CryoSat-2 SARIn mode success to determine lakelevel variations
Sh. Roohi, N. Sneeuw
Institute of Geodesy, University of Stuttgart, Germany
Abstract
To determine realistic lake water level variations, corrupted SARIn waveformsneed to be corrected. We use different retracking algorithms to retrack thewaveforms and to correct water level variations. We assessed the performanceof SARIn mode of CryoSat-2 against external data, e.g. in-situ gauge. In thisstudy we use SARIn level 1 and 2I data from October 2010 to 2014 over Nasserlake, located in Southern Egypt. We analyzed different retracking scenarios overfull and sub-waveforms and based on external validation the performance of eachretracker was evaluated.
Introduction
CryoSat-2 mission was originally designed to study the ice fluctuation in conti-nental ice sheet and ice covered marine of the Earth, especially in Arctic region.But nowadays it is being used to measure water level changes even over in-land water bodies. CryoSat-2 is equipped with the SIRAL radar altimeter thatcollects measurements from the Earth surface in three modes. Over the openocean, flat ice sheets and land it plays the role of pulse limited altimeters, LRMmode. SIRAL is running in SAR mode over sea-ice, coastal zone and some inlandwater bodies. Generally, the mission operates in SARIn mode over continentalsloped ice sheets but it also makes observations over other areas such as lakesand coastal zones. In this mode of observation instead of a pulse, a burst ofpulses is emited from the radar therefore we have more samples during a givenecho, 512 samples. Even though SARIn mode footprint size is small, some ofthe reflected signal back to the radar (waveform) can be contaminated by landand vegetation effects.
CryoSat-2 coverage [CryoSat-2 handbook,2013]
Launch 8 April 2010Mission duration 3 years
Orbit height 717 kmInclination 92 ◦
LRM Ocean and landSAR Coastal zone and sea ice
SARIn Slopped ice-sheet
Retracking
SARIn waveforms were retracked by following retracker algorithms:I OCOG (Offset Center Of Gravity)I Threshold with different threshold valuesI 5-β parameterI Brown model
∆Rretracking = (Gr − G0) ×c
2τ
Gr: Retracked gate G0: Nominal retracking gatec: Light velocity τ : Pulse duration
Retracked water level
Water level anomaly estimation:I Defining water level time series from median values of water level for each
satellite passI Rejecting outliers from the long time series by fitting the following model:
h(ti) = a + bti + ct2i + d sin
(2π
Tti
)+ e cos
(2π
Tti
)a, b, c, d, e : Unknown parameters T : Annual period h : Retrackedwater height
I Validation in front of available in-situ gauge data
Data and area of studying
Data: SARIn mode level 1B and 2I (Oct 2010 – 2014) Area: Nasser lake
31 31.5 32 32.5 33 33.521.5
22
22.5
23
23.5
24CS−2 SARin mode sub−satellite ponits on the Nasser lake
Longitude [deg]
Latit
ude
[deg
]
32.4 32.45 32.5 32.5522.66
22.68
22.7
22.72
22.74
22.76
22.78
22.8 CS−2 SARin mode sub−satellite ponits on the Nasser lake
Longitude [deg]
Latit
ude
[deg
]
Along track waveform variations
0100
200300
400500
600
22.5
23
23.5
24
24.5
25
25.50
1
2
3
4
5
6
7
x 104
Time or Bin Latitude [deg]
Pow
er
0100
200300
400500
600
22.5
23
23.5
24
24.5
25
25.50
1
2
3
4
5
6
7
x 104
Time or Bin Latitude [deg]
Pow
er
Along track waveform variations
Time or Bins
Latit
ude
[deg
]
0 50 100 150 200 250 300 350 400 450 50022.715
22.72
22.725
22.73
22.735
22.74
22.745
0
1
2
3
4
5
6
x 104
Along track waveform variations
Time or Bins
Latit
ude
[deg
]
0 50 100 150 200 250 300 350 400 450 500
22.72
22.725
22.73
22.735
22.74
22.745
22.75
0
1
2
3
4
5
6
x 104
Backscatter coefficient of SARIn mode over Nasser lake
Longitude [deg]
Latit
ude
[deg
]
32.48 32.49 32.5 32.51 32.52 32.53 32.54 32.55 32.5622.7
22.8
22.9
23
23.1
23.2
23.3
23.4
23.5
10
15
20
25
30
35
40
45
50
0100
200300
400500
600
22.5
23
23.5
24
24.5
25
25.50
1
2
3
4
5
6
7
x 104
Time or Bin Latitude [deg]
Pow
er
0100
200300
400500
600
22.5
23
23.5
24
24.5
25
25.50
1
2
3
4
5
6
7
x 104
Time or Bin Latitude [deg]
Pow
er
Along track waveform variations
Time or Bins
Latit
ude
[deg
]
0 50 100 150 200 250 300 350 400 450 500
22.72
22.725
22.73
22.735
22.74
22.745
1
2
3
4
5
6
x 104
Along track waveform variations
Time or Bins
Latit
ude
[deg
]
0 50 100 150 200 250 300 350 400 450 500
22.72
22.725
22.73
22.735
22.74
22.745
0
1
2
3
4
5
6
x 104
Peakiness of SARIn waveform over Nasser lake
Longitude [deg]
Latit
ude
[deg
]
32.48 32.49 32.5 32.51 32.52 32.53 32.54 32.55 32.5622.7
22.8
22.9
23
23.1
23.2
23.3
23.4
23.5
20
40
60
80
100
120
140
Rretracking result
RMS (cm) of different retracking scenarios respect to the in-situ gauge data
retracker full-waveform sub-waveform
mean-all first min-residual
ESA 44 – – –OCOG 89 104 96 94
Threshold 10% 87 93 65 53Threshold 20% 71 92 79 72Threshold 50% 51 92 102 945-β parameter 97 62 40 64
Brown 64 194 64 57
0
50
100
150
200Full−waveform
RM
S [c
m]
ES
A
OC
OG
Th1
0
Th2
0
Th5
0
Bet
a5
Bro
wn
0
50
100
150
200Sub−waveform (mean all)
RM
S[c
m]
OC
OG
Th1
0
Th2
0
Th5
0
Bet
a5
Bro
wn
0
50
100
150
200Sub−waveform (first)
RM
S [c
m]
OC
OG
Th1
0
Th2
0
Th5
0
Bet
a5
Bro
wn
Performance of different retrackers
0
50
100
150
200Sub−waveform (optimized)
RM
S [c
m]
OC
OG
Th1
0
Th2
0
Th5
0
Bet
a5
Bro
wn
2010.5 2011 2011.5 2012 2012.5 2013 2013.5−6
−4
−2
0
2
4
Time [year]
Wat
er le
vel a
nom
aly
[m]
Water level anomaly from CS−2 SARIn mode and in−situ gauge measurements
RMS=40 cmSatelliteIn−situ gauge
Conclusion
I Obviously waveform retracking techniques can improve the quality of altime-try data over inland water bodies.
I The quality of water level is dependent on the waveform retracking techniques.I Backscatter coefficients are changed too much over the lake surface which
leads to variety of waveform variations.I Full-waveform retracking: ESA retracker provides the minimum RMS.I Sub-waveform retracking: First sub-waveform retracked by 5-β parameter is
the best scenario.
Geodetic Week 2014, Berlin, Germany http://www.uni-stuttgart.de/gi [email protected]