Transcript Beckley_et_al_sea_le..

On the ‘cal mode’ correction to TOPEX altimetry and its effect on the global mean sea-level time series
B. Beckley– SGT Inc.
R. Ray – NASA/GSFC
D. Hancock – NASA/WFF
G. Mitchum – U. of South Florida
Abstract
90
80
Sea Height Variation (mm)
70
MEaSUREs v3.2
Global Mean Sea Level Variations
Glacial Isostatic Adjustment applied
1993.0 - 2016.55 linear rate = 3.41 +/- 0.4 mm/yr
60
Watson et al., 2015
Jason-2
50
Jason-1
40
TOPEX
Alt B
30
80
Watson et al., 2015
70 Global Mean Sea Level Variations
20 TOPEX Alt A
10
1993.0 - 2016.55 linear rate = 3.41 +/- 0.4 mm/yr
Sea Height Variation (mm)
Recently Watson et al. [2015] called attention to a possible
spurious (and apparently piecewise linear) drift during the
first six years of the T/P mission. In his analyses he
“calibrates” the 22-year altimeter derived sea surface height
time series by applying drift/bias estimates to each mission
via tide gauge sea level comparisons. The significant
resultant 0.6 mm/y reduction of the GMSL estimate is
reported to be due primarily to the drift/bias observed in
TOPEX Side A. As we show here, it seems likely that the
drift is an artificial one introduced into the system as an
independent correction based on an internal instrument
calibration. It affects only the TOPEX altimeter onboard the
T/P spacecraft. We interrogate the efficacy of the TOPEX
cal1-mode range bias correction and its impact on the GMSL
estimate with and without its application.
0
-10
60 Annual and semi-annual signal removed
Glacial Isostatic Adjustment applied
Jason-2
50
40
Watson et al., 2015
30
Jason-1
20
10
TOPEX Alt A
0
TOPEX
Alt B
-10
NASA/GSFC 09/29/2016
NASA/GSFC 10/04/2016
-20
1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
-20
1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
Global Mean Sea Level (GMSL) estimates from 1993 – mid-2016 derived from the NASA MEaSUREs v3.2 TOPEX/Jason-1&2 sea surface height
time series (update Beckley et al., 2010) are shown in image at left. A similar GMSL estimate (1993 – mid-2014) from Watson et al. is shown above
(solid grey line depicted as “unadjusted GMSL”). A 0.6 mm/y reduction is shown (solid black line) in the “adjusted GMSL” derived from the
application of mission specific bias drift estimates (above table) estimated from altimeter sea level comparisons to the tide gauge network shown
above (red dots indicate selected gauges).
Altimeter-Gauge Height (mm)
Current 64 sites
Expanding to 85
30
Altimetry - Gauge rate = 0.26 mm/y
= 4.5 mm
20 Standard deviation
10
0
-10
-20
0.97 mm/y
5.0 mm
-30
1992
1996
0.61 mm/y -0.33 mm/y
4.2 mm
4.3 mm
2000
2004
Year
-0.22 mm/y
3.6 mm
2008
2012
2016
80 Global Mean Sea Level Variations
1993.0 - 2016.0 linear rate = 3.10 +/- 0.4 mm/yr
70 1993.0 - 2016.0 linear rate = 2.72 +/- 0.4 mm/yr
70 Global Mean Sea Level Variations
60 1993.0 - 2002.0 linear rate = 3.31 +/- 0.4 mm/yr
1993.0 - 2002.0 linear rate = 2.09 +/- 0.4 mm/yr
50 TOPEX cal1-mode
range bias not applied
40
30
20
10
TOPEX cal1-mode range bias
applied
Sea Height Variation (mm)
Sea Height Variation (mm)
80
30
Altimetry - Gauge rate = -0.02 mm/y
= 4.3 mm
20 Standard deviation
10
0
-10
-20
0.19 mm/y
4.6 mm
-30
1992
1996
0.30 mm/y -0.26 mm/y
4.0 mm
4.4 mm
2000
2004
Year
2008
-0.19 mm/y
3.6 mm
2012
2016
Altimeter stability estimates are provided by sea level height variation comparisons to a 64-site tide gauge network (update Mitchum, 2000).
Implementation of recent improvements to vertical land motion (VLM) corrections based on GPS ULR5 series (update Santamaría-Gómez,
2012) have mitigated VLM errors. Drift estimate of altimeter time series with TOPEX cal1-mode correction applied (left figure) reveals
TOPEX Side A cal1-mode quadratic signature in the range bias correction. Improved agreement of altimeter SSH (with TOPEX cal1-mode
correction NOT applied) with tide gauges is shown in above-right image.
The TOPEX instrument had an internal calibration tracking mode that was designed to detect any significant drift in the
altimeter range owing to the internal electronics of the system (left image). TOPEX had two calibration modes for both the Kuband and C-band radars [Marth et al., 1993]. So-called “Cal-1” was designed to detect internal electronic path delays which
would directly translate into errors in range (upper-right image). By design the TOPEX calibration loop bypassed the normal
signal pathways as near as possible to the antenna so it was truly measuring the internal path delay in the altimeter system.
There was a remote possibility that the inferred drift in path delay was actually caused by the calibration loop itself. Were that
to occur the only indication would be through an external assessment, such as comparison of the altimetric sea levels to tidegauge measurements. Based on current altimeter versus recently improved tide-gauge measurements (lower-right image), this
possibility now seems less remote.
90
Altimeter-Gauge Height (mm)
0.82 mm/y
References
Beckley, B.D., N.P. Zelensky, S.A. Holmes, F.G. Lemoine, R.D. Ray, G.T. Mitchum, S. Desai, S.T. Brown, Assessment
of the Jason-2 Extension to the TOPEX/Poseidon, Jason-1 Sea-Surface Height Time Series for Global Mean Sea Level
Monitoring, Marine Geodesy, 33(S1): 447-471, 2010, Supplemental Issue on OSTM/Jason-2 calibration/validation, Vol.
1, DOI: 10.1080/01490419.2010.491029.
Marth, P. C., J. R. Jensen, C. C. Kilgus, J. A. Perschy, J. L. MacArthur, D. W. Han- cock, G. S. Hayne, C. L. Purdy, L. C.
Rossi, and C. J. Koblinsky (1993), Prelaunch performance of the NASA altimeter for the TOPEX/POSEIDON project,
IEEE Trans. Geosci. Remote Sens., 31(2), 315–332.
Mitchum, G.T ., An improved calibration of satellite altimetric heights using tide gauge sea levels with adjustment for
land motion. Marine Geodesy , 23, 145-166., 2000.
2.83 +/- 0.4 mm/yr
1993.0 - 2014.5 linear rates
60 Annual and semi-annual signal removed
Glacial Isostatic Adjustment applied
Nerem, R., D. Chambers, E. Leuliette, G.T. Mitchum and B. Giese, Variations in global mean sea level associated with
the 1997-98 ENSO event. (1999) Geophys. Res. Letters, 26, 3005-3009.
Santamaría-Gómez, A., M. Gravelle, X. Collilieux, M. Guichard, B. Martín Míguez, P. Tiphaneau, and G. Wöppelmann
(2012), Mitigating the effects of vertical land motion in tide gauge records using state-of-the-art GPS velocity field,
Global Planet. Change, 98-99, 6–17.
50
Jason-2
40
Watson, C. S., N. J. White, J. A. Church, M. A. King, R. J. Burgette, and B. Legresy (2015), Unabated global mean sealevel rise over the satellite altimeter era, Nature Clim. Change, x, doi:10.1038/nclimate2635.
3.25 +/- 0.4 mm/yr
30
Jason-1
20
10
0
0
-10
-10
TOPEX
Alt B
-20 TOPEX Alt A
1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015
GMSL estimates derived with (red curve) and without (blue curve) application of the TOPEX cal1-mode range bias corrections are shown in
the left image. Here seasonal signals are retained and GIA has not been applied. The 0.28 mm/y reduction in GMSL over the 1993-2016
period in the estimate with the TOPEX cal1-mode NOT applied is consistent with the overall 0.26 mm/y tide gauge drift estimate shown
above. GMSL estimates (1993 - mid-2014) in the right image have annual and semi-annual signals removed and GIA applied for direct
comparisons to Watson et al solutions. Our solutions show an estimated GMSL reduction of 0.4 mm/y as compared to the 0.6 mm/y reduction
in the Watson et al solutions. Note, our solution does not “calibrate” the altimetry with tide gauge based drift/bias estimates.
-20
1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
Previous studies have shown the high correlation of GMSL variations with ENSO (Nerem et al., 1999). The upper left image shows the
MEaSURES v3.2 GMSL variations (trend removed) versus the NOAA multivariate ENSO index. Note the better agreement during 2010/11 El
Nino/La Nina transition as compared to the 1997-98/99 transition. TOPEX regional sea level trends are shown above for the period 1993-2002
with (left image) and without (right image) application of the cal1-mode correction. Note the more pronounced La Nina signature in the sea
level trends in the solution with Cal1-mode NOT applied.
2016 Ocean Surface Topography Science Team Meeting, La Rochelle, France.