source: anuga_core/source/anuga/config.py @ 7073

"""Module where global ANUGA model parameters and default values are set
"""

import os
import sys

################################################################################
# Numerical constants
################################################################################

epsilon = 1.0e-12                    # Smallest number - used for safe division
max_float = 1.0e36                   # Largest number - used to initialise 
                                     # (max, min) ranges
default_smoothing_parameter = 0.001  # Default alpha for penalised
                                     # least squares fitting
single_precision = 1.0e-6            # Smallest single precision number
velocity_protection = 1.0e-6         # Used to compute velocity from momentum
                                     # See section 7.4 on Flux limiting 
                                     # in the user manual
                            

################################################################################
# Standard filenames, directories and system parameters used by ANUGA
################################################################################

pmesh_filename = '.\\pmesh'
version_filename = 'stored_version_info.py'
default_datadir = '.'
time_format = '%d/%m/%y %H:%M:%S'    # Used with timefile2netcdf
umask = 002  # Controls file and directory permission created by anuga (UNIX)
default_boundary_tag = 'exterior' 

# Major revision number for use with create_distribution
# and update_anuga_user_guide
major_revision = '1.0beta'

################################################################################
# Physical constants
################################################################################

manning = 0.03  # Manning's friction coefficient
#g = 9.80665    # Gravity - FIXME reinstate this and fix unit tests.
g = 9.8
#g(phi) = 9780313 * (1 + 0.0053024 sin(phi)**2 - 0.000 0059 sin(2*phi)**2) micro m/s**2, where phi is the latitude
#The 'official' average is 9.80665

eta_w = 3.0e-3 # Wind stress coefficient
rho_a = 1.2e-3 # Atmospheric density
rho_w = 1023   # Fluid density [kg/m^3] (rho_w = 1023 for salt water)

################################################################################
# Limiters - used with linear reconstruction of vertex 
# values from centroid values
################################################################################

# Betas [0;1] control the allowed steepness of gradient for second order
# extrapolations. Values of 1 allow the steepes gradients while
# lower values are more conservative. Values of 0 correspond to
# 1'st order extrapolations.
#

# There are separate betas for the w, uh, and vh limiters
# I think these are better SR but they conflict with the unit tests!
beta_w      = 1.0
beta_w_dry  = 0.2
beta_uh     = 1.0
beta_uh_dry = 0.2
beta_vh     = 1.0
beta_vh_dry = 0.2

# Alpha_balance controls how limiters are balanced between deep and shallow.
# A large value will favour the deep water limiters, allowing the a closer hug
# to the coastline.  This will minimise 'creep' but at the same time cause
# smaller time steps
# Range:
alpha_balance = 2.0 

# Flag use of new limiters.
# tight_slope_limiters = 0 means use old limiters (e.g. for some tests)
# tight_slope_limiters = 1 means use new limiters that hug the bathymetry closer
tight_slope_limiters = True

# Use centroid velocities to reconstruct momentum at vertices in
# very shallow water
# This option has a first order flavour to it, but we still have second order
# reconstruction of stage and this option only applies in
# balance_deep_and_shallow when
# alpha < 1 so in deeper water the full second order scheme is used.
#
# This option is good with tight_slope_limiters, especially for large domains.
use_centroid_velocities = True
        
# FIXME (Ole) Maybe get rid of order altogether and use beta_w
default_order = 2

################################################################################
# Timestepping
################################################################################

CFL = 1.0  # CFL condition assigned to domain.CFL - controls timestep size
      
# Choose type of timestepping,
#timestepping_method = 'rk2'   # 2nd Order TVD scheme
timestepping_method = 'euler' # 1st order euler

# rk2 is a little more stable than euler, so rk2 timestepping
# can deal with a larger beta when slope limiting the reconstructed
# solution. The large beta is needed if solving problems sensitive
# to numerical diffusion, like a small forced wave in an ocean
beta_euler = 1.0
beta_rk2   = 1.6

# Option to search for signatures where isolated triangles are
# responsible for a small global timestep.
# Treating these by limiting their momenta may help speed up the
# overall computation.
# This facility is experimental.
# protect_against_isolated_degenerate_timesteps = False
protect_against_isolated_degenerate_timesteps = False

min_timestep = 1.0e-6 # Minimal timestep accepted in ANUGA
max_timestep = 1.0e+3
max_smallsteps = 50   # Max number of degenerate steps allowed b4
                      # trying first order

# Perhaps minimal timestep could be based on the geometry as follows:
# Define maximal possible speed in open water v_max, e.g. 500m/s (soundspeed?)
# Then work out minimal internal distance in mesh r_min and set
# min_timestep = r_min/v_max
#
# Max speeds are calculated in the flux function as
#
# lambda = v +/- sqrt(gh)
#
# so with 500 m/s, h ~ 500^2/g = 2500 m well out of the domain of the
# shallow water wave equation
#
# The actual soundspeed can be as high as 1530m/s
# (see http://staff.washington.edu/aganse/public.projects/clustering/clustering.html),
# but that would only happen with h>225000m in this equation. Why ?
# The maximal speed we specify is really related to the max speed
# of surface pertubation
#
# v_max = 100 #For use in domain_ext.c
# sound_speed = 500

################################################################################
# Ranges specific to the shallow water wave equation
# These control maximal and minimal values of quantities
################################################################################

# Water depth below which it is considered to be 0 in the model
minimum_allowed_height = 1.0e-3 

# Water depth below which it is *stored* as 0
minimum_storable_height = 1.0e-5

# FIXME (Ole): Redefine this parameter to control maximal speeds in general
# and associate it with protect_against_isolated_degenerate_timesteps = True
maximum_allowed_speed = 0.0 # Maximal particle speed of water
#maximum_allowed_speed = 1.0 # Maximal particle speed of water
                             # Too large (100) creates 'flopping' water
                             # Too small (0) creates 'creep'
                            
maximum_froude_number = 100.0 # To be used in limiters.

################################################################################
# Performance parameters used to invoke various optimisations
################################################################################

use_extensions = True # Use C-extensions
use_psyco = True      # Use psyco optimisations

optimise_dry_cells = True # Exclude dry and still cells from flux computation
optimised_gradient_limiter = True # Use hardwired gradient limiter
use_edge_limiter = False  # The edge limiter is better, but most runs have been
                          # using vertex limiting. Validations passed with this
                          # one True 9th May 2008, but many unit tests need
                          # backward compatibility flag set FIXME(Ole).

points_file_block_line_size = 500 # Number of lines read in from a points file
                                  # when blocking

################################################################################
# Dynamically-defined constants.
################################################################################

# Determine if we can read/write large NetCDF files
netcdf_mode_w = 'w'
netcdf_mode_a = 'a'
netcdf_mode_r = 'r'

# Code to set the write mode depending on
# whether Scientific.IO supports large NetCDF files
s = """from Scientific.IO.NetCDF import NetCDFFile; fid = NetCDFFile('tmpfilenamexx', 'wl')"""

# Need to run in a separate process due an
# error with older versions of Scientific.IO
if sys.platform == 'win32':
    null = 'NUL'
else:
    null = '/dev/null'
cmd = 'python -c "%s" 2> %s' % (s, null)
err = os.system(cmd)

if err != 0:
    # The Python script s failed e.g. with a segfault
    # which means that large file support is
    # definitely not supported
    pass
else:    
    # Try the import within this process
    try:
        exec(s)
    except IOError:
        # NetCDFFile does not segfault but it does not
        # support large file support    
        pass
    else:
        # Set the default mode to large file support
        netcdf_mode_w = 'wl'