def test_prem(self):
model_data = TestFunctionTaup.gen_test_model("prem")
model_prem = SeisModel(model=model_data)
source_prem = SourceModel(sdep=16.5)
# * note, use larger distance will integrate more, the waveform of only calculating 10km and calculating to 100km will be grealy different
config_prem = Config(model=model_prem,
source=source_prem,
npt=512,
dt=0.1,
receiver_distance=np.arange(10, 20, 10))
gf = calculate_gf(config_prem)
# # * calculate sync_prem_gcmt for the event
event = obspy.read_events(
join(dirname(__file__), "../data/sync_prem_gcmt/test_gcmt"))[0]
source_prem.update_source_mechanism(event)
# # * generate a source time function
source_time_function = generate_source_time_function(
4, 0.5, gf[0][0].stats.delta)
sync_result = calculate_sync(gf, config_prem, 30, source_time_function)
# * test if the cc value is large enough
for index, component in enumerate(["z", "r", "t"]):
sac_path = join(dirname(__file__),
f"../data/sync_prem_gcmt/prem.{component}")
sac_wave = obspy.read(sac_path)[0].data
coef = np.corrcoef(
sac_wave,
sync_result[0][index].data,
)[0, 1]
assert coef > 0.9999
def test_hk():
# * perl fk.pl -Mhk/15/k -N512/0.1 10 20 30
model_path = join(dirname(__file__), f"../data/hk")
model_data = np.loadtxt(model_path)
model_hk = SeisModel(model=model_data, use_kappa=True)
source_hk = SourceModel(sdep=15)
config_hk = Config(model=model_hk,
source=source_hk,
npt=512,
dt=0.1,
receiver_distance=[10, 20, 30])
result = calculate_gf(config_hk)
# * for all the gf in data/hk_gf, test if the results are close (in FK, it uses float but we are using double)
for irec, each_rec in enumerate([10, 20, 30]):
for icomn in range(9):
hk_gf_data = obspy.read(
join(dirname(__file__),
f"../data/hk_gf/{each_rec}.grn.{icomn}"))[0]
coef = np.corrcoef(
hk_gf_data.data,
result[irec][icomn].data,
)[0, 1]
if np.isnan(coef):
coef = 1.0
assert coef > 0.99
def test_big_array(self):
# model_data = TestFunctionTaup.gen_test_model("prem")
# there is a possibility that we write x=f(x) where x is a memoryview in the code
# this might cause segmentation fault
model_data = np.loadtxt(join(dirname(__file__), f"../data/hk"))
model_hk = SeisModel(model=model_data)
source_hk = SourceModel(sdep=16.5)
config_hk = Config(model=model_hk,
source=source_hk,
npt=512,
dt=0.1,
receiver_distance=np.arange(10, 40, 10))
_ = calculate_gf(config_hk)
def test_static_source(self):
model_data = TestFunctionTaup.gen_test_model("prem")
model_prem = SeisModel(model=model_data)
source_prem = SourceModel(sdep=16.5, srcType="dc")
config_prem = Config(model=model_prem,
source=source_prem,
npt=1,
dt=1,
receiver_distance=[50])
gf = calculate_gf(config_prem)
ref_gf = [
-0.242E-06, -0.103E-05, 0.000E+00, 0.236E-06, 0.118E-05,
-0.548E-07, -0.942E-07, -0.156E-05, 0.285E-06
]
coef = np.corrcoef(
gf,
ref_gf,
)[0, 1]
assert coef > 0.99999
def test_prem_sf(self):
model_data = TestFunctionTaup.gen_test_model("prem")
model_prem = SeisModel(model=model_data)
source_prem = SourceModel(sdep=16.5, srcType="sf")
config_prem = Config(model=model_prem,
source=source_prem,
npt=512,
dt=1,
receiver_distance=[50])
gf = calculate_gf(config_prem)
for index, comnname in enumerate(range(6)):
gf_data = obspy.read(
join(dirname(__file__),
f"../data/sync_prem_sf/50.grn.{comnname}"))[0]
coef = np.corrcoef(
gf_data.data,
gf[0][index].data,
)[0, 1]
if np.isnan(coef):
coef = 1.
assert coef > 0.99
def test_exceptions(self):
model_data = TestFunctionTaup.gen_test_model("prem")
model_prem = SeisModel(model=model_data)
source_prem = SourceModel(sdep=16.5)
# * note, use larger distance will integrate more, the waveform of only calculating 10km and calculating to 100km will be grealy different
# ! note receiver_distance can not be 0
config_prem = Config(model=model_prem,
source=source_prem,
npt=512,
dt=0.1,
receiver_distance=np.arange(1, 10))
gf = calculate_gf(config_prem)
event = obspy.read_events(
join(dirname(__file__), "../data/sync_prem_gcmt/test_gcmt"))[0]
source_prem.update_source_mechanism(event)
source_time_function = generate_source_time_function(
4, 0.5, gf[0][0].stats.delta)
# * the main tests
with pytest.raises(PyfkError) as execinfo:
_ = calculate_sync(gf, config_prem, [30], source_time_function)
assert str(execinfo.value) == "az must be a number"
with pytest.raises(PyfkError) as execinfo:
_ = calculate_sync(gf, config_prem, 30, None)
assert str(execinfo.value) == "must provide a source time function"
with pytest.raises(PyfkError) as execinfo:
source_time_function_abnormal = generate_source_time_function(
4, 0.5, 1.2)
_ = calculate_sync(None, config_prem, 30,
source_time_function_abnormal)
assert str(execinfo.value) == "check input Green's function"
with pytest.raises(PyfkError) as execinfo:
source_time_function_abnormal = generate_source_time_function(
4, 0.5, 1.2)
_ = calculate_sync(gf, config_prem, 30,
source_time_function_abnormal)
assert str(
execinfo.value
) == "delta for the source time function and the Green's function should be the same"