def read_nsmn(filename):
"""Read the Turkish NSMN strong motion data format.
Args:
filename (str): path to NSMN data file.
Returns:
list: Sequence of one StationStream object containing 3 StationTrace objects.
"""
header = _read_header(filename)
header1 = copy.deepcopy(header)
header2 = copy.deepcopy(header)
header3 = copy.deepcopy(header)
header1['standard']['horizontal_orientation'] = 0.0
header1['channel'] = get_channel_name(header['sampling_rate'], True, False,
True)
header2['standard']['horizontal_orientation'] = 90.0
header2['channel'] = get_channel_name(header['sampling_rate'], True, False,
False)
header3['standard']['horizontal_orientation'] = 0.0
header3['channel'] = get_channel_name(header['sampling_rate'], True, True,
False)
# three columns of NS, EW, UD
# data = np.genfromtxt(filename, skip_header=TEXT_HDR_ROWS,
# delimiter=[COLWIDTH] * NCOLS, encoding=ENCODING)
data = np.loadtxt(filename, skiprows=TEXT_HDR_ROWS, encoding=ENCODING)
data1 = data[:, 0]
data2 = data[:, 1]
data3 = data[:, 2]
trace1 = StationTrace(data=data1, header=header1)
trace2 = StationTrace(data=data2, header=header2)
trace3 = StationTrace(data=data3, header=header3)
stream = StationStream(traces=[trace1, trace2, trace3])
return [stream]
def read_nsmn(filename, **kwargs):
"""Read the Turkish NSMN strong motion data format.
Args:
filename (str):
path to NSMN data file.
kwargs (ref):
Other arguments will be ignored.
Returns:
list: Sequence of one StationStream object containing 3
StationTrace objects.
"""
header = _read_header(filename)
header1 = copy.deepcopy(header)
header2 = copy.deepcopy(header)
header3 = copy.deepcopy(header)
header1['standard']['horizontal_orientation'] = 0.0
header1['standard']['vertical_orientation'] = np.nan
header1['channel'] = get_channel_name(header['sampling_rate'], True, False,
True)
header1['standard']['units_type'] = get_units_type(header1['channel'])
header2['standard']['horizontal_orientation'] = 90.0
header2['standard']['vertical_orientation'] = np.nan
header2['channel'] = get_channel_name(header['sampling_rate'], True, False,
False)
header2['standard']['units_type'] = get_units_type(header2['channel'])
header3['standard']['horizontal_orientation'] = 0.0
header3['standard']['vertical_orientation'] = np.nan
header3['channel'] = get_channel_name(header['sampling_rate'], True, True,
False)
header3['standard']['units_type'] = get_units_type(header3['channel'])
# three columns of NS, EW, UD
# data = np.genfromtxt(filename, skip_header=TEXT_HDR_ROWS,
# delimiter=[COLWIDTH] * NCOLS, encoding=ENCODING)
data = np.loadtxt(filename, skiprows=TEXT_HDR_ROWS, encoding=ENCODING)
data1 = data[:, 0]
data2 = data[:, 1]
data3 = data[:, 2]
trace1 = StationTrace(data=data1, header=header1)
response = {'input_units': 'counts', 'output_units': 'cm/s^2'}
trace1.setProvenance('remove_response', response)
trace2 = StationTrace(data=data2, header=header2)
trace2.setProvenance('remove_response', response)
trace3 = StationTrace(data=data3, header=header3)
trace3.setProvenance('remove_response', response)
stream = StationStream(traces=[trace1, trace2, trace3])
return [stream]
def read_nsmn(filename, config=None):
"""Read the Turkish NSMN strong motion data format.
Args:
filename (str):
path to NSMN data file.
config (dict):
Dictionary containing configuration.
Returns:
list: Sequence of one StationStream object containing 3 StationTrace
objects.
"""
header = _read_header(filename)
header1 = copy.deepcopy(header)
header2 = copy.deepcopy(header)
header3 = copy.deepcopy(header)
header1["standard"]["horizontal_orientation"] = 0.0
header1["standard"]["vertical_orientation"] = np.nan
header1["channel"] = get_channel_name(header["sampling_rate"], True, False,
True)
header1["standard"]["units_type"] = get_units_type(header1["channel"])
header2["standard"]["horizontal_orientation"] = 90.0
header2["standard"]["vertical_orientation"] = np.nan
header2["channel"] = get_channel_name(header["sampling_rate"], True, False,
False)
header2["standard"]["units_type"] = get_units_type(header2["channel"])
header3["standard"]["horizontal_orientation"] = 0.0
header3["standard"]["vertical_orientation"] = np.nan
header3["channel"] = get_channel_name(header["sampling_rate"], True, True,
False)
header3["standard"]["units_type"] = get_units_type(header3["channel"])
# three columns of NS, EW, UD
# data = np.genfromtxt(filename, skip_header=TEXT_HDR_ROWS,
# delimiter=[COLWIDTH] * NCOLS, encoding=ENCODING)
data = np.loadtxt(filename, skiprows=TEXT_HDR_ROWS, encoding=ENCODING)
data1 = data[:, 0]
data2 = data[:, 1]
data3 = data[:, 2]
trace1 = StationTrace(data=data1, header=header1)
response = {"input_units": "counts", "output_units": "cm/s^2"}
trace1.setProvenance("remove_response", response)
trace2 = StationTrace(data=data2, header=header2)
trace2.setProvenance("remove_response", response)
trace3 = StationTrace(data=data3, header=header3)
trace3.setProvenance("remove_response", response)
stream = StationStream(traces=[trace1, trace2, trace3])
return [stream]
def test_channel():
rate = 50
tchannel1 = get_channel_name(rate, is_acceleration=True,
is_vertical=False, is_north=True)
assert tchannel1 == 'BN1'
tchannel2 = get_channel_name(rate, is_acceleration=True,
is_vertical=False, is_north=False)
assert tchannel2 == 'BN2'
tchannel3 = get_channel_name(rate, is_acceleration=True,
is_vertical=True, is_north=False)
assert tchannel3 == 'BNZ'
rate = 100
tchannel4 = get_channel_name(rate, is_acceleration=True,
is_vertical=False, is_north=True)
assert tchannel4 == 'HN1'
tchannel5 = get_channel_name(rate, is_acceleration=True,
is_vertical=False, is_north=False)
assert tchannel5 == 'HN2'
tchannel6 = get_channel_name(rate, is_acceleration=True,
is_vertical=True, is_north=False)
assert tchannel6 == 'HNZ'
tchannel4 = get_channel_name(rate, is_acceleration=False,
is_vertical=False, is_north=True)
assert tchannel4 == 'HH1'
def _get_channel(angle, sampling_rate):
if angle == 500 or angle == 600 or (angle >= 0 and angle <= 360):
if angle == 500 or angle == 600:
channel = get_channel_name(
sampling_rate, is_acceleration=True, is_vertical=True, is_north=False
)
elif angle >= 315 or angle < 45 or (angle >= 135 and angle < 225):
channel = get_channel_name(
sampling_rate, is_acceleration=True, is_vertical=False, is_north=True
)
else:
channel = get_channel_name(
sampling_rate, is_acceleration=True, is_vertical=False, is_north=False
)
else:
errstr = (
"Not enough information to distinguish horizontal from "
"vertical channels."
)
raise BaseException("DMG: " + errstr)
return channel
def _get_channel(angle, sampling_rate):
if angle == 500 or angle == 600 or (angle >= 0 and angle <= 360):
if angle == 500 or angle == 600:
channel = get_channel_name(sampling_rate,
is_acceleration=True,
is_vertical=True,
is_north=False)
elif angle > 315 or angle < 45 or (angle > 135 and angle < 225):
channel = get_channel_name(sampling_rate,
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
channel = get_channel_name(sampling_rate,
is_acceleration=True,
is_vertical=False,
is_north=False)
else:
errstr = ('Not enough information to distinguish horizontal from '
'vertical channels.')
raise GMProcessException('DMG: ' + errstr)
return channel
def read_fdsn(filename):
"""Read Obspy data file (SAC, MiniSEED, etc).
Args:
filename (str):
Path to data file.
kwargs (ref):
Other arguments will be ignored.
Returns:
Stream: StationStream object.
"""
logging.debug("Starting read_fdsn.")
if not is_fdsn(filename):
raise Exception('%s is not a valid Obspy file format.' % filename)
streams = []
tstream = read(filename)
xmlfile = _get_station_file(filename, tstream)
inventory = read_inventory(xmlfile)
traces = []
for ttrace in tstream:
trace = StationTrace(data=ttrace.data,
header=ttrace.stats,
inventory=inventory)
location = ttrace.stats.location
trace.stats.channel = get_channel_name(
trace.stats.sampling_rate, trace.stats.channel[1] == 'N',
inventory.get_orientation(trace.id)['dip'] in [90, -90]
or trace.stats.channel[2] == 'Z',
is_channel_north(inventory.get_orientation(trace.id)['azimuth']))
if trace.stats.location == '':
trace.stats.location = '--'
network = ttrace.stats.network
if network in LOCATION_CODES:
codes = LOCATION_CODES[network]
if location in codes:
sdict = codes[location]
if sdict['free_field']:
trace.stats.standard.structure_type = 'free_field'
else:
trace.stats.standard.structure_type = sdict['description']
head, tail = os.path.split(filename)
trace.stats['standard']['source_file'] = tail or os.path.basename(head)
traces.append(trace)
stream = StationStream(traces=traces)
streams.append(stream)
return streams
def test_channel():
rate = 50
tchannel1 = get_channel_name(rate,
is_acceleration=True,
is_vertical=False,
is_north=True)
assert tchannel1 == 'BN1'
tchannel2 = get_channel_name(rate,
is_acceleration=True,
is_vertical=False,
is_north=False)
assert tchannel2 == 'BN2'
tchannel3 = get_channel_name(rate,
is_acceleration=True,
is_vertical=True,
is_north=False)
assert tchannel3 == 'BNZ'
rate = 100
tchannel4 = get_channel_name(rate,
is_acceleration=True,
is_vertical=False,
is_north=True)
assert tchannel4 == 'HN1'
tchannel5 = get_channel_name(rate,
is_acceleration=True,
is_vertical=False,
is_north=False)
assert tchannel5 == 'HN2'
tchannel6 = get_channel_name(rate,
is_acceleration=True,
is_vertical=True,
is_north=False)
assert tchannel6 == 'HNZ'
tchannel4 = get_channel_name(rate,
is_acceleration=False,
is_vertical=False,
is_north=True)
assert tchannel4 == 'HH1'
def test_channel():
rate = 50
tchannel1 = get_channel_name(
rate, is_acceleration=True, is_vertical=False, is_north=True
)
assert tchannel1 == "BN1"
tchannel2 = get_channel_name(
rate, is_acceleration=True, is_vertical=False, is_north=False
)
assert tchannel2 == "BN2"
tchannel3 = get_channel_name(
rate, is_acceleration=True, is_vertical=True, is_north=False
)
assert tchannel3 == "BNZ"
rate = 100
tchannel4 = get_channel_name(
rate, is_acceleration=True, is_vertical=False, is_north=True
)
assert tchannel4 == "HN1"
tchannel5 = get_channel_name(
rate, is_acceleration=True, is_vertical=False, is_north=False
)
assert tchannel5 == "HN2"
tchannel6 = get_channel_name(
rate, is_acceleration=True, is_vertical=True, is_north=False
)
assert tchannel6 == "HNZ"
tchannel4 = get_channel_name(
rate, is_acceleration=False, is_vertical=False, is_north=True
)
assert tchannel4 == "HH1"
def _get_header_info(filename, any_structure=False, accept_flagged=False, location=""):
"""Return stats structure from various headers.
Output is a dictionary like this:
- network
- station
- channel
- location (str): Set to floor the sensor is located on. If not a
multi-sensor array, default is '--'. Can be set manually by
the user.
- starttime
- sampling_rate
- npts
- coordinates:
- latitude
- longitude
- elevation
- standard
- horizontal_orientation
- instrument_period
- instrument_damping
- process_level
- station_name
- sensor_serial_number
- instrument
- comments
- structure_type
- corner_frequency
- units
- source
- source_format
- format_specific
- vertical_orientation
- building_floor (0=basement, 1=floor above basement,
-1=1st sub-basement, etc.
- bridge_number_spans
- bridge_transducer_location (
"free field",
"at the base of a pier or abutment",
"on an abutment",
"on the deck at the top of a pier"
"on the deck between piers or between an
abutment and a pier.")
dam_transducer_location (
"upstream or downstream free field",
"at the base of the dam",
"on the crest of the dam",
on the abutment of the dam")
construction_type (
"Reinforced concrete gravity",
"Reinforced concrete arch",
"earth fill",
"other")
filter_poles
data_source
"""
stats = {}
standard = {}
format_specific = {}
coordinates = {}
# read the ascii header lines
with open(filename) as f:
ascheader = [next(f).strip() for x in range(ASCII_HEADER_LINES)]
standard["process_level"] = PROCESS_LEVELS[VALID_HEADERS[ascheader[0]]]
logging.debug(f"process_level: {standard['process_level']}")
# station code is in the third line
stats["station"] = ""
if len(ascheader[2]) >= 4:
stats["station"] = ascheader[2][0:4].strip()
stats["station"] = stats["station"].strip("\x00")
logging.debug(f"station: {stats['station']}")
standard["process_time"] = ""
standard["station_name"] = ascheader[5][10:40].strip()
# sometimes the data source has nothing in it,
# most of the time it seems has has USGS in it
# sometimes it's something like JPL/USGS, CDOT/USGS, etc.
# if it's got USGS in it, let's just say network=US, otherwise "--"
stats["network"] = "--"
if ascheader[7].find("USGS") > -1:
stats["network"] = "US"
try:
standard["source"] = ascheader[7].split("=")[2].strip()
except IndexError:
standard["source"] = "USGS"
if standard["source"] == "":
standard["source"] = "USGS"
standard["source_format"] = "smc"
# read integer header data
intheader = np.genfromtxt(
filename,
dtype=np.int32,
max_rows=INTEGER_HEADER_LINES,
skip_header=ASCII_HEADER_LINES,
delimiter=INT_HEADER_WIDTHS,
)
# 8 columns per line
# first line is start time information, and then inst. serial number
missing_data = intheader[0, 0]
year = intheader[0, 1]
# sometimes the year field has a 0 in it. When this happens, we
# can try to get a timestamp from line 4 of the ascii header.
if year == 0:
parts = ascheader[3].split()
try:
year = int(parts[0])
except ValueError as ve:
fmt = (
"Could not find year in SMC file %s. Not present "
"in integer header and not parseable from line "
'4 of ASCII header. Error: "%s"'
)
raise ValueError(fmt % (filename, str(ve)))
jday = intheader[0, 2]
hour = intheader[0, 3]
minute = intheader[0, 4]
if (
year != missing_data
and jday != missing_data
and hour != missing_data
and minute != missing_data
):
# Handle second if missing
second = 0
if not intheader[0, 5] == missing_data:
second = intheader[0, 5]
# Handle microsecond if missing and convert milliseconds to
# microseconds
microsecond = 0
if not intheader[0, 6] == missing_data:
microsecond = intheader[0, 6] / 1e3
datestr = "%i %00i %i %i %i %i" % (
year,
jday,
hour,
minute,
second,
microsecond,
)
stats["starttime"] = datetime.strptime(datestr, "%Y %j %H %M %S %f")
else:
logging.warning(
"No start time provided. "
"This must be set manually for network/station: "
"%s/%s." % (stats["network"], stats["station"])
)
standard["comments"] = "Missing start time."
standard["sensor_serial_number"] = ""
if intheader[1, 3] != missing_data:
standard["sensor_serial_number"] = str(intheader[1, 3])
# we never get a two character location code so floor location is used
if location == "":
location = intheader.flatten()[24]
if location != missing_data:
location = str(location)
if len(location) < 2:
location = location.zfill(2)
stats["location"] = location
else:
stats["location"] = "--"
else:
stats["location"] = location
# second line is information about number of channels, orientations
# we care about orientations
format_specific["vertical_orientation"] = np.nan
if intheader[1, 4] != missing_data:
format_specific["vertical_orientation"] = int(intheader[1, 4])
standard["horizontal_orientation"] = np.nan
standard["vertical_orientation"] = np.nan
if intheader[1, 5] != missing_data:
standard["horizontal_orientation"] = float(intheader[1, 5])
if intheader[1, 6] == missing_data or intheader[1, 6] not in INSTRUMENTS:
standard["instrument"] = ""
else:
standard["instrument"] = INSTRUMENTS[intheader[1, 6]]
num_comments = intheader[1, 7]
# third line contains number of data points
stats["npts"] = intheader[2, 0]
problem_flag = intheader[2, 1]
if problem_flag == 1:
if not accept_flagged:
fmt = "SMC: Record found in file %s has a problem flag!"
raise BaseException(fmt % filename)
else:
logging.warning(
"SMC: Data contains a problem flag for network/station: "
"%s/%s. See comments." % (stats["network"], stats["station"])
)
stype = intheader[2, 2]
if stype == missing_data:
stype = np.nan
elif stype not in STRUCTURES:
# structure type is not defined and should will be considered 'other'
stype = 4
fmt = "SMC: Record found in file %s is not a free-field sensor!"
standard["structure_type"] = STRUCTURES[stype]
if standard["structure_type"] == "building" and not any_structure:
raise Exception(fmt % filename)
format_specific["building_floor"] = np.nan
if intheader[3, 0] != missing_data:
format_specific["building_floor"] = intheader[3, 0]
format_specific["bridge_number_spans"] = np.nan
if intheader[3, 1] != missing_data:
format_specific["bridge_number_spans"] = intheader[3, 1]
format_specific["bridge_transducer_location"] = BRIDGE_LOCATIONS[0]
if intheader[3, 2] != missing_data:
bridge_number = intheader[3, 2]
format_specific["bridge_transducer_location"] = BRIDGE_LOCATIONS[bridge_number]
format_specific["dam_transducer_location"] = DAM_LOCATIONS[0]
if intheader[3, 3] != missing_data:
dam_number = intheader[3, 3]
format_specific["dam_transducer_location"] = DAM_LOCATIONS[dam_number]
c1 = format_specific["bridge_transducer_location"].find("free field") == -1
c2 = format_specific["dam_transducer_location"].find("free field") == -1
if (c1 or c2) and not any_structure:
raise Exception(fmt % filename)
format_specific["construction_type"] = CONSTRUCTION_TYPES[4]
if intheader[3, 4] != missing_data:
format_specific["construction_type"] = CONSTRUCTION_TYPES[intheader[3, 4]]
# station is repeated here if all numeric
if not len(stats["station"]):
stats["station"] = "%i" % intheader[3, 5]
# read float header data
skip = ASCII_HEADER_LINES + INTEGER_HEADER_LINES
floatheader = np.genfromtxt(
filename,
max_rows=FLOAT_HEADER_LINES,
skip_header=skip,
delimiter=FLOAT_HEADER_WIDTHS,
)
# float headers are 10 lines of 5 floats each
missing_data = floatheader[0, 0]
stats["sampling_rate"] = floatheader[0, 1]
if stats["sampling_rate"] >= MAX_ALLOWED_SAMPLE_RATE:
fmt = "Sampling rate of %.2g samples/second is nonsensical."
raise Exception(fmt % stats["sampling_rate"])
coordinates["latitude"] = floatheader[2, 0]
# the documentation for SMC says that sometimes longitudes are
# positive in the western hemisphere. Since it is very unlikely
# any of these files exist for the eastern hemisphere, check for
# positive longitudes and fix them.
lon = floatheader[2, 1]
if lon > 0:
lon = -1 * lon
coordinates["longitude"] = lon
coordinates["elevation"] = 0.0
if floatheader[2, 2] != missing_data:
coordinates["elevation"] = floatheader[2, 2]
else:
logging.warning("Setting elevation to 0.0")
# figure out the channel code
if format_specific["vertical_orientation"] in [0, 180]:
stats["channel"] = get_channel_name(
stats["sampling_rate"],
is_acceleration=True,
is_vertical=True,
is_north=False,
)
else:
ho = standard["horizontal_orientation"]
quad1 = ho > 315 and ho <= 360
quad2 = ho > 0 and ho <= 45
quad3 = ho > 135 and ho <= 225
if quad1 or quad2 or quad3:
stats["channel"] = get_channel_name(
stats["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=True,
)
else:
stats["channel"] = get_channel_name(
stats["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=False,
)
logging.debug(f"channel: {stats['channel']}")
sensor_frequency = floatheader[4, 1]
standard["instrument_period"] = 1 / sensor_frequency
standard["instrument_damping"] = floatheader[4, 2]
standard["corner_frequency"] = floatheader[3, 4]
format_specific["filter_poles"] = floatheader[4, 0]
standard["units"] = "acc"
standard["units_type"] = get_units_type(stats["channel"])
# this field can be used for instrument correction
# when data is in counts
standard["instrument_sensitivity"] = np.nan
# read in the comment lines
with open(filename) as f:
skip = ASCII_HEADER_LINES + INTEGER_HEADER_LINES + FLOAT_HEADER_LINES
_ = [next(f) for x in range(skip)]
standard["comments"] = [
next(f).strip().lstrip("|") for x in range(num_comments)
]
standard["comments"] = " ".join(standard["comments"])
stats["coordinates"] = coordinates
stats["standard"] = standard
stats["format_specific"] = format_specific
head, tail = os.path.split(filename)
stats["standard"]["source_file"] = tail or os.path.basename(head)
return (stats, num_comments)
def _read_header(lines, filename, table):
header = {}
standard = {}
coords = {}
format_specific = {}
# fill out the standard dictionary
standard["source"] = SOURCE
standard["source_format"] = SOURCE_FORMAT
standard["instrument"] = ""
standard["sensor_serial_number"] = ""
standard["process_level"] = PROCESS_LEVELS["V1"]
standard["process_time"] = lines[0].split(":")[1].strip()
# station name line can look like this:
# VI�A DEL MAR CENTRO S/N 675
sparts = lines[5].split()
station_name = " ".join(sparts[0:sparts.index("S/N")])
standard["station_name"] = station_name
# this table gives us station coordinates and structure type
station_row = table[table["Name"] == station_name]
if not len(station_row):
logging.warning("Unknown structure type.")
standard["structure_type"] = ""
else:
row = station_row.iloc[0]
standard["structure_type"] = row["Structure Type"]
standard["corner_frequency"] = np.nan
standard["units"] = "cm/s^2"
standard["units_type"] = "acc"
inst_dict = {}
for part in lines[9].split(","):
key, value = part.split("=")
fvalue_str = re.search(FLOATRE, value.strip()).group()
inst_dict[key.strip()] = float(fvalue_str)
standard["instrument_period"] = inst_dict["INSTR PERIOD"]
standard["instrument_damping"] = inst_dict["DAMPING"]
standard["horizontal_orientation"] = np.nan
standard["vertical_orientation"] = np.nan
standard["comments"] = " ".join(lines[11:13]).replace("\n", "")
head, tail = os.path.split(filename)
standard["source_file"] = tail or os.path.basename(head)
# this field can be used for instrument correction
# when data is in counts
standard["instrument_sensitivity"] = inst_dict["SENSITIVITY"]
standard["volts_to_counts"] = np.nan
# fill out the stats stuff
try:
stimestr = re.search(TIME_RE, lines[3]).group()
except AttributeError:
try:
stimestr = re.search(TIME_RE2, lines[3]).group()
except AttributeError:
logging.warning("Setting time to epoch.")
stimestr = "01/01/1970 00:00:00.000"
# 2/27/2010 2:45:46.000 GMT
stime = datetime.strptime(stimestr, TIMEFMT)
# it appears that sometimes the trigger time is set to Jan 1, 1980
# by default.
if stime.year == 1980 and stime.month == 1 and stime.day == 1:
fmt = "Trigger time set to %s in file %s"
logging.warning(fmt % (str(stime), standard["source_file"]))
header["starttime"] = stime
npts, duration = re.findall(FLOATRE, lines[10])
npts = int(npts)
duration = float(duration)
header["npts"] = npts
header["delta"] = duration / (npts - 1)
header["sampling_rate"] = (npts - 1) / duration
header["duration"] = duration
raw_channel = lines[6][9:11].strip()
if raw_channel in NORTH_CHANNELS:
channel = get_channel_name(header["sampling_rate"], True, False, True)
elif raw_channel in WEST_CHANNELS:
channel = get_channel_name(header["sampling_rate"], True, False, False)
elif raw_channel in VERTICAL_CHANNELS:
channel = get_channel_name(header["sampling_rate"], True, True, False)
else:
raise KeyError(f"Channel name {raw_channel} not defined")
header["channel"] = channel
header["station"] = lines[5].split()[-1]
header["location"] = "--"
header["network"] = NETWORK
# these files seem to have all zeros for station coordinates!
if not len(station_row):
logging.warning(f"Could not find station match for {station_name}")
coordparts = lines[4].split()
lat = float(re.search(FLOATRE, coordparts[2]).group())
lon = float(re.search(FLOATRE, coordparts[3]).group())
if lon == 0 or lat == 0:
logging.warning("Latitude or Longitude values are 0")
if "S" in coordparts[2]:
lat = -1 * lat
if "W" in coordparts[3]:
lon = -1 * lon
else:
row = station_row.iloc[0]
lat = row["Lat"]
lon = row["Lon"]
altitude = 0.0
logging.warning("Setting elevation to 0.0")
coords = {"latitude": lat, "longitude": lon, "elevation": altitude}
header["coordinates"] = coords
header["standard"] = standard
header["format_specific"] = format_specific
return header
def _read_header(hdr_data, station, name, component, data_format,
instrument, resolution):
"""Construct stats dictionary from header lines.
Args:
hdr_data (ndarray): (10,10) numpy array containing header data.
station (str): Station code obtained from previous text portion of
header.
location (str): Location string obtained from previous text portion
of header.
component (str): Component direction (N18E, S72W, etc.)
Returns:
Dictionary containing fields:
- network "NZ"
- station
- channel H1,H2,or Z.
- location
- sampling_rate Samples per second.
- delta Interval between samples (seconds)
- calib Calibration factor (always 1.0)
- npts Number of samples in record.
- starttime Datetime object containing start of record.
- standard:
- station_name
- units "acc"
- source 'New Zealand Institute of Geological and Nuclear
Science'
- horizontal_orientation
- instrument_period
- instrument_damping
- processing_time
- process_level
- sensor_serial_number
- instrument
- comments
- structure_type
- corner_frequency
- source_format
- coordinates:
- lat Latitude of station.
- lon Longitude of station.
- elevation Elevation of station.
- format_specific:
- sensor_bit_resolution
"""
hdr = {}
standard = {}
coordinates = {}
format_specific = {}
hdr['station'] = station
standard['station_name'] = name
if data_format == 'V1':
hdr['sampling_rate'] = hdr_data[4, 0]
sampling_rate = hdr['sampling_rate']
hdr['delta'] = 1 / hdr['sampling_rate']
else:
hdr['delta'] = hdr_data[6, 5]
hdr['sampling_rate'] = 1 / hdr['delta']
# V2 files have been resampled, we need sensor rate for
# channel naming.
sampling_rate = 1 / hdr_data[6, 4]
hdr['calib'] = 1.0
if data_format == 'V1':
hdr['npts'] = int(hdr_data[3, 0])
else:
hdr['npts'] = int(hdr_data[3, 3])
hdr['network'] = 'NZ'
standard['units'] = 'acc'
standard['source'] = ('New Zealand Institute of Geological and '
'Nuclear Science')
logging.debug('component: %s' % component)
if component.lower() in ['up', 'down']:
standard['horizontal_orientation'] = np.nan
hdr['channel'] = get_channel_name(
sampling_rate,
is_acceleration=True,
is_vertical=True,
is_north=False)
else:
angle = _get_channel(component)
logging.debug('angle: %s' % angle)
standard['horizontal_orientation'] = float(angle)
if (angle > 315 or angle < 45) or (angle > 135 and angle < 225):
hdr['channel'] = get_channel_name(
sampling_rate,
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
hdr['channel'] = get_channel_name(
sampling_rate,
is_acceleration=True,
is_vertical=False,
is_north=False)
logging.debug('channel: %s' % hdr['channel'])
hdr['location'] = '--'
# figure out the start time
milliseconds = hdr_data[3, 9]
seconds = int(milliseconds / 1000)
microseconds = int(np.round(milliseconds / 1000.0 - seconds))
year = int(hdr_data[0, 8])
month = int(hdr_data[0, 9])
day = int(hdr_data[1, 8])
hour = int(hdr_data[1, 9])
minute = int(hdr_data[3, 8])
hdr['starttime'] = datetime(
year, month, day, hour, minute, seconds, microseconds)
# figure out station coordinates
latdg = hdr_data[2, 0]
latmn = hdr_data[2, 1]
latsc = hdr_data[2, 2]
coordinates['latitude'] = _dms_to_dd(latdg, latmn, latsc) * -1
londg = hdr_data[2, 3]
lonmn = hdr_data[2, 4]
lonsc = hdr_data[2, 5]
coordinates['longitude'] = _dms_to_dd(londg, lonmn, lonsc)
coordinates['elevation'] = 0.0
# get other standard metadata
standard['instrument_period'] = 1 / hdr_data[4, 0]
standard['instrument_damping'] = hdr_data[4, 1]
standard['process_time'] = ''
standard['process_level'] = PROCESS_LEVELS[data_format]
logging.debug("process_level: %s" % data_format)
standard['sensor_serial_number'] = ''
standard['instrument'] = instrument
standard['comments'] = ''
standard['structure_type'] = ''
standard['corner_frequency'] = np.nan
standard['source_format'] = 'geonet'
# get format specific metadata
format_specific['sensor_bit_resolution'] = resolution
hdr['coordinates'] = coordinates
hdr['standard'] = standard
hdr['format_specific'] = format_specific
return hdr
def _read_header_lines(filename, offset):
"""Read the header lines for each channel.
Args:
filename (str):
Input BHRC file name.
offset (int):
Number of lines to skip from the beginning of the file.
Returns:
tuple: (header dictionary containing Stats dictionary with
extra sub-dicts, updated offset rows)
"""
with open(filename, "rt", encoding="utf-8") as f:
for _ in range(offset):
next(f)
lines = [next(f) for x in range(TEXT_HDR_ROWS)]
offset += TEXT_HDR_ROWS
header = {}
standard = {}
coords = {}
format_specific = {}
# get the sensor azimuth with respect to the earthquake
# this data has been rotated so that the longitudinal channel (L)
# is oriented at the sensor azimuth, and the transverse (T) is
# 90 degrees off from that.
station_info = lines[7][lines[7].index("Station"):]
float_strings = re.findall(FLOATRE, station_info)
(lat_str, lon_str, alt_str, lstr, tstr) = float_strings[0:5]
component = lines[4].strip()
if component == "V":
angle = np.nan
elif component == "L":
angle = float(lstr)
else:
angle = float(tstr)
coords = {
"latitude": float(lat_str),
"longitude": float(lon_str),
"elevation": float(alt_str),
}
# fill out the standard dictionary
standard["source"] = SOURCE
standard["source_format"] = SOURCE_FORMAT
standard["instrument"] = lines[1].split("=")[1].strip()
standard["sensor_serial_number"] = ""
volstr = lines[0].split()[1].strip()
if volstr not in LEVELS:
raise KeyError(f"Volume {volstr} files are not supported.")
standard["process_level"] = PROCESS_LEVELS[LEVELS[volstr]]
standard["process_time"] = ""
station_name = lines[7][0:lines[7].index("Station")].strip()
standard["station_name"] = station_name
standard["structure_type"] = ""
standard["corner_frequency"] = np.nan
standard["units"] = "acc"
period_str, damping_str = re.findall(FLOATRE, lines[9])
standard["instrument_period"] = float(period_str)
if standard["instrument_period"] == 0:
standard["instrument_period"] = np.nan
standard["instrument_damping"] = float(damping_str)
standard["horizontal_orientation"] = angle
standard["vertical_orientation"] = np.nan
standard["comments"] = ""
head, tail = os.path.split(filename)
standard["source_file"] = tail or os.path.basename(head)
# this field can be used for instrument correction
# when data is in counts
standard["instrument_sensitivity"] = np.nan
# fill out the stats stuff
# we don't know the start of the trace
header["starttime"] = UTCDateTime(1970, 1, 1)
npts_str, dur_str = re.findall(FLOATRE, lines[10])
header["npts"] = int(npts_str)
header["duration"] = float(dur_str)
header["delta"] = header["duration"] / (header["npts"] - 1)
header["sampling_rate"] = 1 / header["delta"]
if np.isnan(angle):
header["channel"] = get_channel_name(
header["sampling_rate"],
is_acceleration=True,
is_vertical=True,
is_north=False,
)
elif (angle > 315 or angle < 45) or (angle > 135 and angle < 225):
header["channel"] = get_channel_name(
header["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=True,
)
else:
header["channel"] = get_channel_name(
header["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=False,
)
standard["units_type"] = get_units_type(header["channel"])
part1 = lines[0].split(":")[1]
stationcode = part1.split("/")[0].strip()
header["station"] = stationcode
header["location"] = "--"
header["network"] = NETWORK
header["coordinates"] = coords
header["standard"] = standard
header["format_specific"] = format_specific
offset += INT_HDR_ROWS
offset += FLOAT_HDR_ROWS
return (header, offset)
def _read_header(hdr_data, station, name, component, data_format,
instrument, resolution):
"""Construct stats dictionary from header lines.
Args:
hdr_data (ndarray):
(10,10) numpy array containing header data.
station (str):
Station code obtained from previous text portion of header.
location (str):
Location string obtained from previous text portion of header.
component (str):
Component direction (N18E, S72W, etc.)
Returns:
Dictionary containing fields:
- network "NZ"
- station
- channel H1,H2,or Z.
- location
- sampling_rate Samples per second.
- delta Interval between samples (seconds)
- calib Calibration factor (always 1.0)
- npts Number of samples in record.
- starttime Datetime object containing start of record.
- standard:
- station_name
- units "acc"
- source 'New Zealand Institute of Geological and Nuclear
Science'
- horizontal_orientation
- instrument_period
- instrument_damping
- processing_time
- process_level
- sensor_serial_number
- instrument
- comments
- structure_type
- corner_frequency
- source_format
- coordinates:
- lat Latitude of station.
- lon Longitude of station.
- elevation Elevation of station.
- format_specific:
- sensor_bit_resolution
"""
hdr = {}
standard = {}
coordinates = {}
format_specific = {}
hdr['station'] = station
standard['station_name'] = name
# Note: Original sample interval (s): hdr_data[6, 4]
# Sample inverval (s)
hdr['delta'] = hdr_data[6, 5]
hdr['sampling_rate'] = 1 / hdr['delta']
hdr['calib'] = 1.0
if data_format == 'V1':
hdr['npts'] = int(hdr_data[3, 0])
else:
hdr['npts'] = int(hdr_data[3, 3])
hdr['network'] = 'NZ'
standard['units'] = 'acc'
standard['source'] = ('New Zealand Institute of Geological and '
'Nuclear Science')
logging.debug('component: %s' % component)
standard['vertical_orientation'] = np.nan
if component.lower() in ['up', 'down']:
standard['horizontal_orientation'] = np.nan
hdr['channel'] = get_channel_name(
hdr['delta'],
is_acceleration=True,
is_vertical=True,
is_north=False)
else:
angle = _get_channel(component)
logging.debug('angle: %s' % angle)
standard['horizontal_orientation'] = float(angle)
if (angle > 315 or angle < 45) or (angle > 135 and angle < 225):
hdr['channel'] = get_channel_name(
hdr['delta'],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
hdr['channel'] = get_channel_name(
hdr['delta'],
is_acceleration=True,
is_vertical=False,
is_north=False)
logging.debug('channel: %s' % hdr['channel'])
hdr['location'] = '--'
# figure out the start time
milliseconds = hdr_data[3, 9]
seconds = int(milliseconds / 1000)
microseconds = int(np.round(milliseconds / 1000.0 - seconds))
year = int(hdr_data[0, 8])
month = int(hdr_data[0, 9])
day = int(hdr_data[1, 8])
hour = int(hdr_data[1, 9])
minute = int(hdr_data[3, 8])
hdr['starttime'] = datetime(
year, month, day, hour, minute, seconds, microseconds)
# figure out station coordinates
latdg = hdr_data[2, 0]
latmn = hdr_data[2, 1]
latsc = hdr_data[2, 2]
coordinates['latitude'] = _dms_to_dd(latdg, latmn, latsc) * -1
londg = hdr_data[2, 3]
lonmn = hdr_data[2, 4]
lonsc = hdr_data[2, 5]
coordinates['longitude'] = _dms_to_dd(londg, lonmn, lonsc)
logging.warning('Setting elevation to 0.0')
coordinates['elevation'] = 0.0
# get other standard metadata
standard['units_type'] = get_units_type(hdr['channel'])
standard['instrument_period'] = 1 / hdr_data[4, 0]
standard['instrument_damping'] = hdr_data[4, 1]
standard['process_time'] = ''
standard['process_level'] = PROCESS_LEVELS[data_format]
logging.debug("process_level: %s" % data_format)
standard['sensor_serial_number'] = ''
standard['instrument'] = instrument
standard['comments'] = ''
standard['structure_type'] = ''
standard['corner_frequency'] = np.nan
standard['source_format'] = 'geonet'
# this field can be used for instrument correction
# when data is in counts
standard['instrument_sensitivity'] = np.nan
# get format specific metadata
format_specific['sensor_bit_resolution'] = resolution
hdr['coordinates'] = coordinates
hdr['standard'] = standard
hdr['format_specific'] = format_specific
return hdr
def _read_header_lines(filename, offset):
"""Read the header lines for each channel.
Args:
filename (str):
Input BHRC file name.
offset (int):
Number of lines to skip from the beginning of the file.
Returns:
tuple: (header dictionary containing Stats dictionary with extra sub-dicts,
updated offset rows)
"""
with open(filename, 'rt') as f:
for _ in range(offset):
next(f)
lines = [next(f) for x in range(TEXT_HDR_ROWS)]
offset += TEXT_HDR_ROWS
header = {}
standard = {}
coords = {}
format_specific = {}
# get the sensor azimuth with respect to the earthquake
# this data has been rotated so that the longitudinal channel (L)
# is oriented at the sensor azimuth, and the transverse (T) is
# 90 degrees off from that.
station_info = lines[7][lines[7].index('Station'):]
(lat_str, lon_str, alt_str, lstr, tstr) = re.findall(FLOATRE, station_info)
component = lines[4].strip()
if component == 'V':
angle = np.nan
elif component == 'L':
angle = float(lstr)
else:
angle = float(tstr)
coords = {
'latitude': float(lat_str),
'longitude': float(lon_str),
'elevation': float(alt_str)
}
# fill out the standard dictionary
standard['source'] = SOURCE
standard['source_format'] = SOURCE_FORMAT
standard['instrument'] = lines[1].split('=')[1].strip()
standard['sensor_serial_number'] = ''
volstr = lines[0].split()[1].strip()
if volstr not in LEVELS:
raise KeyError('Volume %s files are not supported.' % volstr)
standard['process_level'] = PROCESS_LEVELS[LEVELS[volstr]]
standard['process_time'] = ''
station_name = lines[7][0:lines[7].index('Station')].strip()
standard['station_name'] = station_name
standard['structure_type'] = ''
standard['corner_frequency'] = np.nan
standard['units'] = 'acc'
period_str, damping_str = re.findall(FLOATRE, lines[9])
standard['instrument_period'] = float(period_str)
standard['instrument_damping'] = float(damping_str)
standard['horizontal_orientation'] = angle
standard['comments'] = ''
# fill out the stats stuff
# we don't know the start of the trace
header['starttime'] = UTCDateTime(1970, 1, 1)
npts_str, dur_str = re.findall(FLOATRE, lines[10])
header['npts'] = int(npts_str)
header['duration'] = float(dur_str)
header['delta'] = header['duration'] / (header['npts'] - 1)
header['sampling_rate'] = 1 / header['delta']
if np.isnan(angle):
header['channel'] = get_channel_name(header['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
elif (angle > 315 or angle < 45) or (angle > 135 and angle < 225):
header['channel'] = get_channel_name(header['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
header['channel'] = get_channel_name(header['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
part1 = lines[0].split(':')[1]
stationcode = part1.split('/')[0].strip()
header['station'] = stationcode
header['location'] = '--'
header['network'] = NETWORK
header['coordinates'] = coords
header['standard'] = standard
header['format_specific'] = format_specific
offset += INT_HDR_ROWS
offset += FLOAT_HDR_ROWS
return (header, offset)
def _get_header_info(int_data, flt_data, lines, cmt_data, location=''):
"""Return stats structure from various headers.
Output is a dictionary like this:
- network (str): Default is '--'. Determined using COSMOS_NETWORKS
- station (str)
- channel (str): Determined using COSMOS_ORIENTATIONS
- location (str): Set to location index of sensor site at station.
If not a multi-site array, default is '--'.
- starttime (datetime)
- duration (float)
- sampling_rate (float)
- delta (float)
- npts (int)
- coordinates:
- latitude (float)
- longitude (float)
- elevation (float)
- standard (Defaults are either np.nan or '')
- horizontal_orientation (float): Rotation from north (degrees)
- instrument_period (float): Period of sensor (Hz)
- instrument_damping (float): Fraction of critical
- process_time (datetime): Reported date of processing
- process_level: Either 'V0', 'V1', 'V2', or 'V3'
- station_name (str): Long form station description
- sensor_serial_number (str): Reported sensor serial
- instrument (str): See SENSOR_TYPES
- comments (str): Processing comments
- structure_type (str): See BUILDING_TYPES
- corner_frequency (float): Sensor corner frequency (Hz)
- units (str): See UNITS
- source (str): Network source description
- source_format (str): Always cosmos
- format_specific
- physical_units (str): See PHYSICAL_UNITS
- v30 (float): Site geology V30 (km/s)
- least_significant_bit: Recorder LSB in micro-volts (uv/count)
- low_filter_type (str): Filter used for low frequency
V2 filtering (see FILTERS)
- low_filter_corner (float): Filter corner for low frequency
V2 filtering (Hz)
- low_filter_decay (float): Filter decay for low frequency
V2 filtering (dB/octabe)
- high_filter_type (str): Filter used for high frequency
V2 filtering (see FILTERS)
- high_filter_corner (float): Filter corner for high frequency
V2 filtering (Hz)
- high_filter_decay (float): Filter decay for high frequency
V2 filtering (dB/octabe)
- maximum (float): Maximum value
- maximum_time (float): Time at which maximum occurs
- station_code (int): Code for structure_type
- record_flag (str): Either 'No problem', 'Fixed', 'Unfixed problem'.
Should be described in more depth in comments.
- scaling_factor (float): Scaling used for converting acceleration
from g/10 to cm/s/s
- sensor_sensitivity (float): Sensitvity in volts/g
Args:
int_data (ndarray): Array of integer data
flt_data (ndarray): Array of float data
lines (list): List of text headers (str)
cmt_data (ndarray): Array of comments (str)
Returns:
dictionary: Dictionary of header/metadata information
"""
hdr = {}
coordinates = {}
standard = {}
format_specific = {}
# Get unknown parameter number
try:
unknown = int(lines[12][64:71])
except ValueError:
unknown = -999
# required metadata
network_num = int(int_data[10])
# Get network from cosmos table or fdsn code sheet
if network_num in COSMOS_NETWORKS:
network = COSMOS_NETWORKS[network_num][0]
source = COSMOS_NETWORKS[network_num][1]
if network == '':
network = COSMOS_NETWORKS[network_num][2]
else:
network_code = lines[4][25:27].upper()
if network_code in CODES:
network = network_code
idx = np.argwhere(CODES == network_code)[0][0]
source = SOURCES1[idx].decode(
'utf-8') + ', ' + SOURCES2[idx].decode('utf-8')
else:
network = 'ZZ'
source = ''
hdr['network'] = network
logging.debug('network: %s' % network)
hdr['station'] = lines[4][28:34].strip()
logging.debug('station: %s' % hdr['station'])
horizontal_angle = int(int_data[53])
logging.debug('horizontal_angle: %s' % horizontal_angle)
if horizontal_angle not in VALID_AZIMUTH_INTS:
logging.warning("Horizontal_angle in COSMOS header is not valid.")
horizontal_angle = float(horizontal_angle)
# Store delta and duration. Use them to calculate npts and sampling_rate
# NOTE: flt_data[33] is the delta of the V0 format, and if we are reading
# a V1 or V2 format then it may have been resampled. We should consider
# adding flt_data[33] delta to the provenance record at some point.
delta = float(flt_data[61]) * MSEC_TO_SEC
if delta != unknown:
hdr['delta'] = delta
hdr['sampling_rate'] = 1 / delta
# Determine the angle based upon the cosmos table
# Set horizontal angles other than N,S,E,W to H1 and H2
# Missing angle results in the channel number
if horizontal_angle != unknown:
if horizontal_angle in COSMOS_ORIENTATIONS:
channel = COSMOS_ORIENTATIONS[horizontal_angle][1].upper()
if channel == 'UP' or channel == 'DOWN' or channel == 'VERT':
channel = get_channel_name(hdr['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
elif horizontal_angle >= 0 and horizontal_angle <= 360:
if (horizontal_angle > 315 or horizontal_angle < 45
or (horizontal_angle > 135 and horizontal_angle < 225)):
channel = get_channel_name(hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
channel = get_channel_name(hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
horizontal_orientation = horizontal_angle
else:
errstr = ('Not enough information to distinguish horizontal from '
'vertical channels.')
raise GMProcessException('COSMOS: ' + errstr)
hdr['channel'] = channel
logging.debug('channel: %s' % hdr['channel'])
if location == '':
location = int(int_data[55])
location = str(_check_assign(location, unknown, '--'))
if len(location) < 2:
location = location.zfill(2)
hdr['location'] = location
else:
hdr['location'] = location
year = int(int_data[39])
month = int(int_data[41])
day = int(int_data[42])
hour = int(int_data[43])
minute = int(int_data[44])
second = float(flt_data[29])
# If anything more than seconds is excluded
# It is considered inadequate time information
if second == unknown:
try:
hdr['starttime'] = datetime(year, month, day, hour, minute)
except Exception:
raise GMProcessException(
'COSMOS: Inadequate start time information.')
else:
second = second
microsecond = int((second - int(second)) * 1e6)
try:
hdr['starttime'] = datetime(year, month, day, hour, minute,
int(second), microsecond)
except Exception:
raise GMProcessException(
'COSMOS: Inadequate start time information.')
if flt_data[62] != unknown:
# COSMOS **defines** "length" as npts*dt (note this is a bit unusual)
cosmos_length = flt_data[62]
npts = int(cosmos_length / delta)
hdr['duration'] = (npts - 1) * delta
hdr['npts'] = npts
else:
raise ValueError('COSMOS file does not specify length.')
# coordinate information
coordinates['latitude'] = float(flt_data[0])
coordinates['longitude'] = float(flt_data[1])
coordinates['elevation'] = float(flt_data[2])
for key in coordinates:
if coordinates[key] == unknown:
warnings.warn('Missing %r. Setting to np.nan.' % key, Warning)
coordinates[key] = np.nan
hdr['coordinates'] = coordinates
# standard metadata
standard['source'] = source
standard['horizontal_orientation'] = horizontal_orientation
station_name = lines[4][40:-1].strip()
standard['station_name'] = station_name
instrument_frequency = float(flt_data[39])
standard['instrument_period'] = 1.0 / _check_assign(
instrument_frequency, unknown, np.nan)
instrument_damping = float(flt_data[40])
standard['instrument_damping'] = _check_assign(instrument_damping, unknown,
np.nan)
process_line = lines[10][10:40]
if process_line.find('-') >= 0 or process_line.find('/') >= 0:
if process_line.find('-') >= 0:
delimeter = '-'
elif process_line.find('/') >= 0:
delimeter = '/'
try:
date = process_line.split(delimeter)
month = int(date[0][-2:])
day = int(date[1])
year = int(date[2][:4])
time = process_line.split(':')
hour = int(time[0][-2:])
minute = int(time[1])
second = float(time[2][:2])
microsecond = int((second - int(second)) * 1e6)
etime = datetime(year, month, day, hour, minute, int(second),
microsecond)
standard['process_time'] = etime.strftime(TIMEFMT)
except Exception:
standard['process_time'] = ''
else:
standard['process_time'] = ''
process_level = int(int_data[0])
if process_level == 0:
standard['process_level'] = PROCESS_LEVELS['V0']
elif process_level == 1:
standard['process_level'] = PROCESS_LEVELS['V1']
elif process_level == 2:
standard['process_level'] = PROCESS_LEVELS['V2']
elif process_level == 3:
standard['process_level'] = PROCESS_LEVELS['V3']
else:
standard['process_level'] = PROCESS_LEVELS['V1']
logging.debug("process_level: %s" % process_level)
serial = int(int_data[52])
if serial != unknown:
standard['sensor_serial_number'] = str(
_check_assign(serial, unknown, ''))
else:
standard['sensor_serial_number'] = ''
instrument = int(int_data[51])
if instrument != unknown and instrument in SENSOR_TYPES:
standard['instrument'] = SENSOR_TYPES[instrument]
else:
standard['instrument'] = lines[6][57:-1].strip()
structure_type = int(int_data[18])
if structure_type != unknown and structure_type in BUILDING_TYPES:
standard['structure_type'] = BUILDING_TYPES[structure_type]
else:
standard['structure_type'] = ''
frequency = float(flt_data[25])
standard['corner_frequency'] = _check_assign(frequency, unknown, np.nan)
physical_parameter = int(int_data[2])
units = int(int_data[1])
if units != unknown and units in UNITS:
standard['units'] = UNITS[units]
else:
if physical_parameter in [2, 4, 7, 10, 11, 12, 23]:
standard['units'] = 'acc'
elif physical_parameter in [5, 8, 24]:
standard['units'] = 'vel'
elif physical_parameter in [6, 9, 25]:
standard['units'] = 'disp'
standard['source_format'] = 'cosmos'
standard['comments'] = ', '.join(cmt_data)
# format specific metadata
if physical_parameter in PHYSICAL_UNITS:
physical_parameter = PHYSICAL_UNITS[physical_parameter][0]
format_specific['physical_units'] = physical_parameter
v30 = float(flt_data[3])
format_specific['v30'] = _check_assign(v30, unknown, np.nan)
least_significant_bit = float(flt_data[21])
format_specific['least_significant_bit'] = _check_assign(
least_significant_bit, unknown, np.nan)
low_filter_type = int(int_data[60])
if low_filter_type in FILTERS:
format_specific['low_filter_type'] = FILTERS[low_filter_type]
else:
format_specific['low_filter_type'] = ''
low_filter_corner = float(flt_data[53])
format_specific['low_filter_corner'] = _check_assign(
low_filter_corner, unknown, np.nan)
low_filter_decay = float(flt_data[54])
format_specific['low_filter_decay'] = _check_assign(
low_filter_decay, unknown, np.nan)
high_filter_type = int(int_data[61])
if high_filter_type in FILTERS:
format_specific['high_filter_type'] = FILTERS[high_filter_type]
else:
format_specific['high_filter_type'] = ''
high_filter_corner = float(flt_data[56])
format_specific['high_filter_corner'] = _check_assign(
high_filter_corner, unknown, np.nan)
high_filter_decay = float(flt_data[57])
format_specific['high_filter_decay'] = _check_assign(
high_filter_decay, unknown, np.nan)
maximum = float(flt_data[63])
format_specific['maximum'] = _check_assign(maximum, unknown, np.nan)
maximum_time = float(flt_data[64])
format_specific['maximum_time'] = _check_assign(maximum_time, unknown,
np.nan)
format_specific['station_code'] = _check_assign(structure_type, unknown,
np.nan)
record_flag = int(int_data[75])
if record_flag == 0:
format_specific['record_flag'] = 'No problem'
elif record_flag == 1:
format_specific['record_flag'] = 'Fixed'
elif record_flag == 2:
format_specific['record_flag'] = 'Unfixed problem'
else:
format_specific['record_flag'] = ''
scaling_factor = float(flt_data[87])
format_specific['scaling_factor'] = _check_assign(scaling_factor, unknown,
np.nan)
scaling_factor = float(flt_data[41])
format_specific['sensor_sensitivity'] = _check_assign(
scaling_factor, unknown, np.nan)
# Set dictionary
hdr['standard'] = standard
hdr['coordinates'] = coordinates
hdr['format_specific'] = format_specific
return hdr
def _get_header_info(int_data, flt_data, lines, cmt_data, location=""):
"""Return stats structure from various headers.
Output is a dictionary like this:
- network (str): Default is '--'. Determined using COSMOS_NETWORKS
- station (str)
- channel (str): Determined using COSMOS_ORIENTATIONS
- location (str): Set to location index of sensor site at station.
If not a multi-site array, default is '--'.
- starttime (datetime)
- duration (float)
- sampling_rate (float)
- delta (float)
- npts (int)
- coordinates:
- latitude (float)
- longitude (float)
- elevation (float)
- standard (Defaults are either np.nan or '')
- horizontal_orientation (float): Rotation from north (degrees)
- instrument_period (float): Period of sensor (Hz)
- instrument_damping (float): Fraction of critical
- process_time (datetime): Reported date of processing
- process_level: Either 'V0', 'V1', 'V2', or 'V3'
- station_name (str): Long form station description
- sensor_serial_number (str): Reported sensor serial
- instrument (str): See SENSOR_TYPES
- comments (str): Processing comments
- structure_type (str): See BUILDING_TYPES
- corner_frequency (float): Sensor corner frequency (Hz)
- units (str): See UNITS
- source (str): Network source description
- source_format (str): Always cosmos
- format_specific
- physical_units (str): See PHYSICAL_UNITS
- v30 (float): Site geology V30 (km/s)
- least_significant_bit: Recorder LSB in micro-volts (uv/count)
- low_filter_type (str): Filter used for low frequency
V2 filtering (see FILTERS)
- low_filter_corner (float): Filter corner for low frequency
V2 filtering (Hz)
- low_filter_decay (float): Filter decay for low frequency
V2 filtering (dB/octabe)
- high_filter_type (str): Filter used for high frequency
V2 filtering (see FILTERS)
- high_filter_corner (float): Filter corner for high frequency
V2 filtering (Hz)
- high_filter_decay (float): Filter decay for high frequency
V2 filtering (dB/octabe)
- maximum (float): Maximum value
- maximum_time (float): Time at which maximum occurs
- station_code (int): Code for structure_type
- record_flag (str): Either 'No problem', 'Fixed', 'Unfixed problem'.
Should be described in more depth in comments.
- scaling_factor (float): Scaling used for converting acceleration
from g/10 to cm/s/s
- sensor_sensitivity (float): Sensitvity in volts/g
Args:
int_data (ndarray):
Array of integer data.
flt_data (ndarray):
Array of float data.
lines (list):
List of text headers (str).
cmt_data (ndarray):
Array of comments (str).
Returns:
dictionary: Dictionary of header/metadata information
"""
hdr = {}
coordinates = {}
standard = {}
format_specific = {}
# Get unknown parameter number
try:
unknown = int(lines[12][64:71])
except ValueError:
unknown = -999
# required metadata
network_num = int(int_data[10])
# Get network from cosmos table or fdsn code sheet
if network_num in COSMOS_NETWORKS:
network = COSMOS_NETWORKS[network_num][0]
source = COSMOS_NETWORKS[network_num][1]
if network == "":
network = COSMOS_NETWORKS[network_num][2]
else:
network_code = lines[4][25:27].upper()
if network_code in CODES:
network = network_code
idx = np.argwhere(CODES == network_code)[0][0]
source = (
SOURCES1[idx].decode("utf-8") + ", " + SOURCES2[idx].decode("utf-8")
)
else:
network = "--"
source = ""
hdr["network"] = network
logging.debug(f"network: {network}")
hdr["station"] = lines[4][28:34].strip()
logging.debug(f"station: {hdr['station']}")
# the channel orientation can be either relative to true north (idx 53)
# or relative to sensor orientation (idx 54).
horizontal_angle = int(int_data[53])
logging.debug(f"horizontal_angle: {horizontal_angle}")
if horizontal_angle not in VALID_AZIMUTH_INTS:
angles = np.array(int_data[19:21]).astype(np.float32)
angles[angles == unknown] = np.nan
if np.isnan(angles).all():
logging.warning("Horizontal_angle in COSMOS header is not valid.")
else:
ref = angles[~np.isnan(angles)][0]
horizontal_angle = int(int_data[54])
if horizontal_angle not in VALID_AZIMUTH_INTS:
logging.warning("Horizontal_angle in COSMOS header is not valid.")
else:
horizontal_angle += ref
if horizontal_angle > 360:
horizontal_angle -= 360
horizontal_angle = float(horizontal_angle)
# Store delta and duration. Use them to calculate npts and sampling_rate
# NOTE: flt_data[33] is the delta of the V0 format, and if we are reading
# a V1 or V2 format then it may have been resampled. We should consider
# adding flt_data[33] delta to the provenance record at some point.
delta = float(flt_data[61]) * MSEC_TO_SEC
if delta != unknown:
hdr["delta"] = delta
hdr["sampling_rate"] = 1 / delta
# Determine the angle based upon the cosmos table
# Set horizontal angles other than N,S,E,W to H1 and H2
# Missing angle results in the channel number
if horizontal_angle != unknown:
if horizontal_angle in COSMOS_ORIENTATIONS:
channel = COSMOS_ORIENTATIONS[horizontal_angle][1].upper()
if channel == "UP" or channel == "DOWN" or channel == "VERT":
channel = get_channel_name(
hdr["sampling_rate"],
is_acceleration=True,
is_vertical=True,
is_north=False,
)
horizontal_angle = 360.0
elif channel == "RADL" or channel == "LONG" or channel == "H1":
channel = get_channel_name(
hdr["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=True,
)
horizontal_angle = 0.0
elif channel == "TRAN" or channel == "TANG" or channel == "H2":
channel = get_channel_name(
hdr["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=False,
)
horizontal_angle = 90.0
else: # For the occassional 'OTHR' channel
raise ValueError("Channel name is not valid.")
elif horizontal_angle >= 0 and horizontal_angle <= 360:
if (
horizontal_angle > 315
or horizontal_angle < 45
or (horizontal_angle > 135 and horizontal_angle < 225)
):
channel = get_channel_name(
hdr["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=True,
)
else:
channel = get_channel_name(
hdr["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=False,
)
horizontal_orientation = horizontal_angle
else:
errstr = (
"Not enough information to distinguish horizontal from "
"vertical channels."
)
raise BaseException("COSMOS: " + errstr)
hdr["channel"] = channel
logging.debug(f"channel: {hdr['channel']}")
if location == "":
location = int(int_data[55])
location = str(_check_assign(location, unknown, "--"))
if len(location) < 2:
location = location.zfill(2)
hdr["location"] = location
else:
hdr["location"] = location
year = int(int_data[39])
month = int(int_data[41])
day = int(int_data[42])
hour = int(int_data[43])
minute = int(int_data[44])
second = float(flt_data[29])
# If anything more than seconds is excluded
# It is considered inadequate time information
if second == unknown:
try:
hdr["starttime"] = datetime(year, month, day, hour, minute)
except BaseException:
raise BaseException("COSMOS: Inadequate start time information.")
else:
second = second
microsecond = int((second - int(second)) * 1e6)
try:
hdr["starttime"] = datetime(
year, month, day, hour, minute, int(second), microsecond
)
except BaseException:
raise BaseException("COSMOS: Inadequate start time information.")
if flt_data[62] != unknown:
# COSMOS **defines** "length" as npts*dt (note this is a bit unusual)
cosmos_length = flt_data[62]
npts = int(cosmos_length / delta)
hdr["duration"] = (npts - 1) * delta
hdr["npts"] = npts
else:
raise ValueError("COSMOS file does not specify length.")
# coordinate information
coordinates["latitude"] = float(flt_data[0])
coordinates["longitude"] = float(flt_data[1])
coordinates["elevation"] = float(flt_data[2])
for key in coordinates:
if coordinates[key] == unknown:
if key != "elevation":
logging.warning(f"Missing {key!r}. Setting to np.nan.", Warning)
coordinates[key] = np.nan
else:
logging.warning(f"Missing {key!r}. Setting to 0.0.", Warning)
coordinates[key] = 0.0
hdr["coordinates"] = coordinates
# standard metadata
standard["units_type"] = get_units_type(channel)
standard["source"] = source
standard["horizontal_orientation"] = horizontal_orientation
standard["vertical_orientation"] = np.nan
station_name = lines[4][40:-1].strip()
standard["station_name"] = station_name
instrument_frequency = float(flt_data[39])
if instrument_frequency == 0:
standard["instrument_period"] = np.nan
logging.warning("Instrument Frequency == 0")
else:
inst_freq = _check_assign(instrument_frequency, unknown, np.nan)
standard["instrument_period"] = 1.0 / inst_freq
instrument_damping = float(flt_data[40])
standard["instrument_damping"] = _check_assign(instrument_damping, unknown, np.nan)
process_line = lines[10][10:40]
if process_line.find("-") >= 0 or process_line.find("/") >= 0:
if process_line.find("-") >= 0:
delimeter = "-"
elif process_line.find("/") >= 0:
delimeter = "/"
try:
date = process_line.split(delimeter)
month = int(date[0][-2:])
day = int(date[1])
year = int(date[2][:4])
time = process_line.split(":")
hour = int(time[0][-2:])
minute = int(time[1])
second = float(time[2][:2])
microsecond = int((second - int(second)) * 1e6)
etime = datetime(year, month, day, hour, minute, int(second), microsecond)
standard["process_time"] = etime.strftime(TIMEFMT)
except BaseException:
standard["process_time"] = ""
else:
standard["process_time"] = ""
process_level = int(int_data[0])
if process_level == 0:
standard["process_level"] = PROCESS_LEVELS["V0"]
elif process_level == 1:
standard["process_level"] = PROCESS_LEVELS["V1"]
elif process_level == 2:
standard["process_level"] = PROCESS_LEVELS["V2"]
elif process_level == 3:
standard["process_level"] = PROCESS_LEVELS["V3"]
else:
standard["process_level"] = PROCESS_LEVELS["V1"]
logging.debug(f"process_level: {process_level}")
serial = int(int_data[52])
if serial != unknown:
standard["sensor_serial_number"] = str(_check_assign(serial, unknown, ""))
else:
standard["sensor_serial_number"] = ""
instrument = int(int_data[51])
if instrument != unknown and instrument in SENSOR_TYPES:
standard["instrument"] = SENSOR_TYPES[instrument]
else:
standard["instrument"] = lines[6][57:-1].strip()
structure_type = int(int_data[18])
if structure_type != unknown and structure_type in BUILDING_TYPES:
standard["structure_type"] = BUILDING_TYPES[structure_type]
else:
standard["structure_type"] = ""
frequency = float(flt_data[25])
standard["corner_frequency"] = _check_assign(frequency, unknown, np.nan)
physical_parameter = int(int_data[2])
units = int(int_data[1])
if units != unknown and units in UNITS:
standard["units_type"] = UNITS[units]
else:
if physical_parameter in [2, 4, 7, 10, 11, 12, 23]:
standard["units_type"] = "acc"
elif physical_parameter in [5, 8, 24]:
standard["units_type"] = "vel"
elif physical_parameter in [6, 9, 25]:
standard["units_type"] = "disp"
standard["source_format"] = "cosmos"
standard["comments"] = ", ".join(cmt_data)
# format specific metadata
if physical_parameter in PHYSICAL_UNITS:
physical_parameter = PHYSICAL_UNITS[physical_parameter][0]
format_specific["physical_units"] = physical_parameter
v30 = float(flt_data[3])
format_specific["v30"] = _check_assign(v30, unknown, np.nan)
least_significant_bit = float(flt_data[21])
format_specific["least_significant_bit"] = _check_assign(
least_significant_bit, unknown, np.nan
)
gain = float(flt_data[46])
format_specific["gain"] = _check_assign(gain, unknown, np.nan)
low_filter_type = int(int_data[60])
if low_filter_type in FILTERS:
format_specific["low_filter_type"] = FILTERS[low_filter_type]
else:
format_specific["low_filter_type"] = ""
low_filter_corner = float(flt_data[53])
format_specific["low_filter_corner"] = _check_assign(
low_filter_corner, unknown, np.nan
)
low_filter_decay = float(flt_data[54])
format_specific["low_filter_decay"] = _check_assign(
low_filter_decay, unknown, np.nan
)
high_filter_type = int(int_data[61])
if high_filter_type in FILTERS:
format_specific["high_filter_type"] = FILTERS[high_filter_type]
else:
format_specific["high_filter_type"] = ""
high_filter_corner = float(flt_data[56])
format_specific["high_filter_corner"] = _check_assign(
high_filter_corner, unknown, np.nan
)
high_filter_decay = float(flt_data[57])
format_specific["high_filter_decay"] = _check_assign(
high_filter_decay, unknown, np.nan
)
maximum = float(flt_data[63])
format_specific["maximum"] = _check_assign(maximum, unknown, np.nan)
maximum_time = float(flt_data[64])
format_specific["maximum_time"] = _check_assign(maximum_time, unknown, np.nan)
format_specific["station_code"] = _check_assign(structure_type, unknown, np.nan)
record_flag = int(int_data[75])
if record_flag == 0:
format_specific["record_flag"] = "No problem"
elif record_flag == 1:
format_specific["record_flag"] = "Fixed"
elif record_flag == 2:
format_specific["record_flag"] = "Unfixed problem"
else:
format_specific["record_flag"] = ""
scaling_factor = float(flt_data[87])
format_specific["scaling_factor"] = _check_assign(scaling_factor, unknown, np.nan)
sensor_sensitivity = float(flt_data[41])
format_specific["sensor_sensitivity"] = _check_assign(
sensor_sensitivity, unknown, np.nan
)
# for V0 files, set a standard field called instrument_sensitivity
ctov = least_significant_bit / MICRO_TO_VOLT
vtog = 1 / format_specific["sensor_sensitivity"]
if not np.isnan(format_specific["gain"]):
gain = format_specific["gain"]
else:
gain = 1.0
if gain == 0:
fmt = "%s.%s.%s.%s"
tpl = (hdr["network"], hdr["station"], hdr["channel"], hdr["location"])
nscl = fmt % tpl
raise ValueError(f"Gain of 0 discovered for NSCL: {nscl}")
denom = ctov * vtog * (1.0 / gain) * sp.g
standard["instrument_sensitivity"] = 1 / denom
standard["volts_to_counts"] = ctov
# Set dictionary
hdr["standard"] = standard
hdr["coordinates"] = coordinates
hdr["format_specific"] = format_specific
return hdr
def _read_header(hdr_data, station, name, component, data_format, instrument,
resolution):
"""Construct stats dictionary from header lines.
Args:
hdr_data (ndarray):
(10,10) numpy array containing header data.
station (str):
Station code obtained from previous text portion of header.
location (str):
Location string obtained from previous text portion of header.
component (str):
Component direction (N18E, S72W, etc.)
Returns:
Dictionary containing fields:
- network "NZ"
- station
- channel H1,H2,or Z.
- location
- sampling_rate Samples per second.
- delta Interval between samples (seconds)
- calib Calibration factor (always 1.0)
- npts Number of samples in record.
- starttime Datetime object containing start of record.
- standard:
- station_name
- units "acc"
- source 'New Zealand Institute of Geological and Nuclear
Science'
- horizontal_orientation
- instrument_period
- instrument_damping
- processing_time
- process_level
- sensor_serial_number
- instrument
- comments
- structure_type
- corner_frequency
- source_format
- coordinates:
- lat Latitude of station.
- lon Longitude of station.
- elevation Elevation of station.
- format_specific:
- sensor_bit_resolution
"""
hdr = {}
standard = {}
coordinates = {}
format_specific = {}
hdr["station"] = station
standard["station_name"] = name
# Note: Original sample interval (s): hdr_data[6, 4]
# Sample inverval (s)
hdr["delta"] = hdr_data[6, 5]
hdr["sampling_rate"] = 1 / hdr["delta"]
hdr["calib"] = 1.0
if data_format == "V1":
hdr["npts"] = int(hdr_data[3, 0])
else:
hdr["npts"] = int(hdr_data[3, 3])
hdr["network"] = "NZ"
standard["units_type"] = "acc"
standard["units"] = "cm/s/s"
standard[
"source"] = "New Zealand Institute of Geological and Nuclear Science"
logging.debug(f"component: {component}")
standard["vertical_orientation"] = np.nan
if component.lower() in ["up", "down"]:
standard["horizontal_orientation"] = np.nan
hdr["channel"] = get_channel_name(hdr["delta"],
is_acceleration=True,
is_vertical=True,
is_north=False)
else:
angle = _get_channel(component)
logging.debug(f"angle: {angle}")
standard["horizontal_orientation"] = float(angle)
if (angle > 315 or angle < 45) or (angle > 135 and angle < 225):
hdr["channel"] = get_channel_name(hdr["delta"],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
hdr["channel"] = get_channel_name(hdr["delta"],
is_acceleration=True,
is_vertical=False,
is_north=False)
logging.debug(f"channel: {hdr['channel']}")
hdr["location"] = "--"
# figure out the start time
milliseconds = hdr_data[3, 9]
seconds = int(milliseconds / 1000)
microseconds = int(np.round(milliseconds / 1000.0 - seconds))
year = int(hdr_data[0, 8])
month = int(hdr_data[0, 9])
day = int(hdr_data[1, 8])
hour = int(hdr_data[1, 9])
minute = int(hdr_data[3, 8])
hdr["starttime"] = datetime(year, month, day, hour, minute, seconds,
microseconds)
# figure out station coordinates
latdg = hdr_data[2, 0]
latmn = hdr_data[2, 1]
latsc = hdr_data[2, 2]
coordinates["latitude"] = _dms_to_dd(latdg, latmn, latsc) * -1
londg = hdr_data[2, 3]
lonmn = hdr_data[2, 4]
lonsc = hdr_data[2, 5]
coordinates["longitude"] = _dms_to_dd(londg, lonmn, lonsc)
logging.warning("Setting elevation to 0.0")
coordinates["elevation"] = 0.0
# get other standard metadata
standard["units_type"] = get_units_type(hdr["channel"])
standard["instrument_period"] = 1 / hdr_data[4, 0]
standard["instrument_damping"] = hdr_data[4, 1]
standard["process_time"] = ""
standard["process_level"] = PROCESS_LEVELS[data_format]
logging.debug(f"process_level: {data_format}")
standard["sensor_serial_number"] = ""
standard["instrument"] = instrument
standard["comments"] = ""
standard["structure_type"] = ""
standard["corner_frequency"] = np.nan
standard["source_format"] = "geonet"
# these fields can be used for instrument correction
# when data is in counts
standard["instrument_sensitivity"] = np.nan
standard["volts_to_counts"] = np.nan
# get format specific metadata
format_specific["sensor_bit_resolution"] = resolution
hdr["coordinates"] = coordinates
hdr["standard"] = standard
hdr["format_specific"] = format_specific
return hdr
def _get_header_info(int_data, flt_data, lines, cmt_data, location=''):
"""Return stats structure from various headers.
Output is a dictionary like this:
- network (str): Default is '--'. Determined using COSMOS_NETWORKS
- station (str)
- channel (str): Determined using COSMOS_ORIENTATIONS
- location (str): Set to location index of sensor site at station.
If not a multi-site array, default is '--'.
- starttime (datetime)
- duration (float)
- sampling_rate (float)
- delta (float)
- npts (int)
- coordinates:
- latitude (float)
- longitude (float)
- elevation (float)
- standard (Defaults are either np.nan or '')
- horizontal_orientation (float): Rotation from north (degrees)
- instrument_period (float): Period of sensor (Hz)
- instrument_damping (float): Fraction of critical
- process_time (datetime): Reported date of processing
- process_level: Either 'V0', 'V1', 'V2', or 'V3'
- station_name (str): Long form station description
- sensor_serial_number (str): Reported sensor serial
- instrument (str): See SENSOR_TYPES
- comments (str): Processing comments
- structure_type (str): See BUILDING_TYPES
- corner_frequency (float): Sensor corner frequency (Hz)
- units (str): See UNITS
- source (str): Network source description
- source_format (str): Always cosmos
- format_specific
- physical_units (str): See PHYSICAL_UNITS
- v30 (float): Site geology V30 (km/s)
- least_significant_bit: Recorder LSB in micro-volts (uv/count)
- low_filter_type (str): Filter used for low frequency
V2 filtering (see FILTERS)
- low_filter_corner (float): Filter corner for low frequency
V2 filtering (Hz)
- low_filter_decay (float): Filter decay for low frequency
V2 filtering (dB/octabe)
- high_filter_type (str): Filter used for high frequency
V2 filtering (see FILTERS)
- high_filter_corner (float): Filter corner for high frequency
V2 filtering (Hz)
- high_filter_decay (float): Filter decay for high frequency
V2 filtering (dB/octabe)
- maximum (float): Maximum value
- maximum_time (float): Time at which maximum occurs
- station_code (int): Code for structure_type
- record_flag (str): Either 'No problem', 'Fixed', 'Unfixed problem'.
Should be described in more depth in comments.
- scaling_factor (float): Scaling used for converting acceleration
from g/10 to cm/s/s
- sensor_sensitivity (float): Sensitvity in volts/g
Args:
int_data (ndarray): Array of integer data
flt_data (ndarray): Array of float data
lines (list): List of text headers (str)
cmt_data (ndarray): Array of comments (str)
Returns:
dictionary: Dictionary of header/metadata information
"""
hdr = {}
coordinates = {}
standard = {}
format_specific = {}
# Get unknown parameter number
try:
unknown = int(lines[12][64:71])
except ValueError:
unknown = -999
# required metadata
network_num = int(int_data[10])
# Get network from cosmos table or fdsn code sheet
if network_num in COSMOS_NETWORKS:
network = COSMOS_NETWORKS[network_num][0]
source = COSMOS_NETWORKS[network_num][1]
if network == '':
network = COSMOS_NETWORKS[network_num][2]
else:
network_code = lines[4][25:27].upper()
if network_code in CODES:
network = network_code
idx = np.argwhere(CODES == network_code)[0][0]
source = SOURCES1[idx].decode(
'utf-8') + ', ' + SOURCES2[idx].decode('utf-8')
else:
network = 'ZZ'
source = ''
hdr['network'] = network
logging.debug('network: %s' % network)
hdr['station'] = lines[4][28:34].strip()
logging.debug('station: %s' % hdr['station'])
horizontal_angle = int(int_data[53])
logging.debug('horizontal_angle: %s' % horizontal_angle)
if horizontal_angle not in VALID_AZIMUTH_INTS:
logging.warning("Horizontal_angle in COSMOS header is not valid.")
horizontal_angle = float(horizontal_angle)
# Store delta and duration. Use them to calculate npts and sampling_rate
# NOTE: flt_data[33] is the delta of the V0 format, and if we are reading
# a V1 or V2 format then it may have been resampled. We should consider
# adding flt_data[33] delta to the provenance record at some point.
delta = float(flt_data[61]) * MSEC_TO_SEC
if delta != unknown:
hdr['delta'] = delta
hdr['sampling_rate'] = 1 / delta
# Determine the angle based upon the cosmos table
# Set horizontal angles other than N,S,E,W to H1 and H2
# Missing angle results in the channel number
if horizontal_angle != unknown:
if horizontal_angle in COSMOS_ORIENTATIONS:
channel = COSMOS_ORIENTATIONS[horizontal_angle][1].upper()
if channel == 'UP' or channel == 'DOWN' or channel == 'VERT':
channel = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
elif horizontal_angle >= 0 and horizontal_angle <= 360:
if (
horizontal_angle > 315
or horizontal_angle < 45
or (horizontal_angle > 135 and horizontal_angle < 225)
):
channel = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
channel = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
horizontal_orientation = horizontal_angle
else:
errstr = ('Not enough information to distinguish horizontal from '
'vertical channels.')
raise GMProcessException('COSMOS: ' + errstr)
hdr['channel'] = channel
logging.debug('channel: %s' % hdr['channel'])
if location == '':
location = int(int_data[55])
location = str(_check_assign(location, unknown, '--'))
if len(location) < 2:
location = location.zfill(2)
hdr['location'] = location
else:
hdr['location'] = location
year = int(int_data[39])
month = int(int_data[41])
day = int(int_data[42])
hour = int(int_data[43])
minute = int(int_data[44])
second = float(flt_data[29])
# If anything more than seconds is excluded
# It is considered inadequate time information
if second == unknown:
try:
hdr['starttime'] = datetime(
year, month, day, hour, minute)
except Exception:
raise GMProcessException(
'COSMOS: Inadequate start time information.')
else:
second = second
microsecond = int((second - int(second)) * 1e6)
try:
hdr['starttime'] = datetime(
year, month, day, hour, minute, int(second), microsecond)
except Exception:
raise GMProcessException(
'COSMOS: Inadequate start time information.')
if flt_data[62] != unknown:
# COSMOS **defines** "length" as npts*dt (note this is a bit unusual)
cosmos_length = flt_data[62]
npts = int(cosmos_length / delta)
hdr['duration'] = (npts - 1) * delta
hdr['npts'] = npts
else:
raise ValueError('COSMOS file does not specify length.')
# coordinate information
coordinates['latitude'] = float(flt_data[0])
coordinates['longitude'] = float(flt_data[1])
coordinates['elevation'] = float(flt_data[2])
for key in coordinates:
if coordinates[key] == unknown:
warnings.warn('Missing %r. Setting to np.nan.' % key, Warning)
coordinates[key] = np.nan
hdr['coordinates'] = coordinates
# standard metadata
standard['source'] = source
standard['horizontal_orientation'] = horizontal_orientation
station_name = lines[4][40:-1].strip()
standard['station_name'] = station_name
instrument_frequency = float(flt_data[39])
standard['instrument_period'] = 1.0 / _check_assign(instrument_frequency,
unknown, np.nan)
instrument_damping = float(flt_data[40])
standard['instrument_damping'] = _check_assign(instrument_damping,
unknown, np.nan)
process_line = lines[10][10:40]
if process_line.find('-') >= 0 or process_line.find('/') >= 0:
if process_line.find('-') >= 0:
delimeter = '-'
elif process_line.find('/') >= 0:
delimeter = '/'
try:
date = process_line.split(delimeter)
month = int(date[0][-2:])
day = int(date[1])
year = int(date[2][:4])
time = process_line.split(':')
hour = int(time[0][-2:])
minute = int(time[1])
second = float(time[2][:2])
microsecond = int((second - int(second)) * 1e6)
etime = datetime(year, month, day, hour, minute,
int(second), microsecond)
standard['process_time'] = etime.strftime(TIMEFMT)
except Exception:
standard['process_time'] = ''
else:
standard['process_time'] = ''
process_level = int(int_data[0])
if process_level == 0:
standard['process_level'] = PROCESS_LEVELS['V0']
elif process_level == 1:
standard['process_level'] = PROCESS_LEVELS['V1']
elif process_level == 2:
standard['process_level'] = PROCESS_LEVELS['V2']
elif process_level == 3:
standard['process_level'] = PROCESS_LEVELS['V3']
else:
standard['process_level'] = PROCESS_LEVELS['V1']
logging.debug("process_level: %s" % process_level)
serial = int(int_data[52])
if serial != unknown:
standard['sensor_serial_number'] = str(_check_assign(
serial, unknown, ''))
else:
standard['sensor_serial_number'] = ''
instrument = int(int_data[51])
if instrument != unknown and instrument in SENSOR_TYPES:
standard['instrument'] = SENSOR_TYPES[instrument]
else:
standard['instrument'] = lines[6][57:-1].strip()
structure_type = int(int_data[18])
if structure_type != unknown and structure_type in BUILDING_TYPES:
standard['structure_type'] = BUILDING_TYPES[structure_type]
else:
standard['structure_type'] = ''
frequency = float(flt_data[25])
standard['corner_frequency'] = _check_assign(frequency, unknown, np.nan)
physical_parameter = int(int_data[2])
units = int(int_data[1])
if units != unknown and units in UNITS:
standard['units'] = UNITS[units]
else:
if physical_parameter in [2, 4, 7, 10, 11, 12, 23]:
standard['units'] = 'acc'
elif physical_parameter in [5, 8, 24]:
standard['units'] = 'vel'
elif physical_parameter in [6, 9, 25]:
standard['units'] = 'disp'
standard['source_format'] = 'cosmos'
standard['comments'] = ', '.join(cmt_data)
# format specific metadata
if physical_parameter in PHYSICAL_UNITS:
physical_parameter = PHYSICAL_UNITS[physical_parameter][0]
format_specific['physical_units'] = physical_parameter
v30 = float(flt_data[3])
format_specific['v30'] = _check_assign(v30, unknown, np.nan)
least_significant_bit = float(flt_data[21])
format_specific['least_significant_bit'] = _check_assign(
least_significant_bit, unknown, np.nan)
low_filter_type = int(int_data[60])
if low_filter_type in FILTERS:
format_specific['low_filter_type'] = FILTERS[low_filter_type]
else:
format_specific['low_filter_type'] = ''
low_filter_corner = float(flt_data[53])
format_specific['low_filter_corner'] = _check_assign(
low_filter_corner, unknown, np.nan)
low_filter_decay = float(flt_data[54])
format_specific['low_filter_decay'] = _check_assign(
low_filter_decay, unknown, np.nan)
high_filter_type = int(int_data[61])
if high_filter_type in FILTERS:
format_specific['high_filter_type'] = FILTERS[high_filter_type]
else:
format_specific['high_filter_type'] = ''
high_filter_corner = float(flt_data[56])
format_specific['high_filter_corner'] = _check_assign(
high_filter_corner, unknown, np.nan)
high_filter_decay = float(flt_data[57])
format_specific['high_filter_decay'] = _check_assign(
high_filter_decay, unknown, np.nan)
maximum = float(flt_data[63])
format_specific['maximum'] = _check_assign(maximum, unknown, np.nan)
maximum_time = float(flt_data[64])
format_specific['maximum_time'] = _check_assign(
maximum_time, unknown, np.nan)
format_specific['station_code'] = _check_assign(
structure_type, unknown, np.nan)
record_flag = int(int_data[75])
if record_flag == 0:
format_specific['record_flag'] = 'No problem'
elif record_flag == 1:
format_specific['record_flag'] = 'Fixed'
elif record_flag == 2:
format_specific['record_flag'] = 'Unfixed problem'
else:
format_specific['record_flag'] = ''
scaling_factor = float(flt_data[87])
format_specific['scaling_factor'] = _check_assign(
scaling_factor, unknown, np.nan)
scaling_factor = float(flt_data[41])
format_specific['sensor_sensitivity'] = _check_assign(
scaling_factor, unknown, np.nan)
# Set dictionary
hdr['standard'] = standard
hdr['coordinates'] = coordinates
hdr['format_specific'] = format_specific
return hdr
def read_knet(filename):
"""Read Japanese KNET strong motion file.
Args:
filename (str): Path to possible KNET data file.
kwargs (ref): Other arguments will be ignored.
Returns:
Stream: Obspy Stream containing three channels of acceleration data
(cm/s**2).
"""
logging.debug("Starting read_knet.")
if not is_knet(filename):
raise Exception('%s is not a valid KNET file' % filename)
# Parse the header portion of the file
with open(filename, 'rt') as f:
lines = [next(f) for x in range(TEXT_HDR_ROWS)]
hdr = {}
coordinates = {}
standard = {}
hdr['network'] = 'BO'
hdr['station'] = lines[5].split()[2]
logging.debug('station: %s' % hdr['station'])
standard['station_name'] = ''
# according to the powers that defined the Network.Station.Channel.Location
# "standard", Location is a two character field. Most data providers,
# including KNET here, don't provide this. We'll flag it as "--".
hdr['location'] = '--'
coordinates['latitude'] = float(lines[6].split()[2])
coordinates['longitude'] = float(lines[7].split()[2])
coordinates['elevation'] = float(lines[8].split()[2])
hdr['sampling_rate'] = float(
re.search('\\d+', lines[10].split()[2]).group())
hdr['delta'] = 1 / hdr['sampling_rate']
standard['units'] = 'acc'
dir_string = lines[12].split()[1].strip()
# knet files have directions listed as N-S, E-W, or U-D,
# whereas in kiknet those directions are '4', '5', or '6'.
if dir_string in ['N-S', '1', '4']:
hdr['channel'] = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
elif dir_string in ['E-W', '2', '5']:
hdr['channel'] = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
elif dir_string in ['U-D', '3', '6']:
hdr['channel'] = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
else:
raise Exception('KNET: Could not parse direction %s' %
lines[12].split()[1])
logging.debug('channel: %s' % hdr['channel'])
scalestr = lines[13].split()[2]
parts = scalestr.split('/')
num = float(parts[0].replace('(gal)', ''))
den = float(parts[1])
calib = num / den
hdr['calib'] = calib
duration = float(lines[11].split()[2])
hdr['npts'] = int(duration * hdr['sampling_rate'])
timestr = ' '.join(lines[9].split()[2:4])
# The K-NET and KiK-Net data logger adds a 15s time delay
# this is removed here
sttime = datetime.strptime(timestr, TIMEFMT) - timedelta(seconds=15.0)
# Shift the time to utc (Japanese time is 9 hours ahead)
sttime = sttime - timedelta(seconds=9 * 3600.)
hdr['starttime'] = sttime
# read in the data - there is a max of 8 columns per line
# the code below handles the case when last line has
# less than 8 columns
if hdr['npts'] % COLS_PER_LINE != 0:
nrows = int(np.floor(hdr['npts'] / COLS_PER_LINE))
nrows2 = 1
else:
nrows = int(np.ceil(hdr['npts'] / COLS_PER_LINE))
nrows2 = 0
data = np.genfromtxt(filename, skip_header=TEXT_HDR_ROWS,
max_rows=nrows, filling_values=np.nan)
data = data.flatten()
if nrows2:
skip_header = TEXT_HDR_ROWS + nrows
data2 = np.genfromtxt(filename, skip_header=skip_header,
max_rows=nrows2, filling_values=np.nan)
data = np.hstack((data, data2))
nrows += nrows2
# apply the correction factor we're given in the header
data *= calib
# fill out the rest of the standard dictionary
standard['horizontal_orientation'] = np.nan
standard['instrument_period'] = np.nan
standard['instrument_damping'] = np.nan
standard['process_time'] = ''
standard['process_level'] = PROCESS_LEVELS['V1']
standard['sensor_serial_number'] = ''
standard['instrument'] = ''
standard['comments'] = ''
standard['structure_type'] = ''
if dir_string in ['1', '2', '3']:
standard['structure_type'] = 'borehole'
standard['corner_frequency'] = np.nan
standard['units'] = 'acc'
standard['source'] = SRC
standard['source_format'] = 'knet'
head, tail = os.path.split(filename)
standard['source_file'] = tail or os.path.basename(head)
hdr['coordinates'] = coordinates
hdr['standard'] = standard
# create a Trace from the data and metadata
trace = StationTrace(data.copy(), Stats(hdr.copy()))
response = {'input_units': 'counts', 'output_units': 'cm/s^2'}
trace.setProvenance('remove_response', response)
stream = StationStream(traces=[trace])
return [stream]
def _read_header(lines, filename, table):
header = {}
standard = {}
coords = {}
format_specific = {}
# fill out the standard dictionary
standard['source'] = SOURCE
standard['source_format'] = SOURCE_FORMAT
standard['instrument'] = ''
standard['sensor_serial_number'] = ''
standard['process_level'] = PROCESS_LEVELS['V1']
standard['process_time'] = lines[0].split(':')[1].strip()
# station name line can look like this:
# VI�A DEL MAR CENTRO S/N 675
sparts = lines[5].split()
station_name = ' '.join(sparts[0:sparts.index('S/N')])
standard['station_name'] = station_name
# this table gives us station coordinates and structure type
station_row = table[table['Name'] == station_name]
if not len(station_row):
logging.warning('Unknown structure type.')
standard['structure_type'] = ''
else:
row = station_row.iloc[0]
standard['structure_type'] = row['Structure Type']
standard['corner_frequency'] = np.nan
standard['units'] = 'cm/s^2'
standard['units_type'] = 'acc'
inst_dict = {}
for part in lines[9].split(','):
key, value = part.split('=')
fvalue_str = re.search(FLOATRE, value.strip()).group()
inst_dict[key.strip()] = float(fvalue_str)
standard['instrument_period'] = inst_dict['INSTR PERIOD']
standard['instrument_damping'] = inst_dict['DAMPING']
standard['horizontal_orientation'] = np.nan
standard['comments'] = ' '.join(lines[11:13]).replace('\n', '')
head, tail = os.path.split(filename)
standard['source_file'] = tail or os.path.basename(head)
# this field can be used for instrument correction
# when data is in counts
standard['instrument_sensitivity'] = inst_dict['SENSITIVITY']
# fill out the stats stuff
try:
stimestr = re.search(TIME_RE, lines[3]).group()
except AttributeError:
try:
stimestr = re.search(TIME_RE2, lines[3]).group()
except AttributeError:
logging.warning('Setting time to epoch.')
stimestr = '01/01/1970 00:00:00.000'
# 2/27/2010 2:45:46.000 GMT
stime = datetime.strptime(stimestr, TIMEFMT)
# it appears that sometimes the trigger time is set to Jan 1, 1980
# by default.
if stime.year == 1980 and stime.month == 1 and stime.day == 1:
fmt = 'Trigger time set to %s in file %s'
logging.warning(fmt % (str(stime), standard['source_file']))
header['starttime'] = stime
npts, duration = re.findall(FLOATRE, lines[10])
npts = int(npts)
duration = float(duration)
header['npts'] = npts
header['delta'] = duration / (npts - 1)
header['sampling_rate'] = (npts - 1) / duration
header['duration'] = duration
raw_channel = lines[6][9:11].strip()
if raw_channel in NORTH_CHANNELS:
channel = get_channel_name(header['sampling_rate'], True, False, True)
elif raw_channel in WEST_CHANNELS:
channel = get_channel_name(header['sampling_rate'], True, False, False)
elif raw_channel in VERTICAL_CHANNELS:
channel = get_channel_name(header['sampling_rate'], True, True, False)
else:
raise KeyError('Channel name %s not defined' % raw_channel)
header['channel'] = channel
header['station'] = lines[5].split()[-1]
header['location'] = '--'
header['network'] = NETWORK
# these files seem to have all zeros for station coordinates!
if not len(station_row):
logging.warning('Could not find station match for %s' % station_name)
coordparts = lines[4].split()
lat = float(re.search(FLOATRE, coordparts[2]).group())
lon = float(re.search(FLOATRE, coordparts[3]).group())
if lon == 0 or lat == 0:
logging.warning('Latitude or Longitude values are 0')
if 'S' in coordparts[2]:
lat = -1 * lat
if 'W' in coordparts[3]:
lon = -1 * lon
else:
row = station_row.iloc[0]
lat = row['Lat']
lon = row['Lon']
altitude = 0.0
logging.warn('Setting elevation to 0.0')
coords = {'latitude': lat,
'longitude': lon,
'elevation': altitude}
header['coordinates'] = coords
header['standard'] = standard
header['format_specific'] = format_specific
return header
def read_cwb(filename, **kwargs):
"""Read Taiwan Central Weather Bureau strong motion file.
Args:
filename (str): Path to possible CWB data file.
kwargs (ref): Other arguments will be ignored.
Returns:
Stream: Obspy Stream containing three channels of acceleration
data (cm/s**2).
"""
logging.debug("Starting read_cwb.")
if not is_cwb(filename):
raise Exception('%s is not a valid CWB strong motion data file.' %
filename)
f = open(filename, 'rt')
# according to the powers that defined the Network.Station.Channel.Location
# "standard", Location is a two character field. Most data providers,
# including CWB here, don't provide this. We'll flag it as "--".
data = np.genfromtxt(filename,
skip_header=HDR_ROWS,
delimiter=[COLWIDTH] * NCOLS) # time, Z, NS, EW
hdr = _get_header_info(f, data)
f.close()
head, tail = os.path.split(filename)
hdr['standard']['source_file'] = tail or os.path.basename(head)
hdr_z = hdr.copy()
hdr_z['channel'] = get_channel_name(hdr['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
hdr_z['standard']['horizontal_orientation'] = np.nan
hdr_z['standard']['vertical_orientation'] = np.nan
hdr_z['standard']['units_type'] = get_units_type(hdr_z['channel'])
hdr_h1 = hdr.copy()
hdr_h1['channel'] = get_channel_name(hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
hdr_h1['standard']['horizontal_orientation'] = np.nan
hdr_h1['standard']['vertical_orientation'] = np.nan
hdr_h1['standard']['units_type'] = get_units_type(hdr_h1['channel'])
hdr_h2 = hdr.copy()
hdr_h2['channel'] = get_channel_name(hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
hdr_h2['standard']['horizontal_orientation'] = np.nan
hdr_h2['standard']['vertical_orientation'] = np.nan
hdr_h2['standard']['units_type'] = get_units_type(hdr_h2['channel'])
stats_z = Stats(hdr_z)
stats_h1 = Stats(hdr_h1)
stats_h2 = Stats(hdr_h2)
response = {'input_units': 'counts', 'output_units': 'cm/s^2'}
trace_z = StationTrace(data=data[:, 1], header=stats_z)
trace_z.setProvenance('remove_response', response)
trace_h1 = StationTrace(data=data[:, 2], header=stats_h1)
trace_h1.setProvenance('remove_response', response)
trace_h2 = StationTrace(data=data[:, 3], header=stats_h2)
trace_h2.setProvenance('remove_response', response)
stream = StationStream([trace_z, trace_h1, trace_h2])
return [stream]
def read_knet(filename):
"""Read Japanese KNET strong motion file.
Args:
filename (str): Path to possible KNET data file.
kwargs (ref): Other arguments will be ignored.
Returns:
Stream: Obspy Stream containing three channels of acceleration data
(cm/s**2).
"""
logging.debug("Starting read_knet.")
if not is_knet(filename):
raise Exception('%s is not a valid KNET file' % filename)
# Parse the header portion of the file
with open(filename, 'rt') as f:
lines = [next(f) for x in range(TEXT_HDR_ROWS)]
hdr = {}
coordinates = {}
standard = {}
hdr['network'] = 'BO'
hdr['station'] = lines[5].split()[2]
logging.debug('station: %s' % hdr['station'])
standard['station_name'] = ''
# according to the powers that defined the Network.Station.Channel.Location
# "standard", Location is a two character field. Most data providers,
# including KNET here, don't provide this. We'll flag it as "--".
hdr['location'] = '--'
coordinates['latitude'] = float(lines[6].split()[2])
coordinates['longitude'] = float(lines[7].split()[2])
coordinates['elevation'] = float(lines[8].split()[2])
hdr['sampling_rate'] = float(
re.search('\\d+', lines[10].split()[2]).group())
hdr['delta'] = 1 / hdr['sampling_rate']
standard['units'] = 'acc'
dir_string = lines[12].split()[1].strip()
# knet files have directions listed as N-S, E-W, or U-D,
# whereas in kiknet those directions are '4', '5', or '6'.
if dir_string in ['N-S', '1', '4']:
hdr['channel'] = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
elif dir_string in ['E-W', '2', '5']:
hdr['channel'] = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
elif dir_string in ['U-D', '3', '6']:
hdr['channel'] = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
else:
raise Exception('KNET: Could not parse direction %s' %
lines[12].split()[1])
logging.debug('channel: %s' % hdr['channel'])
scalestr = lines[13].split()[2]
parts = scalestr.split('/')
num = float(parts[0].replace('(gal)', ''))
den = float(parts[1])
calib = num / den
hdr['calib'] = calib
duration = float(lines[11].split()[2])
hdr['npts'] = int(duration * hdr['sampling_rate'])
timestr = ' '.join(lines[9].split()[2:4])
# The K-NET and KiK-Net data logger adds a 15s time delay
# this is removed here
sttime = datetime.strptime(timestr, TIMEFMT) - timedelta(seconds=15.0)
# Shift the time to utc (Japanese time is 9 hours ahead)
sttime = sttime - timedelta(seconds=9 * 3600.)
hdr['starttime'] = sttime
# read in the data - there is a max of 8 columns per line
# the code below handles the case when last line has
# less than 8 columns
if hdr['npts'] % COLS_PER_LINE != 0:
nrows = int(np.floor(hdr['npts'] / COLS_PER_LINE))
nrows2 = 1
else:
nrows = int(np.ceil(hdr['npts'] / COLS_PER_LINE))
nrows2 = 0
data = np.genfromtxt(filename, skip_header=TEXT_HDR_ROWS,
max_rows=nrows, filling_values=np.nan)
data = data.flatten()
if nrows2:
skip_header = TEXT_HDR_ROWS + nrows
data2 = np.genfromtxt(filename, skip_header=skip_header,
max_rows=nrows2, filling_values=np.nan)
data = np.hstack((data, data2))
nrows += nrows2
# apply the correction factor we're given in the header
data *= calib
# fill out the rest of the standard dictionary
standard['horizontal_orientation'] = np.nan
standard['instrument_period'] = np.nan
standard['instrument_damping'] = np.nan
standard['process_time'] = ''
standard['process_level'] = PROCESS_LEVELS['V1']
standard['sensor_serial_number'] = ''
standard['instrument'] = ''
standard['comments'] = ''
standard['structure_type'] = ''
if dir_string in ['1', '2', '3']:
standard['structure_type'] = 'borehole'
standard['corner_frequency'] = np.nan
standard['units'] = 'acc'
standard['source'] = SRC
standard['source_format'] = 'knet'
hdr['coordinates'] = coordinates
hdr['standard'] = standard
# create a Trace from the data and metadata
trace = StationTrace(data.copy(), Stats(hdr.copy()))
response = {'input_units': 'counts', 'output_units': 'cm/s^2'}
trace.setProvenance('remove_response', response)
stream = StationStream(traces=[trace])
return [stream]
def _get_header_info(int_data, flt_data, lines, volume, location=''):
"""Return stats structure from various headers.
Output is a dictionary like this:
- network (str): 'LA'
- station (str)
- channel (str): Determined using COSMOS_ORIENTATIONS
- location (str): Default is '--'
- starttime (datetime)
- duration (float)
- sampling_rate (float)
- npts (int)
- coordinates:
- latitude (float)
- longitude (float)
- elevation (float)
- standard (Defaults are either np.nan or '')
- horizontal_orientation (float): Rotation from north (degrees)
- instrument_period (float): Period of sensor (Hz)
- instrument_damping (float): Fraction of critical
- process_time (datetime): Reported date of processing
- process_level: Either 'V0', 'V1', 'V2', or 'V3'
- station_name (str): Long form station description
- sensor_serial_number (str): Reported sensor serial
- instrument (str): See SENSOR_TYPES
- comments (str): Processing comments
- structure_type (str): See BUILDING_TYPES
- corner_frequency (float): Sensor corner frequency (Hz)
- units (str): See UNITS
- source (str): Network source description
- source_format (str): Always cosmos
- format_specific
- fractional_unit (float): Units of digitized acceleration
in file (fractions of g)
Args:
int_data (ndarray): Array of integer data
flt_data (ndarray): Array of float data
lines (list): List of text headers (str)
Returns:
dictionary: Dictionary of header/metadata information
"""
hdr = {}
coordinates = {}
standard = {}
format_specific = {}
if volume == 'V1':
hdr['duration'] = flt_data[2]
hdr['npts'] = int_data[27]
hdr['sampling_rate'] = (hdr['npts'] - 1) / hdr['duration']
# Get required parameter number
hdr['network'] = 'LA'
hdr['station'] = str(int_data[8])
logging.debug('station: %s' % hdr['station'])
horizontal_angle = int_data[26]
logging.debug('horizontal: %s' % horizontal_angle)
if (horizontal_angle in USC_ORIENTATIONS
or (horizontal_angle >= 0 and horizontal_angle <= 360)):
if horizontal_angle in USC_ORIENTATIONS:
channel = USC_ORIENTATIONS[horizontal_angle][1].upper()
if channel == 'UP' or channel == 'DOWN' or channel == 'VERT':
channel = get_channel_name(hdr['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
horizontal_angle = 0.0
elif (horizontal_angle > 315 or horizontal_angle < 45
or (horizontal_angle > 135 and horizontal_angle < 225)):
channel = get_channel_name(hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
channel = get_channel_name(hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
horizontal_orientation = horizontal_angle
hdr['channel'] = channel
logging.debug('channel: %s' % hdr['channel'])
else:
errstr = ('USC: Not enough information to distinguish horizontal '
'from vertical channels.')
raise BaseException(errstr)
if location == '':
hdr['location'] = '--'
else:
hdr['location'] = location
month = str(int_data[21])
day = str(int_data[22])
year = str(int_data[23])
time = str(int_data[24])
tstr = month + '/' + day + '/' + year + '_' + time
starttime = datetime.strptime(tstr, '%m/%d/%Y_%H%M')
hdr['starttime'] = starttime
# Get coordinates
lat_deg = int_data[9]
lat_min = int_data[10]
lat_sec = int_data[11]
lon_deg = int_data[12]
lon_min = int_data[13]
lon_sec = int_data[14]
# Check for southern hemisphere, default is northern
if lines[4].find('STATION USC#') >= 0:
idx = lines[4].find('STATION USC#') + 12
if 'S' in lines[4][idx:]:
lat_sign = -1
else:
lat_sign = 1
else:
lat_sign = 1
# Check for western hemisphere, default is western
if lines[4].find('STATION USC#') >= 0:
idx = lines[4].find('STATION USC#') + 12
if 'W' in lines[4][idx:]:
lon_sign = -1
else:
lon_sign = 1
else:
lon_sign = -1
latitude = lat_sign * _dms2dd(lat_deg, lat_min, lat_sec)
longitude = lon_sign * _dms2dd(lon_deg, lon_min, lon_sec)
# Since sometimes longitudes are positive in this format for data in
# the western hemisphere, we "fix" it here. Hopefully no one in the
# eastern hemisphere uses this format!
if longitude > 0:
longitude = -longitude
coordinates['latitude'] = latitude
coordinates['longitude'] = longitude
logging.warning('Setting elevation to 0.0')
coordinates['elevation'] = 0.0
# Get standard paramaters
standard['units_type'] = get_units_type(hdr['channel'])
standard['horizontal_orientation'] = float(horizontal_orientation)
standard['vertical_orientation'] = np.nan
standard['instrument_period'] = flt_data[0]
standard['instrument_damping'] = flt_data[1]
standard['process_time'] = ''
station_line = lines[5]
station_length = int(lines[5][72:74])
name = station_line[:station_length]
standard['station_name'] = name
standard['sensor_serial_number'] = ''
standard['instrument'] = ''
standard['comments'] = ''
standard['units'] = 'acc'
standard['structure_type'] = ''
standard['process_level'] = PROCESS_LEVELS['V1']
standard['corner_frequency'] = np.nan
standard['source'] = ('Los Angeles Basin Seismic Network, University '
'of Southern California')
standard['source_format'] = 'usc'
# this field can be used for instrument correction
# when data is in counts
standard['instrument_sensitivity'] = np.nan
# Get format specific
format_specific['fractional_unit'] = flt_data[4]
# Set dictionary
hdr['standard'] = standard
hdr['coordinates'] = coordinates
hdr['format_specific'] = format_specific
return hdr
def read_cwb(filename, **kwargs):
"""Read Taiwan Central Weather Bureau strong motion file.
Args:
filename (str): Path to possible CWB data file.
kwargs (ref): Other arguments will be ignored.
Returns:
Stream: Obspy Stream containing three channels of acceleration
data (cm/s**2).
"""
logging.debug("Starting read_cwb.")
if not is_cwb(filename):
raise Exception('%s is not a valid CWB strong motion data file.'
% filename)
f = open(filename, 'rt')
# according to the powers that defined the Network.Station.Channel.Location
# "standard", Location is a two character field. Most data providers,
# including CWB here, don't provide this. We'll flag it as "--".
data = np.genfromtxt(filename, skip_header=HDR_ROWS,
delimiter=[COLWIDTH] * NCOLS) # time, Z, NS, EW
hdr = _get_header_info(f, data)
f.close()
head, tail = os.path.split(filename)
hdr['standard']['source_file'] = tail or os.path.basename(head)
hdr_z = hdr.copy()
hdr_z['channel'] = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
hdr_z['standard']['horizontal_orientation'] = np.nan
hdr_h1 = hdr.copy()
hdr_h1['channel'] = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
hdr_h1['standard']['horizontal_orientation'] = np.nan
hdr_h2 = hdr.copy()
hdr_h2['channel'] = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
hdr_h2['standard']['horizontal_orientation'] = np.nan
stats_z = Stats(hdr_z)
stats_h1 = Stats(hdr_h1)
stats_h2 = Stats(hdr_h2)
response = {'input_units': 'counts', 'output_units': 'cm/s^2'}
trace_z = StationTrace(data=data[:, 1], header=stats_z)
trace_z.setProvenance('remove_response', response)
trace_h1 = StationTrace(data=data[:, 2], header=stats_h1)
trace_h1.setProvenance('remove_response', response)
trace_h2 = StationTrace(data=data[:, 3], header=stats_h2)
trace_h2.setProvenance('remove_response', response)
stream = StationStream([trace_z, trace_h1, trace_h2])
return [stream]
def _get_header_info(int_data, flt_data, lines, volume, location=""):
"""Return stats structure from various headers.
Output is a dictionary like this:
- network (str): 'LA'
- station (str)
- channel (str): Determined using COSMOS_ORIENTATIONS
- location (str): Default is '--'
- starttime (datetime)
- duration (float)
- sampling_rate (float)
- npts (int)
- coordinates:
- latitude (float)
- longitude (float)
- elevation (float)
- standard (Defaults are either np.nan or '')
- horizontal_orientation (float): Rotation from north (degrees)
- instrument_period (float): Period of sensor (Hz)
- instrument_damping (float): Fraction of critical
- process_time (datetime): Reported date of processing
- process_level: Either 'V0', 'V1', 'V2', or 'V3'
- station_name (str): Long form station description
- sensor_serial_number (str): Reported sensor serial
- instrument (str): See SENSOR_TYPES
- comments (str): Processing comments
- structure_type (str): See BUILDING_TYPES
- corner_frequency (float): Sensor corner frequency (Hz)
- units (str): See UNITS
- source (str): Network source description
- source_format (str): Always cosmos
- format_specific
- fractional_unit (float): Units of digitized acceleration
in file (fractions of g)
Args:
int_data (ndarray): Array of integer data
flt_data (ndarray): Array of float data
lines (list): List of text headers (str)
Returns:
dictionary: Dictionary of header/metadata information
"""
hdr = {}
coordinates = {}
standard = {}
format_specific = {}
if volume == "V1":
hdr["duration"] = flt_data[2]
hdr["npts"] = int_data[27]
hdr["sampling_rate"] = (hdr["npts"] - 1) / hdr["duration"]
# Get required parameter number
hdr["network"] = "LA"
hdr["station"] = str(int_data[8])
logging.debug(f"station: {hdr['station']}")
horizontal_angle = int_data[26]
logging.debug(f"horizontal: {horizontal_angle}")
if horizontal_angle in USC_ORIENTATIONS or (horizontal_angle >= 0 and
horizontal_angle <= 360):
if horizontal_angle in USC_ORIENTATIONS:
channel = USC_ORIENTATIONS[horizontal_angle][1].upper()
if channel == "UP" or channel == "DOWN" or channel == "VERT":
channel = get_channel_name(
hdr["sampling_rate"],
is_acceleration=True,
is_vertical=True,
is_north=False,
)
horizontal_angle = 0.0
elif (horizontal_angle > 315 or horizontal_angle < 45
or (horizontal_angle > 135 and horizontal_angle < 225)):
channel = get_channel_name(
hdr["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=True,
)
else:
channel = get_channel_name(
hdr["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=False,
)
horizontal_orientation = horizontal_angle
hdr["channel"] = channel
logging.debug(f"channel: {hdr['channel']}")
else:
errstr = ("USC: Not enough information to distinguish horizontal "
"from vertical channels.")
raise BaseException(errstr)
if location == "":
hdr["location"] = "--"
else:
hdr["location"] = location
month = str(int_data[21])
day = str(int_data[22])
year = str(int_data[23])
time = str(int_data[24])
tstr = month + "/" + day + "/" + year + "_" + time
starttime = datetime.strptime(tstr, "%m/%d/%Y_%H%M")
hdr["starttime"] = starttime
# Get coordinates
lat_deg = int_data[9]
lat_min = int_data[10]
lat_sec = int_data[11]
lon_deg = int_data[12]
lon_min = int_data[13]
lon_sec = int_data[14]
# Check for southern hemisphere, default is northern
if lines[4].find("STATION USC#") >= 0:
idx = lines[4].find("STATION USC#") + 12
if "S" in lines[4][idx:]:
lat_sign = -1
else:
lat_sign = 1
else:
lat_sign = 1
# Check for western hemisphere, default is western
if lines[4].find("STATION USC#") >= 0:
idx = lines[4].find("STATION USC#") + 12
if "W" in lines[4][idx:]:
lon_sign = -1
else:
lon_sign = 1
else:
lon_sign = -1
latitude = lat_sign * _dms2dd(lat_deg, lat_min, lat_sec)
longitude = lon_sign * _dms2dd(lon_deg, lon_min, lon_sec)
# Since sometimes longitudes are positive in this format for data in
# the western hemisphere, we "fix" it here. Hopefully no one in the
# eastern hemisphere uses this format!
if longitude > 0:
longitude = -longitude
coordinates["latitude"] = latitude
coordinates["longitude"] = longitude
logging.warning("Setting elevation to 0.0")
coordinates["elevation"] = 0.0
# Get standard paramaters
standard["units_type"] = get_units_type(hdr["channel"])
standard["horizontal_orientation"] = float(horizontal_orientation)
standard["vertical_orientation"] = np.nan
standard["instrument_period"] = flt_data[0]
standard["instrument_damping"] = flt_data[1]
standard["process_time"] = ""
station_line = lines[5]
station_length = int(lines[5][72:74])
name = station_line[:station_length]
standard["station_name"] = name
standard["sensor_serial_number"] = ""
standard["instrument"] = ""
standard["comments"] = ""
standard["units"] = "cm/s/s"
standard["structure_type"] = ""
standard["process_level"] = PROCESS_LEVELS["V1"]
standard["corner_frequency"] = np.nan
standard[
"source"] = "Los Angeles Basin Seismic Network, University of Southern California"
standard["source_format"] = "usc"
# these fields can be used for instrument correction
# when data is in counts
standard["instrument_sensitivity"] = np.nan
standard["volts_to_counts"] = np.nan
# Get format specific
format_specific["fractional_unit"] = flt_data[4]
# Set dictionary
hdr["standard"] = standard
hdr["coordinates"] = coordinates
hdr["format_specific"] = format_specific
return hdr
def read_knet(filename, config=None, **kwargs):
"""Read Japanese KNET strong motion file.
Args:
filename (str):
Path to possible KNET data file.
config (dict):
Dictionary containing configuration.
kwargs (ref):
Other arguments will be ignored.
Returns:
Stream: Obspy Stream containing three channels of acceleration data
(cm/s**2).
"""
logging.debug("Starting read_knet.")
if not is_knet(filename, config):
raise Exception(f"{filename} is not a valid KNET file")
# Parse the header portion of the file
with open(filename, "rt") as f:
lines = [next(f) for x in range(TEXT_HDR_ROWS)]
hdr = {}
coordinates = {}
standard = {}
hdr["network"] = "BO"
hdr["station"] = lines[5].split()[2]
logging.debug(f"station: {hdr['station']}")
standard["station_name"] = ""
# according to the powers that defined the Network.Station.Channel.Location
# "standard", Location is a two character field. Most data providers,
# including KNET here, don't provide this. We'll flag it as "--".
hdr["location"] = "--"
coordinates["latitude"] = float(lines[6].split()[2])
coordinates["longitude"] = float(lines[7].split()[2])
coordinates["elevation"] = float(lines[8].split()[2])
hdr["sampling_rate"] = float(
re.search("\\d+", lines[10].split()[2]).group())
hdr["delta"] = 1 / hdr["sampling_rate"]
standard["units_type"] = "acc"
standard["units_type"] = "cm/s/s"
dir_string = lines[12].split()[1].strip()
# knet files have directions listed as N-S, E-W, or U-D,
# whereas in kiknet those directions are '4', '5', or '6'.
if dir_string in ["N-S", "1", "4"]:
hdr["channel"] = get_channel_name(hdr["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=True)
elif dir_string in ["E-W", "2", "5"]:
hdr["channel"] = get_channel_name(
hdr["sampling_rate"],
is_acceleration=True,
is_vertical=False,
is_north=False,
)
elif dir_string in ["U-D", "3", "6"]:
hdr["channel"] = get_channel_name(hdr["sampling_rate"],
is_acceleration=True,
is_vertical=True,
is_north=False)
else:
raise Exception(
f"KNET: Could not parse direction {lines[12].split()[1]}")
logging.debug(f"channel: {hdr['channel']}")
scalestr = lines[13].split()[2]
parts = scalestr.split("/")
num = float(parts[0].replace("(gal)", ""))
den = float(parts[1])
calib = num / den
hdr["calib"] = calib
duration = float(lines[11].split()[2])
hdr["npts"] = int(duration * hdr["sampling_rate"])
timestr = " ".join(lines[9].split()[2:4])
# The K-NET and KiK-Net data logger adds a 15s time delay
# this is removed here
sttime = datetime.strptime(timestr, TIMEFMT) - timedelta(seconds=15.0)
# Shift the time to utc (Japanese time is 9 hours ahead)
sttime = sttime - timedelta(seconds=9 * 3600.0)
hdr["starttime"] = sttime
# read in the data - there is a max of 8 columns per line
# the code below handles the case when last line has
# less than 8 columns
if hdr["npts"] % COLS_PER_LINE != 0:
nrows = int(np.floor(hdr["npts"] / COLS_PER_LINE))
nrows2 = 1
else:
nrows = int(np.ceil(hdr["npts"] / COLS_PER_LINE))
nrows2 = 0
data = np.genfromtxt(filename,
skip_header=TEXT_HDR_ROWS,
max_rows=nrows,
filling_values=np.nan)
data = data.flatten()
if nrows2:
skip_header = TEXT_HDR_ROWS + nrows
data2 = np.genfromtxt(filename,
skip_header=skip_header,
max_rows=nrows2,
filling_values=np.nan)
data = np.hstack((data, data2))
nrows += nrows2
# apply the correction factor we're given in the header
data *= calib
# fill out the rest of the standard dictionary
standard["units_type"] = get_units_type(hdr["channel"])
standard["horizontal_orientation"] = np.nan
standard["vertical_orientation"] = np.nan
standard["instrument_period"] = np.nan
standard["instrument_damping"] = np.nan
standard["process_time"] = ""
standard["process_level"] = PROCESS_LEVELS["V1"]
standard["sensor_serial_number"] = ""
standard["instrument"] = ""
standard["comments"] = ""
standard["structure_type"] = ""
if dir_string in ["1", "2", "3"]:
standard["structure_type"] = "borehole"
standard["corner_frequency"] = np.nan
standard["units"] = "acc"
standard["source"] = SRC
standard["source_format"] = "knet"
head, tail = os.path.split(filename)
standard["source_file"] = tail or os.path.basename(head)
# these fields can be used for instrument correction
# when data is in counts
standard["instrument_sensitivity"] = np.nan
standard["volts_to_counts"] = np.nan
hdr["coordinates"] = coordinates
hdr["standard"] = standard
# create a Trace from the data and metadata
trace = StationTrace(data.copy(), Stats(hdr.copy()))
response = {"input_units": "counts", "output_units": "cm/s^2"}
trace.setProvenance("remove_response", response)
stream = StationStream(traces=[trace])
return [stream]
def _get_header_info(filename,
any_structure=False,
accept_flagged=False,
location=''):
"""Return stats structure from various headers.
Output is a dictionary like this:
- network
- station
- channel
- location (str): Set to floor the sensor is located on. If not a
multi-sensor array, default is '--'. Can be set manually by
the user.
- starttime
- sampling_rate
- npts
- coordinates:
- latitude
- longitude
- elevation
- standard
- horizontal_orientation
- instrument_period
- instrument_damping
- process_level
- station_name
- sensor_serial_number
- instrument
- comments
- structure_type
- corner_frequency
- units
- source
- source_format
- format_specific
- vertical_orientation
- building_floor (0=basement, 1=floor above basement, -1=1st sub-basement, etc.
- bridge_number_spans
- bridge_transducer_location ("free field",
"at the base of a pier or abutment",
"on an abutment",
"on the deck at the top of a pier"
"on the deck between piers or between an abutment and a pier."
dam_transducer_location ("upstream or downstream free field",
"at the base of the dam",
"on the crest of the dam",
on the abutment of the dam")
construction_type ("Reinforced concrete gravity",
"Reinforced concrete arch",
"earth fill",
"other")
filter_poles
data_source
"""
stats = {}
standard = {}
format_specific = {}
coordinates = {}
# read the ascii header lines
with open(filename) as f:
ascheader = [next(f).strip() for x in range(ASCII_HEADER_LINES)]
standard['process_level'] = PROCESS_LEVELS[VALID_HEADERS[ascheader[0]]]
logging.debug("process_level: %s" % standard['process_level'])
# station code is in the third line
stats['station'] = ''
if len(ascheader[2]) >= 4:
stats['station'] = ascheader[2][0:4].strip()
logging.debug('station: %s' % stats['station'])
standard['process_time'] = ''
standard['station_name'] = ascheader[5][10:40].strip()
# sometimes the data source has nothing in it,
# most of the time it seems has has USGS in it
# sometimes it's something like JPL/USGS, CDOT/USGS, etc.
# if it's got USGS in it, let's just say network=US, otherwise "--"
stats['network'] = 'ZZ'
if ascheader[7].find('USGS') > -1:
stats['network'] = 'US'
try:
standard['source'] = ascheader[7].split('=')[2].strip()
except IndexError:
standard['source'] = 'USGS'
if standard['source'] == '':
standard['source'] = 'USGS'
standard['source_format'] = 'smc'
# read integer header data
intheader = np.genfromtxt(filename,
dtype=np.int32,
max_rows=INTEGER_HEADER_LINES,
skip_header=ASCII_HEADER_LINES,
delimiter=INT_HEADER_WIDTHS)
# 8 columns per line
# first line is start time information, and then inst. serial number
missing_data = intheader[0, 0]
year = intheader[0, 1]
jday = intheader[0, 2]
hour = intheader[0, 3]
minute = intheader[0, 4]
if (year != missing_data and jday != missing_data and hour != missing_data
and minute != missing_data):
# Handle second if missing
second = 0
if not intheader[0, 5] == missing_data:
second = intheader[0, 5]
# Handle microsecond if missing and convert milliseconds to microseconds
microsecond = 0
if not intheader[0, 6] == missing_data:
microsecond = intheader[0, 6] / 1e3
datestr = '%i %00i %i %i %i %i' % (year, jday, hour, minute, second,
microsecond)
stats['starttime'] = datetime.strptime(datestr, '%Y %j %H %M %S %f')
else:
logging.warning('No start time provided. '
'This must be set manually for network/station: '
'%s/%s.' % (stats['network'], stats['station']))
standard['comments'] = 'Missing start time.'
standard['sensor_serial_number'] = ''
if intheader[1, 3] != missing_data:
standard['sensor_serial_number'] = str(intheader[1, 3])
# we never get a two character location code so floor location is used
if location == '':
location = intheader.flatten()[24]
if location != missing_data:
location = str(location)
if len(location) < 2:
location = location.zfill(2)
stats['location'] = location
else:
stats['location'] = '--'
else:
stats['location'] = location
# second line is information about number of channels, orientations
# we care about orientations
format_specific['vertical_orientation'] = np.nan
if intheader[1, 4] != missing_data:
format_specific['vertical_orientation'] = int(intheader[1, 4])
standard['horizontal_orientation'] = np.nan
if intheader[1, 5] != missing_data:
standard['horizontal_orientation'] = float(intheader[1, 5])
if intheader[1, 6] == missing_data or intheader[1, 6] not in INSTRUMENTS:
standard['instrument'] = ''
else:
standard['instrument'] = INSTRUMENTS[intheader[1, 6]]
num_comments = intheader[1, 7]
# third line contains number of data points
stats['npts'] = intheader[2, 0]
problem_flag = intheader[2, 1]
if problem_flag == 1:
if not accept_flagged:
fmt = 'SMC: Record found in file %s has a problem flag!'
raise Exception(fmt % filename)
else:
logging.warning(
'SMC: Data contains a problem flag for network/station: '
'%s/%s. See comments.' % (stats['network'], stats['station']))
stype = intheader[2, 2]
if stype == missing_data:
stype = np.nan
elif stype not in STRUCTURES:
# structure type is not defined and should will be considered 'other'
stype = 4
fmt = 'SMC: Record found in file %s is not a free-field sensor!'
standard['structure_type'] = STRUCTURES[stype]
if standard['structure_type'] == 'building' and not any_structure:
raise Exception(fmt % filename)
format_specific['building_floor'] = np.nan
if intheader[3, 0] != missing_data:
format_specific['building_floor'] = intheader[3, 0]
format_specific['bridge_number_spans'] = np.nan
if intheader[3, 1] != missing_data:
format_specific['bridge_number_spans'] = intheader[3, 1]
format_specific['bridge_transducer_location'] = BRIDGE_LOCATIONS[0]
if intheader[3, 2] != missing_data:
bridge_number = intheader[3, 2]
format_specific['bridge_transducer_location'] = \
BRIDGE_LOCATIONS[bridge_number]
format_specific['dam_transducer_location'] = DAM_LOCATIONS[0]
if intheader[3, 3] != missing_data:
dam_number = intheader[3, 3]
format_specific['dam_transducer_location'] = DAM_LOCATIONS[dam_number]
c1 = format_specific['bridge_transducer_location'].find('free field') == -1
c2 = format_specific['dam_transducer_location'].find('free field') == -1
if (c1 or c2) and not any_structure:
raise Exception(fmt % filename)
format_specific['construction_type'] = CONSTRUCTION_TYPES[4]
if intheader[3, 4] != missing_data:
format_specific['construction_type'] = \
CONSTRUCTION_TYPES[intheader[3, 4]]
# station is repeated here if all numeric
if not len(stats['station']):
stats['station'] = '%i' % intheader[3, 5]
# read float header data
skip = ASCII_HEADER_LINES + INTEGER_HEADER_LINES
floatheader = np.genfromtxt(filename,
max_rows=FLOAT_HEADER_LINES,
skip_header=skip,
delimiter=FLOAT_HEADER_WIDTHS)
# float headers are 10 lines of 5 floats each
missing_data = floatheader[0, 0]
stats['sampling_rate'] = floatheader[0, 1]
coordinates['latitude'] = floatheader[2, 0]
# the documentation for SMC says that sometimes longitudes are
# positive in the western hemisphere. Since it is very unlikely
# any of these files exist for the eastern hemisphere, check for
# positive longitudes and fix them.
lon = floatheader[2, 1]
if lon > 0:
lon = -1 * lon
coordinates['longitude'] = lon
coordinates['elevation'] = 0
if floatheader[2, 2] != missing_data:
coordinates['elevation'] = floatheader[2, 2]
# figure out the channel code
if format_specific['vertical_orientation'] in [0, 180]:
stats['channel'] = get_channel_name(stats['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
else:
ho = standard['horizontal_orientation']
quad1 = ho > 315 and ho <= 360
quad2 = ho > 0 and ho <= 45
quad3 = ho > 135 and ho <= 225
if quad1 or quad2 or quad3:
stats['channel'] = get_channel_name(stats['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
stats['channel'] = get_channel_name(stats['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
logging.debug('channel: %s' % stats['channel'])
sensor_frequency = floatheader[4, 1]
standard['instrument_period'] = 1 / sensor_frequency
standard['instrument_damping'] = floatheader[4, 2]
standard['corner_frequency'] = floatheader[3, 4]
format_specific['filter_poles'] = floatheader[4, 0]
standard['units'] = 'acc'
# read in the comment lines
with open(filename) as f:
skip = ASCII_HEADER_LINES + INTEGER_HEADER_LINES + FLOAT_HEADER_LINES
_ = [next(f) for x in range(skip)]
standard['comments'] = [
next(f).strip().lstrip('|') for x in range(num_comments)
]
standard['comments'] = ' '.join(standard['comments'])
stats['coordinates'] = coordinates
stats['standard'] = standard
stats['format_specific'] = format_specific
head, tail = os.path.split(filename)
stats['standard']['source_file'] = tail or os.path.basename(head)
return (stats, num_comments)
def add_channel_metadata(tr, inv, client):
"""
Adds the channel metadata for each channel in the stream.
Args:
tr (obspy.core.trace.Trace):
Trace of requested data.
inv (obspy.core.inventory):
Inventory object corresponding to to the stream.
client (str):
FDSN client indicator.
Returns:
trace (obspy.core.trace.Trace): Trace with metadata added.
"""
time = tr.stats.starttime
id_string = tr.stats.network + '.' + tr.stats.station + '.'
id_string += tr.stats.location + '.' + tr.stats.channel
if tr.stats.location == '':
tr.stats.location = '--'
metadata = inv.get_channel_metadata(id_string, time)
coordinates = {
'latitude': metadata['latitude'],
'longitude': metadata['longitude'],
'elevation': metadata['elevation']
}
standard = {
'horizontal_orientation': metadata['azimuth'],
'instrument_period': np.nan,
'instrument_damping': np.nan,
'process_level': 'V0',
'station_name': tr.stats.station,
'sensor_serial_number': '',
'instrument': '',
'comments': '',
'structure_type': '',
'corner_frequency': np.nan,
'units': 'raw',
'source': client,
'source_format': 'fdsn'
}
tr.stats['coordinates'] = coordinates
tr.stats['standard'] = standard
if metadata['dip'] in [90, -90, 180, -180]:
tr.stats['channel'] = get_channel_name(
tr.stats['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
else:
ho = metadata['azimuth']
quad1 = ho > 315 and ho <= 360
quad2 = ho >= 0 and ho <= 45
quad3 = ho > 135 and ho <= 225
if quad1 or quad2 or quad3:
tr.stats['channel'] = get_channel_name(
tr.stats['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
tr.stats['channel'] = get_channel_name(
tr.stats['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
return tr
def _read_header(filename):
# read in first 88 lines
with open(filename, "rt") as myfile:
header = [next(myfile) for x in range(HEADER_LINES)]
header_dict = {}
lastkey = ""
for line in header:
if not len(line.strip()):
continue
if ":" in line:
colidx = line.find(":")
key = line[0:colidx].strip()
value = line[colidx + 1:].strip()
if not len(key):
key = lastkey + "$"
lastkey = key
header_dict[key] = value
# create a list of dictionaries for channel spec
channels = [{}, {}, {}]
channels[0]["standard"] = {}
channels[1]["standard"] = {}
channels[2]["standard"] = {}
station = header_dict["CLAVE DE LA ESTACION"]
channels[0]["station"] = station
channels[1]["station"] = station
channels[2]["station"] = station
channels[0]["network"] = NETWORK
channels[1]["network"] = NETWORK
channels[2]["network"] = NETWORK
# unam provides the start *time* of the record, but not the date.
# it is up to us to determine whether midnight has occurred between
# eq time and record start time.
# the hour/min/sec of trigger time
(rhour, rminute,
rsecond) = header_dict["HORA DE LA PRIMERA MUESTRA (GMT)"].split(":")
dtsecs = (int(rhour) * 3600) + (int(rminute) * 60) + (float(rsecond))
startdt = timedelta(seconds=dtsecs)
eqdatestr = header_dict["FECHA DEL SISMO [GMT]"]
eqdate = datetime.strptime(eqdatestr, "%Y/%m/%d")
# the hour, minute and second of origin
eqtimestr = header_dict["HORA EPICENTRO (GMT)"]
try:
eqtime = datetime.strptime(f"{eqdatestr} {eqtimestr}", TIMEFMT1)
except ValueError:
eqtime = datetime.strptime(f"{eqdatestr} {eqtimestr}", TIMEFMT2)
# if the origin time and record start time are more than 10 minutes
# apart (in either direction), then assume that we need to add 1 day
# to the record start time.
starttime = eqdate + startdt
dt = np.abs((starttime - eqtime).total_seconds())
if dt > MAX_TIME_DIFF:
starttime = eqdate + timedelta(days=1) + startdt
channels[0]["starttime"] = starttime
channels[1]["starttime"] = starttime
channels[2]["starttime"] = starttime
# get record durations for each channel
durstr = header_dict["DURACION DEL REGISTRO (s), C1-C6"].lstrip("/")
durations = [float(dur) for dur in durstr.split("/")]
channels[0]["duration"] = durations[0]
channels[1]["duration"] = durations[1]
channels[2]["duration"] = durations[2]
# get deltas
delta_strings = header_dict["INTERVALO DE MUESTREO, C1-C6 (s)"].split("/")
deltas = [float(delta) for delta in delta_strings[1:]]
channels[0]["delta"] = deltas[0]
channels[1]["delta"] = deltas[1]
channels[2]["delta"] = deltas[2]
# get sampling rates
channels[0]["sampling_rate"] = 1 / deltas[0]
channels[1]["sampling_rate"] = 1 / deltas[1]
channels[2]["sampling_rate"] = 1 / deltas[2]
# get channel orientations
azstrings = header_dict["ORIENTACION C1-C6 (rumbo;orientacion)"].split("/")
az1, az1_vert = _get_azimuth(azstrings[1])
az2, az2_vert = _get_azimuth(azstrings[2])
az3, az3_vert = _get_azimuth(azstrings[3])
channels[0]["standard"]["horizontal_orientation"] = az1
channels[1]["standard"]["horizontal_orientation"] = az2
channels[2]["standard"]["horizontal_orientation"] = az3
channels[0]["standard"]["vertical_orientation"] = np.nan
channels[1]["standard"]["vertical_orientation"] = np.nan
channels[2]["standard"]["vertical_orientation"] = np.nan
az1_north = is_channel_north(az1)
az2_north = is_channel_north(az2)
az3_north = is_channel_north(az3)
channels[0]["channel"] = get_channel_name(channels[0]["sampling_rate"],
True, az1_vert, az1_north)
channels[1]["channel"] = get_channel_name(channels[1]["sampling_rate"],
True, az2_vert, az2_north)
channels[2]["channel"] = get_channel_name(channels[2]["sampling_rate"],
True, az3_vert, az3_north)
# get channel npts
npts_strings = header_dict["NUM. TOTAL DE MUESTRAS, C1-C6"].split("/")
npts_list = [float(npts) for npts in npts_strings[1:]]
channels[0]["npts"] = npts_list[0]
channels[1]["npts"] = npts_list[1]
channels[2]["npts"] = npts_list[2]
# locations
channels[0]["location"] = "--"
channels[1]["location"] = "--"
channels[1]["location"] = "--"
# get station coordinates
coord1 = header_dict["COORDENADAS DE LA ESTACION"]
coord2 = header_dict["COORDENADAS DE LA ESTACION$"]
if "LAT" in coord1:
latitude = float(re.search(FLOATRE, coord1).group())
longitude = float(re.search(FLOATRE, coord2).group())
if coord1.strip().endswith("S"):
latitude *= -1
if coord2.strip().endswith("W"):
longitude *= -1
else:
latitude = re.search(FLOATRE, coord2)
longitude = re.search(FLOATRE, coord1)
if coord1.strip().endswith("W"):
longitude *= -1
if coord2.strip().endswith("S"):
latitude *= -1
elevation = float(header_dict["ALTITUD (msnm)"])
cdict = {
"latitude": latitude,
"longitude": longitude,
"elevation": elevation
}
channels[0]["coordinates"] = cdict
channels[1]["coordinates"] = cdict
channels[2]["coordinates"] = cdict
# fill in other standard stuff
standard0 = channels[0]["standard"]
standard1 = channels[1]["standard"]
standard2 = channels[2]["standard"]
standard0["units_type"] = "acc"
standard1["units_type"] = "acc"
standard2["units_type"] = "acc"
standard0["source_format"] = SOURCE_FORMAT
standard1["source_format"] = SOURCE_FORMAT
standard2["source_format"] = SOURCE_FORMAT
standard0["instrument"] = header_dict["MODELO DEL ACELEROGRAFO"]
standard1["instrument"] = header_dict["MODELO DEL ACELEROGRAFO"]
standard2["instrument"] = header_dict["MODELO DEL ACELEROGRAFO"]
standard0["sensor_serial_number"] = header_dict[
"NUMERO DE SERIE DEL ACELEROGRAFO"]
standard1["sensor_serial_number"] = header_dict[
"NUMERO DE SERIE DEL ACELEROGRAFO"]
standard2["sensor_serial_number"] = header_dict[
"NUMERO DE SERIE DEL ACELEROGRAFO"]
standard0["process_level"] = PROCESS_LEVELS["V1"]
standard1["process_level"] = PROCESS_LEVELS["V1"]
standard2["process_level"] = PROCESS_LEVELS["V1"]
standard0["process_time"] = ""
standard1["process_time"] = ""
standard2["process_time"] = ""
standard0["station_name"] = header_dict["NOMBRE DE LA ESTACION"]
standard1["station_name"] = header_dict["NOMBRE DE LA ESTACION"]
standard2["station_name"] = header_dict["NOMBRE DE LA ESTACION"]
standard0["structure_type"] = ""
standard1["structure_type"] = ""
standard2["structure_type"] = ""
standard0["corner_frequency"] = np.nan
standard1["corner_frequency"] = np.nan
standard2["corner_frequency"] = np.nan
standard0["units"] = "cm/s/s"
standard1["units"] = "cm/s/s"
standard2["units"] = "cm/s/s"
periods = _get_periods(header_dict["FREC. NAT. DE SENSORES, C1-C6, (Hz)"])
standard0["instrument_period"] = periods[0]
standard1["instrument_period"] = periods[1]
standard2["instrument_period"] = periods[2]
dampings = _get_dampings(header_dict["AMORTIGUAMIENTO DE SENSORES, C1-C6"])
standard0["instrument_damping"] = dampings[0]
standard1["instrument_damping"] = dampings[1]
standard2["instrument_damping"] = dampings[2]
with open(filename, "rt") as myfile:
header = [next(myfile) for x in range(HEADER_PLUS_COMMENT)]
clines = header[89:102]
comments = " ".join(clines).strip()
standard0["comments"] = comments
standard1["comments"] = comments
standard2["comments"] = comments
head, tail = os.path.split(filename)
source_file = tail or os.path.basename(head)
standard0["source_file"] = source_file
standard1["source_file"] = source_file
standard2["source_file"] = source_file
standard0["source"] = SOURCE
standard1["source"] = SOURCE
standard2["source"] = SOURCE
decfactor = float(header_dict["FACTOR DE DECIMACION"])
standard0["instrument_sensitivity"] = decfactor
standard1["instrument_sensitivity"] = decfactor
standard2["instrument_sensitivity"] = decfactor
standard0["volts_to_counts"] = np.nan
standard1["volts_to_counts"] = np.nan
standard2["volts_to_counts"] = np.nan
return channels
def _read_header(filename):
# read in first 88 lines
with open(filename, 'rt') as myfile:
header = [next(myfile) for x in range(HEADER_LINES)]
header_dict = {}
lastkey = ''
for line in header:
if not len(line.strip()):
continue
if ':' in line:
colidx = line.find(':')
key = line[0:colidx].strip()
value = line[colidx + 1:].strip()
if not len(key):
key = lastkey + '$'
lastkey = key
header_dict[key] = value
# create a list of dictionaries for channel spec
channels = [{}, {}, {}]
channels[0]['standard'] = {}
channels[1]['standard'] = {}
channels[2]['standard'] = {}
station = header_dict['CLAVE DE LA ESTACION']
channels[0]['station'] = station
channels[1]['station'] = station
channels[2]['station'] = station
channels[0]['network'] = NETWORK
channels[1]['network'] = NETWORK
channels[2]['network'] = NETWORK
# unam provides the start *time* of the record, but not the date.
# it is up to us to determine whether midnight has occurred between
# eq time and record start time.
# the hour/min/sec of trigger time
(rhour, rminute,
rsecond) = header_dict['HORA DE LA PRIMERA MUESTRA (GMT)'].split(':')
dtsecs = (int(rhour) * 3600) + (int(rminute) * 60) + (float(rsecond))
startdt = timedelta(seconds=dtsecs)
eqdatestr = header_dict['FECHA DEL SISMO [GMT]']
eqdate = datetime.strptime(eqdatestr, '%Y/%m/%d')
# the hour, minute and second of origin
eqtimestr = header_dict['HORA EPICENTRO (GMT)']
try:
eqtime = datetime.strptime('%s %s' % (eqdatestr, eqtimestr), TIMEFMT1)
except ValueError:
eqtime = datetime.strptime('%s %s' % (eqdatestr, eqtimestr), TIMEFMT2)
# if the origin time and record start time are more than 10 minutes
# apart (in either direction), then assume that we need to add 1 day
# to the record start time.
starttime = eqdate + startdt
dt = np.abs((starttime - eqtime).total_seconds())
if dt > MAX_TIME_DIFF:
starttime = eqdate + timedelta(days=1) + startdt
channels[0]['starttime'] = starttime
channels[1]['starttime'] = starttime
channels[2]['starttime'] = starttime
# get record durations for each channel
durstr = header_dict['DURACION DEL REGISTRO (s), C1-C6'].lstrip('/')
durations = [float(dur) for dur in durstr.split('/')]
channels[0]['duration'] = durations[0]
channels[1]['duration'] = durations[1]
channels[2]['duration'] = durations[2]
# get deltas
delta_strings = header_dict['INTERVALO DE MUESTREO, C1-C6 (s)'].split('/')
deltas = [float(delta) for delta in delta_strings[1:]]
channels[0]['delta'] = deltas[0]
channels[1]['delta'] = deltas[1]
channels[2]['delta'] = deltas[2]
# get sampling rates
channels[0]['sampling_rate'] = 1 / deltas[0]
channels[1]['sampling_rate'] = 1 / deltas[1]
channels[2]['sampling_rate'] = 1 / deltas[2]
# get channel orientations
azstrings = header_dict['ORIENTACION C1-C6 (rumbo;orientacion)'].split('/')
az1, az1_vert = _get_azimuth(azstrings[1])
az2, az2_vert = _get_azimuth(azstrings[2])
az3, az3_vert = _get_azimuth(azstrings[3])
channels[0]['standard']['horizontal_orientation'] = az1
channels[1]['standard']['horizontal_orientation'] = az2
channels[2]['standard']['horizontal_orientation'] = az3
channels[0]['standard']['vertical_orientation'] = np.nan
channels[1]['standard']['vertical_orientation'] = np.nan
channels[2]['standard']['vertical_orientation'] = np.nan
az1_north = is_channel_north(az1)
az2_north = is_channel_north(az2)
az3_north = is_channel_north(az3)
channels[0]['channel'] = get_channel_name(channels[0]['sampling_rate'],
True, az1_vert, az1_north)
channels[1]['channel'] = get_channel_name(channels[1]['sampling_rate'],
True, az2_vert, az2_north)
channels[2]['channel'] = get_channel_name(channels[2]['sampling_rate'],
True, az3_vert, az3_north)
# get channel npts
npts_strings = header_dict['NUM. TOTAL DE MUESTRAS, C1-C6'].split('/')
npts_list = [float(npts) for npts in npts_strings[1:]]
channels[0]['npts'] = npts_list[0]
channels[1]['npts'] = npts_list[1]
channels[2]['npts'] = npts_list[2]
# locations
channels[0]['location'] = '--'
channels[1]['location'] = '--'
channels[1]['location'] = '--'
# get station coordinates
coord1 = header_dict['COORDENADAS DE LA ESTACION']
coord2 = header_dict['COORDENADAS DE LA ESTACION$']
if 'LAT' in coord1:
latitude = float(re.search(FLOATRE, coord1).group())
longitude = float(re.search(FLOATRE, coord2).group())
if coord1.strip().endswith('S'):
latitude *= -1
if coord2.strip().endswith('W'):
longitude *= -1
else:
latitude = re.search(FLOATRE, coord2)
longitude = re.search(FLOATRE, coord1)
if coord1.strip().endswith('W'):
longitude *= -1
if coord2.strip().endswith('S'):
latitude *= -1
elevation = float(header_dict['ALTITUD (msnm)'])
cdict = {
'latitude': latitude,
'longitude': longitude,
'elevation': elevation
}
channels[0]['coordinates'] = cdict
channels[1]['coordinates'] = cdict
channels[2]['coordinates'] = cdict
# fill in other standard stuff
channels[0]['standard']['units_type'] = 'acc'
channels[1]['standard']['units_type'] = 'acc'
channels[2]['standard']['units_type'] = 'acc'
channels[0]['standard']['source_format'] = SOURCE_FORMAT
channels[1]['standard']['source_format'] = SOURCE_FORMAT
channels[2]['standard']['source_format'] = SOURCE_FORMAT
channels[0]['standard']['instrument'] = header_dict[
'MODELO DEL ACELEROGRAFO']
channels[1]['standard']['instrument'] = header_dict[
'MODELO DEL ACELEROGRAFO']
channels[2]['standard']['instrument'] = header_dict[
'MODELO DEL ACELEROGRAFO']
channels[0]['standard']['sensor_serial_number'] = header_dict[
'NUMERO DE SERIE DEL ACELEROGRAFO']
channels[1]['standard']['sensor_serial_number'] = header_dict[
'NUMERO DE SERIE DEL ACELEROGRAFO']
channels[2]['standard']['sensor_serial_number'] = header_dict[
'NUMERO DE SERIE DEL ACELEROGRAFO']
channels[0]['standard']['process_level'] = PROCESS_LEVELS['V1']
channels[1]['standard']['process_level'] = PROCESS_LEVELS['V1']
channels[2]['standard']['process_level'] = PROCESS_LEVELS['V1']
channels[0]['standard']['process_time'] = ''
channels[1]['standard']['process_time'] = ''
channels[2]['standard']['process_time'] = ''
channels[0]['standard']['station_name'] = header_dict[
'NOMBRE DE LA ESTACION']
channels[1]['standard']['station_name'] = header_dict[
'NOMBRE DE LA ESTACION']
channels[2]['standard']['station_name'] = header_dict[
'NOMBRE DE LA ESTACION']
channels[0]['standard']['structure_type'] = ''
channels[1]['standard']['structure_type'] = ''
channels[2]['standard']['structure_type'] = ''
channels[0]['standard']['corner_frequency'] = np.nan
channels[1]['standard']['corner_frequency'] = np.nan
channels[2]['standard']['corner_frequency'] = np.nan
channels[0]['standard']['units'] = 'cm/s/s'
channels[1]['standard']['units'] = 'cm/s/s'
channels[2]['standard']['units'] = 'cm/s/s'
periods = _get_periods(header_dict['FREC. NAT. DE SENSORES, C1-C6, (Hz)'])
channels[0]['standard']['instrument_period'] = periods[0]
channels[1]['standard']['instrument_period'] = periods[1]
channels[2]['standard']['instrument_period'] = periods[2]
dampings = _get_dampings(header_dict['AMORTIGUAMIENTO DE SENSORES, C1-C6'])
channels[0]['standard']['instrument_damping'] = dampings[0]
channels[1]['standard']['instrument_damping'] = dampings[1]
channels[2]['standard']['instrument_damping'] = dampings[2]
with open(filename, 'rt') as myfile:
header = [next(myfile) for x in range(HEADER_PLUS_COMMENT)]
clines = header[89:102]
comments = ' '.join(clines).strip()
channels[0]['standard']['comments'] = comments
channels[1]['standard']['comments'] = comments
channels[2]['standard']['comments'] = comments
head, tail = os.path.split(filename)
source_file = tail or os.path.basename(head)
channels[0]['standard']['source_file'] = source_file
channels[1]['standard']['source_file'] = source_file
channels[2]['standard']['source_file'] = source_file
channels[0]['standard']['source'] = SOURCE
channels[1]['standard']['source'] = SOURCE
channels[2]['standard']['source'] = SOURCE
decfactor = float(header_dict['FACTOR DE DECIMACION'])
channels[0]['standard']['instrument_sensitivity'] = decfactor
channels[1]['standard']['instrument_sensitivity'] = decfactor
channels[2]['standard']['instrument_sensitivity'] = decfactor
return channels
def _get_header_info(int_data, flt_data, lines, volume, location=''):
"""Return stats structure from various headers.
Output is a dictionary like this:
- network (str): 'LA'
- station (str)
- channel (str): Determined using COSMOS_ORIENTATIONS
- location (str): Default is '--'
- starttime (datetime)
- duration (float)
- sampling_rate (float)
- npts (int)
- coordinates:
- latitude (float)
- longitude (float)
- elevation (float)
- standard (Defaults are either np.nan or '')
- horizontal_orientation (float): Rotation from north (degrees)
- instrument_period (float): Period of sensor (Hz)
- instrument_damping (float): Fraction of critical
- process_time (datetime): Reported date of processing
- process_level: Either 'V0', 'V1', 'V2', or 'V3'
- station_name (str): Long form station description
- sensor_serial_number (str): Reported sensor serial
- instrument (str): See SENSOR_TYPES
- comments (str): Processing comments
- structure_type (str): See BUILDING_TYPES
- corner_frequency (float): Sensor corner frequency (Hz)
- units (str): See UNITS
- source (str): Network source description
- source_format (str): Always cosmos
- format_specific
- fractional_unit (float): Units of digitized acceleration
in file (fractions of g)
Args:
int_data (ndarray): Array of integer data
flt_data (ndarray): Array of float data
lines (list): List of text headers (str)
Returns:
dictionary: Dictionary of header/metadata information
"""
hdr = {}
coordinates = {}
standard = {}
format_specific = {}
if volume == 'V1':
hdr['duration'] = flt_data[2]
hdr['npts'] = int_data[27]
hdr['sampling_rate'] = (hdr['npts'] - 1) / hdr['duration']
# Get required parameter number
hdr['network'] = 'LA'
hdr['station'] = str(int_data[8])
logging.debug('station: %s' % hdr['station'])
horizontal_angle = int_data[26]
logging.debug('horizontal: %s' % horizontal_angle)
if (horizontal_angle in USC_ORIENTATIONS or
(horizontal_angle >= 0 and horizontal_angle <= 360)):
if horizontal_angle in USC_ORIENTATIONS:
channel = USC_ORIENTATIONS[horizontal_angle][1].upper()
if channel == 'UP' or channel == 'DOWN' or channel == 'VERT':
channel = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
horizontal_angle = 0.0
elif (
horizontal_angle > 315 or
horizontal_angle < 45 or
(horizontal_angle > 135 and horizontal_angle < 225)
):
channel = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
channel = get_channel_name(
hdr['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
horizontal_orientation = horizontal_angle
hdr['channel'] = channel
logging.debug('channel: %s' % hdr['channel'])
else:
errstr = ('USC: Not enough information to distinguish horizontal from '
'vertical channels.')
raise GMProcessException(errstr)
if location == '':
hdr['location'] = '--'
else:
hdr['location'] = location
month = str(int_data[21])
day = str(int_data[22])
year = str(int_data[23])
time = str(int_data[24])
tstr = month + '/' + day + '/' + year + '_' + time
starttime = datetime.strptime(tstr, '%m/%d/%Y_%H%M')
hdr['starttime'] = starttime
# Get coordinates
lat_deg = int_data[9]
lat_min = int_data[10]
lat_sec = int_data[11]
lon_deg = int_data[12]
lon_min = int_data[13]
lon_sec = int_data[14]
# Check for southern hemisphere, default is northern
if lines[4].find('STATION USC#') >= 0:
idx = lines[4].find('STATION USC#') + 12
if 'S' in lines[4][idx:]:
lat_sign = -1
else:
lat_sign = 1
else:
lat_sign = 1
# Check for western hemisphere, default is western
if lines[4].find('STATION USC#') >= 0:
idx = lines[4].find('STATION USC#') + 12
if 'W' in lines[4][idx:]:
lon_sign = -1
else:
lon_sign = 1
else:
lon_sign = -1
latitude = lat_sign * _dms2dd(lat_deg, lat_min, lat_sec)
longitude = lon_sign * _dms2dd(lon_deg, lon_min, lon_sec)
# Since sometimes longitudes are positive in this format for data in
# the western hemisphere, we "fix" it here. Hopefully no one in the
# eastern hemisphere uses this format!
if longitude > 0:
longitude = -longitude
coordinates['latitude'] = latitude
coordinates['longitude'] = longitude
coordinates['elevation'] = np.nan
# Get standard paramaters
standard['horizontal_orientation'] = float(horizontal_orientation)
standard['instrument_period'] = flt_data[0]
standard['instrument_damping'] = flt_data[1]
standard['process_time'] = ''
station_line = lines[5]
station_length = int(lines[5][72:74])
name = station_line[:station_length]
standard['station_name'] = name
standard['sensor_serial_number'] = ''
standard['instrument'] = ''
standard['comments'] = ''
standard['units'] = 'acc'
standard['structure_type'] = ''
standard['process_level'] = PROCESS_LEVELS['V1']
standard['corner_frequency'] = np.nan
standard['source'] = ('Los Angeles Basin Seismic Network, University '
'of Southern California')
standard['source_format'] = 'usc'
# Get format specific
format_specific['fractional_unit'] = flt_data[4]
# Set dictionary
hdr['standard'] = standard
hdr['coordinates'] = coordinates
hdr['format_specific'] = format_specific
return hdr
def _get_header_info(filename, any_structure=False, accept_flagged=False,
location=''):
"""Return stats structure from various headers.
Output is a dictionary like this:
- network
- station
- channel
- location (str): Set to floor the sensor is located on. If not a
multi-sensor array, default is '--'. Can be set manually by
the user.
- starttime
- sampling_rate
- npts
- coordinates:
- latitude
- longitude
- elevation
- standard
- horizontal_orientation
- instrument_period
- instrument_damping
- process_level
- station_name
- sensor_serial_number
- instrument
- comments
- structure_type
- corner_frequency
- units
- source
- source_format
- format_specific
- vertical_orientation
- building_floor (0=basement, 1=floor above basement, -1=1st sub-basement, etc.
- bridge_number_spans
- bridge_transducer_location ("free field",
"at the base of a pier or abutment",
"on an abutment",
"on the deck at the top of a pier"
"on the deck between piers or between an abutment and a pier."
dam_transducer_location ("upstream or downstream free field",
"at the base of the dam",
"on the crest of the dam",
on the abutment of the dam")
construction_type ("Reinforced concrete gravity",
"Reinforced concrete arch",
"earth fill",
"other")
filter_poles
data_source
"""
stats = {}
standard = {}
format_specific = {}
coordinates = {}
# read the ascii header lines
with open(filename) as f:
ascheader = [next(f).strip() for x in range(ASCII_HEADER_LINES)]
standard['process_level'] = PROCESS_LEVELS[VALID_HEADERS[ascheader[0]]]
logging.debug("process_level: %s" % standard['process_level'])
# station code is in the third line
stats['station'] = ''
if len(ascheader[2]) >= 4:
stats['station'] = ascheader[2][0:4].strip()
logging.debug('station: %s' % stats['station'])
standard['process_time'] = ''
standard['station_name'] = ascheader[5][10:40].strip()
# sometimes the data source has nothing in it,
# most of the time it seems has has USGS in it
# sometimes it's something like JPL/USGS, CDOT/USGS, etc.
# if it's got USGS in it, let's just say network=US, otherwise "--"
stats['network'] = 'ZZ'
if ascheader[7].find('USGS') > -1:
stats['network'] = 'US'
try:
standard['source'] = ascheader[7].split('=')[2].strip()
except IndexError:
standard['source'] = 'USGS'
if standard['source'] == '':
standard['source'] = 'USGS'
standard['source_format'] = 'smc'
# read integer header data
intheader = np.genfromtxt(filename, dtype=np.int32,
max_rows=INTEGER_HEADER_LINES,
skip_header=ASCII_HEADER_LINES,
delimiter=INT_HEADER_WIDTHS)
# 8 columns per line
# first line is start time information, and then inst. serial number
missing_data = intheader[0, 0]
year = intheader[0, 1]
jday = intheader[0, 2]
hour = intheader[0, 3]
minute = intheader[0, 4]
if (year != missing_data
and jday != missing_data and hour != missing_data
and minute != missing_data):
# Handle second if missing
second = 0
if not intheader[0, 5] == missing_data:
second = intheader[0, 5]
# Handle microsecond if missing and convert milliseconds to microseconds
microsecond = 0
if not intheader[0, 6] == missing_data:
microsecond = intheader[0, 6] / 1e3
datestr = '%i %00i %i %i %i %i' % (
year, jday, hour, minute, second, microsecond)
stats['starttime'] = datetime.strptime(datestr, '%Y %j %H %M %S %f')
else:
logging.warning('No start time provided. '
'This must be set manually for network/station: '
'%s/%s.' % (stats['network'], stats['station']))
standard['comments'] = 'Missing start time.'
standard['sensor_serial_number'] = ''
if intheader[1, 3] != missing_data:
standard['sensor_serial_number'] = str(intheader[1, 3])
# we never get a two character location code so floor location is used
if location == '':
location = intheader.flatten()[24]
if location != missing_data:
location = str(location)
if len(location) < 2:
location = location.zfill(2)
stats['location'] = location
else:
stats['location'] = '--'
else:
stats['location'] = location
# second line is information about number of channels, orientations
# we care about orientations
format_specific['vertical_orientation'] = np.nan
if intheader[1, 4] != missing_data:
format_specific['vertical_orientation'] = int(intheader[1, 4])
standard['horizontal_orientation'] = np.nan
if intheader[1, 5] != missing_data:
standard['horizontal_orientation'] = float(intheader[1, 5])
if intheader[1, 6] == missing_data or intheader[1, 6] not in INSTRUMENTS:
standard['instrument'] = ''
else:
standard['instrument'] = INSTRUMENTS[intheader[1, 6]]
num_comments = intheader[1, 7]
# third line contains number of data points
stats['npts'] = intheader[2, 0]
problem_flag = intheader[2, 1]
if problem_flag == 1:
if not accept_flagged:
fmt = 'SMC: Record found in file %s has a problem flag!'
raise Exception(fmt % filename)
else:
logging.warning(
'SMC: Data contains a problem flag for network/station: '
'%s/%s. See comments.' % (stats['network'], stats['station']))
stype = intheader[2, 2]
if stype == missing_data:
stype = np.nan
elif stype not in STRUCTURES:
# structure type is not defined and should will be considered 'other'
stype = 4
fmt = 'SMC: Record found in file %s is not a free-field sensor!'
standard['structure_type'] = STRUCTURES[stype]
if standard['structure_type'] == 'building' and not any_structure:
raise Exception(fmt % filename)
format_specific['building_floor'] = np.nan
if intheader[3, 0] != missing_data:
format_specific['building_floor'] = intheader[3, 0]
format_specific['bridge_number_spans'] = np.nan
if intheader[3, 1] != missing_data:
format_specific['bridge_number_spans'] = intheader[3, 1]
format_specific['bridge_transducer_location'] = BRIDGE_LOCATIONS[0]
if intheader[3, 2] != missing_data:
bridge_number = intheader[3, 2]
format_specific['bridge_transducer_location'] = \
BRIDGE_LOCATIONS[bridge_number]
format_specific['dam_transducer_location'] = DAM_LOCATIONS[0]
if intheader[3, 3] != missing_data:
dam_number = intheader[3, 3]
format_specific['dam_transducer_location'] = DAM_LOCATIONS[dam_number]
c1 = format_specific['bridge_transducer_location'].find('free field') == -1
c2 = format_specific['dam_transducer_location'].find('free field') == -1
if (c1 or c2) and not any_structure:
raise Exception(fmt % filename)
format_specific['construction_type'] = CONSTRUCTION_TYPES[4]
if intheader[3, 4] != missing_data:
format_specific['construction_type'] = \
CONSTRUCTION_TYPES[intheader[3, 4]]
# station is repeated here if all numeric
if not len(stats['station']):
stats['station'] = '%i' % intheader[3, 5]
# read float header data
skip = ASCII_HEADER_LINES + INTEGER_HEADER_LINES
floatheader = np.genfromtxt(
filename,
max_rows=FLOAT_HEADER_LINES,
skip_header=skip,
delimiter=FLOAT_HEADER_WIDTHS)
# float headers are 10 lines of 5 floats each
missing_data = floatheader[0, 0]
stats['sampling_rate'] = floatheader[0, 1]
coordinates['latitude'] = floatheader[2, 0]
# the documentation for SMC says that sometimes longitudes are
# positive in the western hemisphere. Since it is very unlikely
# any of these files exist for the eastern hemisphere, check for
# positive longitudes and fix them.
lon = floatheader[2, 1]
if lon > 0:
lon = -1 * lon
coordinates['longitude'] = lon
coordinates['elevation'] = 0
if floatheader[2, 2] != missing_data:
coordinates['elevation'] = floatheader[2, 2]
# figure out the channel code
if format_specific['vertical_orientation'] in [0, 180]:
stats['channel'] = get_channel_name(
stats['sampling_rate'],
is_acceleration=True,
is_vertical=True,
is_north=False)
else:
ho = standard['horizontal_orientation']
quad1 = ho > 315 and ho <= 360
quad2 = ho > 0 and ho <= 45
quad3 = ho > 135 and ho <= 225
if quad1 or quad2 or quad3:
stats['channel'] = get_channel_name(
stats['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=True)
else:
stats['channel'] = get_channel_name(
stats['sampling_rate'],
is_acceleration=True,
is_vertical=False,
is_north=False)
logging.debug('channel: %s' % stats['channel'])
sensor_frequency = floatheader[4, 1]
standard['instrument_period'] = 1 / sensor_frequency
standard['instrument_damping'] = floatheader[4, 2]
standard['corner_frequency'] = floatheader[3, 4]
format_specific['filter_poles'] = floatheader[4, 0]
standard['units'] = 'acc'
# read in the comment lines
with open(filename) as f:
skip = ASCII_HEADER_LINES + INTEGER_HEADER_LINES + FLOAT_HEADER_LINES
_ = [next(f) for x in range(skip)]
standard['comments'] = [next(f).strip().lstrip('|')
for x in range(num_comments)]
standard['comments'] = ' '.join(standard['comments'])
stats['coordinates'] = coordinates
stats['standard'] = standard
stats['format_specific'] = format_specific
head, tail = os.path.split(filename)
stats['standard']['source_file'] = tail or os.path.basename(head)
return (stats, num_comments)
