Filtering data in 1D

Contents

5. Filtering data in 1D#

[ ]:

# %load_ext autoreload
# %autoreload 2

import cmocean
import pandas as pd
import verde as vd

import airbornegeo

[ ]:

data_df = pd.read_csv("data/AGAP_gravity_survey_processed.csv")
data_df = data_df[
    [
        "easting",
        "northing",
        "line",
        "unixtime",
        "grav_disturbance",
        "distance_along_line",
    ]
]
data_df = data_df.sort_values(["line", "unixtime"]).reset_index(drop=True)
data_df.head()

	easting	northing	line	unixtime	grav_disturbance	distance_along_line
0	1.000024e+06	226237.330771	1	1.229507e+09	1186.4	0.000000
1	1.000083e+06	226246.631269	1	1.229507e+09	342.1	59.842447
2	1.000142e+06	226255.809132	1	1.229507e+09	-1965.9	119.693401
3	1.000201e+06	226264.969079	1	1.229507e+09	820.0	179.545645
4	1.000260e+06	226274.156809	1	1.229507e+09	3198.0	239.285174

5.1. Filter a line#

[7]:

# extract a single line from the survey
line_df = data_df[data_df.line == 4]

5.1.1. Spatial filtering#

We can perform 1D filtering based on the distance along the flight line.

[10]:

line_df["grav_disturbance_filtered_spatial"] = airbornegeo.filter_line(
    line_df,
    filter_type="g19000+l",  # 19 km low pass gaussian filter
    data_column="grav_disturbance",
    filter_by_column="distance_along_line",
    pad_width_percentage=10,
)

[17]:

ylim = vd.minmax(line_df.grav_disturbance, min_percentile=2, max_percentile=98)
ax = line_df.plot.line(
    "distance_along_line",
    "grav_disturbance",
    style="bp",
    ms=0.6,
    ylim=ylim,
)
ax = line_df.plot.line(
    "distance_along_line",
    "grav_disturbance_filtered_spatial",
    style="rp",
    ms=0.6,
    ax=ax,
    ylim=ylim,
)
ax.set_ylabel("mGal")

[17]:

Text(0, 0.5, 'mGal')

_images/filtering_data_in_1D_7_1.png

[18]:

line_df = line_df.sort_values("unixtime")

line_df["grav_disturbance_filtered_temporal"] = airbornegeo.filter_line(
    line_df,
    filter_type="g400+l",  # 400 sec low pass gaussian filter
    data_column="grav_disturbance",
    filter_by_column="unixtime",
    pad_width_percentage=10,
)

[19]:

ylim = vd.minmax(line_df.grav_disturbance, min_percentile=2, max_percentile=98)
ax = line_df.plot.line(
    "distance_along_line",
    "grav_disturbance",
    style="bp",
    ms=0.6,
    ylim=ylim,
)
ax = line_df.plot.line(
    "distance_along_line",
    "grav_disturbance_filtered_temporal",
    style="rp",
    ms=0.6,
    ax=ax,
    ylim=ylim,
)
ax.set_ylabel("mGal")

[19]:

Text(0, 0.5, 'mGal')

_images/filtering_data_in_1D_9_1.png

[27]:

ylim = vd.minmax(
    line_df[
        ["grav_disturbance_filtered_spatial", "grav_disturbance_filtered_temporal"]
    ],
)
ax = line_df.plot.line(
    "distance_along_line",
    "grav_disturbance_filtered_spatial",
    style="bp",
    ms=0.6,
    ylim=ylim,
)
ax = line_df.plot.line(
    "distance_along_line",
    "grav_disturbance_filtered_temporal",
    style="rp",
    ms=0.6,
    ax=ax,
    ylim=ylim,
)
ax.set_ylabel("mGal")

[27]:

Text(0, 0.5, 'mGal')

_images/filtering_data_in_1D_10_1.png

5.2. Filter all lines in a survey#

5.2.1. Spatial filtering#

[33]:

# plot the unfiltered gravity disturbance data
max_abs = vd.maxabs(line_df.grav_disturbance, percentile=95)
ax = data_df.plot.scatter(
    "easting",
    "northing",
    c="grav_disturbance",
    s=0.1,
    cmap=cmocean.cm.balance,
    vmin=-max_abs,
    vmax=max_abs,
)
ax.set_aspect("equal")

_images/filtering_data_in_1D_13_0.png

[34]:

data_df["grav_disturbance_filtered_spatial"] = airbornegeo.filter_line(
    data_df,
    filter_type="g19000+l",  # 19 km low pass gaussian filter
    data_column="grav_disturbance",
    filter_by_column="distance_along_line",
    groupby_column="line",
    pad_width_percentage=10,
)

Segments: 100%|██████████| 100/100 [00:03<00:00, 31.87it/s]

[35]:

# plot the filtered gravity disturbance data
max_abs = vd.maxabs(line_df.grav_disturbance_filtered_spatial, percentile=95)
ax = data_df.plot.scatter(
    "easting",
    "northing",
    c="grav_disturbance_filtered_spatial",
    s=0.1,
    cmap=cmocean.cm.balance,
    vmin=-max_abs,
    vmax=max_abs,
)
ax.set_aspect("equal")

_images/filtering_data_in_1D_15_0.png

5.2.2. Temporal filtering#

[36]:

data_df["grav_disturbance_filtered_temporal"] = airbornegeo.filter_line(
    data_df,
    filter_type="g500+l",  # 500 sec low pass gaussian filter
    data_column="grav_disturbance",
    filter_by_column="unixtime",
    groupby_column="line",
    pad_width_percentage=10,
)

Segments: 100%|██████████| 100/100 [00:04<00:00, 23.94it/s]

[37]:

# plot the filtered gravity disturbance data
max_abs = vd.maxabs(line_df.grav_disturbance_filtered_temporal, percentile=95)
ax = data_df.plot.scatter(
    "easting",
    "northing",
    c="grav_disturbance_filtered_temporal",
    s=0.1,
    cmap=cmocean.cm.balance,
    vmin=-max_abs,
    vmax=max_abs,
)
ax.set_aspect("equal")

_images/filtering_data_in_1D_18_0.png