Path to the directory where the Clawpack output is that is to be
plotted.
-
-
-plotdir:str
+
+
+plotdir : str
Path to the directory where hardcopy files of plots are to be put.
-
-
-overwrite:bool
+
+
+overwrite : bool
Ok to overwrite old plotdir? Default is True
-
-
-afterframe:strorfunction
+
+
+afterframe : str or function
Function or string to be executed after producing all plots for each
frame. If a string, this string is executed using exec. If a
function, it should be defined to have a single argument
@@ -130,9 +127,9 @@
Clear all plot parameters. Useful as the first command in a setplot
function to make sure previous parameters are cleared if the file is
changed and the function is re-executed in an interactive session.
Return ClawPlotAxes object with the specified name.
If figname==None then search over all figures and return None if
it is not found or the name is not unique.
Return ClawPlotItem object with the specified name.
If axesname==None and/or figname==None
then search over all figures and/or axes and return None if
it is not found or the name is not unique.
Figure number, by default the next unused number will be used (starting
at 1001).
This is usually set as an argument to the new_plotfigure function
of a ClawPlotData object
-
-
-kwargs:dictionary
+
+
+kwargs : dictionary
A dictionary of keyword arguments for the figure command.
For example:
Special attributes for plot_type = ‘1d_from_2d_data’¶
+
+
+map_2d_to_1d : function
Example: In a 2d computation where the solution q[:,:,0] should be
radially symmetric about (x,y)=(0,0), the following will result in a
scatter plot of the cell values q[i,j,0] vs. the radius r(i,j):
@@ -293,62 +290,62 @@
Special attributes for plot_type = ‘1d_fill_between’current_data for a description of the current_data argument.
-
-
-
Special attributes for all 2d plots, plot_type = ‘2d…’¶
-
-
-celledges_show:bool
+
+
+
Special attributes for all 2d plots, plot_type = ‘2d…’¶
+
+
+celledges_show : bool
If True, draw the cell edges on the plot.
The attribute ‘amr_celledges_show’ should be used for AMR computations
to specify that cell edges should be shown on some levels and not
others. See AMR Attributes.
-
-
-patchedges_show:bool
+
+
+patchedges_show : bool
If True, draw the edges of patches, mostly useful in AMR computations.
In general you should specify pcolor_cmin and pcolor_cmax to
specify the range of q values over which the colormap applies. If they
are not specified they will be chosen automatically and may vary from
@@ -407,16 +404,16 @@
Many attributes listed above also have a second related attribute with the
same name pre-pended with amr_. If this attribute is set, it should be a
list whose elements are of the type specified by the original name, and the
@@ -445,28 +442,27 @@
Clawpack stands for “Conservation Laws Package” and was initially developed
for linear and nonlinear hyperbolic systems of conservation laws, with a
focus on implementing high-resolution Godunov type methods using limiters in
@@ -112,14 +109,14 @@
Navigation
and resulting plots.
See the Bibliography for some pointers to papers describing Clawpack and
the algorithms used in more detail.
A new approach to flagging cells for refinement was introduced in Clawpack
5.6.0 – using the solution to an adjoint problem to determine what cells in
the forward solution should be refined because these cells may have an impact
@@ -223,8 +219,8 @@
Navigation
simply inner products of the forward and adjoint solutions that are computed
and a cell is flagged for refinement in cells where the magnitude of the
inner project is greater than amrdata.flag2refine_tol.
-
-
in non-conservation form and is solved using the q-wave formulation in which
jumps in the the solution vector are split into eigenvectors, rather than
jumps in the flux. See the comments in the rpn2 solver for more details.
-
-
+
The basic adaptive refinment strategy used in AMRClaw Description and Detailed Contents is
to refine on logically rectangular patches. A single Level 1 grid covers
the entire domain (usually — if it is too large it may be split into
@@ -134,8 +131,8 @@
Navigation
originally written by Marsha Berger for gas dynamics, and merged in Clawpack
in the early days of Clawpack development by MJB and RJL. The algorithms
used in AMRClaw are described more fully in [BergerLeVeque98].
-
-
Consider a Level k > 1 grid for which we need ghost cells all around the
boundary at the start of each time step on this level. The same procedure
is used at other levels.
@@ -178,9 +175,9 @@
Navigation
the same refinement level at the opposite side of the domain to copy ghost
cell values from, and if so which grid and what index corresponds to the
desired location.
-
-
-
Every few time steps on the coarsest level it is generally necessary to
revise modify the regions of refinement at all levels, for example to follow
a propagating shock wave. This is done by
@@ -221,24 +218,23 @@
Navigation
interpolation is performed based on the Level k grids.
The AMRClaw version of Clawpack provides Adaptive Mesh Refinement (AMR)
capabilities in 2 and 3 space dimensions. (The two-dimensional code can
also be used for 1-dimensional problems, see AMRClaw for 1d problems.)
AMRClaw has been extended to support one-dimensional AMR directly.
The setrun.py file has the same form as for 2d AMRClaw, with the obvious
changes to eliminate the y-direction. See Sample setrun.py module for AMRClaw.
For some examples, see the 1d examples in $CLAW/amrclaw/examples/ and in
gallery_classic_amrclaw.
The two-dimensional code can also be used for 1-dimensional problems, by
setting num_cells[1] = 1 so there is only one cell in the y direction.
Also set refinement_ratios_y = [1,1,1, …] so that refinement is not done
@@ -106,11 +103,10 @@
In Fortran versions of Clawpack-5, the main documentation page
for an application is often found
in a file named README.rst, which can be written using the markup language
@@ -105,18 +102,17 @@
Navigation
which also converts code files into html files for easy browsing.
A list of code files in the directory is automatically generated by
convert_readme.py and links inserted in README.html.
The directory apps/notebooks contains a number of notebooks that illustrate
various aspects of Clawpack. Many of these are also visible in rendered html
form in the
Gallery of Notebooks.
Select the Clawpack image and then
click on the Launch button on this page to start launching an instance
based on this AMI. This means a virtual machine will be started for you,
@@ -140,9 +137,9 @@
Click Close on the next page to
go back to the Management Console. Click on Instances on the left menu
and you should see a list of instance you
@@ -168,9 +165,9 @@
If you run Clawpack on your instance then you will probably want to view the
results. There are at least three possible approaches (see Plotting with Visclaw
for general information about plotting in Clawpack):
Once you are done computing for the day, you will probably want to stop your
instance so you won’t be charged while it’s sitting idle. You can do this
by selecting the instance from the Management Console / Instances, and then
select Stop from the Instance Actions menu.
You can restart it later and it will be in the same state you left it in.
But note that it will probably have a new Public DNS!
If you add additional software and want to save a disk image of your
improved virtual machine (e.g. in order to launch additional images in the
future to run multiple jobs at once), simply click on Create Image (EBS
AMI) from the Instance Actions menu.
first checked and if there is no such file in this directory then
the environment variable B4RUN is checked, which can be set to
the full path of a b4run.py file that you wish to use globally, for example.
-
+
Below are linkes to the default b4step library routines for Classic and AMRClaw.
By default these do nothing. If you wish to specify aux arrays you will
need to copy one of these files to your application directory and modify it
@@ -89,8 +86,8 @@
Navigation
(Note: these links are for the version checked in to the master branch.
You can select a different branch or tag from the GitHub page.)
In GeoClaw, there is a library routine that sets aux(1,i,j) to the cell
average of the bathymetry, aux(2,i,j) to the ratio of the true cell area
to dx*dy (the capacity function), and aux(3,i,j) to the length ratio of
@@ -99,11 +96,10 @@
Boundary conditions are imposed each time step by filling ghost cells
adjacent to the edge of each grid patch. See Chapter 4 of [LeVeque-FVMHP]
for more details.
@@ -139,8 +136,8 @@
Navigation
best to put the modified version of bcN.f in the application directory and
modify the Makefile to point to the modified version.
See User-defined boundary conditions below.
-
-
When AMR is used, any interior patch edges (not at a domain boundary) are
filled automatically each time step, either by copying from adjacent
patches at the same level or by interpolating (in both space and
@@ -159,9 +156,9 @@
Navigation
within the domain (in which case ghost cell values are automatically set by
the AMR code and the user bcNamr routine should leave these values
alone.
-
-
-
For tsunami modeling or other geophysical flows over topography the
computational domain has artificial boundaries that are placed sufficiently
far from the region of interest that any flow or waves leaving the domain
@@ -175,9 +172,9 @@
Navigation
The library routine $CLAW/geoclaw/src/2d/shallow/bc2amr.f is modified from
the amrclaw version only by extrapolating the depth at the boundaries
into ghost cells.
-
-
-
Boundary conditions for clamshell grids on the sphere¶
+
+
+
Boundary conditions for clamshell grids on the sphere¶
In 2D AMRClaw and GeoClaw, an additional option is available for bc_lower
and bc_upper that is implemented in the library routines:
@@ -192,9 +189,9 @@
Navigation
See also [BergerCalhounHelzelLeVeque] for further examples.
If none of the boundary conditions described above is suitable at one or
more boundaries of the domain, then you will have to modify the library
routine to implement the desired boundary condition.
@@ -216,11 +213,10 @@
Navigation
cells extend outside the physical domain and which are filled automatically
from adjacent grid patches or by interpolation from coarser patches if they
are interior to the domain.
-
-
+
Papers describing the Clawpack software and algorithmsYear ={2002}}
-
-
-[BergerColella89]
-
M. J. Berger and P Colella. 1989. Local adaptive mesh refinement for
+
+
BergerColella89
+
M. J. Berger and P Colella. 1989. Local adaptive mesh refinement for
shock hydrodynamics. J. Comput. Phys. 82, 64–84.
-
-
-[BergerGeorgeLeVequeMandli11]
-
M. J. Berger, D. L. George, R. J. LeVeque and K. M. Mandli,
+
+
BergerGeorgeLeVequeMandli11
+
M. J. Berger, D. L. George, R. J. LeVeque and K. M. Mandli,
The GeoClaw software for depth-averaged flows with adaptive refinement,
Advances in Water Resources 34 (2011), pp. 1195-1206.
Papers describing the Clawpack software and algorithmsYear ={1998}}
-
-
-[BergerOliger84]
-
M. J. Berger and J. Oliger. 1984. Adaptive mesh refinement for
+
+
BergerOliger84
+
M. J. Berger and J. Oliger. 1984. Adaptive mesh refinement for
hyperbolic partial differential equations. J. Comput. Phys. 53,
484–512.
-
-
-[BergerRigoutsis91]
-
M. J. Berger and I. Rigoutsos. 1991. An Algorithm for Point Clustering
+
+
BergerRigoutsis91
+
M. J. Berger and I. Rigoutsos. 1991. An Algorithm for Point Clustering
and Grid Generation. IEEE Trans. Sys. Man & Cyber. 21, 1278–1286.
-
-
-[Davis2018]
-
B. N. Davis, 2018. Adjoint-Guided Adaptive Mesh Refinement for
+
+
Davis2018
+
B. N. Davis, 2018. Adjoint-Guided Adaptive Mesh Refinement for
Hyperbolic Systems of Equations,
PhD Thesis, University of Washington.
link
-
-
-[DavisLeVeque2016]
-
B. N. Davis and R. J. LeVeque. 2016. Adjoint Methods for Guiding
+
+
DavisLeVeque2016
+
B. N. Davis and R. J. LeVeque. 2016. Adjoint Methods for Guiding
Adaptive Mesh Refinement in Tsunami Modeling, 2016.
Pure Appl. Geophys. 173 (2016), pp. 4055-4074.
More info
-
-
-[DavisLeVeque2018]
-
B. N. Davis and R. J. LeVeque. 2018. Analysis and Performance Evaluation of
+
+
DavisLeVeque2018
+
B. N. Davis and R. J. LeVeque. 2018. Analysis and Performance Evaluation of
Adjoint-Guided Adaptive Mesh Refinement for Linear Hyperbolic PDEs
Using Clawpack, Preprint,
More info
Papers describing the Clawpack software and algorithmsYear ={2011}}
-
-
-[KetParLev13]
-
D. I. Ketcheson, Matteo Parsani, and R J LeVeque, 2013,
+
+
KetParLev13
+
D. I. Ketcheson, Matteo Parsani, and R J LeVeque, 2013,
High-order Wave Propagation Algorithms for Hyperbolic Systems,
SIAM Journal on Scientific Computing, 35(1):A351-A377 (2013)
@article{KetParLev13,
@@ -269,10 +251,9 @@
Papers describing the Clawpack software and algorithmsYear ={2013}}
-
-
-[KetchesonMandliEtAl]
-
David I. Ketcheson, Kyle T. Mandli, Aron J. Ahmadia, Amal Alghamdi, Manuel
+
+
KetchesonMandliEtAl
+
David I. Ketcheson, Kyle T. Mandli, Aron J. Ahmadia, Amal Alghamdi, Manuel
Quezada de Luna, Matteo Parsani, Matthew G. Knepley, and Matthew Emmett, 2012,
PyClaw: Accessible, Extensible, Scalable Tools for Wave Propagation Problems,
SIAM Journal on Scientific Computing, 34(4):C210-C231
@@ -289,10 +270,9 @@
Papers describing the Clawpack software and algorithmsYear ={2012}}
-
-
-[MandliEtAl2016]
-
Kyle T. Mandli, Aron J. Ahmadia, Marsha Berger, Donna Calhoun, David L.
+
+
MandliEtAl2016
+
Kyle T. Mandli, Aron J. Ahmadia, Marsha Berger, Donna Calhoun, David L.
George, Yiannis Hadjimichael, David I. Ketcheson, Grady I. Lemoine, Randall J. LeVeque,
Clawpack: building an open source ecosystem for solving hyperbolic PDEs
PeerJ Computer Science 2 (2016), e68:
Follow the links to see changes that have been made to the master branch of
each repository since the last release (v5.9.0) on August 26, 2022.
These changes should appear in the next release. If you need them now,
@@ -86,43 +83,43 @@
Navigation
To see documentation that has already been developed to accompany any new
features listed below, click on the “dev” branch of the documentation, in
the menu on the left hand side of this page.
For users who want to migrate code from 4.3 to 5.0, this may be a useful
intermediate step, followed by migration from 4.6 to 5.0 using the script
described at Converting from Clawpack 4.6 to 5.0.
@@ -106,10 +103,9 @@
Navigation
This migration script from 4.3 to 4.6 only works for classic
(single grid) codes in 1d and 2d, not AMRClaw codes.
The rest of this page concerns the Fortran components of Clawpack.
There is no complete list of changes since it has evolved to be very
different from the 4.x version of Clawpack in organization, but some of
@@ -191,11 +188,11 @@
Path to the directory where the Clawpack output is that is to be
plotted.
-
-
-plotdir:str
+
+
+plotdir : str
Path to the directory where hardcopy files of plots are to be put.
-
-
-overwrite:bool
+
+
+overwrite : bool
Ok to overwrite old plotdir? Default is True
-
-
-afterframe:strorfunction
+
+
+afterframe : str or function
Function or string to be executed after producing all plots for each
frame. If a string, this string is executed using exec. If a
function, it should be defined to have a single argument
@@ -130,9 +127,9 @@
Clear all plot parameters. Useful as the first command in a setplot
function to make sure previous parameters are cleared if the file is
changed and the function is re-executed in an interactive session.
Return ClawPlotAxes object with the specified name.
If figname==None then search over all figures and return None if
it is not found or the name is not unique.
Return ClawPlotItem object with the specified name.
If axesname==None and/or figname==None
then search over all figures and/or axes and return None if
it is not found or the name is not unique.
Figure number, by default the next unused number will be used (starting
at 1001).
This is usually set as an argument to the new_plotfigure function
of a ClawPlotData object
-
-
-kwargs:dictionary
+
+
+kwargs : dictionary
A dictionary of keyword arguments for the figure command.
For example:
Special attributes for plot_type = ‘1d_from_2d_data’¶
+
+
+map_2d_to_1d : function
Example: In a 2d computation where the solution q[:,:,0] should be
radially symmetric about (x,y)=(0,0), the following will result in a
scatter plot of the cell values q[i,j,0] vs. the radius r(i,j):
@@ -293,62 +290,62 @@
Special attributes for plot_type = ‘1d_fill_between’current_data for a description of the current_data argument.
-
-
-
Special attributes for all 2d plots, plot_type = ‘2d…’¶
-
-
-celledges_show:bool
+
+
+
Special attributes for all 2d plots, plot_type = ‘2d…’¶
+
+
+celledges_show : bool
If True, draw the cell edges on the plot.
The attribute ‘amr_celledges_show’ should be used for AMR computations
to specify that cell edges should be shown on some levels and not
others. See AMR Attributes.
-
-
-patchedges_show:bool
+
+
+patchedges_show : bool
If True, draw the edges of patches, mostly useful in AMR computations.
In general you should specify pcolor_cmin and pcolor_cmax to
specify the range of q values over which the colormap applies. If they
are not specified they will be chosen automatically and may vary from
@@ -407,16 +404,16 @@
Many attributes listed above also have a second related attribute with the
same name pre-pended with amr_. If this attribute is set, it should be a
list whose elements are of the type specified by the original name, and the
@@ -445,28 +442,27 @@
Clawpack stands for “Conservation Laws Package” and was initially developed
for linear and nonlinear hyperbolic systems of conservation laws, with a
focus on implementing high-resolution Godunov type methods using limiters in
@@ -112,14 +109,14 @@
Navigation
and resulting plots.
See the Bibliography for some pointers to papers describing Clawpack and
the algorithms used in more detail.
A new approach to flagging cells for refinement was introduced in Clawpack
5.6.0 – using the solution to an adjoint problem to determine what cells in
the forward solution should be refined because these cells may have an impact
@@ -223,8 +219,8 @@
Navigation
simply inner products of the forward and adjoint solutions that are computed
and a cell is flagged for refinement in cells where the magnitude of the
inner project is greater than amrdata.flag2refine_tol.
-
-
in non-conservation form and is solved using the q-wave formulation in which
jumps in the the solution vector are split into eigenvectors, rather than
jumps in the flux. See the comments in the rpn2 solver for more details.
-
-
+
The basic adaptive refinment strategy used in AMRClaw Description and Detailed Contents is
to refine on logically rectangular patches. A single Level 1 grid covers
the entire domain (usually — if it is too large it may be split into
@@ -134,8 +131,8 @@
Navigation
originally written by Marsha Berger for gas dynamics, and merged in Clawpack
in the early days of Clawpack development by MJB and RJL. The algorithms
used in AMRClaw are described more fully in [BergerLeVeque98].
-
-
Consider a Level k > 1 grid for which we need ghost cells all around the
boundary at the start of each time step on this level. The same procedure
is used at other levels.
@@ -178,9 +175,9 @@
Navigation
the same refinement level at the opposite side of the domain to copy ghost
cell values from, and if so which grid and what index corresponds to the
desired location.
-
-
-
Every few time steps on the coarsest level it is generally necessary to
revise modify the regions of refinement at all levels, for example to follow
a propagating shock wave. This is done by
@@ -221,24 +218,23 @@
Navigation
interpolation is performed based on the Level k grids.
The AMRClaw version of Clawpack provides Adaptive Mesh Refinement (AMR)
capabilities in 2 and 3 space dimensions. (The two-dimensional code can
also be used for 1-dimensional problems, see AMRClaw for 1d problems.)
AMRClaw has been extended to support one-dimensional AMR directly.
The setrun.py file has the same form as for 2d AMRClaw, with the obvious
changes to eliminate the y-direction. See Sample setrun.py module for AMRClaw.
For some examples, see the 1d examples in $CLAW/amrclaw/examples/ and in
gallery_classic_amrclaw.
The two-dimensional code can also be used for 1-dimensional problems, by
setting num_cells[1] = 1 so there is only one cell in the y direction.
Also set refinement_ratios_y = [1,1,1, …] so that refinement is not done
@@ -106,11 +103,10 @@
In Fortran versions of Clawpack-5, the main documentation page
for an application is often found
in a file named README.rst, which can be written using the markup language
@@ -105,18 +102,17 @@
Navigation
which also converts code files into html files for easy browsing.
A list of code files in the directory is automatically generated by
convert_readme.py and links inserted in README.html.
The directory apps/notebooks contains a number of notebooks that illustrate
various aspects of Clawpack. Many of these are also visible in rendered html
form in the
Gallery of Notebooks.
Select the Clawpack image and then
click on the Launch button on this page to start launching an instance
based on this AMI. This means a virtual machine will be started for you,
@@ -140,9 +137,9 @@
Click Close on the next page to
go back to the Management Console. Click on Instances on the left menu
and you should see a list of instance you
@@ -168,9 +165,9 @@
If you run Clawpack on your instance then you will probably want to view the
results. There are at least three possible approaches (see Plotting with Visclaw
for general information about plotting in Clawpack):
Once you are done computing for the day, you will probably want to stop your
instance so you won’t be charged while it’s sitting idle. You can do this
by selecting the instance from the Management Console / Instances, and then
select Stop from the Instance Actions menu.
You can restart it later and it will be in the same state you left it in.
But note that it will probably have a new Public DNS!
If you add additional software and want to save a disk image of your
improved virtual machine (e.g. in order to launch additional images in the
future to run multiple jobs at once), simply click on Create Image (EBS
AMI) from the Instance Actions menu.
first checked and if there is no such file in this directory then
the environment variable B4RUN is checked, which can be set to
the full path of a b4run.py file that you wish to use globally, for example.
-
+
Below are linkes to the default b4step library routines for Classic and AMRClaw.
By default these do nothing. If you wish to specify aux arrays you will
need to copy one of these files to your application directory and modify it
@@ -89,8 +86,8 @@
Navigation
(Note: these links are for the version checked in to the master branch.
You can select a different branch or tag from the GitHub page.)
In GeoClaw, there is a library routine that sets aux(1,i,j) to the cell
average of the bathymetry, aux(2,i,j) to the ratio of the true cell area
to dx*dy (the capacity function), and aux(3,i,j) to the length ratio of
@@ -99,11 +96,10 @@
Boundary conditions are imposed each time step by filling ghost cells
adjacent to the edge of each grid patch. See Chapter 4 of [LeVeque-FVMHP]
for more details.
@@ -139,8 +136,8 @@
Navigation
best to put the modified version of bcN.f in the application directory and
modify the Makefile to point to the modified version.
See User-defined boundary conditions below.
-
-
When AMR is used, any interior patch edges (not at a domain boundary) are
filled automatically each time step, either by copying from adjacent
patches at the same level or by interpolating (in both space and
@@ -159,9 +156,9 @@
Navigation
within the domain (in which case ghost cell values are automatically set by
the AMR code and the user bcNamr routine should leave these values
alone.
-
-
-
For tsunami modeling or other geophysical flows over topography the
computational domain has artificial boundaries that are placed sufficiently
far from the region of interest that any flow or waves leaving the domain
@@ -175,9 +172,9 @@
Navigation
The library routine $CLAW/geoclaw/src/2d/shallow/bc2amr.f is modified from
the amrclaw version only by extrapolating the depth at the boundaries
into ghost cells.
-
-
-
Boundary conditions for clamshell grids on the sphere¶
+
+
+
Boundary conditions for clamshell grids on the sphere¶
In 2D AMRClaw and GeoClaw, an additional option is available for bc_lower
and bc_upper that is implemented in the library routines:
@@ -192,9 +189,9 @@
Navigation
See also [BergerCalhounHelzelLeVeque] for further examples.
If none of the boundary conditions described above is suitable at one or
more boundaries of the domain, then you will have to modify the library
routine to implement the desired boundary condition.
@@ -216,11 +213,10 @@
Navigation
cells extend outside the physical domain and which are filled automatically
from adjacent grid patches or by interpolation from coarser patches if they
are interior to the domain.
-
-
+
Papers describing the Clawpack software and algorithmsYear ={2002}}
-
-
-[BergerColella89]
-
M. J. Berger and P Colella. 1989. Local adaptive mesh refinement for
+
+
BergerColella89
+
M. J. Berger and P Colella. 1989. Local adaptive mesh refinement for
shock hydrodynamics. J. Comput. Phys. 82, 64–84.
-
-
-[BergerGeorgeLeVequeMandli11]
-
M. J. Berger, D. L. George, R. J. LeVeque and K. M. Mandli,
+
+
BergerGeorgeLeVequeMandli11
+
M. J. Berger, D. L. George, R. J. LeVeque and K. M. Mandli,
The GeoClaw software for depth-averaged flows with adaptive refinement,
Advances in Water Resources 34 (2011), pp. 1195-1206.
Papers describing the Clawpack software and algorithmsYear ={1998}}
-
-
-[BergerOliger84]
-
M. J. Berger and J. Oliger. 1984. Adaptive mesh refinement for
+
+
BergerOliger84
+
M. J. Berger and J. Oliger. 1984. Adaptive mesh refinement for
hyperbolic partial differential equations. J. Comput. Phys. 53,
484–512.
-
-
-[BergerRigoutsis91]
-
M. J. Berger and I. Rigoutsos. 1991. An Algorithm for Point Clustering
+
+
BergerRigoutsis91
+
M. J. Berger and I. Rigoutsos. 1991. An Algorithm for Point Clustering
and Grid Generation. IEEE Trans. Sys. Man & Cyber. 21, 1278–1286.
-
-
-[Davis2018]
-
B. N. Davis, 2018. Adjoint-Guided Adaptive Mesh Refinement for
+
+
Davis2018
+
B. N. Davis, 2018. Adjoint-Guided Adaptive Mesh Refinement for
Hyperbolic Systems of Equations,
PhD Thesis, University of Washington.
link
-
-
-[DavisLeVeque2016]
-
B. N. Davis and R. J. LeVeque. 2016. Adjoint Methods for Guiding
+
+
DavisLeVeque2016
+
B. N. Davis and R. J. LeVeque. 2016. Adjoint Methods for Guiding
Adaptive Mesh Refinement in Tsunami Modeling, 2016.
Pure Appl. Geophys. 173 (2016), pp. 4055-4074.
More info
-
-
-[DavisLeVeque2018]
-
B. N. Davis and R. J. LeVeque. 2018. Analysis and Performance Evaluation of
+
+
DavisLeVeque2018
+
B. N. Davis and R. J. LeVeque. 2018. Analysis and Performance Evaluation of
Adjoint-Guided Adaptive Mesh Refinement for Linear Hyperbolic PDEs
Using Clawpack, Preprint,
More info
Papers describing the Clawpack software and algorithmsYear ={2011}}
-
-
-[KetParLev13]
-
D. I. Ketcheson, Matteo Parsani, and R J LeVeque, 2013,
+
+
KetParLev13
+
D. I. Ketcheson, Matteo Parsani, and R J LeVeque, 2013,
High-order Wave Propagation Algorithms for Hyperbolic Systems,
SIAM Journal on Scientific Computing, 35(1):A351-A377 (2013)
@article{KetParLev13,
@@ -269,10 +251,9 @@
Papers describing the Clawpack software and algorithmsYear ={2013}}
-
-
-[KetchesonMandliEtAl]
-
David I. Ketcheson, Kyle T. Mandli, Aron J. Ahmadia, Amal Alghamdi, Manuel
+
+
KetchesonMandliEtAl
+
David I. Ketcheson, Kyle T. Mandli, Aron J. Ahmadia, Amal Alghamdi, Manuel
Quezada de Luna, Matteo Parsani, Matthew G. Knepley, and Matthew Emmett, 2012,
PyClaw: Accessible, Extensible, Scalable Tools for Wave Propagation Problems,
SIAM Journal on Scientific Computing, 34(4):C210-C231
@@ -289,10 +270,9 @@
Papers describing the Clawpack software and algorithmsYear ={2012}}
-
-
-[MandliEtAl2016]
-
Kyle T. Mandli, Aron J. Ahmadia, Marsha Berger, Donna Calhoun, David L.
+
+
MandliEtAl2016
+
Kyle T. Mandli, Aron J. Ahmadia, Marsha Berger, Donna Calhoun, David L.
George, Yiannis Hadjimichael, David I. Ketcheson, Grady I. Lemoine, Randall J. LeVeque,
Clawpack: building an open source ecosystem for solving hyperbolic PDEs
PeerJ Computer Science 2 (2016), e68:
Follow the links to see changes that have been made to the master branch of
each repository since the last release (v5.9.0) on August 26, 2022.
These changes should appear in the next release. If you need them now,
@@ -86,43 +83,43 @@
Navigation
To see documentation that has already been developed to accompany any new
features listed below, click on the “dev” branch of the documentation, in
the menu on the left hand side of this page.
For users who want to migrate code from 4.3 to 5.0, this may be a useful
intermediate step, followed by migration from 4.6 to 5.0 using the script
described at Converting from Clawpack 4.6 to 5.0.
@@ -106,10 +103,9 @@
Navigation
This migration script from 4.3 to 4.6 only works for classic
(single grid) codes in 1d and 2d, not AMRClaw codes.
The rest of this page concerns the Fortran components of Clawpack.
There is no complete list of changes since it has evolved to be very
different from the 4.x version of Clawpack in organization, but some of
@@ -191,11 +188,11 @@
frequent times than the standard AMR output has been introduced in v5.9.0,
and these are now called fgout grids to complement the fgmax grids.
These fgout grids are described further below.
-
-
The GeoClaw Fortran code reads in one or more files that specify fgout grids
grid(s) for writing out the solution on a fixed grid throughout
the computation.
@@ -160,9 +157,9 @@
Navigation
to each fgout grid using the fgno attribute, described below.
If you do not assign numbers, they will be numbered sequentially (1,2, etc.)
based on the order they are specified in the setrun.py file.
-
-
-
Since the files have the same format as the usual fort.t, fort.q, and
fort.b files for Clawpack output, it is possible to use a setplot.py
file to set up plotting this sequence of fgout frames in exactly the same
@@ -236,9 +233,9 @@
Alternatively, since every output frame consists of only a single uniform
grid of data, it is much easier to manipulate or plot this data directly than
for general AMR data. The fgout_tools.py module described at
@@ -278,9 +275,9 @@
Note above that fgout points are specified by setting e.g. fgout.x1,
fgout.x2 and fgout.nx in setrun.py. For consistency with the way the
finite volume computational grid is specified in setrun.py
@@ -292,9 +289,9 @@
The fgout grid need not be aligned with any computational grid, and in general
it may overlap several grids at different AMR resolutions. At each fgout time
requested, the solution is interpolated from the finest available AMR grid
@@ -321,11 +318,10 @@
Class to hold a single frame of fgout data at one output time.
Several attributes are defined as properties that can be evaluated
and stored only when needed by the user.
Read input info for fgout grid number fgno from the data file
fgout_grids.data, which should have been created by setrun.py.
This file now contains info about all fgout grids.
Read a netCDF file and return a list of FGoutFrame instances.
This will only be possible if the netCDF file contains at least
the qoi’s ‘h’,’hu’,’hv’,’eta’ required to reconstruct the q array
as output by GeoClaw.
Write a list of fgout frames (at different times on the same rectangular
grid) to a single netCDF file, with some metadata and the topography,
if desired.
@@ -375,11 +372,10 @@
Navigation
time the file was created and the path to the directory where it was made.
In addition to simple rectangles, more general ruled rectangles can also be
used as flagregions. These are a restricted set of polygons for which it is
easy to test if a point is inside or outside, as described in more detail in
@@ -171,11 +167,10 @@
The array force_dry_init can now be saved in the same format as topo
files, using topo_type=3 and specifying Z_format=’%1i’ so that
the data values from the array, which are all either 0 or 1, are printed
@@ -349,9 +346,9 @@
To use a force_dry_init.data file of the sort created above, when
setting up a GeoClaw run the setrun.py file must be modified to
indicate the name of this file along with a time tend. The array is
@@ -367,8 +364,8 @@
Users should set the environment variable FC to point to the correct
compiler, e.g. in bash via:
$ export FC=gfortran
@@ -105,9 +102,9 @@
Navigation
f77 and if so resets it to gfortran since much of Clawpack is not f77
compliant. However, it is best to set the FC environment variable
yourself, e.g. in your .bashrc file.
-
-
-
The LFLAGS environment variable is used to provide flags that are needed when
linking the final binary. The most likely use for this flag would be to link a
particular library with the binary (such as a NetCDF library) or provide a path
to a compiled module. If this variable is not set in the environment then
LFLAGS defaults to the relevant flags in FFLAGS.
-
-
-
Pre-Processor and the PPFLAGS environment variable¶
+
+
+
Pre-Processor and the PPFLAGS environment variable¶
Compilers often provide a pre-processor that can scan source code before
compilation providing some ability to define variables at compile time or
transform the code. Currently the pre-processor is always called before
Clawpack compilation to support optional dependencies, such as NetCDF support,
and some testing abilities. The PPFLAGS environment variable is meant to
provide further control of the pre-processor.
With AMRClaw in two space dimensions and GeoClaw
it is possible to specify gauge locations as points (x,y) where the values of all
components of q should be output every time step during the computation over some
@@ -116,8 +112,8 @@
Navigation
can be spatially interpolated from the fgout grid(s) captured in the run).
The temporal resolution will be that specified for the fgout grids.
See Fixed grid output for more details.
-
-
Setting up, running, and plotting a GeoClaw application follows the same pattern
as other AMRClaw applications, which in turn use many of the same
conventions as the classic single grid Clawpack code, in particular:
To simulate flow over topography it is of course necessary to specify
the topography. This is usually done by providing one or more files of
surface elevation (relative to some reference, e.g. sea level) at a set of
@@ -119,25 +116,24 @@
Navigation
Several file formats are recognized by GeoClaw. See Topography data for more
information on how to specify topography and some on-line resources for
obtaining topography.
Fetch water levels and tide predictions at given NOAA tide station.
The data is returned in 6 minute intervals between the specified begin and
end dates/times. A complete specification of the NOAA CO-OPS API for Data
@@ -183,14 +180,14 @@
Navigation
$CLAW/geoclaw/scratch
-
Required Arguments:
+
Required Arguments
station (string): 7 character station ID
begin_date (datetime): start of date/time range of retrieval
end_date (datetime): end of date/time range of retrieval
-
Optional Arguments:
+
Optional Arguments
time_zone (string): see NOAA API documentation for possible values
datum (string): see NOAA API documentation for possible values
@@ -199,7 +196,7 @@
Navigation
verbose (bool): whether to output informational messages
-
Returns:
+
Returns
date_time (numpy.ndarray): times corresponding to retrieved data
water_level (numpy.ndarray): preliminary or verified water levels
x0,y0 is assumed to be a point (or an array with the same shapes as x1,y1)
x1,y1 is a point or two arrays of points (of the same dimension)
returns array with same shape as x1 and y1 containing distance of each point
@@ -220,20 +217,19 @@
Navigation
two points, but this is not suggested since the notation is not consistent.
Invert the Haversine function to find dx given a distance and point.
Invert the haversine function to find dx given distance d and (x1,y1) and y2.
The corresponding x2 can be x1+dx or x1-dx.
May return NaN if no solution.
As with all of Clawpack, the GeoClaw code is provided as a research
and teaching tool with no guarantee of suitability for any particular
purpose, and no liability on the part of the authors. See the
@@ -95,8 +92,8 @@
Navigation
geophysical hazards, but it is the responsibility of the user to fully
understand the model and its limitations and validate it for the intended
purpose.
-
-
GeoClaw is currently in use for tsunami hazard assessment
by several research groups. Version 4.6.1 of the code was approved in 2012
by the US National Tsunami Hazard
@@ -170,11 +167,10 @@
Navigation
seafloor leads to the propagation of acoustic waves that result in a
surface displacement
-
-
+
With the default style, matplotlib axis labels are often hard to read when
plotting in latitude and longitude. It may help to disable offset labeling
and to rotate the longitude labels:
The Google Earth browser is a powerful visualization tool for
viewing georeferenced data and images. The VisClaw visualization suite includes
tools that will produce georeferenced plots and associated KMZ files needed for
@@ -100,8 +97,8 @@
Navigation
critical decisions on conclusions drawn from the Google Earth
visualization alone. Nevertheless, these realistic visualizations are
useful for setting up simulations and communicating your results.
-
-
To get started, you will need to install the Python packages lxml and
pykml. These libraries can be easily installed through Python
package managers PIP and conda:
To create a pyramid of images for faster loading in Google Earth, you
will also want to install the Geospatial Data Abstraction Library
(GDAL). The GDAL library (and associated Python bindings)
@@ -153,16 +150,16 @@
This test will create an image pyramid in the directory frame0005fig1 and an associated
doc.kml file which you can open in Google Earth.
NOTE (8/1/16) The latest release of the Anaconda GDAL library is not fully version compatible with
its dependencies. If you receive an error that ends with:
The Chile 2010 tsunami is included as an example in the GeoClaw module
of Clawpack. Once you have run this simulation, you can create the
KMZ file needed for visualizing your data in Google Earth by using the
@@ -188,14 +185,12 @@
The plotting parameters needed to instruct VisClaw to create plots
suitable for visualization in Google Earth are all set as attributes
of instances of the VisClaw classes ClawPlotData and ClawPlotFigure.
@@ -203,9 +198,9 @@
The following plotdata attributes apply to all VisClaw figures to be
visualized in Google Earth. These attributes are all optional and
have reasonable default values.
Name used in the Google Earth sidebar to identify the simulation. Default : “GeoClaw”
-
-
-kml_starttime:[Y,M,D,H,M,S]
+
+
+kml_starttime : [Y,M,D,H,M,S]
Start date and time, in UTC, of the event. The format is [year,month,day,hour, minute, second].
By default, local time will be used.
-
-
-kml_timezone:integer
+
+
+kml_timezone : integer
Time zone offset, in hours, of the event from UTC. For example, the offset for Chile is +3 hours,
whereas the offset for Japan is -9 hours. Default : no time zone offset.
-
-
-kml_index_fname:string
+
+
+kml_index_fname : string
The name given to the KMZ file created in the plot directory. Default : “_GoogleEarth”
-
-
-kml_publish:string
+
+
+kml_publish : string
A URL address and path to a remote site hosting a
KMZ file you wish to make available on-line. Default : None
A function that maps computational coordinates (in meters, for
example) to latitude/longitude coordinates. This will be called to
position PNG overlays, gauges, patch boundaries, and regions boundaries to the
@@ -273,18 +268,18 @@
The following attributes apply to an individual figure created for visualization in Google Earth.
The first three attributes are required. The remaining attributes
are optional.
Indicates to VisClaw that the PNG files created for this figure should be suitable for
visualization in Google Earth. With this set to True, all titles, axes labels, colorbars
and tick marks will be suppressed. Default : False.
-
-
-kml_xlimits:[longitude_min,longitude_max]
+
+
+kml_xlimits : [longitude_min, longitude_max]
Longitude range used to place PNG images on Google Earth. This setting will override
any limits set as plotaxes attributes. Required
-
-
-kml_ylimits:[latitude_min,latitude_max]
+
+
+kml_ylimits : [latitude_min, latitude_max]
Latitude range used to place the PNG images on Google Earth.
This setting will override any limits set as plotaxes attributes. Required
-
-
-kml_use_for_initial_view:boolean
+
+
+kml_use_for_initial_view : boolean
Set to True if this figure should be used to determine the initial
camera position in Google Earth. The initial camera position will
be centered over this figure at an elevation equal to approximately
@@ -360,17 +355,17 @@
plotfigure attributes
-
-
-kml_figsize:[size_x_inches,size_y_inches]
+
+
+kml_figsize : [size_x_inches,size_y_inches]
The figure size, in inches, for the PNG file. See Removing
aliasing artifacts for tips on how to set the figure size and dpi
for best results. Default : 8 x 6 (chosen by Matplotlib).
-
-
-kml_dpi:integer
+
+
+kml_dpi : integer
Number of pixels per inch used in rendering PNG figures. For best
results, figure size and dpi should be set to respect the numerical
resolution of the the simulation. See Removing aliasing
@@ -378,18 +373,18 @@
Set to True if you want to create a pyramid of images at different
resolutions for faster loading in Google Earth. Image tiling
requires the GDAL library. See Optional GDAL library, above,
for installation instructions. Default : False.
All figures created for Google Earth are rendered as PNG files using
the Matplotlib backend. So in this sense, the resulting PNG files are
created in a manner that is no different from other VisClaw output
@@ -417,8 +412,8 @@
The plotaxes attributes
colorbar, xlimits, ylimits and title will all be ignored by the KML plotting.
For best results, the attribute scaled should be set to its default value False. The
@@ -426,9 +421,9 @@
The most useful plotitem type will probably be the 2d_pcolor type, although other
types including the filled contour contourf can also be used to good effect.
Colormaps that are designed to work well with Google Earth are
A colorbar can be associated with each figure in the Google Earth
browser by setting the figure attribute kml_colorbar to point to a function
that creates the colorbar:
There are no particular attributes for gauge plots and so they
can be created in the usual way. In the Google Earth browser, gauge locations
will be displayed as Placemarks. Clicking on gauge Placemarks will bring
up the individual gauge plots. The screenshot below shows the gauge plot
that appears when either the gauge Placemark or the gauge label in the sidebar is
clicked.
If the computational coordinates are not in latitude/longitude coordinates, then you must
set the plotdata attribute map_topo_to_latlong to specify a mapping between the computational
coordinates and latitude longitude box which Google Earth will use to visualize your data.
See plotdata attributes.
VisClaw has additional plotdata attributes indicating which figures and frames
to plot and which output style to create. When plotting for Google
Earth, one additional output parameter is necessary.
KML files are very similar to HTML files in that they contan
<tags>…</tags> describing data to be rendered by a suitable
rendering engine. Whereas as standard web browsers can render content
@@ -543,9 +536,9 @@
If you create several frames with relatively high dpi, you may find
that the resulting KMZ file is slow to load in Google Earth. In
extreme cases, large PNG files will not load at all. You can improve
@@ -559,9 +552,9 @@
You may find that the transparent colormap leads to unappealing visual
artifacts. This can happen when the resolution of the PNG file does
not match the resolution of the data used to create the image. In the
@@ -573,12 +566,10 @@
Aliasing effects resulting from default kml_dpi and kml_figsize settings¶
-
-
+
Aliasing effects resulting from default kml_dpi and kml_figsize settings¶
+
We can remove these aliasing effects by making the resolution of the
PNG file a multiple of 30*2*6 = 360. This can be done by setting the
figure size and dpi appropriately:
Aliasing effects removed by properly setting kml_dpi and kml_figsize¶
-
-
+
Aliasing effects removed by properly setting kml_dpi and kml_figsize¶
+
This baseline dpi=12 is the minimum resolution that will remove
striped artifacts from your images. However, you may find that this
resolution is unacceptable, especially for close-up views of
@@ -621,9 +610,9 @@
Creating multiple figures at different resolutions¶
You can create several figures for visualization in Google Earth.
Each figure you create will show up as a separate named folder in the Google
Earth sidebar. The name will match that given to the VisClaw plotfigure.
@@ -662,9 +651,9 @@
Creating multiple figures at different resolutions
Mapping topography data to latitude/longitude coordinates¶
+
+
+
Mapping topography data to latitude/longitude coordinates¶
In many situations, your computational domain may not be conveniently
described in latitude/longitude coordinates. When simulating overland
flooding events, for example, topographic data may more easily be
@@ -728,9 +717,9 @@
Creating multiple figures at different resolutions
You can easily share your results with collaborators
by providing links to your archive KMZ file in HTML webpages. Collaborators can
download the KMZ file and open it in a Google Earth browser.
For netCDF files the
data points are generally interpreted as pointwise values at the points
specified in the lat and lon arrays included in the file (or as
cell-averaged values with these points as the cell centers).
Older versions of the documentation used to be tagged for each minor
release, e.g. v5.6.1. Starting with v5.7.0, these are now only tagged
with the major release, e.g. v5.7.x.
When updating a minor version, e.g. from v5.7.0 to v5.7.1, we will continue
to use the same branch v5.7.x. You should just make sure the v5.7.x and dev
are up to date with each other at the time of release.
Any files placed in $CLAW/doc/doc/extra_files will be copied verbatim
(recursively for subdirectories) to the directory
$CLAW/doc/doc/_build/html when Sphinx is used to build the documentation.
@@ -343,9 +340,9 @@
Some webpages are created within other clawpack repositories.
For example, the page http://www.clawpack.org/geoclawdev-2020/
is modified by pushing changes to the master branch of the repository
@@ -357,11 +354,10 @@
Make sure all your subrepositories are up to date and clean:
cd $CLAW
git co master
@@ -105,16 +102,16 @@
PreparationCLAW/doc/gallery/README.md
as required for building the galleries, and check all the resulting plots to
make sure everything looks correct.
-
-
-
Make sure all repositories are checked out to the master branch and are up to
date with the main Clawpack repositories and clean, as described in the
preparation step above.
We generally do this step for a release candidate first, and then
do the same for the final release. For release candidates, modify the
version number to include 5.4.1rc-alpha, for example.
After the release has been finalized, tag all repositories. In the commands
below, it is assumed the remote upstream points to the main Github repos
(not your fork) and that you have push permission. Change 5.x.x to the
@@ -187,9 +184,9 @@
Final releaseChanges to master since v5.9.0 which generate diffs between the latest version and
the current master branch of each repository.
-
-
-
To release on the pypi server (for installation via pip) we have to do a bit
of trickery. We can’t just follow the directions at https://packaging.python.org/tutorials/packaging-projects/
because we have a very non-Pythonic directory structure; in particular,
@@ -239,9 +236,9 @@
See Guide for updating this documentation for general instructions on building the documentation
and galleries using Sphinx, and for how to push changes to Github so they
show up on the web.
Ideally all the apps in the Clawpack Applications repository should be rerun with the new release
and any issues fixed. If old apps are modified, add a note to the
README.rst file in the directory that indicates when it was last updated
@@ -305,14 +302,14 @@
Note that unlike the tar file for a new release, the docker image includes
a clone of the apps repository, so it would be best to first update that
@@ -327,11 +324,10 @@
This page describes ways to download and “install” Clawpack in a way that
the Fortran versions of the PDE solvers (Classic, AMRClaw, and GeoClaw)
can be used, along with Python codes that support these solvers, including
@@ -93,8 +90,8 @@
You can download the most recent (or certain previous versions) of Clawpack
as a tar file. After untarring this, you can set environment variables
to point to this version.
You can clone the git repositories from https://github.com/clawpack.
This is particularly useful if you
want the latest development version or a branch that is not in a release yet,
@@ -126,7 +123,7 @@
gitclonehttps://github.com/clawpack/clawpack.gitcdclawpack
-gitcheckoutv5.9.0# or an older version; `git tag -l` to list options
+gitcheckoutv5.9.0# or an older version; `git tag -l` to list optionsgitsubmoduleinit# for repositories pyclaw, clawutil, visclaw, etc.gitsubmoduleupdate# clones all the submodule repositoriesexportCLAW=/full/path/to/clawpack# in bash
@@ -141,19 +138,18 @@
Navigation
See Clawpack components for a list of what is generally included
under the top level clawpack directory when using any of these approaches.
(And what is not included, e.g. the Clawpack Applications repository.)
-
-
Please see Which Clawpack solver should I use? before installing, particularly
if you are not sure whether you will
be using the Fortran versions of the PDE solvers
@@ -104,7 +101,7 @@
Navigation
or, more specifically,
-
pipinstall--userclawpack==v5.9.0
+
pipinstall--userclawpack==v5.9.0
or you can choose a previous version from the PyPi history.
@@ -113,9 +110,9 @@
Navigation
If you think you might want to use the Fortran packages as well
(Classic, AMRClaw, GeoClaw) and/or want easier access to the Python source
code, it is recommended that you follow the pip instructions below.
The recommended way to install the latest release of Clawpack, for
using both PyClaw and the Fortran packages, is to use pip, e.g. with the
following command
@@ -170,8 +167,8 @@
Using pip to download and install actually clones Git repositories from
https://github.com/clawpack/clawpack. If you are comfortable with
Git you can use the same top repository to update Clawpack or switch
@@ -208,9 +205,9 @@
Next steps:Python path if you are having problems with the wrong version
being imported.
-
-
-
Experimenting with code in the examples directories¶
+
+
+
Experimenting with code in the examples directories¶
We suggest that if you want to experiment extensively with examples or
modify an example to solve your own problem, you first copy a directory out
of the source code tree to a different location, in order to minimize
@@ -218,9 +215,9 @@
Convert float to string in fixed point notation with at most
num_digits digits of precision and trailing zeros removed,
for printing nicely in kml description boxes.
Produce kml files for the computational domain, all gauges and regions
specified, and all topo and dtopo files specified in rundata.
This can be used, e.g. by adding the lines
@@ -331,9 +328,9 @@
Navigation
from the setrun.py file in the current directory.
-
-
-clawpack.geoclaw.kmltools.pcolorcells_for_kml(X, Y, Z, png_filename=None, dpc=2, max_inches=15.0, verbose=True, **kwargs)¶
+
+
+clawpack.geoclaw.kmltools.pcolorcells_for_kml(X, Y, Z, png_filename=None, dpc=2, max_inches=15.0, verbose=True, **kwargs)¶
Wraps pcolormesh in a way that a png file is created that can be viewed
on Google Earth with proper alignment and with sharp grid cell edges.
Works if X,Y are cell centers or edges, and X,Y can be 2d or 1d arrays.
@@ -361,43 +358,43 @@
Navigation
png file, as described below.
Internally the value dpi (dots per inch) for the png file is
determined so that it is at least 16 and so that:
Create kmz file showing onshore and offshore topography as separate layers.
Currently not very flexible but is useful in quickly creating kmz files
to open on Google Earth showing the onshore topography and offshore
@@ -508,7 +505,7 @@
Navigation
cells should be properly rendered to ease exploration of DEMs.
A colorbar png file is also created and included in the kmz file.
-
Input:
+
Input
topo should be a topotools.Topography object,
zlim is the elevation z limits for choosing the color map
@@ -519,23 +516,22 @@
Navigation
close_figs to close the pyplot figures after making png files
One can also use the Fixed grid output capabilities added to
GeoClaw in Version 5.9.0 in order to capture the solution over a specified
fixed grid at frequent output times. If this output is frequent enough,
@@ -154,11 +150,10 @@
An alternative using fgout gridsFixed grid output for more details.
-
-
+
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
-
+
Makefiles for the Fortran code in many repositories
use the common Makefile found in $CLAW/clawutil/src/Makefile.common,
so you must have the clawutil repository.
A number of variables are defined in the Makefiles of application
directories. For example, output is directed to the subdirectory specified
by the variable OUTDIR. To change this, simply modify the Makefile before
@@ -173,9 +170,9 @@
Compiler flags can be changed by modifying the FFLAGS variable in the
Makefile. If you change compiler flags you will generally need to recompile
all the Fortran files and the Makefile dependencies will not detect this.
@@ -185,9 +182,9 @@
Compiler flagsFortran Compilers for more about compiler flags.
-
-
-
Fortran source files with the same base name but different suffixes can cause
unexpected source files to be compiled. This occurs as the Makefiles are
structured to use the free-format Fortran source files .f90* before the
@@ -200,12 +197,11 @@
See Clawpack Makefiles for general information on using the Makefile found
in application directories for the Fortran codes.
New in 5.4.0. The Makefile also typically refers to a common list of
@@ -152,8 +149,8 @@
Navigation
For the classic 1d code there are no modules, only subroutines.
-
-
Replacing files with the same name as library files¶
+
+
Replacing files with the same name as library files¶
Note that the list of default library routines above contains both
qinit.f90 and setprob.f90. The default versions of these files contain
subroutines that have the correct calling sequence but return without doing
@@ -172,9 +169,9 @@
Navigation
(See Fortran 77 vs. Fortran 90 files for an important cautionary note on what
might go wrong if you have both a .f file and a .f90 file with
the same base name in the same directory.)
-
-
-
Using a modified library routine with a different name¶
+
+
+
Using a modified library routine with a different name¶
Suppose we want to use a local version of bc1.f in order to
implement some new boundary condition not included in the default version.
If we call the local file bc1.f then we can simply add it to the list
@@ -204,11 +201,10 @@
Using a modified library routine with a different name
(It doesn’t matter what order the files are listed in each section.)
-
-
+
+
-
@@ -219,17 +215,15 @@
Using a modified library routine with a different name
When using GeoClaw to model inundation, it is important to include an
appropriate bottom friction term in the equations. This takes the form of a
source term added to the right hand side of
@@ -134,10 +130,9 @@
Navigation
everywhere, the effect of this bug on the resulting accuracy was probably
small, but users may want to test this.
The array pt_chosen can be used to create a file that is read in by
GeoClaw to identify points where Ztopo is below MHW but where there
is dry land because of protection by dikes or levies. This is done by
@@ -203,23 +199,23 @@
The output array could also be used to define an AMR flag region as a
ruled rectangle, using the function
region_tools.ruledrectangle_covering_selected_points described in
Ruled Rectangles. This would give a
minimal ruled rectangle covering all the chosen points. An example is
given below.
This argument is useful if you want to apply a sequence of different
criteria to choose points. For example, suppose you first want to choose
all grid points within 5 points of the coast (as can be done using
@@ -240,9 +236,9 @@
First we choose all points with elevation below 15 m and that are
connected to the coast over this topography extent. Note the latter
requirement will eliminate the low lying area at the bottom of the
@@ -326,9 +322,9 @@
Write out the masked array indicating fgmax points¶
+
+
+
+
Write out the masked array indicating fgmax points¶
One we have selected the desired fgmax points, these can be output in
the style of a topo_type=3 topography file, with a header followed
point values at all points on a uniform grid. The values are simply the
@@ -434,9 +430,9 @@
Note the blue region in the above plot that is inland from the coast and
behind a green barrier. Examining Google Earth shows that this is
wetland area that probably should not be initialized as a lake. We can
@@ -505,11 +501,10 @@
Before version 4.3, Clawpack used Matlab (Mathworks, Inc.) for plotting and visualizing
results of simulations. For this purpose, an extensive set of plotting
tools were developed. These are still available in
@@ -105,8 +102,8 @@
Navigation
visualization assumes finite volume data, and individual plot
“patches” use cell-averaged values to color mesh cells directly. No
graphical interpolation is done when mapping to the colormap.
-
-
To use the Matlab plotting tools with Clawpack, the user will need to
first be sure that the necessary Matlab routines are on the Matlab
search path. This can be done by explicitly setting the MATLABPATH
@@ -119,9 +116,9 @@
Once output files, e.g. ‘fort.q0000’, ‘fort.q0001’, and so on, and
corresponding fort.t0000’ or ‘fort.t0001’ files have been created, the
user can plot them in Matlab by entering one of
@@ -192,9 +189,9 @@
Navigation
is returned to the plot prompt, a file afterframe is called. This
file contains user commands to set various plot properties. See below
for more details on what the user might wish to include in this file.
-
-
-
The properties of the Matlab plot are controlled through two main
user-defined files located, typically, in the current working
directory. The first of these files, the ‘setplot’ file (e.g. setplot1.m, setplot2.m or
@@ -240,9 +237,9 @@
Navigation
Each of the examples in Clawpack include a ‘setplot’ file which you
can browse to get an idea as to what can be put in the file.
The ‘afterframe.m’ script is the second file which control aspects of the
plot and is called after the plot has been created. The following are
commonly set in the afterframe file:
@@ -296,21 +293,21 @@
Navigation
The user is encouraged to browse the ‘afterframe.m’ file available
with each Clawpack example to get a better idea as to what one might
include in this file.
The graphics are controlled to a large extent using variables that are
set in the Matlab base workspace. This can lead to unpredictable results
when switching between Clawpack examples.
The interested user is encouraged to browse the Matlab Gallery for
examples of the types of plots that can be created with the Clawpack Matlab
plotting routines.
Several Fortran routines in GeoClaw interpolate from the computational grids
to other specified points where output is desired
(typically using the finest AMR resolution available nearby at each desired
@@ -138,10 +134,9 @@
Navigation
Since the point is apparently wet (h > 0), one might conclude that the sea
surface at this point is 27 meters above the initial sea level, whereas in
fact the sea level is never more than 6 meters above sea level.
-
+
Earthquake sources: Fault slip and the Okada model¶
+
+
Earthquake sources: Fault slip and the Okada model¶
To initiate a tsunami from an earthquake, it is necessary to generate a model of
how the seafloor moves, which is generally specified in a dtopo file as
described in Topography displacement files. This is often done by starting with a
@@ -108,8 +105,8 @@
Navigation
the results together (valid by linear superposition of the solutions to the
linear elastic halfspace problems).
See dtopotools module for moving topography for more documentation and illustrations.
-
-
For historic earthquakes, it is generally possible to find many different
models for the distribution of slip on one or more fault planes,
see for example the pointers at Earthquake source models.
@@ -211,9 +208,9 @@
Navigation
See dtopotools module for moving topography for more details, and the Jupyter notebook
$CLAW/apps/notebooks/geoclaw/dtopotools_examples.ipynb for more examples.
The slip on the fault plane(s) must be translated into seafloor deformation.
This is often done using the “Okada model”, which is derived from
a Green’s function solution to the elastic half space problem, following
@@ -231,9 +228,9 @@
Navigation
modeling of the resulting seafloor deformation may not be justified.
In addition to the parameters above, the Okada model also requires an elastic
parameter, the Poisson ratio, which is usually taken to be 0.25.
It is also possible to set a rupture_time and a rise_time for each
subfault in order to model a time-dependent rupture process. This is called
a “kinematic rupture” since the these values are specified.
The Clawpack Fortran Classic 3d code, AMRClaw 2d and 3d code,
and GeoClaw codes include
OpenMP directives for making use of multicore shared memory machines.
The code in AMRClaw and GeoClaw is parallelized by splitting the list of
patches that must be advanced in time between threads, and then each grid
patch is handled by a single thread. For this reason good performance will
@@ -148,18 +145,17 @@
In Specifying classic run-time parameters in setrun.py, the style of specifying output times can be
specified by setting the parameter output_style. This can be illustrated
by a typical example from a setrun.py file:
In AMRClaw and GeoClaw, the format for the output data (solutions) can be
specified by setting the parameter output_format to ‘ascii’,
‘binary64’, or ‘binary32’. (The single-grid
@@ -137,14 +134,14 @@
Two output files are created at each output time (each frame). The frames
are generally numbered 0, 1, 2, etc. The two files, at frame 2, for
example, are called fort.t0002 and fort.q0002.
New in v5.9.0: Previously the user could specify output_format=’binary’
for binary format. Starting in v5.9.0, the user can specify either
output_format=’binary64’ or output_format=’binary32’. For backward
@@ -217,14 +214,14 @@
The contents of aux arrays can also be output along with each time frame.
Which components are output is controlled by the setrun variable
clawdata.output_aux_components, which can be ‘none’ (default) or ‘all’
@@ -233,11 +230,10 @@
Clawpack includes a number of related hyperbolic PDE solvers:
@@ -99,8 +96,8 @@
Navigation
ideas, make use of the same set of Riemann solvers, and can be used with VisClaw
for visualization. If you’re not sure which solver to use, here you will find
the main differences between them.
-
-
The Classic, AMRClaw, and GeoClaw solvers are Fortran-based
packages and rely on Makefiles and environment variables. Problems are
specified partially through Python scripts at run time (setrun.py) and partially
@@ -118,9 +115,9 @@
Installation and user interfacef2py, a step
that sometimes causes problems and is not required if you are only using
Fortran versions of the solvers.
-
-
-
All of the Clawpack solvers include the classic algorithms described in
[LeVeque-FVMHP]; if you only require those, it’s easiest to use Classic or
PyClaw. Most of the packages contain additional algorithms:
Running a Fortran version of Clawpack produces output files that can then be
read in to a graphics package as a post-processing step. If you understand
the format of the output files, you can write custom graphics routines using
@@ -184,15 +181,14 @@
Navigation
plots. We are developing tools to simplify this process.
This is normally determined by the outdir attribute of
the ClawPlotData object directing the plotting. But see the next FAQ
for the option of using different directories for some plot items (e.g. to
@@ -221,9 +217,9 @@
If you want one plot item on an axis to use the default plotdata.outdir
while another to take data from a different directory (in order to compare
two computations, for example), you can set the outdir
@@ -250,9 +246,9 @@
By default, a figure is created of the default matplotlib size, with a tan
background. Any desired
keyword arguments to the matplotlib figure command can
@@ -287,12 +283,12 @@
See Plotting with Visclaw for general information about plotting Clawpack results.
GeoClaw results can be viewed using these same tools. Some addition
functions and useful colormaps are available in the visclaw module
@@ -100,10 +97,9 @@
Clawpack includes utilities for plotting using Python. Most of these
are defined in the clawpack.visclaw module.
In order to use these you will need to insure that you have the required
modules installed (see python-install).
See Python Hints for more information on Python and pointers to many tutorials.
Typically each applications directory contains a file setplot.py, see for
example Plotting examples.
This file should define a function setplot(plotdata) that sets various
@@ -117,9 +114,9 @@
If you are viewing plots in using Iplotclaw and want to explore the data for
some frame or make plots directly in your Python shell, the data that is
being plotted is available to you in attributes of the Iplotclaw instance.
@@ -226,10 +223,10 @@
The first step in specifying how to plot is to create a ClawPlotData
object to hold all the data required for plotting. This is generally done
by creating a file setplot.py (see below).
If you are installing via pip using the command in
Quick Installation of all packages with pip, or via git clone, then you need Git.
You may already have it; in particular the Xcode tools on
Mac OSX contains Git. If you need to install it, see the Git book.
Many people have contributed to PyClaw, some of them very substantial parts of
the package. Their work is greatly appreciated: no open source project can
survive without a community. The following people contributed major parts of
@@ -116,9 +113,9 @@
