Physics‐based RS algorithms for water quality parameters:Status and challenges
Environmental, Earth and Ocean SciencesUniversity of Massachusetts, Boston, MA
NASA Workshop for Remote Sensing of Coastal & Inland Water, June 20 – 22, 2012
Optical Oceanography Lab
EEOSUMassB
ZhongPing Lee
Water Quality Parameters (WQPs):
Chemical:
Biological:
‘Geological’:
Physical:
Can these be sensed passively?
SST, Turbidity, Clarity, IOPs
SPM (TSM, TSS), POM, PIM
Phytoplankton (chlorophyll, phycocyanin, …), virus, bacteria
pH, CDOM, heavy metals, pesticides, O2, nutrients
ocean (water) color
)0,()0,()(
d
wrs E
LR
Remote‐sensing reflectance (sr‐1):
)0()0(
d
urs E
Lr
EdLw
Ed(0-)Lu(0-)
Quantify water color (radiance spectrum)
Wavelength [nm]400 450 500 550 600 650 700 750 800
R rs [
sr-1]
0.000
0.003
0.006
0.009
0.012
0.015
CZCS
SeaWiFS
MODIS
MERIS
The spectral form of water (ocean) color
Rrs WQPs
Algorithms for chlorophyll (and other WQPs):Gordon and Morel (1983), empTassan (1993), empBukata et al (1995), semiGitelson et al (1996), ‘emp’Hoge and Lyon (1996), semiCarder et al (1999), semi Gower et al (1999), semiKahru and Mitchell (1999), empOC4 (O’Reilly et al 2000), empOC3, empDoerffer (1999), NN Dekker et al (2002), empMaritorena et al (2002), semiBrando et al (2003), semiKutser et al (2005), empDall’Olmo et al (2003), ‘emp’Devred et al (2005), semiLe et al (2009), empGilerson et al (2010), ‘emp’Binding et al (2012), semiHu et al (2012), emp……
Rrs(λ)
IOPs
Physical‐biogeochemical properties (WQPs)
AOP‐IOP link
IOP‐WQP link
Basics of physics‐based algorithms
empirical
Semi‐analytical(algebraic or optimization)(explicit or implicit)
b
b
b
brs ba
bba
bggr
10
AOP‐IOP link (Rrs model )
rs
rsrs
w
LErs r
rR
rn
ttR7.11
52.012
b
brs ba
bGR
The ultimate question: How to adequately solve the Rrs equation for WQPs?
Two basic strategies:
1. Bottom‐up strategy (BUS): Assume we know the spectral shapes of the optically
active components
2. Top‐down strategy (TDS):Only need the spectral shape information when it is
necessary
1. Bottom‐up strategy (BUS):
)()()()(
b
brs ba
bGR
)()()( bpbwb bbb
a(λ) = aw(λ) + axi(λ) bb(λ) = bbw(λ) + bbxi(λ)
Build‐up an Rrs spectrum block‐by‐block:
)()()()( dgphw aaaa
)()()()( 21 dgphw aMaMaa
)()()()( 43 bobmbwb bMbMbb
)()()()()()()()()(
)(4321
43
bobmbwdgphw
bobmbwrs bMbMbaMaMa
bMbMbGR
M1‐4 (or M1‐3) are wavelength independent variables!Then they could be derived by comparing the modeled Rrs spectrum with the measured Rrs spectrum.
The blue‐green domain: e.g., Hoge and Lyon (1996), Carder et al (1999)The red‐infrared domain: e.g., Binding et al (2012)The entire spectrum (spectral optimization): e.g., Bukata et al (1995), Lee et al (1994,1996,1999), Maritorena et al (2002), Boss and Roesler(2006), Brando et al (2012)
Spectral ranges used for solutions (e.g. examples of BUS):
Key: Assume the spectral shapes of the optically active components are well characterized!
Ed
Lw d
wrs E
LR
H
Rrs() = F[a(), bb(), (), H]
Water property bottom property
bottom
a: absorptionbb: back-scattering: bottom albedoH: bottom depth
BUS for optically shallow waters
() = M5 )(bott
Rrs() = F[M1,M2,M3,M4,M5, H]
Water properties aftercorrecting bottom
[m-1][mg m-3]
Water properties beforecorrecting bottom
(Lee et al 2007)
2. Top‐down strategy (TDS):
b
brs ba
bGR
Remote sensing measures the total effect:Water clarity (or turbidity) is also a measure of total effect.
Rrs bb&a ax
Clarity (Secchi depth, light depth, TSM/SPM, etc)
Examples of TDS:QAA (Lee et al, 2002); Smyth et al (2006); Doran et al (2007).
wavelength (nm)400 450 500 550 600 650
abso
rptio
n co
effic
ient
(m
-1)
0.0
0.1
0.2
0.3
0.4
pure water[C] = 0.03[C] = 1.0[C] = 5.0
For a reference wavelength, λ0, variation of a(λ0) is limited.
λ0
λ0
Basics of TDS:
Known a(λ0), enables calculation of bb(λ0) from Rrs(λ0); propagate bb(λ0) to bb(λ), then enables calculation of a(λ) from Rrs(λ).
No need of spectral model of ax(λ) in this process!
Applications of bb&a:
Secchi depth:
Euphotic‐zone depth:
Zeu =f(bb&a)
0
0bpbp )()( bb
rrs()
)()()( 00w0 aaa
)(),(),(F)( 0bw0020bp barb rs
)(),(),(F)( bwbp3 bbra rs
η
The data flow of QAA:
)443()443()443()443()443()443()411()411( w
dgphw
dgph
aaaaaaaa
)443()411()443()411(
dg
dg
ph
ph
aaaa
(Lee et al 2002)
BUS derives every component first, then (simultaneously) derives the total optical property.
Contrast:
BUS relies more on forward bio‐optical model
TDS relies more on Rrs measurement
TDS derives total first, then decompose to separate components.
We should not expect the same error bars for the three pixels.
Quantification of product uncertainty
)1:;( Nixfz i 2
1
ii
N
i
xxzz
Error propagation theory:
TDS makes the uncertainty estimation straightforward.
Issues/challenges?Rrs(λ)
IOPs
Physical‐biogeochemical properties (WQPs)
AOP‐IOP link: Nearly universal
IOP‐WQP link: Transfer coefficients are Regional/temporal dependent
1. Get Rrs right!
(Lee et al 2002)
After we get the right Rrs,different level of products facing different challenges/requirements
a(λ0)&bb(λ0)
Minimum requirement: spectral shape of bbp(λ)
ZSD; Zeu
SPM
Require: spectral shape of aph(λ) and ag(λ)
Require: mass-specific optical properties
)(,)(,)(,)( dgphbobm aabb
How to best parameterize
?
A. Derive them on the fly? e.g. Brando et al (2012)
B. Parameterize them based on regional/temporal characteristics?
Thank you!
