→
NCEO/CEOI-ST Joint Science Conference. University of Sheffield, June 25, 2014
FLEX Mission: Measuring Fluorescence from Space
Luis Alonso, Jose Moreno
LEO Group, LPI, University of Valencia (Spain)
and the FLEX Team
ESA’s living planet programme
Explorer 8 mission
candidates under
study since 2012
•
CarbonSat
Monitoring CO2
content of the
atmosphere (sources,
sinks, hot spots)
•
FLEX
FLuorescence
EXplorer
Monitoring vegetation
fluorescence –
the breath of the
Earth
Slide 2
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
FLEX – mission objectives
•
Primary scientific objectives
 Accurate global mapping of vegetation
fluorescence (emission spectrum)
 Estimate the actual global photosynthetic
activity via related light absorption and
emission
 Vegetation health status / stress identification
 Enhance knowledge on anthropogenic
impacts associated to land use changes
•
Secondary scientific objectives
•
Understanding of the terrestrial component of
•
Improvement of GPP modelling
Slide 3
the CO2 cycle
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
-1
-2
Radiance (W m sr
What is fluorescence?
0.20
0.15
WR with filter
0.10
0.05
B
0.12
-1
Radiance (W m sr nm )
leaf without filter
0.10
-1
-2
Incident solar light can be either reflected,
absorbed or transmitted
by a leaf.
0.00
0.14
0.08
0.06
0.04
leaf with filter
Fs
0.02
0.00
400
500
600
700
Wavelength (nm)
The absorbed light photons can be used for
photosynthesis or can be dissipated.
Dissipation can occur as
Non-Photochemical Quenching (e.g. heat)
or emitted as fluorescence (SIF) photons.
Slide 4
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
800
900
Dynamical fluorescence emission
Fluorescence emission spectra from two different species
exposed at two levels of air pollution
0.000020
Platanus
•
Total Fluo Yield [1/nm/sr]
Low Pollu on
High Pollu on
0.000015
0.000010
650
0.000020
700
750
Wavelength (nm)
800
850
Phoenix Canariensis
Total Fluo Yield [1/nm/sr]
Low Pollu on
High Pollu on
0.000015
Environment
–
Plant structure
–
Stress
Two peaks of the
fluorescence emission
are highly variable
•
Difference in shape
between pollution
levels, but also
between species
Ref: Van Wittenberghe et al. Upward
and downward solar-induced
chlorophyll fluorescence yield
indices of four tree species as
indicators of traffic pollution in
Valencia
0.000010
0.000005
0.000000
650
Slide 5
–
•
0.000005
0.000000
Fluorescence is
influenced by
700
750
Wavelength (nm)
800
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
850
ESA’s FLEX Earth Explorer
Environ Pollut, Vol 173, 2013
http://dx.doi.org/10.1016/j.envpol.201
characteristics have been created to test the
atmospheric correction algorithm.
Measuring a faint signal FSIF
+
W. Verhoef, H. Back. Simulation of hyperspectral and directional radiance
Images using coupled biophysical and atmospheric radiative transfer models.
•
TOA
LTOA
–
Light seen at the top of the
atmosphere (instrument input radiance)
LTOA
•
•
P
é (Edir × cos(J ill ) + Ediff )× rSurf
ù
+ FSIF ú × t atmosphere
ê
p
û
= L0 + ë
1- Salbedo × rSurf
L0
–
S
S
P
BOA
Atmospheric path radiance
S
Edir/diff
–
P
Diffuse and directly flux reflected by
Earth and transmitted through the
atmosphere
•
ρSurf
•
τatmosphere
•
Salbedo
–
–
–
Slide 6
O2-A
Water Vapor
Signal is modulated by the Earth surface
reflectance
O2-B
Sensitivity to atmospheric changes
Atmospheric pherical albedo
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Spectral resolution
Finer resolution: lower SNR
Coarser resolution: loss of spectral features
Slide 7
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
FLORIS band configuration
FLORIS-HR
FLORIS-LR
Band
PRI
Chlorophyll
absorption
 [nm]
500-600
600-677
O2-B
Red-edge
677-686
686-697
697-740
SR (FWHM)
3.0
0.7
0.3
2.0
SSI
2.0
0.5
0.1
1.0
SNR
300
400
300
600
Slide 8
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
O2-A
740-755
755-759
759762
0.7
0.5
Linear from
1200
600 to 1200
ESA’s FLEX Earth Explorer
762-769
0.3
200
0.1
Linear from
200 to 1200
769-780
0.7
+REP
0.5
+Cab
1200
FLEX / S3 tandem concept
o
FLEX will fly in tandem with
Copernicus’ satellite Sentinel3. Synergy between the
different instruments for:
FLEX
Atmospheric characterization
and correction
Vegetation biophysical
parameters
Surface temperature
o
o
o
Sentinel-3
OLCI
TOA Rad
FLEX
SLSTR
TOA Rad
FLORIS
TOA Rad
Dual View
Coregistration
Crosscalibration
Spectral
Calibration
FLORIS
TOC Etot
Atmospheric
State
Atmospheric
Correction
Cloud
Mask
Slide 9
FLORIS
TOC Refl
OLCI /
SLSTR
TOC Refl
Fluorescence
Retrieval
Biophysical
Retrieval
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
L2 Fluo
Map
L2 Photosynthesis
L2
BIOPHYS
Params
Altitude: ~815 km
Local solar time: 10:00 LTDN
ESA’s FLEX Earth Explorer
Temporal co-registration: < 6s (G) / 15s (T)
Two-Step full-spectrum
fluorescence retrieval
Slide 10
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Two-Step full-spectrum
fluorescence retrieval
Full Spectrum Retrieval
Initial SIF estimates at O2 absorptions
Requires (from Atmospheric Correction):
•
TOC Total Irradiance Etot and Êtot (modelled with no O2)
•
TOC Apparent Reflectance ρapp (includes SIF)
Fluorescence
True Reflectance
n-parameters to be fitted
initialization is required
Irradiance [mW/m2/nm/sr]
1200
Estimated ρapp
Interpolated no-O2 ρapp
True ρ
1000
800
600
400
200
0
680
Estimated Etot
Modelled no-O2 Etot
700
720
740
Wavelength [nm]
SIF687 & SIF760
ρ687 , ρ760
𝜆
760
780
𝐹𝑠 = 𝜌𝑎𝑝𝑝 − 𝜌𝑎𝑝𝑝 ∙
𝐸𝑡𝑜𝑡 ∙ 𝐸𝑡𝑜𝑡
𝜋 𝐸𝑡𝑜𝑡 − 𝐸𝑡𝑜𝑡 )
No assumptions about true reflectance ρ
(large source of error in other retrieval method)
Slide 11
𝐿𝑇𝑂𝐶 𝜆 − 𝐹 𝜆 − 𝜌 𝜆 𝐸𝑡𝑜𝑡 𝜆 ) 2
𝑚𝑖𝑛
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ρ errors <1% and SIF errors <10%
ESA’s FLEX Earth Explorer
End-to-End Mission Simulator
Scene location
FLORIS
OLCI
SLSTR (nadir)
SLSTR (oblique)
Latitude [deg]
45.17
45.16
FLORIS
Instrument
Module
45.15
45.14
SEN3
Instrument
Module
45.13
45.12
-0.46
-0.44
-0.42
-0.4
Longitude [deg]
Scene
Generator
SEN3
Geometry
Module
Class map
Slide 12
FLORIS
Geometry
Module
Biophysical
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
Level-1
Processing
Performance
Module
Level-2
Retrieval
Atmosphere
ESA’s FLEX Earth Explorer
Spacecraft and
payload accommodation
•
S/C
– ‘Small satellite’ – recurrent S/C bus
– Wet mass <1000 kg
– Power consumption <500 W
•
P/L
– Power consumption ~120W
– Weight ~120 kg
– Data rate ~150 Mbps
•
Launcher
•
Coverage
– Vega
– Global (land + major islands,
-56 to 75 degree latitude,
coastal zones 50 km
– Revisit time up to 19 days
Slide 13
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Instrument configurations
•
Phase B1 running
– Instrument concepts frozen
– Spectrometer refractive or catadioptric
– Reflective or transmissive grating
– One or two telescopes or spectrometers
FLORIS Concept 1
Slide 14
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
FLORIS Concept 2
ESA’s FLEX Earth Explorer
FLORIS breadboards: Objectives
•
Demonstration of the
manufacturability of the flight
FLORIS Concept 1
representative optical elements (lenses,
mirrors, coatings, filters, gratings, slits)
according to the specifications
•
Demonstration of the assembly
accuracy of the optical elements
according to the accuracies as defined in
the tolerance analysis
•
Verification of the alignment concept,
contamination budget, general
FLORIS Concept 2
accuracies and adequacy of the
measurement signals
•
Demonstration/measurement of the
imaging & spectral performance at
focal plane
•
Tolerance of the instrument to thermal
cycling
Slide 15
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Summary
•
FLEX will provide global mapping of terrestrial vegetation
•
Retrieval and measurement configuration will make use of
•
End-to-End Simulator with detailed implementation of every
fluorescence at moderate spatial resolution
full information content provided by FLORIS and Sentinel-3
aspect of the mission, allowing to test performance of each
module
•
Accuracy of fluorescence measurement expected to be 10%
•
•
FLORIS breadboards in construction and under testing
Slide 16
(as demonstrated by campaigns and by simulations)
Phases A/B1 of FLEX to be completed ~03/2015
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Thanks to the
FLEX Mission Advisory Group
•
•
•
•
•
•
•
•
•
Slide 17
Yves Goulas
Andreas Huth
Gina Mohammed
Franco Miglietta
Jose Moreno
Ladislav Nedbal
Wouter Verhoef
Uwe Rascher
Betsy Middleton (observer)
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Questions?
Would you like to know more?
www.esa.int
FLEX/PARCS
http://ipl.uv.es/flex-parcs/
Slide 18
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
FLEX/PHOTOSYN
http://www.flex-photosyn.ca/
ESA’s FLEX Earth Explorer
System / instrument requirements
Requirement
Instrument type
Specification
Pushbroom Imaging Spectrometer
Mission lifetime
Data latency
3.5 years (T) / 5 years (G)
5 h (T) / 24 h (G)
Coverage
-56o < latitude < 75o
Pitch angle
Swath width
Spatial Sampling Distance
Integrated energy
<5o
150 km
300 m
>70% over an area of 1.1 SSD
Spectral band coverage
Spectral Resolution and Sampling
500 nm – 780 nm
0.3 nm / 0.1 nm (HR) 2 nm / 0.65 nm (LR)
Signal to noise ratio
Spectral stability
> 200 at 761 nm
1nm
Over mission lifetime
10 pm for FLORIS-HR
During observational time of 1 orbit
Straylight sensitivity
0.1 SSI nm for FLORIS-LR
Specified for black & white scene
To maintain radiometric requirement
Spectral co-registration
Spatial co-registration
Knowledge of ISRF
Absolute radiometric accuracy
<0.1 SSD
<0.1 SSD
Better than 1%
2% (G) 3% (T)
Smile
Keystone
Relative accuracy
Polarisation sensitivity
1%
2% (FLORIS-LR)
Spectral and spatial
Temporal co-registration
1% (FLORIS-HR)
6s (G) / 15s (T)
Inter-channel temporal co-registration
<2s
Geo-location accuracy
0.3 SSD (with ground control points
Slide 19
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
Comment
Sun Zenith Angle < 75o
Observation Zenith Angle < 15o
Land area including coastal zones (50km)
Preferably Nadir pointing
Enables global coverage
@ Nadir
Excluding uncertainty of Sun radiance
To avoid significant cloud movements
ESA’s FLEX Earth Explorer
Disentangling fluorescence from TOA
radiance
•
Sensitivity analysis performed to
establish spectral ranges and
spectral resolutions
•
Competing effects
–
–
•
Slide 21
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
Atmospheric state variables

Aerosol Optical Thickness (AOT)

Aerosol height and model

Surface Pressure

Water Vapor
Vegetation reflectance
Disentanglement requires
–
High resolution for Oxygen B and
Oxygen A
–
Medium resolution red-edge
–
Medium resolution elsewhere
ESA’s FLEX Earth Explorer
Retrieval evolution
•
Promising method for fluorescence
retrieval:
–
independent decoupling of reflectance/fluorescence
effects and atmospheric/topographic effects
–
Apparent reflectance

ro’ = ro + Fs / LWLR

true reflectance derived by elimination of the
Oxygen absorption (no free fitting parameter anymore)

Incoming radiance LWLR from radiative transfer programs
•
Other retrieval methods, where reflectance is a free model parameter,
are currently under investigation
•
•
Reflectance expected to be a smooth function of wavelength
Retrieval accuracy expected to be about 10% for both
–
relative accuracy at the two emission peaks
–
total integral of the fluorescence radiance across the whole spectral range
(total energy emission)
Slide 22
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
FLORIS / S-3 band configurations
•
FLORIS (300 m SSD)
•
S-3 OLCI (300 m SSD)
–
Fluorescence
–
PRI
–
Camera 4 – 250 km swath
–
O1 to O4 (400 to 490 nm)
–
Atmospheric correction

Aerosols
–
Apparent reflectance

Chlorophyll

Reflectance
–
O9 to O16 (510 to 779 nm)
–
Context information

Slide 23
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
Cross calibration
ESA’s FLEX Earth Explorer
•
SLSTR (500m – 1 km SSD)
–
VIS S1 to S3 and F1/F2

–
SWIR S4 to S6

–
Aerosols
Reflectance
TIR S7 to S9

Temperature

Clouds (water vapour)
Breadboard 1: system test equipment
Slide 24
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Radiometric performance
•
High signal to noise (SNR) required 
•
Due to relatively large aperture, the
•
Other contributors have been
large aperture (~80 mm)
system is photon noise limited
minimised
SNRSample 
•
N ph ( )  bana  bdig
2
2
( N ph ( )  N dark  N smear )  bana  bdig  ( N readout
 N quant
)  bdig
Grating efficiency is essential
Slide 25
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Instrument response function
Spectral (approximation)
•
Spatial
Due to scattering of the optics and
optical elements, the IRF is more
sophisticated
•
Finite response at larger angles
Slide 26
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Straylight behavior – test case
•
Test of straylight sensitivity on black & white scene
(Lref & LTOC)
Slit
Slide 27
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Radiometric error due to straylight
•
•
Slide 28
Radiometric error ~4% at absorption line
Without correction
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Breadboard 1: telescope test equipment
Telescope illumination system
Lens barrel alignment tool
First lens under inspection
Knife edge scanner
Slide 29
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
Field scan mechanism
ESA’s FLEX Earth Explorer
Breadboard 1: manufacture completed
•
•
BB fully assembled
Telescope and
spectrometer line
shapes measured
•
Very good
performance achieved
so far
Slide 30
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Breadboard 2: system test equipment
BB optics fully
representative


BB and test equipment OGSE under definition

Detailed design review approaching
First grating samples tested
Slide 31
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer
Detector breadboard - CCD
•
Same detector for Concepts 1
•
Split frame transfer CCD,
•
Line transfer times of
and 2
1020 x 440 pixels (TBC)
~1.25 s, frame time of
45.8 ms
•
Pixel size 28  m x 42  m, full
•
Operating temperature 263 K,
•
Read noise <20e-, 2.5 MHz
•
ITT issued, under evaluation,
well capacity ~2 Medark current of ~0.1 nA/cm2
readout rate
start expected in June
Slide 32
NCEO/CEOI-ST Joint Science Conference 2014, Seffield
ESA’s FLEX Earth Explorer