seisgo.types

The types module defines the core data container classes used throughout SeisGo: FFTData, CorrData, and DvvData.

---

Class Summary

---

FFTData

Frequency-domain representation of one station–component time series, with all processing (windowing, normalization, FFT) already applied.

Constructor

from seisgo.types import FFTData

fftdata = FFTData(
    trace,           # obspy.Trace
    win_len=3600,
    step=1800,
    stainv=None,     # obspy.Inventory or None
    freqmin=0.05,
    freqmax=2.0,
    time_norm="one_bit",
    freq_norm="rma",
    smooth=20,
    smooth_spec=None,
    taper_frac=0.05,
    df=None,
)

Key attributes

Attribute	Description
data	Complex 2-D array, shape (n_windows, n_freqs)
std	Per-window standard deviation, shape (n_windows,)
time	Window start times (UTC)
freq	Frequency vector (Hz)
dt	Sampling interval (s)
win_len	Analysis window length (s)
step	Window step / overlap (s)
net, sta, loc, chan	SEED channel identifiers
lat, lon, ele	Station coordinates

---

CorrData

Time-lag cross-correlation (or auto-correlation) between a station pair over multiple time windows.

Constructor

CorrData is normally created by seisgo.noise.correlate() or loaded with seisgo.noise.extract_corrdata(). Direct construction is also supported.

Key attributes

Attribute	Description
data	2-D float array, shape (n_windows, n_lags)
dt	Sampling interval (s)
lag	Maximum lag (s)
dist	Inter-station distance (km)
az, baz	Forward and back azimuths (°)
cc_comp	Component pair string, e.g. "ZZ", "RR", "TT"
side	"a" both, "n" negative, "p" positive, "o" one-sided
sta	[source_name, receiver_name]
net	[source_net, receiver_net]
time	Window timestamps
misc	Auxiliary metadata dictionary

Key methods

# Bandpass filter in-place
corrdata.filter(fmin=0.05, fmax=2.0, corners=4, zerophase=True)

# Collapse to a single stacked trace (returns new CorrData)
stacked = corrdata.stack(method="pws")

# Stack in-place (overwrites data)
corrdata.stack(method="linear", overwrite=True)

# Save to ASDF
corrdata.to_asdf("output.h5")

# Quick plot
corrdata.plot()

---

DvvData

Stores dv/v measurements produced by seisgo.monitoring.get_dvv().

Key attributes

Attribute	Description
dvv	dv/v values, shape (n_windows,) or (n_windows, n_freqs)
err	Measurement uncertainty (same shape as dvv)
cc	Waveform cross-correlation coefficient per measurement
time	Timestamp of each measurement window
freq	Frequency vector (wts) or scalar band (ts)
method	"wts" or "ts"
cc_comp	Component pair
sta	Station pair names

Key methods

dvvdata.plot()
dvvdata.to_asdf("dvv_output.h5")