United States
              Environmental Protection
              Office of
              Radiation. Programs
              Washington, DC 20460
EPA 520/1-83-017
June 1983
1978 Atlantic 3800-Meter
Radioactive Waste
Disposal Site Survey:
Micromorphologic and
Geophysical Analyses

                         EPA REVIEW NOTICE
     This report has been reviewed by the Office of Radiation Programs,
U.S. Environmental Protection Agency (EPA) and approved for publication.
Approval does not signify that the contents necessarily reflect the
                                                 EPA 520/1-83-017
           David H. Hanselman and William B. F. Ryan

             Lamont-Doherty Geological Observatory
                    of  Columbia  University
                   Palisades,  New York   10964
                      Prepared June 1979
                      Revised  June  1983
   This report was prepared as an account of work sponsored
  by  the Environmental  Protection  Agency  of  the  United  States
           Government under Contract No.  68-01-483(5
                       Project Officer
                        Robert S.  Dyer
                Analysis  and  Support  Division
                 Office of  Radiation  Programs
             U.S.  Environmental  Protection  Agency
                   Washington, D.C.  20460


       In response to the mandate of Public Law 92-532, The Marine
 Protection  Research  and Sanctuaries Act of 1972, as amended, the
 Environmental Protection Agency (EPA) has developed a program to
 promulgate regulations and criteria to control the ocean fisposal of
 radioactive_ wastes.  As part of that program, the EPA Office of Radiation
 Programs initiated feasibility studies in 1974 to learn whether present
              Cbe ^                         °f -^active wastes
      In 1978, the advanced technologies represented by the manned
 !     The  present  report provides a detailed description of the geological
 and  topographical characteristics of the site.  Two dives were made in
 Sosits8 S«1U  ,   ^'^.radioactive waste dumpsite.  The sediment
 deposits, bottom  topography, biota, currents, and the differences in
 these parameters between the two areas within the dumpsite are discussed
 The  appearance of radioactive waste drums found in the dumpsite area is
 analyzed in terms of the localized geological and physical Vocess^

             s^biSy oThrC°nClUdeS ^ ' — ^ ussion of
                                   s       tisu
Division, Office of Radiation Programs (ANR-461),  Environmental
Protection Agency, Washington,  D.C.   20460.
                                         Len  L.  S job lorn,
                                      Office  of  Radiation Programs


Five dives with the DSRV ALVIN in the extension of the Hudson Canyon Channel
on the continental rise off Long Island were undertaken to  investigate the
physical, biological and chemical environment of an area formerly utilized  as
a radioactive waste disposal area.  Observations within a depth range of
3985-3830 m revealed angular blocks and piles of displaced  channel wall rock,
boulder and cobble olistoliths of Eocene-age chalks derived from higher
elevations on the slope, and bedforms such  as ripples and scour marks which
imply the existence of periodic strong currents.  Local benthic fauna are
sparse; bioturbation and burrows are uncommon.  Three waste drums were located
and one was subsequently recovered for analyses.  Photographic and visual
evidence suggest that downslope transport of objects such as talus blocks,
olistoliths and waste drums has occurred  in this area.

                                TABLE OF CONTENTS
 INTRODUCTION [[[            I

 DIVE DESCRIPTIONS ...............................................               6

      Dive  812 [[[          6

     ! Dive  813 [[[         12

 RADIOACTIVE  WASTE DRUM  SITE  AT 3970  METERS  DEPTH .........................    18

      Description  of  Site .............................................        18

      Coring  Program  and Field  Description of  Cores .......... .............    24

 INTERPRETATION AND DISCUSSION .......................................         26

 CONCLUSIONS  AND RECOMMENDATIONS ..........................................    31

                                LIST OF TABLES

 I.    Summary  of samples taken  during ALVIN  Dives 679, 812 and 813 ........     5

                                LIST OF FIGURES
 1.   I Regional bathymetric map  of the continental shelf, slope and
       rise seaward of  Long  Island,  New York ................. .............     2

 2.    Bathymetry of the 3800-meter Radioactive Waste Disposal Site ........     3

 3.    Detailed bathymetry in the vicinity of the 1978 ALVIN dives .........     4

 4.    Light tan marl or claystone angular talus blocks ....................     7

 5.   Outcrop of semiconsolidated tan marl  or claystone ...................     8

 6.   : Heavily bored, white Eocene (?) chalk boulder with a
     i  prominent scour moat exposing a gravel  bottom ....... . .............   11

7.   Gravel and cobbl e bottom ............................................   14

                          LIST OF FIGURES (Continued)
10.  Rounded chalk boulder 2-3 meters long, setting on a hard
       gravel bottom that is exposed in a prominent scour moat	    I/

11.  Photograph of the first drum located during Dive 813	    19

12.  Photograph of first drum located on Dive 813	    20

13.  Position and field descriptions of tube cores taken
       down-current from second drum located (#953)	    22

14.  Photograph of second drum (953) located	    23

15.  Conical rock pile near drum 953	    25

16.  Hypothetical model for large scale slumps as a
       canyon filling mechanism	

17.  Stratigraphy and-seismic reflection profile for  DSDP Site 106	    30

18.  Seismic lines showing probable  positions of
       Eocene and Miocene horizons at  dump site	   -^


    , During the period July 22 to 28, 1978, a series of five dives with the
DSRV ALVIN was made at the Atlantic 3800 meter Radioactive Waste Disposal Site.
This site is located about 380 km off the coast of Long Island on the lower
continental rise, 260 km seaward of the edge of the continental shelf, near the
main! channel of the Hudson submarine canyon system (Fig. 1).  The waste
consists of approximately 15,000 55-gallon drums that are filled with low-level
radioactive waste and concrete.  These are distributed over an area of a few
tens: of square kilometers in water depths of 3700 to 4100 m.  The purpose of
the diving program was to study the physical, chemical and geologic setting of
the disposal site and the local biological conditions of the substrate in order
to better assess the feasibility of future disposal of radioactive waste in the

     The study was conducted in an area bounded by coordinates 37°40'N to
38°10'N and 70°24'W to 70°40'W, covering an area on the lower continental rise
near the confluence of the main channels of the Hudson and Block submarine
canyon systems (Fig. 2).  In this region, the channel of the Hudson Canyon is
characterized by an approximately 1 km wide canyon floor that contains a
narrow, deeper, meandering thalweg or axis.  The walls that bound the canyon
floor exhibit slopes from gentle gradients of a few degrees to vertical and
stand as high as 200 m (Fig. 3).

     The primary mission of the project was to locate radioactive waste drums
and to identify and recover a suitable drum for laboratory analysis.  Secondary
goal's were to take sediment and water samples near the recovered drum.  The
geological work consisted of a description of bottom topography and sediments,
the nature of bedrock exposures,  and the sedimentary and erosional processes,
including bioturbation, that affect the stability of the substrate.  Geological
samples were taken for laboratory analysis.  Selected samples are listed and
described in Table I and in Figure 13.  The presentation and interpretation of
these data are the basis of this report.

                                   Figure 1
Regional bathymetric map of the continental  shelf,  slope  and  rise seaward of
Long Island, New York.  A star marks the 1978  dive  positions  within the 3800
meter Radioactive Waste Disposal Site,  enlarged  as  text Figure 2.   Stratigraphic
control for the area is provided by the COST B-2 well, drilled in 1976 to a
depth of approximately 5000 m.  Contours in  meters.
                                       — O —

                                                                                                                  CONTOURS IN METERS   °
                                                                     Fig.  2

      Bathymetry  of the 3800  meter Radioactive  Waste Disposal Site,  contoured  in uncorrected  meters  (750 m =  1  sec. two-way
      reflection time). Solid  line A indicates  ship's track  of the  R/V  Vema  during seismic reflection  (airgun) survey.  Solid  line B
      indicates  locations of  seismic  reflection  profile  illustrated in  Figure  18. Thin dashed  lines  are  echosounding  tracks  of R/V  Lulu.

                                   Figure 3
Detailed bathymetry in the vicinity of the 1978 ALVIN dives.  Contour  interval
is 25 m.  Dives 812 and 813 explored the eastern wall of Hudson Channel.
Dive 814 surveyed and recovered a waste drum within the small circled  region.
Dashed lines are survey tracks of R/V Lulu.

                                              TABLE  I
Sample #

Alvin 812
Ski    ,

Sandy mud
                                Setting     Species  *

                                Ski         Cyclococcolithus leptoporus
                                sample      Coccolithus pelaqious
                                            Helicopontosphaera kamptneri
                                            Helicopontosphaera inversa
                                            Rhabdosphaera clavigera
                                            Coccolithus doronocoxdes
                                            Gephyrocapsa sp.
                                            Thoraco'sphaera sp.
                                            Discoaster brouweri
                                            Discoaster asymmetrioas

Alvin 813
Ski-1   i
Alvin 813
Ski 2
Alvin 8l3
Chalk   >
Greyish-tan mud
w/iron staining

white,  consoli-
dated chalk


Alvin 815
Ski     :
                               No age determination

                               Same as in 812-ski
                               Chiasmolithus solitus
                               Zyqrhablithus bijugatus
                               Cyclicargolithus floridanus
                               Sphenolxthus moriformis
                               Sphenolithus radians
                               Discoaster barbadiensis
                               Braarudosphaera rosa

                               Similar as in 812-ski and
                               813-ski 2
                               Rare Emiliania huxleyi
Eocene, probably
mid Eocene
Alvin 679
             White chalk
             collected in
                   Boulder     Cyclicargolithus florida mis
                               Chiasmolibus expansus
                               Chiasmolithus solitus
                               Discoaster barbadiensis
                               Discoaster deflandrei
                               Coccolithus pelagicus
                               Zygrhab1ithus bijugatus
                               Reticulofenestra umbilica
                               Sphenolithus radians
                               Helicopontosphaera seminuJ.um
                               Chiasmolithus gigas (?)
                               Braarudosphaera bigelowi
                               Braarudosphaera discula
                               Zygolithus dubius
                               Discolithus cf. segmenta
                               Tribrachiatua orthostylus
                                                                               Upper Middle
312-Core #1  Near Core #10

812-Core#10  Core taken into
             scarp face near
        !     bottom of face
                   Core not available

                   Core not available
813-Core #2  Corrosion product  Core not available
             near first drum

813-Core #3  Taken in sediment  Core not available
             build up downcur-
        •     rent from first
             drum, 8 cm long

*Species identification by Gretchen Blechschmidt,  L-DGO

                               DIVE DESCRIPTIONS

Dive 812

     Dive 812 on June 23, 1978, touched down at 1241 hours at a depth of 3924 m
(Fig. 3).  The submersible ALVIN then slid down a 10-30 degree mud slope to a
depth of 3958 m.  According to preliminary bathymetric mapping (Fig.  3) ALVIN's
deepest position during this dive (3958 m) coincided with the base of the east
wall of the Hudson Canyon channel.  From this point an upslope traverse was
initiated first in a northeasterly direction and later in a southeasterly
direction.  ALVIN left the bottom at a time of 1746 and a depth of 3827 m.  No
radioactive waste drums were located during this dive but many interesting and
important geological observations were made.

     The bottom topography around 3958 m is quite variable with slopes ranging
from 0 to 20 degrees.  Small angular mud-blocks  lie on the sea bed and appear
to  have  been derived from upslope sites (Fig. 4).  Feeding trails and current
lineations are  common.  The bottom here could best be described as blocky  and

     Between 3950 and  3925 m,  along the wall of  the channel,  small fluidized
sediment flows  were  initiated  by  the submersible1s contact with the  sea  bed.
The associated  mud clouds moved downslope  at a  rate estimated to  be  about
125 to  300 cm/second.  The water  was turbid and  the visibility was poor.
Southwesterly flowing  bottom  currents  were persistent  in  the  channel wall
region  as well  as down in the  channel  floor.  The velocity  of the currents
was estimated at about 25 cm/second.   Such cross-channel  flow is  not unusual
in submarine  canyons (see Shepard,  et  al., 1979) and  the  presence of such
currents and  associated turbidity in  the  Hudson channel  is  attributed  to
flow associated with the Western  Boundary Undercurrent (Heezen and Hoi lister,
1971;  Eittreim  and  Ewing,  1972).   In  regions  of smooth topography along the
 channel wall,  a light tan  sediment layer about 1 to  5 cm thick covers  a firm,
 light tan,  semi consolidated marl  or clay.   Five to ten meter high marl  or
 clay scarps are present (Fig.  5).  These are separated by flat areas or

                                   Figure 4
3830 m.  Light tan marl or claystone angular talus blocks.  These blocks appear
to be locally derived by the slumping of adjacent wall rock.   Planar surfaces
are;interpreted as joints produced in semiconsolidated strata.   In this and
other bottom photography the scale is approximately 2-3 m for the lower border
of each frame.

                                          1 fill!!:ii:!;ll!!T " •,,' Ji' • "
                                           kllii'jiiiiilliilljSir	ijiiiililij^   ,„ '«' i"'1iii'i»iiii,i:i' " ''" ' :i , ",' '• , •,,":,:, "" :',;/' '^''!'MI«IVV;;^
                                                '•"•'HI'"[!!'• I".!!!!:1'11':!	i	'"'»" n 'is"!'"';'"
                                                  ',:, '^r\r:'"*	.'.,	
                     .  ••••.	V  ...  .,	y*g
                    ...  ,  	...I' • ••': 	 n	I/Ml »
                               ' 	ir-V'j^'
                                                         ?•	L*:A.:".:J::":'»:"-' :.:•"*-.;.."-:'«::' •"•*«*•:•.:•;:
                                                     in™,,,,	 	!'.	l!*1'	,„.,	int	^SfcufSili*';''	,,,;.,:,n,iii	i^i,,,,;	; „ „«	i.,.ma,,!,• '„,**=--ji
                                                     ",illuy; ',;,/,"	Hi/ 'ii'.iii'jiliaiiiHiiiii1' i"! ,':"*iii!i!i'"l'i iiiii',,1'1''1	jiikJinil:!!! ""iiG*",;.''•': ''"i'Sfililli!.11'":!, tfSS .IT1!!1'1'1!!!™!,
                                                                                 ,•	iiMji.
                                              Figure  5
3870  m.'   Outcrop of semi consolidated tan marl  or  claystone.    This  is  a  small
slump scarp,  bedding  is indicated by horizontal  ledge  on  left.   This  scarp  is
about 3  m high  and  is  typical  of the entire  channel  wall  area.   The  flat bench
begins at base  of outcrop.

 benches.   Sediment ripples with a 10 cm wavelength are present on the current-
 swept flats.   A possible boulder slide mark was observed and appeared to be
 very recent.   In this region on the lower channel  wall, a sparse brittle star
 population was observed along with plant debris and a few horizontal  burrows.
    ;  From 3925 to 3900 m,  further up the channel wall,  small  outcrops of soft,
 semiconsolidated marl  or claystone were frequently observed.   The outcrops
 have vertical  faces  1 to 3 m in height, and are separated by flat current-swept
 benches.   Symmetrical  ripples with 15 to 20 cm wavelengths are present,
 suggesting the presence of oscillatory flow at this site.   The flat bench areas
 often exhibit  elongate tensional  cracks,  suggesting that the  steep faces of the
 outcrops  may have had  their origin by slumping and thus the steep faces  may
 represent slump scars.   Subparallel  crenulations just below the thin  sediment
 drape suggest  creep  phenomena that may precede or  be  associated with  slumping.
 In this depth  interval,  the brittle stars  and  the  number of burrowers was seen
 to increase but the  significance  of this observation  is not. known.

    !  Most of the  outcrop faces  between  3900  and 3875  m  are  covered with  a thin
 sediment  drape,  but  because  of  variable resistance  to submarine  weathering,
 layering  or bedding  is  apparent beneath the  cover  (Fig.  5).   Beds  of marl or
 claystone  are  flaggy to massive,  ranging from  10 to 30  cm thick.   Occasional
 cut land fill structures are  present.   Ripples  are again  present  at this  depth
 and occur  on the  flat  benches.  A  few  ripples  displaying wavelengths of  10  to
 20 cm are  oriented parallel  to  the  slope and are also of symmetrical  form.  A
 small slump was observed in  action, probably initiated  by the ALVIN.

    | Similar scarp and bench topography occurs from 3875 to 3850 m.  Lineations
trend downslope on the faces of some of the marl or claystone outcrops.   These
appear to  have been  caused by small talus blocks sliding downslope.  Numerous
smal;l to  large angular talus blocks that appear to have fallen from higher
outcrops litter the  flat areas between  scarps.   Many of the talus blocks are
plate-like slabs of marl that appear to have slid downslope, possibly on a mud
or water cushion.  These flat slabs are about 10 to 50 cm in diameter and
several cm thick.  Larger angular slumped blocks are also present.  These vary

greatly in size and are often bounded by smooth faces that may be joint planes
(Fig. 4).   Tool marks, believed to have been formed by small pebbles sliding
downslope, are present in areas where the bottom is smooth but sloping.  In
this depth interval (3875-3850 m), so many of the scarp and bench features
were observed that it is possible that this whole area is part of a very large
slump complex.  The small scarps and benches are believed to have formed as
small dislocations and pullaparts and are the surface expression of much larger
scale sliding and slumping along more extensive fault planes at depth  (Fig. 5).
The  largest slump scarp  observed was about 15 m in height and dipped 45 to
60 degrees.  Angular  talus blocks were present at the base  of'the scarp.  As
the  submersible slid  down the  face of the scarp, observers  noted that  the
plates  and slabs of rock lying on the scarp  itself seemed to  be  in  a state  of
creep  or  instability.   Scarps  1 to 5 m  in height are the most common.   Steep  to
vertical  angles are the rule for  the scarp  faces and a  smoothed  20-45  degree
slope  at  the  base  of  the scarps is common.   A  rough  estimate  of  the spacing
between scarps or  the width  of intervening  benches  is  about 5 to 20 m.  The
 scarps often  seem  to  die out laterally.   Aim high  scarp,  for  instance,  might
 decrease in  height until it  blends  into a gentle  slope over a distance of
 approximately 20  m.   The megafauna  between  3875 and 3850 m are  generally  sparse
 and there is  a lack of burrowing activity.   This  region appears  to be  a
 biologically impoverished  environment with only occasional  ripples and a  smooth
 to dimpled bottom.

      The upper part of the channel  wall from 3850 to 3828 m still exhibits
 slump  features but the benches between slump scarps are wider and there is a
 general reduction in the average slope.  A small basin-like feature occurs at
 about  3830 m and appears to be a bowl-shaped depression open on the downslope
 side  (Fig. 6).  ALVIN  then  slid down the wall which was about 10 m high and was
 covered with  a light sediment drape.  Currents appear to be  stronger  at this
 depth  as evidenced by  the formation of elongate sediment tails  downcurrent from
 vertical burrows.  These small sediment  accumulations are  dune-shaped, and are,
 on  the average, 10 to 30 cm in length, 5 to 8  cm wide,  and 2 to 4  cm  in  height.
 Long  thin current lineations  (grooves) are also present.   Patches  of  pebbles
 and plant material  of unknown origin  are exposed  in  areas  where the current  has

                                   Figure 6
3964 m.  Heavily bored, white Eocene (?) chalk boulder with a prominent scour
moat exposing a gravel bottom.

swept away the top layer of sediment.   Empty snail and clam shells are present
but not abundant.  The visibility is poor at these depths, 3 to 5 m being
typical.  Animal life is generally sparse along the canyon wall between 3850
and 3828 m.  Brittle stars occur sporadically and increase slightly upslope.
Occasional sea pens, sea whips, sponges, bryozoans, and ear-shaped sponges or
Coelenterates form the rest of the sessile fauna.  Cigar-shaped hollow tubes
with an opening on one end were quite common on the bottom at depths of about
3850 m.  Possibly these represent some kind of discarded egg case or molt.
Occasional rattail fish and red shrimp are present but are not common.  Small
tan crabs  occur on outcrops toward the upper part of the wall.  Long sinuous
horizontal burrows with impressive downcurrent dune-shaped sediment mounds  are
common  where the  currents are  stronger.  Small conical mounds  are more common
toward  the upper  part of the canyon wall.

     At 3828 m,  the  shallowest point  during this  dive, brown  talus blocks 30  by
15 by  3 cm were observed that  were  similar  in  appearance  to brown  siltstone
lithologies observed in New  England  submarine  canyons  during  previous  dives
(Ryan  et  a!.,  1978).  Before terminating this  dive,  a  short core  (Core #10) was
taken  from the base of  a  small fault scarp.  Another short core was  taken for
radioisotope  analysis  (Core  #1).   See Table I  for core descriptions.   All of
the sediment  encountered  and cored during  this dive  was  soft  to  semi consolidated
as compared  to the abundance of hard rock  encountered on the  next day, June 24,

 Dive 813

      ALVIN dive 813 on,June 24, 1978, in the same general area as Dive 812
 (Fig.  3), reached bottom at 1305 hours at a depth of 3985 m.   ALVIN touched
 down on a relatively flat terrain on the floor of the Hudson Canyon channel
 (Fig.   3).  A 700 m long traverse was made in an easterly direction in search of
 waste  drums.   The course followed the break in slope between the generally flat
 floor  of  the canyon and a relatively steep (20 degrees to vertical) canyon wall.
 Radioactive waste drums were  located at a depth  of  3970 m and the dive was
 terminated from  this depth at a time of 1733  hours.

   j  Bottom topography in the region of this dive is characterized by irregular,
boulder-strewn and hummocky areas separated by broad flat areas.   This region
is being swept by southwest-flowing currents similar to currents  noted during
Dive 812.   The water was turbid but slightly clearer than the water higher on
thei channel wall.  Slumped blocks of all sizes and shapes litter  the bottom but
seem to be concentrated in groups and piles (Fig. 7).   These are  separated by
featureless broad flat areas that are 30 to 50 meters wide.   The  blocks and
boulders can be grouped into three distinct classes.  (1) Rounded cobbles and
boulders of hard crystalline rock are common.   These have been carried into
deeper water by ice-rafting and have probably undergone some later concentration
in the submarine canyon system.  Quartzite, granitic and metamorphic rocks
probably make up the bulk of these glacially derived cobbles (Fig.  8).  (2) The
most common pebble and cobble lithology is a soft tan marl or claystone.   Tan
mar,! blocks are very abundant and often occur in piles (Fig. 8).   The blocks are
very angular and seem to have been shaped by intersecting joint planes.  These
pilbs of blocks appear to be fresh and are generally not coated with sediment.
Based on these observations and those of Dive 812, these blocks are interpreted
to have been locally derived by slumping and sliding of semiconsolidated
sediment from further up the canyon wall.  Therefore, local  gravitational
instability along the canyon walls could easily explain the origin of this type
of jtalus.   (3) Rounded pebbles, cobbles and boulders of white chalk are abundant
together with lesser occurrences of what appear to be blocks of brown siltstone.
One boulder of apparently interbedded white chalk and probable brown siltstone
was; observed (Fig. 9).  The average size of the white chalk boulders is smaller
than the softer marl talus blocks, but 2-3 m diameter blocks were observed
(Fig. 10).  The chalk blocks exhibit a variety of shapes and sizes, but equant
to Cylindrical and tabular shapes are the most common.  Relatively smooth
surfaces are most common but some blocks, especially the larger boulders, are
heayily bored (Fig. 6).  Burrows are subparallel and probably occur parallel to
original bedding in these rocks.  The bored fabric appears to be  a relict
feature, and may have taken place in a different setting.  The white chalk
blocks look foreign to this environment, particularly since none  of these were
observed further up the channel wall during Dive 812, although blocks of the
softer tan marl were common.  The white chalk does not occur in the wall

                                   Figure 7
3960 m.  Gravel and cobble bottom.  Quartz pebbles with manganese coatings.
White Eocene (?) chalk cobble protrudes from bottom at right center.  Rounded
marl or claystone talus is derived from the wall on the left.

                                   Figure 8
3967 m.  Gravel and mud bottom.  Currents in this region are 25 to 30 cm/sec.
Rounded cobble in foreground is glacially derived.  White Eocene (?) chalk
boulder in background is believed to represent slumped material derived from
exposures farther up the Hudson canyon or from the adjacent slope.   Soft marl
or claystone talus is derived from wall at left corner.


                                   Figure 9
3960 m.  Semiconsolidated marl or claystone talus derived from the wall  at
right front of photograph.  White chalk and brown siltstone (?) boulder at
center is embedded in soft marl or claystone.   This material is thought to be
derived from exposures at shallower depths in the canyon or on the slope.

  !                                 Figure 10
39^1 m.   Rounded chalk boulder 2-3 m long, resting upon a hard gravel bottom
that is exposed in a prominent scour moat.  Such excavations and the general
absence of sediment dusting on boulders implies fairly strong and recent
current activity at the site.

exposures either as bedded deposits or as resedimented clasts.   The white chalk
blocks were probably derived from further up the continental slope where
similar rocks have been observed in place (Heezen and Dyer, 1977; Ryan et al.,
1978; Musick, pers. comm., 1978).  A sample of this rock type was collected at
the drum recovery site and was determined to be of mid-Eocene age based on its
contained microfauna (Table I).  A sample of the same lithology was taken during
investigation of the 2800 m dump site (Rawson and Ryan, 1978) and has also been
dated as mid-Eocene (Sample 679, Table I).  The white chalk appears to have
originally accumulated in a shallower and more landward environment in water
depths of 1500 to 2500 m (G. Blechschmidt, pers. comm.) and was  redeposited
at a later date in the Hudson  Canyon system as blocks, boulders, cobbles and
pebbles.  Thus it  is concluded that these rocks are not locally  derived but
have been transported down the canyon by some type of slide or mudflow mechanism.

     Scour moats are generally present around the bases of boulders in this
region of the Hudson channel.  These current scours are present  on the upcurrent
side of  the  boulders and a  sediment buildup or  drift  is usually  present  on the
downcurrent  side.   Scour moats (Fig. 10)  have a variety of sizes and  shapes but
are  5 to 10  cm  deep on the  average.  A  large quartzite boulder  of glacial
origin was observed that was  surrounded  by  an  impressive  scour  moat.   An older
mud  burial  line was still  visible  on  the boulder about 30 cm  above the exposed
base of  the  boulder.   This  provides  a  good  example  of recent  erosion.


 Description of Site

      The first radioactive waste drum was spotted during Dive 813 at 1540 hours
 and a depth of 3970 m (Fig. 11).  This drum was lying on a hummocky,  irregular
 sea bottom.   The three types  of talus blocks described above surrounded the
 drum.  Although these blocks  varied greatly in size many of them were about the
 size of the drum (Fig. 11).   A prominent scour moat was present on the upcurrent
 side of the drum, and a mound of granule-sized sediment had been built up on
 the downcurrent side of the drum (Fig. 12).  The drum was  intact, but highly

    :                               Figure 11
3970 m.  Photograph of the first drum located during Dive 813.  Currents
have kept the upper surface of this heavily corroded drum free of sediment
accumulation.   Soft marl talus litters the local area of this drum.   Corrosion
has obliterated identification markings.

                                   Figure 12
Photograph of first drum located on Dive 813.   This drum is resting on a hard
substrate.  Corrosion of the drum has streaked the sediment surface in a
downcurrent direction.  The corrosion products are coming from the upper right
hand corner of the drum.  Note large marl or claystone talus blocks lying about.

corroded and blistered.  Trails of corroding material were observed coming  from
the, drum (Fig. 12).  A core was taken in the corrosion-rich sediment and another
in |the sediment buildup on the downcurrent side of the drum (Table I).  The
sediment drift consisted of foraminiferal sand.  The core only penetrated about
8 cm and a hard light tan clay or marl substrate was observed.  Thus the
bottom in this area does not appear to be a product of recent deposition, but
instead probably represents an erosional surface covered with a thin biogenic
or hemipelagic sediment layer (Fig. 13).

     Another waste drum (No. 953) was located nearby (Fig. 14).  This drum was
surrounded by a well-defined scour moat similar to that around the first drum.
One end of the drum had a wedge of sediment sloping away from it.  This wedge
appears to have formed when the drum slid into its final resting position-,
plowing the sediment as it moved (Fig. 14).   Various sizes,, shapes, and types
of talus were lying around drum No. 953.  Abundant rounded white and brown
pebbles were present in the drum area (Fig.  7).  These gravelly deposits appear
to have been sorted by currents and are exposed in patches.  Elsewhere the
gravels are obscured by thin, hemipelagic surface deposits.  These gravels are
probably glacial  in origin and were later concentrated by currents flowing in
the canyon channel.  Small white anemones were attached to larger rocks near
the drum and it is interesting to note that these organisms;,  which favor hard
substrates, were absent from the drum surface.   Brittle stars and a few small
scorpion-like crabs and large rattail  fish were the dominant inhabitants of the
drum area.   As noted, no organisms were observed on the drum itself.   The
currents seemed to be stronger at this drum site than those of the previous
site,  Dive 812.   The velocity was estimated at 25 to 30 cm/sec.   When the mud
bottom was disturbed by the submersible, the sediment was quickly swept away
downcurrent after a few minutes.   The largest and deepest scour moats were
observed around the large talus blocks in this area (Fig.  10).   The site of
this drum is characterized by a highly variable, complex depositional  and
erosional topography.   Thirteen sediment cores, a water sample and a box core
containing a brittle star were taken from this site for various geochemical  and
radiochemical laboratory analyses (Fig.  13).

                       3970m ...j.,.v..... ..v..i-.i.....:i"^[-rC^..^
                             -^•v?^5ls^s^-'-:^i-••.-••"'•'	7
                                            FORAM OOZE (SAND)
                                            WITH CLAY AND SILT

       LT= LIGHT TAN
       G =GREY
                                                                         Figure 13

                                   Position and field descriptions of tube  cores taken downcurrent from Drum 953.

                                   Figure 14
3970 m.   Photograph of second drum (953) located.   Soft mud bottom is disturbed
by ALVIN.   Sediment wedge at the end of the drum may have been plowed up when
the drum slid into its present position.

     Before leaving the bottom, rock specimens were taken from a conical  pile
of unsorted rubble near the drum.   This rock pile is about 3 to 5 m high  and
consists of rounded white cobbles and boulders of chalk, glacial cobbles  and
locally derived tan marl talus blocks (Fig. 13, Fig. 15).  This mound of  rocks
was like others observed while searching for the waste drums.  Two samples were
taken from this pile.  The white cobbles (ALVIN 813 chalk) here are believed to
be representative of all the white blocks of boulder to cobble size encountered
along the bottom during Dive 813.   The mode of origin of these piles is unclear
at this point.  Perhaps these represent the toe of a circular slump that has
pushed this rock mixture to the surface, forming a small cone that has had-Us
fine-grained sedimentary component stripped away by recent currents.

Coring Program and Field Description of Cores

     The first waste drum discovered was a 55-gallon drum.  A scour moat was
present on the upcurrent side of the drum  and  a sediment buildup occurred on
the downcurrent side.   The drum was sitting on a firm bottom as evidenced by
its lack of burial (Fig. 11).  Two cores (Table I) were  taken with 40 cm-long
plastic core tubes,  but because of the  hard bottom  only  a few inches of sediment
were recovered in  each  core.  One core  (Core #2) was taken  in the area stained
by the corrosion  from the barrel near the  east corner of the drum and the other
core (Core #3) was taken in  the sediment buildup on the  downcurrent  side of  the

     A  series  of  13  tube cores, 10  of which are  discussed here,  were taken  at
the  recovery  drum site  (No.  953).   The  10  cores were  taken  at  approximately
2 m  intervals  starting  at  a  distance of about 14 m on the downcurrent  side  of
the  drum  (Fig. 13).   These cores were  to  be  used for  various radiochemical
and  geochemical  analyses  and only  the  field  description made through the
plastic core  tubes is available for this  report.   These cores  sampled  two
 distinct lithologies,  or sedimentary layers.   The  first layer consists of a
 4 to 10 cm thick foraminiferal  rich marl.   The percentage of clay and silt
 increases downward within this unit.   This layer is soft and porous and  is tan
 to light brown in color.   Near the drum and in the sediment buildup on the
 downcurrent side of the drum a significant amount  of corrosion products  (rust)

                                   Figure 15
3970 m.   Conical rock pile near drum 953.  Two cobbles were sampled from this
pile.   One is a glacial erratic, the other is Eocene-age chalk (Table I.)

from the drum is mixed into this layer, creating a reddish discoloration of the
sediment (Fig. 13, Core #1).   The upper layer is generally in sharp contact
with the lower layer which consists of semiconsolidated light tan, silty marl
or claystone.  This layer is quite hard as evidenced by the lack of core
penetration.  This layer is mottled in some of the cores and vertical burrows
exist in others.  The surface sediment has filtered down into the burrows.   A
pebble assumed to be quartz was noted along the contact between these two units
in Core 6 and probably represents a lag deposit formed during a period of strong
currents.  Core #1 was taken at an angle under the drum to test the oxidation
state of the  sediments under the drum.  This core contained considerable black
colored sediment, the cause of which is uncertain but may be due to the
formation of  sulfides under anoxic conditions.  This was the longest core taken
from this series  and was 28 cm long.  Average penetration of the cores was
about 15 cm.

     The contact  between these two lithologic units probably represents an
erosional surface or a local unconformity.  This erosional surface or
unconformity  is probably still forming in  some parts of the channel, while  in
other parts  of  the  channel foraminifera-rich  surface sediments accumulate above
it, particularly  behind rocks or other topographic  obstructions that break  up
the constant  southwesterly current flow and possible tidal (and/or inertial)
current  flow which  may be present.

                          INTERPRETATION AND DISCUSSION

     The radioactive waste  dump  site  studied  encompasses  an  area  of  a  few
tens of square  kilometers  centered around 70°35'W to 37°50'N.  According to
available records,  about  15,000  55-gallon drums  of low-level  radioactive
waste  embedded  in concrete  are  reported  to make up the  contents  of the dump
site  (Dyer, 1976).   Only  three  drums  were located during  Dives 812 and 813
and one drum (No. 953)  was  recovered  on  Dive  814.   The  drums located on these
dives  appear to be lying  near the base of the eastern  side of the Hudson
 submarine canyon channel  on the lower continental  rise  at a depth of 3970  m
 (Fig.  3).   The main axis  or thalweg of the channel appears to lie about 1  km  to

the west based on a limited bathymetric survey (Fig. 3).  There is some
evidence that this deeper axis meanders across a relatively flat channel floor
(Figl 3).  Where the channel thalweg hugs the channel wall, tidal and/or
contour currents may be concentrated to produce higher velocity currents than
those observed on the channel flats (15-30 cm/sec).  Higher velocity currents
might cause the thalweg to cut laterally into the soft marl or claystone
bedrpck thus oversteepening the slope.   Oversteepening or undercutting at the
basejof the slope would allow slumps or gravity slides to occur.  Based on the
observations made during Dives 812 and 813, there seems to be little doubt that
this; is a region characterized by major slumping and that the oversteepening
mechpnism could account for the observed scarp and bench topography.   The scale
of this slumping is difficult to assess.  Small slumped blocks and scarps are
easijly observed, but larger features are difficult to assess from a submersible.
It is most probable that the smaller features observed indicate that slumping
on ajmuch larger scale is taking place (Fig.  16).   These data indicate that the
canyon axis is presently being filled along its margins through the process of
slumping from the walls.
    JA possible model for the earlier incision and later partial filling of the
Hudsbn Canyon channel is suggested below and illustrated in Figure 16.  During
the ;last low sea level stand some 15 to 20,000 years ago, a much deeper, steep
walled canyon was cut into the deposits of the continental rise.  Many previous
workers have both suggested and documented that the submarine canyons of the
continental slope were deeply incised during this period (Shepard, 1952;
Stetson, 1936, 1949).  The Pleistocene age Hudson Canyon channel or seaward
extension of the Hudson Canyon may have been a deep vee-shaped valley cut or
formed by the action of sediment-laden turbidity currents or submarine debris
flows as they passed seaward through this region towards the Hudson Fan Complex.
As the Wisconsin-age glaciers receded, sea level rose and the sediment supply
to the continental shelf, slope, rise and abyssal plain was diminished.  At
this time, the Hudson Canyon channel began to backfill from the seaward side.
Two possible mechanisms, operating together, could account for this backfilling:

                                                             Figure 16

                             Hypothetical  model  for large scale slumps as a canyon filling mechanism.

 with the many documented slump scarps along the channel walls, indicate that
 the channel floor is a site of net accumulation.

  ;    If these postulated geologic processes are indeed operating in this area,
 some geologic recommendations in terms of waste disposal can be made.  First,
 it  is clear that this is a very unstable area as indicated by slump scarps,
 avalanche deposits,  and a lack of biological  activity.  Concrete filled drums,
 once introduced into the canyon system might  behave as any other sedimentary
 particle, that is,  these drums may move down  the channel just as the chalk and
 siltstone cobbles and boulders have moved.  The barrels observed on Dive 813
 were probably in their original  dumped position but some evidence of movement
 was  indicated (plowed sediment)  (Fig.  14).  It is  also possible that these
 drums may have slid  down the  channel  walls  to their present position,  which
 would account for their position  at the base  of the channel.   Any drums present
 on  the channel  floor could  be  subsequently  moved or buried by slides or slumps
 from the  channel  walls  which may  plow  into  the channel  area with  great force.

  :    Many speculations  and  assumptions  have been presented in this  report,  but
 these are considered  reasonable based  on  the  available  data.   Further  study is
 needed  to support, alter  or disprove the  suggested  geologic processes  operating
 in. the  lower  reaches  of  the Hudson  submarine  canyon.   Based on  this  study,  this
 area  is classified as a  dynamic deepsea environment  whose  agents  and processes
 are  presently poorly  understood.  With  respect  to future, waste  disposal
 activities  in this and other canyon environments, material  dumped on the flat
divides between channels may eventually be  displaced or  buried  if the
 interpretation of successive slumping and filling of channels  is correct (Fig.
 16).   If  low-level radioactive waste material   is to  be dumped  in the North
Atlantic, the flat and featureless abyssal  plain areas to  the northeast would
be a geologically more stable area to consider for the possibility of future


     11.  i
Cacchione, D.A., G.T. Rowe, A. Malahoff, 1978,  "Submersible  Investigation
of Outer Hudson Submarine Canyon," in Stanley,  D.J. and G. Kelling,  eds.,
Sedimentation  in Submarine Canyons, Fans, and Trenches, Dowden,
Hutchinsen & Ross, p. 42-50.

Dyer, R.S., 1976, "Environmental Surveys of Two Deepsea Radioactive  Waste
Disposal Sites Using Submersibles," in Proceedings, International
Symposium on Management of Radioactive Wastes from the Nuclear Fuel  Cycle,
International  Atomic Energy Agency, Vienna, v.  2, p. 317-338.

Eittreim, S. and M. Ewing, 1972, "Suspended Particulate Matter in  the  Deep
Waters of the  North American Basin," in Studies in Physical  Oceanography,
A.L. Gordon, ed., Gordon & Breach, NY, v. 2, p. 123-168.

Emery, K.O. and E. Uchupi, 1972, "Western North Atlantic Ocean Topography,
Rocks, Structure, Water, Life and Sediments," Amer. Assoc. Petrol. Geol.
Mem., 17, 532  pp.

Heezen, B.C. and C.D. Hollister, 1971, The Face of the Deep, Oxford
University Press, NY, 659 pp.

Heezen, B.C. and R.S. Dyer, 1977, "Meandering Channel on the Upper
Continental Rise of New York," EOS, Transactions, American Geophysical
Union, v. 58,  p. 410.

Hollister, C.D., J.I. Ewing, et al., 1972, Initial Reports of the  Deep
Sea Drilling Project, v. XI, U.S. Government Printing Office, Washington,
DC, p. 313-319.

Rawson, M.D. and W.B.F.  Ryan, 1978, "Geologic Observation of the Atlantic
2800-Meter Radioactive Waste Disposal Site," U.S. Environmental Protection
Agency Report  520/1-83-018, 86 p.

Ryan, W.B.F.,  M.B. Cita, E.L. Miller, D. Hanselman, W.D. Nesteroff,
B.  Hecker, and M. Nibbelink, 1978, "Bedrock Geology in New England
Submarine Canyons," Oceanologia Acta, v. 1, p.  233-254.

Shepard, F.P., 1952, "Composite Origin of Submarine Canyons," Jour.  Geology,
v.  60, p. 84-96.

Smith, M.A., R.V. Amate, M.A. Furbush, D.M. Pert, M.E. Nelson, J.S.  Hendrix,
L.D. Tamm, G. Wood, Jr., and D.R. Shaw, 1976, "Geological and Operational
Summary, Cost  No. B-2 Well, Baltimore Canyon Trough Area, Mid-Atlantic DCS,"
U.S. Geological Survey Open File Report 76, p.  774-779.

Stetson, H.C., 1936, "Geology and Paleontology  of the Georges Bank
Canyons," Geol. Soc. Amer.  Bull., v.  47, p. 339-366.
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE


                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 EPA 520/1-83-017
              3. RECIPIENT'S ACCESSION NO.
 1978 Atlantic 3800 Meter Radioactive Waste Disposal  Site
 Survey:  Sedimentary,  Micromorphologic and Geophysical
              5. REPORT DATE
                 June 1983
  David H. Hanselman,  Ph.D.
  William B.F. Ryan, Ph.D.
  Lamont-rDoherty Geological Observatory
  of Columbia University
  Palisades, New York  10964
              10. PROGRAM ELEMENT NO.
              11. CONTRACT/GRANT NO.
               Contract No. 68-01-4836
 Office of Radiation Programs
 U.S. Environmental Protection Agency
 401-M Street, S.W.
 Washington, D.C.  20460	
              14. SPONSORING AGENCY CODE

 16. ABSTRACT            :            ~~~	'	—	——	
  During jthe period of 22-28  July 1978, five dives were made  in  the manned submersible
  ALVIN into the Atlantic Ocean 3800-meter depth radioactive  waste disposal site located
  in the [Hudson Canyon channel  approximately 320 kilometers from the Maryland-Delaware
  coast. | A geological description of the site was made by direct examination of the
  bottom (topography, bedrock  exposures, sedimentary and erosional processes, and sedi-
  ment cojres collected from the dumpsite area.  Observations  within a depth range of
  3985-38f30 meters revealed angular blocks and piles of displaced channel wall rock,
  boulder; and cobble olistoliths of Eocene-age chalks derived from higher elevations on
  the slope, and bedforms such  as ripples and scour marks which  imply the existence of
  periodic strong currents.   Local benthic fauna, were sparse.  Three low-level radio-
  active waste drums were examined from the submersible, and  one was subsequently
  recoverjed for corrosion and concrete deterioration analyses.   Photographic and visual
  evidence suggest that downslope transport of objects such as talus blocks, olistoliths
  and radioactive waste drums has occurred in this area.
                                KEY WORDS AND DOCUMENT ANALYSIS
 ocean disposal
 low-level radioactive waste disposal
 deepsea geology
 Hudson Canyon geology
 marine foraminifera

  Unlimited Release
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE
                                              b.IDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLASS (This Report)
                                              2O. SECURITY CLASS (This page)
                                                                        c.  COSATI Field/Group
                           22. PRICE



United States
Environmental Protection
Washington DC 20460
Official Business
Penalty for Private Use $300
Postage and Fees Paid
Environmental Protsction Agsncy