partial_x},\\ \\overleftarrow{\\partial_y},\\ \\overleftarrow{\\partial_z}$ | - shifted first derivative wrt $x,y,z$ | shifted 1/2 sample backward in space |\n",
"| $\\omega_c = 2 \\pi f$ | center angular frequency | constant |\n",
"| $b(x,y,z)$ | buoyancy $(1 / \\rho)$ | function of space |\n",
"| $Q(x,y,z)$ | Attenuation at frequency $\\omega_c$ | function of space |\n",
"| $m(x,y,z)$ | Total P wave velocity ($m_0+\\delta m$) | function of space |\n",
"| $m_0(x,y,z)$ | Background P wave velocity | function of space |\n",
"| $\\delta m(x,y,z)$ | Perturbation to P wave velocity | function of space |\n",
"| $u(t,x,y,z)$ | Total pressure wavefield ($u_0+\\delta u$)| function of time and space |\n",
"| $u_0(t,x,y,z)$ | Background pressure wavefield | function of time and space |\n",
"| $\\delta u(t,x,y,z)$ | Perturbation to pressure wavefield | function of time and space |\n",
"| $q(t,x,y,z)$ | Source wavefield | function of time, localized in space to source location |\n",
"| $r(t,x,y,z)$ | Receiver wavefield | function of time, localized in space to receiver locations |\n",
"| $F[m; q]$ | Forward nonlinear modeling operator | Nonlinear in $m$, linear in $q$: $\\quad$ maps $m \\rightarrow r$ |\n",
"| $\\nabla F[m; q]\\ \\delta m$ | Forward Jacobian modeling operator | Linearized at $[m; q]$: $\\quad$ maps $\\delta m \\rightarrow \\delta r$ |\n",
"| $\\bigl( \\nabla F[m; q] \\bigr)^\\top\\ \\delta r$ | Adjoint Jacobian modeling operator | Linearized at $[m; q]$: $\\quad$ maps $\\delta r \\rightarrow \\delta m$ |\n",
"| $\\Delta_t, \\Delta_x, \\Delta_y, \\Delta_z$ | sampling rates for $t, x, y , z$ | $t, x, y , z$ | "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## A word about notation \n",
"\n",
"We use the arrow symbols over derivatives $\\overrightarrow{\\partial_x}$ as a shorthand notation to indicate that the derivative is taken at a shifted location. For example:\n",
"\n",
"- $\\overrightarrow{\\partial_x}\\ u(t,x,y,z)$ indicates that the $x$ derivative of $u(t,x,y,z)$ is taken at $u(t,x+\\frac{\\Delta x}{2},y,z)$.\n",
"\n",
"- $\\overleftarrow{\\partial_z}\\ u(t,x,y,z)$ indicates that the $z$ derivative of $u(t,x,y,z)$ is taken at $u(t,x,y,z-\\frac{\\Delta z}{2})$.\n",
"\n",
"- $\\overleftarrow{\\partial_t}\\ u(t,x,y,z)$ indicates that the $t$ derivative of $u(t,x,y,z)$ is taken at $u(t-\\frac{\\Delta_t}{2},x,y,z)$.\n",
"\n",
"We usually drop the $(t,x,y,z)$ notation from wavefield variables unless required for clarity of exposition, so that $u(t,x,y,z)$ becomes $u$."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## The Nonlinear operator\n",
"\n",
"The nonlinear operator is the solution to the self adjoint scalar isotropic variable density visco- acoustic wave equation shown immediately below, and maps the velocity model vector $m$ into the receiver wavefield vector $r$.\n",
"\n",
"$$\n",
"\\frac{b}{m^2} \\left( \\frac{\\omega_c}{Q}\\overleftarrow{\\partial_t}\\ u + \\partial_{tt}\\ u \\right) =\n",
" \\overleftarrow{\\partial_x}\\left(b\\ \\overrightarrow{\\partial_x}\\ u \\right) +\n",
" \\overleftarrow{\\partial_y}\\left(b\\ \\overrightarrow{\\partial_y}\\ u \\right) +\n",
" \\overleftarrow{\\partial_z}\\left(b\\ \\overrightarrow{\\partial_z}\\ u \\right) + q\n",
"$$\n",
"\n",
"In operator notation, where the operator is nonlinear with respect to model $m$ to the left of semicolon inside the square brackets, and linear with respect to the source $q$ to the right of semicolon inside the square brackets.\n",
"\n",
"$$\n",
"F[m; q] = r\n",
"$$"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## The Jacobian operator\n",
"\n",
"In this section we linearize about a background model and take the derivative of the nonlinear operator to obtain the Jacobian forward operator. \n",
"\n",
"In operator notation, where the derivative of the modeling operator is now linear in the model perturbation vector $\\delta m$, the Jacobian operator maps a perturbation in the velocity model $\\delta m$ into a perturbation in the receiver wavefield $\\delta r$.\n",
"\n",
"$$\n",
"\\nabla F[m; q]\\ \\delta m = \\delta r\n",
"$$\n",
"\n",
"#### 1. We begin by simplifying notation \n",
"To simplify the treatment below we introduce the operator $L_t[\\cdot]$, accounting for the time derivatives inside the parentheses on the left hand side of the wave equation. \n",
"\n",
"$$\n",
"L_t[\\cdot] \\equiv \\frac{\\omega_c}{Q} \\overleftarrow{\\partial_t}[\\cdot] + \\partial_{tt}[\\cdot]\n",
"$$\n",
"\n",
"Next we re-write the wave equation using this notation.\n",
"\n",
"$$\n",
"\\frac{b}{m^2} L_t[u] =\n",
" \\overleftarrow{\\partial_x}\\left(b\\ \\overrightarrow{\\partial_x}\\ u \\right) +\n",
" \\overleftarrow{\\partial_y}\\left(b\\ \\overrightarrow{\\partial_y}\\ u \\right) +\n",
" \\overleftarrow{\\partial_z}\\left(b\\ \\overrightarrow{\\partial_z}\\ u \\right) + q\n",
"$$\n",
"\n",
"#### 2. Linearize\n",
"To linearize we treat the total model as the sum of background and perturbation models $\\left(m = m_0 + \\delta m\\right)$, and the total pressure as the sum of background and perturbation pressures $\\left(u = u_0 + \\delta u\\right)$.\n",
"\n",
"$$\n",
"\\frac{b}{(m_0+\\delta m)^2} L_t[u_0+\\delta u] =\n",
" \\overleftarrow{\\partial_x}\\left(b\\ \\overrightarrow{\\partial_x} (u_0+\\delta u) \\right) +\n",
" \\overleftarrow{\\partial_y}\\left(b\\ \\overrightarrow{\\partial_y} (u_0+\\delta u) \\right) +\n",
" \\overleftarrow{\\partial_z}\\left(b\\ \\overrightarrow{\\partial_z} (u_0+\\delta u) \\right) + q\n",
"$$\n",
"\n",
"Note that *model parameters* for this variable density isotropic visco-acoustic physics is only velocity, we do not treat perturbations to density.\n",
"\n",
"We also write the PDE for the background model, which we subtract after linearization to simplify the final expression. \n",
"\n",
"$$\n",
"\\frac{b}{m_0^2} L_t[u_0] =\n",
" \\overleftarrow{\\partial_x}\\left(b\\ \\overrightarrow{\\partial_x} u_0 \\right) +\n",
" \\overleftarrow{\\partial_y}\\left(b\\ \\overrightarrow{\\partial_y} u_0 \\right) +\n",
" \\overleftarrow{\\partial_z}\\left(b\\ \\overrightarrow{\\partial_z} u_0 \\right) + q\n",
"$$\n",
"\n",
"#### 3. Take derivative w.r.t. model parameters\n",
"Next we take the derivative with respect to velocity, keep only terms up to first order in the perturbations, subtract the background model PDE equation, and finally arrive at the following linearized equation:\n",
"\n",
"$$\n",
"\\frac{b}{m_0^2} L_t\\left[\\delta u\\right] =\n",
" \\overleftarrow{\\partial_x}\\left(b\\ \\overrightarrow{\\partial_x} \\delta u \\right) +\n",
" \\overleftarrow{\\partial_y}\\left(b\\ \\overrightarrow{\\partial_y} \\delta u \\right) +\n",
" \\overleftarrow{\\partial_z}\\left(b\\ \\overrightarrow{\\partial_z} \\delta u \\right) + \n",
" \\frac{2\\ b\\ \\delta m}{m_0^3} L_t\\left[u_0\\right]\n",
"$$\n",
"\n",
"Note that the source $q$ in the original equation above has disappeared due to subtraction of the background PDE, and has been replaced by the *Born source*: \n",
"\n",
"$$\n",
"q = \\frac{\\displaystyle 2\\ b\\ \\delta m}{\\displaystyle m_0^3} L_t\\left[u_0\\right]\n",
"$$\n",
"\n",
"This is the same equation as used for the nonlinear forward, only now in the perturbed wavefield $\\delta u$ and with the Born source."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## The adjoint of the Jacobian operator\n",
"\n",
"In this section we introduce the adjoint of the Jacobian operator we derived above. The Jacobian adjoint operator maps a perturbation in receiver wavefield $\\delta r$ into a perturbation in velocity model $\\delta m$. In operator notation:\n",
"\n",
"$$\n",
"\\bigl( \\nabla F[m; q] \\bigr)^\\top\\ \\delta r = \\delta m\n",
"$$\n",
"\n",
"#### 1. Solve the time reversed wave equation with the receiver perturbation as source\n",
"The PDE for the adjoint of the Jacobian is solved for the perturbation to the pressure wavefield $\\delta u$ by using the same wave equation as the nonlinear forward and the Jacobian forward, with the time reversed receiver wavefield perturbation $\\widetilde{\\delta r}$ injected as source. \n",
"\n",
"Note that we use $\\widetilde{\\delta u}$ and $\\widetilde{\\delta r}$ to indicate that we solve this finite difference evolution time reversed.\n",
"\n",
"$$\n",
"\\frac{b}{m_0^2} L_t\\left[\\widetilde{\\delta u}\\right] =\n",
" \\overleftarrow{\\partial_x}\\left(b\\ \\overrightarrow{\\partial_x}\\ \\widetilde{\\delta u} \\right) +\n",
" \\overleftarrow{\\partial_y}\\left(b\\ \\overrightarrow{\\partial_y}\\ \\widetilde{\\delta u} \\right) +\n",
" \\overleftarrow{\\partial_z}\\left(b\\ \\overrightarrow{\\partial_z}\\ \\widetilde{\\delta u} \\right) + \n",
" \\widetilde{\\delta r}\n",
"$$\n",
"\n",
"#### 2. Compute zero lag correlation \n",
"\n",
"We compute the perturbation to the velocity model by zero lag correlation of the wavefield perturbation $\\widetilde{\\delta u}$ solved in step 1 as shown in the following expression: \n",
"\n",
"$$\n",
"\\delta m(x,y,z) = \\sum_t \n",
" \\left\\{ \n",
" \\widetilde{\\delta u}(t,x,y,z)\\ \n",
" \\frac{\\displaystyle 2\\ b}{\\displaystyle m_0^3} L_t\\bigl[u_0(t,x,y,z)\\bigr]\n",
" \\right\\}\n",
"$$\n",
"\n",
"Note that this correlation can be more formally derived by examining the equations for two Green's functions, one for the background model ($m_0$) and wavefield ($u_0$), and one for for the total model $(m_0 + \\delta m)$ and wavefield $(u_0 + \\delta u)$, and subtracting to derive the equation for Born scattering.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Implementation\n",
"\n",
"Next we assemble the Devito objects needed to implement these linearized operators."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Imports \n",
"\n",
"We have grouped all imports used in this notebook here for consistency."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"from examples.seismic import RickerSource, Receiver, TimeAxis\n",
"from devito import (Grid, Function, TimeFunction, SpaceDimension, Constant, \n",
" Eq, Operator, solve, configuration, norm)\n",
"from devito.finite_differences import Derivative\n",
"from devito.builtins import gaussian_smooth\n",
"from examples.seismic.self_adjoint import setup_w_over_q\n",
"import matplotlib as mpl\n",
"import matplotlib.pyplot as plt\n",
"from matplotlib import cm\n",
"from timeit import default_timer as timer\n",
"\n",
"# These lines force images to be displayed in the notebook, and scale up fonts \n",
"%matplotlib inline\n",
"mpl.rc('font', size=14)\n",
"\n",
"# Make white background for plots, not transparent\n",
"plt.rcParams['figure.facecolor'] = 'white'\n",
"\n",
"# We define 32 bit floating point as the precision type \n",
"dtype = np.float32\n",
"\n",
"# Set logging to debug, captures statistics on the performance of operators\n",
"# configuration['log-level'] = 'DEBUG'\n",
"configuration['log-level'] = 'INFO'"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Instantiate the Devito grid for a two dimensional problem\n",
"\n",
"We define the grid the same as in the previous notebook outlining implementation for the nonlinear forward."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"shape; (301, 301)\n",
"origin; (0.0, 0.0)\n",
"spacing; (10.0, 10.0)\n",
"extent; (3000.0, 3000.0)\n",
"\n",
"shape_pad; [401 401]\n",
"origin_pad; (-500.0, -500.0)\n",
"extent_pad; (4000.0, 4000.0)\n",
"\n",
"grid.shape; (401, 401)\n",
"grid.extent; (4000.0, 4000.0)\n",
"grid.spacing_map; {h_x: 10.0, h_z: 10.0}\n"
]
}
],
"source": [
"# Define dimensions for the interior of the model\n",
"nx,nz = 301,301\n",
"dx,dz = 10.0,10.0 # Grid spacing in m\n",
"shape = (nx, nz) # Number of grid points\n",
"spacing = (dx, dz) # Domain size is now 5 km by 5 km\n",
"origin = (0., 0.) # Origin of coordinate system, specified in m.\n",
"extent = tuple([s*(n-1) for s, n in zip(spacing, shape)])\n",
"\n",
"# Define dimensions for the model padded with absorbing boundaries\n",
"npad = 50 # number of points in absorbing boundary region (all sides)\n",
"nxpad,nzpad = nx + 2 * npad, nz + 2 * npad\n",
"shape_pad = np.array(shape) + 2 * npad\n",
"origin_pad = tuple([o - s*npad for o, s in zip(origin, spacing)])\n",
"extent_pad = tuple([s*(n-1) for s, n in zip(spacing, shape_pad)])\n",
"\n",
"# Define the dimensions \n",
"# Note if you do not specify dimensions, you get in order x,y,z\n",
"x = SpaceDimension(name='x', spacing=Constant(name='h_x', \n",
" value=extent_pad[0]/(shape_pad[0]-1)))\n",
"z = SpaceDimension(name='z', spacing=Constant(name='h_z', \n",
" value=extent_pad[1]/(shape_pad[1]-1)))\n",
"\n",
"# Initialize the Devito grid \n",
"grid = Grid(extent=extent_pad, shape=shape_pad, origin=origin_pad, \n",
" dimensions=(x, z), dtype=dtype)\n",
"\n",
"print(\"shape; \", shape)\n",
"print(\"origin; \", origin)\n",
"print(\"spacing; \", spacing)\n",
"print(\"extent; \", extent)\n",
"print(\"\")\n",
"print(\"shape_pad; \", shape_pad)\n",
"print(\"origin_pad; \", origin_pad)\n",
"print(\"extent_pad; \", extent_pad)\n",
"\n",
"print(\"\")\n",
"print(\"grid.shape; \", grid.shape)\n",
"print(\"grid.extent; \", grid.extent)\n",
"print(\"grid.spacing_map;\", grid.spacing_map)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Define velocity, buoyancy and $\\frac{\\omega_c}{Q}$ model parameters\n",
"\n",
"We have the following constants and fields to define:\n",
"\n",
"| Symbol | Description |\n",
"|:---:|:---|\n",
"| $$m0(x,z)$$ | Background velocity model |\n",
"| $$\\delta m(x,z)$$ | Perturbation to velocity model |\n",
"| $$b(x,z)=\\frac{1}{\\rho(x,z)}$$ | Buoyancy (reciprocal density) |\n",
"| $$\\omega_c = 2 \\pi f_c$$ | Center angular frequency |\n",
"| $$\\frac{1}{Q(x,z)}$$ | Inverse Q model used in the modeling system |\n"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Operator `WOverQ_Operator` run in 0.01 s\n"
]
}
],
"source": [
"# NBVAL_IGNORE_OUTPUT\n",
"\n",
"# Create the velocity and buoyancy fields as in the nonlinear notebook \n",
"space_order = 8\n",
"\n",
"# Wholespace velocity\n",
"m0 = Function(name='m0', grid=grid, space_order=space_order)\n",
"m0.data[:] = 1.5\n",
"\n",
"# Perturbation to velocity: a square offset from the center of the model\n",
"dm = Function(name='dm', grid=grid, space_order=space_order)\n",
"size = 10\n",
"x0 = shape_pad[0]//2\n",
"z0 = shape_pad[1]//2\n",
"dm.data[:] = 0.0\n",
"dm.data[x0-size:x0+size, z0-size:z0+size] = 1.0\n",
"\n",
"# Constant density\n",
"b = Function(name='b', grid=grid, space_order=space_order)\n",
"b.data[:,:] = 1.0 / 1.0\n",
"\n",
"# Initialize the attenuation profile for Q=100 model\n",
"fpeak = 0.010\n",
"w = 2.0 * np.pi * fpeak\n",
"qmin = 0.1\n",
"qmax = 1000.0\n",
"wOverQ = Function(name='wOverQ', grid=grid, space_order=space_order)\n",
"setup_w_over_q(wOverQ, w, qmin, 100.0, npad)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Define the simulation time range and the acquisition geometry \n",
"\n",
"#### Simulation time range: \n",
"\n",
"In this notebook we run 3 seconds of simulation using the sample rate related to the CFL condition as implemented in ```examples/seismic/self_adjoint/utils.py```. \n",
"\n",
"We also use the convenience ```TimeRange``` as defined in ```examples/seismic/source.py```.\n",
"\n",
"#### Acquisition geometry: \n",
"\n",
"**source**:\n",
"- X coordinate: left sode of model\n",
"- Z coordinate: middle of model\n",
"- We use a 10 Hz center frequency [RickerSource](https://github.com/devitocodes/devito/blob/master/examples/seismic/source.py#L280) wavelet source as defined in ```examples/seismic/source.py```\n",
"\n",
"**receivers**:\n",
"- X coordinate: right side of model\n",
"- Z coordinate: vertical line in model\n",
"- We use a vertical line of [Receivers](https://github.com/devitocodes/devito/blob/master/examples/seismic/source.py#L80) as defined with a ```PointSource``` in ```examples/seismic/source.py```"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"def compute_critical_dt(v):\n",
" \"\"\"\n",
" Determine the temporal sampling to satisfy CFL stability.\n",
" This method replicates the functionality in the Model class.\n",
" Note we add a safety factor, reducing dt by a factor 0.75 due to the\n",
" w/Q attentuation term.\n",
" Parameters\n",
" ----------\n",
" v : Function\n",
" velocity\n",
" \"\"\"\n",
" coeff = 0.38 if len(v.grid.shape) == 3 else 0.42\n",
" dt = 0.75 * v.dtype(coeff * np.min(v.grid.spacing) / (np.max(v.data)))\n",
" return v.dtype(\"%.5e\" % dt)"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Time min, max, dt, num; 0.000000 1200.000000 2.100000 572\n",
"time_range; TimeAxis: start=0, stop=1201.2, step=2.1, num=573\n",
"src_coordinate X; +752.0000\n",
"src_coordinate Z; +1505.0000\n",
"rec_coordinates X min/max; +2257.0000 +2257.0000\n",
"rec_coordinates Z min/max; +0.0000 +3000.0000\n"
]
}
],
"source": [
"t0 = 0.0 # Simulation time start\n",
"tn = 1200.0 # Simulation time end (1 second = 1000 msec)\n",
"dt = compute_critical_dt(m0)\n",
"time_range = TimeAxis(start=t0, stop=tn, step=dt)\n",
"print(\"Time min, max, dt, num; %10.6f %10.6f %10.6f %d\" % (t0, tn, dt, int(tn//dt) + 1))\n",
"print(\"time_range; \", time_range)\n",
"\n",
"# Source at 1/4 X, 1/2 Z, Ricker with 10 Hz center frequency\n",
"src_nl = RickerSource(name='src_nl', grid=grid, f0=fpeak, npoint=1, time_range=time_range)\n",
"src_nl.coordinates.data[0,0] = dx * 1 * nx//4\n",
"src_nl.coordinates.data[0,1] = dz * shape[1]//2\n",
"\n",
"# Receivers at 3/4 X, line in Z\n",
"rec_nl = Receiver(name='rec_nl', grid=grid, npoint=nz, time_range=time_range)\n",
"rec_nl.coordinates.data[:,0] = dx * 3 * nx//4\n",
"rec_nl.coordinates.data[:,1] = np.linspace(0.0, dz*(nz-1), nz)\n",
"\n",
"print(\"src_coordinate X; %+12.4f\" % (src_nl.coordinates.data[0,0]))\n",
"print(\"src_coordinate Z; %+12.4f\" % (src_nl.coordinates.data[0,1]))\n",
"print(\"rec_coordinates X min/max; %+12.4f %+12.4f\" % \\\n",
" (np.min(rec_nl.coordinates.data[:,0]), np.max(rec_nl.coordinates.data[:,0])))\n",
"print(\"rec_coordinates Z min/max; %+12.4f %+12.4f\" % \\\n",
" (np.min(rec_nl.coordinates.data[:,1]), np.max(rec_nl.coordinates.data[:,1])))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Plot the model \n",
"\n",
"We plot the following ```Functions```:\n",
"- Background Velocity\n",
"- Background Density\n",
"- Velocity perturbation\n",
"- Q Model\n",
"\n",
"Each subplot also shows:\n",
"- The location of the absorbing boundary as a dotted line\n",
"- The source location as a red star\n",
"- The line of receivers as a black vertical line"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"