ESTIMATING OF CHROMOPHORIC DISSOLVED ORGANIC MATTER (CDOM) WITH IN-SITU AND
SATELLITE HYPERSPECTRAL REMOTE SENSING TECHNOLOGY
Yong Q. Tian
1
, Qian Yu
2
, and Weining Zhu
1
1
Department of Geography/Institute for Great Lakes Research
Central Michigan University, Mt. Pleasant, MI. 48859, USA
E-mail: yong.tian@cmich.edu
2
Department of Geosciences, University of Massachusetts
Amherst, MA. 01003, USA
Keywords: Chromophoric dissolved organic matter
(CDOM), dissolved organic carbon (DOC), EO-1
Hyperion, Hyperspectral data, Ocean color, Coastal
Environmental, QAA-CDOM
ABSTRACT
The influence of land surface characteristics and
climate conditions on the source, quantity, quality, and
timing of dissolved organic carbon (DOC) fluxes to
coastal waters is not well understood. The significant
correlation between DOC and chromophoric dissolved
organic matter (CDOM) leads to an increasing need of
CDOM monitoring for understanding the DOC land-
water dynamics. This study is to report the results of
using hyperspectral in-situ data and satellite images to
estimate riverine CDOM from rivers to oceans and Great
Lakes. A major research goal is to demonstrate that
advancement of hyperspectral and high spatial resolution
remote sensing technology is vital to the study of
interactive processes between terrestrial ecosystems and
aquatic environments. Our research results confirm that
hyperspectral remote sensing is effective in extracting
riverine CDOM loading.
INTRODUCTION
Chromophoric dissolved organic matter (CDOM)
absorbs ultraviolet and visible light [1]. Due to the tight
connection with dissolved organic carbon (DOC),
CDOM is often used as a DOC tracer [2-6]. Both DOC
and CDOM in nature primarily derive from the litter and
decaying of vegetation detritus; they are transported to
estuarine and coastal waters through terrestrial runoff and
stream flow [7, 8]. DOC has significant implications on
our environment and climate change due to its important
role in the carbon cycle [6, 9-11]. Studying the
concentration and distribution of CDOM in aquatic
ecosystems, particularly the estuarine and coastal regions,
will greatly improve our understanding of the land-water
dynamics of DOC, terrestrial-oceanic carbon cycle, and
the impact of anthropogenic activities on water quality.
METHODOLOGY
A method for the inversion of hyperspectral remote
sensing is developed to determine the absorption
coefficient for CDOM in river-coastal margins. The
algorithm was based on above-surface measurements
with a ship-mounted spectroradiometer. Simultaneously
to above-water srface, water absorption and attenuation
coefficients, CDOM and chlorophyll fluorescence,
turbidities and other related water properties were also
measured at very high resolution (meters) using in situ,
underwater and flow-through (shipboard, pumped)
optical sensor. Rather than using a
dg
440 (the total
absorption coefficient for CDOM and non-algal particles)
as an indication of the “concentration” of CDOM in most
previous ocean color study, we separate a
g
440 (the
absorption coefficient for CDOM) from a
dg
440 based on
two absorption-backscattering relationships. The first is
between a
d
440 (the absorption coefficient for non-algal
particles) and b
bp
555 (total particulate backscattering
coefficient at wavelength 555 nm) and the second is
between a
p
440 (the absorption coefficient for total
particles) and b
bp
555. These two relationships are
respectively referred to a
d
-based and a
g
-based methods.
Consequently, based on Lee’s QAA algorithm, we
developed the so-called Extended Quasi-Analytical
Algorithm (QAA-E) to decompose a
dg
440, using both a
d
-
based and a
g
-based methods. The absorption-
backscattering relationships and the QAA-E are tested
using International Ocean Color Coordinating Group
(IOCCG) synthetic and in situ data as well as field data
from our study site. And then the algorithm was
calibrated to estimate CDOM distributions in river plume
areas from EO-1 Hyperion satellite images. The
algorithm was tested with measurements in river
dominated coastal margin: Atchafalaya River and the
Mississippi River. The test was also extended to
Saginaw Bay of the Lake Huron recently.
RESULT AND DISCUSSION
The resulted accuracy indicates the a
d
-based method
is relatively better than the a
g
-based method. The
accuracy of CDOM estimation is significantly improved
2040 978-1-4673-1159-5/12/$31.00 ©2012 IEEE IGARSS 2012