Statistics computation
Notebook setup
[1]:
# %matplotlib notebook # does not work in JupyterLab
%matplotlib inline
%load_ext autoreload
%autoreload 2
[2]:
import sys
sys.path.append('../..')
[3]:
import numpy as np
from matplotlib import pyplot as plt
[4]:
import s1etad
Product introspection and navigation
[5]:
filename = '../../sample-products/S1B_IW_ETA__AXDH_20200127T113414_20200127T113858_020002_025D72_0096.SAFE'
[6]:
product = s1etad.Sentinel1Etad(filename)
[7]:
product
[7]:
Sentinel1Etad("../../sample-products/S1B_IW_ETA__AXDH_20200127T113414_20200127T113858_020002_025D72_0096.SAFE") # 0x7fce4fd88390
Number of Sentinel-1 slices: 11
Sentinel-1 products list:
S1B_IW_SLC__1ADH_20200127T113414_20200127T113444_020002_025D72_FD42.SAFE
S1B_IW_SLC__1ADH_20200127T113442_20200127T113510_020002_025D72_FC8F.SAFE
S1B_IW_SLC__1ADH_20200127T113507_20200127T113534_020002_025D72_78DF.SAFE
S1B_IW_SLC__1ADH_20200127T113532_20200127T113559_020002_025D72_1ABB.SAFE
S1B_IW_SLC__1ADH_20200127T113557_20200127T113624_020002_025D72_6706.SAFE
S1B_IW_SLC__1ADH_20200127T113622_20200127T113649_020002_025D72_78F0.SAFE
S1B_IW_SLC__1ADH_20200127T113647_20200127T113715_020002_025D72_11FB.SAFE
S1B_IW_SLC__1ADH_20200127T113712_20200127T113739_020002_025D72_FCC6.SAFE
S1B_IW_SLC__1ADH_20200127T113737_20200127T113804_020002_025D72_1FED.SAFE
S1B_IW_SLC__1ADH_20200127T113802_20200127T113829_020002_025D72_C434.SAFE
S1B_IW_SLC__1ADH_20200127T113827_20200127T113858_020002_025D72_E443.SAFE
Number of swaths: 3
Swath list: IW1, IW2, IW3
Azimuth time:
min: 2020-01-27 11:34:14.475597
max: 2020-01-27 11:38:58.167033
Range time:
min: 0.005371080957826855
max: 0.006416814907377812
Grid sampling:
x: 8.081406101630269e-07
y: 0.028777788199999974
unit: s
Grid spacing:
x: 200.0
y: 200.0
unit: m
Processing settings:
troposphericDelayCorrection: True
ionosphericDelayCorrection: True
solidEarthTideCorrection: True
bistaticAzimuthCorrection: True
dopplerShiftRangeCorrection: True
FMMismatchAzimuthCorrection: True
[8]:
swath = product['IW1']
[9]:
burst = swath[1]
[10]:
t, tau = burst.get_burst_grid()
print('t.shape', t.shape)
print('tau.shape', tau.shape)
t.shape (111,)
tau.shape (403,)
[11]:
correction = burst.get_correction(s1etad.ECorrectionType.SUM, meter=True)
[12]:
rg_correction = correction['x']
az_correction = correction['y']
print('rg_correction.shape', rg_correction.shape)
print('az_correction.shape', az_correction.shape)
rg_correction.shape (111, 403)
az_correction.shape (111, 403)
[13]:
plt.figure()
plt.imshow(correction['x'], extent=[tau[0]*1e6, tau[-1]*1e6, t[0], t[-1]], aspect='auto')
plt.xlabel(r'$\tau\ [\mu s]$')
plt.ylabel('t [s]')
plt.colorbar().set_label('[{}]'.format(correction['unit']))
plt.title('{} correction in range direction'.format(correction['name']))
[13]:
Text(0.5, 1.0, 'sum correction in range direction')
Collect data statistics
NOTE: statistics are also available in the XML annotation file included in the S1-ETAD product
[14]:
import pandas as pd
Initialize dataframes
[15]:
x_corrections_names = ['tropospheric', 'ionospheric', 'geodetic', 'doppler', 'sum']
xcols = [('', 'bIndex'), ('', 't')] + [
(cname, name)
for cname in x_corrections_names
for name in ('min', 'mean', 'std', 'max')
]
xcols = pd.MultiIndex.from_tuples(xcols)
xstats_df = pd.DataFrame(columns=xcols, dtype=np.float64)
y_corrections_names = ['geodetic', 'bistatic', 'fmrate', 'sum']
ycols = [('', 'bIndex'), ('', 't')] + [
(cname, name)
for cname in y_corrections_names
for name in ('min', 'mean', 'std', 'max')
]
ycols = pd.MultiIndex.from_tuples(ycols)
ystats_df = pd.DataFrame(columns=ycols, dtype=np.float64)
Collect statistics
[16]:
for swath in product:
for burst in swath:
az, _ = burst.get_burst_grid()
t = np.mean(az[[0, -1]])
# range
row = {
('', 'bIndex'): burst.burst_index,
('', 't'): t
}
for name in x_corrections_names:
data = burst.get_correction(name, meter=True)
row[(name, 'min')] = data['x'].min()
row[(name, 'mean')] = data['x'].mean()
row[(name, 'std')] = data['x'].std()
row[(name, 'max')] = data['x'].max()
xstats_df = xstats_df.append(row, ignore_index=True)
# azimuth
row = {
('', 'bIndex'): int(burst.burst_index),
('', 't'): t
}
for name in y_corrections_names:
# NOTE: meter is False in this case due to a limitation of the
# current implementation
data = burst.get_correction(name, meter=True)
row[(name, 'min')] = data['y'].min()
row[(name, 'mean')] = data['y'].mean()
row[(name, 'std')] = data['y'].std()
row[(name, 'max')] = data['y'].max()
ystats_df = ystats_df.append(row, ignore_index=True)
Inspect results
Correction statistics in range direction
[17]:
xstats_df.head()
[17]:
| tropospheric | ionospheric | ... | geodetic | doppler | sum | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| bIndex | t | min | mean | std | max | min | mean | std | max | ... | std | max | min | mean | std | max | min | mean | std | max | |
| 0 | 1.0 | 1.582778 | 2.155326 | 2.714470 | 0.145982 | 2.858699 | 0.009107 | 0.009528 | 0.000177 | 0.009969 | ... | 0.001659 | 0.119038 | -0.388467 | -0.001669 | 0.220652 | 0.387346 | 1.977618 | 2.848050 | 0.281908 | 3.358570 |
| 1 | 4.0 | 4.331057 | 2.163874 | 2.707724 | 0.147308 | 2.858642 | 0.008750 | 0.009168 | 0.000172 | 0.009595 | ... | 0.001654 | 0.118861 | -0.390278 | 0.000689 | 0.222749 | 0.393396 | 2.117201 | 2.843175 | 0.259000 | 3.363231 |
| 2 | 7.0 | 7.093725 | 2.298610 | 2.738260 | 0.095753 | 2.858575 | 0.008407 | 0.008809 | 0.000167 | 0.009222 | ... | 0.001668 | 0.118819 | -0.391369 | 0.000025 | 0.223233 | 0.393376 | 2.247333 | 2.872574 | 0.241518 | 3.364020 |
| 3 | 10.0 | 9.856392 | 2.329711 | 2.735871 | 0.099814 | 2.860330 | 0.008076 | 0.008454 | 0.000161 | 0.008853 | ... | 0.001664 | 0.118681 | -0.392313 | -0.001560 | 0.222806 | 0.390904 | 2.073561 | 2.868118 | 0.254228 | 3.360668 |
| 4 | 13.0 | 12.604671 | 2.220333 | 2.681844 | 0.142236 | 2.860439 | 0.007739 | 0.008107 | 0.000154 | 0.008488 | ... | 0.001647 | 0.118491 | -0.386082 | 0.000791 | 0.220615 | 0.389624 | 2.029879 | 2.815948 | 0.275655 | 3.360558 |
5 rows × 22 columns
Maximum value of corrections in range direction
[18]:
xstats_df.abs().max()
[18]:
bIndex 306.000000
t 282.079880
tropospheric min 3.076796
mean 3.189554
std 0.227448
max 3.305507
ionospheric min 0.014389
mean 0.014750
std 0.000205
max 0.015121
geodetic min 0.113129
mean 0.116042
std 0.001812
max 0.119038
doppler min 0.475521
mean 0.006117
std 0.268563
max 0.471421
sum min 2.730788
mean 3.306536
std 0.378128
max 3.867100
dtype: float64
Correction statistics in azimuth direction
[19]:
ystats_df.head()
[19]:
| geodetic | bistatic | fmrate | sum | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| bIndex | t | min | mean | std | max | min | mean | std | max | min | mean | std | max | min | mean | std | max | |
| 0 | 1.0 | 1.582778 | 0.008805 | 0.009118 | 0.000159 | 0.009418 | -3.657967 | -3.109659 | 0.317352 | -2.561351 | -0.569971 | 0.040845 | 0.152048 | 1.046240 | -4.288769 | -3.395230 | 0.388531 | -1.899021 |
| 1 | 4.0 | 4.331057 | 0.008865 | 0.009175 | 0.000158 | 0.009475 | -3.657982 | -3.109672 | 0.317354 | -2.561361 | -1.120739 | -0.003720 | 0.152041 | 0.735728 | -4.235378 | -3.439752 | 0.328276 | -2.288562 |
| 2 | 7.0 | 7.093725 | 0.008918 | 0.009232 | 0.000159 | 0.009533 | -3.657997 | -3.109685 | 0.317355 | -2.561372 | -0.823152 | 0.000050 | 0.091578 | 0.471964 | -4.237398 | -3.435940 | 0.326020 | -2.424142 |
| 3 | 10.0 | 9.856392 | 0.008980 | 0.009290 | 0.000159 | 0.009590 | -3.658013 | -3.109698 | 0.317356 | -2.561383 | -0.598072 | 0.000486 | 0.096002 | 0.764596 | -4.335532 | -3.435461 | 0.346501 | -2.123276 |
| 4 | 13.0 | 12.604671 | 0.009036 | 0.009348 | 0.000158 | 0.009646 | -3.658029 | -3.109712 | 0.317358 | -2.561394 | -0.627781 | 0.010036 | 0.143159 | 0.808039 | -4.390991 | -3.425868 | 0.363987 | -2.126139 |
Maximum value of corrections in azimuth direction
[20]:
ystats_df.abs().max()
[20]:
bIndex 306.000000
t 282.079880
geodetic min 0.013753
mean 0.013995
std 0.000167
max 0.014231
bistatic min 3.661955
mean 3.113049
std 0.376610
max 2.564143
fmrate min 1.957247
mean 0.127447
std 0.906995
max 1.909569
sum min 5.042909
mean 3.481153
std 0.979676
max 2.857225
dtype: float64
Plot
[21]:
iw3 = product['IW3']
iw3_df = ystats_df[ystats_df[('', 'bIndex')].isin(iw3.burst_list)]
t = iw3_df[('','t')]
fmrate_df = iw3_df.loc[:,'fmrate']
fmrate_df.insert(0, 't', t)
fmrate_df = fmrate_df.sort_values(by='t')
plt.figure()
fmrate_df.plot(x='t', style='o-')
plt.ylabel('t [s]')
plt.ylabel('[m]')
plt.title('IW3 - FM Rate correction (azimuth)')
plt.grid()
<Figure size 432x288 with 0 Axes>
[22]:
xstats_df.iloc[xstats_df['sum', 'max'].abs().argmax()]
[22]:
bIndex 306.000000
t 282.079880
tropospheric min 3.076796
mean 3.189554
std 0.065287
max 3.303904
ionospheric min 0.014389
mean 0.014750
std 0.000194
max 0.015121
geodetic min 0.088957
mean 0.091113
std 0.001214
max 0.093352
doppler min -0.471009
mean 0.000176
std 0.266898
max 0.468082
sum min 2.723540
mean 3.305272
std 0.274704
max 3.867100
Name: 305, dtype: float64
[23]:
ystats_df.iloc[ystats_df['sum', 'max'].abs().argmax()]
[23]:
bIndex 241.000000
t 222.236469
geodetic min 0.012263
mean 0.012542
std 0.000145
max 0.012807
bistatic min -3.660727
mean -3.112005
std 0.317592
max -2.563283
fmrate min -0.203718
mean -0.006160
std 0.029448
max 0.167373
sum min -4.187425
mean -3.441411
std 0.318898
max -2.857225
Name: 80, dtype: float64
[24]:
xstats_df.loc[xstats_df['']['bIndex'] == 306]['sum']
[24]:
| min | mean | std | max | |
|---|---|---|---|---|
| 305 | 2.72354 | 3.305272 | 0.274704 | 3.8671 |
[25]:
ystats_df.loc[ystats_df['']['bIndex'] == 241]['sum']
[25]:
| min | mean | std | max | |
|---|---|---|---|---|
| 80 | -4.187425 | -3.441411 | 0.318898 | -2.857225 |