Overview of the Calibration/Validation ofOverview of the Calibration/Validation of

Pascal Bonnefond1, Bruce Haines2

1Observatoire de la Côte d’Azur, Grasse, France2JPL, CALTECH, Pasadena, USA

Ocean Surface Topography Science Team MeetingJune 19, 2008

Buellton, California

2

Jason-2CALVAL

Overview

OSTM/Jason-2, OSTST meeting, June 19, 2008

CONTEXT

Long-term observations of the Mean Sea Level

  Long, accurate and homogeneous time series is needed  Lifetime of satellites is nominally about five years=> The link (calibration and validations) between missions is essential

Jason-1 launch

End of T/P mission

Jason-2 launch

— TOPEX/Poseidon alone— TOPEX/Poseidon and Jason-1— Jason-1 alone

Slope = +3.1 ±0.4 mm/yr

Source: A. Cazenave, CLS/LEGOS

-50

0

50

100

150

200

250

1992 1994 1996 1998 2000 2002 2004 2006 2008

Global Sea Level from T/P and Jason-1: UNCALIBRATED Record for 1992—2008

Mea

n G

loba

l SS

H (

mm

)

T/P: ALT A SideRate = +2.7 mm/yr

T/P: ALT B SideRate = +3.9 mm/yr

Jason-1: Old (A) GDR Rate = –1.9 mm/yr

143 mm13 mm

Jason-1: Current (B) GDRRate = +2.7 mm/yr

26 mm

Source: B. Haines, JPL

FROM SPACE TO SEA

Orbit (gravity field, tracking data, …)

Range (bias, retracking, …)

Corrections

Ionosphere

Troposphere (Dry and Wet)

Sea State Bias

Other corrections

Calibration/Validation and data consistencyCalibration/Validation and data consistency

Toward a better consistency between grand father, father and sonToward a better consistency between grand father, father and son

But also in terms of stability (models, instruments, …)But also in terms of stability (models, instruments, …)

In terms of Sea Surface Height biasIn terms of Sea Surface Height bias

Goal is to link altimetric missions at few mm and below 1 mm/yr level

4

Jason-2CALVAL

Overview

OSTM/Jason-2, OSTST meeting, June 19, 2008

INTRODUCTION

Definition of altimeter bias calibration:sea height bias = altimeter sea height - in situ sea height  

Sea height bias < 0 meaning the altimetric sea height being too low (or the altimeter measuring too long)Sea height bias > 0 meaning the altimetric sea height being too high (or the altimeter measuring too short)

Historical overviewAbsolute calibration of radar altimeters has been used previously for numerous missions such as:

- Seasat in Bermuda (1978)- ERS-1 in Venice (1991)- TOPEX/Poseidon in Lampedusa (1993)- TOPEX/Poseidon and Jason-1 on the Harvest platform (since 1992)- TOPEX/Poseidon, Jason-1 and EnviSat in Corsica (since 1998)

ConceptThe calibration principle is to compute the difference between the Sea Surface Height (SSH) measured with the altimeter and the SSH recorded by local sensors (tide gauge, GPS-buoy, …) or other satellites. The Calibration/Validations activities are not only focused on the altimeter bias value but also on its origin through the validations of all the instruments and models (radiometer, orbit, …).

ABSOLUTE CALIBRATIONABSOLUTE CALIBRATION(from in-situ measurements)(from in-situ measurements)

•Provide absolute biases Provide absolute biases

•Enable separation of error sources Enable separation of error sources using information from multiple in using information from multiple in situ sensors (e.g., tide gauge, GPS, situ sensors (e.g., tide gauge, GPS, radiometer)radiometer)

•Highly sensitive to Geographically Highly sensitive to Geographically Correlated ErrorsCorrelated Errors

RELATIVE CALIBRATIONRELATIVE CALIBRATION(from other satellites)(from other satellites)

•Statistically significantStatistically significant

•Less sensitive to Geographically Less sensitive to Geographically Correlated ErrorsCorrelated Errors

•Provide only relative behaviorProvide only relative behavior

5

Jason-2CALVAL

Overview

OSTM/Jason-2, OSTST meeting, June 19, 2008

• CNES calibration site established in 1998

• Supports continuous monitoring of Jason-1 (and formerly T/P)

• Employs distributed configuration

– Fiducial point near Ajaccio equipped with GPS/FTLRS/DORIS.

– Senetosa coastal site (along ground track) equipped with tide gauges.

– Open-ocean verification point for GPS buoy deployments.

• Open-ocean altimeter readings connected to tide gauges via detailed local geoid model

– Derived from intensive GPS buoy and catamaran surveys along ground track.

• Extension to Ajaccio (2005) and Capraia (2004)

– EnviSat, ERS, GFO, Jason-1.

Corsica Calibration SiteHarvest Calibration Site

• Established by NASA in 1992.– Ten years of T/P overflights ended 8/13/2002

(Cycle 365)– Continuous monitoring of Jason-1 since 2002

• Active drilling platform (Arguello). – Lies directly on Jason-1 ground track.– Located in open ocean (~10 km from land) at

western entrance to Santa Barbara (CA) channel.

• Employs concentrated configuration.– All instrumentation (e.g., GPS, tide gauges,

radiometer) collocated at platform.– Jason-1 radar altimeter illuminates the

platform as s/c passes overhead.

33  

34 

35 

36 

221 (Asc)

43 (Asc)

119 (Asc)130 (Des)

206 (Des)

HarvestLos Angeles

HARVEST

JASON TRACK

LOS ANGELES

CALVAL

SITES

Why Distributed Calibration Sites?Geographically Correlated Errors

CORSICA(Bonnefond et al.)

HARVEST(Haines et al.)

GAVDOS(Pavlis, Mertikas et al.)

BASS STRAIT(Watson et al.)

Absolute Calibration Experiments

6

Jason-2CALVAL

Overview

OSTM/Jason-2, OSTST meeting, June 19, 2008

T/P

&

JASON|1

BIASES

Absolute Long-Term Sea-Surface Height Bias Estimates~16 Years of Monitoring at Harvest, 4 Altimeter Systems(Haines et al.)

~10 Years of Monitoring in Corsica, 4 Altimeter Systems(Bonnefond et al.)

-200

-100

0

100

200

300

400

500

1993 1995 1997 1999 2001 2003 2005 2007

Bia

s (m

m)

LAD Drift(mm/yr)

LSQ Drift(mm/yr)

LAD Bias(mm)

LSQ Bias(mm)EpochAltimeter

+3.2+1.2 ± 1.5–2.6+4.0 ± 4.71993.0TOPEX-A–2.8–0.6 ± 2.7+2.8–1.7 ± 10.81993.0POSEIDON-1–4.3–1.8 ± 3.3+4.5+7.3 ± 5.42002.0TOPEX-B–0.6–0.1 ± 1.3+90.8+88.4 ± 4.52002.0JASON GDR "C"

T/P (ALT-A) T/P (ALT-B)T/P (POSEIDON-1)JASON-1 (POSEIDON-2)

Relative altimeter biases are also monitored at a global scale through:- Comparisons to a selected tide gauges network (Dorandeu et al., Miller et al., Nerem et al., …)- Comparisons with other altimetric missions (Bosch et al., Dorandeu et al., …)

7

Jason-2CALVAL

Overview

OSTM/Jason-2, OSTST meeting, June 19, 2008

WET TROPOSPHERE

Calibration of Vertical Wet Path DelayUsing data from in-situ GPS receivers and local meteorological sensors

This study has permitted to confirm the unexpected behaviorof Jason-1 Microwave Radiometer

(in the first release of Jason-1 data, GDR-A)

Wet troposphere is one of the key corrections (highly variable in time and space):=> Lots of other Calibration/Validations activities are dedicated to Wet Path Delay monitoring at the global and local scales.(Brown et al., Eymard et al., Fernandes et al., …)

8

Jason-2CALVAL

Overview

OSTM/Jason-2, OSTST meeting, June 19, 2008

FORMATION FLIGHT

A unique opportunity to cross compare all the corrections A unique opportunity to cross compare all the corrections and the derived Sea Surface Heightand the derived Sea Surface Height

Systematic sea-surface height errors revealed by flying Jason-1 in formation with TOPEX/POSEIDON (for ~200 days)

Harvest Corsica

Bass Strait

129 141 153 165 177 189

(mm)

129 141 153 165 177 189

(mm)

ASC. TRACKS = 16 mm

DES. TRACKS = 12 mm

70 seconds

(~400 km)

Same strategy for the first six month of

OSTM/Jason-2 mission

Jason-2 will be placed ~60 seconds

behind Jason-1

9

Jason-2CALVAL

Overview

OSTM/Jason-2, OSTST meeting, June 19, 2008

DATA

CONSISTENCY

No SSB, old orbits, old ranges

No SSB, new orbits, old ranges

No SSB, new orbits, new range for T/P (LSE)

No SSB, new orbits, new ranges for T/P (LSE) and Jason-1 (MLE4)

New SSB for T/P and Jason-1, new orbits, new ranges for T/P (LSE) and Jason-1 (MLE4)

New SSB for T/P and Jason-1, consistent orbits for T/P and Jason-1, new ranges for T/P (LSE) and Jason-1 (MLE4)

Sea Surface Height differences - mean bias (cm)

T/P – Jason-1 Sea Surface Heights during the Formation Flight Phase

Source: M. Ablain, CLS

10

Jason-2CALVAL

Overview

OSTM/Jason-2, OSTST meeting, June 19, 2008

CONCLUSION

Calibration/Validations Activities:- A joint effort involving large and various scientific and technical communities- Knowledge acquired over 16 years through T/P and Jason-1- But also lots of things to better understand

32 selected OSTM/Jason-2 proposals include Cal/Val studies (CNES and NASA):

3 on the orbit

3 on the retracking

4 on the Sea State Bias, waves and winds

5 on the Wet tropospheric correction

6 on the absolute calibration

10 on the global validations and data consistency (multi-missions)

11 on various modeling (Mean Sea Level, Ocean Circulation, Tides, …)

11 on coastal, shelf and inland waters

This particularly addresses the performance of the OSTM/Jason-2 measurement

system over inland waters thanks to the new POSEIDON-3 altimeter tracker mode

Improvement in these fields are very important for the scientific communities but will also help to prepare missions such as SWOT