United States
Environmental Protection
Agency
Environmental
Research Laboratory
Narragansett, Rl 02882
Research and Development
EPA/600/S3-87/009 Sept. 1987
&EPA Project Summary
Report on the Usefulness of
AVHRR and CZCS Sensors for
Delineating Potential Disposal
Operations at the 106-Mile Site
Peter Cornillon
Data from the AVHRR sensor for 18
TIROS-N series satellite passes were
examined for signs of ocean disposal at
106-Mile Disposal Site off the North-
east Coast of the United States. The
passes selected occurred within five
days following the actual disposal. All
spectral channels were analyzed. In
addition, the difference and the ratio of
channels 1 and 2 were used.
No dumping is evident in any of the
images. If dumping is actually visible,
the signal must be very weak at the 1
km resolution of the AVHRR sensor.
Because of this, the images would have
to be absolutely clear. Small scattered
clouds or thick haze add too much
variability to permit detection of a weak
signal which renders AVHRR data of
marginal value for such work.
Because the exact location of the
dumps was not known, it is possible
that the signal is detectable. For this to
be true, the dumps would have had to
occur quite far from the location pro-
vided in the Coast Guard records. These
records give only the site (e.g., dumpsite
106), not actual coordinates.
This Project Summary was developed
by EPA's Environmental Research
Laboratory, Narragansett, Rl, to an-
nounce key findings of the research
project that Is fully documented In a
separate report of the same title (see
Project Report ordering Information at
back).
Introduction
As part of its ongoing monitoring
activities for the 106-Mile Ocean Disposal
Site, the Environmental Protection
Agency (EPA) contracted to the University
of Rhode Island (URI) to investigate the
possiblity of detecting dumped waste in
AVHRR satellite data. The primary focus
was to be on the use of the visible
channels (Channel 1: 0.55 - 0.9/um,
Channel 2: 0.725 - 1.1/urn for TIROS-N;
Channel 1: 0.58 - 0.6^m, Channel 2:
0.725 - 1.1 Mm for NOAA-6 to NOAA-9)
although consideration was also to be
given to the IR channels (Channel 4; 10.5
- 11.5Mm for TIROS-N, NOAA-6 and
NOAA-8; and Channel 4: 10.3 - 11.S^m,
Channel 5: 11.5 - 12.5/um for NOAA-7
and NOAA-9). Because the signal from
the dumped material is thought to be
very weak, and because of the substantial
noise in Channel 3, that channel was not
considered.
The reason that the EPA wished to
consider AVHRR data for their monitoring
program is the frequent coverage of the
106-Mile Site area by satellites carrying
this sensor. In general, each satellite
passes the area twice a day and there are
generally two satellites collecting data;
i.e., the dumpsite is covered between two
and four times a day. NOAA-6 and
NOAA-8 are in dawn-dusk orbits. Such
orbits overfly any point at about 7:00 a.m.
and 7:00 p.m., local sun time. TIROS-N,
NOAA-7 and NOAA-9 are in orbits which
overfly any point on the globe at 2:00
a.m. and 2:00 p.m., local sun time. This
means that only one of the NOAA-7 and
one of the NOAA-9 orbits overfly the
dumpsite in daylight hours. NOAA-6 and
NOAA-8 generally have both passes dur-
ing daylight but at times corresponding to
very low sun angles. This means that
-------
very little sunlight penetrates the water
column. Therefore, unless the dumped
material is near the surface (top few
centimeters) there will be little sunlight
reflected from it; at best, it will be difficult
to detect.
Discussion
No disposal plumes were observed in
any of the images analyzed. Many of the
images which appeared to be clear when
the data were initially scanned, often had
light scattered clouds in them when pro-
cessed at full resolution. Such clouds
make it difficult to impossible to detect a
weak signal in the visible channels. It is
clear from this work that the disposal
plumes, if apparent in the AVHRR visible
data, will be very weak. This means that
to observe these dumps the area will
have to be absolutely clear. Furthermore,
there will have to be significant solar
penetration in the water column. The
very clear dawn-dusk images showed
nothing.
The question arises why disposal
plumes are detectable in LANDS AT data
and not in AVHRR data. The primary
reason relates to the spatial resolution.
LANDSAT MSS data has a resolution of
80 m compared to the 1.1 km resolution
of the AVHRR. In the LANDSAT data,
plumes at the 12-Mile Site in New York
Bight although 10's of kilometers long,
are generally less than 1km wide. To see
such a plume in AVHRR data would re-
quire a substantial difference in reflect-
ance with that of the surrounding waters.
To detect ocean disposal plumes from
AVHRR data one would have to know
exactly where to look. In this study, only
the date of the dump and the dumpsite
were provided, not the precise location of
the dump. Unless the dumps actually
occurred far from the 106-Mile Site loca-
tion used in this analysis, the signal even
if detectable, is far too weak to be of
value in a monitoring program.
The LANDSAT Thematic Mapper data
should be analyzed to determine the size
and spectral signature of the disposal
plumes. With such data, to determine
whether or not AVHRR visible data has
any promise at all would be fairly straight-
forward.
The assumption is that CZCS data would
provide a slightly better chance of dis-
criminating dispoal plumes because
NIMBUS-7 passes overhead at approxi-
mately local noon when the penetration
of solar radiation into the water column
is at maximum, and because more spectral
channels are available in the visible. Also,
these channels have higher radiometric
resolution than the AVHRR channels.
However, because it was of very low
quality due to degradation of the sensors,
CZCS data were not considered.
Peter Cornillon is with University of Rhode Island, Kingston, Rl 02881.
John F. Paul is the EPA Project Officer (see below).
The complete report, entitled "Report on the Usefulness of A VHRR and CZCS
Sensors for Delineating Potential Disposal Operations at the 106-Mile Site,"
(Order No. PB 87-168 829/AS; Cost: $9.95, subject to change) will be available
only from:
National Technical Information Service
5285 Port Royal Road
Springfield, v'A 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
Narragansett, Rl 02882
Jnited States
'.nvironmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
)fficial Business
'enalty for Private Use $300
EPA/600/S3-87/009
0C0Q329
------- |