SNOW ROAD CONSTRUCTTON
A TIIESIS SUBMITTED TO THE FACULTY OF GRADUATE STUDIES UNIVERSITY OF MANTTOBA IN PARTIAL FULFILMENT OF THE REQU]REMENTS FOR THE DEGREE MASTER OF SCIENCE TN CIVTL ENGTNEERING
þv
MTCHAEL ZENON KOVÙALCHUK
Winnipeg, l4ani'toba December l-976 ''SNOhI ROAD CONSTRUCTION''
by
MICHAEL ZENNON K0Ì¡IALCHUK
A rlissert¡tion subnritted to the Faculty of Graduate Studil':s of the University of Manitob:r irr partial fulfillmcnt ol'the rcquin:ments ott tlrt' tlegrce of
MASTER OF SCIENCE
@ 1977
Pernrissiort lus lrec¡r grurttcd to tlto LIBRARY OI¡ T¡ltj L,NlVUlì' slTY Otj MANITOtJA to lr:¡rd or scll copies of this dissert:ltiorr, to tlrr.' NATIONAL LIBRAIì,Y Ot¡ CANADA to ¡lticn¡fitm this dissertatio¡r and to lend c¡r scll ctlpics of tfte filnt, and UNIVtiRSITY MICROFTLMS to publish i¡¡¡ abstruct of this dissertation.
The author reserves othcr ¡rublication rights, and ncithor tl¡e ABSTRACT This study investígates the application of snow in the con- struction of snow roads and the suitability of the snow road'as a trans- 'i portation facility. The sngw properties are reviewed and the current status : of'snow road téchnology is identified. In the fíeld studies, where vari- ations in snow road construction procedures are evaluated., a conventional I snow bl-ower is utilized as a snolar processing implement. The resul-ts-in- ' l:t-::": dicate that an equidimensional particle size is produced with no significant ì / chahge occùring in either the particle size or distribution with repeated processing.. The results fur:ther identify the operational difficulties in . frandfing blower-processed snow during snow road construction. Increasing i the snow road density by íntroduction of, free moisture into the pore f spaces, through surface heat applícation, iS also investigated. For this purpose a prototype wood-fired heater is utílized. The test results give i promising indlcation that this method of surface heat application, in ì combination with heavy compaction, could have considerable practical merit. The snow road strength is índicated in terms of snow lhardnessr determined i .' by the Rammsonde cone penetrometer. .The conventional- iìammsond.e hardness I eguation is re-evaluated and the hardness index is correlated to snow temperature. i---. I l-l- . ACKNOWI,EDGEMENTS I wish to express my sincere appreciation and gratitude to Dr. K.M. Adam for his guidance and assistance in the preparation of this .generous thesis and for his and unstinting help during all phases of the , .: degree progran. Sincere appreciation is al.so éxtended to the members of the thesis examination committee, Dr. J. Graham and Dr. G.E. Laliberte, ,l ..' fortheirinvaluablesuggestions. \ r::,' : : r: :i :: I give full- acknowledgement of my inde.btedness to the University : ,: :' r of Manitoba Maintenance Department for making avaitable, on a rental basis, '.,'.,, the necessary equipmênt and personnel, and to the National Reséarch Council provided grate- I of Canada who financial support for the project. I al-so ]ful1yacknowledgethepersonaIsourceSofassistancewhichduringthe ;: . course of the degree program incl-uded: a University of Manitoba Graduate , ' i Research Assistantship, a University of Manitoba Graduate Fellowship, and a National Research Council Post-graduate Scholarship. il I ! To Ms. Janet dourlay, I extend my warmest thanks. for doing a f.ine l i : job in typing this manuscript 4nd for offering timely editorial advice ': Finally, I wish to express my deepest gtriitrd. to my parents for ',,,',' :: ..; .' guided mr¿ , institl-ing in me the value of education that has endeavours. ,.: ,ì , ;l- --- - l_ l_ J_ TABLE OF CONTENTS .i Chapter Ee+ ABSTRACT i. ACKNOWLEDGEMENTS ii. TABLE OF. dONTENTS iii. LTST OF TABIES . vii. LIST OF FIGURES viii. ì LIST OF PHOTOGRAPHS x.. LIST OF SYMBOLS xi. I TNTRODUCTION t. ÏI SNOVü AS A MATERIAI a A-. SNOüI METAMORPI{ISM 3. Destructive Metamorphism 3. Constructive Metamorphism 4- Melt Metamorphism A Pressure Metamorphism 5. B. , SNOW AS .A POROUS MEDIU}I Porosity .. Pore-size Distribution 8. 'o Saturated Permeability and Conductivity t0- Relative Permeability and Saturation 13. Relationships C. WATER FLOI,V THROUGH SNOVü L7. General Flow Equations 17. Capillary-pressure, Saturation Relationships 20- ]v. Chapter Page D. HEAT AND MASS TR.âNSFER 1a E. SNOVí MECHANICS : Densification 27. Response to Rapid Loading zo-ôô Response to Sustained Loading 31. F. INDEX PROPERTIES 34. Density 34. ' Grain Size . ?¿. / I Graín Shape 36. Temperature 37. Hardness 37- TII hTTNTER ROADS 43. System of Classification 43. Route Selection 44. Measures to Increase the Bearing Capacity of Snow 45. A. SNOW ROAD CONSTRUCTION 46. Initial Site Preparation 46. Dis agþre gati on/Proce s si ng 46. .Leve ling-Compacting 47. . PreBaring Final Surface 47. B. CO]iSTRUCTION METHODS AND RESULTS 49. IIhe Scandinavian Experience 49. 52. 1. Camp Hale, Colorgdo 52. 2. Point Barrow, Alaska 53. ,3. Churchill , Ma¡itoba 56. .v. Chapter Page 4. Greenland 58.t ' 5. Kapuskasing, ontario 63. ' 6. McMurdo Station, Antarctica 66. 7. Norman Vüell-s, N.W.T. IV F1ELD STUDIES 73. A. EXPERIMENTAL TECHNIOUES 73- Cofistruction of Slrow Road Test Lanes 73. .Surface Compaction at OoC 74. Ramrnsonde Hard.ness Tests 74. Temperature t5- Grain Size 75. 16. . Density Ttre Prototype V'food-fired Surface Heater B. RESULTS AND DÏSCUSSION 78. . Construction of Test Lanes 78. Porosity, Density and Grain Size 79. Performance Characteristics of the Sicard Snow 81t Blower 8r. Rammsonde ttardness - Temperature Relationstrips 82. Burn Tests 83. V CONCLUSIONS AND RECOMMENDATIONS 85: BIBLIOGRAPHY 88. APPENDIX A A.1 RAMMSONDE HARDNESS DATA AND OBSERVATIONS ' :. vl_ . Page APPENDIX B 8.1 EXPERI¡4ENTAL RESULTS FROM BURN TESTS APPENDÍX L c.l THE RAMI4SONDE CONE PENETROMETER APPEnorx b D.r DETERMINATION OF SNOW DENSITY USTNG A RA,DIATION MEASURE¡4ENT TECHNIQUE r'1 APPENDIX E coNSTRUcrÍoN rgureMENT AND'MATERTALS . I V]-I. LIST.OF TABLES / Ta-ble Page 2-L Effecti_ve thermal conductivity* (K ) 24. of dry snow. e. 2-2 Snow Classification. 35. 2-3 Summary of regression and correlation 4L. coefficients. _: 3'L summary of traffic tests 65 A-l Summary of Rammsonde test results. A.2 A-2 Summary bf air temperature, énow road tem- A.16 perature anil re1ative humiditlz for duration of Rammsonde testing. A-3 Aver-age äensiti,es of single- and multiple- A.26 processed snow roads. : B-1 . Summary of snow road density profíles before 8.2 B-2Summaryofsnow'roadd.ensityprofilesafterB.3 lburn tests.. i I --iir::":1 vt¡_r_ . LIST OF FIGURES l Figurê Page 2-L Capillary pressure as a function of 20. :. liquid saturation.. 2:2 Relationship between dynamic Youngr s 28. modulus and snow density. 2-3 Unöonfined compressive strength versus . 30. densíty for naturally compacted snow: l 2-4 Tensile strength of high density snow 30. at a Èemperature of - 10oC 2-5 Strength ín doubl-e shear plotted against 30. density. I 2-6 Required ram hardness for supporting 38. various wheel loads 2-7 Required hardness (or.strength) of a 39. snow pavement for various wheel- Ioad conditions. 2-8 Average tensile strength as a function of 40- ram nesistance and age 3-1 Normal hardness growth in a síngle layer 60. of polar sno\^r. t 3-2 Typical vertical hardness in compacted 61. I snow compared with its redistribution after surface hardening. 1 3-3 Tentative minimum hardness guide for 62. ia::) compacted snow runways on deep- snow. r:, ¡: A-1 Typical snow road hardness profiles as A. 15 indicated by the corrected and uncorrected versions of the Rammsonde hardness data. A-2 Relationship between' Rammsonde hardness, N.23 age-hardening time and air temperature. A-3 Scatter diagram of RaJnmsonde.hardness and A..25 s.no\^r road temperature data. l_x. LIST OF FIGURES (cont'd) Figure Page ! A-4 Í\zpical snow road hardness increase due A.27 to surface compaction at OoC. B-t Snow road density profiles before and 8.5 after burn tests c-1 Diagrarn of Rammsonde cone penetrometer c.2 showing ttre components of the instrument. D-tr Diagram showing the arrangement and basic D.7 components of the radiation/scintillation appar.atus. \ ,'r' .,--r..:, : E-l Diagram showing the construction of the E-5 prototype wood-fired surface heater. I: ::.: :' LÏST OF P'HOTOGRAPHS I ++*j4 Page .E-t Windrowing the snow with the Champion E.6 Motor grader in preparation for proces- sing with the Sicard snowblower E-2 A view of the snorry-blower/tractor unit E-6 'processing the windrowed snow and de- positing it along the proposed ::oadway E-3 The roadway after final- snow processing 8.7 and deposition E-4 Road surface after leveling and dragging 8.7 with the steel drag. E-5 Station-wagon traveling on the snow road 8.8 \^rith no detrimental- effect on the road s.urface ti- o The Champion motor grader blading minor 8.8 snow accumulation from the snow road .surface E-7 The prototype surface heater fully 8.9 fueled with one (I) cord of seasoned pine E-8 A, .view of the surface heater in 8.9 operation . xl-. LIST OF SYMBOLS SymboI Dímensíons * a Empirical constant in strain-rate equation . none a Subscript indicating "air" none a Subs,cript indicating "apparent,, none a Empirical constant in equation fgr hydraulic conductivity none b Ernpírical constant in creep-rate equation no,ne b Subscript indicating "bubbling" none ¡ C Hydraulic conductivity, L/t ,C Specific heat capacity , .H/MT '2 Vapour concentration M/L. C Volumetric heat capacity u.1t3t C Correction factor for effective conductivity none c Subscript índicating i'crit.ical saturation" none ! c Coefficient of cohesion F/L. c Subscript indicating '!sapil1ar1z" none 'Diffusivity i ', : L. /T d Grain diameter L ) E Youngrs Mqdulus. F/L- e Subscript ind.icating "effective" . none e Coefficient of restitution none e Efficiency none a g Gravitational acceleration L/L. H Vertical distance i A counting integer none { LIST OE SYMBOLS (cont'd) Symbol Dj mensionst i Sulscript indicating "icerr none K Thermal conductivity . H/LTL t- ) k Permeability D k A measure of con.fidence in units of "number of standard deviations" L Latent heat of fusion \' L Characteristic lengtki L ,r -: I Subscript indicating "liquidrr none .t / m Mass M N Empirical.cònstant in equation for hydraulic conductivity n A counting integer none :' o : Subscript indiöatin9: "sêturated" none P .Pressure F/L2 p .Subscript indicaÈing "constànt.pressure" none , 9. fÍeight of cone pcnetrometer F q Intensity of evaporation oi cond.énsation vln3t a q Heat flux H/r,'L R Rammsonde hardness number F ': 'r' Correlation coefficient none r Subscript indicating "relative permeability" irone r Subscript indicating ":residual" none S Saturation Penetralion L s Subscript indicating "sölid" none xl-l-I . LIST OF SYMBOLS (cont'd) .t SymboI llr mens l-ons T s Subscrípt, indicating "snow" none S Specific surface 12 /t3 S Shear strength F/L2 T Temperature T T Detector resolul-iorr time t t. Time t: V Veloci!y L/t Îl Vüeight F I w Subscript indicating "wetting phase" none x Cone penei:r'tion L Spatial cooráinate L z. Spatial coordinat L .c Thermal diffusivity t2 /t Y WeighÈ clensity F/L3 A Denotes a difference none.. I Exponent in .relative permeability equation none -l E Strain-rate i t : A Volumetric water content ¡,3 /n3 .À e91g-s-rz3 listribution index; -d (loqS-e-c ) ,zd ( loqP ) none Dynamic viscosity EL/L2 u ; v Poisson's ratio L/L p Mass density M/L3 ç Summation none xrv. LIST OF SYMBOLS (cont'd) Symbol Dimensions f .2 o Stress F/L .0 Porosity none ó Angle of shearing resistance none ú Gra{ient cperator I/L div Divergencè operator L/L Superscript indi-cating resultant of \ '.,. ,. , surface and body forces. none ' ': i :.: .:; . : i r' * ' Fundamental dimensions: force ...... F 'heat ...... H length :...L iååi.;;;;;; ::::i time ...... t 1. CHAPTER I INTRODUCTION J ' With íncreasing o-il and gas exploration and the subsequent interest generated by this activity in Canada's North, the potential ,.::,:: .:::, damágeresultingfromthemovementofgoodsandmachinesbecomeSin_ . creasingllr important. In particul-ar, consideration of a Northern pipeline development has brought tå tfre forefront a recognition of \ ::::r:.:. i -- many niulti-disciplinary- -i -a ^-t- problems" l ,'.::,,, i The studies by Adarn'[1,2] point out the applicability of snow ' : ,,..,.. ''' '' lt:' roads and ice-capped snow roads in the transportation of men and rnaterials during winter pipeline construction. The studies also in- dicate that while the idea of compacting snow as a means of improving procedure .l oversnow trafficability is èxtremely old, the for constructing i ' snow pavements capable of carrying heavy wheeled tr.affic, is not well- established. This is partly due to the fact that in the past no ap- . 'l parent standard for the identification and treatment of the snow pro- i ¡ perties has been fol-low.ed and that the existing information, though ' 1 ;. extensive, is widely disperse;d.. .i.:'..:;..;;;¡,i : ..::iì:.., In Chapter II and III of this study, pertinent facts are .;,:.,.,.¡]'1:;,, , "pooled" to form a single comprehensive review. The review includes discussíon of the porous characterístics and the thermodynamic and mechanícal behaviour of snow. A procedural guide for the fietd identi- , ,.. , :, .,- ..- i , fication of the index properties is also presented. Chapter ïfr is a ',,'' ":':. surnmary of'the snow road construction experience where construction 2. equipment, procedures and results are discusçed. 'I Previous stud.ies indicate that lnow compaition alone will not produce the strength required. of snow roads subjected to intensive wheeled traffic. Disaggregation, mitlíng and mixing of the snow (prior to final grading and compaction), to produce a more non-uniform particle-size distribution becomes necessary. For this purpose, morl,i fied agricultural- earth tiffers and speçial pulvimixers have been used, but with varied success. In this:study, the conventional snor4 1,, blower is evaluated as a snow'processing ímp'lement and the significance ' :.: 't-- ofi multiple processing is determined. r-: ì The ice-capped snow road is a processed and compacted snowbed with ,water applied to the surface . to increase the road density and strength. Although this class of winter road offers some significant advàntages, i ts use may be seriously limited in permafrost regions where the large quantities of fresh water required may not be available. The feasibility of utilizing a portable wood-fired surface heater to iproduce the free water.is investigated as an alternative to the direct water application mettrod.. The wood burner would have the dual- advantage of utilizing right-of:way släsh material as fuel , and at the same tíme provide a facility for the disposal of the waste material. i:!ii ) CHAPTER TI ! SNOW AS A IVIATERIAL Snow is a porous medium composed of ice crystals and/or aggregatgs of ice grains. The pore spaces contain air, and water vapour andrin the case ôf "v/et" snow, liquid water as well. Thus the snow pack forms a heterogeneous system which exhibits a. complex and continuous interaction b.etween thq phases. ' In this ctrapter, the properties of snow and the þrocesses thaþ occur due to the thermodynamic instability of snow are reviewed. The review cítes pertinent theory and includes qualitative explan- ation of.''' governîng principles and terminology. A. SNOW METAMORPHIS}4 Destructive Metamorphism . After the depcsition of fresh snor¡r, very local- transfor-, mations, calÌed destructive or isothermal metamorphism, are first observed. During this process the sharp points of the branches of individual- snow fl-akes become blunt. This is achieved partly by evaporation flue to the higher vapour pressures associated with small lradii of curvature, and partly by the migration of molecules in the quasi-liquid surface layer (Bader [6] ). The crystals become rounded and the larger spheres then grow at the expense of the smaller ones exclusively by vapour transfer. After a few days, crystal shape is al-most completely lost. Each grain is a single crystal ranging from 0.5 to I mm in size with only one or two smaLl crystallographic faces.Thegrainsarejweaktybondedandthedensityofthisfine_ ï: .3 snov¡'usual-ly ranges from 0.15 tor O.25 qm/cm' . .grained Following the original deposition, of snowr a temperature gradient through the snow cover develops (colder at the top than at the bottom), condtructive or gradíent metamorphism sets in.. .and 1 Constructive metamorphism involves heat and mass transfèr and is' .characterized by growth of .selected crystals; de.cre.ase of crystal :j.: 'number per unit bulk volume; and development of crystallographic / faces, edges and vertices (de Quervain t18l). In dry snow, the mass transfer occurs primarily over the vapour phase by diffusion androwing to the.temperature grad.ient, is accelerated by convective ? At low snow densíties (beIow 0.3 gm/cm'), constructive metamorphism is very rapíd and results in production of a distin- ctive sno\^/ type known as "depth hoar". Here grain size is between 2 and I mm an{ singlg crystal-s as large as 15 rmn have been observed (Bader I2l). Depth hoar extribits very poor bonding, and has a high viscosity. Tt is the major cause of aval-anching. and is most difficul-t to compact to a hard. snow pavement. Melt ['letamorphism '.:...: ence of liquid water. As the weather becomes \^¡armer, and the snow becomes'moist, temperature gradients and their effect vanish (Yen f44l). Crystallographic elements also quickly vanish due q to the high surface tension of the water iilm covering all the ! grains. Itrus crystals become rounded and. cl-usters of grains coalesce to form larger polycrystalline grains. As meltwater percolates through the snow, crystals grow to a.maximu¡n size of about 3 mm and composite grains to about t5 mm. The bondíng between the grains is very weak and within a few days, the well known t'rotten,, snow of the thaw season is produced. Upon refreezing however, the wet i snow.aquires a high strength dUe to the growth in bond area. . .::i Ì, ., Due to the inhomogeniety of the snow pack, the percolating i tt,' meJ-t¡vater is not equalty distributed in all the layers. Some layers retain very little water while others absorb water almost to ' saturation whích convert to dense ice lenses on refreezing. Pres sure lr{etamorphism 'Pressure metamorphism characterizes the densification of dry ,r.rrJ (þranulated sriow accumulated and subsequently compacted to glacier ice). This process is very sl-orrr and may ta.Jce several '- - i decades to change snow of density{-rz 0.45a, oa gm/cm'to^- /*3 +-n icei ¡a ofn€ density.{anc.i +.r ?ì 0.83 gm,/cmJ. (At the density'of 0.83 9m/cm3,. air permeability decreases to zero and the snow changes to ice by definition. Pure 2 ,icehasadensityofO.9L7gm/cm-).Duringthisperiod.,cjfs1ow change, grain and grain-bond growth by vapour or surface migration appears to be of secondary importance (Bader t6l ) . The densification is apparently d.ominated. by processes of mechanicaf defòrmation under pressure. At the surface of the snow pack, hard crusts generally fqrm. 1.. 6. Bader [6] attributes crust formation to:, refreezing of surface layers previously wetted by thaw, rainfall or wet fog fall=out, surface condensation from moist air, wind packing; and vapour migration from lower sno\^r layers. i ,il 7. B. SNOhT AS A POROUS MEDIUM The parameteris pred.ominantly used to describe porous media systems are porosity, the pore-size distribution, and the relation- ships between capíltary pressure and the permeability and saturation : : .1.., ì.: of the fluid pháses. In treating snow åyste*s (wet or dry), -.1 of snow. As evidenced earlier in the discussion of snow metamorphism \ . ,,, ,: ': . : i .'""''-" the physical properties of s.now are very strongly tíme and temperature . :. .l I extensive experimental analysis of the parame.ters. porosity 'l , 'Porosity is the ratio of volume of voids to the snow bulk I volume. It is calculated. from the measured snow density using the following relationship ; .0.917-p P.-PLss =- -'l-'lnorì^ .vrv Y (.2.L) .1 p. o-9L7 S i where . 0 = porosi-ty (sometímes cal1ed "absolute" porosi.ty) , j d.ensity of pure g^/cm3, .l_0, ice = .O.9I7 p = density of snow, gm¡cm3 .S The term "absol-ute" porosity is used when dealing with .high:density :::;::..: r;i:: ' .-:-.--- ..-t t :- , ?:' Snow (above 0.7 gmrlcm') and is distinguished from "relative porosÍty' which refers to the volume of communicating pores only. apparently ? at densities greater than o.7 qm/cm", the volume of the isorated Ij.-:...tt' ; :: :.1 -'---"---'I:a': 8. pores becomes sígnificant. The relative and absolute porosities are approximately,equal for low and medium density sno\^¡. No .measurement of relative porosity qf high-density snow was found in the reviêw of pertinent l-iterature. Pore-Size Distribution :., The hydraulic.behaviour of a porous medium is affected by , the porosity and the. distribution of pore:sizes from point to point l ,l in¡the pore space. The pore-sj-ze distribution indexi À, is derived '.! from a log-1og plot of effective saturation S., versus capiltar! pressure head expressed as P^/pS, (Laliberte I25l). The negative slope of the straight líne portion of the qurve represents the pore- , sizè distribution index, À. Qualitatively, the larger the index'' the more.uniform is the pore-size distribution. Capillary'¡fressure i .. for the air-water system is defined as the pllessure difference ' ,across the interface of the two immiscible fluids, given as. Pcaw =P -P (2.2) where P- = capillary pressure, dynes,/c*2, .nd .'. .c P_, P- = prêssure of the air and water phases respect- a' \¡t ? ively, dynes,/cm-. Effective saturation. S^, is defi.ned by; s -. s-^ c : _ .r (2.3) .te l_-S where ri:i:i:::|4 9- S = saturation, (volume of water expressed as a decimal 'fraôtion of the volume of voids),. and S-r = residual saturation, (saturation corresponding to a capillary pressure at which the saturation de- ., creases very little with large .increases in capi.lla,ry pres sure.) Fries,en l22I obLained values of À ranging from 6 .2 ta 9.9 for snow.densities betwe'en O'.40 and 0.48 gm/cm3. Colbeck (noted by Adán and.Wilson t3l) obtained ),: 4.9 for snow densities of a 0.55 and 0.59 gmlcmi. Latiberte t25l indicates a pore-size distri- bution index of. 7.3 for glass beads (uniform pore size) and 3.7 for a fine sand ',] Pubbling Pressure The bubbling pressure head, Pb/pg , is defined as the inter- lcept of the straight.line in the log-1og plot of S, versus te P^/pS (Laliberte c , t25l). The bubbling pressure is a function of .the largest continuous pores,of the porous rnedium u.rrd'ir normally very close to the minimum capillary pressure head where the non- wetting fluid permeability can be measured during drainage. The lower the bubbling pressurer' the larger are the continùous pores in the porous mediutn Colbeck (noted by Adam and Wilson t3l ) obtained a bubbling pre'ssure head of 5 cm for a snow-kerosene system and 3..7 cm for a snow-water system. Friesen l22l estimated. a bubbling pressure head r0. of 8 cm and 9 cm for.sno\rr densities of p.48 gm/cm3 and 0.40 gm/crn3 respectirzelyl in snow-so1trol C*syçtems ------_Saturated Permeability and Conductivity Fluid permeability.is one of the basic quantities that describe the physical properties of a porous medirrln . sat.urated "permeabirity" is distinguished from saturated 'iconductivity" the former being a quantity depend.ent on the porous.matrix and the latter .:'. being a quantity dependent on the porous matrix and the fluid I The relationship between volume flux and cond.uctivity (for stable porous máterials furly saturated. with one liquid at a con- stant temperature) was discovered exper.imentally by D'Arcy. A popular form of the Darcy equation for 'one- dimensional flow is: n:t A (s-) ' oct n=-Õ 'L (2.4) where ! C = conductivíty, cm,/sec , ' n* = piezometric potential, dynes/cm-2 (represents the resultånt of the.normal surface and body ' forces per unit vol-ume acting on a .volume element in a fluid of uniform density and has dimensions . of energy .per unit volume), .1 pg .=specific weight, dynes,/cm', Pi piezometric pg = head, cfr , ìT, = 1.nnth over which the incrementaL change in piezometric head occurs in the direction of flow, . .l cm , .and t- l , Soltrol C is a light hydrocarbon oil 1r. volume flux, cm/sec (the component of the vol_ume I of discharge per unit time per unit of bul_k area in the direction L). The Dârcy equation re-written in terms of permeability , .-i:, ''l::': ''' becomes '\. q=-+ uL (2.5) \ ., r--. where t , '. dyne-gec dynamic , ,,, U = vis-cosity of the fluid ', and ./ : 2') ...:: : 'z ô Cm r:i: K = pêrmeability, cnr-. The coefficients C and K are related by C=K-. Ocf (2.6) u Calculation*s of permeability and conductivity based on structural parameters have been proposed by various authors. The ffozeny-Carman equation (Laliberte t25l ) approximates the saturated permeabilíty of- a porous medium with uniform poie size and low i eccentricity: .3 *^-.= * (2.7) ow. 5"2 where t .owK = saturated permeability, cÍl- (permeabilily of the wetting phase at saturation S = 1) , 0 =.porosity, dimensionLess, and 21, s = specÍfic surface, cm'¡crrr', (ratio of the surface i area of solid pore boundaries to the bulk volume of the nedium). ' !'l L2. The equation yields poor approxim-ationsi however, for media exhibiting Second.ary porosity. Arthough snor¡7 usuarly possesses a smarl- range of pore size, it exhibits some secondary porosity or structure. Shimizu (Colbeck llf1 ¡ related the saturated permeabitity of snow to grai,n size and density by: K^-- = 1.7 x to-4a2 exp(-7.e x to-3 p-) (2.8) O\¡t - :S : where d - grain d.iameter, cln , and 3 Ps = snow density, gmfcml , Kurioì^/a (Colbeck tlOl ) expresses the permeability of snow as a function of porosity: l ç) K = 1.17 x I0-o - cm' (2.e) ow "*p(15.90) .For.a snow-keroseRe system¡ Kuriowa l'241 relates saturated con- fductivity to porosity as follows: : a N. c - o (2.10) ortr N-0 ' where C = saturated cond.uctivity cm,/sec , and O\^7-- ârN.= emperical. constants depend.ent on snow structùre. For snow-wþter systems, Moskalev (noted by de expresses j:,:- ::':: ,i Quervain tla1 ¡ i:.r::-;:.r::.: :,. saturated conductivity as a function of both porosity and. grain size by the relationship: ' i ' ::':i. : i: 13. 'ì 63 co, = 2'88 0 d*'"" cm,/sec.I (2.11) where d = mean grain diameter, mln. de: Quervain I18l obtained the following values of saturated conductivity for \,¡ater in snow. Grain Size Snow Density(gm/cm3) C-- (cmrlsec) - . O\^7 \ ;'' d < 1 mm 0.35 O.57 ': : I g.8 < d < l-.5 mm 0.38 I.20 d > 2 ¡om 0.385 2:24 Colbeck and. Davídson (summarized by Adam and lV.ilson t3l ) observed the following permeabilities during vrater percolation in ' snow. snow Dènsity (g*,/.*3) Porosity (O) k^..,(cm2t ow' : 0.653 O.32 1.2 x LO-6 -^" 0."623 0.36 3.2 { 10 : ,: : l:' r'- l' ,l 'r': .:. . . '.:,:: ::' ReLative Permeability and Satüration Rglationships ;:'.::::;r :'r. The concept of relative permeabítitv has been applied exten- ' sively in dealing with porous media systems. The relative permea- :..: bility of, the wetting phase, k.__, is defined by the ratio: .:.. . fW'::::j: .,,,,: k 1r \¡t rwk=- (2.L2). .ow 14. \.iihere effective peremabiliay, (permeability "*', of the wetting phase at a particufar saturation S), and .a saturated. permeabilíEy, cm', (permeability of the wetting phase at saturation S = 1). Similar-ly,. the relative permeability of the non-wetting phase, k---,-. is defined by the ratio rnw k k -4!¡- (2.13) / "rnw = k Qnw where k is the to the non-wettilng phase at onw ¡lermeability saturation S = O. Note that the subsctipts w and nr¡/ ref er: to the wetting and non-wettí.rg pt."." respeb¡ively. : The relationship between relative permeability and saturation is affected by the resi-dual saturation and pore-size distribution of : the porous medium. For isotropic media, the relative permeability iof the wetting'phase (dúring the drainage cycle) and the non-wettinÇ phase can be apprgximated by the Burdine equations, (r-,aliberte t25l ). '¡s ds l-l2 S-S^IPt.¿ .c tl-S' (2.14) ^r\nl 'tl rL ldst- ' o) P2 ic and dS 2 S-S ^ P. ,I' (2.1s) rnw tr - 'Se-4-ll' -S .Cr ds 2 P "f' c 15. where i S, = critical saturation, (saturation at which the non- wetting phase becomes discontinuousntinuous and whichwh cor- responds apptoximately to the saturation at the bubbling -9pressure, P,_)'. ,, Corey (noted by Laliberte t251 ) found experimentally that for relatively homogeneous and isotropic medià having an average pore- \" sizê distribution (tr= 2 appro*.), equations (2.L4) and (2.15) could ': (2.16) and Q.L7) Laliberte l25l sho\¡rs 'that "e " in the more general equation, kr\Á/. =sE -e : (2'rB) could be approximated by the relation 'Ì1 2+3^ ¡ e 'r Q.Lg) Porous media of cbmpletely uniform pore-size distribution have an exponent e = 3.0' and. for unconsolidated sands having a single grain structure, e = 3.5. ; Dealing with air-water-snor^r systems, Col eck [10] mentions the use of e = 3 in the relative permeability-effective saturation relationship (eqn.2.18): Ho\¡rever, he uses e = 2. In later pubti- .r.:1:j..-::¡:1.r i'ji.l.ri.;_-':.::..::..::.:.Ìj:..-l;l!.t:::-t:::'fr-:.i:::t.:.1::1 L6. cations [13], [15] r [16] , Colbecl< uses e = 3. Friesen lZZl obtained ! e = 3.3 and ¿ = 3.2 for snow densities of O.4O gm/cm3 and o,48 gm/c*3 in the snow-soltrol- C systems. L7. C. WATER FLOW THROUGH SNOV'] '1. An understanding of the water flow mechani.sms in snow is of particular importance in the development of predictive capability :. :: ,,i regarding avalanche formation ancl the seasonal release of mel-t ::,' water. Knowledgee of the mechanismsmechani also finds application in the: ice-capping process of snôw roads. rn recent years, general theories of f luid flow through porous media have b-een appried. --¿ to the air- \ ;,.,,:... 'r water-snow system and a theoretical basis for understanding the , ' ;::'t flpw mechanisms has been developeã. ïn thís section, frow equations i::: are reviewed and the results of pertinent studies are summarized. General. Elow Equations : Flow through porous med.ia may often be considered from a macroscopic point of view and treated as irrotational fl_ow. In additionr where the inert.ial forces are negligible as compared to I ithe viscous forces d.uring flow through an isotropic medíum, Darcy's Iaw may be applied.. The general differentiaL form of Darcy,s 'i 'equation is 'k __. un* (2.2O) u where q = \¡ohme flux , r "^3)"^2,:sec. 'a k - permeability of the fluid, cn', u=dynamicviscosit1zofthef1uid.,ry,'and .')cm P* = piezomi:tric potential, dynes)'cñ'. 18. The piezonÞtric potential, P*, repres9nts the resultant of the normal- surface and body forces per unit vol1e acting on a volume element in a fl:uid of uniform densíty. p* has the d.imensions' of energy per unit volume. If adsorptive forces are ignored , p* becomes the summation of the pressure potential P, and gravitational pote¡¡i¿1 pg". That is P*=P+pgz. (2.2L) \ ' :.::r: .j :' : ¡ i: ..,- :.:. Equation (2.2L) applies separately to the wetting and non-wetting /.''' ¡ phases. For air-water systems where the air phase is at atmospheric pressure and. static, equation (2.2O) may be re-written as '\^tk s,\^r. .= - + V(p + o-_sz) (2.22) U'w \Á¡ -vt'-' where .a cL- = volume flux of \^iater, , '.' "m'7cm'-sec k-- effective water permeability, cúr2 w = , ,'::. .:. u-.-w = dynamic viscosityLJ ofvr water,wquç! 4yls-;Jsg-- r 2 ;.'trtr".,',i.r,:t,,... l cR . ::. P: water pressure, ::::... t¡I- = dynes/cm¿', .:.:. :r.:-:-:i:":::: .j.'1 .: i' rwo = water density, Yrlt/gm/cm3 urtr , g =gravitatíonal constant, cm/sec",2 and z . = relative vertical position, positiwe up\^7ards, Çh . .. ::.:;: 1;; :.; ::':::'t:;1 : ': I,he associat.ed continuity equation (water is assurned to be incom- pressible) is: as- - .divq-=-ö-;".-w,'iiÈ e.23) :t tt i :i'tt..-': '" 'tt ' 19. where 0 = porosity, dimen.sionlesç . S water saturation, dimen.sionless, and w = t = tíme, sec , .. .. , Çombining equations (2.22) and (2.23) gives ,^ge kãs div tüv(Pw + ewsr)l = O # e.24) : :..: - \ '""' Equation (2.24) is known as the Richards equation (Laliberte t25l ). r.-: : CapíIlary pressure, as defined earlierris the preSsure dif- ,,, - I . t;_:.:j-tt;i ference across the interface of the two immiscible fl-uids. For the air-water system P =P :P (2.25) C a .\^7 where . 2 P pressure, dynes/crh-, and c = capillary D.a, Pw = pressure of the air and water phases respectively 2 (P^ gauge pressure). dlmes/cni-i '. a = constant = O ' 'i Combrnrng equations .(2.24) and (2.25) gi-ves k as div tüv(Pc - e*sz)l = - O ¡q Q.26) For one dimensional vertic.al flow, equation (2.26) becomes AS ' ^ âP .: *J=-til-Ltn'(#-Pws)l .(2'27) The Ðarc1t and continuity equations for the air phase are similar to equatíon (2.22) and (2.23) 20. Capiltarv-pressuie, Saturation Relationships As an approximation to the snow-water system, Colbeck t13l performed capillary-pressure experiments (for the d.rainage cycle) with snow and kerosenerat, a temperature of - lOoC. A fine-grained snow (approximately I mm dia.) compacted to a density of 0.56 gm/cm', vras used. The plotted resufts are shown in Figure 2.L from which several observations may be madÞ . 4000 3200 zE 2400 o øf, 'o ,Ø o- r600 .=I 'õ- oo * 800 SF o . o,2 0.4 0.6 0.8 t.o . Líquid Soturoiion ; Figure 2.I - Capillary pressure as a function of liquid.saturation (after Colbeck, S.C; [13], The capitlary effects on water percolation in homogeneous snovtr Journal of Glaciology, VoI. f3). The curve indicates a rapid desaturation which'is typical of matèrials 2r' with..a uniform pore size. Another important featu.re is the value of : residual saturation S-- = 0.07. Since usually a narrow range of pore size qccurs in snow, the relationship described.in Figure 2.I may be a good estimate for most types of sno¡¡. Colbeck t13l indi- cates however, that only a small- par! of the curve is significarit ,,,,, with regarQ to'v¡ater drainage in homogeneous snow sinie water sat- uqations generally exist in the narrow range, S = 0.1 to S = 0:2. Fríesen l22l performed capillary pressure experiments with \ ,.,, . ... .:' . ::.:: snow. and soltrol C at a temperaLure of -15oC. Residual saturations ':i_ of/ St = .::0.035 .and Sr = 0.094 weré obtained for snow densities of ,',,., t O.4O gn/cm3 and 0.48 þrnlcm3, respectively. )) D. HEAT AND MASS TRANSFER I There are three modes by which heat is transfered from one point to another: conduction" convection, and radiation. Because of the 1ow temperatureg associated with snow, the radiant heat com- ,,,.-.,,: ponent.is usually negligible (Yen l44l). Also, according to yen 1441, thermal conductiviÈy through interstitial air is often insignifi'cant. Therefore, conduction through the ice skeleton, convection, and heat \ ,.,.,.,i, ,.-,..,. transfer due to vapour diffusion are the significant,contributqrs ::: l ,' ' to/the overal-l heat flow. In dry snow a net mass transfer and re- .,,'., ': distribution of density occurs (constructive metamorphism) owing to tte processes. of sublimation,. diffusion and condensation of the water väpour afong the tempeiature gradient. r The general Fourier heat conduction equation for a medium : containing no heat sources or sinks (no sublimation, vapour tran'sfer r condensation, etc.) is: äT _2 c !" T (2.28) -=dt where i .T = temperature, oC, t = timer sec , e = K/pC^:p thermal diffusivit:y, cm2¡sec , K = thermal conductivity, cal/cm-sec-"C, p = density, gm/cm3, arrd : Cp specific heat, cal/gm-oC. For one-dimensional heat conduction, equation (2.28) reduces to: aa AT ð2r (2.2e) at. dz^¿ The heat transfer'rate in the z directíon for a particular time t, or for the steady-stàte condition, is d,escribed by the Fourier rate equation; o=-K=- dT (2.30) d,z where cf=*z heat flux in the z direct íon, caL/.*'-=." , dT temperature gradient in the z direction, dz oC/cm, and K : thermal conductivj-icy., caL/cm-sec-oC , Sulakvelidze (noted by Yen 144.J) formulated a heat-transfer equation to account for. the evaporation and condensation occur:ringl porous media contaíning vapour; water or ice at temperatures close to the transition temperatures. This was done by including an I additional term in the Fourier heat cgnduction equation. For.one- dimensional heat transfer, Sulakvelidzè's equation .is i L ðT ð2r s ct (.2 . 31) âr 2 ö.0 dz AS where ) _Ac_r.,ð-c.* or '.2 condensation, gm,/cm3-sec, .a =.vapour concentration, gn/cy , l L latent heat cal/gm, .s = of subtimation, c. = speqific heat of ice or sno\¡¡, caL/gm-"C , l_ :aji1;Í,1-.1j 24 p-.S density of snorv, g^/"^3, and .l p= mass d.iffusivity, "^2¡t"". rn dealing with snow-air systems. the concept of effective thermal conductivity. has been applied extensively. Effective thermal conductivity (designated K") represents the net effect. (exclud.ing rad.iant transfer) of all the mechanisms acting in the heat flow. A general expression defining K^ e is: g i K" 'grad T (2.32) 'J where q heat f1ux, caL/cmz-sec r :car/cm=sec-bc' 'eK conductiv íLv' and T ,-".r"."r.:":::t Measureme4ts of effective thermal conductivity have been obtained by several invesLigators. where the results have been. rempirically correlated.to structural properties, slrot¡r density, was the sole parametets. Yen [44] summarized the results of severar inves- i tigations shov¡.n in Table 2.1'. .(p^.S = snow density, 9rn7cm3¡. TABLE 2.1 Effective Thermal Conductivity (K.) of Dry Snow ri 'r, . fnvestigator Empirical expression Density range Abels 0.0068p"2 0.14 < p"< 0.34 Jansson 0.00005 * 0.0019p, * 0.006p"a 0.08 (after Yen, Y.C. 1441, Recent studies on snow properties, Advánces in Hydroscience, Academic press, New Voif) The significant heat transfer mechanisms in dense sno\^/ are conduction through the ice skeleton and pore : the spâce;. The effective ,,.,,,.,,, thermalconduct'ivityofdensesnowisexpressedbySchwerdtfeger (Yen L44l)as, 2Ks + :Ku- 2 0 (Ks - *.) ,' , ,... \ ,'j K.-e-2Ks+Ka+ô(Ks-Ka) ^sK (2.33) ".' where.i 'e K = effective thermal conductivity, caL/cm-sec-oC t : ' :..' .)a,v Ks = thermal conductivity of air and snov¡ respect- ively , cal,/cm-sec-"C t ,ó I r = f - @=/Or): porosity of sno\¡¡, dimensionless, and -.3 g_, p. = density of snow and resBectively, gm/cm . '.s' ]- ice . For a liquid saturated porous medium, Somerton (noted by iAdam and' !{ilson. t3l ) proposed'the expression K- : : 1¿' co K.re = ..sK^ 'j'rK(;-) / (2.34) ,S wher:e K^ e = .effective thermal conductiví1..y, caL/cm-sec-oC. K K, : thermal conductivity . s', l- of the solid and tiquid respectively, cal/em-sec-oC, C = empirically derived correction (= 1.0) 26. ' In the.case of a flowing liquid phase, the heat transfer ij process is comp.licated by convection. Based. on the comprehensive review by Adam and Vüilson [3], the problem may be formulated by the heat balance eqgation: (2.3s) where K = thermal cond.uctivity I cal/cm-sec-oC, : I .v - fluid flux, cm/sec t I ? C..\^¡ = vol-umetric heat capacity of water,. cal,/cmr-dçi ae u"a","'iaT = n - r.^ ----: ' "apparent" volumetric heat capacity of the system, .cal,/cmt-oar .r c = volurnetric heat capacity, cal/cm3-oç, L = latent heat of fusion, caL/gm , p. = ice densi5y, gm/cm3 , arrd l_ tl- 0. = volumetric ice content, r*3,r"*3. ; 27. E. SNOW MECHANTCS Snow mechanics is a relatively new subject of scientific- technical'research. Conventionally, the study of snow mechanics has been based on analogy to soil.mechanics; however, dif,ficuLties arise due to the strong time and temperature dependence of many snor^/ properties. : Bader t6l and Mellor [3O] provide an extensive review of \ :;1,,,,,;, the mechanical- properties - of snov¡. Their review forms the basis of .,,,',',1 the fotlowing summary. '.;..:.: ( :':-:: I Densification Undisturbed snow which has undergone destrgctive metamor- phism (a few days following initial deposition) has a density l? usually above O.I5 gm/cm' and a grain size typically of al¡out I mm. The struct-ure of t'his low density dry snow can be conceptualized as a packing of granufes with some of the grains removéd. Thus there are relatively l'arge pore spaces into which grains can move yhen the structure is defo.rmed. Densification occurs to shearing i.,,,,,,,.,,. {ue ' :_ t,-._.t-- of the contact areas without deformation of the individual grains. ,., . :...-':': : At a densit¡z of around 0.5 gmrlcm', th. condition of close'packing ,' , ' r is approached and further densj-fication is not possibfe without grain defonqralion. The rounded grains (due to destructive metamor- phism) achieve relatively small contact areas and thus contribute : ': tg Low strengt-tr, Constructive metamorphism further contributes to a decrease in'strength.- of the lower layers. 28. By reconstituting snow (mechanicalty disaggregating, milling, mixinþ, etc), a densitlr increase accomprnr"U by a significant in- crease in strength is achieved. The increase in strength is due to the growth in conlact area promoted by the more intimàte contact of fractured grains of different temperatures. Response to Rapid Loading A well bonded, dense snow behav€s as a compressibl-e vis- coelastic material, That is, for a very short time interval it can res¡iond elastical-ly under moderate load, with sträins r^¡hich are ' proportionaL to stress and which are recoverable on re-movat of the l-oad. Vfith sustained loads, hovever, deformation is largely viscoplastic and is. irrecoverabl-e. The reLationship between .. and density (Greenland snow) was investigated by lounøls.mcjdulus Bentley, Lee and Nakalra (Mellor t3ol) and is shor¿n in Figure 2.2. Ê ø,uG o .(/, f )J oo4 E I zI oÒf o.4 o.6 DENSITY (g/cmõ) Figure 2.2.. - Rel-ationship between dynamic Youngts modul-us and snow density for a range of Greenland snow types at -9oC. (4fter MeIIor [3,0], Pofar snor¡r - A summarlz, of. Engineering Properties, MIT Press, 1963) 29. For the naturally compacted snows of Figure 2.2, Youngrs mod.ulus (E) ., ! follows the relationship 62 E = 6.3 x I0' exp 14.6 Q dyne,/cm-, (2.36) '? for the density range O.27 < p < O-5 qm/cm', and (I6.4p s = - 7.2o) " t010 dyne/cm2, (2.37) 2 \ --'r. \ ,, i for the density range 0.5 < p < 0.9 gm/cm". It was also found ',, : :t': :_ : ' that for a given density, E tended to increase with a decrease j-n .: ' r:.: .t :r : i ' temperature and decrease with grain size. The inalrility at present, : : to assign numerical val-ues to textural characteristícs malces *^ .i The investigations in- ^e^^-fGreenland ^-; by Bentley, Pomeroy and j i Dorman (Mellor t30J) have shown that Poisson's ratio varies very I little with changes in density. For snow with density between 0.4 i?. i and 0.7 gm/cm", Poisson's ratio is given by: r '! I' rI , v = 0.15p + O.2 (2.38) l | . :_ I : Wfl€fê . :.... ,,.i,. i v = poissonrs ratio, cm,/cm, and i:r, I . 1: p = snow density, gm/cm3. The tests performed by Lee on high-density processed sn'ow (noted by Mellor t30l) are in agreement with equation (2.38). i Butkovich, Ramseier and Gow carried out extensive inves- tigation into the strength of snow. The relationships between 30- kg/cmz 6 10 ts= Figure 2-3 - Unconfined compressive z U E strenþth versus Figure 2.4 . Tensíle strength of high density snow at a temper- f aqure of (-10oC has been ì- -]OoC 0 adopted by the U.S. Army Cold Uz É ¡- Regions .Research and Engínegring ØtO u- Laboratory as the standard test J õ temperature) . z U F o. o,? .o.4 0.6 0.8 oE NSITY (g/cm3) psi \g 1cm" 2Qc -UNCONFTNED Figure 2.5.- Strength in double I shear plotted against density for I TI the unconfined case and for ul9l normal pressures of 30 and 60 psi. tl :lF tol- (Temperature - I0eC) 'lÉl I oL DENSITY (9/cm3 ) l-.r - r' . 'Many attempts have been made to apply the Coulomb equation sno\^/. Thè Coulomb equationr.as applied in soil mechanics rel-ates shear strength with normal pressu.re in the linear form; i:l ìr ':l 3I. s=c+ptanQ (2.3e) \^7here s - shear strength, psi, P = normal pressure, PSi, c = coefficient of cohesion, psi,. and 0 = angle of shearing resistánce. In the case .of sno\4r., the plot of shear strength against normal pressure is not linear. Ho\,veverf in studies of oversnow vehicles (Digmond 119.1, Nuttall, Thomson 1341, et a1) where. normal pressures lie in a narror¡/ range (Q to 10 psi), the portion of the curve is often approximated with a straight line. In this connection, studies made over a wide range of surface sno\^¡ densities and tem- peratures indicate variations in 'c' ranging from O to 1.6 psi and vaLues of tÖt ranging frorn 2l.o to 55" Response to Sustained Load.ing ! In most construction materials, a detectable rate of creep is often considered to be a failure. In sno\^r, where rélatively high creep rates in engineering practice are tolerable, failure is identified.with collapse rather than with creep. The .strain raie due to sustained J-òading does not bear a l-ine:ar dependence on.stress but rather exhibits an exponential type of relationship. It has been found that the creep rate of snow can be represented by the hyperbolic sine function; ¡o Ê=Asinh- (2.40) o o J¿.aa where e = strain rate, sec-l -, 2t o = stress, gm/cm', (wheie stress is expressed in mass units, the force associated with the i mass is implied), o-o: = à constant havingr'.i hñ the+l^^ stressô+e^^^ units,"-ì +^ and^*J A = A val-ue. dependent on the snow density. ,,'1, The creep rate is hi.ghly dependent on density and the value of 1A' in equation (2.4A) varies as the snow is strained. Bader [6] indicates ?. \. :-:. that when stresses are always above 800 grm/cm-, equation (2.4o) may be ,.,' simplified to :t t:,' / : r: I -o . e=o_-o sinh: (2.4L) .o Õ At a 10? error }imit, equation (2.4L) may be further.simptified to o rOO e= ¿:o^ exp-t- . Q.42') .o where .2 o^: 700 gmlcm-.. 0 = '2 Irihere stresses do not .exceed 600 grm/cm' , a ð.rrect proportionalitlr be- tween stress and'strain rate mal!¡ be assumed. ; Creqp rate can also be expressed as an exponentiaf of the density by the relationship; -b(o-I - o'0 ) el=eOe ( 2'43) : where -I e0 = strain rate at time t = O r sec - -t .l_;. = strain rate at some later time t = I, =".-f, ' g^ro' p.I =snowdensities att=0 andt=l respectively, 3 grmlcm , and . ?? = a parameter which behaves nearly.as a constant wiÈh a value of 21-.05 most fre{uently found. It is believed that 'b'. is dependent somewhat on snow type, time sequences'and temperature. The range of creep rate within which the relationship of equation (2.43) I is valid ii: not known. Bader t6l indicates that the theory of equations '(2.4O) and (2.43).U for which o^ = 70O gmlcm- and b = 21.05r faifs at : stresses (o) ttrat are greater than 5 kg/cm2 and at densities (p) greater e. than O.84 gm/cm' . It is interesting to note that at these densitíes a :i!,i i a 1.,:.:a.a . ..,. ¡ ::.: .:. l-Íi:::,aj:' i 34. F. ]NDEX PROPERTTES '! Field identification and classification of snow type for any given purpose is not a.wel-1-developed. subject. Although no standard for field investigation exists, review of published material suggests a consensus that at l-east the determination of density, grain síze, graín shape, temperature and hardness be the minimum criterion. These properties will-be re.ferred to heneeforward, as the index properti.es of snow. \' oeúsity ? The density (mass per unit volurne , gm,/cm') is perhaps the most significant index property of snow; The average density is usually determined by simpty weighing a known vol-ume of snow, or melting a known volume of snow and measuring the vol-ume of melt water. St;ìn- less stqet tubes are generally used to obtain the snow sample. I{ore sophisticated te.chniques such as radiation scanning and attenuat.ion .measuremenL are used in more precise laboratory work. Bader 12) notes that for the purpose of correlation with other properties, the deterrirination of d.".r"ity to two significant .figures i.s often ".no*' insufficient. Grain Size !{ith reference to grain size, snow is usually described qualr-tatively as being either fine, medium, coarse', or very coarse- grained. A system of classification, presented below in Table 2.2, l:¡:.:'-. i.,' .-. l.- ' -: 35. was proposed by schaefer, K1êin, and deeuervain. The predominating grain diameter iF deternúned by d.irect ínspection with the aid of a hand lens and plate having a l-mm grid. TABLE.? 2.2 now Classifícation Predominating grain Ðesignation diameter (mm) < 0.5 very fine grained (vfg) 0.5 to 1.0 fine grained (fS) 1.0 to 2 .0 med.ium grained (mg) > 2.O to 4.0 coarse grained (çS) '4.O very coarse grained. (vcg) (after Bader [6], The physics and rechanics of snow as a material, CRREL Research Report, 1962) ,,Dependingonthenatureoftheinvestigatíon,.9rainSizemay be measured'under a microscope or by screening or elutriation. lfhen .:. .. ..: :. . .; ::-':':-t' '' snow is screened into fractions, Bader [6] proposeà the following ::' . f:..:.:...:...: definition of mean grain diameter r '":: : d = /-l,lm (2.44) where d.= mean grain diameter of a fract.ion, mm , M = side length of'square mesh openings through which the fraction passes, . mm . and i "-!. 36, m = sid.e length of square mesh ope.nings on which the .! f.raction is retained, mm. The mean grain diameter of the mixture is:. n I (I^r. d. ) II t=l .d sn (2.45) x (vr. ) '.1-=l-l_ where , ds = mean grain diameter of the mixture, fltm., I n = number of fractions, di = niean grain diameter of ith fraction, mm , and wi =lt.ight of íth fraction, g*/.m3. Grain Shape A large.variety of grain shapes occur in freshly deposited sno\¡¡. The reason being that the snow flake sh¿pe flepends on the /atmospheric conditions in which the ice crystals grow. New snow is :commonllt classified; stellar,. column, plate, needle, gapped column, or spacial dendrite, depending on the geometric featuies of the snow flakès. Ttreqe distinct features., however, soon disappear following i the initial deposition. Grain shape is observed. generally to determine the stage of metamorphism lidentification of depth hoar, etc.) and for decidíng . it:.:.::1.,, : whether a give¡r dry snow has or has not been previously.wet: 37. Temperature In the classification of materials, temperature is not usually considered to be an index property. However, since so many properties of snow are highly temperature dependent, thmperature must be specified. Snow can exist at temperatures < OoC . Hardness ,'.Evaluation of the strength parameters by direct means is usually laborious and timeaconsuming, and hence of timited utility in ' thé field. Attempts have been made to devise indirect methods which l pr:ovide a quantitative correlation to the funad.mental strength para- meters. These indirect method.s are generatly based on some form of penetration test; r Perhaps the most popular of the field testing proced.ures is the de.te.rmination of a hardness number by means of the Rammsonde con-.e penetrometer. Ihe.measure of "hardness" (kg) is based on the pene- a tcone under an impact of known energy (discussed in /tration ,of .Appendix C). The Rammsonde instrument is used in the study of'un- 'I disturbed sno\Âr covers as well as the compacted snow surfaces of snow roads and runways. The Rarnmsonde (Rarn) hardness necessafy for supporting various wheel loads and tire Þressures for a specified number of wheel coverages (Figure 2:6) has been determined by Wuori- ['431 . The hardness values i:..,.:,:,::: :rt :t:i';t' i:" '¡ represent the lowest hardness permissible in the top 15 cm of the prepared'Sno\^7surface:Anadditiona1requirementisthattheunder- 38. Iying layers have a hardness of at least 75e" of the. surface layer. The Rammsonde hardness of a snow pavement. has also been correlated to the unconfined compressive strength. Clark, Abele, and Wuori [9] presenÈ a nomograph (Figure 2.7) in which the un- confined compressive strength is related to various wheel loads, contact pressurps and number-of wheel coverages. The use of the BCC r 7AC èF 600 o 500 o 400 ,1, g o 30o '@ o ¿oò - 3 Þ-õ3 Figure 2.62 Required ram hardness for supporting various wheel l-oads. The abcissa índicates wheel loads at various tire inflation pressures. (after Wtrori [43], Supporting capacity of processed snow ruru^rays, ccR.REL, 1962) Rairmsonde penetrometer in conjunction with a prepared nomograph, as shown .in Figure 2.7, facifitates guality control during construction of snow pavements as well as the'monitoring of perfortnance, during operatron 39. ; :$ ,r{. 5 li< --*l ."91 l,l.),l ì' -r" çll" l' J '"{5""^*,1'-:,,r1 fw'\lrb't". t.. I \\\\\ I .çr-c'f'í"t 1', rr.J,L,ð"* I Figure 2.7 z Requíred hardness (or strength) of a snow pavement for various wheel load conditions. Examples in the use of.the nomograph are shown for various ãircraft. (after Cl-ark, Abel-e, and lVuori [9], Expedient sn:\^r arrstrip construction technique, CRREL, 1973) In Figure 2.8, deQuervain (noted by Martinelli ¡ZeJ), illustrates a good correlatión betweer rpm resistance (hardness) ånd the tensile strength'of new snor¡/. The results ,indicate age hardening, änd show 'the tensile strength fo¡: snow four (4) days old or older to be =; 1.5 times greater than the tensile strength of younger snow with the same r¿ìm resistance. Age hardeníng in snow is highty temperature dependent, however no reference to temperature history is made. . .40. 2000 ,+ looo I log (tensile) = 1.69317 + 0.658tB log (rom) f. r = O.93 og.e > 4 doys 500 log (lensile) = 1.45416 + O.7Bl09 log (rom) (r¡ or+ I r= o.9l E() @ G \, r> qt g, / at ctJ) e roo' 3c' log (lensile) 1.20818 O.A9237 log (rom = @ a4 < ,/ r = O.95 oge 3 doys ø)// 50 Nu.mbers ore oge of snow (doys ) $ lnitiol hord slob o Typicol og.od snow À Persislenl sofl snow Romrnumbers of I cnd strength. volues < l5 excludsd f rom regressions. Q¿ l^ = correlacion coefticient. to 5to 50 t00 Rom resislonco (kg) Figure 2.8: Average tensile strength as a funct.ion of ram resistance and age (after deQuervain, Martinelli 1291, Physical properties of alpine sno\^r as related to weather and aväIanche conditions, U.S. Department of Agriculture,, Forest Servicq, 1971) 4L. A number of investigators have correlated Ra¡runsonde hardness to snow density. Martinelli 1291 summarized the regression and cor- relation coefficients for several sets of data fitted to the regression logR=a+bg (2.46) in Table 2.3 where R = rammsonde hardness'number, kg, and p = snou/. density , krJ/^3 ' : .. \ :.::.:,.. ABLE 2.3 l -1 1:.. i 1 :_'.: i . Summary of'Regression and Correlation Coefficients Correlation Age of Denbici' Source a .b ccefíicient sn-or*r range (r) -'t tss--E: rBull (1956) -0.6107" 0;00531 0.80 Keeler & h'eeks - .S44ó ,C0640 .9!+ <4 rnonEhs lO0 - 5i.0 (Le 67 ) Keeler (1968) - .7482 .,.0C599 .89 <4 nonths 150 - 430 : .I I - .428 .00543 .84 <14 days 40 - 4-<0 '- z .305 .00343 .54 Èervice, L}TI) : ' r' l:i.:..-.- ' Polar sno\^r of Greenland was studied by Bull (1956) and dry seasonaL SnowsofMontanabyKeeIerand!{eeks(Ig6],t968).A1pinesnow '.. (source 1) and sno\^7 from a sheltered opening'near source I (source 2, 3), were studied by Martinellí. No reference to temperature is made 42- in these studies. Ager 141 studied the influence df temperature on the density and hardness of snow in the temperature range from -loC to - 23"C . He found that for a given compactive effort, highest densities were achieved _.r -.t,. : at - loC. Subsequently, highest hardness values were obtained when the : '.r.j'__, : -.t , compacted sample vTas cooled to - 23oC. He also found that the density and hardness \^ras .much more sensitive to temperature variation in fine- grained snow coarse-grained snoh¡. ; ' than.in \' r:':"t:"' ": The studies of Abele, !ûuori, Clark, €t. al.r.suggest that ''': correlation'of snow hardness to- temperature (at particular .d.ensíty) ,' .' '.. is perhaps .a mþ.¡e meaningful approach to the density, hardness, temperature interrelationship of sno\^¡. ^) CHAPTER IIT VIINTER ROADS J The idea of àompacting snow as a means of improving oversnow trafficabitity, is extremely old. The snow road of the horse and sled age was compacted by.primitive'me.ans - today, accomodatión of heavy-wheeled traffic . necessitates improved methods in the construc- tíon of snoÌ^r pavements Many studies. directed towards obtaining a better traffic foundation, have been car¡ried out. The literature prodtrced by a I I number of these investigations, is reviewed. in .Chapter IIf. Salient observations and conclusions are cited ín order that a proced.ural l guide, foF the construction of snow pavements, frây be d.eveloped. Sysiem of Çlassification To dispel any confusion_ regarding the terminology used in this 7tudy, a complete classification of winter roads i's presented. The system of classificatíon proposed by adam [1], is adopted herein. Winter Trail - "a trail used gnly in winter between ireeze-up and break-up, and which is established by a single pass of a wheeled or tracked 't¡ehicle using a 'blade' if necessary to gain access. " Snow Road - "a road built primarily by using s¡row as cut and fill material to establish some resemblance to a construct- ed road grade. Compaction is part of the construction proced.ure. . IA srrow road can. be further classified by tprocessed' .i whether the snow is agitated or before 44- compaction: l i) _Compacted Snow Road the snow is compaôted without tBrocessingt. ii) Procêssed Snow Road - the snow is processed before compaction. " Ice Road - "a snow road with the physical addition of water to form a bonding agent, between snow particles in order to give .:. . t added stabilíty to the roadway itself. Ice roads are " ¡1"-' similar to snow roa Ice. Brid.ge . - "an artificially thickened ice cover that provides the required weight capacity at rdver. crossings or other þodies of water." Route Selection the main thrust of.this: study, the, importance of the subject justifies a few brief comments; As pointed out by Adam ll,2f, performancq evaluation of winter roads must include trafficability:aspects as well as environmental'considerations. Trafficabitity can be improved, and '', operatíon and maintenance costs reduced, if shallow gradients and fl-at curvatures are maintaineã. AIso, the degradation of underlying perma- r : ::: ,,. . : i.. i' t:.: ' 45' : frost in permafrost areas, and drainage and erosíon probrems may be ' minimized or avoided through pr'oper foutp sel-ection Measures.to Increase the Bearing Capacity of Snow The successful construction of winter roads requires a .::..:r. -.: :- t.I -l knowledge of the basiô mechanisms by which the strength of snow increases. To gain a qualitative perspective, the mechanisms are ' reviewed. l t""' \ 'tt Studies by Ager, Clark, Abele, Wuori, et al , have shown that- ' .",,"', snow strength íncreases with an inc.rease in density and a decrease in ltemperature. In addition, bond growth occurring between the individua.l snow .grains (age hardening or sintering) , makes a signif- icdnt contribution to strength increase It hps bçel found that by,nechanically agitating (processing) , the snow, a.more non-uniform particle size distríbution is produced. This atrlows (upon compaction) more intimate grain-to-grain contact which results in an increase in density and an acceleration of the .l àge hardening process. Ttre disaggregation and compaction of snow is most effeetive if performed at higher temperatures (below freezing), however after that, fower temperatures are desirable to achieve high strength. The age-har'dening of snow is highly temperature depend.ent. Hardening rates are higher at higher temperatures (below freezíng), however snow which .Ls age-hardened at a lower'temperature will. ultimately reach a hígher strength ì 46- A. SNO!{ ROAD CONSTRUCTTON ! snow roads are basically constructed of compacted in-place snov¡ or compacted processed snow. The avai.labl-e literature d.escri-bing the construction experience is extensive r äs is the variation in ,'.'., construction teshnique and construction equipment used. An outline of the basic steps in snow road construction is presented forlowed by a'review (limit-ed,to a selèctive cross-section of the líterature) \ ,:.:: ': ' of construction. methods and results. :': ';/ À thorough literature research (wínter roads, einvironmental. ,'.', impLicationsr etc.) and catalogue of abstracts is pres_ented by Adam [I] l. Initial Site preparàtion FoLlowing the first- major snowfall of the seasonf an initial dragging operation is performed. A simple wooden drag is towed by a low-pressure tracked vehícle to promote frost penetration and to smooth out minor surface irregularities. V'then a sufficient sno\^r cover has accumulated, and the undertrying ground is sufficiently frozen,. snow road construction may be started. In permafrost regions (where this preparatory stage is 'especially applicable) caution must be exercised not to remove too much cover vegetation and thus cause degradation of the underlying . permafrost in subsequent years. . 2. Disaggregation/Processing There is gene,ral agreement ín the literature that compaction 47- alone, on in-pl-ace snow, will not be sufficient to achieve the strength required of snow surfaces subjected to intensive wheeled traffic. That isr.the effects of compaction al-one are usually too limited in depth to provide a sufficiently thick snor¡/ pavement. Depth processing is there,fore required and the types of equipment used for this purpose have been rotary earth tillers, versions of the farm. harrow, rotary snow plows, and pulvimixers. \ ,:: 3. Leveling - Compacting ¡ For maximum effectiveness, the compaction should be performed immediately after disaggregation, that is, before the prgcessed snow has a chance to rset-up' (age harden). Any levelinj required is performed prigr to or during the compaction. . r The leveling is usually performed by towing a sieel or wooden : . dfag. The drag rnay be a simple frame-like apparatus fixed. with . cutting edges or A ballasted pontoon-type assembly. Depending on the weight and design. the drag may serve a dual purpose of leveLing and I com.pacting. Where major leveling of irregularities or shapinÇ,is ,' . required, backblading with a bulldozer is performed. ' ComBaction is'usually performed by dragging and/or rolling ,with large diameter (8 or 10 ft) rol-trers. An additional compactive effort is provided by the tracks of the tow vehicle. Vibratory compaction equipment have also been studied, however, this type of equipment is not extensivety used. 4. PrepaTing ,Fina1 Surface The finat grading is usually performed using a finishing drag 48. or snor^¡ pl-ane, followed by a final rolling. The finishing drag or J snow plane is fixed with adjustable cutting blades permitting a smoother surface to be aqhíeved. This tlpe of equipment is pre- dominantly used in snow airstrip construction. rf the roadway is sufficiently age hardened, the ccnventional road grader may be used to remove high spotg and fill in depressions. An effort must be made to achieve as smooth a surface as þossibl-e during the initial compaction as the low spots filred by the grader, at this stage,.generally remain weaker even after the final rolling i lce-capping of snow road.s, as a means of achieving a foad surface cap'a-ble of withstanding â high intensíty of heawy wheeled traffic, has been stud.ied. Ttre ice-capped snow road'is a processed and compacteä snowbed which has water appried to the surface to in- crease its density and strength. Methods of increasing road strength and surface durability, by heat apprication during d.isaggregation and surface finishing, have also been investigated. 1. j i,.l:-l 49. : B. CONSTRUCTTON METHODS AND RESULTS The Scandinavian Experience Ager t5l discusses the snow compaction tests carried out in Lycksele (1956), by the Forestry Socíety and Royal Domain Admini- . stration (S.D.A.) of Sweden. Although the methods described are applicable only to horqe and tracked vehicle roads, the tests iden- .; tif,y the significance of snovr processing in snow road construction. The resuits of the tests were compared to the results produced by the traditional method of snow road construction, where the snow is I first processed throughout the roadbed, with a deeply peneÈrating track-type tractor, and then dragged. . The followi ng equipment was used in the tests: a.wooden .) drag.(-3 bl-ades) having a contact pressure of 0.O35 kg/cm1 ¡ a rail-- roller'(open type) having an:apprîoximate weight of 450 kg per I.5 n length, a.closed roll-er which was constructed by covering the rail- rol'Ier with corrugated sheet metal, a pontoon drag 2 m long and l weighing 180 kg, a rotary tiller, and an Oliver OC 3 tractor equip- ped with standard tracks. { The test results, for various combinations of the equipment, compaçed as follows. Roads constructed by a single snow processing, : using the rail-roller, followed by a single compaction pass'wÍth the'pontoon drag or. the closed roller, were equívalent in quality to the roads constructed by the traditional method. A double processing with the rail-rollerr followed by a single cornpaction with the pontoon drag or closed roller, produced roads with hard- , 50. ness (strength) values approximately 40% greater. A double processing with the rotarl¡ till-er, however., followe'b ry r single compaction with the pontoon drag, yielded considerably higher bearing strength val-ues . rience in the con- struction of snow roads lntended for horse:drawn and tracked. .vehic1etraffic.Agricu1tura]'equipmentsuchasthespringharrow, .l the disc harrow and the rotary tiller were :: ' used to process the sno\¡¡. \ ',.., The, best processing. results were achieved using a modified rotary , :: :.:l: l ti{-er where the original rigid tínes wäre replaced by chain fl-ails. i I,eijonhufvud.a1sopointsoutthesuperiordeptheffectproducedby rolrer compaclion qs compared to drag compaction. The d.rag, however, . . ] prod.uces a harder snow surface due to the crushing and grindir,rg of i , the snow p4rticies in the sùrface layer. i The.conditions necessaryân r forf ¡r hard.eninghar¡lani n- ofÀ€ a- snów--Á'.' ^^.'cover may vary depending on the type of snow and on the. temperature.. To . achieve a desirable hardness in snow roads. a fine grained snow . i l compacted to a high density is initialty required. The fusion of ' ''' : : ' .-.-:..i snow particles (sintering), however, is dependent on the avail-- ',". ";l ' ,.", ability of free water or water vapour. Agier [5] carried out experi- ,;-,,,, ments on the hardening process of compacted snow duiing which he insulated'the road bed either from the atmosphege or the ground 'He surface or both. found that a newly compacted snow road hardens : :: : ::i:;1..:j. -upward from the bot,tom independentry of the prevailing weather conditions, but the hardening of the'surface rayers depends largely on the low temperatures of the air and on the penetration of fow 51. temperatures deep into the snow. rt has been found that hardening whiah starts from the bottom is faster on peat soils than on mineral soils (Putkisto t3S1 ¡. Putkisto t35l states that the direct effect of the relative humidity of the a'ir on the hard.ening of snow is probably compar- itively smatl. 'However, an exceptionally low relative humidity may cause intense sublimation of snow, softening the upper layers and restraíning the hardening process. It i's also pointed out that an r :. accr:rnulation of. fresh snow on a snow road ínsulates the road bed from the¡cold atmosphere and softening from below will occur. As indicated by Ager [5], Putkisto [35] and Eriksson l"2ll, the afternoon is generalJ-y the most favourabl-e time for compacting sno\^r. .During this time of ,day, the snow has a relatively high temperature and thus has sufficient water vapour between the snow partictes. AIso, during the níght when the temperature drops, freezing will occur. Putkisto t35l considers the most favourable meterological conditions for snow road construction'to occur irnmed- I iately following. the cessation of low-pressure periods. That is, the cool air currents and lack of precipitation of high-pressure periods are most desirable. Furthermore, the construction should proceed d.uring eally wÍnte.r as the size of snow particLes (as pointed out in Chapter II), is more advantageous as regards to disaggre- gation, compaction and hardening. : The next sectioh deals with the North American experience where a much heavier variety of equipment is used to construct snow pavements capa.ble of withstanding a high intensity of wheeled traffic'. It will be noted:that the basic. construction techníques and variables remain the same. l:'i::j, 52. TLre North American Experience 1.' Camp Halen Colorado ! In 1950, the U.S. Bureau of Yards and Docks carried out a series of.tests on depth processing and compacting of snow, at Camp Hale, Coilorado {Urt. Navy Tech. Pub. t40l). During' the test period., (fate January, 1950, to mid-March, 1951), the snow depths and ter.nperature data vrere as follows: average snow depth l-7 in; \i maximurç snovi depth 30 in; . { average temperature IT)E¡ maximum temperature 64"F¡ and minimum temperâture -28oF. The following equipment was used: 1) 8-ft:diameter, 8-ft-wide, A-Lon, hollow-shel1, fixed-,fâÇê, corrugated-stee1 roller; 2) pontoon barge drag., weighted to 7 tons with I bearing pressure of 0.88 psi i , 3) Seaman's 6-ft-wide pulvimixer; ' 4) disc harrowi' and : 5) 250,000-Btu/hr gasoline heater with a bearing pressure of 0.2 psi. . : . Based on the snow road. densities obtäined, the various techniques tried, rank in the following order. .1. Drag, pulvimixer, roller .-. 2. Roller, disc harrow, drag 3. Roll-er 53. 4. Pulvimixer, roller 5 - Rol-Ier 6. Pulvimixer, drag 7. Pulvimixer a .:- : ':i : 8. Pulvimixer, heater ¡,,¡.,,,,. 9. , Drag harrow 10. bisc harrow ': sections éonstructed by methods 1. and 2. withstood :; .::' - \ 'l'' only limited traffic of a 4-ton 6 x 6 Army \n/recking truck. The .:l:-. othçr test sections were not traffic-tested. rr'r:i:' T4e report conchides that a multiple-pass procedure is more effectivethanthesing1e-pasSforincreasingthedensityandin.most cases for increãsing the hardness. The concl-usions aiso state that i the multiple-pass procedure beyond the third or fourth pass is im- practical. It is inferred that doùble or triple passes of the equip- i i ment. performed were on the I0 test sections referred to. I 2. Point Barrow, Alaska : ' 'theI95O-51winterSeaSon,seve'ra1'teSt]]StÏipswere built using ,:"'., ' various bombinations of snow stabilization equipment at ::' . ,'.i, .'l Point Barrow, Alaska. The U.S. Navy Technical Publication t40l gives 'account of the 4 most effective combinations tested. The equipment . and method.s used, and the r."lrìa" obtained from th'e 4 test roads: :.-: .:... alle suInma rlzed : '": :' a Seaman's, 6-ft-wide pulvimixer and immediately fôllowed þy an 8-ft-dianeter,8-ft-wide,.4-ton,'ho11ow:-shel1.fixed'-fa'ce,corrugated steetr roll-er. fÍve pâsses of this combination, at 2-hour intervals, l,i''l,,.;; 54. was performed.. The pulvimixer operated with ease, cutting about 12 in into the snow on each pass, and the rollèr, exhibitíng a hígh diameter to weight ratio, rolled freely without plowing. .Subse- quent tests indicated that the density of the entíre depth of snow increased uniformly with no stratification. Hardness values, however, are not gíven,. The road section, subjected to a traffic test, withstood 20 continuous trips with a Jeep and l-0 with an empty GMC 2 L/2-Lon, 6 x 6 cargo truck. The resulting minor surface deterioration was easillr repaired by a single pass of a light fixed-screed steel drag. I A low-bed trailer, with a gross weight of 26 tons was then towed over the roadway, by a Caterpiller Model-f2 motor gradef. This caused shallow tire depre.ssions along the fine of travel whích could be removed quickly byl . pass of the drag and roller. Section If - The equipment combination used.in this Èoad section was the pulvimixer, immediately follow.d þy a surface heater and sheepsfoot roller. As ín Section ï, 5 passes'of the equipment i combination were performed at 2-hour interval-s. The test results showed a marked increase in the': density of the entire snow cover 'over'that obtained in Section I, with the greateqt increase near Èhe . surface. When subjected to the traffic test, however, 10 trips with the Jeep and 10 with the empty GMc 2 L./2-Lon', 6 x 6 truck necessi- tated a maj-ntenance pass with the steel snow drag. A single pass over the. section with the motor grader and foadeð,26-ton-gross low- bed trailer causedsed vervvery shallow tire marks on thet road. Firm conclusions regarding the bènefits of using a surface heater.ì wei:e not possible wíth .the data available. The study points out, howevòr, that ,the method of surface heating used on Section II ís very ineffi-cient because of the large'quantity of heat lost to the atmosphere- Also, the surface heater towing speed, being approx- imately 0.85 mptr (other equipment, 1.5 mph), increases overall- construction time section,rrr - The sno\¡r v¡as processed with the puivimixer, fbllowed by a water carrier and the 8-ft rorler. Five passes of the equipment cornbination were made at a 15-hr interval between passes. I :t-t.-... I-. : The water car.rier moved at a speed of l.o mph, discharging water at the, rate of 50 9a1l1in ft (0.5 tblsq ft ), through a spray bar t 8 ft long and 2 L/2 in. ín.diameter. To prevent adhqrence of,.wet .snow to the roller,.a 3O-minute intervar was arrowed between water apprication and rotling. The surface density of the finished road- \^ray was very high,. however for the, application raté used, the water penetration r^¡as not complete, and the lgwer layers remained at a low density. I The study concruded that although the water apprication had some effect, the test section wäS not significantly better than the one built using only the pufvÉmixer and rol-ler (seetion r). rt was also observed that in regions such as the Arctic plains, iack of adequate wâter supplies wourd make the water application technique infeasible. As regards to trâfficabitity, high ambient temperatures I prevented adequate traffic testing of Section III Section IV - The snow was processed with the pulvimixer. forlowed immediatery by á pontoon barge drag with a 6,300-1b 1oad. I'ive passes of the equipment combination v¡ere performed at a 24-hour 56. interval between passes. rt was found that the density írrcreased uniformly with each additional pass,. arthough the actual values are not given Due to high ambient temperatures the traffic tests were inconclusive. rtre study pointed out, however, that the pontoon barge drag was not satisfactory as a piece of snow stabilization or.maintenance equipment because of its butk and. lack of man- 'l euverability. . The Point Barro+w Studies concluded; ,,a combínation of ag- itatíon and rolting appears to be the most practical and effectirr. I means of stabilizing a snow'road on the Arctic coastal plain". Furthermore¡ ...."!he best method of maintaining a snow'road is with a veighted steel or: wood f¡ame .drag. Either drag followed by the 8-ft:roller is very effective in restoring the surface to good condition". 3. Churchill. Manitoba '! i In 19.51, a l2-mite section of snow road, 45 miles from Chlrrchil1r Manitoba and Z mif;s from the shore of Hudsgn Bay was I Engineers (U.S. Navlr Teöh. Puþ. t40l). In this, reøion, high winds are common, and ¿ the snow conditions are described..to be analogous to those occurring on f1at, exposed. Arctic coasts.. The temperatures during the test period (Mid-February to early Ap¡il), ranged from -3OoF to 15oF atthough higher tempe.iatures'of 20oF and 30oF were experienced in April. Thre equipment used, consisted primarily of; I ski-mounted Seaman's pulvimixer equipped with a ch4in flail consisting of L2O '"::: 57. chains (5/8 in. diam.), each 15 in. ih length; 3 variable-weight (3,L75 Ib to 5,80O Ib by 75 ]b incrementb), corrugated-steel rollers, 6-ft in diameter and 8-ft in width; and 5 tracked tractors. The snow road, with a 15-ft. surface width, \¡/as constructed by an initiat buildup process, followed by the pulvimixer treatment and roller. T\,ro tractôrs, initially spaced 45 ft apart, on opposite sides of the right-of,-way, windrowed the. snow in towards the center- Line. The crests of the 2 final windrows, after 3 passes of each tractorr were lO to 13 ft apart- The contraction in snow volume at s.tage, starting with alsnow density of O.: gm/cm3. (18.7 lblftl¡ fhis ¡ was in the order of. 40 or 50 percent. A rough .levelinþ was then performed with.the tracked tractors, followed by 5 passes, at l-hour intervals, of the pulvimixer-roller combination. No satisfactory : answer was obtained iegarding the optimum number of passes of the equipment combinationi however, it was concludeci that the time interval between passes should be kept to a minimum and in no case should be allowed to exceed 2 hours. The buitd-up process was accomplishedatthê'rateot.|mi1esperday;however,theprogress on the finished rg.ad dêpended, on the single putrvimixer, limiting'the construction rate to 3 miles'per 24-hour working day, under favour- able conditions. Favourable conditions are defined as terrain open and fl-at, sno\¡r depth'under 3 ft , wind chill factor less than 2,IOO, and visibil-ity at least half a miLe. l.:' 3 The densities achieved ranged between 0.4 q,m/cm and 'a 0.5 grm/cm' 4nd upon completion of a road. section, a 36-hour curr-ng. period was allowed- before the road was subjected. to traffic. The 58. . traffic consisted of: 3-ton trucks loaded to a gross weight of 9 tons, developing axle loads of 1I,5OO lb ¡ 2O-ton loaded sleds (moving aL 25 mph) and tracked tractors. The actual- traffic volunie is not given in reference [4O]; however, iL is inferred that the ' traffic \^/as more or less continuous over the test peri.od. rNo faiLures occurréd over,any part of the road, and no evidence of .deterioration,of the road surfac was to improve the bearing qualities of thê snow pavement over Èhe durption of the traffic tes.ting 4. Greenland The basic technique of snow stabilization consists of depth processing followed by compressive compaction. In 1953, trials on l variations of the basic technique were condueted by the U.S. Navy on the Greenland Ïce Cap. The variations included doubte-depth proces- sing and layer compaction. The constrUctíon of a snow-compacted ! funrttay during these trials, is described by Moser [31] and summarized below. i a. Precompaction Preparations - It was found that considerab1e grading of the wind-driven undulating snow surface {sastrugi) was 'a'necessary to produce a leve1 surface for compaction. As a result, a ski-mounted snow plane for grading and leveling snow was developed. Tþis 6,I2O-lb snow plane is mounted on 4 skis giving a bearing pressure of 2.5 psi, and is equipped with a l-2-ft-wide combination planer bowl and grader blade- Approximately L2 to 24 hours after initial rolling with a LO,24}-lb, 8-ft-diameter, 8-ft-wide steel 59. roller, the area was l-eveled with the sno\^r plane. Two independent rolling passes were performed during thei initial rolling (1 to 4 hours between passes) and 2 Lo 3 passes of the plane \¡tere. required to achieve a smooth and l-evel surface. Allowing 24 to 48 hours of age-hardening after the levelling operation, the entire surface was double-roIled. ApproximateLy 2 days later, the first depth processing was performed o.ri the prepared snowbed. b. Depth Processing - Depth processing of the prepared snowbed was l : ,.: performed using the Navy Model. 42 s.now mixer. The snow mixer is a modified engine-driven earth pulverizer, I ft wíde,. with a 42-in I r:otor. The modifícations permitted an increase in the rotor per- ipheral.speed from 2,480 to 51665 fpm, and included a:balanced ski mounting eliminating porpoising and providing initial comþaction to the processed sno\^I . At an average air temperature of OoF, 3 mixing passes (at 3 32-in depth of cufl were perf,ormed at L-hour intervals, immediately followed by-3 roller passes. The snow par- tioles ranged in size from I to 5 mm after one depth processing I '(3 mixer passes) and regardless of the numbér of additional mixer passes,. further pulverizing was limited. The 3-pass mixing followed by 31pass rolling produced a 16-in thick snow-mat that.was 46% more dense than the natural snow (0.33 to 0.47 gm/cm3) and within 3 days was 6 times as hard (39 to 231 R*). It was concl-uded that the number of mixer.passes is dependent on the type of sno¡¡ being processed and that old, perennial sno\¡r' quch as that found in Greenland a:nd Antarctica, requires three passes of the snow mixet l-:or good pulver- ization and blending. * The quantity R is the snow hardness index obtained with the Rammsonde cone penetrometer. i'The penetromeÈer is discussed in detail in Appendix c. . 60, c. Double Depth Pr:oce.ssing - Duríng the triafs, it was reasoned that by reprocessing a well-bonded, once-processed snow, r more thorough pulverizaLion and consequently a smaller sno\^r particle size might be ' attained. A doubl-e depth processing (3 mixer passes followed by 3 roller passes) was then apptried to the top 10 in of the once- processed sno'hr. The reprocessing produced smalLer indivi dual particle sizes, mostly ranging 'from L Eo 2 mm, raising the.average sno\^r density to 0.54 gm/em3. After 30 days o.f age hardening in ân average air temperature of l4oF, the average hardness was raised.to. - 635 R. Aircraft tests showed that the mat could genêrally-support ¡ whgql loads up to lOO psi except for sporadic soft spots where the mat hardness was as loq as 30OR. The occurrence of.sofÈ spots result- ed from poor quality control during the depth processing. The hard.ness growth due to depth processing is iltustrated in ) Figure 3.1. t¡orul âiI tenptrÀturc$ gr4tltr lÊ bclor ?Ol --.a- lÍt - c:or--:r' reiúlcd aiI ler;e-tw¡3 - lcq ? lcs ,_*)_J I I ¡ \_ cr;*tt' d.layed ir ri r ' 'tcrpcaottr¡'es abor.: 25F i:: o5ro15202530 ' Fígure 3.1 - Normal hårdness growth in a.single layér of polar sno$r. Navv cold-processr-ns sno\^¡-comtrìaction techniques, ,liåT:"i::"îrfiil, 6r. d. Surface Hardening - Ranmsonde hardness tests performed on the double-depth processed snow mat, indicated that the hardness was not unifòrm with depth. That ís, the bufk of the hardness occurred in the middte two-thirds of the mat. To improve the finish and hardness ì of the upper layer, a standard commercial 13-ton pneumatic:tired, 13-wheel roller,wTs used followed by a 2'830-l-b finishing drag. The fínishing drag was constructed with cylindrical bottom skids and measured 20 in high, 7 L/2 ft long and l-2 ft'wide. Artypical I\" ::: hardness distribution, occur:ring before and after sur.face hardening, . .: r-s snor,vn an Ï'actureJ LLç 3.2-J.L.. I Figure 3.2 Typical vertica.l hardness in compacted snow compa.red with its redtstribution after surface hardeníng. (after Moser [31], Nav]¡ cold-processing snow-comPaction techniques, MIT Press ' 1963.) .. Layered Compaction : T\^ro heavy laygrs of drift were dlosited du.ring the operation of the sriow ruru^¡ay. The layers of drift were double-depth processed and a final mat consisting of the original l6-in layer plus a 9-1" layer and an 8-in layer \¡¡as produced. After t-esting with aircr¿ift wÍ.th various wheel inflation pressures, a ten- tative mj-.nimum hardneSs gruide for compacted runways on deep snow was i:i.1-.-::::l-:..- :..".,:' .-';::-:":::r:::::;:-i::i-r:r..:i'-:-i- iiiii-ii;';.1:l:.: 62. developed; The guide is shown.in Figure 3.3. Flrst leter Second rl6ter Thfrd'1ôt€r ' O l0 20 30 LC ¿3 Snou pat thlck¡êss (tnchcF) Eigure 3.3 - Tentative minimum hardness guide for compacted-snow run¡¡ays on deep snow. (after Moser [31], Navy cold-processing sno\^i- compaction techniques, MfT Press, 1963.) The layered compaction te.chnique offered several advantages. Firstly, the runwa)¡ was.elevated abóve the natural snow surface, elimínating any fruther drift problems. Secondly, a more uniform Ioad:bearing surface v/as achieved as the probaìrility of coincidence of low-strength areas in the dndividual layers \^7as very' sma1l. A logical conclusion is that the layered compaction technique allows a is required to produce a relaxation ..in the degree o.f control that uniform hardness .in a single layer of'snow pavement. Howevêr:, the ': method generally requíres transportation of borrowed material which couldìbe both costly and time consuming In the same p'ublication, Moser cites other examples, where the cold-procqssing methods described, hàve been used successfully. These includq thq 125-acre'¡rarking lot and l-rnile sno\^7 road at.squaw Valley, 63. California (1960 V'Iinter Olympic Games) and the 3-mile snow road. at McMurdo Station; Antarctice (1960)'. Further detait on the U.S. Navy cold-processing technique and equipment is given by Moser Í321, U.S. Nawy [39], CaÍm f8l and Easton t2Ol. i '.'::1: 5. Kapuskasing, Ontario (1952-53) I : The reporÈ entitled U.S.-Canad.i4n Joint Compactign Trials [38J providesacomprehensiveSuÛnaryoftheco1d-processinganddryh.eat- processing.techniqueSexperimentedwithatvariouâCanadian.andU.S.- locations; The experience gained at Kapuskas'ing, Ontari.o is represen-. j,,..:,:.:..,.:: ' l :l ::,: :' tat¡ve .:" effort ": r":". During the course of the triaf.s at KapuskasinE, Ontario, the foI1owíngcombínationsofequipmentwereinvestigated: . or without heat; 2. a lane processed from 1 to l0 times by a pulvimixer r^¡ith : orwithoùtheatfo1}owedbyaro11eroriombinationof i |.,rol1erSproceSsingthesame1anefromIto10times; .3. a lane processed from l to 1O times by a roller or . i.',,i.:,:.; ' \ combination of rollers followed at varying intervals of 't',:.::. "t"tt :. :: ...- t:. :': r:; time (t to 24 hours) by a pulvimixer with or without heat, processinS thq lane from I to 10 times and followed finally by a rolter or combination of rollers processing the lane ' ... :. ' from I to 10 times i iti..ìr,' . a lane processed from I to l-0 times by a 'roller or cornbinati.on of rollers i 5. A lane processed from I te 10 times by a V'foods Preparizer; and followed by a roller or combi-nation of roll-ers processing the lane from 1 to 10 times The basic equipment referred to is detailed as follows: ' Rotlers -. Several rollers of different type and weight \^rere used. the rollers used most extensively and successfully were: a segmented steel roller (4, z-ft sègments j-ndividually mounted),.10 ft in diameter 4nd weighing 22,¿OO f¡s; a lO-ft diameter corrugated roller 8 ft long and weighing 7,7OO lbs ; a V,filliam Bros. rubber-tired . . ' roLler weighted to a gross weight of 38,000 lbs giving a unit tire I pressure of 110 psí. Pu1vimixerS-Seaman'spu1vimixerswithmaximumrotoroperatingspeeds of 275 rpm were used. T\¡/o of the pulvimixers were modified by the attachment of heaters to the qanopies. In this. modification, 2 sizes of heaters were used - one heater capable of turning out approximately 1,.000,000 Btu/hr and the other 5,O0O,OOO.gtu/hr. gloods preparizer - The lrioods prepáriåer is a standard plvement breaker with a mechanical action similar to that of. the Seamanrs pulvimixer. The Vfoods preparizer however ís equipped with a:more pôwerful engine and a higher speed of the rotors is possible. Traffic testing of all:laneé was carríed out using standard a Canadian and U.S. Military vehicles. Lugged tires and commercial .. pattern tires. were used at ínflation pressures ranging from 15 psi to , . 70 psí.' The lanes.c-apable of carryinE the traffic had a minimum density .? . of O.53.to 0.55 gm/an'and a minimum Rammsonde hardness of 350. The foflowing table shows the percentage of lanes laid by each'method 65. which were abfe to carry the traffic. TABLE 3..1 Summary of Traffic Tests Percentage of lanes placed Method by method indicated which car.ríed traffic Pulvimixer with heat and rollers 82% :.: .l' (segrmented 6LZ , : ;'r Rollers and rubber tired) alone \ ".. ...1 Pulvimixers without heat and:with rotrl-ers 56s" Pulvimixeis alone with heat 7eo / Pul-Vimixers alone without heat 12'.5"ó Tn the above analysis, no. particular attention was given to such variables as ambient air. temperature, temperature of processing, period of age-hardening or hardened snor¡r temperature. H9wever, the analysis'strongly indicates that regardless of the method of.snow agitation, \nrith or without'heat, heavy rollers are essential to pro- duce hard surfaèes. Furthermore, whereas the maximum Rammsonde hardness produced Without heat with hqavy roller was 531, the maximum 1 Rammsonde hardness achieved wíth heat and with roll-er was 2631. It can be concluded that the probability of successful lanes where heat and heavy rollers are employed is greater than when heat is not used. It was reäsoned that to produce a sno$l surface that "definitely" carry atl- mitritarlz traffic, a minimum density of ) 0.6 gm/cm' would be required. It was also confirmed. that a density '? greater than 0 .54. gry/cm" very difficult to achieve by cornpaction '' -was alone, eVen when using extremely heavy rollers. Therefore the increase in density has to come from a decrease ín voids due to the addition 66. of r^rater. The.addition of 10.4e. by weight of water to snow already compacted to 0.54 En/cn3 wil-l bring about the required increase. For a J'2-in depth of snow, this represents the addition of 3.3 lb of : water per square foot: To qatisfy this requirement the following processing schedule \^7as developed- ,,' 1. 2 passes wít=l:' 22,400*lb steel roller - density approachi4g 0.54 - 0.55 qmlcmr 2. 5 passeé of pulvimixer with heater - l0eo moisture added (hãater outp-,rt of 2,560,000 Btu/hr, : l.'.,-:,,: ': : forward speed of 1.9 rnph) \ ::' 'rolLer 3, I pass of 22,4OO-L! o ' 56 -- O ' 60 gm/cmj :'1"" '' The method described,above allows construction of a finished 8-ft roadway (under favourable conditions) to proceed at the rate of 1.9 mph. Dieset fuel consumption is estimated at 233.3 gaL/lnr. 6. McMurdo Station, Antarctica (1972-73) AlOmile test section of snor,rr road, over.the Ross ïce Shelf betweenMcMurdo'StationandWiI1iamsFie1d'waSconstructedusinga 1ayeredcompactionprocedure.Thepurposeoftheexperimentwas.to simplify the existing cold-prooessing technique as developed by the i,,..,,. Naval, Engineering Laborêto-ry (NCEL) NCEL cold-processing Civil : The . ,,,,:,,. . ...,:..; ' technique. has been discussed in Section B of this Chapter under the : heading'4. Greenland' . The simplified snôw road construction procedure as outlined by Thomas and Vaudrey f36l , consisted of the following steps: I. Site preparation - The snow surface along'the proposed snow road, alignment was initially rolted wíth an 8-ft diameter, 8-ft wide, 10,,000-1b steel roller. The rolling was necessary to produce a 'l . '.:j :: t 67- faírly uniform, compacted surface for d.epositing the layers of blown snow. i . 2. Containment berms - By a single pass along the outer edges of the proposed roadr .the snowbl-ower deposited 2 berms of snow 24 to 30 inches high. .The berms contained the layers of snow sub- sequently deposited by the blower, and prevented a laterat åisptace: ment during the compactíon, and leveling of the indívidual layers. 3. Constr:ucting the snow layers - Snow was deposíted between the containment berms as evenly as possi.ble and in sufficien.t quantity so as to produce a 4-in layer after leveling and compact.ion had been i.: r' :.' ( completed. Leveling of the btown sno\¡r, inunediately following the ' deposition, was attempted with 1ow ground pressure (LGP) D-4 and D-8 dozers, however, with no success. It became necessary to use an '80-ft snowplane,'towed'by an LGP D-4 dozer, to accomplish the leveling. : . To compact'the 4-in layers, 3 walking passes of the LGP D-8 tractor were used. Attempts to .use the 10,000-1b roller $¡ere un- successfuf as the roller.tended to plow creating a rough and humpy burface. !. 4. Preparing finished surface -.Once the desired road height rras achieved, the 80-ft snowplane was used for finish leveling'the surface. After finish leveling, the road was allowed to set:up for '3 days. Then to surface harden the snow road a rubber-tired, wobbly-wheeled roll-er was used as'in the conventional NCEL cold- processing method.. Following the rolling, a timber drag was towed by a l-ton picl¡up to smooth the road surface- After 4 days of age hard.ening, the road was opened to normal traffic 68. The study conöIuded that roads built by the modified method would give satisfactory service for the movement of cargo and personnel by heavy, wheeled'-transpor:tation equipment. Test results have shown that the densities and shear strengths obtained by the layered method compare favourably with those obtained by the conventional ".,.,'. cold-processíng method (Pulvimixitrg) . Furthermore, the layered method allows a red.uction in overall construction time by more than 40 percent and eliminates the need for special equipment such as the ski- . ,.:::.. \ ,, ' mounted sno\¡/ mixer. \ . :: .::: 7. Norman !Ìells, N.l,V.T. I, During March and Aprit of 1973, a winter road.research study was conducted (Adam'[2]) near Norman VteLls, N.W.T. to evaluate con- struction methods-and. environmental effects of various types of winter : roads. The research study included the construction and trafficking of a.compacted snow road, an ice-capped snour road, and a solid ice road. The evaluation of the winter roads, as regards to construction pethod and trafficabilityr êrg summarized. l. Snow Road - Snow was first bladed in towards the centerline of the right-of:way to form about a 3O-ft wide base'for the snow road ,., ':: Amakeshiftleveler-compactorr,^7asdevised(asmoreadequateequipment was not available at the site) using a large diamêter log with a Eucfid tire tied on for added weight. Thís leveler-compactor was to\^/ed ', ,:: b1t a D-7-l-74, tractor and 2 com¡¡teté passes over the road surface \¡/ere l',: performed. The snow road was allowed to age-harden for I day before it was processed to a depth of 10 inches using a disc tiller. To ì permit control over the depth of penétration, the disc tiller was -. __. mounted to the front-gnd loader assembly of a IHC-TD6- The snow road i,,,. 69. \^/as processed by 2 passes of the disc tiltèr and then comþacted by 4 passes of'a D-7 tractor, each pass covgring the road surface l track-to-track. The road was allowed to age-harden for 2 days before it wãs traffic tested to destruction Príor to the traffic testing, the snow road had a density of ? 0.53-0.54 gm/cml and a Rammsonde hardness of 25I. In one lane, the snow road .failed beyond repair after 25 passes of a Fargo, B-30 ser-ies, lS-passenger bus \,\rith â,GVlv of 7,7O0 lb, and tire inflation pressure of,,29 psi. In the other lane, the snow road failed beyond repair after 4 passes of a 651000 lb, FWD, 6-wheel drive water tank I truck. the traffic tests support the earlier,studies carried out at' Kapuskasing, Ontario where it was determjned that a minimum Rammsonde hardnôss of 350 was required before a snow road could be trafficked by wþeeled vehícLes. 2. Ice-capped snow road - The snow road of 1. was reshaped and then compacted in 4 walking passes of a D-7 tractor. This was followed with an.additional 4 passes of the leveler-compactor. The road surface v/as finalfy leveled with a single pass of an I-Beam drag. (18 '1 The l-Beam dîag consisted of ,6 I-beams in x 7 in x 7 ft ) arranged in a V-shape with 3 l-beams hooked one behind the other per ?, s1de. The snow density attained was 0.50 gmr/cm over an averáge road thickness of, 37 cm and the averlage Rammsonde hardness after 23 hours of age-hardening (in an ambient temperature of approximately -10oF) was 75. . Approximately 24 hours after the reshaping and compacting of the snow road was completed,'water was applied at the rate'of 70; .) O.49 gaL/ft' by hose. The density of the ice-capped snow road : 3 averaged 0.63 gm/cm3 over a 20-cm thicknpss, O.7g g /c over a lO-cm.thickness, and 0.85 gm/c*3'orr"t a 5-cm thickness. After 12 trours of age-hardening, the average Rammsonde hardness increased to 646. One day,after the ice-capping, the ice-capped snow road was traff,ic tested.'After.25 passes of the.Fa¡go lS-passenger bus, minor failures were observed along the outer edges of the road. surface. During Êurther traffickingi, numerous more failures (2.52 by area) occurred. These failures -were easily repaired by hand and the road ! remained serviceable on.a continuous basis. The oceurrence of weak sglots was attributed to the lack of uniformity attained by the hose- end method of water application. . r The traffic testing. remained more or lèss continuous from March15toApriII.Thetestvehic1esinc]uded: 1 Fargo, 15 passenber bus, GVhl 7,7OO Ib., tire preåsure 29 psi; l Fargo 4 x 4 Crew Cab, G\/!,I 8rO00 1b., tire presl;ure 40 psi; , 1 Ford tandem T804 Series, GVW 43,000 lb., tiré pressure 65 ,psi; I IHC-IOO Series, GWü 5,400 lb.¡ tire pressure 28 psi; 2 Gl4C tandemsf GVw 39,000 lb., tire predsure 80 psi; l- Ford F-500, cwt 2O,0OO Ib., tire pressure 75 psi; I Ford F-350, GVhl 1O,OO0 lb., tire pressure 60 psi;'and 1 Ford F-IOO, GVI^I 5,000 1b., tire pressure 32 psi: A total of 35,924 vehicle passes hlere accomplÍshedcomplished by ththese vehicles travelling an estimated 8,160 miles over a 1'200-ft test loop. ' The test loop included the ice-capped snow road and an ice. road. The ice road portion of the test loop ís discussed next. 3. Ice road - A road bed was prepared by an initial'leveling operation using the Euclid tire clrag towed by the D-7 tractor. The ice road was then constructed by spraying water from a spray bar mounted on the back of the water tank trucks. By successive appli- cations of water, an ice layer was built up to an approximate thickness of 5.35 inches (13.5 cm). As already noted, the ice road portion of the test loop-was trafficked at the same time as the ice- capped sno\^r road. The ice roâd held up well to the traffic, requiríng I tittle maintenànce except for occasional sanding to improve traction. The study makes several concluding coruircnts regarding thg 3 tlpes of .winter roads eval,uated. These are srmmarized as follows: - Unless sufficient.moisture is available in the snow.it is ímpossible to construct a traffícable snow road without the use of heavy drags and rollers. Even with sufficient moisture the use of tight equipment is questionabfe I 2 l - epplication of 0.5 to O.8.gaL/ft- of water to a minimally constructed.snow road will produce a tnaffica-bíe ice-capped .snow road ovêrnight,if temperatures are well below freezing. Maintenance of ice-capped. snow roads is subsbantial if the hose-end method of .water application is used. It is pos- sible that by a more uniform application of water, a large portion of t.tte maintenance could be efiminated. - Ice roads may be trafficked befor'e completion on a timited . basis. No tr4ffic, however', cogld be tolerated on'a sno\^7 or ice-capped snow road during its construction. 72. - The large quantities of water required for ice road con- struction may prohibit extensi.ve,use of this type of wínter road. - The traffii applied to the test 1oop, as'regards to total number of passes and total tonnage, simulates the actual traffic anticipated in 48" pipeline co4struction; Final Comment Ì:ì"; 'influence vrinter. rõad constructioh practice. These inctutle: Çêo- i/.: graphic location¡ meteorofogicäl conditions; building material pro- perties; structural requirements; and other multi-disciplinary concerns (environmental, etc.). fn order to adapt a tqchnique of winter road construction to the requirements, something must be knor¡¡n about the range of.results obtained by the different construction methods and the variation in results under'various conditions. , In the review (Construction Methods and Results), two phases l ín the field of winter road reseaich can be distinguished. Ttre first phase includes: fundamental studies on the properties ánd behaviour' of snowi tthe securing of information on the variations of the bearing properties under different cond.itions; and the determination of the demands made on th.e snor,f substratum by the different types of: trans- port. The. second phase belongs to'.the category of applied research and is aimed at the development of .construction and maintenance '', techniques that satísfy the requirements from the viewpoint of traffic demand, economy., and environmental impact. The fiterature : Índicates that although the first phase is fairly well deveLoped, the I second is still in its adolescent stages. 73. CHAPTER IV FIELD STUDIES .''; The field studies were performed with the following primary objectives: l. To evaluate the performance of a Sicard snow btrower : aq a snovl Processíng imPlement. 2. To study the effects of ambient and snow temperatures . on Rammsonde hardness. .: 3. To investigatê the potential of utilizJ.ng a tow-type surface heater in the construction of snow roads. / The cgnstruction equipment \^¡as provided by the Universj-ty of ¡,lanitoba Mainte-nance Department. Tkre fimited equipment selection de- termined the course'of the field studies A. EXPERTMENTAL TECHNIQUES Construction of Snorv Road Test Lanes as a snow processor, 5 snow road test l-anes were constructed. Three of. the test lanes (200 ft long and 1g ft wide each), weïe constructed end- i,:', - 'fwo to-end using 2; 4 and 6 complète blower passes (,Jan. 26/76). molîe fanes (350 ft long and 12 ft wide, each) | \rrere constrqcted.separately 26/76 and Mar . 18/76) using a singJ-e processing. , iüan. Sufficient snow was brought in and positioned adjacent to the t¡ roarlh'ed by windrowing the snov¡ on either side of the proposed roadway with'a Champion D-562D motor grader. The positioning wás necessary as the snow blower was equipped.with a standard loading chute and could not eject the snow to the rear of, the machine, only to the side.' Immediately' .i 74. following the windrowing, the snow \^ras processed and deposited adjacent to the windrow if further processing was to be performed, or spread as : j uniformly. as possible over the roadbed surface if the final processins yas being completed- The snow blower was kept at ful-l load, moving at a. forward speed of 60 ft /min , with an output of about 630 tons of snow ': per hour. 'The auger and impeller speed was maintained at about 700 r.p.m. Fhotograpþs E-1 to E-6 illustrate the construction procedure. Following the final processing and depositign, the road s.urfaces 'h¡ere smoothed and dragged with A 6" x fO" x Il-' wide-flange, steel beam ,' in I complete passes each. Further compaction \^ras atteinpted with a I C.sá 1l5O crawler, following a 2-hour period of age-harderìing (Jan. 26/76). A 4-ft diameter, 2-Lon, steel roller was'used in constructing the single .processed test lane on Marcþ 1,8/1 6. . Surface Comþaction at 0"C (Feb. 9/76) Duríirg the period January 26fi6 to Marc-h Ig/76, unusually high I tempgratures. weie experienced. During one of the warm speIls,. the snow ' temperature clirirl¡ed to OoC, simulating a surface condition as míght be produced by a surfacelheater. During this warm period, a portion of the single-processed test lane wab surface compacted with the Case I15O crawler tractor. The ground pnessure delivered by the tractor tracks was . about 9 psi. Rammsonge Hardness Tests The snow road hardness in the,O - 5 .cm.., 5 - I0 cm , l0 - 15 cm , I and 15 - 20 cm 7 lalreis \tras observed daily, starting inrmediately after construction and continued up until permanent thaw in lvlarch. Ttre 1360 -.t'i, i á ; j{iti ili : ;:i.ia i:; : ì.1 ;'5,t: 75- readings obtained were used as indicators as regards to road bearing strength., with high and low Rammsond.e hardness corresponding to high and low strength respectively. The Rammsonde cone penetròmeter is discussed in detai.l in Appendix C. The snow hardness of the clifferent lanes was compared and the significance of single- and mulÈíple-processing using the snow blower, was evaluated. ,The age-hardening phenomenon and the lemperaLure/ hardness relationship. in the test lanes was observed. llhe correlation bqtween snow temperature and hardness was de.termined and a finear re- gressiop equation fitted to the data. The analysis involved 700 snow temperature/hardness data pairs. : Temperature Air temperatures \^rere o,bserved on a continuous basis for the dufation.of the field studies. The proximity of lhe field test area to ''l theMàteorological Station at the Vlinnipeg International'Airport, per- mitted filling in of missing data from the published monthly Meteorologiçal Sr:nrmary- The dail1z relative humidity for the testing period was derived l from the same source. Snow road temperatures \^rere observed gnly during the time of Rammsonde testing and were taken at a lO-cm depth. Grain Size The grain ri".'of the snow, before andafterprocessing, was estimated by visual inspection using a milfimeter scal-e.. The snow was randomly Sampled and an apparent average particle size determined. No statistical interpretation was applied to the sampling and no sievê analysis was performed to establish the particle-size distribution. :;)rf,ì{ }!:':{':ilì'i';_;i.:t : : - ,.:1 1,\-. r\. rra.,i'.i-.._i{:.ì.i.:.ii: 76, Density A snovr sampling tube was used in determining the average density of the undisturbed and reconstituted snow. The snow cores measured 5 cm in diamêter arid 19 cm (max. ) in length and the. average mass density (S*/.p3) was.detennined by weighing the kúown volume of snow. The density of the individual test lanes was based on the mean of 6 core samples in each lane. The densi.ty .r¡¡rofile' ín a total of 34 core sampl-es was deter- mined (prior to and after the burn test) using a Nuclear-Chicago 8775 : scaler/timer, and a Harshaw DS-200 crystal scintillation detector. / Cobätt-60.was used as a ganma-ray source which was directe.d as a beam through a pin-hole in the machined lead shielding bricks. The core sample was placed between the radioactive source and detector"and at least 10 counts of O.2-minute durátion were made at each point along 'l the core. The pY:ocedure is presented and evaluated.in Appendix D. IÏIg_Pr"tqtyp" ur"pd- f it The good rèsulis obtained from. the snow road surface compaction tqst at OoC encouraged an investigation into the possibility and practi- i cability of artificial surface heating. For this purpose a wood-fired heater \^tas considered to be potentially the rnost advantageous. That is, , along wi.th performing its primary function, the hea.ter would also provide a disposal facil-ity for right-of-way slash material. To, evaluate the potential of the heating method, a prototype heater \¡¡as constructed and Six, 45-ga1 , oil drums were cut in half longitudinally and the halves were welded together in a configuration as shohrn in Figure E-I of '. 77. Appendix E. .The assembly was adeqirately reinforced anil tow hooks were provided :as shown. The weight of the heater, 'as constructed, was 550 lbs- I Seasoned pine was selected as the fuel for the burn tests and one (1) cord.of the seasoned wood was used. The wood was randomly piled rn the burner and after setting it on fire, 15 minute allowed for the fire to spread evenly to al} parts of the pite. The wood burner was thenltowed by the Case 1150 crawler over the nultiple-processed snow road -. r": The sl.owest speêd possible with the crawler 20 .tow .was ft/mLn. To study the effects of the heating at speeds less ihan 20 f.E/nin , a / I start-stop towing procedure was necessary. During the stop intervals, the heater was held stationary for I, 2, 4, 6,8, 12 and 20 minute durations. l '.i.::.a':.:a ??i,? 78. B. RESULTS AND DISCUSSION Cons.truction of Test Lanes After the final snov¡ processing and deposition by the blower, the snowbed varied. in depth fr.om f.5 to 3.0 feet (Photograph E-3) This depth of' sno\^r proved. e-xcessive for, the equipment used, and man- euverabilÍty probtems were encountered. During the leve1ing and dragging, ope.ratíoq, th9 crawler tended to bog d.own churning up the '-! ..ìi 'i snowbed r¡ther than compacting it. Also, the steel drag was too light anä unstable to be very effective " With each sucòessive pass of the ,:rl cra¡vler and draçf combination, the snowbed became increasingly spread out and uneven. It became il-,ecessary to re-shape ùhe snowbed with the grader and start ane\^r. This time, the leveling was achieved by using two,tow vehicLes, one on each side of the roadbed w'ith the steel drag in the middle. This ga\re sone stability to the drag and a smooth and even road surface was achieved in I passes. Compaction of, the snow road was neglígíble by Èhis dragging nethod. , ' Two hours after completion of the 2, 4 anð,6 blower-pass test t Lanes an attempt was made at compacting'the roads by walking the Case 1150 crawler over: the road sqrfaces. As it turned out, age-hardening had. not advanced sufficiently and the tractor broke through the road surfaÇe on the first pass. It waç observed that the double-processed road section carried the tractor with no difficulty whereas the tractori inmediately broke through on the 4 and 6 blower-pass test lanes. Since i: ' it is.lggical to expect a finer particle size to be produced þy repeated processing, the observation suggests a slower age-hard.ening response in 79. finer s.no\,¡. Ager [4] reports simil-ar obse:ivations regarding hardness sensitivity to temperature and grain size. The hardness/temperaLure/ 'l grain-size interrelatiohship is discussed further in later sections. Construction of the single-processed test lane (March 18/76) . using the 2-Eon, A-fi- diameter steel rolier, wêÇ unsuccessfùl. To : avoid the maneuverahil-ity problems previously experienced., a layered compaction procedure \^¡as attempted. It was found. that even with shallow ì snow laye.rs of 3 to 4 inches, a competent road could not be consttucted. j Although'the reduction in snow depth improved equipment maneuveralcility ì / ' Due'to the small 'roller diameter, skidding tended to occur producing I an uneven burnpy surface. Furthermore, with the roller width and diameter being approximately equal, a side-to-side tipping action persisted. After 2 or 3 passes of the rioller, all resemblence of a roadbed was deStroyed. The present experience, with btower-processed snow, lends justification for the use of specially designed long-span snow þla-nes '(t4cMurdolst.tiorr, such. as recommended by Thomas and Vaudrey [36], I An'tarctica) Poro.sity, Density and Grain Size ' The minÍmum porosity obtainable with an aggregat.e of 'equidimen- r sionalt particle5, . all Èouching one another is 40.2e". This corresponds to a snow density of approximately 0.53 gm/cm3. Recalling that a snow road density of 0.6 gm/cm3 hrr been established as the acceptabfe minimum (u.s. - Canadian Joint Compaction Triáts [38] ), the premise upon which blower-processing is based, is thaÈ the required. d.ensity can be accom- 80. plished by a filling of the void spaces withlparticl-es smaller than those creating the voids. That is to say, the milling and mixing.action of the snow-blower produces a particl-e size distribution such that the physical condition of optímal close packing is appqoached. This condition, in addition to satisfying the density requirement would also be most advantageous for maximum bond (strength) growth. Unf,ortunately, as wiLl be seen in the following discussion, the field observations permit only a qualititive evaluation of the cold-processing technique. The average: densily of the l, 2, 4 and 6 blower-pass test lanes ? 0.48L,. and 0.5L2 gm/en- respectively'(Table A-3, .was.0..458, -0..511 / . -' eppindix A). A 5e" gain in density was achieved by increasing the blo¡¡er- processing from L ro 2 Basses and a further 6% gaín by increasing the processing from 2 to 4.passes. Repeatêd. processing beyond. 4 passes, produced no further density inerease- The apparent average grain size before and after processing:. was estimated At 2' mm'and 0.6 mm, respectively.' No visual difference was detectable in the single- and multiple-processed snow. :3 i Failure to achieve densities that approach the acceptable 0.6 gm/cm- can be attributed, at least in part, to inadequate compaction of the processed såowbed. liowever, without knowledge of the particle-size gradation in the processed snow, the effeqtívenêss of'the processing umethod.remains qncertain. Tn this regard, the hypothesis is adopted that repeatgd processing tends to produce an 'equidimensional' material with no' significant decrease in the average síze of the srrow particles. ,As the prepent density results imply, the condition of minimum porosity, produced by a close packing of equidimensional particles. is approached : through the repeated. processing. Previous studies lend further credence 81. to the. hypothesis in that they show repeatedly,. that regardless of the 3 intensity of agitationi a sno\^r density of approximately d.SA gm7cm" ! remains the upper limit. Although, in concept, tlie sno\^r processing in the present fíeld studies was performed to achieve non-unif,or,mity in grarn srze, tne results make strong claim thqt an equl,dimensional particle size was produced, ': Performance Characteristics of the.sicard Snow Bl-ower At full load. the Sicard. snowblower (powered by an IHC-345, V-8 engine), processed the freshly windrowed snow at an average rate of . 630i tons per hour. At a full depth of cut (blower intake hood, 78" x. 30") and an impelter speed of 700 r.p.m., a forward speed of al¡out 60 tJ/mLn (0.68 m.p.h.) could be maintained. The density of the windrowed' ì l snow was approximately 0.30 to 0.35 gm/cm'. '. The discharge characteristics and spreading uniformity (photograph E-3) depended on the macliine operator's abílity to mai¡rtain a constant load on the intakê êugers.; Overloading or 'chokingt occurred frequentfy.r i 1eøuiring temporary stops so that eng:ine and ímpeller r.p.m:'s could be regaíned. This' behavioür predominated when dguble-depth processing (see Chapter III, '4.. Greenland') was attempted Rarnnpnde Hardnesl_of Test Lanes The Rammsonde hardness data are presented in Apþendix A. The hardness numbers are given in both the uncorrected (conventional) and corrected forms. The two methods of presentatíon are used in Figure A-1 to illustrate the typical snow road hardness obtained. o,) The hardness data indicate very low strength in the snow roads. : In fact, for a1l 'practicaf! purposes, none of the snow roads could be I classified as tkaffica-ble. On the basis of previous 'investigations, in particular the èxperience gained at Kapuskasing, Ontario'[38.], and the studies by Adam lL,2i, an average hardness in excess of *350 is required. :':'.:.:.:'.'i : On the basis of this criteríon, only the test lane compacted at the surface temperature of OoC was acceptable " Figure A-4 illustrates the hardness : increase due to the surface compaction. :.,,,',,,, I : r..::-.-:. A dístinction between 2, 4 or 6 blowei-pass pr:ocessing could.not : \ , ,, ''.' be made on the basis of the.Rammsonde hårdness data,.although, the síngle- ¡':ì:'::1';ì' I proó.="ed lane (without surface compaction at O"C)'showed some tendency to be softer. It.was generalllr observed, however, tliat with an increase 'I in hardness (decrease in temperature), the deviation from the mean 1 increased as welI.. Rammsonde Hardness-Temperaturê Relati onships The .relationship between Rammsonde hardness and air temperature, . for the duration of the testing period, is shown in Figure'A-2. It is I. observed that dufing the initial age hardening period (approx. 2 days), 'i the snow road hardness increased independently of the fluctuation (+ or -) in air temperature. After that, with some lag however, the hardness paralteled the variations in air temperature. In numerous previous attempts, no correlati,on between Rammsonde hard.ness and air temperature was found. Íhis is likely due to this 'lag' phenomenon where consequently the handLing gf data near the temperature inflection points becomes i i * The average RaÍmsonde hardness of 350 is equivafent, in the corrected form, to an average hardness of tO50 in the top I.Q çrn.' . ': I 83' problematical. In the present study, correlation of Rammsonde hardnesb and snow .t road temperatufe was at,tempted. The correlation coefficients found for the O-5 cm, 5-lO cm, lO-15 cm and L5-2O cm depths \Árere O.78; 0.65, 0.43 and O.37, respectively. A data scatter diagram and regression line, for the O-5 cm layer is represented in Figure A-3. Burn T:ests '.: : :.; As a means of increasing snow density, ' the filling of void spaces,..ì \ j., with free moisture through surface heating, was investigated. The results . .of Çhe burn tests, using the prototype surface heater, are given in : I Appendix B. Figure B-l- shows the snow road density profiles. before.and : after the heat applicatíon. The practicability of the heating method may be appraised in termgofefficiency.Forthepresentana11zsis,efficiencyisdefined as follows: (Energy theoretically required to achieve the observed density increase) e=( t.. 'The resufts indicatê that the heating rnethod is about O-5u efficient. .i.: The l-ow efficiency is attributable to two factors. Firstly, the. design of ' I :- the heater permitted large quantities of heat to be 1ost to the -,atmosphere, and secondllr, as indicated by the efficiency equation, the density increase depended entirely, on the generation of free water to : lr,, occupy the porè spaces.. Clearly, surface heating in this fashion becomes. onllz a poor substitute for the direct water appfication method Once the sno\,r ís vi'ar,rned to O'C, the additional heat apþlied to the'snow mass is used for conversion of the snow to the liquid state 84. The compaction test laÊ OoC þreviougly discussed), indicates that a high moisture saturati-on,.even with minirqal compaction, is not required to produce A trafficabfe pavement. A significant improvement in results can be exBected throggh the combination of surface heating immediately followed by heavy compaction. An obviôus and essential modífication to the prototype heater would be the addition of, a suitable canopy over the heating unit. Appurtenant features may include a fanrand soot trap. Ag. indicatåd in PhotographlE-g, the containment of soot and ash particles would be a ' rnajo¡ consideration. I A sample calcutation :of the heating efficiency just discussed is presented in Appendix B. 85- CHAPTER V, CONCLUSIONS AND RECOMMENDATIONS l The literature rêview in this study identified two distinct areas of research pertaining to sno\¡/. One area of .research deals with the theoretical- aspects of the sno\¡¡ properties and the.othe,r: with the practical appliqations of snov¡ as an engineering material. Evidently, .. the two areas have evolved more or less independently and thus much óf the present practical knowledge regarding snow roads is. based on judge- ì. mental observation. Although the more recent studies have generally been based.on evaluabl-e daÈa, many of the conclusions have been drawn .¡. /- on the sheer weight of accumulated experience. By virtue of its vast implications, the single most important factor influencing s4ow road construction is.geogrq.ptric location.' Im- nlic+t are the influences of meteorotogical conditions, teriain features, environmental stability, .:. etc. The technicaL or procedural aspects specifictothepresent.studyaredeaItwithbelow. The resúl-ts of the fietd studies indicate th'at regardless of the ¡.! method or intensity of snow processing, heavy cômpactors are essentiaL for compaction of the snowbed. If heavy rollers are.used, they must be : of large diameter in order that free rolling can be maintained. I'Ihen' dealing with blower-processed snow, it appears that a layered-compaction procedure, as proposed by Thomas and Vaudrey [36] ís necessary. Depending on ,' the type and size of construction equipment used, the fine-grained snow pro-. i.,,: reate mobility 'problems even in snow depths -l of 4 or 5 inches. .The use of containment berms would eliminate the ?irr.1.3.¡:!ìi.' 86' problems of lateral displacement duri-ng leveling and compaction. .i Multipl-e-processing, using the Sicard snow blowerr.tended to produce snow with an'equidimensional paçticle size (æ O.6 *I). . No significant decrease in the average par:ticle size or increase in strength (hardness) was observed with repeated processing. Apparently¡ the benefits attributable to procelssing result largely from the sintering phenomenon which is enhanced by the mixing of fine snow particles of different The results'of the snow road age-hardening observations confirmed the findings of prevíous studies. For the first 48 hours (approx.) 1 ¡ after the snow has been reconstituted, an increase in. hardness will : continue regardless of the fl-.uctuation (+ or -) in temperature..: After that. there appears to be a st.rong correlati.on between snow hardness and In the field studies, where the snow temperature data -\^/erenoten!ire11zadequate'acorre1ationcoefficientof0.78wasfound for the top 5 cm of the snow roads. The resul-ts of the corirpaction and burn tests indieate that a : combination of.surface heating and surface compaction may be a practícal means of íncreasing:snow road density. Surface heating, alone, .however, is only,a poor substitute for the direc! water application method where in both cases the' increase in density is due exclusively to the intro- 'duction of free water into the pore spaces. Modif,ications, to minimize heat loss are necessary before further testing of the wood-fired heater would be practical. A canopy, which may include a fan, a hea.t discharge chute.and a soot trap.are essential . . 87. . Unquestionably, continued fíeld ínvqstigations wil.I contribute to the present snor^r.road construction tknow-how'. Future studies should I incruàe: 1. A detail-ed investigation into the particle-size gradation ,.. 2. Fulf scale field t:sts utilízing aïroved versions of the wood-fired surface heater in combination with roller compaction. 3. Further investigations into the relationship between snorr¡ ha¡dness and snow temperature. 4. Development of. a rstandardt field snow-testing kit and data documentation procedure. The testing eguipment must include practical. devices by which actual- grain size and arrangement can studied. Also¡ recording devices which . be temperature cou1d be permanently installed through a cross-section of snow road. 5. Documentation, by interview, of unreported construction experience. I! is expected that from these interviews I i wifl emerge avenues of investigation which otherwise might not be uncovered. 88. BÏBLIOGRAPHY tIl Adam, K.M., !'linter Ro¿id Study, Prepared lor the Environment Protection Board, Sponsored by Canadian Arctic Gas Study Ltd., Winnipeg, Manitoba, L97.2. l2l Adam, K.M. , Normal Wells lfinter Road Research Study, to Canadian Arctic Gas Study Ltd., Interdisciplinary Systems, Ltd., Winnipeg, Manitoba, 1973. t3l A{am, K.M., The ínter-relationships of water appli- Wi lson, T.M., cation rate, ambient temperature, sno\Ar road propefties and íce-cap depths during the construction of winter roads, De¡lt. of Civil Engineering, University of Man- itoba, A Research Report to Canadian Arôtic Gas Study Limited, L975. t4t Aqer, B.H:son, Some tests on the compactibility and hardness after compaction of different types of snow, Journal of Gafciology, L964. t5l Ager, B-H:son, Compacted snow as a transport substratum, National Research Council of Canada, Tech. Trans. 865, Ottawar'Canada, 1956- t6l Bader , H. , Ttre physics and mechanics of snow as a material, U.S. Army Cold Regions Research and. Engineering Laboratory, Research Report, .Hanover, New'Hampshire, 1962. t71 Benson, C.S;¡ Fie1d measurements on the flux of water Trabant, D.C. n vapour through dry snow,, The Role of ' Snow and Ice in Hydrology Symposium, Vol. l-, 29L-298, UNESCO-WMO-IAHS, 1973. t8l camm, J.B., Snow-Compaction Equipment., Snow drags, Tech. Report IO9, U.S.. Nav1z Civil Engin- eering Laboratory, Port Hueneme, California, 1960. tel Clark, E.F. , Expedient snow airstrip construction Army Regions .Abele , G. ¿ technique, U.S. Cold Wuori, A.F. , Research and Engineering Laboratory, Research Report, Hanoverr New Hampshire, L973. 89- I10l Colbeck. S.C., One dimensional water flow through snow, U.S. Army Cold Regions Research and Engin ;ï:i3,:i:;";ïi;åi:::";ä,1:n""'n u' t1ll Colbeck, S.C,, Effects of stratigraphic layers on water flow thróugh snow, U.S. Army Cold Regions Research and Engineering Labor- atory, Research--tött Report 311, Hanoverl *"*-áå*p"ii'" , . ll.zl Colbeck, d.C... Theory of metamorphism of wet snow, U.S. Army Cold Regions Research and Engin- eering Laboratory, Research Report 313, t13l Colbeck. S..C., The capittary effec-.s.on \^rater percolation Ín homogeneous sno\^r, Journal of Glaciology, Vol, 13, No. 67, 85-97, L974 I t14l ColJceek, S.C., Vrlater flow through snow overlying an impermeable boundary, Water Resources Research, Vol. 10, No. 1, LL9-L23, 1974.. {r5]Co1beck:S.C.,Ana1ysisofhydro1ogicresponSe'torain- on-sno\¡¡, U.S. Army CoId Regions Research . and Engineering Laboratory¡ Research Report 34o, Haiover, New frampshire, Lg75. rlol colbeck, s.'c- r;:.:;.i:=:*:"ff::3:,:n:.ru*u , l":ffi:l, vol. 11, No. 2, 26'I:266, L975.. t17l Combarnous, M.4., Dispersion in Porous Media, Advances .in Fried, J.J., Hydroscience (V.f . Chow, ed.); Vo1. 7, :: : :. L69-2AO, Academic Press, New Yor15, L97L. , '.'.'', tISl de Quervain, M.R., .'Snow structure, heat,. and rtrass flux through . snow, The Role of Snow and lce in Hydrology . Symposium, Vol. I | 203-223, UNESCO- wMo-rAHs, L973. tf9 l Diamond, M., Studies on vehicular trafficability of snow (Part 1) ?irctic Institute of North America, april 1956. l2ol Easton, R., Nav)¡ sno\^¡¡compaction equipment-. Military Engineer L964, Vol. 56, No. 372, 4L7-4I9. 91. 1,321 l4oser, E.H. Jt., Roads on snow and ice, Symposiun on .Section of ASCE and Univ. of Alaska, Vol. l, Univ. of Alaska, 1971 l33l Muclear-Chícago Corp. Nuclear-Chicago laboratory manual (r,a¡. Scaler Model 8775'), Púb. No. 7139tO, Des. Plaines, Illinois, 1966. t34l Nuttall, C.J., A study of some phases of the snow/ Thompson, J.G., vehícle interaction, Defence Research Board, Report No. DR 127, Ottawa, Canada, 1958. t35l Putkisto, K., Sno\{ âs a road-buitding material, National Research Council of, Canada, Tech. Trans. 822, Ottawa, Canada, Ig5g. t36l Thomas, M.W. r Snow road construction technique by / snow, l{ava} Civil. Engineering Laboratory, , Tech..Ñote N-1305, Port Hueneme, California, L973- f371 Veda, H., USA CRREL I Snow .and lce Testing Equipment, Sellmann, P., Special Report L46, lJ.S. Army Cold Abele , G., Regions Research and Engineering J - -r- --- [38] U.S: - Canadian Joint Report on snow compaction trials, Winter Compaction Trials, L952-1953, DRB,/BSIS Documents, L25 Elgin Streetr' Ottawa, Canada, 1953. t39l U.S; Navy Snow-compaction equipment, Snow rollers. - Tech. Report 107, U.S. Navy Civil Engin- i' Port Hueneme' cari- :::lï3,"Ë;î:t"'o' t40l U:SJ $awy , Arctic Engineering, Tech: Pub,l. NavDocks TP-PW-II, Dept. of the Navy, Bureau of Yards and Docks, lrtashington 25, D.C., 1955. t41l Waterhouse, R. r. Re-eval-uation of the Rammsonde hardness equation, Special Report 100, U.$. Army Cold Regions Researih and Engineering Laboratory, Hanover, New Hampshire, L966. l42l Ìvelty, J.R., .Fundamentals of Momentum, Heat .and Mass wiòks, C.E., Transfer, John Vtiley and Sons, Inc., Wilson, R.8., New York I 1969 l-431 wuori, A.F. , Supporting capacity of processed snow rtrnivays, U.S. Army Cold Regions Research and Engineering La_beratory, tect Report, 82,iHanover, New Hampshire, Lg62. [4q] Yen, y.C. Recent studj-es on snoÌ^r properLies Advances ,'. in Hydroscience tV..t. Chow, ed.) Vo1 . 5, I73-2I4, Academic press, .New York, 1969. l ' .- ,:.' ; I t; i .:1 n1 ': APPENDIX A RAMMSONDE HARDNESS DATA AND OBSERVATIONS TABLE A.1 Summary of Rammsonde Test Results shown Note: - Rammsonde hardness values coärec+-ed as i-ndicated by- Veda, Sellman, Abe]e t37l are in brackets. (see Appendix C) .. . - * identifies lane with surface recompactecl at snow terperature of OoC. Rariunsonde Hardness No.. Age Hardening No. of B.Iower Date Ti.me Hours (Days ) Passes 0.5 cm. 5-10 cm. 10-15 cm. L2-2O cm. Jan.26 17:00 2. s (0. r) I 23 (107) 31(4e) 1 rs (69) 31(49) z 31(1'44) ss (88) 2 19 (88) 63 (r00) 4 le (88) 63 (rOO) 22:OO 7.5(0.3) I 27 (L26) 35 (56 ) 19 43 I 27 (L26) 43 (68) 51 5I 2 43 (20L) 75 (r20) 67 51 2 43 (201) se (94) 9t- 107 4 39(r82) L29 (2'06) r35 207 4 35 (163) 107 (171) 75 140 6 43 (2Or) ee (r58) 99 150 6 43 (2Or) :t5 (r20) rcis 63 Jan.27 09:30 18.5(0.8) I 83 (38e) 30 (48) 48 51 1 3s (1s0) el (r45) 67 67 ? 3s (163) 67 (r07) t15 203 2 43 (20r) el (14s ) 139 163 4 27 (L26) 43 (68) 67 75 4 43 (20r) 163 (260) 183 153 6 43 (201) r47 (235) 2LL 253 6 43 (201 ) l.63 (260) 2L3 L23 Þ N) TABI,E A.l (continueC) Date 'Time Age Hardening -'-No. of Blower Ranunsonde Hardness No. (Days) - Padses Hours 0-5 crn" 5-10 cm" 10-15 cn. 15-20 c¡n. Jan.27 17 :00 26 .'0 (1. 1) I 27 (rI3) 67 (rO2) 9l LO7 l 27 (r13) 3s (s6 ) 5I 67 2 ,5r(226) 163 (260) a 43 (2O1-) 99 (ls8) 131 l_ 3t 4 43 (zOL) sr (8r) f:s L23 4 51 (238) 7s (120) 139 'o7 6 43 (20r) r07 (171) 195 195 6 43 (201) 99 (1s8) 73 43 Jan. 28 IO :00 43. 0 (1.8) t_ 43 (201) s1 (81) 83 91 I 5e (276) 67 (l-A7) 75 ot 2 s1 (238) s1(81) 5l 115 2 83 (3Be) 235 (376 403_ 393 2 83 (38e) L23 (re6) 133 233 4 sr (238) 131 (209 ) 93 ÕJ 4 83 (389) i63 (260) 83 73 6 75 (3sr) r39 (222) 153 173 6 er (426) r55 (248) 2L3 413 Jan.29 l0:00 67.OQ.e) l_ 67 (T,4) 67 (rO7) 83 83 I 107 (502 ) r39 (222) r87 t87 2 er (426) 163 (260) 283 523 :¿ L39 (6s2) L81 (29e) 153 133 4 sL(426) L53 (244) 153 2L3 4 131 (614) 223 (356) 1S3 153 6 1r5 (s39 ) 323 (s16) 293 153 6 L23 $77) 3s3 (564) 6.40 640 Þ (, TABI.E A.I (continueC) Date Time Àge Har{ening -- No. of Blower Rarunsonde Hardness No. (Dalts) Passes Hours 0-5 cn. 5-I0 cm, 10-15 cm" l5-2O cm. Jan. 30 16 :00 s7 .o (4.L) I 83 (38e) 83 (132) 73 83 I 67 (3L4) 93 (r48) 113 L23 2 sr (426) 7s {.r20)' 83 aiJ 2 r07 (502 ) 223 (3s6) 183 L73 A 7s (3sr) 213 (340.) 2L3 273 4 5e'(276) 73 (116) r33 r03 6 9r('426) )ot ßzs) 163 2L3 6 'ee (464) 23s (372) 313 363 Jan. 31 17 :00 L22.0 (5 . 1) I 43 (2Or) 43 (68) 53 63 : I 43 (201) s3 (84) 113 123 '67 2 (3L4) 83 (r32) 83 103 2 67 (3r4) e3 (148) r33 153 4 67 (3L4) L43.(228) I8 t5 4 83 (38e) 103 (164) 133 l23 4 9r(426) r83 (292) 153 123 6 83 (38e) r83 (292) 243 263 6 9L(426) 113 (180) i3 53 6 7s (351) 113 (180) r33 183 Feb. l 14:00 143.0(6.0) 1 ee (464) LV3 (276) L73 r73 I 131 (614) 203 (324) lq3 73 2 115 (539) 283 (452) 193 183 2 T'79 (840,) 283 (452) 383 423 A 17e (840) 293 (468) r73 133 4 179 (840) 343 (s48) 193 143 6 171 (802) 503 (804) 503 493 o7 23s (1103) 213 (34C) 303 313 Þ è TABI.E . A.1 (continueC) Date Time Age Hardening No. of Blower Ranunsonde Hardness No. (Days) Passes Hours 0-5 cm. 5-10 cm" 10-15 cm. 15-20 cm. Feb.2 16:00 170.0(7.1) I 67 (3r4) 113 (180) 163 153 I er(426) t_63 (260) 163 163 2 13r- ( 614 ) 193 (308) 193 163 2 ee (464) 223 (3s6) -403 4C3 4 75 (351) r53 (244) L73 2L3 4 ss (464) 263 (420) 443 503 '253 6 107 (502 ) (4O4) 603 573 6 l]s (s3e) 2O3 (324) 733 .482 Feb.5 tr5:00 24O.0(f0.0) I ee (464) 163 (260) 223 223 .s2 t (386) ]83 (2e1) 222 290 2 r70 (800) 362 (679) 932 800 2 L7s (840) 603 (964) 943 943 4 136 (640) r:.2 (I7s) '94 L66 4 r84 (86s) Le4 Qe5) L84 202 4 227 (1066) 363 (580) 263 r83 6 r47 (6eO) 433 (692) 353 393 6 82 (386) 2e2 (467) 292 252 Feb.7 12z3O 285.5(11.9) I 50 (236) 72 (11s) 52 72 1 43 (2Or) 73 (116) 103 113 2 86 (40s ) rl-z (r7e) L42 152 2 63 (zes) 73 (116) 113 143 4 98 (46r) IEz QgI) 322 532 A 9e (464) 2e3 (468) 433 593 (3s8) ( 6 76 196 3r4 ) \tz 196 6 77 (36L) 197 (31s) L97 L97 Þ (¡ TABI,E e.1 (continued,) -- Date Time Age Hardening " No. of Blower Ranunsonde Hardness No. Hours (Days) Passes o-5 ¿m. - 5-10 cm- 10-15 cm. l-5-20 cm. Feb. I 16 :00 313.s (r3.r) I 3s (163) .83 (132) 83 lr3 1 5s (276) r23 (L96) 203 183 I 43 (2OL) 83 (132) r03 103 'L73 ¿ '17 (36L) 101 (16r) . 22I 2 s9 (2'76) 77 (L23) 53 L25 2 83 (i8e) 223 (356) 393 393 2 7s (35r) L43 (228) 183 153 4 77 (36L) 77 (L23) L25 10t 4 77 (36L) r49 (238) l4e L49 4 es (44s) L49 (238) 10r L49 6 77 (36L) l67 (267) 77 113 6 77 (36L) rr3 (r80) 113 L49 Feb. 9 12:30 334.0(13.9) I ls (6e) 6e (110) 123 131 I e (4r) 57 (e1) 99 1r5 I 1s (6e) 7s (r20) r39 r47 1 e (4r) 45 (72) 59 75 I e (4r) 93 (148) LO7 L23 1* 19 (88) 43 (68) 93 143 1* 27 (L26) 103 (164) 193 133 f* 3s (163) r53 (244) 303 reg rtb r8 (88) 93 (14.8) 293 ¿¿5 2 rs (6e ) ee (L58) 315 303 2 e (41) 83 (r32 ) 263 263 L e (41) 103 0-64) 273 443 4 e (41) 123 (re6) 193 l-23 6 e (41) e3 (r48) 153 163 6 ls (69) r03 (164) 133 133 Þ Or ,.::ii ,ii j.-'l :lÌ: ,. t. :i, -.''- TABI-E e.1 (continueC) Date Time Age Hardening - No. of Bl,ower Rammsonde Hardness No. Hours (Dalts) : Passes 0-5 cm. 5-10 cm. 10-15 cm. 15-20 cm. Feb. l-0 16: O0 361 .5 (15 .l) I 113 (5 30) r94 (3r0).) 329 248 I e5 (445). L67 (261) L94 140 1* 47 3 (2222) 653 (1044) 545 r67 I* 38.3 ( r79e ) 626 (rOOr) 329 r67 I* 383 (1799) 828 (518) 193 113 ) rr3 (s30) 22L(3s3) l-94 329 2 11-3 (s30) 27s (440)' 464 464 4 r13 (s30) 3s6 (570) 275 22L 4 r3r (614) 2.75 (440) 194 22L 6 131(614) 2'75 (44O) 302 302 6 r31 (614) 3s6 (s69), 356 4LO Feb.12 13:15 406.8(f7.0) I 95 (44s) 146 (234) 113 r40 I 77 (36L) 86 (137) II3 95 l* 329 ('rs45) 680 (1088) .275 113 1* 356 (1672) 383 (612) l-40 ll3 I* 27s (L29r) 248 (396) tol 248 l* 302 (1418) 437 (6e9) 275 L67 2 113 (530) 194 (310) 302 437 2 es (44s) 1r3 (180) 140 'r13 4 113 (s 30), 275 (440) 275 L94 4 131 (614) r94 (310) L40 248 6 L49 (699) 22r (3s3) L67 32 6 r49 (69e) 302 (483) 4LO 5I8 ' : Feb . 15 14 : 30 480.0 (20 .0) 1 95 (445) L40 (224) 113 86 I 77 (36r) 86 (137) 86 86 1* 329 (Ls45) 545 (872). 572 329 l* 32e (rs4s) 275 (440) , t_13 140 (872) 1* 437 (ZOs3) s45 248 L40' Þ (s69) 2 L4e (6e9) 356 329 302 .-J 2 9s (445) r4o (224) 410 49L l.r'1. TABIÐ. a. I (continueC) Date Tíme Age Hardeníng - No. of Blower R.ammsonde Hardness No. Hours (Days) Passes 0-5 cm. 5-1,0 cm. t0-15 cm. 15-20 cm. 4 e5 (445) 22L(3s3) 113 86 4 113 (s30) 22L (353) r6:7 22r o r13 (s30) 194 (310) Lg4 248 6 1r3 (s30) 27s (44O) 275 22L Feb. 16 16 :35 so6.o (2L.r) I 77 (36'r) :-67 Q67) 113 86 I 77 (36r) 86 ( 137) 32 .59 1 se (276) 5e (94) 32 86 1t( 302 (1418) 464 (742) L67 59 1* 275 (L29L) 248 (2e6) r13 59 1* 275'(r2eL) 464 (7 42) L94 59 ¿a r13 (s30) 410 (656) 464 356 2 113 (s30) L40 (224) 248 383 2 71 (36L) 113 (180) 86 r13 4 77 (36L) r94 (310) L94 194 4 ir.¡ (sso) 22L(353) 22L l-94 6 95 (445) r94 ( 3r0) 248 329 6 rr3 (530) 275 (440) 410 518 6 77 (3çL) L4O (224) L46 248 Feb. l-7 13:00 526 .5 (2L.9) I s9 (276) r31 (2oe) L67 L67 I 41 (r91) 77 (r23) 77 77 I se (276) 59 (94) 59 5g l* 18s (868) 22r (353) 86 32 l* L67 (7e4) 22L (3s3) i:t 113 2 '77 (36r) 113 (180) r40 22L 2 7'7 (36r) 194 (310) 140 86 4 77 (36L) 1e4 (310) 22r 140 4 59 (276) se (e4) 1t3 L94 4 77 (36L) 328 (526) 275 140 Þ 6 77 (36r) 328 (526) 518 626 @ 6 77 (36L) LAO (224) 86 22L b 5e (276) 113 (r80) 248 r6'7 TAEI.E 4.1 (continued.) Date Tinie Age Hardeníng- - No. of Blower Ranunsonde HardrÍeSs No. (Days) :'- Passes Hours 0-5 qn. 5-10 cm. 10-15 cm. 15-20 cm. Feb. 19 17 :30 579 .O (24:l) l' 77 (36t) 1r3 (180) 140 rl3 I se (216) L4O (224) l.67 86 .l* 27'5 (r29L) 3s6 (s6e) 248 ]40 1* 27s (r2sr) 2'75 (44O) L94 I4O 2 113 (530) 324 (s26) 437 410 2 e5 (44s) L67 (267) 140 24A 2. s8 (276) 113 (r80) 140 113 4 9s (44s) 22L (3s3) L67 248 4 113 ( s3o ) L67 (26'7) 59 59 77 (36L) r40 (224) 113 22L 6 77 (36L) ]-67 (267) L94 22L 6 e5 (44s) 1e4 (310) '275 L67 (276) se (e4) 86 86 Feb.20 17r30 603.0(25.1) I s9 . 1 se (276.) 86 (137) 86 86 I es (44s) 194 (310) 22L 1.40 l* 275 (zLeL) 383 (612 ) l-67 86 l* 329 (Ls4s) 32e (s26) 113 86 2 77 (36L) 86 (r37) 113 I40 2 .77 (36r) 86 (137) rl3 1t3 4 113 (s30) 194 ( 3r0) t4o r40 4 es(445). 113 (r80) 22I 356 6 95 (445) L4O Q24) 86 59 6 131(614) re4 (3ro) 22I 22L 6 113 (s30) 275 (440) 464 464 Feb.21 15:15 624.8(26.0) I ss (44s) 1e4 (3rO) 140 r40 I e4 (44s) 275 (440) r4o 140 't* '1* 32e (L545) 356 (s6e) t6l 113 302 (1418) 1e4 (3r0) l-40 140 Þ (o TABI.E 4.1 (continueC) Date Time Age Hardening= No. of Blower Rarnnsonde Hardness No. Hours (Da1zs) Passes 0-5 sm. 5-I0 cn" t0-15 crn. 15-20 cn. 1 131 (614) 275 (440) 356 356 2 95 (445) rr3 (r80) 86 59 2 113 (s30) 167 (267) 194 L94 4 167 (784) 27s (44O) 4LO 248 ' A 113 (530) 86 (137) 5g 86 À 9s (44s) rr3 (180) L67 L94 6 13r (al-4) 22L (3s3) L94 329 6 L67 (784) 248 (3e6) 2t-l 140 Feb.23 16:30 674.O (28.1) I 77 (36L) L6:7 (267) L94 140 1 4r (191) se (e4) rl3 l-94 t 77 (36L) se (e4) 59 86 1* 113 (s30) 410 (6s6) 140 113 I* 131 (614) 572 (9L5) 383 95 l* 239 (LL22) 437 (6e91 302 rl3 2 es (44s) lr3 (180) t13 140 : 2 77 (36L) 86 (r37) r40 113 ,:1)j: 4 77 (36L) 22L(353) 22L l-94 .:':; ir.: 4 77 (36L) 22r (353) 329 329 ':;., :.¡i 6 77 (36t) 194 (3rO) L40 L67 .il¡ 6 e5 (44s) :-67 (267) 22r 22I îeb.24 16:00 697.5 (20.1) I 23 (L07) 4r (6s) 77 77 1. 4L(LgL) 4r. (65) 4L 4L 1t( 248 (Lr-64) 680 (1088) 464 248 1* 1e4 (9rO) 302 (483) 140 86 1* r4-a rcs7). 329 (s26) 140 59 2 59 (276) 27s (440) Lg4 383 2 77 (36r) 302 (483) ' 329 464 'L67 4 59 (276) L67 (267)' L67 Þ 4 77 (36L) 22L(3s3) l67 140 H 4 77 (36L) 113 (180) 22I L67 o o 77'(36L',1 248 (3e6) 22L 22L 6 es (44s) L40 Q24). 140 .r67 rAurü;. A. r (oontanuec, Date Tíme Age Hardeníng - -. No. of Blower R.anunsonde Hardness No" Hours (Days) FaSseS .0:5 cm. 5-10 cm- 10-15 cm. t5-20 cm. Feb.26 17 :00 746.5(3r.1) I 4r (rel) se (94) 95 113 I 5e (276) se (94) 77 77 1* 27s (L29r) 248 (396) l-61 86 l* l67 (784) LAO (224) 32 59 1;t 383 (1799) s99 (9s8) 545 275 r1 86 (403) 86 (137) 1{3 86 2 59 (276) L67 (267) 194 1t3 I 77 (36L1 L4o (22+) Lg4 1r3 4 .76 (36L) rr-3 (r80) 86 86 4 95 (445) e5 (L52) r67 302 4 77 (36L) L6'7 (267) 22L 140 6 . e5 (445) 194(3ro) L94 22L 6 4r (rel) L6t7 (261), 49L 518. 6 77 (36L) 194 (310) 22L 302 Feb.29 15:00 816.s (34.0) I 113 (s30) L67 (267) rt3 113 I 95 (445) LAO (224) r67 r4o 1 77 (36L) L40 (224) L94 86 1* 248 (t 165) 572.(9Ls) ,302 r13 lx 302 (r4r8) Lr39 (L822) 410 r40 1* 653 (3068) 8ls (1304) 733 22r 2 r31 (614) 518 (828) 653 599 2 203 (953) 464 (7 42) 626 76L 2 r49 (699) 383 (612). 49r 599 4. r8s (868) 410 {656) 4LO 275 L 131 (614) 302 (483) 248 248 4 167 (784) 383 (612) 329 248 6 r31 (614) 32e (526) 275 22r 6 r49 (6ee) 383 (612) ' 545 329 Þ ts P TABIE A.l (continueC) Date Time Age Hardening' - No. of Blower Rammsonde Hardness No. (Days) Passes Hours 0-5 cn" - 5-10 cm" 10-15 cm" t5-20 cm. Mar.. 4 . I7:OO 9L4 .5 (38. r) I se (276) 86 (137) 86 113 I 95 (445) rr3 0-80) L67 1t3 f* 545 (2560) 788 (1260) 248 86 l* 4sL(no6) 4eL(78s) 22I 86 I* ssg (28L4) 13òr (208r) _ s99 L94 l* 356 (L672) 599 (9s8) 248 86 2 r3r (614) 410 (6s6) 599 545 2 l3r (614) 302 (483) L67, J_40 2 r3r (614) 194 (3r0) L40 I40 4 .13r (614) 248 (396) L94 L94 4 es.{44s) 464 (7 42) 275 L67 6 13r (614) 302 (483) 22I L94 6 r49 (699) 2V5 (44O) 275 ¿ t5 Mar. 5 17:00 938.5 (39;f) I .95 (445) L61 (267) r40 113 I 77 (36L) le4 (3ro) L6l It3 1* 4ro (1926) s45 (872) 383 L94 1* 383 (1799) t=l39 (1822) 1058 329 1* 32e (1545 ) 437 (699) 329 22I l* 4ro (re26) s23 (r476) 572 194 2 rLe (699) 383 (612) 248 194 2 r13 (s30) 41e (78s ) 5'72 329 2 r-31(614) 27s (440) .22L l-94 4 13r (614) 302 (483) 383 437 A 13r (614) 54s (8i2) sl8 250 6 LAs (6ss) 383 (612) 356 22L .6 LAg rcsg) 410 (6s6) 545 707 Þ ts N) TABI.E A.l" (continueC) Date Time Age Hardening No" of Blower Ransnsond,e Hardness No. (Days) Passes , uours O-5 cm. 5-10 cm. 10-15 cm. 15-20 cm. Mar. 7 14: 30 984..0 (4r-.0) I 77 (36L) 167 (267) t94 140 1 9.4 (445) 167 (.26V) tr3 140 1* 437 (20s3) 1895 (3032) 977 680 (2306) 1t 49L r-19'734(Lr74) 3 11908 ) a37 194 1* 3s6 (L672) 62þ 356 2 e5 (44s) s18 (828) 680 653 2 13r (614) 437 (6ss) 518 6.26 4 r31(614) 32s (526) L67 248 4 r3r (614) 248 (3s6) 22L L67 6 l-67 (784) 491 (785) 599 545 6 - 113 (530) L67 (267) 356 L94 6 L49 (6ee) 302 (483) 248 302 Mar, 9 17:30 1035.0 (43. r) I 5s (276) 22L (353) l-94 140 I . 59 (276) 86 (137) 59 59 t 41 (re1) r40 (224) 86 trô I* 464 (2L79) 49t (785) 22L 86 l* 302 (r4r8) 383 (612 ) I40 !40 l* 32s (Ls4s) 896 (1433) 545 275 2 77 (36r) 167 (267) L40 140 2 13r (614)! 275 (440) 140 113 4 77 (36r) 22r (3s3) 59 32' 4 95 (44s) 248 (396) 275 330 4 77 (36L) 27s (440) 248 L67 6 13r (614) 248 (3e6) 302 r6:1 6 1r3 (s 30 ) 356 (569) 302 302 Þ ts UJ -;:.j.'i.. .. TABL,S 4.1 (continueC) Age Hardening No. of Blower Ranunsonde Hardness No. (Days) Passes Hours 0-5 cm. 5-10 cm. 10-15 cm. 15-20 cnt. Mar.12 12:30 LLO2.0 (45.9) .t 58 (273) rr2 (180) 139 85 I 76 (3s8) r-e3 ( 30e ) 85 58 1* 328 (L542) 760 (L2L6) 247 LI2 1* r48 (6e6) s44 (87r) 328 139 1* 184 (86s ) 679 (1087) 24'7 L12 2 13r (612) 247 (3ss) 220 L66 ¿ 131 (612) Aeo (7e4) 409 409 2 131 (612) 382 (611) 463 382 4 131 (612) 247 (395) 166 274 4 r3r (612) 22O (352) r66 193 o Lrz (s27) 193 (30e) 139 139 6 LL2 (527) 274 (439) 220 193 Þ ts 'S RAMMSOT\iOE HARDNESS N9 (X 102 ) 3 10 uncorrected .a" 6 ,)' d8 o 4 r- 10 CL t¿¡ ct 12 14 16 18 20 Þ È (tl Figure A-l: T\zpical snow road hardness profiles as indiqated by the corrected and uncorrected versions of the Rammsonde hardness data. TABLE A.2 Summary of'air temperáture,- snow road temper,ature and relative humidity for duration of namlnsonde testing. (eii temperatufes reflect maximum and minimum fluctuations. ;;;;;ã";d;;;;;;" ;;;"-;. 10 ;;'aepttr .. .i*" orìestins). REL. HUMÏDITY % DATE TTME AIR TEMPERATURE OC ,SNOV\I ROAD TEMPERATURE ."C MAx. .MIN. Jan. 26 08:00 -29.9 86 64 17 :00 -26.O -16.0 22zOO :30: 0 -19. 0 Jan. ¿ l' 00:00 -28.6 91 65 09: 30 -22.O -22.O l-7:00 -L2.O -16.0 ìlan.28 00 :0O - 4.1- 89 67 05 :00 3.3 10 :00 -10.0 - 9.0 17:O0 :13 .9 ,J Arr. ¿Y 0l:00 -20.4 a7 72 l-O:00 -L5.2 -14 .0 15:00 -11 .6 2l- :00 -14.3 Jan. 30 01:00 ' -rg.'1 93 77 06:00 ' -ts.z 16:00 - 9.4 -10 .0 " 21:00 - 3.1 Jan.3t 06:00 8.9 93 1I l0 :00 5.4 14:00 1.1 17 :00 5.6 6.0 - P ts ol TABTE 4.2 ('coñ:trnued) OC REL. HUMTDITY A DATE AI.R TEMPERÀTURE OC SNOV.I ROAD TEMPERATURE TIME MAX. MIN. Feb. 1 OO:0O -20.6 7.9 64 07:00 -30.4 14:00 -22.8 .17. 0 20:00 -2L.8 Feb. 2 02:00 -20. 0 89 77 l0 :00 -14;1 16:.00 -10. r -11.0 22:OO -L2.L Feb. 3 05: O0 -14.0 85 62 08:OO -L4.4 14 :00 -20.7 Feb. 4 0O :00 -25.6 e7 60 07: OO -28.4 14:00 -L9.4 22 zOO -23 -8 Feb. 5 00 :00 -26 -3 87 7I 08 :00 -29.7 15:00 -18.1 -19. 0 Feb. 6 OO: OO -L4 -6 78 58 04:00 -13.8 O8:OO -Li.L 16 :00 -10.1 Feb. 7 00:00 -1I. I 90 69 06:00 -2) L2 |30 - 2.O : 6.0 20 :00 - 4.r Þ ts ! '.:.::, TABL¡j -A. 2' (ConËrnuectj REL. HUMIDITY B AIR TEMPERATURE OC SNOV,I ROAD TEMPER.ATURE OC DATE TIME MÀX. MIN. .b'eþ. u 02 :00 - 8.6 90 76 10: OO - 6.3 16:00 - r'.7 - 5.0 22:OO - 0.9 Feb. 9 04:00 '4.93.8 77 57 i2:3O 0.0 20 0.6 :0o -Feb. 10 04:,00 - 9.5 82 59 09 :00 -13.3 16:00 -16.0 -10.0 22 zOO -23.I Feb. 1I 07 :00 -24.8 88 7L L2:QO -ls.5 18: 00 - 9.2 Feb. L2 02:00 -r2.o. 93 75 08:00 -10.3 l3:15 -10.0 ì11.0 20:00 -15.5 Feb. 13 00:00 -18. l 85 6L 08:00 -r7'.2 17 :00 + 9.4 22:OO -r3 -4 Feb. I4 06 :00 5.0 90 75 16:00 o.4 22zOO 2..8 Feb. 15 06:00 5.0 93 77 Þ ts 14 :00 7.4 - 7.O @ 20:00 8.6 ieeLn À.2 (êõntinued) REL. HUMTDITY 8 .TIME OC SNOW ROAD TEMPERÀTURE OC DATE ATR TEMPERATURE MÀX. MIN. Feb- 16 0O: OO '11.0 98 83 06:00 -15.6 16 :35 . - 2.O - 8.0 22zOO Feb. I7 02: OO o.o I00 9t 13:00 ,0.6 - 5.0 18:00 0.0 Feb. l8 00:00 - 6-8 96 89 08:00 - 7.2 16 :00 - ._ 5.6 22:OO - 4.4 Feb. 19 02: O0 - 5.0 vö 90 13: ÔO - 2.r 17i30 - 3.0 - 3.0 Feb- 20 0O :00 - 4.9 95 74 08:00 -L6.6 14 :00 -10.4 iz::o -11.6 - 6.0 Feb. 2I óo,oo . -16.7 93 70 08 :00 - -24.9 15: l-5 ' -iz.t - 9.0 21:00 -1',0.r0 Feb. 22 02 :00 -20.6 93 78 08:00 -24.6 16: OO -L2.2 Þ ts \o . ì:ì TABLE 4.2 (conÈt-nuéd) : .: i.t' ':;i;i ', i!, 8 .:ì,1 OC ôC REL. HUMÍDITY ÐATE TTME AIR TEMPERATURE SNOT¡T ROAD TEMPEB.ATURE MAX. MIN. :lii':,ilr '',.ii '.1,: Feb. 23 Q2 :00 4.6 90 60 :,; 07;00 6.g : l;,j 16:30 7.O lt:a\i 1:u - :..\.,: ,:;::ii 89 62 ,)ii Feb. 24 00:00 L.¿ ': ';1r 07 :00 6.L 12:00 0.9 riii 16 :00 3.0 - 6.0 '. ti 22 zOO L.6 ,rj:,.,1i: :::;:.ì 90 67 ' Feb. 25 02 :00 0.0 :"11.: r 1(ì) 08:00 o.J ,..i.jj ...!j.:iiÌ 15:00 I.2 ;:-:l ',,i1 Feb. 26 00¡00 t2 94 69 ','ti 12:00 5.1 :jli:i]:; - .:ì;jtl 17: OO O 4.O -II. - r,ii: :'1,. Feb. 27 OO:00 -13 .8 89 71 i): :;iir 08:00 -ll:9 ìi: . :;, ì:ali l-8:00 -13 .9 .t::i: l:i. I j Feb. 28 00:00 -22.7 88 69 ,:j 'l: 04:00 -2V .2 'rii:i :;i.i'i 12 :00 -r7 .2 ì: i.;, 20:00 -:l-7 . B ..ir; .: i.l' Feb. 29 00:00 -r1 0 86 62 jj) 07 :00 -30.7 'rl;il l-5:00 -2L.8 -15.0 :iiiì 1:: 22zOO -26 -2 :.tlriì ' :1.'iì Þ .f N) l':ji tr:,i..J: 'i'Ì ':rì:j :.1'jì.;¡,: :i.." l;,:,: .ii ''',:i 'lÌ,ì, :;Ìì: 'i:ìl r ;lr, :i¡ LæÞÞ N. A T OC OC REL. HUMTDITY DATE TIME AIR TEMPER.ATURE SNOW ROAD TEMPER.ATURE MÃX. MIN. Mar. I O2:00 -zo-o 86 69 08: OO -28.8 19':00 - 8.7 Mar. 2 00: OO -rr.1 9I 82 16:00 - e.9 Mar. 3 00:00 -I0.8 91 63 09 :0Q -20.4 rQ i00 -16:1 Mar. 4 00:00 -22.O 95 64 08:00 -24.4 17 :00 -18.0 -L2.O Mar. 5 00:00 -28.7 82 6I 06:00 -31.6 17:00 -22.0 :f7.0 Mar. 6 O0 :00 -19.'I a7 67 I0 :0.0 -ro.7 16:00 - 5¿6 Mar. 7 OO: O0 -L2.7 84 63 06:00 -L9.2 l-4:30 - 7.O -13.0 2l:00 -1r.1 Mar. I 0I:00 _oo 87 oø 06 :00 -L6.9 l-3 :00 - 8.3 20 :00 -L2.2 Þ NJ P TABLE 4.2 (continued) REL. HUMIbTTV Z DATE TIME A'IR TEMPERATURE OC SNOI¡I ROAD TEMPERATURE "C MAX. MIN. Mar. 9 02 :00 :l_2.8 85 66 l0:00 -L7.8 17:00 :16. o -r0.0 Mar. f0 00:00 -26.L 86 59 05 :00 -29.6 16:00 -r3.2 Mar. l-t 00: 00 -L7:I 89 7I 06:00 ,L4.3 16 :00 -11.8 Mar.12 00:00 -14. I 86 55 _ 08:00 -20.L 12 :30 -13 .0 -L4.O 17 :00 - 9.8 Þ N) N) A-23 (\¡e \ -Ètq)- sf '\ N lr \ ..{ \ \. a^ .¿t (\¡ ! dÉE rd l.{ o .o: \q{ ...{ÉrJ o ÐÐo) U bì(l) ç $.1 I'.1 ! (,UËo .ú trO . rd.( t = j S lJ. .t oÉ !" ul'.r. : ) i d. I õ tlt \o -\ ^ 'lJ ')), :.L I :'om--Èiu ''rôg-o ct. '^. ô. g o(I) c¡t Ð tt z, v(d z úo t¡¡ ul c, -.I (\¡ cr 0)5 ä 'çrE -l :t. ' rd .Cu(t) r¿¡ 0) o) (9 õÉ c¡ oH É.(u.d rdoE Érú doÉ \ oÉoül I 3E {J(Úoú --'/' ô.v ' \ ^-'¿ Or. \_> ¿_-_._._r_ ..1 .duc) .- _E- o IDã . ordÉ+r ..J $ / lJ c) .rErd gl úÐoo ñ I 'o jt{ .bl L._ ..1 , 'Êl oo oo o o. 6r oo li att ql îâ + lr I 'd 0o l,ìl I I (zolx)öN sslN0uvH t0N0sllt/ì¡vu + t0 0 tj .Ét \ /v \ d-10 \rt\ l- I \Ui"uir-.. = ,í\. i \ j"t I y -20 tr' , / t I \u' tV I t \u,na. , I \ I -30 J v 10 C\¡ I o x I c'¡ z, 7 '-. /' ^.\ i.i,l cn \.-.---'-'-. ,t;i v, /' -r:.: .t.i: 6 _-_. z -- _ E¡ E /\ -'\. ! .'?' t¡¡ É¡ I z,o ,\. câ \ i = \.! = \/ aÊ teb.20 \/ Mar. I Mar..l0 ú I ,t ,l '38 44 46 48 24 26 28 30 32 34 36 40 42 Þ AcE HARDENING (Daysl ¡È Figure A-22 Continued; :,::': ',,.'1;,, :,..1., l4 t2 RAM- NP = 215.65-30.32 T Corielation coef. ( r) = 0.78 l0 a\¡o x '. z8ò. q v) ' l¡¡ .3 oz, fr0É r¡J c¡ o= v,E4 .É= -2O -t8 :16 -r4 -12 -10 -8 -6 -4 -2 SNOW BOAD TEMPERATURE(T) OC Þ (toP 5 data. NJ Figure A-3: Scatter diagram of, Rammsonde hardne'ss cm. ) and snow road temperature LN (Rammsonde hardness , values are presented in' the corrected form. ) TABLE A.3 Averagie.densities of the single- and Multiple-þroeessed snow roads a No. of Blower Snow-Core Density (gm/cm-) Averäge Passes 1 o.452 0.450 0 -454 o.463 0,459 o .47I 0 .458 2 0.481 o.4a7 0.489 o-482 O,479 o,.47r o.4a2 0.520 0. 502 o.532 . o.494 0.506 0.512 0.511 0.510 0.530 0.508 0.502 0.517 0.506 o.5r2 P t\J Or :,-.;:,'.:.. )::":... :,,:.ri ':r' : ::". ' . : . )... RAMMSONDE HABDNESS N9 (* ro2 ) t2 ¡6 20 24 28 32 .¿ -'o- oc no compact¡on at 0 / -t' ./ ..? .1 / -/ 2\ /a / oC / \previousiy compacted at 0 o .a' (¡ l ! '/ .a CL t0 I ci 1., t2 Jli .'1 l4 li' t6 ot/ ( t8 20 Þ N){ Figure A-4: Tlpical increase in snow road hardness due to surface gompaction at OoC. (Rammsonde hardness values are presented Ìn the corrected form. ) .: : .- - ---.-t-.-. lii 'I.,:: .il:.l.: :. t.. B.I APPENDIX B EXPERI¡4ENTAL RESULTS FROI4 BURN TEST rii j iii TABLE B-I r,,.1 ,;i; ,i¿i Sumlnary of dnow-road density profiles before burn test (Mar. L7/76) ¡¿l '? : t:ii ¡,,i Snow Road Density (gm/crllr) i.:) .:) , ii, i:i li. Il, Core\ Depth 1.5 2.5 5.0 7.5 10. 0 L2.5 14.0 iii ¡:\ No. \cm) ' 'li i. ,: l. I':, l:j: 0.535 o.522 0.480 0.525 0.480 0. 390 :¡i ,:.1. 0.513 0.525 0.510 o.460 0. 385 0. 360 ,l.i ì;,1 ,rl o.496 o.525 0'.560 0.430 0 .425 o.420 . i:, .': il rii 0.485 0.500 o.4g5 0.450 o.475 0.375 ::¿;: ,ìi) i;ir ii o 360 5 0.500 0.530 0.530 o .480 .473 0. lrìj iir: \l b 0.405 0.510'; 0.545 o .4go o.425 0. 370 .;ìÌi:r rl: 7 o..475 0.463 0.385 0.410 I'i Q.481 0.460 i:iì ¡ii 0. 515 0.500 o.420 o. ¿ro o 0.431 0.440 i,,i j::ì Ì.."i:.'!ì Average 0.481 0.502 0 .514 o .475 o.434 0. 387 t:, r.) t.ti ': ' t;¡ rli ,::t, 'ì;ì, ti!, I./. 'a: il: /:i: Ed ' ..:,1 N) lÌ. ::, l'.t i:l il ' iì, i:il i:ii ¡:, ::11 iii !iì .,.;,i ila'l-.t.1 .riì TABLE .B-2 'Srunmary of snow road. density profiles after burn test (Mar. L7/76) .. 2 Snow Road. Density (smlcql) I min. burn I\ 0.550 0.661 o.sez 0 .600 0.401 o.394 .2 o.645 0 .590 o. 545 o.495 0. 500 0.450 3 0.525 0.600 0.520 0. 550 0.520 0.435 4 0.565 0 .585 0.555 o. ozo 0.490 o.4'75 5 0. 620 0.580 o.525 o. 480 o.420 0.380 6 0. 600 0.580 0.590 0.525 0.45s 0.375 Average 0.584 0,599 0.55 3 0.554 o.464 0.418 2 min. burn I 0. 590 0.680 9. 585 o.495 0.490 o .440 2 0.61_5 0.630 0,550 0. s05 0. 500 0.410 3 0.550 o.570 0. 485 0.455 0.370 0.480 4 0.615 0.615 ô. szo 0.525 0.435 o .475 Average 0.593 o.624 0.548 0.495 o.449 o-451 4 min. burn 1 0.640 0. 640 0.540 0.435 0.430 0.415 0. 315 tú ?. 0.620 0.595 0.580 0.535 c.555 0. 480 0.455 UJ a o .415 0 .510 o.640 0.585 9. 505 0. 460 0.370 Average 0.578 0.582 0.587 0 .518 o.497 o.452 0.380 TABLE B-2 (continued) Core\'Depth 1._5 2.5 5.0 7.5 10.0 72.5 14.o No. \cm) 6 min. burn I 0.615 o.645 0;6.00 0. 515 0.450 0. 415 0. 400 2 .o.630 o.675 o -625 0.520 0.465 0. 490 0.500 3 0.745 0. 735 0.450 0.420 0.405 0.345 0. 3lC 4 0. 615 0.705 0.660 0.535 c. 490 o. ¿rs o.4r2 Average 0: 651 0.690 0.584 0.498 0.453 o.4L6 0.406 8 mi.n. burh :t 0.635 o.675 0. 590 0. 405 0. 360 0. 415 0.458 l z o.670 0. 680 0.585 0. 535 0.480 o.460 0.450 ) J 0. 680 0.735 o. 700 0.490 0.435 0. 380 o.405 Average o.662 o.697 o.625 o .477 o..425 o. 418 0.438 12 min. burn I o.625 0.680 0.580 0 .550 o.495 0. 450 0. 450 ) 0.630 O:675 o .495 0.435 0. 415 0. 360 O.3OO 5 O. 7OO 0.710 0. 680 o.625 0.570 0.525 o.465 Average o.652 0.688 0.585 0.537 o.493 o.445 0.405 20 rnin. ,burn 1 0.655 0.760 o.645 0.615 0.570 0.525 0.670 Ed 2 0.600 0.585 0.590 9.585 0.Þr0 0.450 0.375 . È 3 o.670 o -640 0.515 0.490 9..480 0.415 o. 390 Average o -642 o.662 0.583 0.563 0.520 o. 463 o.478 '')'l¡. -:.::. 0ENSITY (gm.zc.c.) 0.0 0.1 0.2 0.3 0.4 0.5 : -- 0.6 0.7 0.8 0.9 1.0 0 1 o 2 \ o F 3 4 5 ì (,E6 -7 CL Êr8t¿¡ g 10 Burn timg; o 2 min. 1l o 6 tnil. t l2 min. 12 d'ensity jrange .13 lnitial 1+ 15 bJ ctl Figure B:1: Snow road density profiles before and after búrn tests. 8.6 Sample Celculation of Heating Effíci.ency General datá: Snow temperature -9.00c Air temperâture -9.00c .: Effective depth of density change 14.Q. cm Thermal'value of seasoned pine 8000 Btu/Ib Averagte burn rate L9 Lb/mln ) HeaXer/snow contact area 10,O00 cm- l Gross heater output, 3,800 caL/mhn/cm2 Specific heat of snow O -5 caI/gm/oC Latent heat of snow 80 callgm The efficiency calcufation is based on the average densily increase observed in a column of snow road, I .*2 in surface area by L4 cm deep. In this particular calcul-ation, the 6 minute burn interval ís considered. This burn interval, for the surface heater used, is equivalent to a forward speed of approximately I ft/min. The average density in a snow col,umn l-4 cm deep, before and after the burn test (6 min.), was.O.48 .i??i gm/cm' and 0.55 gm/cm,' respectively. Therefore, to bring about the densíty increase. the following heat exchange vras required (1) Heat required to raise the average snohr temperature from -9."C to OoC; O.5 cal/g4t/?ë x O.4B gm/cm3 x 14 cm x 9oC = 30.24 caL/cm2 .t I Q) Density increlase (o .48 g*1"*3 to 0.55 gmTcrn3¡ is equivalent to the addition of I grm of 'water to the, 14 cm coluinn. Ttrus, the heat rèquired = 80 ca]-,/cm¿, Therefofe by the definition; l:'- t-: 8.7 (Energy theoretically required to achieve the observed density increase) (ToËal energy expended during the burn interval) 30.24 +' AO 3,800 x 6 = 0.005. c.1 i , APPENDIX C THE R.AMMSONDE CONE PENETROMETER :::.:.. c.2 THE RAMMSONDE CONE PENETROMETER The Rammsonde cone penetrometer consists of a hollow 2 cm-dianieter aluminum shaft with a 6Oq conical probe, a guide rofl and a drop hammer. The. total length of the standard probe (to the beginning of the shaft) is 10 cm; the cone lip having a diameter of 4 cm and a height of 3.5 cñ ..The stand¿rd Rammsonde penetrometer 'A kit contains two drop hammers, 1 kg and 3 kg in weight. proper combination of hammer weight and drop height'(range 0 to 50 óm) allows a suitable'rate of penetration in a variety of snows. l Figure C.1: Diagram of Rammsonde cone penetrometer showing the various .components of the instrument. The hammer is raised by hand to a certain height (H) whích is read in centimeters on the guide rod (Fig. C.1) and.then dropped c.3 freety. The depth of penetration is read from the centimeter scale on the shafÈ. . The resistance to penetration (hardness) of '! snow can be determ:i-ned by observing either the amount of the pene- tration after each hammer drop or the number of hammer drops (blows) necess-ary to obtain. a certain penetration. The hardness reading at any depth, obÈained'when the tip of the cone is at that depth, 'i ' represents the mean hardness through the depth incr:ement between this and.the previous reading. The Rammsonde hardness is computed i-.:-.t :.: from the foll-owing êxpression. R =lwHn + w +. e X where R = Rammsonde (ram) hardness number, kg., W = weíght of drop hammer, kg. , ' :. H = height of drop, cln;¡ n = number of hammer¡ blows, tnt x = penetration after blowsr crlt., and ,,.,..,,'i Equatíon (1) i-s a simplifícation of the real case and is j,,,,'.;,',;:.,;,',,.,, , not completely satisfactory for êvery situation encountered. Re- eval-uation of the equation by Bender, Brunke and Niedringhaus (Veda; Sellman and Abele t37l ) tras shown that because of the conical i ri shape of the penetrometer head and the proximity of a free surface, the resistance number obtained requires correction. The'resistance ' number (R), for the 0 to 5 cm depth has to be multiplied by 4-.7, l i c.4 that for the 5 to 10 cm depth by 1.6,. or that for the O to lO cm depth by 3.0. It has not been stand.ard. practice, however, to I apptry these corrections, and therefore, the râm hardness data and correlations to the strength parameters as presented'in the liter- ature, are generally in the uncorrected form. A further limitation of the Rammsonde cone penetrometer in its present shape, is.that it becomes unreliable as an indicator when hardness values exceed 800 kE.. Although R has units of kiJ-ograms, the magnitude of R, - \ ,,,,,.,, l.::': : t:'::: : , except in relative terms, has no real significance. The foli-owing reûiew of Rammsonde hardness equations will serve to illustrate , the rather arbitrary nature of equation (I) and. its limitations. I I The Original Rammsonde Haijþ-ess Equation .:, The original ram hardnesè equatíon was developed by reasoning that the kinetic energy of the falling hammer equals the mean force ' Rrtímes the penetration S. That is; ' 2 ;! !úv1r Rs- zs,Q) where i !,I = weight of hammêT t. V, velocíty of hammer uFon impact, I. = S = penetration into the snow, ,and g = gravitational constant. By substituting the identity. C.5 i*i?= m eH (3) where ' i. . m = mass of hammer, and H ='height of hammer fall. into equation (2), the original equation was derived as R=- WH (4) S Equatíon (4) proved unsatisfactory as the impact effect is not accounted for. That is, W is not the weight and V- is not the 1 velocity of the sysúem in motion during the penetration. rt is interesting to note, however., .that the right hand side of equation (4) is also the first term in equation (1), Nie*ringhaus equation fer ram hardness The momêntum of the system after impact was accounted for l-'-'.-'¡that, W Ul = WrV2 (s) where I VrI_ weight W * T =.total = Q, 'Y ô = weight of pehetrometer, and . \/: .conmon ' 2 = velocity.after impact,*-:-s-- -' of the ' ' penetrometer and hammer. The kinetic energy of the system' following the impact of the ,::'i') c.6 hammer is therefore' 2 ,wrV2 RS !. (6) ¿g Substilutin9 V, of (s) equation (6) gives; .equation -into '1) ütTv[-v1- (7) RS -) 2sw; and since t v; = 2sH (3) equation (7) becomes I WH VT (8) ft= SWr Equation (8) (Niedringhaus equation) was evalgate.d as follows (waterhouse [41] ) : Since a conìmon velocity for the penetromeþêr ,and hapmer (after impact) \^tas assur4ed, tþe collision is treated as ínelastic. .Later tests showed that this assumption r¡/as not valid and thaË R varied considerably with changes in hammer weight. t I Equation (8) was consequently discarded. : i Haef eli rqm hargggss_s.guat¿eg .' In an attempt to correct the Rammsonde hardness equation deficiencies, ttaefeli (1963) developed an equation thát included the coefficient of restitution for the harnmer-penetrometer collision. The ram hardness was expressedÂüñrôêêô,{ as;âc. w+e2w VÍH T (e) s w+wT c.7 where e = coefficient of restitution, and Other symbols as previously defined. .l AS with equation (8), Haefeli's equation could not be generally applied as it too was sensitive to variation ín hammer and total weight. tliley riam hardness. equation The.Hiley equation is bäsed on pi.1e driving technology and (i4 the preserit nomenclature ) is written as: ) WH W+e-Q R=-;r (10 ) t v{+Q Comparing to equation (9) we can see that w: \^ras replaced by Q. T : Waterhoùse t41l indicates that utilization of equation (10) ,for:the ram number would permit future correlation over a broad 'r.ange of materials and conditions; r . As was evidenced in the review,,the limitations of the dynamic formulae resuft from the inabili-ty to account for energy: foss.es ouring conê penetration. Since aecorlntability of atl losses is impossible, a certain margin of error is inevitable. l.-: !.:.....' j''"I ' I D.l APPENDIX D DETERMINATION OF SNOI/ü DENSITY USING RADIATTON MEASUREMENT TECHNIQUFì D-2 DETERMINATTON OF SNOW DENSITY USfNG A RADIATION MEASUREMENT TECHNIgUE Equipment and Materials Used I 1 Nuclea.r-Chicago, Model 8'175, scaler/timer Scal.er: - 5 decade read.out 99999 maximum count caPacitY ': preset ceunt capability Timer.: - 4 decade readout ' '-' .i '- preset time capabilíty I l- Harshaw, Model DS-200, crystal scintillation detector Crystal: -sodium iódide, th'allium, activated, NaI (Tl) - 2 in-diameter by 2 in-thick - resolutíon better than l0% (Nuclear: Chicago Corp. t331) 1 Cobalt-6o, 1 milli-curie.galnma-ray source ! Machined lead brick shielding Eguipment Setup and Systern Chärâcteristics , The scintíllation detector was aligned and set 12 cm away from the face of the lead shielding, thus making the distance to the radioactive'source 17 cln (see rigure D.1). The equipment, in this cónfiguration, had the fotlowing characteristics. . 1) System. efficiency - The "system efficiency" hras defined by the relationship; CPM - BPM (1) DPM where ' e = system'efficiency, CPM = recorded counts per minute, BPM = background count per mínute determíned when all radioactive sources \^¡ere removed from the vicinity of the detector, and , disintigration per m.inute of, the radioactive ,.oPM ¡ 'sourcê (1 curie = 3.7 x tolO disintegrations { per second). isourcewasused,theefficienòywas.O.015%.Alsothroughoutthe : I tests¡ BPM \^ias always fess.than 2% of CPM. The low.system efficiency i'.. lead shielding. 2) Coincidence loss - Because of the random nature of I 'emissionsI f,rom a r,adioactive sanple i.t is possible for a partiele emission to enter the detector while the detector is still "para- -.'t lized" from a previous pulse. Incorrectly, in this c.ase, ontry one puls'e would be detected by the counting'system' Thê significance of this counting error depends on the resolution time of the ! i.: '. t ' - i':' "' detector.. The correction for this source of error is reasoned as follows . CPM (T) (2) 60 . D.4 where t = time in minutes during which the system is ,i insensitive per each l'minute count interval, CPM =. rqÇorded counts per minute, and T = the..resolution tirne of the detector' in seconds. Since there is as much'chance for emissions to occur during time .l 't' as during any other period of equal duration, CPM (T) % of counts lost = -ãõ--- x IoU (3) / I and therefore number.of counts lost cPM x t (4) per ,mj nute Testing Procedure The snow core samples were liositioned midway between the :- lead shielding and detector on an adjustable stand r¡hich could be lelevated or lowered as desired. A minimum of 10 counts of. O.2 minute duratíon weïe made at each point (spaced from 1.0 to 2.5 cm apart) a'long the l.ength of the core. The count/density relation- ship for the snow \^7as assumed to be linear and the slope of the ; relationship was determined from the average count for a container The inside filled with water and that for the enpty container. ì4.1, ': diameter of the pfastic container (5 cm) was the same as the out- side diameter of the snow core samples. Having accounted for the l Ioss in. recorded çount due to the plastíc contàiner' a chart for D.5 the count/density relationship was developed. .Prior to each round of testing, holrrever, the chart was corrected for fluctuatíon in background nôise. .Furthermore, all reciordêd counts \^¡ere corrected for coincidence loss as described by eguation (4). ' Statistical Irtetpt"tatíot of Da ' Thet0,countsof0.2-minutedurationateachpointa1ong ' t:: ::l::: population. ,,'.,,..r'-,,',, the core r^¡ere each treated as randori samples of a larger : :: .r '': distri- The sample nean (m-) was computed and assuming a normal , . S' \ ::':.:, ' bution fór the çount rate, the.standard deviation for the sample ' . r'' : . was'--- calculated. over 250, lO-count samples were óbtained,and in l ---. - . every case the standard deviation was less than 1% of the sample .l exprsssed as 1% at 68.3å probab- mean. The. result could,.: be "*o t ility". r To estimate the populati.on mean (m ), however, the error associated with the mean value determined from a sample of random counts must be .:consideród- The error is inversly proportional , tó the squate root of the total number of counts (regardless of I the time required) and is. expressed as fo1lows.: k .error (s) ,ri- where k measure of confidenôe, and : ':':'::: =.¿l i: I :.t. r ì_. : n = number of counts observed. j D.6 .': If k is expressed in terms of "number'of standard deviations" from the'sample mèan, than the deviation that is statistically expected in the sample mean is; I % devíatior, &- x IOO (6) = y'n where k = confidence constant (expressed as : :: 'f'nuinber of standard deviatiòns" from the ;;:r.--':--:" ' 'r: ....' gampLe mean), and '\ ,:. ::.:::: f n. = nùrdber: of counts observed. ì-..:.r.,.:-i:';;;,: . Ch6osing k = I as the lequirgd confidence, and for n = 10, the statistical error in m- as an approximation of m-, is 32>". is a 68.3% probábility that the population mean ' (m To reduce the statistical pÞ) falls within the range m^ I 32e". error, the sample síze wòuld have to-be increased. For example., fora68.3%probabi1ity(k=1)that'm.wouldfa11withintherangep m- 10%, sample size..'n' would have to be IOO. s t the (10 of i In the tests described, a sma1l bample s.ize counts i,,;,'.,-t1, '.i,-'., chosen as testing '-r'., O.2-minute duration each) was necêssarily the .' ' ': -:. i . -:.'..-:.:.:ì- facilitieswerenottemperatürecontrol1ed.Ídithalargersamp1e size, significant mel-ting of the snow would have occurred. i:l;::.;.::: 5cm.-dià, snow core st and support lead shield scintillation detector rad ioact ive I' s0u rce adiustable suPPort Seale, 1,3 l/g cm. componeÀts of the radiation,/scintillation Figure Ð:l: Diagram showing the arrangement and basic o' apparatus - { E.L \ / , i, APPENDIX -E CONSTRUCTION EOÛIPMENT AND MATERÍALS E.'2 DESCRiPTTOÑ, OF CONSTRUCTTON EQUTPMENT ANÚ MATERTALS Champion motor grader, Model D-562D i V,Ieight: 26t495 pounds Engine: 672.eu. in. Cummíns diesel Vüidth: 7feetginches Turning radiusJ 40 feet ' Tires.: 14-00 x 24, 12 PIY' Tire pressure-: 20 Psi Allis-Chatmers (loader) tractor, Model TLI4D I Engine: Mod.el 600.0 diesel Transmission: Power shift Çab: Tires I I4.0o x 24, 12 PIY :r Tire pressrlre: 20 Psi TL14D tracÈor equipped with SÍcard snow blower and IHC power unit. t I Sicard snow blower "Snov¡master Baby'l , Model BMG4-27 Blower-frame: 3/16 in.-thick plate steel intake hood, 78 in' wide and 30 in' high mounted on 3 : L/2 1:n' thick 'steel skates' conveyor assenùbly: 2 - 25 in. diameter, 3/L6 in.-thick ' ' steel augers. Left and right auger E'3 sections weldêd át center to form imPeller Discharge chute: cásting direction variable l80o with 10 to 50 foot casting I distance caPabi'IitY' : Clutch: Twin disc : Engine R.P.M- r:e- duction gear box: 3'6:I gear ratio : IIIC snow blower po\^7er- q4¡!,"-i {- I4odel W-345 Engine: I cYlinder, 3 7/.8 in' bore ' .t / c 2I/32 i:n' stroke .t. Full- Ioad R.P.M.: 2500 t I00 J. I.: Case load.fr_ l4o991- 1159 Engine: 6 cYlinder 451 c.í.d, case diesel operatingweight:approximateIy24,OOO1bwith bucket and counterweight Track i 15 inches ¡¡¡ide, ground pressure (approx.) 9 psi. Stee Diameter: :4 feet 2 inches V'Iidth: 4feet4inches 'aPProximateiY V{eight¡ 2 tons 8.4 Steel I-beam drag cross-section: W10 x 29 'I Length: . 11.feêt Weight: 320 lbs. Seasoned pine fife-wood QuantitY used: 1 cord lfood density: 27 Ib/fL3 . Volume of scilid wood 85 fr3 Weight per cord: 2295 Lb. ( Thermat value: 800O Btullb. Average burning rate during burn test: 19 tblmin: i1;:.ì :.: :i;::..,:.. --.:!- a .: i tow hooks Figure E-I: Diagrarn showing the construction o-f the prototype wood-fired surface heater' (si;, 45 gaL , gil drums arè cut and welded together as shown.) H (tt :i':.i:'!i ::,- . .ì;i 'il Il.6 Photograph B-l: Vlindrorvì-ng the sno¡,v with the Champion moLor grader in preparation for processinq wi.th the Sicard snorvblower - lrlrotograph Ti-2: A vj-erv of the snorvl¡ 1.ower/trac.Lor uni t processing the r,vindror';r:d snor,., and depositing it alonq t-hc ¡rroposed roadrualr. E-7 ,,' ri . r:'r$' i:.. :i ì i :;!$¡.i}iì11:¡l*r.:,,,'..' Photograph E-3: The roadvÌay after finaf snor¡/ processing' and deposition. (Note: Photo. E-3 illustrates the spreading capability of the Sicard snowblower. The snow depth varies from 1.5 9o 3.C feet) Photogra¡rh E-4: Road surface af1,er leveling and draggj-nq with the steel drag. tr_B ia. ir r{ ;l Photograph E-5: Station-\Â/agón \,üith no detrimental- effect on inflation pressure 28 psi. ) Photograph E-6: The Champion motor grader blading minor snow accumul-ation from the snov¡ road surface. n.9 Photograph E-7: The prototype surface heater shown fully Íueled with one (1) cord of seasoned pine. ' 1 .ìa :l'"1'¡¡;r. Þ f tl :SS-lN$Y.&Ìi:i]l"qE ìa,* i ' r +.s&å"+ --ilTL 1 lv * : I' . g--, r-r'' '. Photograph E-8: A view of the surface heater i-n operation. (Note trail of ash and cinders on road surface. )