def precompute(self):
stations = list(self.stations.iterrows())
channel = self.config['wavefield_channel']
if channel == "all":
channels = ['E', 'N', 'Z']
else:
channels = [channel]
for i, station in stations[self.rank:len(self.stations):self.size]:
for cha in channels:
self.function(station, channel=cha)
if self.rank == 0:
wfile = glob(
os.path.join(self.args.project_path, 'greens',
'*' + station['sta'] + '.*.h5'))[0]
ofile = os.path.join(self.args.project_path,
'wavefield_example.png')
with WaveField(wfile) as wf:
wf.plot_snapshot(self.npts / self.Fs * 0.3,
outfile=ofile,
stations=[(station['lat'], station['lon'])])
def get_ns(wf1, source_conf, insta):
# Nr of time steps in traces
if insta:
# get path to instaseis db
#ToDo: ugly.
dbpath = json.load(
open(os.path.join(source_conf['project_path'],
'config.json')))['wavefield_path']
# open
db = instaseis.open_db(dbpath)
# get a test seismogram to determine...
stest = db.get_seismograms(source=instaseis.ForceSource(latitude=0.0,
longitude=0.0),
receiver=instaseis.Receiver(latitude=10.,
longitude=0.0),
dt=1. / source_conf['sampling_rate'])[0]
nt = stest.stats.npts
Fs = stest.stats.sampling_rate
else:
with WaveField(wf1) as wf1:
nt = int(wf1.stats['nt'])
Fs = round(wf1.stats['Fs'], 8)
# Necessary length of zero padding for carrying out
# frequency domain correlations/convolutions
n = next_fast_len(2 * nt - 1)
# Number of time steps for synthetic correlation
n_lag = int(source_conf['max_lag'] * Fs)
if nt - 2 * n_lag <= 0:
click.secho('Resetting maximum lag to %g seconds: Synthetics are too\
short for a maximum lag of %g seconds.' % (nt // 2 / Fs, n_lag / Fs))
n_lag = nt // 2
n_corr = 2 * n_lag + 1
return nt, n, n_corr, Fs
def get_ns(all_conf, insta=False):
# Nr of time steps in traces
if insta:
# get path to instaseis db
dbpath = all_conf.config['wavefield_path']
# open
db = instaseis.open_db(dbpath)
# get a test seismogram to determine...
stest = db.get_seismograms(source=instaseis.ForceSource(latitude=0.0,
longitude=0.),
receiver=instaseis.Receiver(latitude=10.,
longitude=0.),
dt=1. / all_conf.config
['wavefield_sampling_rate'])[0]
nt = stest.stats.npts
Fs = stest.stats.sampling_rate
else:
any_wavefield = glob(os.path.join(all_conf.config['project_path'],
'greens', '*.h5'))[-1]
with WaveField(any_wavefield) as wf1:
nt = int(wf1.stats['nt'])
Fs = round(wf1.stats['Fs'], 8)
n = wf1.stats['npad']
# # Necessary length of zero padding
# # for carrying out frequency domain correlations/convolutions
# n = next_fast_len(2 * nt - 1)
# Number of time steps for synthetic correlation
n_lag = int(all_conf.source_config['max_lag'] * Fs)
if nt - 2 * n_lag <= 0:
n_lag_old = n_lag
n_lag = nt // 2
warn('Resetting maximum lag to %g seconds:\
Synthetics are too short for %g seconds.' % (n_lag / Fs, n_lag_old / Fs))
n_corr = 2 * n_lag + 1
return nt, n, n_corr, Fs
def get_ns(wf1, source_conf):
# Nr of time steps in traces
with WaveField(wf1) as wf1:
nt = int(wf1.stats['nt'])
Fs = round(wf1.stats['Fs'], 8)
print('Sampling rate (Hz) %g' % Fs)
# Necessary length of zero padding for carrying out frequency domain correlations/convolutions
n = next_fast_len(2 * nt - 1)
# Number of time steps for synthetic correlation
n_lag = int(source_conf['max_lag'] * Fs)
if nt - 2 * n_lag <= 0:
click.secho('Resetting maximum lag to %g seconds: Synthetics are too\
short for a maximum lag of %g seconds.' % (nt // 2 / Fs, n_lag / Fs))
n_lag = nt // 2
n_corr = 2 * n_lag + 1
return nt, n, n_corr
def g1g2_kern(wf1str, wf2str, kernel, adjt, src, source_conf, insta):
measr_conf = json.load(
open(os.path.join(source_conf['source_path'], 'measr_config.json')))
bandpass = measr_conf['bandpass']
if bandpass == None:
filtcnt = 1
elif type(bandpass) == list:
if type(bandpass[0]) != list:
filtcnt = 1
else:
filtcnt = len(bandpass)
ntime, n, n_corr, Fs = get_ns(wf1str, source_conf, insta)
# use a one-sided taper: The seismogram probably has a non-zero end,
# being cut off whereever the solver stopped running.
taper = cosine_taper(ntime, p=0.01)
taper[0:ntime // 2] = 1.0
########################################################################
# Prepare filenames and adjoint sources
########################################################################
filenames = []
adjt_srcs = []
adjt_srcs_cnt = 0
for ix_f in range(filtcnt):
filename = kernel + '.{}.npy'.format(ix_f)
filenames.append(filename)
#if os.path.exists(filename):
# continue
f = Stream()
for a in adjt:
adjtfile = a + '*.{}.sac'.format(ix_f)
adjtfile = glob(adjtfile)
try:
f += read(adjtfile[0])[0]
f[-1].data = my_centered(f[-1].data, n_corr)
adjt_srcs_cnt += 1
except IndexError:
print('No adjoint source found: {}\n'.format(a))
break
adjt_srcs.append(f)
########################################################################
# Compute the kernels
########################################################################
with NoiseSource(src) as nsrc:
ntraces = nsrc.src_loc[0].shape[0]
if insta:
# open database
dbpath = json.load(
open(os.path.join(source_conf['project_path'],
'config.json')))['wavefield_path']
# open and determine Fs, nt
db = instaseis.open_db(dbpath)
# get receiver locations
lat1 = geograph_to_geocent(float(wf1[2]))
lon1 = float(wf1[3])
rec1 = instaseis.Receiver(latitude=lat1, longitude=lon1)
lat2 = geograph_to_geocent(float(wf2[2]))
lon2 = float(wf2[3])
rec2 = instaseis.Receiver(latitude=lat2, longitude=lon2)
else:
wf1 = WaveField(wf1str)
wf2 = WaveField(wf2str)
kern = np.zeros((filtcnt, ntraces, len(adjt)))
########################################################################
# Loop over locations
########################################################################
for i in range(ntraces):
# noise source spectrum at this location
# For the kernel, this contains only the basis functions of the
# spectrum without weights; might still be location-dependent,
# for example when constraining sensivity to ocean
S = nsrc.get_spect(i)
if S.sum() == 0.:
# The spectrum has 0 phase so only checking absolute value here
continue
####################################################################
# Get synthetics
####################################################################
if insta:
# get source locations
lat_src = geograph_to_geocent(nsrc.src_loc[1, i])
lon_src = nsrc.src_loc[0, i]
fsrc = instaseis.ForceSource(latitude=lat_src,
longitude=lon_src,
f_r=1.e12)
s1 = np.ascontiguousarray(
db.get_seismograms(
source=fsrc,
receiver=rec1,
dt=1. / source_conf['sampling_rate'])[0].data * taper)
s2 = np.ascontiguousarray(
db.get_seismograms(
source=fsrc,
receiver=rec2,
dt=1. / source_conf['sampling_rate'])[0].data * taper)
else:
s1 = np.ascontiguousarray(wf1.data[i, :] * taper)
s2 = np.ascontiguousarray(wf2.data[i, :] * taper)
spec1 = np.fft.rfft(s1, n)
spec2 = np.fft.rfft(s2, n)
g1g2_tr = np.multiply(np.conjugate(spec1), spec2)
c = np.multiply(g1g2_tr, S)
#######################################################################
# Get Kernel at that location
#######################################################################
corr_temp = my_centered(np.fft.ifftshift(np.fft.irfft(c, n)),
n_corr)
#######################################################################
# Apply the 'adjoint source'
#######################################################################
for ix_f in range(filtcnt):
f = adjt_srcs[ix_f]
if f == None:
continue
for j in range(len(f)):
delta = f[j].stats.delta
kern[ix_f, i, j] = np.dot(corr_temp, f[j].data) * delta
#elif measr_conf['mtype'] in ['envelope']:
# if j == 0:
# corr_temp_h = corr_temp
# print(corr_temp_h)
# if j == 1:
# corr_temp_h = hilbert(corr_temp)
# print(corr_temp_h)
#
# kern[ix_f,i,j] = np.dot(corr_temp,f[j].data) * delta
if i % 50000 == 0:
print("Finished {} source locations.".format(i))
if not insta:
wf1.file.close()
wf2.file.close()
for ix_f in range(filtcnt):
filename = filenames[ix_f]
if kern[ix_f, :, :].sum() != 0:
np.save(filename, kern[ix_f, :, :])
return ()
def run_preprocessing(source_config):
configfile = os.path.join(source_config['project_path'],
'config.json')
config = json.load(open(configfile))
files = glob(os.path.join(config['wavefield_path'],'*.h5'))
processed_path = os.path.join(source_config['source_path'],
'wavefield_processed')
if not os.path.exists(processed_path):
os.mkdir(processed_path)
# very simple embarrassingly parallel loop
comm = MPI.COMM_WORLD
size = comm.Get_size()
rank = comm. Get_rank()
files = files[rank::size]
for file in files:
newfile = os.path.join(processed_path, os.path.basename(file))
if os.path.exists(newfile):
print("File {} was already processed, skipping.".format(os.path.basename(file)))
continue
else:
print("Preprocessing {}".format(os.path.basename(file)))
if source_config['preprocess_truncate_sec'] is not None:
# truncating
with WaveField(file) as wf:
wf.truncate(newfile,float(source_config['preprocess_truncate_sec']))
if source_config['preprocess_decimation_factor'] is not None:
# Already truncated file?
if os.path.exists(newfile):
newfile_temp = newfile + '.temp'
with WaveField(newfile) as wf:
wf.decimate(decimation_factor=source_config['preprocess_decimation_factor'],
outfile=newfile_temp,
taper_width=0.005)
os.system("mv {} {}".format(newfile_temp,newfile))
else:
with WaveField(file) as wf:
wf.decimate(decimation_factor=source_config['preprocess_decimation_factor'],
outfile=newfile,
taper_width=0.005)
if source_config['preprocess_filter_kind'] == 'bandpass':
# The file has been written previously by wavefield.truncate
if os.path.exists(newfile):
with WaveField(newfile,w='a') as wf:
wf.filter_all(
source_config['preprocess_filter_kind'],
overwrite=True,
freqmin=source_config['preprocess_filter_params'][0],
freqmax=source_config['preprocess_filter_params'][1],
corners=source_config['preprocess_filter_params'][2],
zerophase=source_config['preprocess_filter_params'][3])
else:
# The file still has to be written
with WaveField(file) as wf:
wf.filter_all(
source_config['preprocess_filter_kind'],
overwrite=False,
freqmin=source_config['preprocess_filter_params'][0],
freqmax=source_config['preprocess_filter_params'][1],
corners=source_config['preprocess_filter_params'][2],
zerophase=source_config['preprocess_filter_params'][3],
outfile=newfile)
def compute_correlation(input_files, all_conf, nsrc, all_ns, taper,
insta=False):
"""
Compute noise cross-correlations from two .h5 'wavefield' files.
Noise source distribution and spectrum is given by starting_model.h5
It is assumed that noise sources are delta-correlated in space.
Metainformation: Include the reference station names for both stations
from wavefield files, if possible. Do not include geographic information
from .csv file as this might be error-prone. Just add the geographic
info later if needed.
"""
wf1, wf2 = input_files
ntime, n, n_corr, Fs = all_ns
ntraces = nsrc.src_loc[0].shape[0]
correlation = np.zeros(n_corr)
if insta:
# open database
dbpath = all_conf.config['wavefield_path']
# open
db = instaseis.open_db(dbpath)
# get receiver locations
station1 = wf1[0]
station2 = wf2[0]
lat1 = geograph_to_geocent(float(wf1[2]))
lon1 = float(wf1[3])
rec1 = instaseis.Receiver(latitude=lat1, longitude=lon1)
lat2 = geograph_to_geocent(float(wf2[2]))
lon2 = float(wf2[3])
rec2 = instaseis.Receiver(latitude=lat2, longitude=lon2)
else:
wf1 = WaveField(wf1)
wf2 = WaveField(wf2)
station1 = wf1.stats['reference_station']
station2 = wf2.stats['reference_station']
# Make sure all is consistent
if False in (wf1.sourcegrid[1, 0:10] == wf2.sourcegrid[1, 0:10]):
raise ValueError("Wave fields not consistent.")
if False in (wf1.sourcegrid[1, -10:] == wf2.sourcegrid[1, -10:]):
raise ValueError("Wave fields not consistent.")
if False in (wf1.sourcegrid[0, -10:] == nsrc.src_loc[0, -10:]):
raise ValueError("Wave field and source not consistent.")
# Loop over source locations
print_each_n = max(5, round(max(ntraces // 5, 1), -1))
for i in range(ntraces):
# noise source spectrum at this location
S = nsrc.get_spect(i)
if S.sum() == 0.:
# If amplitude is 0, continue. (Spectrum has 0 phase anyway.)
continue
if insta:
# get source locations
lat_src = geograph_to_geocent(nsrc.src_loc[1, i])
lon_src = nsrc.src_loc[0, i]
fsrc = instaseis.ForceSource(latitude=lat_src,
longitude=lon_src,
f_r=1.e12)
Fs = all_conf.config['wavefield_sampling_rate']
s1 = db.get_seismograms(source=fsrc, receiver=rec1,
dt=1. / Fs)[0].data * taper
s2 = db.get_seismograms(source=fsrc, receiver=rec2,
dt=1. / Fs)[0].data * taper
s1 = np.ascontiguousarray(s1)
s2 = np.ascontiguousarray(s2)
spec1 = np.fft.rfft(s1, n)
spec2 = np.fft.rfft(s2, n)
else:
if not wf1.fdomain:
# read Green's functions
s1 = np.ascontiguousarray(wf1.data[i, :] * taper)
s2 = np.ascontiguousarray(wf2.data[i, :] * taper)
# Fourier transform for greater ease of convolution
spec1 = np.fft.rfft(s1, n)
spec2 = np.fft.rfft(s2, n)
else:
spec1 = np.ascontiguousarray(wf1.data[i, :])
spec2 = np.ascontiguousarray(wf2.data[i, :])
# convolve G1G2
g1g2_tr = np.multiply(np.conjugate(spec1), spec2)
# convolve noise source
c = np.multiply(g1g2_tr, S)
# transform back
correlation += my_centered(np.fft.fftshift(np.fft.irfft(c, n)),
n_corr) * nsrc.surf_area[i]
# occasional info
if i % print_each_n == 0 and all_conf.config['verbose']:
print("Finished {} of {} source locations.".format(i, ntraces))
# end of loop over all source locations #######################################
return(correlation, station1, station2)
from matplotlib.mlab import griddata
import matplotlib.tri as tri
#################################
v = 1.
stations = [(0., 0.)]
lonmin = -120.
lonmax = 120.
latmin = -60.
latmax = 60.
latc = 0.0
lonc = 0.0
resolution = 4
fps = 0.5
wf = WaveField(sys.argv[1])
t_min = float(sys.argv[2])
t_max = float(sys.argv[3])
t_step = float(sys.argv[4])
filename = sys.argv[5]
#################################
FFMpegWriter = manimation.writers['ffmpeg']
metadata = dict(title='Wavefield',
artist='Matplotlib',
comment='Movie support!')
writer = FFMpegWriter(fps=fps, metadata=metadata)
fig = plt.figure()
plt.subplot(111)
config = json.load(open(os.path.join(projectpath, 'config.json')))
source_config = json.load(open(os.path.join(sourcepath, 'source_config.json')))
print 'Loaded config files.'
if source_config['preprocess_do']:
ext = '*.h5'
wavefield_path = os.path.join(sourcepath, 'wavefield_processed')
else:
ext = '*.h5'
wavefield_path = config['wavefield_path']
wfs = glob(os.path.join(wavefield_path, ext))
if wfs != []:
print 'Found wavefield.'
with WaveField(wfs[0]) as wf:
df = wf.stats['Fs']
nt = wf.stats['nt']
# The number of points for the fft is larger due to zeropadding --> apparent higher frequency sampling\n",
n = next_fast_len(2 * nt - 1)
freq = np.fft.rfftfreq(n, d=1. / df)
taper = cosine_taper(len(freq), 0.01)
print 'Determined frequency axis.'
def get_distance(grid, location):
def f(lat, lon, location):
return abs(gps2dist_azimuth(lat, lon, location[0], location[1])[0])
def compute_kernel(input_files,
output_file,
all_conf,
nsrc,
all_ns,
taper,
insta=False):
ntime, n, n_corr, Fs = all_ns
wf1, wf2, adjt = input_files
########################################################################
# Prepare filenames and adjoint sources
########################################################################
adjt_srcs = open_adjoint_sources(all_conf, adjt, n_corr)
if None in adjt_srcs:
return (None)
else:
if all_conf.config["verbose"]:
print("========================================")
print("Computing: " + output_file)
# Uniform spatial weights. (current model is in the adjoint source)
nsrc.distr_basis = np.ones(nsrc.distr_basis.shape)
ntraces = nsrc.src_loc[0].shape[0]
# [comp1, comp2] = [wf1, wf2] # keep these strings in case need to be rotated
if insta:
# open database
dbpath = all_conf.config['wavefield_path']
# open and determine Fs, nt
db = instaseis.open_db(dbpath)
# get receiver locations
lat1 = geograph_to_geocent(float(wf1[2]))
lon1 = float(wf1[3])
rec1 = instaseis.Receiver(latitude=lat1, longitude=lon1)
lat2 = geograph_to_geocent(float(wf2[2]))
lon2 = float(wf2[3])
rec2 = instaseis.Receiver(latitude=lat2, longitude=lon2)
else:
wf1 = WaveField(wf1)
wf2 = WaveField(wf2)
# Make sure all is consistent
if False in (wf1.sourcegrid[1, 0:10] == wf2.sourcegrid[1, 0:10]):
raise ValueError("Wave fields not consistent.")
if False in (wf1.sourcegrid[1, -10:] == wf2.sourcegrid[1, -10:]):
raise ValueError("Wave fields not consistent.")
if False in (wf1.sourcegrid[0, -10:] == nsrc.src_loc[0, -10:]):
raise ValueError("Wave field and source not consistent.")
kern = np.zeros(
(nsrc.spect_basis.shape[0], all_conf.filtcnt, ntraces, len(adjt)))
if all_conf.source_config["rotate_horizontal_components"]:
tempfile = output_file + ".h5_temp"
temp = wf1.copy_setup(tempfile, ntraces=ntraces, nt=n_corr)
map_temp_datasets = {0: temp.data}
for ix_spec in range(1, nsrc.spect_basis.shape[0]):
dtmp = temp.file.create_dataset('data{}'.format(ix_spec),
temp.data.shape,
dtype=np.float32)
map_temp_datasets[ix_spec] = dtmp
# Loop over locations
print_each_n = max(5, round(max(ntraces // 5, 1), -1))
# preload wavefield and spectrum
S_all = nsrc.get_spect_all()
wf1_data = np.asarray(wf1.data)
wf2_data = np.asarray(wf2.data)
for i in range(ntraces):
# noise source spectrum at this location
# For the kernel, this contains only the basis functions of the
# spectrum without weights; might still be location-dependent,
# for example when constraining sensivity to ocean
#S = nsrc.get_spect(i)
S = S_all[i, :]
if S.sum() == 0.:
# The spectrum has 0 phase so only checking
# absolute value here
continue
if insta:
# get source locations
lat_src = geograph_to_geocent(nsrc.src_loc[1, i])
lon_src = nsrc.src_loc[0, i]
fsrc = instaseis.ForceSource(latitude=lat_src,
longitude=lon_src,
f_r=1.e12)
dt = 1. / all_conf.source_config['sampling_rate']
s1 = db.get_seismograms(source=fsrc, receiver=rec1,
dt=dt)[0].data * taper
s1 = np.ascontiguousarray(s1)
s2 = db.get_seismograms(source=fsrc, receiver=rec2,
dt=dt)[0].data * taper
s2 = np.ascontiguousarray(s2)
spec1 = np.fft.rfft(s1, n)
spec2 = np.fft.rfft(s2, n)
else:
if not wf1.fdomain:
s1 = np.ascontiguousarray(wf1_data[i, :] * taper)
s2 = np.ascontiguousarray(wf2_data[i, :] * taper)
# if horizontal component rotation: perform it here
# more convenient before FFT to avoid additional FFTs
spec1 = np.fft.rfft(s1, n)
spec2 = np.fft.rfft(s2, n)
else:
spec1 = wf1_data[i, :]
spec2 = wf2_data[i, :]
g1g2_tr = np.multiply(np.conjugate(spec1), spec2)
# spectrum
for ix_spec in range(nsrc.spect_basis.shape[0]):
c = np.multiply(g1g2_tr, nsrc.spect_basis[ix_spec, :])
###################################################################
# Get Kernel at that location
###################################################################
ctemp = np.fft.fftshift(np.fft.irfft(c, n))
corr_temp = my_centered(ctemp, n_corr)
if all_conf.source_config["rotate_horizontal_components"]:
map_temp_datasets[ix_spec][i, :] = corr_temp
###################################################################
# Apply the 'adjoint source'
###################################################################
for ix_f in range(all_conf.filtcnt):
f = adjt_srcs[ix_f]
if f is None:
continue
for j in range(len(f)):
delta = f[j].stats.delta
kern[ix_spec, ix_f, i,
j] = np.dot(corr_temp, f[j].data) * delta
if i % print_each_n == 0 and all_conf.config['verbose']:
print("Finished {} of {} source locations.".format(i, ntraces))
if not insta:
wf1.file.close()
wf2.file.close()
if all_conf.source_config["rotate_horizontal_components"]:
temp.file.close()
return kern
# preparations:
os.mkdir('test/testdata/testsrc/step_0/corr')
os.system('cp -R test/testdata/testsrc/wavefield_processed_archived \
test/testdata/testsrc/wavefield_processed')
os.system('cp test/testdata/config_archived.json \
test/testdata/config.json')
os.system('cp test/testdata/testsrc/measr_config_archived.json \
test/testdata/testsrc/measr_config.json')
os.system('cp test/testdata/testsrc/source_config_archived.json \
test/testdata/testsrc/source_config.json')
m_a_options = {'g_speed': g_speed, 'window_params': window_params}
m_func = rm.get_measure_func(mtype)
wf = WaveField('test/testdata/wavefield_vel/NET.STA1..CHA.h5')
nlocs = wf.stats['ntraces']
# create perturbation
d_q = 2 * (np.random.rand(nlocs, ) - 0.5)
# evaluate original misfit and load original gradient
m_a_options = {'g_speed': g_speed, 'window_params': window_params}
m_func = rm.get_measure_func(mtype)
# open the files....
obs = read('test/testdata/testsrc/observed_correlations/*.sac')[0]
syn = read('test/testdata/testsrc/step_0/corr_archived/*.sac')[0]
syn.stats.sac = {}
syn.stats.sac['dist'] = obs.stats.sac.dist
msr_o = m_func(obs, **m_a_options)
def g1g2_kern(wf1, wf2, corr_file, kernel, adjt, src, source_conf, scale=1.0):
#ToDo: Take care of saving metainformation
#ToDo: Think about how to manage different types of sources (numpy array vs. get from configuration -- i.e. read source from file as option)
#ToDo: check whether to include autocorrs from user (now hardcoded off)
#ToDo: Parallel loop(s)
#ToDo tests
ntime, n, n_corr = get_ns(wf1, source_conf)
taper = cosine_taper(ntime, p=0.05)
with WaveField(wf1) as wf1, WaveField(wf2) as wf2:
if wf1.stats['Fs'] != wf2.stats['Fs']:
msg = 'Sampling rates of synthetic green\'s functions must match.'
raise ValueError(msg)
# initialize new hdf5 files for correlation and green's function correlation
#with wf1.copy_setup(corr_file,nt=n_corr) as correl, NoiseSource(src) as nsrc:
#with wf1.copy_setup(corr_file,nt=n_corr) as correl:
with NoiseSource(src) as nsrc:
correlation = np.zeros(n_corr)
kern = np.zeros(wf1.stats['ntraces'])
# Try to use a trick: Use causal and acausal part of f separately.
f = read(adjt)[0]
f.data = my_centered(f.data, n_corr)
n_acausal_samples = (f.stats.npts - 1) / 2
specf = np.fft.rfft(f[n_acausal_samples:], n)
# Loop over source locations
#with click.progressbar(range(wf1.stats['ntraces']),\
#label='Correlating...' ) as ind:
for i in range(wf1.stats['ntraces']):
#print(i)
s1 = np.ascontiguousarray(wf1.data[i, :] * taper) * scale
#print(s1.sum())
spec1 = np.fft.rfft(s1, n)
#print(spec1.sum())
T = np.multiply(np.conj(spec1), nsrc.get_spect(i))
# plt.plot(np.abs(spec1)/np.max(np.abs(spec1)))
# plt.plot(np.abs(T)/np.max(np.abs(T)))
# plt.show()
# it would be cleaner to use ifftshift here!
T = np.fft.ifftshift(np.fft.irfft(T,
n))[0:len(s1)] #[-len(s1):]
# if i in [1,2,3,4,5]:
# plt.plot(T[::-1]/np.max(np.abs(T)))
# plt.plot(s1/np.max(np.abs(s1)),'--')
# plt.show()
# Get s2 in the shape of f
# we need to add half of the length of f before G to replace the
# acausal part that G does not have to start with:
#s2 = np.zeros(n)
s2 = np.ascontiguousarray(wf2.data[i, :] * taper) * scale
#print(s2.sum())
#s2[n_acausal_samples:n_acausal_samples+ntime] = s2_caus
spec2 = np.fft.rfft(s2, n)
# plt.plot(s2_caus/np.max(np.abs(s2_caus)))
# plt.plot(f.data/np.max(np.abs(f.data)))
# plt.plot(s2/np.max(np.abs(s2)))
# plt.plot(n_acausal_samples,0.5,'rd')
# plt.show()
# transform both f and s2 to fourier d
# (again, zeropadding but just to avoid circular convolution)
# plt.plot(np.abs(spec2)/np.max(np.abs(spec2)))
# plt.plot(np.abs(specf)/np.max(np.abs(specf)))
# plt.show()
g2f_tr = np.multiply(np.conj(spec2), specf)
#print(specf.sum())
#plt.plot(n_acausal_samples,0.5,'rd')
#plt.plot(n,0.5,'gd')
# it would be cleaner to use ifftshift here!
u_dagger = np.fft.ifftshift(np.fft.irfft(
g2f_tr, n))[0:len(s1)] #[-len(s1):]
# plt.plot(u_dagger/np.max(np.abs(u_dagger)))
# plt.plot(T/np.max(np.abs(T)))
# plt.show()
# The frequency spectrum of the noise source is included here
## corr_temp = my_centered(np.fft.ifftshift(np.fft.irfft(c,n)),n_corr)
# A Riemann sum -- one could actually build in a more fancy integration here
#print(f.stats.delta)
kern[i] = np.dot(u_dagger, T) * f.stats.delta
#print(kern[i])
if i % 50000 == 0:
print("Finished {} source locations.".format(i))
#np.save(kernel,kern)
return (kern)
def g1g2_corr(wf1, wf2, corr_file, kernel, adjt, src, source_conf, kernelrun):
#ToDo: Take care of saving metainformation
#ToDo: Think about how to manage different types of sources (numpy array vs. get from configuration -- i.e. read source from file as option)
#ToDo: check whether to include autocorrs from user (now hardcoded off)
#ToDo: Parallel loop(s)
#ToDo tests
ntime, n, n_corr = get_ns(wf1, source_conf)
taper = cosine_taper(ntime, p=0.05)
with WaveField(wf1) as wf1, WaveField(wf2) as wf2:
if wf1.stats['Fs'] != wf2.stats['Fs']:
msg = 'Sampling rates of synthetic green\'s functions must match.'
raise ValueError(msg)
# initialize new hdf5 files for correlation and green's function correlation
#with wf1.copy_setup(corr_file,nt=n_corr) as correl, NoiseSource(src) as nsrc:
#with wf1.copy_setup(corr_file,nt=n_corr) as correl:
with NoiseSource(src) as nsrc:
correlation = np.zeros(n_corr)
if kernelrun:
#if not os.path.exists(adjt):
# print('Adjoint source %s not found, skipping kernel.')
# return()
kern = np.zeros((wf1.stats['ntraces'], len(adjt)))
f = Stream()
for adjtfile in adjt:
if adjtfile == '-':
return
f += read(adjtfile)[0]
f[-1].data = my_centered(f[-1].data, n_corr)
# Loop over source locations
#with click.progressbar(range(wf1.stats['ntraces']),\
#label='Correlating...' ) as ind:
for i in range(wf1.stats['ntraces']):
# noise source spectrum at this location
# if calculating kernel, the spectrum is location independent.
S = nsrc.get_spect(i)
if S.sum() == 0.: # The spectrum has 0 phase anyway
continue
s1 = np.ascontiguousarray(wf1.data[i, :] * taper)
s2 = np.ascontiguousarray(wf2.data[i, :] * taper)
spec1 = np.fft.rfft(s1, n)
spec2 = np.fft.rfft(s2, n)
g1g2_tr = np.multiply(np.conjugate(spec1), spec2)
# if i%50000 == 0:
# print(g1g2_tr[0:10],file=None)
# print(g1g2_tr.max(),file=None)
c = np.multiply(g1g2_tr, S)
# if i%50000==0:
# print(c[0:10],file=None)
# print(c.max(),file=None)
if kernelrun:
corr_temp = my_centered(
np.fft.ifftshift(np.fft.irfft(c, n)), n_corr)
##if i%50000 == 0:
# ##print(corr_temp[0:10],file=None)
#print(corr_temp.max(),file=None)
# A Riemann sum
for j in range(len(adjt)):
kern[i, j] = np.dot(corr_temp,
f[j].data) * f[j].stats.delta
else:
correlation += my_centered(
np.fft.ifftshift(np.fft.irfft(c, n)), n_corr)
if i % 50000 == 0:
print("Finished {} source locations.".format(i))
if kernelrun:
np.save(kernel, kern)
else:
trace = Trace()
trace.stats.sampling_rate = wf1.stats['Fs']
trace.data = correlation
trace.write(filename=corr_file, format='SAC')
def setup_source_startingmodel(self, args):
# plotting:
colors = ['purple', 'g', 'b', 'orange']
colors_cmaps = [
plt.cm.Purples, plt.cm.Greens, plt.cm.Blues, plt.cm.Oranges
]
print("Setting up source starting model.", end="\n")
with io.open(os.path.join(args.source_model, 'source_config.yml'),
'r') as fh:
source_conf = yaml.safe_load(fh)
with io.open(os.path.join(source_conf['project_path'], 'config.yml'),
'r') as fh:
conf = yaml.safe_load(fh)
with io.open(source_conf['source_setup_file'], 'r') as fh:
parameter_sets = yaml.safe_load(fh)
if conf['verbose']:
print("The following input parameters are used:", end="\n")
pp = pprint.PrettyPrinter()
pp.pprint(parameter_sets)
# load the source locations of the grid
grd = np.load(os.path.join(conf['project_path'], 'sourcegrid.npy'))
# add the approximate spherical surface elements
if grd.shape[-1] < 50000:
surf_el = get_spherical_surface_elements(grd[0], grd[1])
else:
warn('Large grid; surface element computation slow. Using \
approximate surface elements.')
surf_el = np.ones(grd.shape[-1]) * conf['grid_dx_in_m']**2
# get the relevant array sizes
wfs = glob(os.path.join(conf['project_path'], 'greens', '*.h5'))
if wfs != []:
if conf['verbose']:
print('Found wavefield stats.')
else:
pass
else:
raise FileNotFoundError('No wavefield database found. Run \
precompute_wavefield first.')
with WaveField(wfs[0]) as wf:
df = wf.stats['Fs']
n = wf.stats['npad']
freq = np.fft.rfftfreq(n, d=1. / df)
n_distr = len(parameter_sets)
coeffs = np.zeros((grd.shape[-1], n_distr))
spectra = np.zeros((n_distr, len(freq)))
# fill in the distributions and the spectra
for i in range(n_distr):
coeffs[:, i] = self.distribution_from_parameters(
grd, parameter_sets[i], conf['verbose'])
# plot
outfile = os.path.join(args.source_model,
'source_starting_model_distr%g.png' % i)
if create_plot:
plot_grid(grd[0],
grd[1],
coeffs[:, i],
outfile=outfile,
cmap=colors_cmaps[i % len(colors_cmaps)],
sequential=True,
normalize=False,
quant_unit='Spatial weight (-)',
axislabelpad=-0.1,
size=10)
spectra[i, :] = self.spectrum_from_parameters(
freq, parameter_sets[i])
# plotting the spectra
# plotting is not necessarily done to make sure code runs on clusters
if create_plot:
fig1 = plt.figure()
ax = fig1.add_subplot('111')
for i in range(n_distr):
ax.plot(freq,
spectra[i, :] / spectra.max(),
color=colors[i % len(colors_cmaps)])
ax.set_xlabel('Frequency / Nyquist Frequency')
plt.xticks([
0,
freq.max() * 0.25,
freq.max() * 0.5,
freq.max() * 0.75,
freq.max()
], ['0', '0.25', '0.5', '0.75', '1'])
ax.set_ylabel('Rel. PSD norm. to strongest spectrum (-)')
fig1.savefig(
os.path.join(args.source_model,
'source_starting_model_spectra.png'))
# Save to an hdf5 file
with h5py.File(
os.path.join(args.source_model, 'iteration_0',
'starting_model.h5'), 'w') as fh:
fh.create_dataset('coordinates', data=grd)
fh.create_dataset('frequencies', data=freq)
fh.create_dataset('model', data=coeffs.astype(np.float))
fh.create_dataset('spectral_basis', data=spectra.astype(np.float))
fh.create_dataset('surface_areas', data=surf_el.astype(np.float))
# Save to an hdf5 file
with h5py.File(os.path.join(args.source_model, 'spectral_model.h5'),
'w') as fh:
uniform_spatial = np.ones(coeffs.shape) * 1.0
fh.create_dataset('coordinates', data=grd)
fh.create_dataset('frequencies', data=freq)
fh.create_dataset('model', data=uniform_spatial.astype(np.float))
fh.create_dataset('spectral_basis', data=spectra.astype(np.float))
fh.create_dataset('surface_areas', data=surf_el.astype(np.float))
import matplotlib.tri as tri
#################################
v = 1.
stations = [(0.,0.)]
lonmin=-120.
lonmax=120.
latmin=-60.
latmax=60.
latc=0.0
lonc=0.0
resolution = 4
fps = 0.5
wf = WaveField(sys.argv[1])
t_min = float(sys.argv[2])
t_max = float(sys.argv[3])
t_step = float(sys.argv[4])
filename = sys.argv[5]
#################################
FFMpegWriter = manimation.writers['ffmpeg']
metadata = dict(title='Wavefield', artist='Matplotlib',
comment='Movie support!')
writer = FFMpegWriter(fps=fps, metadata=metadata)
fig = plt.figure()
plt.subplot(111)
map_x = wf.sourcegrid[0]
def g1g2_corr(wf1, wf2, corr_file, src, source_conf, insta):
"""
Compute noise cross-correlations from two .h5 'wavefield' files.
Noise source distribution and spectrum is given by starting_model.h5
It is assumed that noise sources are delta-correlated in space.
"""
#ToDo: check whether to include autocorrs from user (now hardcoded off)
#ToDo: Parallel loop(s)
#ToDo tests
# Metainformation: Include the reference station names for both stations
# from wavefield files, if possible. Do not include geographic information
# from .csv file as this might be error-prone. Just add the geographic
# info later if needed.
with NoiseSource(src) as nsrc:
ntime, n, n_corr, Fs = get_ns(wf1, source_conf, insta)
# use a one-sided taper: The seismogram probably has a non-zero end,
# being cut off whereever the solver stopped running.
taper = cosine_taper(ntime, p=0.01)
taper[0:ntime // 2] = 1.0
ntraces = nsrc.src_loc[0].shape[0]
print(taper.shape)
correlation = np.zeros(n_corr)
if insta:
# open database
dbpath = json.load(
open(os.path.join(source_conf['project_path'],
'config.json')))['wavefield_path']
# open and determine Fs, nt
db = instaseis.open_db(dbpath)
# get receiver locations
lat1 = geograph_to_geocent(float(wf1[2]))
lon1 = float(wf1[3])
rec1 = instaseis.Receiver(latitude=lat1, longitude=lon1)
lat2 = geograph_to_geocent(float(wf2[2]))
lon2 = float(wf2[3])
rec2 = instaseis.Receiver(latitude=lat2, longitude=lon2)
else:
wf1 = WaveField(wf1)
wf2 = WaveField(wf2)
# Loop over source locations
for i in range(ntraces):
# noise source spectrum at this location
S = nsrc.get_spect(i)
if S.sum() == 0.:
#If amplitude is 0, continue. (Spectrum has 0 phase anyway. )
continue
if insta:
# get source locations
lat_src = geograph_to_geocent(nsrc.src_loc[1, i])
lon_src = nsrc.src_loc[0, i]
fsrc = instaseis.ForceSource(latitude=lat_src,
longitude=lon_src,
f_r=1.e12)
s1 = np.ascontiguousarray(
db.get_seismograms(
source=fsrc,
receiver=rec1,
dt=1. / source_conf['sampling_rate'])[0].data * taper)
s2 = np.ascontiguousarray(
db.get_seismograms(
source=fsrc,
receiver=rec2,
dt=1. / source_conf['sampling_rate'])[0].data * taper)
else:
# read Green's functions
s1 = np.ascontiguousarray(wf1.data[i, :] * taper)
s2 = np.ascontiguousarray(wf2.data[i, :] * taper)
# Fourier transform for greater ease of convolution
spec1 = np.fft.rfft(s1, n)
spec2 = np.fft.rfft(s2, n)
# convolve G1G2
g1g2_tr = np.multiply(np.conjugate(spec1), spec2)
# convolve noise source
c = np.multiply(g1g2_tr, S)
# transform back
correlation += my_centered(np.fft.ifftshift(np.fft.irfft(c, n)),
n_corr) * nsrc.surf_area[i]
# occasional info
if i % 50000 == 0:
print("Finished {} source locations.".format(i))
###################### end of loop over all source locations ###################
if not insta:
wf1.file.close()
wf2.file.close()
# save output
trace = Trace()
trace.stats.sampling_rate = Fs
trace.data = correlation
# try to add some meta data
try:
sta1 = wf1.stats['reference_station']
sta2 = wf2.stats['reference_station']
trace.stats.station = sta1.split('.')[1]
trace.stats.network = sta1.split('.')[0]
trace.stats.location = sta1.split('.')[2]
trace.stats.channel = sta1.split('.')[3]
trace.stats.sac = {}
trace.stats.sac['kuser0'] = sta2.split('.')[1]
trace.stats.sac['kuser1'] = sta2.split('.')[0]
trace.stats.sac['kuser2'] = sta2.split('.')[2]
trace.stats.sac['kevnm'] = sta2.split('.')[3]
except:
pass
trace.write(filename=corr_file, format='SAC')