Tie line levelling

5. Tie line levelling#

[1]:

%load_ext autoreload
%autoreload 2


import logging

import geopandas as gpd
import pandas as pd
import plotly.io as pio

import airbornegeo

logging.getLogger("airbornegeo").setLevel("INFO")
logging.basicConfig()
pio.renderers.default = "notebook"

/home/airbornegeo/airbornegeo/.pixi/envs/default/lib/python3.14/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html
  from .autonotebook import tqdm as notebook_tqdm

5.1. Load data#

This is a subset of the BAS AGAP survey over Antarctica’s Gamburtsev Subglacial Mountains. The file is download and subset in the notebook AGAP_magnetic_survey, and the BAS processing steps are repeated in the notebook processing_AGAP_magnetic_survey.

[2]:

data_df = pd.read_csv("data/AGAP_magnetic_survey_processed_blocked.csv")
data_df = data_df[
    [
        "easting",
        "northing",
        "height",
        "line",
        "unixtime",
        "distance_along_line",
        "mag",
    ]
]

# for run faster, limit number of lines
# for run faster, limit number of lines
data_df = data_df[~data_df.line.between(133, 142)]
data_df = data_df[~data_df.line.between(168, 176)]
data_df = data_df[
    (data_df.line.isin(data_df.line.unique()[::2])) | (data_df.line >= 143)
]

# define flight lines vs tie lines with column 'tie' which is True or False
data_df["tie"] = False
data_df.loc[data_df.line >= 142, "tie"] = True

data_df.head()

[2]:
easting northing height line unixtime distance_along_line mag tie
0 621099.093988 159056.748193 4110.45 1 1.229500e+09 27.181565 -33.385 False
1 621206.517953 159071.301512 4114.50 1 1.229500e+09 135.587530 -35.120 False
2 621313.794221 159085.113599 4117.90 1 1.229500e+09 243.749367 -36.875 False
3 621420.722903 159099.184618 4120.80 1 1.229500e+09 351.600337 -38.585 False
4 621527.158177 159114.303863 4123.15 1 1.229500e+09 459.104857 -40.205 False 
[3]:

airbornegeo.plotly_points(
    data_df[data_df.tie],
    color_col="line",
    hover_cols=["distance_along_line"],
)

[4]:

airbornegeo.plotly_points(
    data_df[~data_df.tie],
    color_col="line",
    hover_cols=["distance_along_line"],
)

5.2. Find intersections#

[5]:

# convert dataframe into geodataframe
data_df = gpd.GeoDataFrame(
    data_df,
    geometry=gpd.points_from_xy(data_df.easting, data_df.northing),
    crs="EPSG:3031",
)

[6]:

# calculate theoretical intersection points
inters = airbornegeo.create_intersection_table(data_df)
inters

Line/tie combinations: 100%|██████████| 3630/3630 [00:00<00:00, 7796.49it/s]
Potential intersections: 100%|██████████| 670/670 [00:06<00:00, 99.87it/s]
INFO:airbornegeo:found 662 intersections

[6]:
line tie geometry max_dist easting northing
0 1 143 POINT (1158153 254083) 16.441255 1158153.0 254083.0
1 1 144 POINT (1190820 259865) 24.124887 1190820.0 259865.0
2 1 145 POINT (1223524 265689) 17.269650 1223524.0 265689.0
3 1 146 POINT (1256193 271497) 31.608179 1256193.0 271497.0
4 1 147 POINT (1288901 277249) 50.174377 1288901.0 277249.0
... ... ... ... ... ... ...
657 125 199 POINT (1318165 302398) 33.689493 1318165.0 302398.0
658 125 200 POINT (1350929 308239) 36.343660 1350929.0 308239.0
659 127 148 POINT (1319894 292699) 22.093103 1319894.0 292699.0
660 127 199 POINT (1319922 292702) 23.297236 1319922.0 292702.0
661 127 200 POINT (1352642 298507) 36.743873 1352642.0 298507.0

662 rows × 6 columns

_images/tie_line_levelling_8_3.png

5.3. Add intersections as rows to the dataframe#

[7]:

data_df, inters = airbornegeo.interpolate_intersections(
    data_df,
    inters,
    to_interp=["mag", "height"],
    window_width=500,
    method="cubic",
    extrapolate=False,
)

Line 206: 100%|██████████| 121/121 [00:12<00:00,  9.79it/s]

[8]:

# see which lines dont have intersections
airbornegeo.lines_without_intersections(data_df, inters)

[8]:

[np.int64(107),
 np.int64(109),
 np.int64(129),
 np.int64(131),
 np.int64(188),
 np.int64(189),
 np.int64(190),
 np.int64(192),
 np.int64(193),
 np.int64(194),
 np.int64(203)]

5.4. Calculate initial cross-over errors#

[9]:

inters = airbornegeo.calculate_crossover_errors(
    data_df,
    inters,
    data_col="mag",
    plot_map=True,
)
_images/tie_line_levelling_13_1.png 
[10]:

inters.head()

[10]:
line tie geometry max_dist easting northing dist_along_flight_line dist_along_flight_tie flight_interpolation_type tie_interpolation_type flight_height tie_height mistie_0
0 1 143 POINT (1158153 254083) 16.441255 1158153.0 254083.0 545768.072695 138071.419338 interpolated interpolated 4168.038983 4008.397691 100.458661
1 1 144 POINT (1190820 259865) 24.124887 1190820.0 259865.0 578952.394781 131585.125284 interpolated interpolated 4174.500813 4010.840823 32.151483
2 1 145 POINT (1223524 265689) 17.269650 1223524.0 265689.0 612181.972720 147253.496664 interpolated interpolated 3544.149925 3995.961216 72.936553
3 1 146 POINT (1256193 271497) 31.608179 1256193.0 271497.0 645372.035087 136211.666619 interpolated interpolated 3564.579994 3578.174277 84.125118
4 1 147 POINT (1288901 277249) 50.174377 1288901.0 277249.0 678598.170740 155115.224990 interpolated interpolated 3536.160719 3545.947314 -7.731396

5.5. Level lines to ties#

Now we will level only the lines, holding the ties constant. We will just use a trend degree of 0, which allows only a DC shift of the lines. We will save the levelled data to a new column mag_levelled_lines_trend0. If you don’t want to keep track on new columns, you can just use the same name as the data column.

[11]:

data_df, inters = airbornegeo.line_levelling(
    data_df,
    inters,
    lines_to_level=data_df[data_df.tie == False].line.unique(),
    data_col="mag",
    levelled_col="mag_levelled_lines_trend0",
    degree=0,
)
inters.head()

[11]:
line tie geometry max_dist easting northing dist_along_flight_line dist_along_flight_tie flight_interpolation_type tie_interpolation_type flight_height tie_height mistie_0 mistie_1
0 1 143 POINT (1158153 254083) 16.441255 1158153.0 254083.0 545768.072695 138071.419338 interpolated interpolated 4168.038983 4008.397691 100.458661 96.945621
1 1 144 POINT (1190820 259865) 24.124887 1190820.0 259865.0 578952.394781 131585.125284 interpolated interpolated 4174.500813 4010.840823 32.151483 28.638444
2 1 145 POINT (1223524 265689) 17.269650 1223524.0 265689.0 612181.972720 147253.496664 interpolated interpolated 3544.149925 3995.961216 72.936553 69.423514
3 1 146 POINT (1256193 271497) 31.608179 1256193.0 271497.0 645372.035087 136211.666619 interpolated interpolated 3564.579994 3578.174277 84.125118 80.612078
4 1 147 POINT (1288901 277249) 50.174377 1288901.0 277249.0 678598.170740 155115.224990 interpolated interpolated 3536.160719 3545.947314 -7.731396 -11.244436 
[12]:

airbornegeo.plot_line_and_crosses(
    data_df,
    line=77,
    x="distance_along_line",
    y=["mag", "mag_levelled_lines_trend0"],
    y_axes=[1, 1],
    plot_inters=True,
)

[13]:

inters = airbornegeo.calculate_crossover_errors(
    data_df,
    inters,
    data_col="mag_levelled_lines_trend0",
    plot_map=True,
)
_images/tie_line_levelling_18_1.png 
[14]:

airbornegeo.plot_levelling_convergence(inters)
_images/tie_line_levelling_19_1.png

From the above profile, we can see line 77 has been shifted down to try and minimize the cross-over errors. The map, histogram and levelling convergence figures show we have reduced the cross-over errors, bringing the RMSE from ~43nT to ~24nT.

5.6. Level ties to lines#

We can also level the tie lines to the flight lines. We do this be supplying the lines_to_level parameter with the names of the tie lines. We will give this levelled data a new name mag_levelled_ties_trend0

[15]:

data_df, inters = airbornegeo.line_levelling(
    data_df,
    inters,
    lines_to_level=data_df[data_df.tie == True].line.unique(),
    data_col="mag_levelled_lines_trend0",
    levelled_col="mag_levelled_ties_trend0",
    degree=0,
)
inters.head()

[15]:
line tie geometry max_dist easting northing dist_along_flight_line dist_along_flight_tie flight_interpolation_type tie_interpolation_type flight_height tie_height mistie_0 mistie_1 mistie_2
0 1 143 POINT (1158153 254083) 16.441255 1158153.0 254083.0 545768.072695 138071.419338 interpolated interpolated 4168.038983 4008.397691 100.458661 96.945621 72.958517
1 1 144 POINT (1190820 259865) 24.124887 1190820.0 259865.0 578952.394781 131585.125284 interpolated interpolated 4174.500813 4010.840823 32.151483 28.638444 19.088634
2 1 145 POINT (1223524 265689) 17.269650 1223524.0 265689.0 612181.972720 147253.496664 interpolated interpolated 3544.149925 3995.961216 72.936553 69.423514 49.368210
3 1 146 POINT (1256193 271497) 31.608179 1256193.0 271497.0 645372.035087 136211.666619 interpolated interpolated 3564.579994 3578.174277 84.125118 80.612078 60.446633
4 1 147 POINT (1288901 277249) 50.174377 1288901.0 277249.0 678598.170740 155115.224990 interpolated interpolated 3536.160719 3545.947314 -7.731396 -11.244436 -2.791611 
[16]:

inters = airbornegeo.calculate_crossover_errors(
    data_df,
    inters,
    data_col="mag_levelled_ties_trend0",
    plot_map=True,
)
_images/tie_line_levelling_23_1.png 
[17]:

airbornegeo.plot_levelling_convergence(inters)
_images/tie_line_levelling_24_1.png

We can repeat these steps, increasing the trend order if we need more levelling, and alternating between levelling lines or ties. The next notebook shows how to automatically perform multiple iterations of levelling.