MODIS Chlorophyll
Fluorescence
Ricardo Letelier, Mark Abbott,
Jasmine Nahorniak
Oregon State University
Outline
• Photosynthesis and fluorescence
• Measurement and validation of
chlorophyll natural fluorescence from
space
• Fluorescence and the estimation of sea
surface chlorophyll concentration
• Using fluorescence to estimate
variability in phytoplankton quantum
yield
Weekly FLH
Weekly chl_MODIS
Use of FLH
• Improve estimates of ocean Primary
Production through:
- Improving [chl] estimates in case
II waters
- Improving our capability to monitor the
variability in the mean physiological
state of algal assemblages in
surface waters
Light Harvesting, Fluorescence and
Photosynthesis
Light energy not used for
photosynthesis is lost as
heat and fluorescence
Fp + Ff + Fh = 1
u
2 e-
LHC
ADP+P ATP
Fluorescence
ff
heat
fh
NADP +
2H+
NADPH2
Blue light induced chlorophyll fluorescence in Tobacco leaf. A. photographed
in white light. B. taken in the low steady state of fluorescence, 5 min after the
onset of illumination. The bright red fluorescing upper part of the leaf is where
photosynthesis has been blocked by the herbicide duiron (DCMU).
(From Krause and Weis, 1988)
u
LHC
e-
PSI
(ATP & NADPH2)
L683
heat
u
DCMU
PSI
LHC
(ATP & NADPH2)
L683
heat
Fp + Ff + Fh = 1
0.016
0.014
Chlorophyll absorption
0.012
0.008
Increase in fluorescence
0.006
0.004
0.002
450
500
550
600
Wavelength, nm
FLH = Lu683 – Baseline
Baseline = Lu1 - [(Lu1-Lu2)/(lLu2-lLu1)]*(683-lLu1)
650
700
0.8
Lu683
0.7
0.6
Lu1
0.5
FLH
0
400
Exitance, W m-2 µm-1
Lu/Es
0.01
0.4
0.3
0.2
Lu2
0.1
0
600
620
640
660
680
700
Wavelength, nm
720
740
760
Absorption spectra for water, CDOM, and
phytoplankton
absorption spectra
0.08
3.0
pure water
pigments
gelbstoff
2.0
0.04
1.5
1.0
0.02
0.5
0.00
0.0
400
500
600
wavelength (nm)
700
800
aw (m-1)
ap, ag (m-1)
0.06
2.5
MODIS FLH bands: avoid oxygen absorbance at 687 nm
Weighting factor used to
compensate for offcenter FLH
0.6
0.6
9.9
9.9
0.55
Radiance, W m-2 µm-1 sr-1
D TOA spectra
(10 - 0.01 mg)
9.4
9.4
0.5
0.5
0.01 mg
0.45
8.9
8.9
0.4
0.4
0.35
8.4
8.4
Band #14
0.3
0.3
7.9
7.9
0.25
1.4%
7.4
7.4
660
660
665
670
670
675
680
680
685
690
690
Wavelength, nm
695
700
700
705
0.2
710
0.2
710
Normalized band transmittance and
difference between TOA Spectra, W m-2 m-1 sr-1
10 mg
0.14
100%
0.12
Radiance, W m-2 µm-1 sr-1
80%
0.1
70%
FASE FLH
60%
0.08
50%
0.06
40%
Actual Band 14 CW
0.04
30%
% difference
20%
0.02
10%
0
0%
674
675
676
677
678
679
Wavelength, nm
MODIS Specified Band 14 CW
680
681
682
% Difference between FASECODE FLH
and LOWTRAN FLH
90%
LOWTRAN FLH
SNR sensitivity =
0.012 W m-2 mm-1 sr-1
Requires correction at low chl concentrations due to
the convex behavior of the TOA signal between 667 and 683 nm
(From Frank Hoge
and Paul Lyon)
FLH vs.
chlorophyll
FLH vs.
CDOM
GLOBEC NEP AUGUST 2002
GLOBEC NEP AUGUST 2002
In situ chl a, mg m-3
(July 31st – August 19th)
MODIS chl a, mg m-3
Day x
Day 218
MODIS_FLH day x+1
Day 217
Day 212
Day 213
Day 211
Day 212
MODIS_FLH day x
GLOBEC NEP AUGUST 2002
FLH, W m-2 µm-1 sr-1
All cruise data
Only pixels of passes within 5 hrs of sampling time
[chl] = .021 + 43.4 FLH1.866
In situ chlorophyll, mg m-3
GLOBEC NEP AUGUST 2002
In situ chl
chlFLH empirical
(this study)
chlFLH semi-analytical
(Huot & Cullen
assuming ff = 0.006)
-Both FLH derived chl algorithms appear to slightly overestimate chl a fields.
-They do not seem to reproduce the low values observed in situ.
-Some of the differences between in situ and FLH derived could be due to time
differences and sampling depth (in situ = 5 m depth)
Natural (passive) Fluorescence
• F [chl](PARa*)FF
where
F = fluorescence
[chl] = chlorophyll concentration
PAR = photosynthetically available
radiation
a* = chlorophyll specific absorption
fF = fluorescence quantum yield
• Absorbed Radiation by Phytoplankton
ARP = a* x [chl] x PAR (calculated independently from [chl])
• F/ARP = Chl Fluor. Efficiency (CFE)  fF
• ARP / ([chl] x PAR) = a*
Huot & Cullen’s approach
• FLH = EPAR(0) [chl] aj*(512) Qa* j Cf
Kfabs + af
Where Qa* = aj*(678) / asol*(678)
We have to assume j constant
OSU Direct Broadcast October 04, 2001
MODIS_Chl
MODIS data shows
chl not always in
spatial
correspondence
with fluorescence
MODIS_FLH
MODIS_CFE
MODIS_ARP
Physiological parameters also vary spatially
m-2 mm-1 sr-1
,W
offshore
Fisher and Kronfeld
(1990) Assuming
CFE = 0.003
inshore
, mg m-2
5.0E-02
(Lu683-backscatter) / Chlest,
(µW cm-2 nm-1) / mg m-3
4.5E-02
4.0E-02
3.5E-02
3.0E-02
2.5E-02
2.0E-02
1.5E-02
1.0E-02
5.0E-03
0.0E+00
0
10
20
30
40
Ed490, µW cm-2 nm-1 sr-1
50
60
In Situ Observations of F/[chl] suggest it can
be a proxy for ff
Initial slope proportional to fF
0.1
0.08
Qnp
Ff
0.06
0.04
Ek
0.02
0
0
400
800
PAR, µmol quanta m-2 s-1
1200
1600
chlFLH empirical
(this study)
MODIS ARP
Huot & Cullen ARP
using in situ chl to
Derive an average ff
Thalassiosira weissflogii
Chemostat results 2001-2002
0.6
0.4
After 3 days of constant
cell counts
0.3
After 14 days
Fv/Fm, n.d.
0.5
0.2
0.1
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0.05
0.045
9 AM CFE, r.u.
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
m/mmax , n.d.
Where do we stand?
•
•
•
•
Field observations suggest that MODIS FLH is a
robust product.
Preliminary comparison of [chl]field vs FLHMODIS
suggest that FLH may prove of use to derive [chl] in
turbid waters. However:
- CFE was almost constant in the set of samples
used in this study. Is this the range of
values we should expect to see in natural
environments?
CFE validation requires that of FLH and ARP.
In order to interpret CFE we need field and
laboratory based work that explores the effect of
environmental variability and phytoplankton specific
composition.
In other words:
• FLH and CFE are very different MODIS
products in terms of validation.
- FLH is based on nLw at 678 nm after
baseline correction
- CFE is a proxy for Ff (a physiological
parameter) that requires the previous
validation of ARP ([chl] x a*).
- Further use of Ff to infer Fp requires
the characterization of the variability
in energy distribution within the
photosystem.
Acknowledgments
• Wayne Esaias (NASA/GSFC)
• Bob Evans, Kay Kilpatrick & Howard
Gordon (Univ. Miami)
Fluorescence Product Flags
•
•
•
•
•
Bit
Bit
Bit
Bit
Bit
6
7
8
9
10
FLH/chloro_MODIS > 1
FLH/chloro_MODIS > 0.5
FLH > 2
FLH > 1
chloro = -1
• Bit 11 ARP quality ≥ 2
• Bit 12 ARP quality = 1
• Bit 13 CFE > 0.1
Flags
• For FLH
- 0 if 6-10 are clear
- 1 if 7 and 9 are set but 6, 6, and 10 are clear
- 2 if 6, 8, or 10 are set
- 3 if any common flags are set
• For CFE
- 0 if common flags are clear and bits 11 and 12 are clear
- 1 if bit 12 is set or FLH quality is 1
- 2 if bit 11 or bit 13 is set or FLH quality is 2
- 3 if any common flags are set or if CFE > 0.15 or if
FLH quality is 3
End of talk
Photoprotective:Photosynthetic pigment ratio
127 ° W
45 ° N
126 ° W
125 ° W
124 ° W
123 ° W
New port
19
18
1.2
1.33871
23
24
22
1.13065
17
Heceta Head
Latitude
Coos Bay
43 ° N
1
PP/PS
0.8
Cape Blanco
0.6
16
Temperature (C)
°
44 N
0.922581
25
15
14
0.714516
13
23
0.506452
12
42 ° N
0.4
11
0.2
°
41 N
Longitude
0.298387
24
10
9
30
30.5
31
25
31.5
32
32.5
Salinity (ppt)
26
33
33.5
34
0.0903226
PP/PS
Other alternatives :
- Changes in ARP (We just finished analyzing the filter
pad particulate absorption samples)
- Heat dissipation processes not accounted for
In situ Fv / Fm (proxy for Fp)
FLH/chl vs. Fv/Fm as Function of SST
0.55
SST, °C
17.30
Fp + Ff + Fh= 1
0.50
16.34
0.45
15.39
0.40
14.43
0.35
13.47
0.30
12.51
0.25
0.20
0
0.05
0.1
0.15
0.2
0.25
0.3
MODIS FLH / chl, W m-2 µm-1 sr-1 (mg m-3)-1
0.35
11.55
(proxy for Ff)