HomeMy WebLinkAboutAPA2689c
0 n c., g) 3J:CD' m~ . zco f)
-t~
n2'
03
Z-t -to
:::0
0 r
FIELD INVESTIGATIONS OF A HANGING ICE DAM
S. Beltaos, Research Scientist
(Formerly -Research Officer
National Water Research lnstieute
AJberta Rnearch Council
A.M. Dean, Jr •. Electrlcal Enslneer U.S. Army Cold. Resions
Research and Ensineerlns Laboratory
ABSTRACT
u.s.A.
A hansln& Ice dam that forms annually in the lower Smoky Riwer, Alberta, ha~ been the
object of contlf'lJed lnvest!satlon mrlrc the period ""-1979. The study almt at documentlrc
physical dimensions and material properties of the dama elucldatlnc the rnechanlsmi of Its
formation and removal; and assessi"A Its effects on the prosress of breakup In the rl~r. Thjs
paper pre~nts a summary of the results obtained to date.
INTRODUCTION
A hanslns lee dam Is a downward projection of river Ice, produced by deposition of frazlt
slush ooder an exlstlf1B Ice cover { I''· Typic•lly, • han~il18 dam forms at • low lpftd tee tim of
a stream, located lmmedlatl!ly downstream of a hl•h spftd teetlon. Durin~ erwn .. , the lltMr
remains open while ae;t Ice cover forms at the former section. Frazil lm praduc:ed In the rliplctr
flow Jeetlon aglomerates Into slush n pens tNt are transported UJ1Cieo:r ·the CO'flet' \\f .._,
tranquil section and deposit where the flow speed is sufficiently low. Depositi"J'1 contlflues until
either the lJ!)Stream supply Is discontinued or tM-flow velocity under the cc.umulatlon lncreaMSI
to a value capable of transportln& the entire amount of incomlns Ice. Thl! llmltl,. velocity•
varies between I m/s lnd '·' m/s dependh'l on the CIOrnpo!lltton Mel dimensions of the·
trransported material ( ••• 16).
Hansins dams are often mentioned In ke tenslneerlnt Uteratur:e ( J, '9 111, I,, 171 but ther~
exists little documentation of their behaviour und effects. Gold and WUif,.amt I Ill described a
90 m deep and 1200 m lon& han&lns dam In the Ottawa River. Such musi~ accumulations or ice
can obstruct the sprlns Ice r1.11 and ln!!!~ie ma~ lee jams u well as beina capable or lnterfer·ln~
with river structures.
The ~~~biiity of a han&lns dam accutrrlng 'In the Smoley R:lwer, about -o km above ittl
cenfiuence with Peace River (see Fisure I) wu first detected In 19n by Brlltlsh Columbia HJdrCil
staff; the existence of the dam was confirmed by ~i"IS in earfy 197 '· Because of pMsible
effects of this hansins darn on breakup near tM town of Peace River (Fisure 1). a Jonu-~erm
lnvestl&atlon was initiated by Alberta Retearch Council. The main objectivtos of this study1are to
document the formation of the darn, assess its effect en the hrealcUp proceu In the Smalley and
Peace Rivers and examine whether Impact forces by movlns frasments of such dams ne:ed be
considered in the desi1n of river structures. This paper presents a summary of thi~ t,•esults
obtained durins the period 197,-79.
e -!~I
Locatloo map.
F1 1 • 2 Oblique alr photo of hahsins dam site flc~a~rins
upstream, Dec. 2, 19"' note hummocked 1ice
surface and open water lead In ~rapids
upstream).
SMOKY RIW!R HANGING DAM
The stream c:onltl&Uratlon In the vicinity of the hans!ns dam site consists of a deep and1wide
section preceded by a section of rapids upstream; this sequence exhibits features conduci,,e to
hanslf11 dam formation .,d, to a de&ree, Is Illustrated In Flsure 2. Also shown In Fisure 2 is the
hummocked md perceptibly elevated surface of the hanll"l dam. A lonKitudlnal profile ol the
dam, obtained In March 197,, Is shown in Flsure 3. The frazil accumulation is ro~hly trian1~ular
with a base of JOO m Wid a maximum depth of iJ m below the water suface. These dimen:slons
vary from year to year. In January 1976, the lensth and maximum depth of the dam were JI)O m
1nd 16.) m .ttlle correspondln& values for February 1977 were 700 m and 11.0 m. River t:ross
teetlans, located u lndlcat~ In Flsure J, are shown In Flsure 4 where the deepenlns and
wldenlrc of the ~lver near section 0-0 (deepest section) are well illustrated.
Open-water flow conditions at the dam site wer'! documented In July 19,, It was found
that, due to d\annel exp~nslon, two larse eddies were present ~·ar the river b5nks, as sketche~d in
Fisore '; there was no evidence of the channel bed depression havlns bl!f!n filled in by sedimt"nt
476
• . !it II.,
II'\
p lllr-N
~ ·;, --1 :::E ..... .. --E .. .a c ··-!' a;-.. !" ··-! ....... ·;,
!,!:
'[SF-i 1 ..
~~ Ctlr-i --,., 0 --I ... _
=' .!! _,.., ... _
~ ;;:: o-I ••-. K !! '6 'ill ,I -iiE -II'\ ··-:! ....
\1 ~~ ·-CP\ 'B-: ~ I
" .. ..,
\ 'iON ··-~1 i II'\ . .., ·~ 0 I
I u: --\ ., • ·-
"" r··· ..,
tUtJ H0111111rn .QMftiiiW ii:
477
The Slime .,., rei'C"ted for tM Ottawa River dam ( II J. A vertical velocity profile, tal~en In the
live stream at J!C!Ction 0-0, irdicated an averase velocity of 1.2 m/s which Is comparable to
values me•sur~ In the rapid! section '-"stream of the dam site. This explains why the1river bed
deprenlon is not filled with ~lment In IM summ~: as flow velocities and, hence, :transport
rates are comparable upstream of and at the dam ~ te, depo!li tlon Is not lllcely to occur.
FORMA noN
Formation of the hanslnz dam was document~d in November JCJ7~ by means of 1an I mm
movie camera, programmed to expose one frame per minutf! &Jr-ing dayllgh" hours and ~~nstalled:
near the top or the west valley wall. The resultln; 1~im shows rr~ze up events for a 1:1eriod of'
six dayts and provides a fair description o! th~ dam formation mechanism. At thE dam 1~1te, the
ice co;ver Is Initiated at the edcf¥ :Sreas where frull floes and pans recirculate and e~entually
becotr1e shore last. Cr.e:!u&lly, tM firm Ice cover extends outward from t~ banks to~v11rd the
midstrearn and ~mewhat upstream. As this occurs, the eddies .abo move upstream which 1ehables
~tl'nl!t~ build up of the cover. Eventually, ooly a narrow strip of open water l!s left at
r;-.id!itream, correspondlns roughly to the Jive stream under open-water conditions Csee Fllgure ,,,
Thit; strip Is finally bridged by an arching mechanism similar to that studied by Calfl!Jns and
Ash/ton ( 6 J. For the 1971 freeze up, surface flux of lee pans be san durlns the nlzht of Nc!Vemb~r
I flo 19 and a ~mplete ice bridge across the river formed by the morning of Novt~:m1ber 14
Te·mperature r~cords at Watino r II ir1dicate that November ·I was the first day of sust11T.n1!d frost
w&!lle the average air temperattm! during the formation period was about -IJ,,°C.
It was mentioned earlier that the limiting frull slush deposition vdoclty Is In thl!• range 1.0
to 'I,, m/s ( 1", 161. Neglectins seepag~ through the dam and using the flow areas f,17Ciml Figure
~. avera~:e velocities under the dam are about 0.12 m/s for March 19n (dlschargei~:J)I m3 /s).
AUowfng for a freeze~ discharge of about 100 m1/s {Nov~mber 1974), the corres~•dins•,freeze
up velocity is estimatt!!lf as 0.20 m/s. This is much less than the limiting deposition 'ltalue· which
suzgests that vertical growth of the dam i!i limited .by a discontinuation of ice s~ply 1due to
freezln« over of the r.11pids usptream. Vertic~ll velocity profiles under the dam were me;1sr1red in
1977 and 1971 uslns a magnetic flow meter. With the exception Cl~ one profile, the rneasured
values ('I are well below 1.0 m/s which reinforces the above sugestlon.
The measured velocity profiles showed further that the absolute ro!J8hness of. ~! dam
~ersurhce fs highly variable, being sometimes less and sometim~ more than that of llhe! river
bed. Frorn temi·Josarlthmlc plots of the velocity data, the averase friction factor and
equivalent sand roushneu were estimated asD.OI and 0.8 m respectively.
MATERII\L PROPERTIES
At the time this study was Initiated, riO information could be found on material prop!rtlesl
of frazll accumulations. Such information was thought important ln engl~erl"k applications,
478
\uth as assessins effects en Ice brea~~ !G'ces on river structures ... d flow lflrqh
«:curnulations. The followins is a wrr:rn•y ~f pertii'W!nt f~ndi"Bs to da~e.
Compositior1
The non·submerseti portion of the han&lns dam ("overburden") consists of a '-.mt7Mdcy
«:cumulation of snow 11\d weak f!ranular lee with a m;,;dmum thldcne• of 2 m1 the iA~~~· -~ '0.
cl.usifled as 5' Ice "dtalned con&ealed frazll sluzh", usln& the termlnolop of M!chel [I' I· A,
6 ~m thick layer of solid Ice topped the overburden near the rl~er benks, e•ttndl,. 101
approximately the Jive stream boundaries Wtder open-water conditions. These flndlnp .,.
qualitatively similar to those concernlns the Ottawa River dam 1,11 J. .
·Near the free water strfece, there is a 0.) to 0.9 m thlc:lc la,er of •lid lte, underlain !If
the main (submeried) accumulatlm of frazll. The latter Is fairly dense fra&ll IIUih, llmll• in
cornpositlon to the o~rburdm materhih its pores An! satwated lnd its CDhetlon lt much let! !!=;
that of the ~rburden. The overburden ori&ln~tes from saturated fraaJI that rtHt lt.oft tht
water surface tnd :dralnt as the accumulation RfOWI In depth. The •lid lee la,er near the water
strface forms from the slush as Its aptals hive random orlentatlCII\ lnd •aa CDmpwable II
those of the submerpd frazil. The conductJvlty of this layer was fad'd ID ..,., in 1he wertiall
direction (1976), bel" 7.1 and 1).7 a~mho/cm at respective depths of 0.1 and 0.) m whidr
~ests Jmpudty mliratlon (total sample depth•O.l7 ma see allo ( • J ). The thin .tDp layer of let!
near the blinks Is ordinary Ice that forms at the eddy weas prior eo sll"lficant lra&ll depolltlclll
tnderneath .,d oonsequent emerr,nce aboYe the ~~~rfece. That no such la,er h11 betn· found in11
midstream sugests that frazll ..:cumulation In the live stream wea Is much faster 1hln In tfle1
eddY! areas.
The Ice particles ·Jn the saturat~ slush we :between spheroid and ditCDid In npe with II
major diameter of 14 mm. The size cfistrlbutlcn by wel&flt 11 &S~Pfox!matelyr &0 percent In h!
ran~ 1.1 to 2.4 mm; l'. percent in the ranse 2.• to •.s mmr ... d 'percent In the ranp '' tel
6 mr.n.
Shear Strensth and Bearlrw C!f!!;ity
The shear strensth of ~ slush was matured by means of sheer va'Wt attacheciiD a, terift
of i., m lens eJ~ter.slons:. Torque was applied lnd maaswed with a commercially available torque
wrerteh. Flsure 6 shoq shear strenath values h 1) maasured In 1976, Plotted ..,..,. depth at
ttree locations1 two vane sizes were used for comparl~~an (two holes Sf*ed I m·apert W8e
drUied at each location). The scatter in Flpre 6 Is typical and lllustrat~ both the crudeness of
the meast~ement ter;h~1ique and the natural Ylrlablllty of the strensth. No eoNistent 'ftlrlaticn
of T f is evident In Fl11ure 6 but later data have shown that T f Increases ..-rally \lrith hel&flt
tibove the bottom of tl/'le accumulation (see, fa~· example, 197' data In Flpre na this trend 11
sometimes obscured by1 scatter. Fisure I S(Yes a summwy of !3hear stren5th me.surements,
479
plotted in the form of depth-averAged 'tf versus accumulation thickness. The shear strength Is
seen to vary from year to year.
!
I
" I & .
~
i
I r
Vertical profiles' of shear strength Fl&• 7
(March 10 and 11; 1976)
10 20
Thickness oi AccunUI!tlon (ml
Vertical profiles of shear S'lnm&th
(March 111, 1979; 10.2 em var,M!ll
Depth-averaged shear stretnr~th
versus ;accumulation thicknu~sl'
The results of plate bearln& tests exhibited la:-ge scatter, bot average values lncreasedl ~'ith
h!isftt of obsetvatlon from the bottom of the accumulation (h 1), being 300, UO and 90 kPa1 at
v.tfUH of hf equai b 11.2, ,,) and 2 • .1 m r~spectlvely.
Density~ Porosity
The dry density of the slush (o1) obtained from the drained weights of known volumesp
increased with hi as shown In Figure 9. The porosity of the eccumulatlon (£f) is given by:
(I)
tn which pisdenslty of Ice. From Figure 9, £ f Is calculated as 0.,1 and O.JJ at J\r"'2m and 112m
~spectively.
480
i 7-:=]
~ 4 .,,L_._,~
HlillM ...,. !loiiOI'I rJl ~"'"· hf ""'
Fl!• 9 Dry densl ty venus hei&ftt above
bouom .of accumulation.
Usln& cverap! va(\M of 1916 data On o1 (Fipre 9) and Tf (Fipe 6) at correspondi"l
hel&t~ts h1, an empirical correlaticn was obtained, .s followsr
"\'{ ,. 0.2, ,, ... , (2)
In which Tf Is In lcPa iiiild "r is In Kslm'. Equation 2 applies In the ranses T 1sl0•7' kPa Mel
o1=•'0-'20 KKfm 3•
Flsure 9 may be used further to determine the stress-density relatlanshlp fcw the
accumulation. 1lle ftrtlcal stress sradlent due to buoyancy isr
())
In which p:vertlcal stress; sseccelerarion of srevityJ .net P._•density of water. U~f11 ~tian I
and lntesratlns slves:
An approrlmate calculatlcn osln& sraphlcal intesration (tee Fipe ') resulted i~ the strell·
density relationship depicted In Fisure 10 aiCfll with relevant flndlnp for tnow I UJ. ,ot ~
same stress level, frazU.densiflcatlon ''about 1,, times that of the lower bound of Mellor'• ( UJ
data. This Is prlm•lly caused by differences In temperature and water content and, to a lESser
desree, by particle seometry effects. The mechenlcs of denslftcatlon chanp siplficantly near
0 0 C In a saturated media where pressure melt!ns end reselatlon ( 71 stronsly affect the
deformation of frazll Ice and allow demlflcatlon at a lower stress level than would be fOIM In
cky, co~der media. Colbeck et al [I) discussed this difference and reP«ted test~ results that
comp•e saturated with dry snow at -2°C, If the same denslflcation is considered linear Mel
applied to the J)fetent results, the transformed diata fall much clo!ler to Mellot's (Fipre 10).
Intrinsic Permeebllt!f
Permeebility was Clllculated based on flow rate of a lOW motor oil throtJih • cylindrical
sample mder a fiiCed head {JO). The equation used ls (2h
•
•
ci
____ ......,,..,..,
[O· .... c]
• 0 • 0
0
0
Oll'rSHRI Dele
• set.,..., .. o.,. .,_,..,.,..
Fls. 10 Comparison of ntress-
deislty relationship for
the frazll accuf1rtUiatlon
with Mellor's da1tl1 for
snow.
In 111hich QsfJO'II rate; kf2lntrireic permeability; A:cross-sectional area of test cylinder; Lt=Jfluld
vltc:oslty at test temperature; "=fluid density at test temperature1 h':.tM!~adlstance from blip of
Input reservoir to tip of -*'•in tti>e; 1nd L'zlensth of test cylinder. Measured kf valuetJ lftt!!re
16.Jxto-', U.6xto4 and U.Oxlo-' cm2 at h1 values of 2m, 7.6 m .,d 12.2 m. The hanKins !dam
permeability is between those of coar.se sand and fine sravel [ 12) which appears reasonabl•e 9nc:e
the lrull particle size Is consistently between 1 and 6 mm. Snow with 1 to 2 mm particles t•s A
k1 value of 2xiO-' cm2 ( 9).
BREAKUP
Brealcup observations hsve been carried out annually durin& the period 197,·79. Detailed
information may be found in ( •J 1 only a brief summwy will be stven here.
The han&ins dam obstructs the prosreu or the sprins breakup and Initiates Ice Jam!a, moet
of which are major. .Removal of the: dam Is usually forced, that Is, It shears off at the &Ides
(rOIJihly at the live strum bou1daries) and is subsequently broken Into small pieces upon the lflr.l
release ol lhe jam upstream. There was one Instance, however, when the ~tream ice paassed
under the dam; the lattt!l" remained in ;place for several days IW1d was removed sradually b)• 11rater
erosion (1977). Twice !1976, 197~), removal of the dam was followed by surstns ice Nu that
were only ~rrested 2 ~m upstream of the Smoky River mouth (about .)8 km downstream olr the
ct.m site)J en both occasions, major jams formed there and sradually broke throush Into: P'HCe
River. The effect or the dam on breakup near the town of Peace River (Fisure I) can b«~ either
ISZ
beneficial or detrimental depend!,. on prevalll"l Ice eondltlans In the Peece lltlver ltMif.
Continued annual observations are deemed desirable JO as to obtain a more compl~t• record of,
lnd assisn frequencies to, various events of Interest.
To develop a crit9!rlon 'for the removal of the dam, an ;srprc•lmate farce 'SftaiJSitwa!l
carried out ( fl ), as outlined briefly below. Upon releaw of the jam upstream, the main
horizontal force on the dam is ~ net hydrostatic presS\We caused bf the advMCi"l w5ter w .. e
CFipre II); other hwces, e.s. hydrodynamiC Ioree and pressure of advMdnl ice jams ate
relatively very small in thh case. The dam shears off when the .'applied fcrce e•ceeds Its
resist.,Ce on two vertical surfaces which separate the srounded portions at thf .:cumulation
near the banks from the floatlns portion In midstream. Analysis hat shown that the •m wUI be
removed when
,,,
In which W=dlstance between tt-e two shear Sl.zorf.cesl :'f:averap shear strns ewer thf tht•ed
area; m ST=tce slope of the •.~pstream prn fust prior to release. Oet.alled breallup·••• taken In
191' indicated that ST was In the ranse 0.004) to 0.007. Uslns W')217C~ m (tee Plaure •), Equatien
'ftlves T=l to fi.S kPa which is generally lower than measured midwinter values shown ~, Fipre
I. It is noted that a decrease in strensth is likely cllrins die !lpi"lf'll breakup If the water
ternperature rises above 0°C. If t and W do not change appreclal»>y from year co year, Equation
' would 5tUest that there Is a limitln& value of ST' between 10~00• .,d 0.007, that mutt be
at1ained before the dam can be removed. This Is c:onslstent with the 1977 fcndlf11, I.e. thet the
dam did not "break" m the upstream Ice passed mder Itt the avalla~fe data fer 19771ndicated
that ST could not have exceeded 0.00)9.
Prier to jam releaset
negligible horizontal component
of net hydrostatic pressure
Fis. II Sketch of assumed mechanism of dam removal
Shortly after iatn releuea
slsnlficant horizontal CQmponent
of net hydrostatic pressure.
SUMMARY AND CONO.USIONS
A hanKin& lee dam that forms in the lower Smoky River has been the obJect of MniP.l\1 flt·ld
observations Md the results tave been reported in the previous sect!Cils.
The hln&i"l dam site Is a depression of the river bed, preceded by a section of rapids. The
mod! of dam formation is essentially •s has already been described by others: slte-tpeclflc
pecullwltles have been Identified~ based on a visual record obtained by means of an .. utomatic,
time-lapse Jflotolf'aphy apparatus.
The streamwlte prollle of the dam !t roU&hly trl.,sular, with a bue of 300 te» 700 m .-.d a.
depth oi II to 16 fi1J ~he dam consist! of porous frazil slll!h with lee particles 1 to 6 mm in size.
The In situ shear strensth of this material varies from ye~r to rearaln ~ny one year, lt lncr.eases
with helsht above the bottom or the <lccumulatlon and senerally does not exceed ao kPla. A
similar -narlatlon was f~ for the d-y density of the material. The Intrinsic permeability ()f the
dam is about U.,x 10~ cm2 and decreases siiWttiY with helsflt above the bottom of \he
accumulation. Velocity measurements undl!f the dam Indicated avera&e values of O.OS and o.a m
for. the friction fiCtor lnd equivalent s.nd roughness heisht of the dam tn:ferslde respectively.
Ourlns sprl" breakup, the dam initiates VI Ice jam '-"stream. Usually, final release ;of this
Jam Is followed by removal of the dam and occasional lee surses that are only arrested rn~ar the
river mouth, 31 km downstream. 0., one occasion, jammed lee '-"stream released and was
tral'llported under the dam rather th., dlslodsJns it. To exp!aln the ,;echanism of dam r1emoval,
a preliminary force analysis has ~n carried out and partly documented uslns available data.
The effect of the dam on sprln& water levels near the town of Peace River can bf~ either
beneficial or d!trlmental dependlns on simultaneous ice conditions in Peace Rive•· itself.
Continued observations are deemed desirable In order to develop an adequate statistical record.
ACKNOWLEDGEMENTS
A portion of the Mtrlc repclf'ted herein was carried out as part of a continulns field research
prosram on river lee hydraullcsa this prosram Is conducted by thfe' Transport&tlon and Surface
Water f!il&lneeri"B Dlvls!on of Al~rta Research C~cU In cooperation with Alberta Environ-
ment, Alberta Transportation and University of Alberta.
Oa:aslmal assistance provided by 8. C. Hydro observers (F. Sampson, B. Tutt, M.
Vanderkraan) Jsaratefully ICicnowledsed. G. Childs, G. Putz and T. Rldsway of Alberta Research
Ccmcll pertlc:l~tee In the field work. Review comments by T. M. Dick, Y. L. tau and G. Tsans
of Environment Canada •e ..,preclated.
REFERENCI!S
(I) Atm01pherlc !nvlronment, 1971. Monthly Record-Meteorolo&ical Observatlonsi..,Westem
Clftada". November, Vol. 63, No. 11, Part 1.
[ Z) Amyx, J, w., Ban, o. M. Met Whltlnz, R. L, 1960. "Petroleum Reservoir Enslneerl"!:
Properties". McGr4w-Hlll, New York.
I :U Barnes, H. T., tnl. "Ice !"llneertns•. Renouf Publlshitli Cot,.,_.,.,, Mantreal.
h J Beltaos, S. and Diean, A.. M. Jr., 1,.1. "Field lnvestlaatlons of a H.,si"' Dam•. NWRI
Hydraulics Dlvlslon Unpublished Report (in prep.).
U 1 Bolsensa, S. J., 1968. "River Ice lams-A Literature P.evlew•. Research Rf:90't ,_,,U.S.
Corps of !flllneers, Lake Survey District.
16) Calk!ns, D. J. and Ashton, G. D., 19,. "Archl"' of Frasmented lee Covers". CIMdl.,
Journal of Civil Enslneerins, Vol. I, No. '• pp. 392-399.
I 7 J CGlbec:k, S., 1976. "Thermodynamic Deformation of Wet Snow". U.S. Army CRR!L Repcwt
76 .....
111 Colbeck, S. C., Shaw, K. ·A. and Lemieux, G~ 1971. "The Compression of Wet sno.. • u:s.
Army CRREL Report 78-10.
t 9 J Colbeck, S. C. and Davidson, G., 191). "Water Percolation thrqh Homopneous Snow".
lnternat. Sympo5lum m the Role of Snow and Ice In HydroloaY, Vol. I, pp. 2•2-2,7.
I 10) Dean,A. M., 1976. "A Method for Oeterminins the Permeability of Frazlllc:e". u.s. Army
CCR!L, Technical Note (unpublished).
Ill J Gold, k ll. and Williams, G. P., 1963. "An Unusual Ice Formation on the Ottawa Ri¥er•.
Journal of Glaclolocy, Vol. If, No. 31, pp. '"·"3.
(12) Hqh, B. K., 1969. "Soil Moisture". In Basic: Soils £f11lneerl"', Ronald Press, New York.
Chapter 3, 76 p.
ill) Mellor·, M, 197ft. "Mecsnlque de Ia neiae". Proc. Cirlnde!wald Sympoelum, A,wll, lAMS-
lASH Publ. No. II ••
lltt) Michel, B, 1971. "Winter R~ime of Rivers ~~nd L11m". U.S. Army CRREL Monocraph H!-
Bla.
I UJ Michel, 8., 1975. "The Formation of Ice Covers". Unlver!!te Laval Report CiC5-7,·09·0'·
tl6) Michel, 8., 1,8. ,.lice Accumulations at Freeze-Up or Break-Up". Proc. IA.HR Symposium
on Ice Problems, Lulf!a, Sweden, Part 2, pp. 301-317.
ll71 Sampson, F., 19iJ.. "The Ice Re&lme of the Peace River In the VIcinity of Portate
Mountain Development, Prior to and ~rln& Diversion". Proc:. Seminar on Ice Jams
In C4nada, University of Alberta, Edmonton, ~ublished as NRC Technical M•mnrmn•
dum No. 107, PP• US-IS&.
485