Active-Passive Spaceborne Data Fusion for
Mapping Nearshore Bathymetry
Nicholas A. Forfinski-Sarkozi and Christopher E. Parrish
Abstract
"In anticipation of the National Aeronautics and Space
Administration (
NASA
) ICESat-2 mission, which will employ
the Advanced Topographic Laser Altimeter System (
ATLAS
),
a 532 nm photon-counting Light Detection and Ranging
(lidar), we demonstrate a spaceborne data-fusion approach
that has the potential to significantly shrink the global
nearshore data gap often referred to as the “white ribbon”.
Bathymetry algorithms relying on multispectral imagery are
conventionally limited by the availability of
in situ
refer-
ence depths, particularly in remote or difficult-to-map areas.
Therefore, a completely spaceborne approach could greatly
extend the usefulness of such algorithms. The approach is
tested with data from
NASA’s
airborne I
CESat-2 ATLAS
simula-
tor, Multiple Altimeter Beam Experimental Lidar (
MABEL
),
and passive optical imagery from Landsat-8 using an exist-
ing spectral-ratio algorithm. The output bathymetric data
set agrees with high-resolution Fugro
LADS MK II
bathymetric
data to within an
RMS
difference of 1.1 m. The spatiotem-
poral variability of areas where this spaceborne data-fusion
approach will potentially be useful is evaluated, based on
worldwide coastal water clarity as interpreted from Visible
Infrared Imaging Radiometer Suite (
VIIRS
) Kd(490) data.
Introduction
Despite a large variety of remote sensing techniques that mea-
sure or estimate water depth (Gao 2009; Jawak
et al.
2015),
there remains a global lack of nearshore bathymetry data (
IHO
2018; NRC 2004). The hydrographic community has called
this data gap the “white ribbon”, referring to the correspond-
ing along-shore empty space on many nautical charts (Leon
et
al.
2013; Mason
et al.
2006). Inadequate or nonexistent chart
data is particularly an issue in the Arctic, which has seen
an increase in ship traffic (
NOAA
2016), and in the southwest
Pacific (
IHO
2018). The white ribbon affects numerous re-
search domains and coastal-management applications beyond
navigation, such as coral-reef studies (Miller
et al.
2011), river
geomorphology (Legleiter and Overstreet 2012), reservoir
management (Moses
et al.
2013), inundation modeling, and
broader efforts to generate seamless topographic/bathymetric
digital elevation models (
DEMs
) (Eakins and Grothe 2014).
Even in countries with dedicated coastal-mapping agencies,
populating and maintaining coastal bathymetric databases
is challenging, given environmental constraints, limited
resources, and difficulties and hazards of working in remote
and dangerous areas. Global and regional bathymetric datasets
do exist, but high-resolution datasets are limited, and much of
the global-scale data is derived from altimeter/gravity tech-
niques that have significant limitations in coastal areas (Smith
and Sandwell 2004; Weatherall
et al.
2015). These methods
generally produce low-resolution (nominally five to 100 km
and, therefore,
ons. While these
for physical
ad spatial extents,
patial resolution
in the nearshore domain.
Relying solely on conventional acoustic (i.e., ship- or
boat-based echosounder) methods to map the nearshore zone
would be expensive and time-consuming, as well as danger-
ous in the shallowest areas and near reefs, rocks, and other
dangers to navigation. According to Weatherall
et al.
(2015),
approximately 900 ship-years of sonar-based acquisition time
(ignoring logistics, data processing, and repeat-survey require-
ments) are needed to map Earth’s oceans, with the zero to
200 m depth range requiring two thirds of that time. Another
technology for Bathymetric mapping that merits discussion
is Airborne Lidar Bathymetric (
ALB
). The use of airborne
(i.e., airplane- or helicopter-mounted) laser mapping systems
extends back at least as far as 1968 and a study done by Hick-
man and Hogg (1969), but widespread use of
ALB
only began
in the 1990s through early 2000s, as global navigation satel-
lite system (
GNSS
), aided inertial navigation systems (
INS
) and
related technologies matured to the point of facilitating broad,
operational use. Current, commercial
ALB
systems include the
Optech Coastal Zone Mapping and Imaging lidar (
CZMIL
) and
Titan, Leica Chiroptera 4X and HawkEye 4X, and the Riegl
VQ-880-G II. Despite the widespread use of
ALB
, it remains
difficult to deploy—especially in remote regions—and requires
significant resources. Hence, it is a great option for localized
acquisition, but not feasible as a sole means of filling the
nearshore data void. Multispectral remote sensing methods of
deriving bathymetry have been appealing ever since the mid-
1960s, when Gilg and McConnell, Jr. (1966) considered using
spaceborne photography for Bathymetric reconnaissance.
Beginning with the first documented spectral-based
approach (Polcyn and Rollin 1969), methods of mapping
bathymetry from multi- and hyperspectral imagery have been
well established in the published literature. However, a major
drawback of most spectrally-based bathymetry retrieval ap-
proaches is that they require
in situ
data, in the form of refer-
ence or “seed” depths. This requirement negates the possibili-
ty of mapping bathymetry solely from satellite-based data and
from using the methods in areas where there are no existing
reference soundings. To overcome this limitation, we propose
an active-passive spaceborne data fusion approach, leverag-
ing satellite-based lidar and multispectral imagery. (Note that,
for purposes of this study, we define data fusion broadly to
include any technique leveraging multiple, complementary
inputs where each provides some information the others do
not (Castanedo 2013; Elmenreich 2002).
Nicholas A. Forfinski-Sarkozi and Christopher E. Parrish
are with the School of Civil and Construction Engineering,
Oregon State University, 101 Kearney Hall, 1491 SW Campus
Way, Corvallis, OR 97331
)
(
).
Photogrammetric Engineering & Remote Sensing
Vol. 85, No. 4, April 2019, pp. 281–295.
0099-1112/18/281–295
© 2019 American Society for Photogrammetry
and Remote Sensing
doi: 10.14358/PERS.85.4.281
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
April 2019
281
