Earth Exploration Satellite Service (passive) at 1.4 GHz ... · PDF fileEarth Exploration Satellite Service (passive) at 1.4 GHz : the SMOS experience. Yann H. Kerr*, Vincent Meens**,

SMOS RFI – June 20161

Earth Exploration Satellite Service (passive) at 1.4 GHz :

the SMOS experience.

Yann H. Kerr*, Vincent Meens**,Philippe Richaume*, Eric Anterrieu*, François Cabot*, Jean Pla**,

Alexandre Vallet***

* CESBIO **CNES ***Agence Nationale des FRequences

SMOS RFI – June 20162

The SMOS mission and observation system in brief

• Radio-interferometer operating in the L-band @ 1413MHz within the ITU protected 1400 –1427 Mhz band (ITU RR 5.340)

• 69 elementary antennae (LICEF) distributed along Y shape arms• 3dB beamwidth= ~70° = whole Earth + sky• The image reconstruction forms synthetic antennas ~ 2.5° beam

= ~40 km on the Earth at 755 km altitude Allows Brightness Temperatures (BTs) images (SNAPSHOTs) over 1500km x 1000

km every 1.2 sec largely overlapping Full polarimetic BTs in polarization XX, YY and XY 2.5K to 5K of radiometric accuracy

• Boresight tilted 32° ahead Asymmetric Field Of View (FOV)

Nadirdirector cosine

SMOS motion

SMOS RFI – June 20163

•Each integration time, (2.4 s) a full scene is acquired (dual or full pol)

•Average resolution 43 km, global coverage•A given point of the surface is thus seen with several angles

•Maximum time (equator) between two

acquisitions 3 days

Principle of operations

SMOS FOV; 755 km altitude

P. Waldteufel, 2003

SMOS RFI – June 20164

SMOSPrinciple and key points

SMOS RFI – June 20165

Why L Band – Why SMOS

L band only protected frequency to access soil moisture and Sea surface salinity

SMOS used for Weather forecast Extreme event prediction (drought index, rainfall estimates) Hurricane monitoring Flood /flash flood alerts Irrigation and water management Sea ice ….

Stay tuned http://www.cesbio.ups-tlse.fr/SMOS_blog/

SMOS RFI – June 20166

The RF Interference environment

frequency

Active band

Unwanted emissions

OoB domain Spurious domain

Passive band

1427 MHz1400 MHz

Wanted signal(Brightness temperature)

Interfering signals

SMOS uses only 19 MHz of the available 27 MHz “to be on the safe side”

SMOS RFI – June 20167

Radio frequency interferences examples as seen by SMOS• Brightness Temperatures usually far too high – but not necessarily all the time• As much important, brightness Incidence angle profile can be severely damaged

Soon after launch the protected L-band did not appear so protected

1000 K TBX1040 K TBY

327 K TBX362 K TBY

320 K TBX306 K TBY

43200 K TBX5683 K TBY

300K

240K

180K40K

-30K0K

64000K

10000K

0K

-2500K

SMOS RFI – June 20168

Tilted antenna plane asymmetric RFI impacts depending on orbit passes

http://www.cesbio.ups-tlse.fr/SMOS_blog Data RFI Monitoring RFI Probability

30%

00%

05%

10%

15%

20%

25%

Ascendingorbits

P Richaume

SMOS RFI – June 20169

http://www.cesbio.ups-tlse.fr/SMOS_blog Data RFI Monitoring RFI Probability

Tilted antenna plane asymmetric RFI impacts depending on orbit passes

P Richaume30%

00%

05%

10%

15%

20%

25%

Descendingorbits

SMOS RFI – June 201610

RFI impact on retrievals total destruction of data to low level (halo) contributions giving erroneous values

This situation started 20120822 and lasted for several weeks

SMOS RFI – June 201611 J. Boutin

SMOS RFI – June 201612 J. Boutin

SMOS RFI – June 201613

SMOS Brightness Temperature observations in Japan

3 Sep 11 21 Sep 11 9 Oct 11 27 Oct 11 9 Oct 11 27 Oct 11

1. SMOS revisits Japan twice every 3 days (one in descending direction + one in ascending direction). A clear change was observed between the

21st September and the 24th September.

24 Sep 11 6 Oct 11

SMOS RFI – June 201614

The Japanese RFI case

0 20 40 60 800

5

10

15

20

25

30

Months from January 2010

Num

ber o

f RFI

RFI #testeventsmax per pass

3G TV-2 4G TestTV-1

SMOS RFI – June 201615

RFI Detection approaches

Sources can be quite different Radars TV Mobile Transmission links

All based on physical signals Brightness temperature level (unphysical)

o Many different criteria (overall scene temperature, variations with surroundings, temporal stability, etc… )

Abnormal angular signatures or polarisations Etc..

And accurate geolocalisation of sources when strong and notdiffused (Between a few 100 m do 4-5 km)

SMOS RFI – June 201616

RFI Action status

Continuous activities Regular updating of RFI Continuous update of maps on the blog Continuous tracking of sources as per requests Monthly meetings High accuracy localisation (EA – IRAP)

Coordination with other entities NASA SMAP and Aquarius satellites Informing through ESA administrations of RFI sources on their

territories Search for improvements

Using the 6 years to get accurate (?) position on permanent sources (PR)

Use of other indicators to fine tune RFI pollution (PR-AlM)

SMOS RFI – June 201617

R. Oliva

SMOS RFI – June 201618

R. Oliva

SMOS RFI – June 201619

Evolution of achievements in Switching Off RFIs (R. Oliva)

- Improved detection algorithms have allowed ESA to provide better information to the national authorities.

- This was reflected in the corresponding improvement in in the situation in Europe. - The current limitation is the lack of response of several countries around the World.

SMOS RFI – June 201620

Conclusion• RFI: a real issue

• At best diminishes data quality• Often makes « holes » in the data set• No geophysical retrievals possible in some areas

• Good news• SMOS « discovered » the amplitude of the problem so Aquarius

and SMAP could capitalise on it and improve detection BUT RFI still present and data quality degraded or data missing

• Several souces were suppressed and progresses are still made

• Bad News• Situation evolves slowly• Detection is tricky considering the number of source types

• Detection algorithms well identified and starting to be accurate

• Coordinated actions with complementary systems i.e., SMOS +SMAP. Work required at National and international level

SMOS RFI – June 201621

Example of Hydrology from SMOS

A. Mialon, S. Bircher

SMOS RFI – June 201622

An example of SMOS Flood  Risk Forecast

RainfallforecastRainfallforecast

SMOS SM L3 products

Rainfallprobabilities

Rainfallprobabilities

SMOS Soilmoisture

probabilities

SMOS Flood Risk

SMOS Flood Risk

PrecipitationFlood Risk

PrecipitationFlood Risk

Flood Risk(Precip + SMOS)

Flood Risk(Precip + SMOS)

Leveraging flood risk based on SMOS soil moisture knowledge

Al Bitar A., Chone A., S. K. Tomer, Kerr Y.  CESBIO

Methodology

SMOS RFI – June 201623

An example of SMOS Flood  Risk Forecast

Al Bitar A., Chone A., S. K. Tomer, Kerr Y.  CESBIO