AD-A122 47 BERNG STRA SEA IC AND THE FAIRWAY RO ICEFOOT(U) 1/1COLD REGIONS RESEARCH AND ENGINEERING LAB HANOVER NH

A KOVACS ET AL. OCT 82 CRREL-82-31UCSSIFIED F/G 8/12 . NL

UNCLASS E

* f l f l f * * f l f END

= h

MICROCOPY RESOLUTION TEST CHARTNATIONAL BUREAU OF STANDARDS-1963-A

S&L5

i +u.

IJ02

7 ADA122 47 7

REPORDTf82'31 US Army CorpsE1E~iP~It~S~i~IFEof Engineers

Cold Regions Research &Engineering Laboratory

Bering Strait sea ice and theFairway Rock ice foot

La

82 12150

CRREL Report 82-31October 1982

Bering Strait sea ice and theFairway Rock icefoot

A. Kovacs, D.S. Sodhi and G.F.N. Cox

'A'al a 1n d/or

i"'PGt V Special

Approved for public release, distribution unlimited.

,, U.I ,il I

! Jnnatif.A

SECURITY CLASSIFICATION OF THIS PAGE (Mnm Date Enteroed

READ INSTRUCTIONSREPORT DOCUMENTATION PAGE BEFORE COMPLETING FORM

1. REPORT NUMBER 2. GOVT ACCESSION NO 3. RECIPIENT'S CATALOG NUMBER

CRREL Report 82-31 - / - 7__ _

4. TITLE (and Subiltl.) S. TYPE OF REPORT & PERIOD COVERED

BERING STRAIT SEA ICE ANDTHE FAIRWAY ROCK ICEFOOT S. PERFORMING ORG. REPORT NUMBER

7. AUTHOR(s) S. CONTRACT OR GRANT NUMBERre)

A. Kovacs, D.S. Sodhi and G.F.N. Cox

9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT. PROJECT, TASK

AREA & WORK UNIT NUMBERS

U.S. Army Cold Regions Research and Engineering LaboratoryHanover, New Hampshire 03755

II. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE

U.S. Army Cold Regions Research and Engineering Laboratory October 1982

Hanvoer, New Hampshire 03755 4414. MONITORING AGENCY NAME & AODRESS(If different &ro Controlling Olfieo) IS. SECURITY CLASS. (of thie report)

Unclassified

ISe. DECLASSIFICATtON/DOWNGRADINGSCHEDULE

Is. DISTRIBUTION STATEMENT (of thl Repopr)

Approved for public release; distribution unlimited.

17. DISTRIBUTION STATEMENT (of the abtract emtered In Block 20, II different from Report)

1a. SUPPLEMENTARY NOTES

Funded by Bureau of Land Management/Gulf Canada Resources Inc. National Oceanic and AtmosphericCalgary, Alberta, Canada Administration

Washington, D.C.IS. KEY WORDS (Cmlnue mmn reverse side it necess.r mid Identify by block mmber)

Ice'Rea iceIce forces

24L AMTRACI(Cknt b an rwes aftb neeemy mod t fy by block m,,ber)

Information on sea ice conditions in the Bering Strait and the icefoot formation around Fairway Rock, located in thestrait, is presented. Cross-sectional profiles of Fairway Rock and the relief of the Icefoot are given along with theoret-ical analyses of the possible forces active during icefoot formation. It is shown that the ice cover most likely fails inflexure as opposed to crushing or buckling, as the former requires less force. Field observations reveal that the FairwayRock icefoot is massive, with ridges up to 15 m high, a seaward face only 20' from vertical, and interior ridge slopesaveraging 33". The icefoot is believed to be grounded and its width ranges from less than 10 to over 100 m.

00, W3 a- 1o c o Nov s1 @5. LETEaSECUIY CLANWFPCATION OP TNIs PAW (o, em Date

/- . ~1 r ,

PREFACE

This report was prepared by Austin Kovacs, Research Civil Engineer, Applied Re-search Branch, Experimental Engineering Division; Dr. Devinder S. Sodhi, Re-search Hydraulic Engineer, Ice Engineering Branch, EED; and Dr. Gordon F.N.Cox, Geophysicist, Snow and Ice Branch, Research Division of the U.S. Army ColdRegions Research and Engineering Laboratory.

Support for the study was provided by Gulf Canada Resources Incorporated, Cal-

gary, Alberta, Canada, and by the Bureau of Land Management through an inter-agency agreement with the National Oceanic and Atmospheric Administration aspart of the Alaska Outer Continental Shelf Environmental Assessment Program.The authors thank Douglas C. (Skip) Echert of Oceanographic Services Incor-porated, who assisted with the aerial reconnaissance.

UI

_____

CONTENTSPage

Abstract...................................................Preface ..................................................... iiIntroduction .................................................. 1IBering Strait.................................................. IField reconnaissance ........................................... 11Estimation of ice forces on Fairway Rock........................... 28

1. Creep deformation ........................................ 282. Crushing failure ........................................... 283. Flexural failure ........................................... 294. Forces required to form floating or grounded pressure ridges along the

rock or to pile ice on the beaches .............................. 305. Buckling failure ........................................... 31

Driving forces................................................ 34Angle of internal friction of sea ice ................................ 35Summary.................................................... 36Literature cited............................................... 37Appendix A: April 1982 field observations at Fairway Rock .............. 39

ILLUSTRATIONS

Figure1. Landsat view of the Bering Strait on 22 June 1973 .................. 22. Representative underwater environment, south side of Fairway Rock 23. Side-looking airborne radar image of 20 February 1980 showing Cape

Prince of Wales and the sea ice in the shear zone north of the BeringStrait.................................................. 3

4. Daily mean values of water transport through the Bering Strait ......... 45. Atmospheric pressure pattern over Siberia and Alaska on 22 Nov 1976. 56. Floe conditions in Bering Strait on 6 March 1973 at beginning of major

rapid southern ice drift .................................... 57. Ice conditions in Bering Strait on 7 March 1973 .................... 68. Area map of the Bering Strait................................. 79. Two thick multi-year sea ice formations and one thinner multi-year sea

ice floe ................................................ 710. Ice field movement from Norton Sound into the Bering Sea and on into

the Chukchi Sea....................................... 811. Fifteen-day mean sea level pressure composite charts ............... 912. South side of Fairway Rock .................................. 1113. Morphology of icefoot on north side of Fairway Rock .............. 1114. South side of Fairway Rock .................................. 1215. West side of Fairway Rock .................................. 1216. North side of Fairway Rock .................................. 1217. East side of Fairway Rock ................ ..,***...............1318. Termination of reconnaissance flgtdue to fuel loss through failed gas

cap seal................................................ 13

uIl

Figure Page

19. Aerial view over "south" side of Fairway Rock .................... 14

20. Aerial view over "north" side of Fairway Rock .................... 1521. Elevation profile of icefoot and Fairway Rock along line IA ......... 1622. Elevation profile of icefoot and Fairway Rock along line 6F .......... 1623. Fairway Rock-icefoot profile 2B ............................... 1724. Fairway Rock-icefoot profile 3C ............................... 17

25. Fairway Rock- icefoot profile 4D ............................... 17

26. Fairway Rock- icefoot profile SE ............................... 18

27. Fairway Rock- icefoot profile 7G ............................... 1828. Fairway Rock-icefoot profile 8H ............................... 1929. Fairway Rock icefoot, April 1981 ................................ 20

30. Icefoot on southeast side of Fairway Rock ........................ 21

31. Icefoot rubble along south side of Fairway Rock ................... 21

32. Icefoot on southeast side of Little Diomede Island .................. 2233. Icing on north side of Little Diomede Island ....................... 2234. Icing on rock surface at the top of Little Diomede Island ............ 23

35. Aerial views of shear ridge fields west of Prince of Wales Shoal ....... 24

36. Shear ridge field profile 30 ...................................... 2537. Shear ridge field profile 31 ...................................... 25

38. Shear ridge field profile 32 ...................................... 2539. Shear ridge field profile 33 ...................................... 2640. Shear ridge field profile 34 ...................................... 2641. Shear ridge field profile 35 ...................................... 26

42. Shear ridge field profile 36 ...................................... 2743. Shear ridge field profile 37 ...................................... 2744. Coefficient C, versus slope angle and friction ...................... 3045. Coefficient C2 versus slope angle and friction ...................... 3046. Sea ice pressure ridge and shore ice pile-up models .................. 3047. Geometry of a wedge-shaped floating ice sheet ..................... 3148. Non-dimensional buckling load for different ice sheet wedge angles a,

R/L values and buckling conditions ............................ 32

49. Ice floes in the Bering Strait ..................................... 3450. Ice wedge developed on the south side of the Diomede Islands, 2 April

1975 ...................................................... 35

5 I. Ice wedge developed on the south side of the Diomede Islands, I I April1975 ...................................................... 35

TABLES

Table PageI. Calculated creep ice loads ...................................... 28

2. Values of H /b ................................................ 293. Buckling pressure for different RIL and boundary conditions ........ 334. Far-field ice pressure required to buckle a semi-infinite ice sheet ...... 34

iiI

A-Yr. f

BERING STRAIT SEA ICE AND THEFAIRWAY ROCK ICEFOOT

A. Kovacs, D.S. Sodhi and G.F.N. Cox

INTRODUCTION the western side and about 60 m near the easternside. The bottom relief is probably the result of

In the design of offshore structures to be placed scouring by strong currents, ice gouging, and an-in arctic waters, major consideration is being cient river scouring (Selkregg 1976). On the west-given to determining the loads developed during ern side of the strait is the formidably steep andice movement and failure against a structure. This rugged mountain topography of Cape Dezhnevacould result in the creation of an ice rubble field and on the eastern side is the bold mountain land-which could increase the effective width of the scape of Cape Prince of Wales. Winds in the straitstructure and the forces transmitted to it. The phe- tend to be funneled and accelerated in northerly ornomena of ice pile-up and over-ride are also major southerly directions by these headlands.design considerations. Within the Bering Strait are three islands (Fig.

A variety of analytical and model studies have 1): Little and Big Diomede Islands and Fairwaybeen made to investigate ice forces on offshore Rock. Fairway Rock, situated about 24 km west-structures, but they are inconclusive, as their re- southwest of Cape Prince of Wales, Alaska, is asuits have not been verified by field measure- 350-in-diameter igneous rock that rises almost ver-ments. This paper presents the results of an inves- tically out of SO-in-deep water (G. Bloom, pers.tigation of the general configuration of ice rubble comm.) to a height of about 165 m. Its top slopesaround Fairway Rock, Alaska. The purpose of the gradually from the northwest to the southeast andinvestigation was to acquire data on the morphol- is covered with a thin growth of hardy tundraogy of the sea ice rubble surrounding the rock that grasses, lichens and mosses. At sea level, the is-would be applicable to offshore structures in gen- land exhibits surfaces of near-vertical rock anderal and to those placed in deep water in particu- steep talus slopes composed of large boulders.lar. Estimates of the relative force levels required From the tidal zone to a depth of about 5 in iceto form the observed rubble are given. The surface abrasion inhibits marine growth (Fig. 2). Belowrelief across a number of sea ice rubble fields this depth abundant flora and fauna exist (Shum-formed in the shear zone along the west side of way et al. 1964). In winter a very high pile of icePrince of Wales Shoal is described for the purpose rubble surrounds the island. The lateral extent ofof documenting the size of ice features which may this icefoot is limited by the abrupt drop-off of theimpact offshore structures placed in the northern island into deep water and by the erosional forcesBering Sea. of drift ice, which pluck, cleave and grind away

the ridged ice. Oceanographic instrument cablesextending downward along the submarine boulder

BERING STRAIT surface of the island have been extensively dam-aged to a depth in excess of 30 m by ice impact and

The Bering Strait is 85 km wide and has an ir- abrasion (Bloom and McDougal 1967).regular bottom, with a depth of about 52 in near Fast ice extends offshore about 1 km at Cape

Farw

Figure 1. Landsat view of the Bering Strait on 22 June 1973.

,

10

520

Figure 2. Representative underwater environment to a depth of 20 mat the south side of Fairway Rock (after Shum way et aL 1964).

2

Prince of Wales and increases in width to the. north. Its maximum thickness is about 1.4 m. All

ice seaward of this shore-fast ice is in nearly con-stant motion. Along the western edge of the fast

. "ice, extending in a general north-northeast direc-A" • tion, very long shear ridge-rubble fields form as a

4' .,result of pack ice pressing in against the stable ice• . 2. ":(Stringer 1978). These linear ice formations are

Chukchi Sea " " " easily visible in Figure 3. Many of the shear ridgefields have been estimated to be over 7 m high(Kovacs 1972). On Prince of Wales Shoal, which

. extends some 60 km in a northward direction fromthe eastern tip of Cape Prince of Wales, large,

- "' 14-m-high, isolated, grounded ice formations have., ' ;also been observed (Kovacs 1972).

Oceanographic cables laid offshore at Cape, .4v Prince of Wales have been consistently cut or

damaged by drift ice, usually during early freeze-up when pressure ridges develop along the fast ice

4 edge or during spring breakup when grounded iceis pushed across the sea floor. Time domain reflec-tometry measurements on the broken cables haverevealed that they are consistently damaged, be-ginning at about the 10-m depth and extendingoutward to about the 25-r. depth. Sounding sur-veys have shown that ice gouging of the sea floorcan be expected to a depth of 4 m, and that ice

Ip .r gouging extends out to about the 25-m waterdepth (Bloom and McDougal 1967).

Strong currents are characteristic of the BeringStrait flow regime. These currents consist of asemipermanent northward flow on the order of0.8 ±0.2 Sverdrup (Sv) (Coachman and Aagaard1979). A downward sea surface slope of about 2.6

Cape Prince of WN x 10- to the north is the primary mechanism re-sponsible for this flow through the strait (Coach-man and Aagaard 1966). However, wind and re-gional atmospheric pressure can cause significantspatial and especially temporal variation (on atime scale of days) in this flow, and major south-erly flow can occur for short periods of time. If itwere not for the strong southerly transport events,the annual northern transport would be higher by

• a factor of about 5. For example, between Sep-tember 1976 and March 1977 there were 20 flowreversals, each lasting 6 to 12 days, for a total of61 days. The transport-time record for this period(Fig. 4) shows that the largest transport through

Figure 3. Side-looking airborne radar image of 20 the strait occurred on 27 October. It was in aFebruary 1980 showing Cape Prince of Wales and southerly (-) direction and had a volume of ap-the sea ice in the shear zone north of the Bering proximately 5.05 Sv, which for the Bering Strait isStrait. equivalent to a mean flow velocity of 4.4 km/hr

(Coachman, pers. comm.). The largest northwardtransport of 3.15 Sv occurred on 7 January; meanflow velocity was 3.0 km/hr. The high flow veloci-

3

ties which occur during these events may be thereason why More (1964) observed the sea floor tobe essentially devoid of sediment for approxi-

mately 45 km northward of Little Diomede Island.Coachman and Aagaard (1979) show that trans-

port through the Bering Strait is well correlatedwith regional east-west atmospheric pressure dif-ferences (see Fig. 5): "When pressure on the eastside (Nome, Kotzebue) is greater than on the west

E side (Provideniya Bukhts, Cape Serdtse-gamen),_t the regional air flow is towards the north, as are

the water transports, and vice versa. SoutherlyQ transport through the strait has a marked episodicE character, and these events are caused by particu-

lar large scale atmospheric conditions. When astrong E-W pressure gradient lies over the strait

0%-. and extends in a north-south direction from theChukchi Sea to the central Bering Sea, entirely

crossing the northern Bering Shelf, extensivet northerly winds move water southward off the

shelf. This produces a sea-level slope down to thesouth which, together with the northerly winds,drives southward transport. The atmospheric

0% pressure pattern causing this condition is always a

strong low located a considerable distance south-N east of the strait (e.g. over Kodiak) together with

the Siberian high, being centered some distanceP west of the strait."

- E -~i Movement of ice southward through the BeringStrait is therefore driven by northerly winds andsoutherly current flow. The resulting coupling ofthe wind and current produces drag forces on theice which exceed its arching strength. Thus thearched or jammed-up ice bridging the strait fails,and moves rapidly southward (Fig. 6 and 7). Twosuch events were studied by Ahlnas and Wendler(1979). One event in January 1977 resulted in amean estimated ice velocity through the strait of4.1 km/hr. A second in March 1978 resulted in amean ice velocity of 2.8 km/hr. These speeds aremuch higher than the typical speed of ice floes inthe Bering Sea that have been tracked by satelliteimagery. The latter always move at less than 1. 1km/hr and generally at less than 0.7 km/hr

' (Muench and Ahlnas 1976).Under strong driving forces acting over a pro-

longed period of time, sea ice in a belt 100 or morekm wide, extending from the strait northwardalong the east side of the Chukchi Sea to Pt. Bar-row and beyond, gradually moves southward (Fig.8). In some years this movement brings multi-yearice floes into the northern Bering Sea. St. Law-rence Island natives interviewed at Gambell inApril 1981 report seeing multi-year ice floes on oc-

AS 'iV~dsNVw if casion. During our reconnaissance flights we have

4

r ____

2''Z

170o

IO. 170' r 'o,,O

Figure 5. Atmospheric pressure pattern over Siberia and Alaska on 22November 1976 (from Coachman and Aagaard 1979).

Prince of al~es Shoo

Figure 6. Floe conditions in Bering Strait on 6 March 19; at beginning of major rapid southern

ice drift.

5

- ,, _ " .. ..- -,,,

,-<-', . , -" r,, : - # ..~A -r

ti

Figure 7. Ice conditions in Bering Strait on 7 March 1973. Note the position of the numbered ice floes in relation totheir locations in Figure 6, the splitting of an icefloe moving past Fairway Rock, and ice movement along the shear zonewest of Prince of Wales Shoal. Also note the turbulent wind wakes downstream of various landforms.

6

__________________________________~1L- ---

,Mw A'. .i

Chukchi-, ' /

Seae

Figure 8. Area map of the Bering Strait. Dashed lines indi-cate zone of major ice stream which occurs during large south-ern ice drift. Stippled area is zone of maximum drift.

-o:

-ZI

Figure 9. Two thick multi-year sea ice formations, center left in photo, andone thinner multi-year sea ice floe, near top center of photo, located off TinCity. Tin City is situated at the bottom on the south side of Cape Mountain (Fig. 3).Note shore ice pile-up on the terraced beach.

7

2/24/ 0 2/ 5/79C H U K C H I/

22142 2192/24/19

2126/79 20462z SEA ~2 114 Z

COP opeOelnavo 2/23/792104Z

Soria# Dolmede XIs

4 2/22/

2 2/9 Foiwe~y ~, COPS PrinCe of W0195154?7Z

2/2117965 1557Z

16042Z

2/19/179161 2/1879 17920472

1504Z

SEA 21/79

1507Z

17164Figure 10. Ice field movemlent from Norton Sound into the Bering Sea and on into the Chkehi SeOfro27 Jaw to 289 March 1979 (datar soumre, Sea Ice Consultants, Camp Sprns Mdal

8rns d)

observed multi-year sea ice on the south side of In any event, once the Bering Strait ice arch col-King Island, offshore of Tin City (Fig. 9) and lapses, in excess of 60,000 km2 of sea ice from thenorth of the Bering Strait. Chukchi Sea can move southward at high veloci-

A number of investigators have studied the Ber- ties in a relatively short period of time (Ahlnasing Strait ice breakout phenomenon and the mag- and Wendler 1979). Fairway Rock is situated innitude of the forces needed to release the ice jam the center of the major ice floe stream through theand allow the ice to flow southward through the strait.constriction (Shapiro and Burns 1975, Sodhi 1977, Movement of ice northward through the BeringPritchard et al. 1979, Reimer 1979). Pritchard et Strait is also driven by wind and current. Figure 10al. (1979) found that if failure is caused by wind shows the movement of an ice field tracked on sat-and current drag alone, i.e. neglecting any contri- ellite imagery by Sea Ice Consultants in February-bution from internal stress transmitted through March 1979 (unpublished data, courtesy W.S.the ice from distant locations, then an order of Dehn). Between 27 and 31 January the ice fieldmagnitude body force of 0.67 Pa is needed to col- moved northwest with the strong coastal currentlapse the arch. The assumptions are that the ice out of Norton Sound. Thereafter, under changinghas a strength of 1.5< 10' N m' and that failure wind fields, as depicted in Figure 11, the ice floesoccurs along a diamond-shaped yield surface, moved first southward to a location east of St.

First Half February (/28/79- 212/79)

Figure )). Fifteen-day mean sea level pressure composite charts. Arrowsindicate wind direction. Numbers are pressure, e.g. 39 is 1039 millibas. Thesaddle area between the high pressure armas (H) is a zone of weak con vergmceand divergence (data source, Sea Ice Consultants, Camp Springs. Md.).

9

ono

IIn

k

Lawrence Island, and then northward through the

Bering Strait. About 75 km north of the BeringStrait the floes changed direction again, first head-ing east and then southward. These ice floe move-

ments are instructive for they represent one of themore complex routes which may be followed bypetroleum introduced into the Norton Soundmarine environment.

It is apparent that ice movement through theBering Strait is both highly variable and quite dy-namic. As a result, Fairway Rock is frequently im-pacted by either southerly- or northerly-movingice floes throughout the winter season.

FIELD RECONNAISSANCE

The 1980 field program consisted of two recon- Figure 12. South side of Fairway Rock. Note lownaissances of the island in a small fixed-wing air- cloud ceiling.craft, during which hand-held camera photog-raphy and visual observations were made. One

overflight in a remote sensing aircraft also ob- quently pushing in against and shearing past the

tained stereo aerial photography of the rock and ice rubble which had formed during various north

portions of the ice rubble in the shear zone along and south ice movement events. On the south side

Prince of Wales Shoal. of the island a large ice floe was driven against the

The first reconnaissance of Fairway Rock was wall of the icefoot by the prevailing northward

made on 15 February 1980. The day was overcast, current flow (Fig. 12). To the north of the island

with a ceiling of about 150 m (Fig. 12). It was was open water. At the base of the island on the

noted that the rock was surrounded by an icefoot north side, a large rock talus area was noted which

with a very high "vertical" face. The piled-up ice upon first observation appeared to be ice rubble

appeared highly consolidated due to floes fre- from a large ice ride-up event (Fig. 13).

;J

Figure 13. Morphology of icefoot on north side of Fairway Rock. Note rock

talus area covered with snow.

II

00

V is

12 I-

1 4 0Figure 17. East side of Fairway Rock. Dark objects on top of island are largepropane gas tanks and a generator installation.

00

Figure 18. Termination of reconnaissanceflight due to fuel loss through failedgas cap seal. (Landing gear destroyed, right wing tip bent, prop and stabilizer bent.)

The second reconnaissance was made on 26 diffused pack were noted, whereas to the north the

April under a broken cloud cover. High winds and pack ice could be seen in an apparent holding line

associated turbulence around the rock caused the between the Diomede Islands and Cape Prince of

small aircraft to be thrown around and prevented Wales. The northerly winds, while high, were not

a close-up inspection of the icefoot. Nevertheless, strong enough to drive the pack ice south against

the photos taken show an impressive accumula- the prevailing southerly current.

tion of pressurized ice forming the icefoot, partic- A view of the south side of the rock is shown in

ularly on the north and south sides. At the time of Figure 14a. Note the steep face of the icefoot and

the reconnaissance, open water surrounded the is- the slope of the top of the island. A similar view of

land. Off to the south significant open water and a the rock taken on 8 April 1962 is shown in Figure

13

-' i ... .. - . . ... i ,m ... .. ... ...... ...... ..

.L Z . " ..

, *r J b * . . "

-.. ,. ,, 1*

ft.)

"b." A"comparisonof e ie 4a a

ai foo t o

controledin.art by sumain reie a wl as by viwo h rcFg,,)sosprtoso h

th genel IP A vi

nore half oterislandis sveen cadowth Foloiniheflb of Fairway Rock, lvto rflswr ae aro strorhc on-

soand along thrembaseerok sulrfceois coy- Inatisncetof pessre ridge eaureas witinle thefigredon w ithatn ayer ofgaeandry ice. goth is is t shear oews fa ice of W aes Sheoal And ofs

S-rr

controlled~ ~ ~ in patb8umrn eif swl sb iwo hc(Fg17shwpotnsfte

the genra shp fteiln.Terde eifvrialrc aea e ee n hc cn cof hesothrn ndnothrn. portion ofteie,"uain oern h okt naprn

nore half oter i slan ie see n cakdou th Foloinideflb of Fairway Rock. alvto prflswr ae rmseo rhc oto

soand al song th a s thear ok siuilar ceois cov- inatispheofo pessre ridge etus re s witineaerdfiurton ina tinerofglazand e ry ice. goThis is t shear ews fa ice of a es Sheoa. and ofs

cotrlld n at y ub arn rlif s el s y ie f herok Fg.17 sow orios1f4h

,,. ..,~ ~ ~J -A•, ..,'.'

,,-.. , ,,k.

• , -. ,,, .,,'

•96" li p"

,, '

Figure 20. Aerial view over "north" side of Fairway Rock. North is to the right. (Phoo ws reduced to about 80%

of ( riginal size, thus 0. 8 in. = 333 ft.)

ice rubble piles on the shoal was planned. How- narrow width of the icefoot on the west and south-

ever, an engine malfunction and a short (30-m) east sides of the rock. In these areas the rock wall

two-bounce sideways forced landing (Fig. 18) pre- seems to drop vertically into the sea. Unlike the

luded accomplishment of this objective. view of the rock shown in Figure 0 the south face

Aerial photography of Fairway Rock and por- of the rock at this time was caked with snow

tions of the Cape Prince of Wales shear zone was and/or ice. The top of the rock is clearly wind-

obtained on 14 March 1980. Two views of Fairway swept; visible are shadows from a cluster of pro-

Rock are shown in Figures 19 and 20. The ridge pane tanks and a small generator used to power

systems composing the southern icefoot are quite salinity, temperature, conductivity and current in-

apparent due to the favorable sun angle. The gen- struments on the sea bottom (G. Bloom, pers. or-

eral bluntness of the icefoot is striking, as is the respondence).

* 15

CM) (f 0)160-

500

120- 400

300

200

40-

0-0o

0 1000 2000CA

0 100 200 300 400 500 600 (m)Distan-

Figure 2). Elevation profile of icefoot and Fairway Rock along line MA.

CM) 00t

500

120 -140

1300

200

40-

100

0 000 WOO0(MI A I I0 Poo 200 30400 WO0 60 m)

Distance

Figure 22. Elevation Profile of icefool and Fairway Rock along line 6F.

Elevation profiles of the ice rubble surrounding cess of 70 *on the seaward side, and in places reststhe island were made from stereopairs. The profile against an apparently near-vertical rock wail. Cal-lines are drawn and numbered in Figure 19. The culated interior ridge slope angles averaged 33 .elevation profiles are presented in Figures 21-28. Field measurements of free-floating first-ye.rThese profiles reveal that the icefoot attains an pressure ridges have shown that their sail height toelevation of at least 14.6 m. has a steep slope in ex- keel draft ratio is of the order of I to 4.5 (Kovacs

16

(m) (ft) (m) (ft)10 110

30 30

90 90

TO 70 -20 20-

C £

o 050 -- 50 -5

10- 30 -1 2- 30 -2

45 1

-10 Ic 1 4

0 0 - -A-- - 0_ 0 .I'-

-3 F -0 - 1 O1/t .- - -0 1- 014015

0 500 (ft) 0 500 ft)I I a I 2 _.

0 100 200 (m) 0 100 tm)Distance Distance

Figure 23. Fairway Rok-icefoot profile 2B. Ice ridge Figure 24. Fairway Rock-icefoot profile 3C. Ice ridge slopes I

slopes 1 through 5 are 73, 30, 35, 37 and 38 degrees, respec- through 4 are 70, 35, 28 and 29 degrees respectively and the rock face

tively. Highest point on icefoot is 14.7 m. slope at 5 is 76 degrees. Highest point on icefoot is 14.7 m.

(m) (ft)

70

20

- 50C3.o0

0 10-rA 30o

10Figure 25. Fairway Rock-icefoot profile 4D.

0 0 ice ridge slopes I and 2 are 77 and 30 degrees re-spectively and the rock face slope at 3 is 62 degree.

-3 -10 _ _ _ _ Highest point on icefoot is 9.4 m.

0 Soo MC) 17

Distance

(M) (ftt)40

10-30

C.2 20

LlJ

03 -0 -

0 1000 (ft)I I II

0 100 200 300 (m)Distance

Figure 26. Fairway Rock-icefoot profile SE. lee ridge slope I is 76degrees. Highest point on icefoot is IL.I m.

110

30-

90

7020- 2

C0

-50

0

30-

30

- 10 I I/l

0 1000 MIt

10Distance260So M

Figure 27. Fairway Rock-cefool profile 7G. Ice ridge slope I is 73 de-gree and rock face slope of 2 is 50 degrees. Highest point on Wceool is 9.6 m.

1s

Cm) (f't)

110

30-

90 3

7020

Z 50

W

0-

1 0 3 0 2

10

0- 0 2 - -

- -I0 0lS/Ol-

0 1000 (ft)

0 100 200 300 m)Distance

Figure 28. Fairway Rock-icefoot profile 8H. Ice ridge slope I is 72 de-grees and rock face slopes 2 and 3 are 66 and 64 degrees respectively. Highestpoint on icefoot is 8. 0 m.

and Mellor 1974). If the 14-m-high ridges compos- Aerial views of Fairway Rock taken during ouring the icefoot at Fairway Rock were free-floating April 1981 visit are shown in Figure 29. The northit could be assumed that their keels would extend side of the rock is again seen to be covered with icedownward to the 63-m depth. Since the water and snow, and the shape of the icefoot is consis-depth surrounding Fairway Rock is at least 10 m tent with previous observations. A surface view ofshallower than this, and given the fact that boul- position o.ie in Figure 29 is shown in Figure 30.der talus composing much of the submarine rock The ice floe on which we landed was relatively sta-face slopes gradually outward with depth (Fig. 2), tionary during our visit, but on the west side of theit is only reasonable to conclude that the icefoot is island the ice was moving briskly northward withfirmly grounded. It is assumed that ice which is the current. The icefoot wall on the left in Figurenot securely grounded is rapidly dislodged by pack 30 was measured to be 16.5 m high and was foundice forces generated during major north or south to be near vertical. A view of the ice rubble look-ice movement events. ing west from the top of this high wall is shown in

In April 1981 an investigation of the Fairway Figure 31. The numbers on this photo correspondRock icefoot sail and keel and the submarine slope to those at the locations shown in Figure 29b. Bearof the rock was planned from the U.S. Coast tracks were noted in the snow alongside the islandGuard icebreaker Polar Sea. However, because and the ice rubble. A walk to the west side of thethis ship became disabled in the ice of the Chukchi island ended where the icefoot became about aSea, our field program was reduced to a short in- meter wide and the rock wall dropped "vertically"spection using a helicopter, into the sea, position S in Figure 29.

19

a. North.

. . . .J f,,

b. South.

Figure 29. Fairway Rock icefoot, April 1981.

Following our Fairway Rock stop we visited Lit- Three aerial views of the rubble fields located intie Diomede Island. The south and east sides of the shear zone west of Prince of Wales Shoal arethis island were also found to be surrounded by a shown in Figure 35. These shear ridge fields aremassive icefoot (Fig. 32). On the north side of the not as massive as some which have been observedisland the rock surface up to 10 m above sea level farther north near the Arctic Circle (Kovacs 1972).was found to be coated with up to a 0.3-m-thick Nevertheless, they are large, and because theyicing (Fig. 33). The icing, while decreasing in formed under shear-compression can be assumedthickness with elevation, did extend to the top of to be reasonably compact. These features wouldthe island (Fig. 34). pose a major obstacle to winter navigation. On oc-

20

I -------r

Fiue7.7e! t nsuhatsdeo ara ok

Figure 3. icefoo rubeong south side of Fairway Rock.

21

Figure 32. Icefoot on southeast side of Little Diomede Island. Big DiomedeIsland is in the background.

a. Aerial view. b. Ground view of position indicated by arrow in (a).

Figure 33. Icing on north side of Little Diomede Island.

22

Figure 34. Icing on rock surface at the top of LittleDiomede Island.

casion these formations are dislodged and driven the root mean square (RMS) elevation for eachsouthward. They therefore represent one of the profile by:major ice features to be considered in the design ofa structure to be placed in the northern Bering RMS = + .h an

Sea. The high points on elevation profiles 30-37 1 nshown in Figures 36-43 are 5.7, 5.8, 4.4, 4.1,4.5,5.9, 4.2 and 4.8 m respectively. While the eleva- where h = individual ice elevation and n = num-tion profiles reveal shear ridge ice relief of over 5 ber of data points.m, ridges on the order of 3 m high are more From this analysis the RMS elevations for pro-typical, files 30 through 37 were found to be 2.53, 2.07,

Rubble field profiles were constructed from ele- 2.23, 2.16, 2.12, 1.93, 1.48 and 1.79 m respec-vation data obtained from stereographic photo tively. The average of these RMS values is 2.0 m.analysis where all ridge peak and trough elevations Using a 1:4.5 sail height to keel depth ratio forwere determined along a profile. Among these fea- first-year pressure ridges (Kovacs and Mellortures additional elevation determinations were 1974) we can deduce an RMS ice thickness of I Imade at random to provide representative inter- m. While the analysis is limited by the data, as pre-mediate ice relief. Typically, each profile coin- viously mentioned, and by the fact that some rub-prised between 100 and 180 data points. This ble profiles include short segments of undeformedtranslates into one data point about every 5 in. ice, it does give some indication of the relative iceThus, there is a paucity and uneven distribution of thickness in the rubble fields.elevation data along each profile. Nevertheless, to In early April 1982, we briefly visited Fairwayobtain a general idea of the ice thickness in the Rock by helicopter and made the observationrubble fields, the data were analyzed to determine given in Appendix A.

23

WXI

-01~ SI

1:7- k.

'7~

VI .1 %

-I kx2

IX' N%±' 4

4 '5

3r..

-W-

Q - 'I. 1.1

Figure 3.5. Aerial views of shear ridge fields west Of Prince of Wales Shoal. Numbered lines represent site of ele-vation profiles in Figures 36-43. North is to right.

24

00

0 00 0 8.

-00

0

0 0 -

- 0

UIOU

0 0~0

0 I

z 0

2..

0p4 80

I I

00 0 000

A-- 0 0 v Imo o

-L OS - A *

UOI&DASI*13

- 202

I ICc

00 ~o 100

IN IVI -Y IO 4* C

I 0 U0 4 A 1UOIJDA*I

I j26

0 0

0I

4! h

000

___ ___ __" 0 ('

ESTIMATION OF ICE FORCES 2~1 .. 2 jA ~ (~ON FAIRWAY ROCK 2= 3 + 2D-) "

An estimate of the ice forces against Fairway andRock was made based on limited observations ofthe icefoot and the elevation of ice pile-ups as 4 = 0.441.revealed by stereographic photo surveying. Themagnitude of the ice forces is dependent upon the If we assume an upper bound ice velocity ofmode of ice-island interaction and available forc- 5 x 10- m/s for this case, and sheet ice thicknesseses. Five ice failure modes were considered and an of 0.5, 1.0 and 1.5 m, we can estimate the totalestimate of the available forces is given below. "creep" ice load on the rock from the reference

strain-rate and the corresponding uniaxial com-

1. Creep Deformation pressive strength determined from the laboratoryUnder situations where the ice is moving slowly tests. Tabulated values are given in Table 1. The

against Fairway Rock and all the ice deformation uniaxial compressive strength of the ice at the ref-is by creep, an estimate of the total ice load on the erence strain-rate was obtained from laboratoryrock can be obtained by applying the reference data for sea ice presented by Wang (1979). Icestress concept. This technique gives approximate pressure on the rock for an ice velocity of 5 x 10-1solutions to non-linear structural problems by m/s is about 1.60 MPa (235 psi). Similar results

combining the simpler linear-elastic and rigid- are obtained from the plane stress case where thereplastic problems. Ponter et al. (1981) have success- is no slip between the ice and structure. It is in-fully applied reference stress theory to low velocity teresting to note that the reference strain rate as

ice indentation problems. For a vertical, flat in- determined by theory is about V/0.SD as com-dentor, the reference stress theory gives pared to the less conservative, semi-empirical

values of V/2D and Vi/4D used by Ralston (1979)

rVvil and Michel and Toussaint (1976), respectively.F = Dt Fo -- J (1) The theoretical reference strain-rate is probably

more appropriate at low ice velocities.

where F = total ice forcD = structure width (-400m for Fairway 2. Crsbingl Failure

Rock) The potential ice force due to the crushing of ant= ice thickness ice sheet against the "vertical" faces of the rock or

Vi = ice velocity the icefoot rubble pile may be calculated from theF(xJ = uniaxial stress corresponding to a Korzhavin formula (Croasdale 1980):

uniaxial strain-rate xand 4 = functions of D11 derived from a so- F = JmkucDt (2)

lution to the plastic and correspond-ing elastic edge indentation prob- where F = total ice forcelems, respectively. k = factor to account for the lack of uni-

form ice contact

In plane stress, where D>> I and there is free slip I = indentation factor, 1 - 2.5 for nar-between the ice and structure, row structure, I = I for wide struc-

ture

Table 1. Cakulated creep Ice loads for U ice velocity of $ x 10'-rol md different ice fhck .

Icethickwu VilDI/# Ice uirmth Ice load

Ino) (in.) 9 (r') (fMP) (pa) (MN) (ib)

0.5 19.7 1.155 2.45 x 10" 1.4 203 325 7.31 x 10'1.0 39.4 1.15 2.45 x 10" 1.4 203 650 1.46 x I0'1.5 59.1 1.15 2.45 x 10" 1.4 203 975 2.19x I0'

28

--- -.-- ,-.rp

m = shape factor, m 1 for circular 385 MN (85 x 10' ib), 770 MN (170x 10' Ib) andstructure 1160 MN (255 x 10' lb) for 0.5-, 1.0- and 1.5-m-

ac = uniaxial compressive strength thick ice sheets, respectively.D = structure width (about 400 m in the

case of Fairway Rock) 3. Flexural Failuret = ice thickness. The following formula can be used to calculate

the ice forces on a sloping wide structure (Croas-The values for m and tare taken to be I because dale 1980):

D/t >> 1. The contact factor k accounts for thenon-simultaneous failure of the ice sheet across H /Qgt' 0.25 +the entire structure width. Kry (1978) presented a E = / C1 +ZigC 2 (3)statistical study in which the variation of thestresses with respect to ice displacement is shown where H = horizontal force required to breakto be reduced due to non-simultaneous failure ac- and push the ice sheet uption. Thus, the effective stress for a wide structure b - ice sheet widthis less than the peak stress for a narrow structure. of ice flexural strengthAlthough there are no recommended or suggested t= ice thicknessvalues of k for a wide structure, a low value of 0.4 E = ice modulus of elasticityhas been suggested by Korzhavin for fast-moving ew= water densityfloes (Michel 1970). i= ice density

The selection of a compressive strength for the Z = height reached by ice on the slopeice is a difficult task, as it depends upon so many g = acceleration due to gravityparameters, such as ice temperature, salinity, C 1, C2 - factors which depend on slope anglegrain size, crystallographic structure, strain rate, a and coefficient of friction, it, asetc. In the Arctic, the average temperature of sea obtained from Figures 44 and 45.ice is about -10*C, and we may use the laboratorya, data determined at -10 0 C. The slope of the beach (if there is any at Fair-

Michel and Toussaint (1976) have proposed an way Rock) is not known; however, taking a - 33 0empirical definition of strain rate as err = Vi/4D, (average measured ice pile-up slope angle). it =where V is the velocity of ice, and D the structure 0.3, of = 700 kPa (100 psi), E = 106 kPa (1.45width. This definition of strain rate is slightly x 10' psi), ew = 1030 kg/m (64.3 lb/ft'), Qi =more unconservative than that proposed by Ral- 920 kg/m' (57.43 lb/ft'), Z = 15 m (49.21 ft), weston (1979), eff = Vi/2D, and that of the refer- get the values of H/b shown in Table 2 for dif-ence stress theory. However, using a denominator ferent thicknesses of ice.of 4D gives better agreement between uniaxial As can be seen from the values in the abovecompression and indentation tests at moderate table, the force required to push the ice blocks upstrain rates. The strain rate for Vi = 0.5 m/s and the ice pile-up slope constitutes the major portionD - 400 m is .f = 3.1 x 10-'. Using the corn- of the force per unit width. The ice pile-ups maypressive strength of sea ice at this strain rate as reach 15 m high but are certainly not this highgiven by Wang (1979), the crushing strength of ice everywhere. Therefore, the estimates given areis found to be on the order of 4830 kPa (700 psi). considered to represent an upper value. In addi-Substituting these values into eq 2, we have p = tion, because flexural failure is not expected to oc-F/Dt = Imkoc = 1930 kPa (280 psi). The total cur simultaneously across the entire width of theforce is given by F = pDt, and it is estimated to be rock, the above breaking force values would be in-

Table 2. Values of H/b.

Ice Fore/wdth to push Totel force/widtIkknmss ke upstiope H/b Preaure P

(fMo) (in.) (kN/M) (b/t) (kNIm) (lbft) (kP*) (ir)

0.5 19.7 149 10,204 161 11,057 322 46.81.0 39.4 296 20,4M 327 22,436 327 47.51.5 59.1 447 30,612 591 40,43 394 57.13M 29

0..

0. 0.1oC, 0.1 C"

2 0.12

00o I I a2 * 4 0 6 0 " S e 2 0 " 4 0 " e lf s "

a, Slope Angle a, Slope Angle

Figure 44. Coefficient C, ver- Figure 45. Coefficient C, ver-sus slope angle and friction sus slope angle and friction(after Croasdale 1980). (after Croasdale 1980).

appropriate to apply uniformly across the entire tained by considering the force required to addface of the island or the icefoot. gravitational potential energy to the ice pile and to

overcome ice block friction as discussed by Par-4. Forces Required to Form Floating or merter and Coon (1973) and Kry (1977). In a simi-Grounded Pressure Ridges Along the Rock lar manner, the force required to add gravitationalor to Pile Ice on the Beaches potential energy to a shore ice pile-up (Fig. 46c)

The average force required to form a floating or was discussed by Kovacs and Sodhi (1980). Thea grounded pressure ridge (Fig. 46a, b) can be ob- equations for calculating the force F required to

a. Model for floating pressure ridge. b. Model for grounded pressure ridge.

c. Model for shore ie pile-up.

30

"I I,-.iI II "--t ... .. -" .. .'IIi'/... - . ..I.. .

add gravitational energy to a floating ridge (fl), a a mode which requires the least force. In the pre-grounded ridge (Sr), and a shore ice pile-up (pl) vious three sections, the average ice pressures re-are as follows: quired to fail an ice sheet in crushing, bending,

pile-up and ride-up modes have been estimated.=1 4Q1 Ht Force associated with buckling failure will now beF 2 (4) investigated where an ice sheet is loaded over a

limited contact area.

For = I In the past, investigators have considered theF 2r "" gH5 t buckling of a floating ice sheet using the buckling

2 analysis of a beam on an elastic foundation. The_____ le \ inherent assumption in the use of this analysis re-

S+ 1j I R+. (5) quires that the forces act uniformly on the edge ofthe ice sheet. The resulting high force levels re-

S l igH( quired to cause failure led the investigators to con-(6) lude that the available driving forces were only

sufficient to allow thin ice sheets to fail in thewhere F = force per unit width buckling mode (Assur 1971, Kry 1980). However,

ei = ice density recent buckling analyses of a floating ice sheet byg = acceleration due to gravity Sodhi and Hamza (1977), Wang (1978a, b) and

H, = ridge sail height or height of ice pile Sodhi (1979) allow for the consideration of thet = ice thickness buckling failure of a large ice sheet when it is in

ew = water density contact with a wide structure or a beach over aD = ridge keel grounded depth limited area.

a = a factor whose value ranges from I To investigate all possible situations for buckl-(for a pile-up on level ground) to 2 ing failure, an idealized wedge-shaped ice sheet is(for an ice ride-up), considered, as depicted in Figure 47. The ice sheet

is shown to be loaded by a total force F actingIt can be shown that F r < Fn because D /H < along the curved leading edge, such that it creates

(Qi/Qw - Qi) (Kry 1977). 7he maximum force perunit width is therefore due to ice pile-up or ride-

up.The frictional forces (Ffr) per unit width of anice sheet required to push through a rubble pile(Fig. 46c) may be estimated by the following ex-pression (Kovacs and Sodhi 1980, Kry 1980):

Ffr = i(! -v) QigfH2 Cote (7)

where Pi = coefficient of friction of ice on icey = porosity of the sail massf = fraction of the sail mass that acts to

create the frictional force, H = height of sail or pile-up

9 = slope angle of sail or pile-up fromthe horizontal.

The combined forces to add gravitationalenergy and to overcome frictional forces in an icerubble are of the same order of magnitude as thatrequired for flexural failure of an ice sheet againsta sloping surface.

Figure 47. Geometry of a wedge-shaped floatingS. Duckling all Ice sheet. Note that the arrows show the assm distlbu-When an fi e sheet moves against a wide struc- don along the area of contact change with the cosine of

ture or a beach, it is generally considered to fail in 0 and thes sum up to . total force F.

31

MI

*UI Ii I I I11111 III l

K =Modulus of Foundation(specific weight of water)L= Characteristic length of ice sheet

9001500

1800a=1.8*

F/BKLI 300

a.1 1 t.R/L

Figure 48. Non-dimensional buckling load for different ice sheetwedge angles a , R/L values and buckling conditions (from Sodhi1979).

a radial (compressive,-) stress field Nrr = a = 180° (i.e. semi-infinite ice sheet). During the- [2F/(cx + sina)I~cosO/r] N/rn in the ice sheet, course of recent experiments (Sodhi et al. 1982) itThe symbols used in this equation are graphically was observed that the non-dimensional bucklingexplained in Figure 47. The non-dimensional loads from theory (Sodhi 1979) and experimentsbuckling loads for wedge-shaped floating ice are in close agreement. it was also observed thatsheets have been given by Sodhi (1979), and these the boundary condition at the ice/structure inter-are presented graphically in Figure 48 for different face corresponds to a situation between the fric-wedge angles, RIL values and boundary condi- tionless and hinged boundary conditions. Unlesstions. We may write the non-dimensional buckling ice freezes to the structure, the fixed boundaryload as follows: condition is not likely to occur. Hence the buck-

ling loads corresponding to the hinged and theF J(R c ndiin frictionless boundary conditions are the upper and

BKL2 = 1k7L, a, boundary onio/(8) lower bounds. The expression for the pressure atthe contact area required to buckle an ice sheet is:

where K = foundation modulus (equal to thespecific weight of water in the case Pb = (F/Bh)of floating ice)

L - IEh3 /[12(i - 2)K~lt, characteristic = KL(L/h)f(R/L, a, boundary con 'ion) (9)length of ice sheet

a= Poisson's ratio Assuming values of E = 10' Pa (0.145 x 10' psi),E = ice sheet modulus of elasticity K ,= ewg =" 10.1 kN/m' (64.3 Ib/ft') ad =h = ice sheet thickness 0.33, we calculated the buckling pressures shownB = 2R sin(u/2) (width of contact area). in Table 3 for different values of RIL, and ice

thicknesses for the frictionless and hinged boun-Since the non-dimensiona buckling load is not dary conditions. From these buckling pressure va-

influenced strongly by the wedge angle a (Fig. 48), ues it can be seen that, for a particular ice thick-our discussion is restricted to a particular value of ness, if the area of contact is small, a large prm-

32

- R//

Figure-- ..... No- -mnioa bukln .. .. ... fo ifeetiche

Table 3. Buckling pressure for different R/L and boun-dary conditions.

RIL

0.! 1.0 /0h L Buckling pressure

(M) (inch) (in) (f) (MPa) (psI) (MPa) (psi ) (MPa) (psi)

Frictionless Boundary Condition

0.3 (11.8) 3.98 (13) 9.23 (1339) 1.36 (198) 0.51 (74)0.5 (19.7) 5.83 (19) 11.92 (1729) 1.76 (255) 0.66 (96)1.0 (39.4) 9.81 (32) 16.86 (2445) 2.49 (361) 0.93 (135)1.5 (59.1) 13.30 (44) 20.65 (2995) 3.05 (442) 1.14 (166)

Hinged Boundary Condition

0.3 (11.8) 3.98 (13) 27.54 (3995) 3.26 (473) 1.18 (171)0.5 (19.7) 5.83 (19) 35.55 (5157) 4.21 (610) 1.52 (221)1.0 (39.4) 9.81 (32) 50.28 (7293) 5.95 (863) 2.15 (313)1.5 (59.1) 13.30 (44) 6.158 (8932) 7.28 (1057) 2.64 (383)

sure is required to initiate ice sheet buckling, off = (2R/L) sin(ct/ 2 )Pb/X (10)whereas the buckling pressure is relatively lowwhen the area of contact is large. where Pb is given by eq 7.

However, if the contact area is small, high stress To illustrate the usefulness of the above equa-can be created along the boundary by a relatively tion, we consider a semi-infinite ice sheet (i.e. a =low far-field stress in the ice sheet. This is due to 1800) and we take R/L = 1.0 because smaller con-the fact that the radial stress decreases as the in- tact areas would likely cause crushing and largeverse of the radial distance from the wedge apex. contact areas are less likely to occur. Based onIn fact it is possible that the ice sheet will fail in the these assumptions the far-field ice pressures need-crushing mode rather than in the buckling mode if ed to cause buckling of a semi-infinite ice sheet arethe effective crushing pressure is less than the listed in Table 4.buckling pressure for small contact area. This was In a short version of this report presented atshown to be the case where ice was shown failing POAC-81 (Kovacs and Sodhi 1981), P.R. Kryagainst a man-made gravel island in a movie pre- brought up the point that the spacing X of the loca-sented by D. McGonigal of Gulf Canada Re- tions where buckling occurs is given as an inde-sources, Inc. at the October 1980 National Re- pendent parameter in Table 4:search Council Workshop on Sea Ice Ridging held "The mean stress presented in the table thus de-in Calgary, Alberta. As the ice is crushed, the con- creases significantly as X increases because it as-tact area becomes enlarged and the effective mod- sumes there is more area over which the drivingulus of elasticity of the ice sheet becomes lower stress is applied OL); whereas the failure load is adue to cracking activity. Both of these effects constant being restricted to single areas separatedlower the pressure required to buckle the ice sheet. by increasing distance."This is questionable on the prounds that theIf the ice sheet buckles, it produces jagged ice spaig is inegenal tegnd th fleedges which in turn promote more crushing andspcn.iingerldtrmedbthfaueprocess. For buckling failures the cusps, givingbuckling at other locations as the ice sheet moves rise to individual subsequent failures, will have atowards a wide object, be it a structure, beach or spacing of only a few action radii (X< < 10). Xrubble pile. Again, this was apparent in the values higher than this may not be physicallyMcGonigal film mentioned above. The phenome- reasonable."non of a "complex combination of continuous We agree that the spacing of contact areas is de-crushing and buckling" was also reported for the termined by the size of cusps or the area of the iceice failing around Netserk F40, an artificial sheet affected by buckling failure. The spacing"gravel" island (Gladwell 1977). parameter was arbitrarily assumed to demonstrate

If the contact areas are regularly spaced at a dis- that low environmental forces can cause bucklingtance of ) times the characteristic length, then the failure in an ice sheet. It should also be pointedfar-field ice pressure (off) required to buckle the out that the above situation refers to the case whenice sheet at all the contact areas simultaneously is: all buckling failures occur simultaneously. If non-

33

Table 4. Far-field ice pressure required to buckle a semi-infialte icesheet when the contact areas are spaced X times the characteristiclength. Each contact area Is twice as wide as the characteristiclength.

5 10 20 50A L Far-field ice pressure

(m) (in.) (m) ft) (MPa) (psi) (MPa) (psi) (MPa) (psi) (MPa) (psi)

Fkcdionlem Boundary Condition

0.3 (11.8) 3.98 (13) 0.55 (79) 0.27 (40) 0.14 (20) 0.05 (8)0.5 (19.7) 5.83 (19) 0.70 (102) 0.35 (51) 0.18 (26) 0.07 (10)1.0 (39.4) 9.81 (32) 1.00 (144) 0.50 (72) 0.25 (36) 0.10 (14)1.5 (59.1) 13.30 (44) 1.22 (177) 0.61 (88) 0.30 (44) 0.12 (18)

Hinged Boundary Conditiom

0.3 (11.8) 3.98 (13) 1.30 (189) 0.65 (95) 0.33 (47) 0.13 (19)0.5 (19.7) 5.83 (19) 1.68 (244) 0.84 (122) 0.42 (61) 0.17 (24)1.0 (39.4) 9.81 (32) 2.38 (345) 1.19 (173) 0.59 (86) 0.24 (35)1.5 (59.1) 13.30 (44) 2.91 (423) 1.46 (211) 0.73 (106) 0.29 (42)

simultaneous failure zones are assumed to occur, DRIVING FORCESthe values given in Table 4 may not be so unrea-sonable. In addition to considering the forces associated

From the above reasoning it is evident that the with the different types of ice failure modesice sheet may fail in either crushing or buckling, if against the rock, it is appropriate to also considerthe ice contact areas are either very small or are the environmental driving forces. Ice forces on thespaced at distances in excess of ten times the char- rock will be governed by the ice thickness, failureacteristic length. mode, and available driving forces.

Dim d sands

Figure 49. Ice floes in the Bering Strait. For reference Big Diomede Is-land is about 10 km long.

34

I AL

The dominant factors affecting ice motion andforces in the Bering Strait are wind and water dragon the ice cover. Internal ice stresses are small be-cause of the open nature of the pack. Consider alarge (10-km-square) floe impinged against thenorth side of Fairway Rock acted upon by south-ward 30-m/s winds and 1.5-m/s currents. Usingthe drag coefficients given by Pritchard et al.(1979) for ice in the Bering Strait, we can calculate X ,I

the ice driving forces on the rock from F[rF = (aCa P. + ewCw Vw)L (11)

where Qa = air densityCa = ice-air drag coefficientVa = air velocityQ, = water densityc w = ice-water drag coefficientV = water velocity A

L = floe size. _t

For QaCa = 1.1 x l0- kg/m and wcw 8 kg/m-,and the assumed winds and currents, a drivingforce of 1900 MN (4.27xi0' lb) is obtained. If . .both the winds and currents were reduced by a fac-tor of 2, we would still obtain a driving force of Figure 50. Ice wedge developed on the south side475 MN (l.07x 10' Ib). As 10-km-square floes are of the Diomede Islands, 2 April 1975. The angle ofnot uncommon (Fig. 49), these calculations sug- internalfricion, 0, is equal to 90°- .gest that all five of the ice failure modes consid-ered are feasible in the vicinity of Fairway Rock.

ANGLE OF INTERNAL FRICTIONOF SEA ICE

As the ice moves past an island in deep water itfails, forming a wedge-shaped rubble formation.If we assume that the ice floes represent a materialwhich can be characterized by a Mohr-Coulombtype failure criterion, we can estimate its angle ofinternal friction. The angle between the wedge side(shear boundary) and the wedge base representsthe angle between the intermediate plane and theplane of major principal stress. Knowing thisangle, the angle of internal friction # at the shearboundary can be determined from 9 = 45 * -/2.However, determination of 0 is not always easybecause both island geometry and differential ero-sion of the ice rubble can distort the wedge shape, A

making it difficult to locate the position of thewedge base. This can be resolved by assuming thatthe base apex angles of the wedge are the same. Itthen follows that 0 is equal to (180 - a)/2, wherea is the angle formed at the intersection of thewedge sides (Fig. 50). Since a is equal to 90 0 -0, it Figure 51. Ice wedge developed on the south sideis then possible to determine 0 by knowing a. This of the Diomede Islands, !! April 1975.

35

SUMMARYmethodology was used by Kovacs (1976) in a studyof snow displaced along piles driven into snow and The information obtained in this study revealedSodhi (1977) in the analysis of a wedge-shaped ice that Fairway Rock was surrounded by a massiveformation north of St. Lawrence Island. icefoot in the winter of 1980. This icefoot was the

The apparent angle a was measured for five result of ice impacting the island, failing, and sub-wedge-shaped formations at Fairway Rock, five sequently piling up, forming ridges up to 15 mformations that formed on either the north or the high. The icefoot varied from less than 10 to oversouth side of the Diomede Islands (Fig. 51), and 100 m wide. The slope of the inner ridges averagedtwo formations that formed on the north side of 33 0 while the slope of the outer face of the icefootSt. Lawrence Island. These angles varied from 500 was in excess of 700. This was apparenty the re-to 700 and averaged 60*. From this average the sult of nongrounded ice rubble having slumped orangle of internal friction was determined to be been cleaved off. The instructive findings are, as300. This angle may be compared to the results of anticipated, that ice rubble formation around aother investigations: model shear box values large structure placed in "deep" water will not ex-found by Keinonen and Nyman (1978) of 470, tend appreciably beyond the width of the struc-Prodanovic (1979) of 47 0 and 53 0, and Weiss et al. ture, and therefore will not add significantly to its(1981) of II 0 to 34 with an average of 22°; the effective diameter. In order for this to be so, theaverage angle of repose of ice blocks in first-year submarine slope needs to be relatively steep. Atpressure ridge sails found by Kovacs (1972) of 240, Fairway Rock, it is reasonable to assume that theGladwell (1976) of 340, and Tucker and Govoni shallowest submarine slope was at or near the(1981) of 260; the average angle of repose of ice angle of repose of the rock talus.blocks in shore ice pile-ups found by Kovacs et al. Fairway Rock and Little Diomede Island were(1975) of 320 and Kovacs (1982) which varied observed encrusted with a thick icing layer that ex-from 30 to 45 0 and averaged 37 0; the average val- tended high above sea level. For offshore struc-ue of 33 0 found in this study for the ice blocks in tures placed in these waters, this phenomenonthe ridges at Fairway Rock; the average angle of needs to be considered from a loading as well as anrepose of ice blocks in first-year pressure ridge operational hindrance standpoint. For detailed in-keels found by Kovacs (1972) of 33 0; and the val- formation on the subject of icing formation inues obtained from triaxial tests on first-year sea general, and ocean structure icing in particular,ice by Panov and Fokeev (1977), which varied see Minsk (1977, 1980) and Stallabrass (1980).from 140 to 55 *. Shear ridge rubble fields observed in this study

Differences in test conditions, ice stress state, near Prince of Wales Shoal were found to reachand ice properties may be invoked to explain much heights on the order of 6 m, with an apparentof the large variation in the above 0 values. Ice average of 2 m. The average ice thickness was esti-stress state and ice properties were clearly shown mated to be I I m. These rubble fields are, on oc-by Panov and Fokeev (1977) to affect 0. However, casion, driven south, and therefore represent onethe lower slope angles of Kovacs (1972) and Tuck- of the major ice features to be considered in theer and Govoni (1981) are the result of these in- design of a structure placed in these waters.vestigators' desire to characterize the relative The ice force levels given here are only estimatesshape of the ridge sail. They were not specifically based upon assumed ice data and geometry con-interested in the angle of repose of the ice blocks. siderations. Larger forces are possible at localHigher values, which average 290, result when the contact points where stress is concentrated. Theslope of the ice blocks is measured near the peak calculated effective pressures due to creep, crush-of the ridges, as studied by Kovacs (1972). In this ing and flexural failure of ice 1.5 m thick againstway the effect of nonuniform ice piling, which can Fairway Rock were estimated to be 1630 kPa (236cause highly irregular ridge slopes, is minimized, psi), 1930 kPa (280 psi), and 394 kPa (57 psi),For purposes of determining the angle of ice block respectively. An analysis of the available drivingfriction, the latter measurement is preferred, pro- forces suggests that all three failure modes arevided a "sufficient" number of ice blocks exist in possible. However, out of the three possiblethe ridge face to establish a representative angle of modes of failure, the ice cover would most likelyrepose. fail in flexure, as opposed to crushing, since it re-

From the above determinations and observa- quires less force.tions, a representative value for the angle of inter- We have also presented expressions to show thatnal friction of sea ice would appear to be between buckling is also entirely possible at areas of local30 and 35 °. sheet ice contact, even when the far-field ice prm.

36

sure is relatively low. Indeed, we envision that all Kovam, A. and D.S. SoMi (1980) Shore ice pile-up and ride-three failure modes will occur randomly during ice up: Field observations, models, theoretical analyses. Ca& Re-

faih~re against a large structure or during ice pack ions Sciece and Tehnoo', vol. 2.

deformation. Kean, A. aid D.S. SnM (1961) Sea ice piling at FairwayRock, Bering Strait, Alaska: Observations and theoretical

Finally, examination of ice wedges against vari- analyses. Proceedings, Sixth International Conference on Port

ous islands in the Bering Sea suggests that the and Ocean Engineering Under Arctic Condition, University of

angle of internal friction of sea ice is about 30°. Laval, Quebec, Canada, vol. i and Ill.Key, P.R. (1977) Ice rubble fields in the vicinity of artificial is-lands. Proceedings, Fourth International Conference on Portand Ocean Engineering Under Arctic Conditions, Memorial

LITERATURE CITED University of Newfoundland, St. John's, Newfoundland, Can-

ada.

Abinas, K. nd G. Weudler (1979) Sea-ice observations by sat- Ky, P.R. (1978) A statistical prediction of effective ice crush.

ellite in the Bering, Chukchi and Beaufort Seas. In Proceed- ing stress on wide structures. Proceedings, InternationalAsso-ings, Fifth International Conference on Port and Ocean Engi- ciation of Hydraulic Research Symposium on Ice Problems,

neering Under Arctic Conditions. Norwegian Institute of Tech- Lulea, Sweden.nology, Trondheim, Norway. Key, P.R. (1980) Ice forces on wide structures. Canadian Geo-Amne, A. (1971) Forces in moving ice fields. In First Interna- technical Journal, vol. 17.

tional Conference on Port and Ocean Engineering Under Arc- Michul, B. (1970) Ice pressure on engineering structures.

tic Conditions, Trondheim, Norway, vol. I. CRREL Monograph Ill-BIb. AD-709625.Eloem, G.L. ad D.E. McDongal (1967) Bering Strait unat- Michel, B. and N. Toumait (1976) Mechanism and theory oftended oceanographic telemetry system utilizing a commercial indentation of ice plates. Symposium on Applied Glaciology,

LCC-25 strontium 90 generator. The New Thrust Seaward. Cambridge, England. Journal of Glaciology. vol. 19, no. 81.

Transactions of the Third Annual Marine Technological Socie- Mia, L.D. (1977) Ice accumulation on ocean structures.

ty Conference and Exhibit, 5-7 June, San Diego, California. CRREL Report 77-17. ADA-044253.Washington, D.C.: Marine Technological Society. Mink, L.D. (1980) Icing on structures. CRREL Report 80-31.Coachmn, L.K. and K. Aganed (1966) On the water ex- ADA-095474.

change through Bering Strait. Limnological Oceanography, More, D.G. (1964) Acoustic-reflection reconnaissance of con-

vol. I, no. 1. tinental shelves: Eastern Bering and Chukchi Seas. Papers inCochmmm, L.K. mul K. Aagand (1979) Re-evaluation of Marine Geology, Shepard Commemorative Volume (R.L. Mil-water transports in the vicinity of Bering Strait. Preprint of ler, Ed.). New York: Macmillan.paper to appear in The Bering Sea Shelf: Oceanography and Mueuel, R.D. md K. Ahlas (1976) Ice movement and distri-Resources (W.D. Hood and J.A. Calder, Eds.). Juneau, bution in the Bering Sea from March to June 1974. Journal of

Alaska: U.S. Dept. of Commerce, NOAA Alaska Office of Geophysical Research, vol. 81.Marine Pollution Assessment. Pamov, V.V. aad N.Y. Fokeev (1977) Compression strength ofCromsdale, K.R. (1980) Ice forces on fixed, rigid structures. In sea ice specimens under complex loading. Probenmy Arktiki iWorking Group on Ice Forces on Structures: A State-of-the- Antarktiki, vol. 49.

Art Report (T. Carstens, Ed.). U.S. Army Cold Regions Peeler, A.R.S., A.C. Palme, DJ. Degman, M.F. Ahby,Research and Engineering Laboratory, CRREL Special Report A.G. Evan and J.W. Hutchlisn (1981) The force exerted by

80-26. ADA-09674. a moving ice sheet on an offshore structure. I: The creep mode.Gladwel, R.W. (1976) Field studies of eight first-year sea-ice Unpublished manuscript, Center for Cold Ocean Resources,pressure ridges in the southern Beaufort Sea. Imperial Oil Lim- Memorial University of Newfoundland, St. Johns, Newfound-

ited, Production Research Division Report IPRT-SME-76, land.APOA Project No. 75, Calgary, Alberta, Canada. Parmurter, R.R. and T.D. Coon (1973) Model of pressure

Gldwefl, R.W. (1977) Ice conditions around artificial islands ridge formation in sea ice. Journal of Geophysical Research,1975 to 1976. ESSO Resources Canada Ltd.. Production Re- vol. 77. no. 33.search Division, Calgary. Alberta, Canada, Arctic Petroleum PrItcmrd, R.S., R.W. Remw sai M.D. Ca. (1979) Ice flowOperators Association Report No. 105-3. through straits. Proceedings, Fifth International Conference

Keloemmm, A. aid T. Nyma (1973) An experimental model- on Port and Ocean Engineering Under Arctic Conditions, Nor-

scale study of the compressibility, friction and cohesive be- wegian Institute of Technology, Trondheim, Norway.

havior of broken ice mass. IAHR Symposium on Ice Problems, Pradnemle, A. (1979) Model teats of ice rubble strength. Pro-

Lulea. Sweden. ceedings, Fifth Interntional Conferenc on Port and OceanKovaes, A. (1972) On pressured i ice. In Proceedings, Inter- Engineering Under Arctic Conditions, Norwegian Institute ofnational Conference on Sea Ice (T. Karason, Ed.). Reykjavik, Technology, Trondheim, Norway.

Iceland: National Research Council. Ralston, T.D. (1979) Sea lee loads. Presented at Technical

lemma, A. (1976) Study of piles installed in polar snow. Seminar on Alaskan Beaufort Sea Gravel Island Design, ExxonCRREL Report 76-23. ADAI029191. Company, Houston, Texas.Keyna, A. (1962) Recent shore ice ride-up and pile-up obser- Rimer, R.W. (1979) Ice breakout in the Bering Strait. Master

vations. Part 1: Beaufort Sea, Alaska. (In press.) of Engineering thesis, University of Washington, Seatt,

Kovacs, A. and M. Mellor (1974) Sea Ic morphology An Ice Washington.as a geologic agent in the southern Beaufort Sea. In The Cony UsbuM& L.L. (1976) Alaska regosial profiles, Yukon region.and Sh of the Bafort Sm (J.C. Reed and J.E. Sater, Eds.). Vol. VI. Arctic Environmental Information and Data Center,

Arlington, Va.: Arctic Institute of North America. University of Alaska.ira, A., D. Dikea amd 5. Wig1t (1975) Multi-ya pres- Shol, L.a =JJ. I. (1973) Satlite obsrvalom of ma

sure ridge and shore ice pile-up. Arctic Petroleum Operators ice movement in the Bering Strait reiom. In Clmast of the

Association Project 9. Arctic (0. Waller and S.A. Bowling, Id.). Proeeudings of the37L "

AAAS-AMA Symposium. Fairbanks, Alaka, August 1973. Sfthgsr WJ. (1976) Morphology Of Beaufort, Chukchi andGeophysical Institute, University of Alaska. Bering Sea nearshore ice conditions by means of satellite andlibuaway, G.. D.G. More ail G.B. Dowling (1964) Fairway aerial remote sensing. Geophysical Institute Report. UniversityRock in Bering Strait. Papers in Marine Geology, Shepard of Alaska, Fairbanks, Alaska.Commemorative Volume (R.L. Miller, Ed.). New York: Mac- Tucker, W., I mnd J.W. Goed (1901) Morphological in-millan. vestigations of first-year sea-ice pressure rige sails. Cold Re-Sedhl, D.S. (1977) Ice arching and the drift of pack ice through gions Science and Technology. vol. S, no. 1.restricted channels. CRREL Report 77-l8. ADA.044216. Wong, Y.S. (1978a) Buckling analysis of a semi-infinite iceSodhl, D.S. (1979) Buckling analysis of a wedge-shaped float- sheet moving against cylindrical structures. Proceedings, infer-ing ice sheet. Fifth International Conerence on Part and natilanal Association of Hydraulic Research Synvposiunr on IcetOcean Engineering Under Arctic Conditions, Norwegian Insti- Problents, Lulea, Sweden.tute of Technology, Trondheim, Norway. Wang. V.S. (1978b) Buckling of a halk ice sheet against abodhi, D.S. and H.E. Rama (1977) Buckling analysis of a cylinder. Proceedings, ASCE, Journal of Engineering Mechmn-semi-infinite ice sheet. Fourth International Conference on ics Division. EMS.Port and Ocean Engineering Under Arctic Conditions, Mem- Wong, T.S. (1979) Sea ice properties. Presented at Technicalorial University ofNewfoundkan, St. John's, Newfoundland. Seminar on Alaskan Beaufort Sea Gravel Island Design, ExxonSedhi, O.S., F.D. Rayona, K. Kate and K. Hkraysoa (1962) Company, Houston, Texas.Experimental determination of the buckling loads of floating Wele, N.Y., A. Prodanovic andl K.N. Wend (1961) Determin-ice sheets. 2nd International Symposium on Applied ation of ice rubble shear properties. JAHI International Sym-Glaciology, CRREL, Hanover, N.H., 23-27 August 1982. posium on Ice, Quebec City, Quebec, Canada.Stoabess, 1.R. (1980) Trawler icing: A compilation of workdone at N.R.C. National Research Council of Canada ReportNRC-19372.

38

APPENDIX A: APRIL 1962 FIELD OBSERVATIONS AT FAIRWAY ROCK

During the first week of April 1982, a short visit A. An ice floe would contact the icefoot and be-to Fairway Rock was made. There was a strong gin to crush or fail in flexure along the contactbreeze coming out of the south. This driving areas. This would be followed by splitting of theforce, coupled with that of the northward current floe into two or more fragments, followed by re-flow, was pushing 1-m-thick sea ice past the rock newed crushing or flexural failure at the icefootat about 1 km/hr. The icefoot was again noted to contact areas.be composed of densely packed sea ice rubble up B. A floe would contact the icefoot, stop, andto 15 m high, and the rock walls were observed in undergo crushing or flexural failure upstream ofplaces to be covered with a thick glazing of ice to the contact area. During this process, ice ridgingan elevation of over 30 m (Fig. Al). would occur. These ridges were typically less than

Ice floes were seen being driven against the ice- 2 m high.foot (Fig. A2), and were undergoing failure in the C. A floe would contact the icefoot, undergofollowing ways: crushing or flexural failure at the impact areas,

_VW

Figure A 1. View of Fairway Rock icfoot and glaze ce formtion. Top photoshows southmst side of rock; bottom photo shows northet side.

39

and then split. A portion of the floe would then be half the width of the icefoot. During crushing, thedriven past the rock, where it would undergo con- resulting ice blocks were typically less than halftinuous crushing and flexural failure as it moved the thickness of the ice in the floe being crushed.along the icefoot shear boundary. During this pro- Standing near the failure zone, we noted that thecess much of the broken ice was reduced to rubble ice sheet did not appear to be moving up or downwith dimensions less than the thickness of the ice during crushing.floes. During our visit the icefoot was not "growing."

D. An ice floe would contact the icefoot and un- The ice which had failed and piled up against thedergo complete disaggregation by crushing and icefoot would from time to time appear to cave inflexural failure. or slump downward. The ice keel of the rubble

We noted that crushing of ice floes at the base was apparently undermined by the strong current,of the icefoot was a common mode of failure and which transported the submerged ice blocks away.during our observations appeared to occur over

4-a IC

Figure A2. Ice floes against the southern Fairway Rock icefoot. The discoloredke blocks are contaminated with biological organisms. Thew blocks resulted fromcrushing of the par"t fio.

40

* -yr- --- A

A facsimile catalog card in Library of Congress MARCformat is reproduced below.

Kovacs, A.Bering Strait sea ice and the Fairway Rock icefoot,

by A. Kovacs, D.S. Sodhi and G.F.N. Cox. Hanover,N.H.: U.S. Cold Regions Research and Engineering Lab-oratory; Springfield, Va.: available from NationalTechnical Information Service, 1982.

iv, 44 p., illus.; 28 cm. ( CRREL Report 82-31. )Bibliography: p. 37.1. Ice. 2. Sea ice. 3. Ice forces. I. Sodhi,

D.S. II. Cox, G.F.N. III. United States. Army.Corps of Engineers. IV. Army-Cold Regions Researchand Engineering Laboratory, Hanover, N.H. V. Series:CRREL Report 82-31.

t.

*U.S. GOVERNMENT PRINTING OFFICE: 1982 - A-2294/251

• ! '