)

1p hans Nahata

jay 44%, 4 DUST PAR

ret i Haan ma yO a BE Pou

My NAR Wd Ade yyese iN oa

i Adal DON ie

eB he ia

> ae) N Maan yr mat 4

yy +) oy

1S ji

fre Ma FeAl i ah it nit ay qt

4 NAPE

Ls] No aey ih 0a0 Tee

Sa Fab pone Rr War ALR AY i Wind AN

‘ ,

Ny ay HN aE

Ren Pa tel Tet

i, i Phish iad bys) tye

horn site a9)

Hoty

Ag ih nh ti fe" if

1 at LAA Bho VENT wens |

wy ji, 1)

Ten) NOU

WY ‘Na,

Hi { aM NON a

HD Ny wait

hiny Ainlits sya) 44

wat

Wee SENN Thea thd 5)

4 whey i! j nai UVa sagy

. ihe \ at ; e) a wo asi ‘ti Ni "4 : LS rants Y

ee) eri " ib ms Wy

4 eR D LUE Sis “0

hist yt eri

i ees qe niles ay

: ssi 7 HS hi eins

ot it

aaa 1 Ry a)

ny,

ne fei i fl Mid Lai Pa i yi

aN be RU

1% uf Hey \

Weal) Ihe 4 1 f anny iF a Rinvvianen ey wel inh ' ih

i Mi ony ot Sa atin i Hy

Kila! Vay OBIE) ATR ac

she ey Nae i i ' ‘ ; r ; 4 ch Ws Nhe? Bu He ;

vey Hite

Dine Wish Of05i hh sh Hern it SHU ) ih nt Mi dit a 1 nh Hi fi Ty yeaah ann ’

Hath

Inti

rani ve welootarh

meter ght NN ivstepreans ead ia

reba a ae

Wyeniyy

Ni at ty iat ul ‘al Sele de ce

Lt OH i tt Mepagtnces

Vat it uhh PN eras hy Vite)

¥ Miles iy Sue AGAIN +n hea Vy bra byy 4

athe: iene

" $e PMS Be ease 4 '

LR UA ih lien rah b ta i Wet ttt Vet

“wh Wnt) cote RICE ELT

eeiee 4 Walle pj ag it ae hat it’at;? ee a

ree AA) 105 i¥8) ie ie Hite bea ; e )

Hy al ni a Hi a Mee i oe ane

abel} petty

be Ae Muth } TUMEUR A esau

i itty htt ap

ABBE UN is BV yeby

vw y ads a Bein

esas Thad rer etiig

beaitiy fe) fy

im if * a ae oa iat

Moker ie iat rah We pees’ ct nit

LeLddaed aay

, dt rane beta i ths : asstierorte

tte ne et

ny AP ye ,

bw eg Ane lth ait ytd as ee ee

tants fhe

Bette Biitrentesaae

iii he ; i Mi ie . _ ae i . uP 4 Hes

ee Pi

ih win ey aie

ya ihe Nor guidand Line

pias 4 i iH Bun Ht peel

“ee

rH}

an iat treed * me Ht

hie

Hh He a

a ue i

th mat a yy

—.

252225552

aaa errs

sose

— =e See ae.

Sr ree ace Soe

SS.

DS i ae Yad py

tak a at ih, Mi an a

vt an OHV aaeS f

i nt ieat

tater! iene

Hit i ave ani

i a if Hie | ft DAA mits i Hy ; ou i Mi ee Kh i i | ih on aeg nf rn ap

ADA eh ase nota

warn nn te asad beasts aS sai Bais

r ‘ ¥ pints seh ent ii fit

- Way) mat Ieee Haat si an eh eae ia ve ih ay) Hi Bait anies ADI ohh) dit pee ints ith Verh Osysagy Ma ee Air — ib ia bh ni

Hes

inn a | rN) i eke YR ‘a

Harirgtans rn Fiver a Weta ne

Af pikes a AM ” a) Ss) t mt J) i ie Hh

Soe ee

is sews ss

1 Me ' ; f HF MruL See SANA TY Pau ea yratey nat za PMN Y SS SAH tenant att nit me i sea ie

iteta?

HU ah

hy rite "Y

avs i ii! Matsa % “ na bpine thinelt bet ath Aa ft bes : 1) - ‘ 4 ’ Nis HE it i aaith w tA hi eb ob rate: ‘ Warahugers ie

" Mare ag in witeaitare N a Nth HH bata | eb eM me A NE fay) sapien it Pitti ales Sts ce ws Ae abs We 8 VIAN eh ae H yates Tats ‘ rete mi NEVE Va they sl { Ve. tNibewsararagany

echt ae hi:

HRB TE Ty

Lipa Wary ihe W's ‘ aw Vik

PM eos sivas

Dit bads Ht aonb

Bose on

aay ¥ na Nita lite

Lyd 4 “ as Woes tah As be: " M3 ta tet Near! ae Orin! at) wee + cr) uae isl absttiraenatsnic et oh hye) ae Rie: Li Tearb nti y %! i Nisity bs eres

Ji an vel LEH Poh Soeh Bitar A Hie fioyhrs

run yrs

hey totbarhed ; Pag N

ieee

| IUBLIC ROADS >

lh. | Ih, Mh, call! ith, cath. al

A JOURNAL OF HIGHWAY =e

UNITED STATES DEPARTMENT OF AGRICULTURE

BUREAU OF PUBLIC ROADS

Wi Lian. IN. COL 1 V PE BRUAR Y. oS 7

ee cee as

; '

ROAD CONSTRUCTION OVER PEAT BOGS HAS PRESENTED A DIFFICULT PROBLEM IN MICHIGAN Lee ee ee — ee

WASHINGTON : GOVERNMENT PRINTING OFFICE : 1927

PUBLIC KOADS

A JOURNAL OF HIGHWAY RESEARCH

U. 8. DEPARTMENT OF AGRICULTURE BUREAU OF PUBLIC ROADS

CERTIFICATE: By direction of the Secretary of Agriculture, the matter contained herein is published as administrative information and is required for the proper transaction of the public business

VOL es, NOcaZ FEBRUARY, 1927 H. S. FAIRBANK, Editor

TABLE OF CONTENTS Fill Settlement in Peat Marshes

Determination of Consistency of Soils by Means of Penetration Tests

California Road Survey Demonstrates the Economic Possibilities of Subgrade Studies Design of a Constant Temperature Moist Closet Effect of Highway Slash on Infestation by Western Pine Bettle Studied

The Action of Calcium Chloride on Cement

THE VU. S. BUREAU OF PUBLIC ROADS Willard Building, Washington, D. C.

REGIONAL HEADQUARTERS Mark Sheldon Building, San Francisco, Calif.

DISTRICT OFFICES

DISTRICT No. 1, Oregon, Washington, Montana, and Alaska. DISTRICT No. 7, Illinois, Indiana, Kentucky, and Michigan. Box 3900, Portland, Oreg. Souih Chicago Station, Chicago, Ill.

DISTRICT No. 2, California, Arizona, and Nevada. DISTRICT No. 8, Alabama, Georgia, Florida, Mississippi, South Carolina, Mark Sheldon Building, San Francisco, Calif. and Tennessee.

DISTRICT No. 3, Colorado, New Mexico, and Wyoming, 301 Customhouse Building, Denver, Colo.

Box J, Montgomery, Ala.

DISTRICT No. 9, Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Rhode Island, and Vermont. DISTRICT No. 4, Minnesota, North Dakota, South Dakota, and Wiscoas n. Federal Building, Troy, N. Y. 410 Hamm Building, St. Paul, Minn. DISTRICT No. 10, Delaware, Maryland, North Carolina, Ohio, DISTRICT No. 5, Iowa, Kansas, Missouri, and Nebraska. Pennsylvania, Virginia, and West Virginia. 8th Floor, Saunders-Kennedy Building, Omaha, Nebr. Willard Building, Washington, D. C.

DISTRICT No. 6, Arkansas, Louisiana, Oklahoma, and Texas. DISTRICT No. 12, Idaho and Utah. 1912 F.& M. Bank Building, Forth Worth, Tex. Fred J. Kiesel Building, Ogden, Utah.

Owing to the necessarily limited edition of this publication it will be impossible to distribute it free to any persons or institutions other than State and county officials actually engaged in planning or constructing public highways, instructors in highway engineering, periodicals upon an exchange basis, and Members of both Houses of Congress. Others desiring to obtain ‘Public Roads” can do so by sending 10 cents for a single number or $1 per year to the Superintendent of Documents,

Government Printing Office, Washington, D. C.

———————— - = rt OC OrUTDTLCUCUC OU OU

FILL SETTLEMENT IN PEAT MARSHES'

By V. R. BURTON, Engineer on Special Assignments, Michigan State Highway Department

HE MICHIGAN State Highway Department has ale just completed a special study of fill settlement in peat marshes as a part of the general soil sur- vey it has undertaken. Settlement in peat marshes has caused serious trouble in many places, because it was unknown how much fill material would be required and even after the fill had been completed there was no assurance that it would maintain the original grade line. It was therefore desired to ascertain the physical characteristics of the peat and decide on the field data necessary to properly locate and estimate the cost of a road over a peat deposit. It was hoped that it would be possible to devise a method of construction more eco- nomical than that then in use, and to adopt better main- tenance methods to restore sunken fills and prevent fur- ther settlement.

A brief description of peat deposits will be necessary if any adequate idea is to be had of what actually exists under the surface of a “‘sink hole.’’ Peat deposits are classified and mapped stratigraphically in essent- ially the same way as mineral upland soils. For general peat studies, peat materials have been described by layers in terms of composition, texture, structure, and color. This gives a fairly good idea of the main differences between separate peat layers and localizes their relative positions.

VARYING CHARACTERISTICS OF PEAT LAYERS AID STUDY OF FILL SETTLEMENT

The classification of peat strata was made in accord- ance with the system largely developed in this country by Dr. Alfred P. Dachnowski, of the Bureau of Plant Industry, United States Department of Agriculture, and he personally supervised a number of surveys. The following description of peat deposits is taken from an article on the ‘Stratigraphic Study of Peat Deposits”? by Doctor Dachnowski. ‘Distinction should early be made between muck and peat. Muck is the thin surface layer of disintegrated peat and is entirely different in appearance and properties to the relatively unaltered peat below it. This distinction is made because many engineers use the terms inter- changeably with no knowledge that there is a difference in the two.”

Table 1 is from the article just referred to, and gives the classification of characteristic peat layering.

Pulpy peat is formed in water basins under conditions of poor drainage from transported organic sediment carried from its origin and redistributed by water

currents. There are no visible bedding planes although the original sediment was laid down in practically horizontal position under conditions differing by rate of deposition, seasonal changes, etc. The plant re- mains are small in size. In texture the peats vary from coarse to very finely divided particles, while in structure a layer may vary widely in its degree of compaction. Fresh from the deposit and while still moist the colors vary from gray, green, and brown to black.

DRAINAGE DitcH THrouGcH Peat Boa

As with most peats, the pulpy variety quickly loses its character by oxidation on exposure to the air and drying. It shrinks enormously and becomes a hard, water-resisting substance when quickly dried out. It will not then absorb anything like the original quantity of water. Doctor Dachnowski states that “it may be accepted as an axiom that undrained deposits of peat contain from about 70 to 95 per cent of water.”

The outstanding characteristic of fibrous layers of peat is their matted or felt-like, porous nature which has its origin in the slightly altered remains of moss, roots, rootlets, and rhizomes of herbs. The peat layers differ widely in character. Their water content is enormously high, the solid matter commonly composing only from 5 to 15 per cent of the peat mass. In texture they vary from coarse to very finely fibered material usually presenting a rather loose, porous appearance, with colors ranging from gray through yellow to red brown and dark brown. The layer may be interbedded or it may grade into a layer of woody of pulpy peat by admixture or 1t may occur in overlapping beds of other kinds of fibrous peat.

TABLE 1.—Structural units of peat deposits and some of their physical characteristics

rare set fag Types of peat Character of peat layers Color of peat layers Texture of peat layers Structure of peat layers ING STN oe Se ne (Oar BRIE Fs ae \Pulpy (sedimentary) -____- Olive green, brown to| Coarse to very finely | Compact, impervious, ae Saale black, divided. stiff, plastic or loose, Reed friable. FS EO DE er PEAS | 7 Marsh....------------------- {sees Sihees as eaeeehe ype Fibrous. -__- Sok, ae Seed pall Gray, red or yellow brown | Coarse to fine fibered_____. Dense, matted, felty, or wes ern me ss to dark brown. porous, spongy. Bog ---.--------------------- tn sek eo hase Willow-alder shrub_ -_--_--_- r 2 . ; WOO Vicwee, <8 Sane Dark brown to blackish | Coarsely fragmented to | Compact and granular, or Swamp....-..--------------- {Peciduous raphe eee Le brown. granular. loose, wicker-like. : ‘oniferous forest._...-.--- |

1 Paper presented at the fifth annual meeting of the Highway Research CounciJ, Washington D. C., December, 1926.

2 Soil Science, vol. 17, No. 2 February, 1924. 80228—27

233

PUBLIC

234

ROADS Vol. 7, No. 12

Woody peat may occur either as the dominant com- ponent or as a prominent admixture in other types. The woody plant remains are broken down partly into eranular débris and partly into irregularly shaped woody fragments, which are almost unaltered by decom- position. The woody peats differ widely in composition and character. They may be as finely divided as sawdust or as coarse as a tangle of waterlogged brush and logs. Several layers may be found as, like the fibrous peats, they are developed on moist flat land under conditions of a rising or fluctuating water table. Land may grow up to forest, be submerged to form a marsh, and then emerge again to become a forest. In one peat deposit examined on this survey three distinct forest layers were plainly visible.

Marly phases of peat are formed from calcareous plant remains and shells of mollusks and are usually found admixed in the macerated types of peat. Exten- sive deposits of marl with peat occur mainly in regions underlain by limestone rock or where the soil adjacent to and underneath a peat area is derived from lime- stone drift. Deposits vary in color from gray to brown and are usually soft plastic masses with particles from fine to comparatively coarse, depending on the amount and character of plant or shell remains present.

The lake clay found at the bottom of most deep peat deposits is structureless, mainly siliceous material from plant and animal remains consisting of diatomaceous shells and drift débris. It is gray to gray blue in color, fairly compact, but with a high percentage of water in it and is quite plastic and sticky. It is, however, a much more compact and weighty material than any of the peats or marls.

In carrying out the study of fill settlement it was decided to cross section marshes accurately by borings, which would show the various peat layerings and the exact position of the filled material with reference to them. Samples of the various classes of peat were to be collected and sent to the laboratory for physical tests in an effort to discover some relationship between their bearing power and the known settlement for various water contents. If a large number of existing fills were thus examined, it seemed logical that general conclusions might be drawn to serve as a guide on future work.

FILL SETTLEMENT NOT AFFECTED BY CHARACTER OF PEAT

Fills over eight different peat deposits were cross sectioned at 83 different points where the depth of the peat ranged from 1 to 66 feet. The shape of the fill material was accurately determined with a 2-inch Empire drill. Cores from the peat were taken with the Davis peat sampler where possible, but if the peat was compacted beyond a certain amount the Empire drill and jet was used. Enough test holes were put down on the cross section chosen to show the fill shape fairly accurately and an effort was made to extend the cross section in both directions from the fill sufficiently far to get out of the zone where the layers were dis- turbed by filling. In a few cases it later became evident that the test holes at the extremities of the cross section were not placed far enough from the center line.

Samples of the different peat layers were collected at various depths and sent to the laboratory for physical tests. Results were very disappointing in this regard. It was impossible to maintain the peat in its original condition on exposure to the air. Peat as it comes from the ground is only partially oxidized and at the

greater depths there is actually some reducing action. Consequently, immediately it is handled in air, oxida- tion begins and if it is attempted to dry it, a totally different material results. Laboratory bearing value tests were not satisfactory. It is certain that the amount of compression under loading in a vessel such as is used in the standard bearing value test which does not permit of a free displacement in all directions will not give results at all indicative of the behavior of the same material in its natural place.

The only way, then, to secure the solution of the various problems involved in this study was by a care- ful study of the cross sections showing the fill shape and the peat layerings in the disturbed position after filling. An effort was made to reconstruct the position of the layering as it originally existed from the position of the layers outside the distorted area of the cross section. A study of the fill cross section and the distortion of the original layers as reconstructed led to the con- clusions drawn from these effects.

The various factors which it was thought should determine the amount of settlement were depth and character of peat layering, height of the water table, slope of the mineral subsoil beneath the peat, height and width of fill, The weight and character of the fill material would also be a factor but the difference in weight of the materials studied was so small as to be neglected. All fills in this investigation were earth and varied from a clay to sand. No rock fills were studied.

So far as could be determined the character of the peat layering is very slightly responsible for differences in fill settlement. There is, to be sure, a decided differ- ence between the top layer of peat a foot or two in depth where the material is fairly well decomposed, but this is due more to a smaller water content than in the actual character of the material itself. The lake clay which occurs in the bottom of most of our deeper peat deposits is enough different in character to make it necessary to consider its effect separately. It is also probable that if any considerable thickness of pure marl with a heavy texture were encountered it should be grouped with the clay or given separate treatment. As it happened in this investigation no very great amounts of marl were found except as they occurred as a peat admixture. <A soft soupy marl should probably be classed as peat in considering its effect on fill settlement. The preliminary work of sounding a peat bog preparatory to estimating fill settlement is thus considerably simplified for the engineer, as where a uniform marl is absent, only two layers, the peat and lake clay, need be considered. No differentiation of peat layering is necessary for estimating fill settlement.

DEPTH OF PEAT AND LAKE CLAY THE MOST IMPORTANT FACTORS

If there were any great variation in the height of the water table compared to the depth of the deposit, this factor should certainly show its effect. It so happened that the road ditches were about the same depth through all of the shallow peat deposits studied and in the deposits of greater depth the depth of the ditch was very small in comparison to the total depth of the peat. The water table in most cases varied from 1 to 3 feet below the fill surface. The varying height of water table, which was to a certain extent dependent on the time of year as well as on the particular situation of the cross section, was not taken into account. Some inconsistencies in the shallower depths are probably due to the neglect of this factor.

February, 1927

PUBLIC ROADS

The height of the fill above marsh surface is of course a factor, although not so great a one in the case of loading over a wide area as one would expect. In peat depths of less than 10 feet the settlement is small and is affected by factors other than height of fill and results are inconsistent. The height of fill produces a

marked effect in medium depths of peat of from 10 to.

25 feet. In depths beyond 25 feet the total depth of fill is so great in proportion to the height above marsh surface that the amount of this additional loading is not important. The above refers to completed fills studied and not to new construction methods in which the height of fill placed in the early stages is of consid- erable importance.

The slope of the mineral subsoil produces a marked effect on settlement and the bottom of the fill nearly always takes the same direction of slope as the marsh bottom. This is due to two reasons; first, the flow of peat beneath the fill is aided by gravity, and, second, the greater depth of peat down slope induces greater settlement. Most of the inconsistencies in amount of settlement from the determined normal occur above sharp slopes in the mineral subsoil. The amounts varied so much, however, that no general law could be observed. It is safe to say that in general if the mineral subsoil does slope quite sharply, the normal amount of settlement will be increased.

One very important factor met in special cases is the effect on fill settlement of the peat compression by neighboring fills. The compressed peat is much more resistant to flow than is the uncompressed peat, and hence flow on the side adjacent to an old fill is very much smaller than on the free side.

The factors of greatest importance and the ones least difficult of determination are simply the depth of the peat layer and the depth of the lake clay beneath it. Other factors undoubtedly have their influence but the depths alone, so far as could be determined from the information available, directly determine the amount of settlement which will occur. No other factors operated with the consistency or effect of these.

Let us consider what happens when a fill is placed over a peat deposit. The top crust is first subjected to compression beneath the interior of the fill and to shear and tension at the edge. The communication of this compression to the interior of the peat is opposed by the resistance to flow of the peat or its internal friction. On account of the inertia of the peat particles, water is squeezed out of the material nearest the fill and the material is compacted, and this process extends with continually decreasing effect as distance from the fill increases until at some distance the peat is practi- cally undisturbed.

Lateral flow of peat begins at an early stage in the placing of a fill and is due to the horizontal component of pressure due to the internal friction of the material and also to the inclined position of the settling load. Peat is forced outward and upward, until in the course of time a condition of equilibrium is reached.

It is easily understood, then, that the width of fill is a function of the amount of settlement which takes place. The fills in this investigation do not vary ereatly in width, viz, from 24 to 30 feet, and so no account is taken of this factor. It is important, however, in considering the cause of the greater settle- ment in deeper deposits.

When deposits are relatively deep in proportion to their width, considerable vertical movement can take

place before any marked reaction is experienced from the mineral subsoil below. In this case, lateral flow is comparatively small at first and a large amount of settlement is possible before there is much resistance; the elastic crust ruptures in tension and shear and the whole weight of the fill is suddenly thrown on the soft material below. This sudden application of weight very quickly compacts the peat below to a degree at which lateral flow begins in a considerable amount and we have a “‘sink hole.’’ <A portion of the peat flowing from beneath the load of earth is forced up at the sides. Due to the lighter weight of the peat as compared to the fill material, a correspondingly YELLOW-GRAY SEDIMENT

WITH CHARA AND SHELL- MARL AOMIXTURE

BROWN WOODY AND COARSE

0 FIBROUS PEAT.

THEORETICAL FILL SHAPE BLACK WOODY PEAT WELL DECOMPOSED BLACK WOOOY PEAT 1D < * RR <a

50 ~~ = ——- 80 70 60 50 40 30 20 10 oO 10 20 30 40 50 60

Fig. 1—Fint or Normau Cross Section tn A Prat Marsy Aspout 22 Freer Derr FRoM SuRFACE TO CLAY

BLACK WELL DECOMPOSED PEAT, BROWN FINELY FIBROUS, PEAT Ny

0 | REEDY BROWN PEAT, =~ - = et eee ee - EORETICAL! FILL BLACK COMPRESSED’ | «ge 20 ta GRAVEL ul AND 30 CLAY BROWN FIBROUS MACERATED PEAT 40 GRAY-GREEN SEDIMENTARY PEAT:

PULPY

FEET

Fia. 2.—A Finut Partty on Uptanp Soit AND PARTLY ON PEAT MarsuH

larger amount of peat must be raised to balance the heavier fill. Subsequent loading increases this heaving until the weight of the heaved peat brings about a condition of equilibrium.

SLOPE OF SOIL BENEATH PEAT BED AFFECTS FILL SHAPE

In the later stages of fill settlement the rate of settle- ment is very slow, but even a rate of a foot a year is decidedly objectionable. The rate progressively de- creases until finally, after a long time, equilibrium is established. The length of time required to reach a stable condition in some cases is known to have been more than five years. We have arbitrarily established the rate of five-hundredths of a foot in 30 days as being the rate of settlement at which filling may be stopped. Since the rate will usually decrease progressively, this should give a settlement of less than 6 inches per year.

Unfortunately, at times, if the early settlement is not rapid,.some later cause may accelerate the rate of settlement. Fills completed in the fall under earlier methods were quite apt to settle rapidly the following

236

PUBLIC

ROADS Vol. 7, No. 12

spring. The vibration of traffic also tends to increase the rate of settlement after a road is first opened. Large ditches cut along an old road are quite apt to start considerable settlement. As a matter of fact, unless settlement during filling has been accelerated in order to insure its completeness, some subsidence is almost sure to take place later.

A number of cross sections illustrating the various shapes taken by fills under different conditions are shown. Figure 1 shows a fill of normal cross section in a peat marsh about 22 feet deep from surface to clay. The theoretical shape of fill is shown by the dashed line and is obtained by a method described later. This deposit is of nearly equal depth on both sides of the fill and therefore the shape of fill is quite regular.

Figure 2 shows a fill which is partly on upland soil and partly over peat marsh. It will be noticed that the mineral subsoil slopes sharply toward the peat deposit and for this reason the proportion of the fill beneath the surface of the peat is considerably increased above that shown on the normal cross section. It will also be noticed that on account of the flow of the material only a very small amount of the peat remains under the fill, and that the layers to the right of the fill are considerably thickened by reason of their flow from beneath the fill and subsequent expansion due to the decrease of pressure.

Figure 3 shows a fill which is still settling into the peat marsh at a rapid rate. The slope of the bottom of the fill is quite noticeably parallel to the slope of the mineral subsoil beneath the light clay and most of the distortion of the peat layers has occurred on the low side of this mineral subsoil. It is quite evident that even with this considerable depth of peat beneath the fill, pressure is communicated throughout the entire mass.

The effect of the slope of the mineral subsoil upon the direction of pressure is still more marked in Figure 4, where it will be noticed that, except immediately adjacent to the fill, practically all of the disturbance of the peat has taken place to the left of the center line, indicating that flow has largely occurred in this direc- tion.

Figure 5 shows a fill which was placed adjacent to a railroad fill at the right, but not shown in the figure. In this instance it will be noted that the compacted peat to the right of the center line reduced the amount of fill necessary in a material manner. <A very con- siderable amount of distortion of the layers is shown on the left of the figure. In this instance the peat is heaved above the normal level of the marsh some 10 or 12 feet and slopes away from the crest of the heaved material in a fairly uniform manner.

Figure 6 shows the deepest cross section encountered in this investigation, a fill with a depth of about 66 feet. The marsh, over which this fill was made, lies between two fairly large lakes, which are at present connected by a small stream. It is quite evident that the location of the highway in this particular instance was very unfortunate in that the center line of the highway at the cross section shown is practically on the center line of the old glacial drainage channel be- tween the lakes in the early stage of their formation. A location of center line 30 feet to the right of its pres- ent location, which would have been possible, would have eliminated about 25 feet of fill. The entire amount of the lake clay which probably nearly filled the old channel, has been forced to one side by the load

of fill above it. The irregular shape of the cross sec- tion on the right is probably due to filling a hole left by the peat heaving 8 or 10 feet at the side during construction.

METHOD OF PREDICTING FILL SETTLEMENT AND SHAPE DEVELOPED

From the whole series of cross sections of the char- acter of those illustrated, most of which, however, are considerably more regular, diagrams giving the amount of settlement and shape of fill have been prepared. Depths of peat and marl on center line of fill were DARK BROWN FIBROUS MACERATED PEAT WITH SEDIMENTRY ADMIXTURE

DARK BROWN FINELY

FIBROUS FELTY PEAT WITH TWIGS

DARK GREEN SEDIMENTARY PEAT AND MARL WITH STREAKS OF DARK GREEN COLLOIDAL MATERIAL

GREY SEDIMENTARY

CHARA SHELL BLACK WOODY PEAT

THEORETICAL FILL

BLACK MEALY PEAT a _

FEE

FEET

Fic. 3.—SHAPE OF FILL IN PEAT CONSIDERABLY AFFECTED BY Storer or MINERAL Sort at Bortom or PEAT BED

GREEN SEDIMENTARY PEAT WITH A TRACE OF CHARA AND SHELL “TOWARD BOTTOM OF LAYER

BROWN FINELY FIBROUS FELTY PEAT WITH AN INCREASE TOWARD MACERATED BROWN FIBROUS PEAT TOWARD BOTTOM

BROWN FINELY FIBROUS FELTY PEAT | BROWN FIBROUS MACERATED PEAT

WITHA TRACE OF MACERATED BROWN WITH SEDIMENTARY ADMIXTURE. (GREEN SEDIMENTARY PULPY PEAT WITH FIBROUS PEAT TOWARD BOTTOM TRACE OF SHELL AND CHARA

AT BOTTOM OF LAYER

LACK WELL BZDECOMPOSED

BROWN FIBROUS MACERATED PULP

FEET

Fig. 4—Distortion or Peat Fiut to tHe Lerr DUE TO SLOPE OF Firm Soi, BENEATH PEAT BED

platted against the total depth to which the fill pene- trated the peat and marl as measured from the original marsh surface. Three curves were then drawn, one representing the average of fills with a height above marsh level of 1 to 3 feet, one for 3 to 4 foot fills, and the third for fills from 4 to 5 feet in height. These three curves, shown in Figure 7, indicate that the amount of settlement increases uniformly but is not great up to a marsh depth of 20 feet. As the marsh depth increases from 20 to 26.5 feet the settlement in- creases rapidly and when the depth exceeds 26.5 feet the fill goes completely through the peat in every in- stance, so that the penetration is equal to the depth and the three curves become common and make a straight line with a slope of 45°. Therefore, for peat and marl depths of less than 26.5 feet the depth of lake clay need not be taken into account, except as a factor should settlement be different from the average, and anticipated settlement is read directly from the

February, 1927 PUBLIC

ROADS

diagram. For depths of peat and marl over 26.5 feet the settlement will probably be the full depth of the deposit plus some penetration into the lake clay. The amount of penetration into the latter material is read directly from the curve marked lake clay, which has been plotted, using the combined depth of peat and clay as abscissas.

Suppose we have a peat marsh consisting of 30 feet of peat and marl and 5 feet of lake clay. The peat curve gives us complete penetration of the full 30 feet of peat and from the 35 feet total depth of the marsh deposit we read off from the clay curve 2 feet for clay penetration. There will then be a distance from bottom of fill to marsh surface of 32 feet. Again, if we had a peat deposit consist- ing of 38 feet of peat and 2 feet of lake clay we should get 38 feet of peat penetration plus 4 feet of clay pene- tration. The total, 42 feet, exceeds our total depth of deposit and the fill therefore stops at the marsh bottom with a 40-foot penetration. This last case is not com- mon except in the very deep deposits, of 50 feet or more. It will usually be found that in marshes of such extreme depth with ordinary proportions of peat and lake clay, the fill goes to the very bottom.

No claim of scientific reason is made for the shape of the curves shown in Figure 7. They are purely empir- ical and represent merely the average behavior of a con- siderable number of fills. Extreme variations of actual from expected settlements as derived by this method have been found to be as much as 100 per cent for an indicated settlement of 5 feet, 60 per cent for 10 feet, 30 per cent for 20 feet, 20 per cent for 25 feet, 10 per cent for 30 feet, and less for greater amounts. Generally accuracy of the method is well within these limits.

The method of estimating fill quantities is illustrated by Figure 8. The typical shape shown was arrived at

WOODY AND FIBROUS MIXTURE WITH TWIGS

OLIVE-GREEN SEDIMENTARY PEAT BROWN FIBROUS PEAT oping WITH SOME STICKS

MATERIAL BROWN WOODY PEAT WITH |

SOME FIBROUS ADMIXTURE

SEDIMENTARY PEAT WITH SHEL!

OLIVE-GREEN COLLOIDAL PEAT

SEDIMENTARY PEAT

FEET

180 170 160 150 140 130 I20 10 100 99 80 70 6 SO 40 30 20 1 0 10 20 30 40

FEET

Fig. 5.—A Fitt ADJACENT TO A RAILROAD FILL AT THE RIGHT But Not SHown. Compactep PEAT ON THE RIGHT SIDE REDUCED THE AMOUNT OF FILL CONSIDERABLY

BROWN WOODY MACERATED PEAT WITH FIBER MIXED f

BLACK WOODY PEAT BROWN COARSLEY BLACK WOODY PEAT | FIBROUS PEAT

BROWN FINELY

FIBROUS PEAT |

BLACK WOODY PEAT

| GREEN SEDIMENTARY PEAT, COMPRESSED

BROWN WOODY AND FIBROUS PEAT, MACERATED

OLIN ITN pan

BROWN FINELY FIBROUS PEAT, PULPY ADMIXTURE

x Ree

BROWN FIBROUS SEDGE PEAT ~

OY Sawn

eewegen rss weer

crass

ibd ddd phere VAS

SYS NN SAT YAN Mi Vi Vag) LR

S 4 BROWN rien PEAT WITH! PARTICLES OF WOOD- MACERATEO

RC LR CRS ARTS ANDERS

FEET

Fria. 6—An Unusvuatity Derr Fitt. Sounpinas WouLp HAvE INDICATED TH1Is CONDITION AND PERMITTED Morr ADVAN- TAGEOUS LOCATION

by drawing on cross-section paper all cross sections of equal depth, superimposed on each other, with the marsh level as a base line, and then fitting an average shape to them. A side slope of embankment above marsh level of 1 to 3 feet was used, since most of the fills studied had this slope. The fills quite generally slope inward from the outermost point of the embank- ment to a point directly beneath the shoulder line. The total area of fill is composed of two trapezoids and can be calculated by the following formula, the sym- bols representing distances as indicated in Figure 8.

Area=(W+3F) (F+ 8)

PENETRATION -FEET

4-5’ FILL

ce) 10 20 ACEC 40 50 60 DEPTH OF PEAT AND MARL-FEET OEPTH OF PEAT,MARL