def test_get_recurrence_simple_characteristic(self):
'''
Tests the function to get the recurrence calculation for a simple
characteristic earthquake
'''
self.mfd_config = {
'MFD_spacing': 0.1,
'Model_Name': 'Characteristic',
'Model_Weight': 1.0,
'Maximum_Magnitude': None,
'Maximum_Uncertainty': None,
'Lower_Bound': -2.,
'Upper_Bound': 2.,
'Sigma': 0.12
}
self.model = RecurrenceBranch(8500., 5., WC1994(), 0., 30.)
self.model.get_recurrence(self.mfd_config)
# Test the same process using just the openquake.hmtk.faults.mfd.characteristic
# Implementation
test_model = Characteristic()
test_model.setUp(self.mfd_config)
test_model.get_mmax(self.mfd_config, WC1994(), 0., 8500.)
_ = test_model.get_mfd(5.0, 8500., 30.)
self.assertTrue(isinstance(self.model.recurrence, IncrementalMFD))
self.assertAlmostEqual(self.model.recurrence.min_mag, test_model.mmin)
self.assertAlmostEqual(self.model.recurrence.bin_width,
test_model.bin_width)
np.testing.assert_array_almost_equal(self.model.recurrence.occur_rates,
test_model.occurrence_rate)
def test_get_recurrence_simple_characteristic(self):
'''
Tests the function to get the recurrence calculation for a simple
characteristic earthquake
'''
self.mfd_config = {'MFD_spacing': 0.1,
'Model_Name': 'Characteristic',
'Model_Weight': 1.0,
'Maximum_Magnitude': None,
'Maximum_Uncertainty': None,
'Lower_Bound': -2.,
'Upper_Bound': 2., 'Sigma': 0.12}
self.model = RecurrenceBranch(8500., 5., WC1994(), 0., 30.)
self.model.get_recurrence(self.mfd_config)
# Test the same process using just the openquake.hmtk.faults.mfd.characteristic
# Implementation
test_model = Characteristic()
test_model.setUp(self.mfd_config)
test_model.get_mmax(self.mfd_config, WC1994(), 0., 8500.)
_ = test_model.get_mfd(5.0, 8500., 30.)
self.assertTrue(isinstance(self.model.recurrence, IncrementalMFD))
self.assertAlmostEqual(self.model.recurrence.min_mag,
test_model.mmin)
self.assertAlmostEqual(self.model.recurrence.bin_width,
test_model.bin_width)
np.testing.assert_array_almost_equal(self.model.recurrence.occur_rates,
test_model.occurrence_rate)
def build_fault_model(area, slip, mmax, config={}, mu=30.0, coupling=1.0):
"""
"""
if not config:
# Simple Dirac function
geological_moment = coupling * (mu * 1.0E9) * (area * 1.0E6) * (slip *
1.0E-3)
mmax_moment = moment_function(mmax)
rate = geological_moment / mmax_moment
mf_dist = EvenlyDiscretizedMFD(mmax, 0.1, [rate])
return get_incremental_cumulative(mf_dist)
d_m = config["MFD_spacing"]
config["Model_Weight"] = 1.0
if config["model"] == "Characteristic":
# Try characteristic
for key in ["MFD_spacing", "Lower_Bound", "Upper_Bound", "Sigma"]:
# Verify everything is present
assert key in config and config[key] is not None
rec_model = Characteristic()
rec_model.setUp(config)
rec_model.mmax = mmax
elif config["model"] == "Hybrid":
# Try Hybrid
for key in ["Minimum_Magnitude", "b_value"]:
# Verify everything is present
assert key in config and config[key] is not None
rec_model = YoungsCoppersmithCharacteristic()
rec_model.setUp(config)
rec_model.mmax = mmax
elif config["model"] == "Exponential":
# Try exponential
for key in ["Minimum_Magnitude", "b_value"]:
# Verify everything is present
assert key in config and config[key] is not None
rec_model = YoungsCoppersmithExponential()
rec_model.setUp(config)
rec_model.mmax = mmax
rec_model.mmin += (config["MFD_spacing"] / 2.)
else:
raise ValueError("Model %s not recognised" % config["model"])
mmin, bin_width, rate = rec_model.get_mfd(slip, area, mu)
mf_dist = EvenlyDiscretizedMFD(mmin, bin_width, rate.tolist())
return get_incremental_cumulative(mf_dist)
class TestSimpleCharacteristic(unittest.TestCase):
'''
Implements the basic set of tests for the simple estimator of the
characteristic earthquake for a fault
:class openquake.hmtk.faults.mfd.characteristic.Characteristic
'''
def setUp(self):
'''
'''
self.model = Characteristic()
self.config = {
'MFD_spacing': 0.1,
'Model_Weight': 1.0,
'Maximum_Magnitude': None,
'Maximum_Uncertainty': None,
'Lower_Bound': -2.,
'Upper_Bound': 2.,
'Sigma': None
}
self.msr = WC1994()
def test_model_setup(self):
'''
Simple test to ensure model sets up correctly
'''
self.model.setUp(self.config)
expected_dict = {
'bin_width': 0.1,
'lower_bound': -2.0,
'mfd_model': 'Characteristic',
'mfd_weight': 1.0,
'mmax': None,
'mmax_sigma': None,
'mmin': None,
'occurrence_rate': None,
'sigma': None,
'upper_bound': 2.0
}
self.assertDictEqual(self.model.__dict__, expected_dict)
def test_get_mmax(self):
'''
Tests the function to get Mmax
Values come from WC1994 (tested in openquake.hazardlib) - only
functionality is tested for here!
'''
# Case 1 MMmax and uncertainty specified in config
self.config['Maximum_Magnitude'] = 8.0
self.config['Maximum_Magnitude_Uncertainty'] = 0.2
self.model = Characteristic()
self.model.setUp(self.config)
self.model.get_mmax(self.config, self.msr, 0., 8500.)
self.assertAlmostEqual(self.model.mmax, 8.0)
self.assertAlmostEqual(self.model.mmax_sigma, 0.2)
# Case 2: Mmax and uncertainty not specified in config
self.config['Maximum_Magnitude'] = None
self.config['Maximum_Magnitude_Uncertainty'] = None
self.model = Characteristic()
self.model.setUp(self.config)
self.model.get_mmax(self.config, self.msr, 0., 8500.)
self.assertAlmostEqual(self.model.mmax, 7.9880073)
self.assertAlmostEqual(self.model.mmax_sigma, 0.23)
def test_get_mfd(self):
'''
Tests the calculation of activity rates for the simple
characteristic earthquake distribution.
'''
# Test case 1: Ordinatry fault with Area 8500 km ** 2 (Mmax ~ 8.0),
# and a slip rate of 5 mm/yr. Double truncated Gaussian between [-2, 2]
# standard deviations with sigma = 0.12
self.config = {
'MFD_spacing': 0.1,
'Model_Weight': 1.0,
'Maximum_Magnitude': None,
'Maximum_Uncertainty': None,
'Lower_Bound': -2.,
'Upper_Bound': 2.,
'Sigma': 0.12
}
self.model = Characteristic()
self.model.setUp(self.config)
self.model.get_mmax(self.config, self.msr, 0., 8500.)
_, _, _ = self.model.get_mfd(5.0, 8500.)
aaae(
self.model.occurrence_rate,
np.array([
4.20932867e-05, 2.10890168e-04, 3.80422666e-04, 3.56294331e-04,
1.73223702e-04, 2.14781079e-05
]))
expected_rate = np.sum(self.model.occurrence_rate)
# Test case 2: Same fault with no standard deviation
self.config['Sigma'] = None
self.model.setUp(self.config)
self.model.get_mmax(self.config, self.msr, 0., 8500.)
_, _, _ = self.model.get_mfd(5.0, 8500.)
aaae(0.0011844, self.model.occurrence_rate)
# As a final check - ensure that the sum of the activity rates from the
# truncated Gaussian model is equal to the rate from the model with no
# variance
aaae(expected_rate, self.model.occurrence_rate, 3)
class TestSimpleCharacteristic(unittest.TestCase):
'''
Implements the basic set of tests for the simple estimator of the
characteristic earthquake for a fault
:class openquake.hmtk.faults.mfd.characteristic.Characteristic
'''
def setUp(self):
'''
'''
self.model = Characteristic()
self.config = {'MFD_spacing': 0.1,
'Model_Weight': 1.0,
'Maximum_Magnitude': None,
'Maximum_Uncertainty': None,
'Lower_Bound': -2.,
'Upper_Bound': 2.,
'Sigma': None}
self.msr = WC1994()
def test_model_setup(self):
'''
Simple test to ensure model sets up correctly
'''
self.model.setUp(self.config)
expected_dict = {'bin_width': 0.1,
'lower_bound': -2.0,
'mfd_model': 'Characteristic',
'mfd_weight': 1.0,
'mmax': None,
'mmax_sigma': None,
'mmin': None,
'occurrence_rate': None,
'sigma': None,
'upper_bound': 2.0}
self.assertDictEqual(self.model.__dict__, expected_dict)
def test_get_mmax(self):
'''
Tests the function to get Mmax
Values come from WC1994 (tested in openquake.hazardlib) - only
functionality is tested for here!
'''
# Case 1 MMmax and uncertainty specified in config
self.config['Maximum_Magnitude'] = 8.0
self.config['Maximum_Magnitude_Uncertainty'] = 0.2
self.model = Characteristic()
self.model.setUp(self.config)
self.model.get_mmax(self.config, self.msr, 0., 8500.)
self.assertAlmostEqual(self.model.mmax, 8.0)
self.assertAlmostEqual(self.model.mmax_sigma, 0.2)
# Case 2: Mmax and uncertainty not specified in config
self.config['Maximum_Magnitude'] = None
self.config['Maximum_Magnitude_Uncertainty'] = None
self.model = Characteristic()
self.model.setUp(self.config)
self.model.get_mmax(self.config, self.msr, 0., 8500.)
self.assertAlmostEqual(self.model.mmax, 7.9880073)
self.assertAlmostEqual(self.model.mmax_sigma, 0.23)
def test_get_mfd(self):
'''
Tests the calculation of activity rates for the simple
characteristic earthquake distribution.
'''
# Test case 1: Ordinatry fault with Area 8500 km ** 2 (Mmax ~ 8.0),
# and a slip rate of 5 mm/yr. Double truncated Gaussian between [-2, 2]
# standard deviations with sigma = 0.12
self.config = {'MFD_spacing': 0.1,
'Model_Weight': 1.0,
'Maximum_Magnitude': None,
'Maximum_Uncertainty': None,
'Lower_Bound': -2.,
'Upper_Bound': 2.,
'Sigma': 0.12}
self.model = Characteristic()
self.model.setUp(self.config)
self.model.get_mmax(self.config, self.msr, 0., 8500.)
_, _, _ = self.model.get_mfd(5.0, 8500.)
aaae(self.model.occurrence_rate,
np.array([4.20932867e-05, 2.10890168e-04, 3.80422666e-04,
3.56294331e-04, 1.73223702e-04, 2.14781079e-05]))
expected_rate = np.sum(self.model.occurrence_rate)
# Test case 2: Same fault with no standard deviation
self.config['Sigma'] = None
self.model.setUp(self.config)
self.model.get_mmax(self.config, self.msr, 0., 8500.)
_, _, _ = self.model.get_mfd(5.0, 8500.)
aaae(0.0011844, self.model.occurrence_rate)
# As a final check - ensure that the sum of the activity rates from the
# truncated Gaussian model is equal to the rate from the model with no
# variance
aaae(expected_rate, self.model.occurrence_rate, 3)