Contents

Demo MAGxLR_1B (magnetic field 1Hz)

Authors: Ashley Smith

Abstract: Access to the low rate (1Hz) magnetic data (level 1b product), together with geomagnetic model evaluations (level 2 products).

%load_ext watermark
%watermark -i -v -p viresclient,pandas,xarray,matplotlib

2021-01-24T15:45:21+00:00

CPython 3.7.6
IPython 7.11.1

viresclient 0.7.1
pandas 0.25.3
xarray 0.15.0
matplotlib 3.1.2

from viresclient import SwarmRequest
import datetime as dt
import matplotlib.pyplot as plt

request = SwarmRequest()

MAGX_LR_1B product information

This is one of the main products from Swarm - the 1Hz measurements of the magnetic field vector (B_NEC) and total intensity (F). These are derived from the Vector Field Magnetometer (VFM) and Absolute Scalar Magnetomer (ASM).

Documentation:

  • https://earth.esa.int/web/guest/missions/esa-eo-missions/swarm/data-handbook/level-1b-product-definitions#MAGX_LR_1B_Product

Measurements are available through VirES as part of collections with names containing MAGx_LR, for each Swarm spacecraft:

request.available_collections("MAG", details=False)

{'MAG': ['SW_OPER_MAGA_LR_1B', 'SW_OPER_MAGB_LR_1B', 'SW_OPER_MAGC_LR_1B']}

The measurements can be used together with geomagnetic model evaluations as shall be shown below.

Check what “MAG” data variables are available

request.available_measurements("MAG")

['F',
 'dF_AOCS',
 'dF_other',
 'F_error',
 'B_VFM',
 'B_NEC',
 'dB_Sun',
 'dB_AOCS',
 'dB_other',
 'B_error',
 'q_NEC_CRF',
 'Att_error',
 'Flags_F',
 'Flags_B',
 'Flags_q',
 'Flags_Platform',
 'ASM_Freq_Dev']

Check the names of available models

request.available_models(details=False)

['IGRF',
 'LCS-1',
 'MF7',
 'CHAOS-Core',
 'CHAOS-Static',
 'CHAOS-MMA-Primary',
 'CHAOS-MMA-Secondary',
 'MCO_SHA_2C',
 'MCO_SHA_2D',
 'MLI_SHA_2C',
 'MLI_SHA_2D',
 'MLI_SHA_2E',
 'MMA_SHA_2C-Primary',
 'MMA_SHA_2C-Secondary',
 'MMA_SHA_2F-Primary',
 'MMA_SHA_2F-Secondary',
 'MIO_SHA_2C-Primary',
 'MIO_SHA_2C-Secondary',
 'MIO_SHA_2D-Primary',
 'MIO_SHA_2D-Secondary',
 'AMPS',
 'MCO_SHA_2X',
 'CHAOS',
 'CHAOS-MMA',
 'MMA_SHA_2C',
 'MMA_SHA_2F',
 'MIO_SHA_2C',
 'MIO_SHA_2D',
 'SwarmCI']

Fetch one hour of MAG data and models, at 10-second sampling

request.set_collection("SW_OPER_MAGA_LR_1B")
request.set_products(
    measurements=["F", "B_NEC"],
    models=["CHAOS-Core", "MCO_SHA_2D"],
    sampling_step="PT10S"
)
data = request.get_between(
    # 2014-01-01 00:00:00
    start_time = dt.datetime(2014,1,1, 0),
    # 2014-01-01 01:00:00
    end_time = dt.datetime(2014,1,1, 1)
)

[1/1] Processing:  100%|██████████|  [ Elapsed: 00:01, Remaining: 00:00 ]
      Downloading: 100%|██████████|  [ Elapsed: 00:00, Remaining: 00:00 ] (0.098MB)

See a list of the input files

data.sources

['SW_OPER_MAGA_LR_1B_20140101T000000_20140101T235959_0505_MDR_MAG_LR',
 'SW_OPER_MCO_SHA_2D_20131126T000000_20180101T000000_0401',
 'SW_OPER_MCO_SHA_2X_19970101T000000_20210526T235959_0705']

Transfer data to a pandas dataframe:

df = data.as_dataframe()
df.head()
F_MCO_SHA_2D B_NEC F Spacecraft Radius Latitude Longitude F_CHAOS-Core B_NEC_CHAOS-Core B_NEC_MCO_SHA_2D
Timestamp
2014-01-01 00:00:00 22874.211509 [20103.5246, -4126.2621, -10086.988800000001] 22867.5503 A 6878309.22 -1.228938 -14.116674 22874.580109 [20113.33871534297, -4127.023256760892, -10083... [20113.62392147383, -4127.463956127047, -10081...
2014-01-01 00:00:10 22820.941425 [19815.0914, -4160.9933, -10514.4074] 22814.5656 A 6878381.17 -1.862521 -14.131424 22821.316363 [19824.843984083705, -4162.664701084325, -1051... [19825.16184358055, -4163.127549318375, -10508...
2014-01-01 00:00:20 22769.369161 [19523.4946, -4195.196800000001, -10926.966400... 22763.2585 A 6878452.05 -2.496090 -14.146155 22769.746082 [19533.200506222718, -4197.0443252580635, -109... [19533.553905492434, -4197.529053749354, -1092...
2014-01-01 00:00:30 22719.238240 [19229.2386, -4228.4747, -11324.8335] 22713.3703 A 6878521.87 -3.129644 -14.160861 22719.613020 [19238.95246531388, -4230.095511676884, -11320... [19239.343572344795, -4230.60181901415, -11318...
2014-01-01 00:00:40 22670.304681 [18932.8807, -4260.8424, -11708.0897] 22664.7202 A 6878590.61 -3.763184 -14.175534 22670.673539 [18942.644962034465, -4261.75308572276, -11703... [18943.075143502585, -4262.280633762369, -1170...

Use expand=True to extract vectors (B_NEC…) as separate columns (…_N, …_E, …_C)

df = data.as_dataframe(expand=True)
df.head()
F_MCO_SHA_2D F Spacecraft Radius Longitude F_CHAOS-Core Latitude B_NEC_N B_NEC_E B_NEC_C B_NEC_CHAOS-Core_N B_NEC_CHAOS-Core_E B_NEC_CHAOS-Core_C B_NEC_MCO_SHA_2D_N B_NEC_MCO_SHA_2D_E B_NEC_MCO_SHA_2D_C
Timestamp
2014-01-01 00:00:00 22874.211509 22867.5503 A 6878309.22 -14.116674 22874.580109 -1.228938 20103.5246 -4126.2621 -10086.9888 20113.338715 -4127.023257 -10083.040213 20113.623921 -4127.463956 -10081.454567
2014-01-01 00:00:10 22820.941425 22814.5656 A 6878381.17 -14.131424 22821.316363 -1.862521 19815.0914 -4160.9933 -10514.4074 19824.843984 -4162.664701 -10510.007809 19825.161844 -4163.127549 -10508.410652
2014-01-01 00:00:20 22769.369161 22763.2585 A 6878452.05 -14.146155 22769.746082 -2.496090 19523.4946 -4195.1968 -10926.9664 19533.200506 -4197.044325 -10922.464628 19533.553905 -4197.529054 -10920.860481
2014-01-01 00:00:30 22719.238240 22713.3703 A 6878521.87 -14.160861 22719.613020 -3.129644 19229.2386 -4228.4747 -11324.8335 19238.952465 -4230.095512 -11320.327547 19239.343572 -4230.601819 -11318.721366
2014-01-01 00:00:40 22670.304681 22664.7202 A 6878590.61 -14.175534 22670.673539 -3.763184 18932.8807 -4260.8424 -11708.0897 18942.644962 -4261.753086 -11703.550793 18943.075144 -4262.280634 -11701.947796

… or to an xarray Dataset:

ds = data.as_xarray()
ds

<xarray.Dataset>
Dimensions:           (NEC: 3, Timestamp: 360)
Coordinates:
  * Timestamp         (Timestamp) datetime64[ns] 2014-01-01 ... 2014-01-01T00:59:50
  * NEC               (NEC) <U1 'N' 'E' 'C'
Data variables:
    Spacecraft        (Timestamp) object 'A' 'A' 'A' 'A' 'A' ... 'A' 'A' 'A' 'A'
    F_MCO_SHA_2D      (Timestamp) float64 2.287e+04 2.282e+04 ... 4.021e+04
    B_NEC_CHAOS-Core  (Timestamp, NEC) float64 2.011e+04 ... 3.557e+04
    F                 (Timestamp) float64 2.287e+04 2.281e+04 ... 4.021e+04
    Radius            (Timestamp) float64 6.878e+06 6.878e+06 ... 6.868e+06
    B_NEC_MCO_SHA_2D  (Timestamp, NEC) float64 2.011e+04 ... 3.557e+04
    Longitude         (Timestamp) float64 -14.12 -14.13 -14.15 ... 153.6 153.6
    F_CHAOS-Core      (Timestamp) float64 2.287e+04 2.282e+04 ... 4.02e+04
    B_NEC             (Timestamp, NEC) float64 2.01e+04 -4.126e+03 ... 3.558e+04
    Latitude          (Timestamp) float64 -1.229 -1.863 -2.496 ... 48.14 48.77
Attributes:
    Sources:         ['SW_OPER_MAGA_LR_1B_20140101T000000_20140101T235959_050...
    MagneticModels:  ["CHAOS-Core = 'CHAOS-Core'(max_degree=20,min_degree=1)"...
    RangeFilters:    []
ds.Sources

['SW_OPER_MAGA_LR_1B_20140101T000000_20140101T235959_0505_MDR_MAG_LR',
 'SW_OPER_MCO_SHA_2D_20131126T000000_20180101T000000_0401',
 'SW_OPER_MCO_SHA_2X_19970101T000000_20210526T235959_0705']

Instead, fetch the residuals directly

Adding residuals=True to .set_products() will instead directly evaluate and return all data-model residuals

request = SwarmRequest()
request.set_collection("SW_OPER_MAGA_LR_1B")
request.set_products(
    measurements=["F", "B_NEC"],
    models=["CHAOS-Core", "MCO_SHA_2D"],
    residuals=True,
    sampling_step="PT10S"
)
data = request.get_between(
    start_time = dt.datetime(2014,1,1, 0),
    end_time = dt.datetime(2014,1,1, 1)
)
df = data.as_dataframe(expand=True)
df.head()

[1/1] Processing:  100%|██████████|  [ Elapsed: 00:01, Remaining: 00:00 ]
      Downloading: 100%|██████████|  [ Elapsed: 00:00, Remaining: 00:00 ] (0.081MB)
Spacecraft Radius F_res_MCO_SHA_2D Longitude F_res_CHAOS-Core Latitude B_NEC_res_MCO_SHA_2D_N B_NEC_res_MCO_SHA_2D_E B_NEC_res_MCO_SHA_2D_C B_NEC_res_CHAOS-Core_N B_NEC_res_CHAOS-Core_E B_NEC_res_CHAOS-Core_C
Timestamp
2014-01-01 00:00:00 A 6878309.22 -6.661209 -14.116674 -7.029809 -1.228938 -10.099321 1.201856 -5.534233 -9.814115 0.761157 -3.948587
2014-01-01 00:00:10 A 6878381.17 -6.375825 -14.131424 -6.750763 -1.862521 -10.070444 2.134249 -5.996748 -9.752584 1.671401 -4.399591
2014-01-01 00:00:20 A 6878452.05 -6.110661 -14.146155 -6.487582 -2.496090 -10.059305 2.332254 -6.105919 -9.705906 1.847525 -4.501772
2014-01-01 00:00:30 A 6878521.87 -5.867940 -14.160861 -6.242720 -3.129644 -10.104972 2.127119 -6.112134 -9.713865 1.620812 -4.505953
2014-01-01 00:00:40 A 6878590.61 -5.584481 -14.175534 -5.953339 -3.763184 -10.194444 1.438234 -6.141904 -9.764262 0.910686 -4.538907

Plot the scalar residuals for each model

… using the pandas method:

ax = df.plot(
    y=["F_res_CHAOS-Core", "F_res_MCO_SHA_2D"],
    figsize=(15,5),
    grid=True
)
ax.set_xlabel("Timestamp")
ax.set_ylabel("[nT]");
../_images/03a1_Demo-MAGx_LR_1B_24_0.png

… using matplotlib interface (Matlab-style)

NB: we are doing plt.plot(x, y) with x as df.index (the time-based index of df), and y as df[".."]

plt.figure(figsize=(15,5))
plt.plot(
    df.index,
    df["F_res_CHAOS-Core"],
    label="F_res_CHAOS-Core"
)
plt.plot(
    df.index,
    df["F_res_MCO_SHA_2D"],
    label="F_res_MCO_SHA_2D"
)
plt.xlabel("Timestamp")
plt.ylabel("[nT]")
plt.grid()
plt.legend();
../_images/03a1_Demo-MAGx_LR_1B_26_0.png

… using matplotlib interface (Object Oriented style)

This is the recommended route for making more complicated figures

fig, ax = plt.subplots(figsize=(15,5))
ax.plot(
    df.index,
    df["F_res_CHAOS-Core"],
    label="F_res_CHAOS-Core"
)
ax.plot(
    df.index,
    df["F_res_MCO_SHA_2D"],
    label="F_res_MCO_SHA_2D"
)
ax.set_xlabel("Timestamp")
ax.set_ylabel("[nT]")
ax.grid()
ax.legend();
../_images/03a1_Demo-MAGx_LR_1B_28_0.png

Plot the vector components

fig, axes = plt.subplots(nrows=3, ncols=1, figsize=(15,10), sharex=True)
for component, ax in zip("NEC", axes):
    for model_name in ("CHAOS-Core", "MCO_SHA_2D"):
        ax.plot(
            df.index,
            df[f"B_NEC_res_{model_name}_{component}"],
            label=model_name
        )
    ax.set_ylabel(f"{component}\n[nT]")
    ax.legend()
axes[0].set_title("Residuals to models (NEC components)")
axes[2].set_xlabel("Timestamp");
../_images/03a1_Demo-MAGx_LR_1B_30_0.png

Similar plotting, using the data via xarray instead

xarray provides a more sophisticated data structure that is more suitable for the complex vector data we are accessing, together with nice stuff like unit and other metadata support. Unfortunately due to the extra complexity, this can make it difficult to use right away.

ds = data.as_xarray()
ds

<xarray.Dataset>
Dimensions:               (NEC: 3, Timestamp: 360)
Coordinates:
  * Timestamp             (Timestamp) datetime64[ns] 2014-01-01 ... 2014-01-01T00:59:50
  * NEC                   (NEC) <U1 'N' 'E' 'C'
Data variables:
    Spacecraft            (Timestamp) object 'A' 'A' 'A' 'A' ... 'A' 'A' 'A' 'A'
    B_NEC_res_MCO_SHA_2D  (Timestamp, NEC) float64 -10.1 1.202 ... 2.782 8.984
    B_NEC_res_CHAOS-Core  (Timestamp, NEC) float64 -9.814 0.7612 ... 2.925 10.1
    Radius                (Timestamp) float64 6.878e+06 6.878e+06 ... 6.868e+06
    Longitude             (Timestamp) float64 -14.12 -14.13 ... 153.6 153.6
    F_res_MCO_SHA_2D      (Timestamp) float64 -6.661 -6.376 ... 3.153 3.108
    F_res_CHAOS-Core      (Timestamp) float64 -7.03 -6.751 ... 5.201 5.203
    Latitude              (Timestamp) float64 -1.229 -1.863 ... 48.14 48.77
Attributes:
    Sources:         ['SW_OPER_MAGA_LR_1B_20140101T000000_20140101T235959_050...
    MagneticModels:  ["CHAOS-Core = 'CHAOS-Core'(max_degree=20,min_degree=1)"...
    RangeFilters:    []
fig, axes = plt.subplots(nrows=3, ncols=1, figsize=(15,10), sharex=True)
for i, ax in enumerate(axes):
    for model_name in ("CHAOS-Core", "MCO_SHA_2D"):
        ax.plot(
            ds["Timestamp"],
            ds[f"B_NEC_res_{model_name}"][:, i],
            label=model_name
        )
    ax.set_ylabel("NEC"[i] + " [nT]")
    ax.legend()
axes[0].set_title("Residuals to models (NEC components)")
axes[2].set_xlabel("Timestamp");
# automatic unit labels will be possible in v0.5.0
../_images/03a1_Demo-MAGx_LR_1B_33_0.png

Note that xarray also allows convenient direct plotting like:

ds["B_NEC_res_CHAOS-Core"].plot.line(x="Timestamp");
../_images/03a1_Demo-MAGx_LR_1B_35_0.png

Access multiple MAG datasets simultaneously

It is possible to fetch data from multiple collections simultaneously. Here we fetch the measurements from Swarm Alpha and Bravo. In the returned data, you can differentiate between them using the “Spacecraft” column.

request = SwarmRequest()
request.set_collection("SW_OPER_MAGA_LR_1B", "SW_OPER_MAGC_LR_1B")
request.set_products(
    measurements=["F", "B_NEC"],
    models=["CHAOS-Core",],
    residuals=True,
    sampling_step="PT10S"
)
data = request.get_between(
    start_time = dt.datetime(2014,1,1, 0),
    end_time = dt.datetime(2014,1,1, 1)
)
df = data.as_dataframe(expand=True)
df.head()

[1/1] Processing:  100%|██████████|  [ Elapsed: 00:01, Remaining: 00:00 ]
      Downloading: 100%|██████████|  [ Elapsed: 00:00, Remaining: 00:00 ] (0.072MB)
Spacecraft Radius Longitude F_res_CHAOS-Core Latitude B_NEC_res_CHAOS-Core_N B_NEC_res_CHAOS-Core_E B_NEC_res_CHAOS-Core_C
Timestamp
2014-01-01 00:00:00 A 6878309.22 -14.116674 -7.029809 -1.228938 -9.814115 0.761157 -3.948587
2014-01-01 00:00:10 A 6878381.17 -14.131424 -6.750763 -1.862521 -9.752584 1.671401 -4.399591
2014-01-01 00:00:20 A 6878452.05 -14.146155 -6.487582 -2.496090 -9.705906 1.847525 -4.501772
2014-01-01 00:00:30 A 6878521.87 -14.160861 -6.242720 -3.129644 -9.713865 1.620812 -4.505953
2014-01-01 00:00:40 A 6878590.61 -14.175534 -5.953339 -3.763184 -9.764262 0.910686 -4.538907 
df[df["Spacecraft"] == "A"].head()
Spacecraft Radius Longitude F_res_CHAOS-Core Latitude B_NEC_res_CHAOS-Core_N B_NEC_res_CHAOS-Core_E B_NEC_res_CHAOS-Core_C
Timestamp
2014-01-01 00:00:00 A 6878309.22 -14.116674 -7.029809 -1.228938 -9.814115 0.761157 -3.948587
2014-01-01 00:00:10 A 6878381.17 -14.131424 -6.750763 -1.862521 -9.752584 1.671401 -4.399591
2014-01-01 00:00:20 A 6878452.05 -14.146155 -6.487582 -2.496090 -9.705906 1.847525 -4.501772
2014-01-01 00:00:30 A 6878521.87 -14.160861 -6.242720 -3.129644 -9.713865 1.620812 -4.505953
2014-01-01 00:00:40 A 6878590.61 -14.175534 -5.953339 -3.763184 -9.764262 0.910686 -4.538907 
df[df["Spacecraft"] == "C"].head()
Spacecraft Radius Longitude F_res_CHAOS-Core Latitude B_NEC_res_CHAOS-Core_N B_NEC_res_CHAOS-Core_E B_NEC_res_CHAOS-Core_C
Timestamp
2014-01-01 00:00:00 C 6877665.99 -14.420068 -10.398368 5.908082 -10.417588 1.895748 -0.353263
2014-01-01 00:00:10 C 6877747.67 -14.434576 -10.026707 5.274386 -10.213451 2.031666 -0.977793
2014-01-01 00:00:20 C 6877828.39 -14.449141 -9.737933 4.640702 -10.128052 1.935491 -1.419892
2014-01-01 00:00:30 C 6877908.15 -14.463755 -9.526265 4.007030 -10.237920 1.539257 -1.955235
2014-01-01 00:00:40 C 6877986.93 -14.478412 -9.306678 3.373371 -10.336332 1.068366 -2.353611

… or using xarray

ds = data.as_xarray()
ds.where(ds["Spacecraft"] == "A", drop=True)

<xarray.Dataset>
Dimensions:               (NEC: 3, Timestamp: 360)
Coordinates:
  * Timestamp             (Timestamp) datetime64[ns] 2014-01-01 ... 2014-01-01T00:59:50
  * NEC                   (NEC) <U1 'N' 'E' 'C'
Data variables:
    Spacecraft            (Timestamp) object 'A' 'A' 'A' 'A' ... 'A' 'A' 'A' 'A'
    B_NEC_res_CHAOS-Core  (Timestamp, NEC) float64 -9.814 0.7612 ... 2.925 10.1
    Radius                (Timestamp) float64 6.878e+06 6.878e+06 ... 6.868e+06
    Longitude             (Timestamp) float64 -14.12 -14.13 ... 153.6 153.6
    F_res_CHAOS-Core      (Timestamp) float64 -7.03 -6.751 ... 5.201 5.203
    Latitude              (Timestamp) float64 -1.229 -1.863 ... 48.14 48.77
Attributes:
    Sources:         ['SW_OPER_MAGA_LR_1B_20140101T000000_20140101T235959_050...
    MagneticModels:  ["CHAOS-Core = 'CHAOS-Core'(max_degree=20,min_degree=1)"]
    RangeFilters:    []
Working with large data volumes Demo MAGxHR_1B (magnetic field 50Hz)