Congelation-Structure and Frost-Heaving Ratio at Assan, Manchuria Sugaya, J

NRC Publications Archive Archives des publications du CNRC

Congelation-Structure and Frost-Heaving Ratio at Assan, Manchuria Sugaya, J.

For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.

Publisher’s version / Version de l'éditeur: https://doi.org/10.4224/20331435 Technical Translation (National Research Council of Canada), 1956

NRC Publications Record / Notice d'Archives des publications de CNRC: https://nrc-publications.canada.ca/eng/view/object/?id=eaecced8-f651-4a9b-b2cb-f7f9c04630a1 https://publications-cnrc.canada.ca/fra/voir/objet/?id=eaecced8-f651-4a9b-b2cb-f7f9c04630a1

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at https://nrc-publications.canada.ca/eng/copyright READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE.

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site https://publications-cnrc.canada.ca/fra/droits LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

Questions? Contact the NRC Publications Archive team at [email protected]. If you wish to email the authors directly, please see the first page of the publication for their contact information.

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à [email protected]. NATIONAL RESEARCH COUNCIL OF CANADA

Technical Translation TT-596

Title: Congelation-structure and frost-heaving ratio at Assan, hhnchuria.

Author: LEji Sugaya, Institute of Low Temperature Science, HokkaidE University, Sapporo, Japan.

Translator: E.R. Hope, Directorate of Scientific Information Service, Defence Research Board, Canada,

Translated, with the permission of the aul;hor, from an unpublished manuscript,

This translation has also been issued by the Defence Research Board as their translation T54J. The Division of Building Research of the National Research Council is keenly interested in permafrost research, having now its ovm Permafrost Research Station at Norman Wells,

N.W.. -.T. Contact is maintained with the Sliow. Ice and Permafrost Research Establishment of the U. S. Departuie6t of Defence in connection with this work and allied snow and ice research, The Division shares this interest with many branches of the Department of National Defence and in particular with the Defence Research Board, Through the Defence Scientific Information Service of the Board this translation was kindly made available to the Division for inclusion in the N.R.C, series of translations, following earlier cooperation of the same order, The translation was kindly carried out by Mr, E.R, Hope of the Defence Research Board staff with the permission of Dr. Sugaya, The Division here records its thanks to Mr, Hope for this further expert translation service, and to the Defence Research Board for their kind agreement to the inclusion of this translation in the N,R. C. series, It may be surprising to some readers to find Japanese research workers interes-te6 in the problems of permafrost, until it is noted that the paper refers to work in Manchuria, The translation shows that permafrost problems are certainly not peculiar to North America, It is hoped that the publication of this translation, as a joint venture by the Defence Research Board and the Division of Building Research, National Research Council, will prove to be a useful contribution to public lsnowledge of this irnportant subject,

OTTAWA, Robert F, Legget, April 1956. Director, CONGELATION-STRUCTURE AND FROST-HEAVING RATIO AT !ASSAN, MANCHURIA

Institute of Low Temperature Science, HokkaidZ University, Sapporo, Japan

Translated (from unpublished manuscript)

E.R. Hope

Directorate of Scientific Information Service DRB Canada January 25, 1956 The Defence Research Board of Canada thanks the author Dr. JGji Sugaya, and Professor Zyungo Yoshida, Director of the Institute of Law Temperature Science, Hokkaid5 University, for permission to publish this translation.

Note-

An effort has been made to adjust the terminology of this translation to conform with that which is likely to bocome standard in Canada, a terminolow agreeing closely with one suggested by the U.S. Army Corps of Engineers. For this reason there is a divergence from the language of previous translations, in particular DRR translation T 25 J (also published as National Research Council Technical Translation 382).

For example, the term "frost layerff of T 25 J, which directly translates the Japanese expression, has been replaced, in T 54 J, by "permafrost", except in a few cases where this is not the correct sense. The reason for this is the conflicting use, in Canada, of Itfrost layern to designate the seasonal freezing at the surface ... a term of low inherent intelligibility which might well be replaced by fftop-frosttf (parallel to fftop-soilff,etc.). CONGELATION-STRUCTURE AND FROST-HEAVING RATIO IN PERMAFROST AT ASSAN, MANCHURIA

JCji Sugaya

Institute of Lon Temperature Science, HokkaidE University, Sapporo, Japan.

1. Argument

During a five-day period from September 29th to October 3rd, 19h3, at a water resources field research station about four kilometers from the village of Assan in the North Manchurian permafrost belt, the author carried out a study of the congelation-structure of the permafrost bed, and collected permafrost samples. In this work he utilized a pit excavated by the Manchurian Railway Underground Plater Resources Survey Unit for geological research purposes. From study of the samples, it was possible to calculate the frost-heaving ratio for the bod; also to suggest, from the state of occurrence of the frozen soil, a theory of the cause of the bed's formation. The results are set forth below.

2. Details of Topography and Time

The Assan Valley is at about 120~40~East Longitude and 49'53 1 North Latitude, on the upper reaches of the Merger River, a tributary of the Khailar, where it penetrates deeply into the Greater Khingan l!ountains. It is an old valley, with a breadth of about four kilometers and extending back about twelve km. The point where the present investigation was made is about half way up the valley.

On the west of the valley are the mountains of the Khingan Range; to the east, across the river-bed plain, the ground rises into a belt of hills, as shown in Fig.2. At about the center of the plain, there is the slaw- flowing Assan River, 10 to 15 m wide. Close to where the foothill-slopes of the mountains descend eastward onto the river plain, we find a swamp [muskeg] about a meter deep, fed by spring water; a good deal of water also came from the borings which were made around this swamp.

The pit where we made our investigation is on the border-line between the plain and the hill which may be seen in theforeground of Fig.2. The slope in this vicinity is about 2.s0. Four pits were dug at intervals of 50 or 100 m along a line running straight dmthis slope, and ours was the third from the upper end. At the time, the Water Resources Survey Unit was making a hydrological and geological stucly at this station, using sixteen borings and these four hand-excavated pits. The author, in the present research, directed his attention mainly to these pits. It was impossible to study the permafrost structure from drill cores, because the water used in drilling thawed the permafrost . The ground in the vicinity of the No.3 pit, where we made our study, was 2 or 3 meters above the water level in the swap which may be seen in the photograph.

As for the pedology of this region, the topsoil was of hums, but forming a comparatively shallow layer, and the grass-cover was rather poor. Around a point 15'0 m up from No.3 pit, the surface soil wm thin, from 70 to 80 cm in depth. Below it was gravel, apparently a weathered sedimentary deposit. Still higher up the slope, there were places where the stone and gravel showed at the ground surface and the grass-cover disappeared.

Excavation of pit No.3 had been started on September 21st. Subsequent progress was as shovm in Table 1.

When the author arrived at the place, he found the pit quite well corrwed with matting. Results of ground temperature distribution measurements (madi. by Mr. Yamada of the Water Resources Survey Unit) during the course of the excavation are shown in Table 2,

The ground. temperatures were measured, as shm in Fig.3 (wall B), by cutting a block out of the wall, 50 cm wide and going back 25 cm; then thermometers were inserted horizontally into the new face to a further depth of 25 cm.

The orientation and dimensions of the excavated pit were as shown in Fig .3; it was dug to 1 x 1.9 m size and taken to depth 3.5 m.

The state of the pit walls on the 29th, the day we began our study, was that the topsoil and the brawn loam layer beneath it were thawed out, and the wall surface had dried and hardened; farther down, the wall surface was permafrost. The permafrost wall surface had thawed; nevertheless, possibly because of intensive sublimation, the earth did not cave in, but kept more or less its original shape. The depth of thawing at the time of our investigation was around 2.1 m, Between depths 1.9 and 2,l m, pract,ically no ice could be found even at a distance of 60 cm back from the pit wall. The permafrost layer below this consisted of bluish-gray clay, with a content of humus. It included a great deal of ice-strata; the structure was to all appearances very like that which is seen in frozen soil where frost-heaving has taken place.* This layer went dmto 3.5 m. Below it was a bed of frozen gravel, an extremely hard-packed foundation. According to the results from borings around the swamp, this gravel foundation, alternating with layers of sand, continues downward. Permafrost was still found even at depths of 1L to 15 meters.

The air temperatures as read in tho pit were +6.0°c at 2 m depth and +IreSOcat 3 .5 m, the external atmospheric: temperature being 14.7O~ at the time, weather fine and bright, no wing. Tho early morning minimum tomperature for several days in succession was -8 C to -9OC, for which reason, no doubt, the frozen soil in the pit was very slow in thawing.

- -- u U. Nakaya, "The Mechanism of Frost-Heaving; Second Year of Field Survey." KishE Shcshi, series 2, vo1.20, NO .h (1942 ) . U. Nnlcaya ,and J .Sugaya. "Perma- frost in the Tundra Zone," Rilcen I!]:, vo1.21, No.8 (19L2). -2 - Table 1 Date Depth of Excavation Working Time Remarks

all day During the day whenever work was no digging not in progress, and at night, no digging the pit was covered with matting 3-4 hrs for heat insulation. all day On the 24th, ground temperatures all day were read down to 2 meters depth. all day Depth of permafrost surface 1.9 m. all day On tho 2Sth, ground temperature at 2.2 m was read. no digging Survey begun.

Table 2

Depth Ground Remarks (m ) Temperature - - 0.20 +3 .S0c Measurements at depths down to 2 meters 0.hO 41.3 were made on the 2bth. The 1.9 m penna- 0.60 4.3 frost level determined by visual obser- 0.80 41.0 vation. (Atmospheric temperature 1.00 +3.8 +7.3OC. ) 1.20 +2.0 1.LO +2 .O 1. +1.8 1.80 +1.0 1.90 0.0 Surface of permafrost. 2 000 -0.3 2.20 -0.2 Measured on the 25th.-- From m.easvresents ~trade by Mr. Yamf~da, 01 kter Resorlrces Survey hit. 3. Thaw-layer and Permafrost-layor Stratification

Detailed determinations were made of the stratification of the thaw- layer and permafrost layer in walls A,B,C ,D (as designated in Fig.3 ) of the pit used by us Fn our investigation, narnely pit No.3. The results are shown in Fig,)l, which is the developed elevation of the four walls of the pit. The depths shown in this figure are measured from a horizontal reference-plane passing through point P of Fig .3, the lovest-lying point of the ground surface.

The soil stratification-structure around the pit walls was as seen in Fig.b. The frozen gravel layer at 3.115 m forms a quite horizontal floor. On it there lies a 1.2 meter thick stratum of blue clay containing a great deal of segregated ice. This is covered with a layer of sandy loam about 10 cm thick. Still higher is a stratum of brawn loam, in a thawed state, about 2.2 m thick. As the top of this loam approaches the ground surface, the amount of humus in it increases, so that it merges into the black humus topsoil. In the present case, this 2.2 m thickness of brm loam constitutes the local activo layer Fond.+:- The point to be noted in Figs.3 and 4 is that in the middle of the blue clay layer in the permafrost there is a layer of peat roughly 10 cm thick, which has a wavy form. Since the stratification of all the other soils is roughly level, the wavy foze.m of .this peat layer is an outstanding feature, It means that this peat layer was uneven from the start, and it got covered over eth the blue clay which formed a level surface. Above this again, and several times in succession, other layers of even thickness accumulated.

We shall now proceed discuss these stratification phenomena in detail.

4. C~n&~ion-.structure of the Permafrost Layer

We made a st,ratification sketch of .the permafrost column at the point [in a pit walg wherVe the sfgregation of ice-strata was most developed, that is, where the amount of frost,-heaving was presumably greatest, and another sketch for the point where there were least ice-strata, that is, where the frost-heaving was pres~mablyleast. These points were, respectivoly, between

Tho surface strata in the permafrost belt thaw out to a certain depth in summer and freeze up again in winter, in most cases freezing solidly to the underlying permanent1.y frozen layers. Those upper strata are called the active layer. In Northern Manchuria the depth of the active layer is about 2 m. According to Shakuniantz, the Manchurian permafrost is the southern limit of the Siberian permafrost. lines 5 and 6 on wall B and between lines 9 and 10 on wall C in Fig.4. From the columns at these points, samples of congelation-structure were taken. These sketches are shm in-Fig.5 (col~mof n~inimumfrost-heaving) and Fig.6 (column of maximum frost-heaving ) . To obtain the profile sections, we cut away the thawed soil on the wall, going a *rther 20 cm inward from the measurement-plane of Fig.4; from this new face, the thiclcnesses of all the layers were measured, and samplos were taken. For this reason the strat~Lficationdiffers a little from that of Fig.b. The squares in Figs. 5 and 6 show the positions fron which the samples were taken, and the figures in the squares are the sample numbers. These positions were selected as representative of the congelation-structure and soil-types for each part of the column.

To describe the congelation-structure, we shall use the nomenclature which has been applied to the case of frost-heaving.*

In the profile of Fig.5, the layer of frozen peat, about 7 cm thick, appears at a depth of 266 cm from the ground surface. Below this we see that the structure shows a gradual variation from the "frostff typo of congelation to the "ice with soil inclusions" tne. This congelationstructure, as in Fig.5, is found on wall C, and also around lines 1 and 2 on wall A, line 7 on wall B, lines 12 and 1h on wall D.

Now we turn to the remarkable coneelationstructure in the blue clay layer under the peat in Fig.6. The soil below the peat layer bhree times in successiog starts from the ffconcretedflstate +w and ~aduallyincreases its content of ice-strata, passing from the fine frost type of congelation to the frost type, then to frozen soil with ice-lenses, then to ice vdth soil inclusions, and finally to pure ice. Around the junction of walls A and B in particular, several pure ice strata, of thicknesses 10 cm and over, were found. In Fig.6, we see three such clearlyaarked progressions, one above the other. This congelation-structure as in Fig.6 is seen, outside of wall B, around line 3 on wall A and around 13 on wall D.

In taking our samples from each of the profiles, the frozen soil was first sliced out in comparatively large blocks and then separated according to each type of congolation. In the case of layers where the congelation- struct~rrevaried in a continuous manner, we took, for convenience in calculating the frost-heaving ratio as discussed infra, a portion exhibiting an intermediate type of congelation. Immediately after thus cutting up the samples, the surfaces were prepared and photographed. u U. Nakaya and C. Magono. The Mechanism of Frost-Heaving; a Field Survey. Kish5 ShEshi, series 2, ~01.18, No .lo, (1940). Nakaya and Magono. An Experimental Study of Frost-Heaving . ~ish6Shkhi, vo1.20, No .S , ( 1942 ) . H That is, a cemented state without visible ice-segregation. ransl slat or.) The photographing was done at times just after sunrise or just before sunset, when the atmospheric temperatures were relatively low. These temperatures being as high as +2-10oC, the taking and photographing of the samples had to be done in haste, and for this reason the error in tho volume determinations (infra) is rather large. For photographing the congelation-structures, we used a Leica camera temporarily fitted with a 135 mm Zeiss Sonnar lens. Representative examples of our photographs of the congelation-structure3 in the frozen soil are shown in Figs. 8 to 16. Fig.8 is a photo of the loam layer at 2 in Fig,S. Fig. 11 is the permafrost in the blue clay layer at 13 in Fig,6. As may be seen in these photos, the frozen soil above the peat layer shows very little segregation of ice-strata. Fig.9 and Fig.10 show the permafrost below the peat layer in ~ig,S;here the ice-strata are comparatively numerous, Figs, 12 to 15 are photographs of the permafrost just below the peat layer; they show in detail the progression exhibited in Fig,7, that is, the content of ice-strata gradually increasing from the llconcretedll structure. Fig. 16 is a specially selected photograph from below tho peat layer at lj-ne 4 on wall B.

Wow when we examine the above photographs, we find that the frozen soil in the permafrost layer has a congclation-structure generally very similar to that seen in the phenomenon of frost-heaving. In particular the soil in Fig,l6 exhibits a truly typical 'Ifro~t'~type of congelation. Therefore we may conclude that this permafrost was formed under conditions quite similar to those in the frost-hewing which occurs close to the ground surface, In all cases, however, a detailed microsco?ic examination of the structure of the ice- strata in the frozen soils shows that they consist of transparent ice, In the ice-strata of annually frozen soil, there are generally vertical air-columns or air bubbles, and the interfaces between the columnar ice crystals are distinctly seen, but in our permafrost layer these characteristics were practically undetec- table. This is evidence that the frozen soil here is permafrost. The ice- strata in the permafrost bed consist of ice which was formed a iong time ago; the crystals, through standing for a long time at temperatures in the neighborhood of zero (centigrade), have been fused together by recrystallization into masses of pure ice.

Immediately after taking the congelation photographs, we cut the frozen soil into cubes of about 7 cm a side, and measured their volumes; then they were stored in hermetically sealed tins, for later weighing, These samples were dried out in the field by heating the tins over hot water, and again in a dryer at 100~~after our return to the Institute, Then the dry weights were determined,

Our original intention, after collecting our samples at the point of maximum frost-heaving, was to excavate the frozen gravel layer in order to determine its limiting thickness. Hmever, the ground was so hard that the work made little progress: two laborers got dom barely 15 cm in a day's digging, At this depth we took permafrost sample No .22.

Although the depth of the pit had now reached 3 ,5 m, the vertical walls stayed up quite safely, in a very stable condition, wl-rilo Inre were thus taking our soil samples, This was because the thaw-layer at the top, approximately 2 m thick, had just the right ~atercontent and was in a very compact state. No "run-out layerft or layer of extren~elyhigh water content was encountered at the interface between thc? frost layer aid the thaw layer, nor indeed between any of the other layers. The so-called l1loonV odor was very strong in the pit, so much so that, it was not easy to remove the smell which adhered to clothing, etc.

Collecting the samples took three full days1 time. The frozen soil weights, relative water contents, and apparent specific gravities as determined for all theee samples are shown in Table 3. A grain-size analysis for the permafrost is shown in Table 4 and a similar analysis for the gravel in Table 5.

As may be seen In Table 3, the frosen soils all contain a very large fraction of water, the maxirmrm water content being about 80$, with averages about 50% and over. The peat was in a fairly advanced state of decomposition, corresponding to what is called a low-grade brown coal. Table 3 Point of minimumfrost-heaving, Wall C, Fig.5

Sample No. Weight of Weight of Water Ratio of Ratio of Frozen soil Apparent Type of Soil type Photo frozen soil, dry soil, content water to water to vol~e(cc) specific congelation ma (w) W (gr) dry soil frozen soil gravity

1 172.7 161.5 11.2 0.07 6 .5% - - Thaxed soil Reddish-brown loamy clay -

2 3U .3 229 .O 83.3 0.36 26.7 192 1.63 Coarse frost Reddish-brovm Fig. 8 sandy loam

3 288.0 199.5 88.5 0.U 30.7 17h 1.66 Concreted Silty blue clay

b 267 .O 161.3 105.7 0.66 39.6 - - Fine frost Blue clay 5 2bl .l 97.6 U3-5 1. 59.6 - 1.U ldassively frozen Peat

6 232 .O 110.6 121.b 1.10 52.11 - - Coarse frost Blue clay, Fig.9 humuscontent

7 220.3 76.2 Ub.1 1.89 65.6 - -. Frost ~Ith Blue clay, Fig .lo ice-lenses humus content

6 121.8 bl.O 09.8 1.97 66.b - - Ice-layer with Blue clay, soil inclusions humus content 9 298.7 19b.Q 10b.7 0.511 35 .l - - Finefrost Brm loan Point of madmumfrost-heaving, WallB, ~i~.6.

10 h9.6 33 .O 16.6 0.50 33.5 - - Fine frost Reddish-broun sandy loam

11 123.3 82 .O h1.3 0.50 33.5 - - Fine frost Brmsilt

12 260 .o 192.7 67.3 0.35 25.9 1h9 1.7h Concreted Silty blue clay

13 193.5 118.5 75 .O 0.63 38.8 - - Pine frost Blue clay Fig.11 lh 86 .O hl.l U.9 1.09 52.2 78 1.09 bkssively frozen Peat

15 306.5 212 .O 9b.S 0.h5 30.e - - Coarse frost Blue clay, Fig .12 humus content

16 305.3 159.0 llt6.3 0.92 h8.0 250 1.22 Frost Blue clay, Fig .I3 humus content

17 30h .O 127.3 176.7 1.39 58.1 220 1.37 Frost with Blue clay, Fig.llr ice-lenses hscontent 18 288.1 120.2 167.9 1. 58.3 233 1.2b Ice-layer with Blue clay, Fig .l5 soil inclusions humus content

19 83.5 h6.5 37 .O 0.80 h4.h 63 1.32 Frost Blue clay, humus content

20 2h9.3 h7.0 202.3 h.30 81.b 2 h6 0.99 Ice-layer with Blue clay, soil inclusions humus content

21 118.8 6b .7 511.1 0.8h b5.6 89 1.3b Fine frost Brown loam 22 239.5 161.7 77.8 0 .L8 32.5 - - Concreted Gravel

2 3 118.2 65.0 53.2 0.82 h5.0 83 1.39 Typical frost- Blue clay Fig.16 state congelation Table Ir Results of Grain-size Analysis, Frozen Soil

Sample No. Fine soil (<2.00 m) , percentages Remarks

2-0.25 / 0.25-0.05 10.05-0.01 10.01-0.005 1< 0.005 1 Thaw layer, clay Permafrost layer (upper part)

Just on top of gravel layer Permafrost layer (upper part)

Blue clay, representative samples

Between blue clay and gravel layers mica1 permafrost

Table 5 Results of Grain-aiee Analysis, Gravel

Diameter (nun) Content in % 5. Amount of Frost-Heaving in the Permafrost Layer

From the survey results which have just been described, we see that the frosen soil in the permafrost stratum has quite the same form as is found in localities where frost-heaving has occurred. Consequently when any building is done on the active layer at the ground surface, the heating or thermal insulating effect of the structures will cause a thawing of the permafrost, and this will naturally cause a considerable subsidence. Let us nuw try to calculate the frost-heaving ratios in the soils above the gravel foundation layer and the actual amount of frost-heaving in the two sectional profiles which we have studied, and from this to Wer the amount of subsidence which will occur when the soil thaws out.*

If V is the frozen soil volume and Vo the original wet soil volume, then according to the findings of the author and his colleagues, i+++the frost- heaving ratio in the frozen soil may in general be represented by the expression: v-v, Fi P -v

Moreover from experimental research, V and Vo in equation (1) are:

Since v,, vi and v, in these equations are, respectively,

vw = makro (11 we see that they can be found experimentally. The notation here is:-

v, v, = volume of dry soil particles

vi = volume of ice

v, = volume of water in the met soil

W = water content in frozen soil

G, = true specific gravity of soil particles u We may here consider that the gravol layer does not undergo any frost- heaving, and thus has no subsidence. w U , Nakaya and J . Sugaya, The Mechanism of Frost-heaving; a Field Survey. Teion Kagaku, 2 (19149). - 11 - ma = weight of dry soil

ro I ratio of saturated water content to weight of dry soil. The coefficient k which modifies ro has in most cases a value of about 0.&-0.9. C1 and C2 are constants to allow for the amount of air-spaces in frozen soil and wet soil respectively; they are generally in the range 1.0 to 1.1. Since C1 = C2, these constants may be dropped from the above equations. Accordingly, when we substitute (2), (3) and (b) in equation (l), we have:

NOW we use (5) to calculate the frost-heaving ratios of the soils in Table 3.

Table 6 shows the measured values of Gs and ro for the soils in Table 3. Appropriate values of k are 0.9 for loam and 1.0 for clay, and the corresponding Values of kro are thus 0.35 and 0.44.

Table 6

Soil type Sample Nos. Gs ro kro dense very dense mean

Brown loam 2 +9 +10 41 2.49 0.42 0.36 0.39 0.35

3 +6+7 +8 Blue clay 11+12+13 2 .I46 O.lr8 0 .39 0.44 0.L4 +15-20

Using the figures in Tables 3 and 6, we now calculate V and Vo, and then the frost-heaving ratio for each soil, with results as shom in Table 7. Then we multiply the values by the frozen soil layer-heights for each type of congelation, as seen in Figs.5 and 6, and thus obtain the amount,^ of frost- heaving. The results are given in Table 8. The total values shom in the last column represent the amounts of frost-heaving in the two vertical profile- sections, namoly 34.7 cm at the point of minim frost-heaving in Fig.b, and 50.2 cm at the point of maximum frost-heaving in Fig.k.

We conclude that when these frozen soil layers thaw, subsidence will take place in amounts roughly oqual to the above amounts of frost-heaving. Consequently when we undertake to put up buildings or other structures in these permafrost areas, it seems obvious that we should first, of all dig test-pits to determine tho depth of the permafrost, whether it is frost-heaved, and what, the amounts of frost-heaving are. Then the work could proceed on a basis of precise calculation and planning. 6. Discussion-- of the Origin of the Permafrost Bed.

Now we come to the question of what form of permafrost the above- described frozen soil layer is, and here we find many questions arising. Smgin +t says that the region of our survey is on the southern boundary of the permafrost sheet, and this is confinred by later research carried out by the bfanchurian Rail.wav. Vle have spoken of finding ice, at Assan, to a depth of around 15 meters below the ground surface, and from r~inter~easor~earth- temperature measurements at Assan village, the thickness of the active layer in this region mas fourid to be about 2 m.Hc Thus there is no doubt that the frozen soil at the point where our study was mda is indeed the permafrost.

Nmwe shall try to draw some conclusions re the origin and formation of the permafrost layer at this point.

i) Classification of Permafrost according to Origin..-

As regards the origin of permafrost, it has been argued that there are two cases. The first is when very lor atmospheric temperatures cause permafrost to form, by the draining-off of heat and gradual penetration of frost into the ground. Tho second is when a surface layer of frozen soil is buried by somo sudden occurrence, so that it never thaws out again.

Now from the Soviet Union we have numerous studies in connection with the first of these cases, but they may be regarded only as general theorizing about the origin of the presently existing permafrost layer. Thus we have still to hear of any evidence, qualitative or q~~antitative,based on actual examples of the causation at individual places, Moreover, because of the lack of observational data, we ar5 still ignorant of whether the permafrost at any particular place is, under the present climatic conditions, in process of forming or in process of disappearing.

As for the second case, we may take it that this can occur from time to time in the permafrost zone. This is amply evidenced by the remains of mamrrioths which, exposed by recent earth-novements, are discovered from time to time at pohts 3x1 the Arctic Ocean coastal region where the permafrost is fairly thick; these cadavers must have been buried suddenly, in a fresh condition ,M

w Me Sumgin. Permafrost in Soviet Territory. Vladivostok, 1927. (Translator's note: See item 17189, Arctic Bibliography, Arctic Institute of North America, 1953. )

wt U, Nakaya and J, Sugaya. A Report on Permafrost Surveying (Manchuria 1943 ), Teion Kagaku, 2 (1.949), 119. Defence Research Board Translation T 25 J; National Research ~ouncnof Canada, Tech. Translation 3 82.

3.w E ,W .Pf itzenmayer . In Search of lfammoths . Trwslated. by Ken j i Hashiguchi , TC~EBunko, Dairen, 1943. However, if we find that instead of dependlngon such accidental discoveries we can reveal the origins of the permafrost by study of the stratification and congelation-structure ln the permafrost as it now exists, then this is something which may furnish important clues in permafrost research. Let us examine the results of our survey from this point of view, ii) Origin of the permafrost at Assan as deduced from the stratification picture.

If we examine the permafrost bed in Fig.4 from the point of view of stratification, we see a group of layers deposited on the nearly level foundation layer of gravel: at the bottom the silty loam layer, then a blue clay layer, and on this the peat layer. Over the peat there is a second layer of blue clay, obviously a later deposit. Above this again there is a third group of layers, reddish-brown sandy loam and brown loaqy clay, representing several successive deposits. That the three groups of soil-layers were laid down at different times is evident from the presence of the peat layer, and from the fact that the second layer of fine-grained blue clay is overlain by layers of coarse-grained loam. Moreover, from the fact that this second blue clay layer and the loam layers over it rise highest above the horizontal plane at line 8 on wall C (Figs. 3 and L), we realise that the deposits here have been washed down, three or more times in succession, from this direction, that is, from up the present ground-slope.

The question here arising is whether the soil below the peat layer was frozen before it was covered over by the earth and sand nuw lying above it. As.we see from Table 8, the amount of frost-heaving in the layer under the peat, for instance, at the point of maximum frost-heaving, is L1.5 cm (in this layer alone). If it is assumed that the freezing took place after this layer was covered over, then the upper surface of the second clay layer would have been pushed up into the same unevennesses as the peat layer. Actually, however, this boundary is nearly level..

The same conclusion follows if we consider the frost-heaving in the second clay layer. As may be seen in Table 7, the amount of frost-heaving in this layer is comparatively small. The fact that the layer is thick at one side and thin at the other cannot be due to differences in the amount of frost- heaving.

All the above circumstances indicate that frozen soil layer under the peat was already frozen before it was covered up, with the same congelation- structure as now. Thus we know that the higher layers were deposited on top of the f'rozen peat layer which had already been buckled up into hills and hollows.

iii) Origin of the frost layer as deduced from the congelation-structure.

The above may also be deduced from the congolation-structure. Vie have seen in Section 3 and in Fig.7 that the frozen soil layer under the poat displays a continuous progression in congelation-structures, from the "fine frost" type to the "frost" type, then from frozen soil with interspersed ice- lenses to a pure ice layer. Moreover, this progression is repeated at different levels. As may be seen from the selected photograph in Pig -16, the permafrost shows quite the same congelation-structures a3 the single-year frozen soils - 1ll - found in frost-heaving surveys in Hokkaid6 and elsewhere. These structures are quite dflferent from the congelation-structures in permafrost formed at similar depths in the soil by steady-state thermal conduction, as seen in another case by the author, during a survey carried out in 1945 in the Manchurian hinterland.*

These points all suggest that the permafrost under the peat layer is soil which was frozen at generally the sane rates of congelation as observed in freeaing which takes place close to the ground surface and under the influence of the external atmospheric temperature variations.

Moreover the fact that these layers of the permafrost have under- gone a good deal of frost-heaving is no doubt duo to tho fact that the gravel bed, sinco we find it frozen in a water-saturated state, must Eefore freezing havo constituted a good source of water.* Table 8 shows us that the amount of frost-heaving of the second blue-clay la.yer over the peat is very sma3.1, in spite of the fact that the character of the soil is about the same as in the blue clay layer below the peat. Again, the fact that the soil, as in the sample of Fig.12, shuvrs frost shrinkage, like that seen in the case of "closed system'' frost-heaving,+= indicates that in this second clay layer the congelation took place under conditions such that the supply of water from belm was cut off, because of the peat layer's being already frozen.

Nuw as regards the freezing under the peat layer, if from Table 8 we calculate the wet-earth heights, before freezing, at the places of maximum and minimum frost-heaving, and if we compare these heights, then in each case the height ho of original wet earth turr~sout to be around 32.6 a 4 .O cm. Thus we may take it that the blue clay under the peat, before it froze, was roughly level in stratflication.

The uneven frost-heaving of the said layer was probably due to the varying thickness of the peat layer which topped it. In fact, if we examine Figoh we find, generally speaking, that at places where the frost-heaving is great the peat layer is thin, and vice versa, Moreover, at the time when the peat layer was at the ground surface and immediately after it got covered over with earth and sand, it had a considerable thickness, a conclusion which we may draw from the fact that density of tho layer is quite high, as ore see in Table 3. At the time when it got buried, this thick peat layer constituted an efficient protection against the frozen earth's thawing.

+t J. Sugaya. Permafrost Survey at Shiichi in the greater Khingan Mountains, 1945. (Unpublished. )

+++ Stephen Taber. Frost-heaving. J. of Geology, (1929), 428: -38 (1930)~301. '?~~~?Oe~"Oe-f"O,?pSOe~aq(??~~ rlrc~u\u\rlr--r--O\ooOrl~rlOCU\Och rlrl3cON rc3 rl(?rlO\(?mcUomcu rl 1 rlrl cu I CUeSe'sc';\?'Z~o*'Z~e-~~oepSoe-?me~q sto\Or\;fLtP--m0\moO\u\rccuO\mchu\ +\OmO\\Om CU\O\OO+~OOmm~ln rlrlrl 3 rlrlrlrlrlrl \

d+\O maChmcOmAN3rlAaO mO\a O\o\rlmm-rlrl~+cOO'uO\azt(Uu\u\ rlrlrl rl rlrlrlrl cu I

...... _ .... I ...... _.,+ ...... __ _C.. .._ ??c';~-i"3.7'?9"?Q.V!?pSo':9~r!CU. m-V\rl=JOA\Orl rcm-j'Q\O eCccvz3m a3aOcva~Orla\OtcO\~tcQmOu\~ rlrlrl rl rlrlrl cu ...... " ...... ,.--. 0 =?'??2='!?993\??9?c".'?9Y? Cumr(O\Or(-~mCUcu-cum+Owrccj~ Cu m\orlCc3O\ma3O\rlrlV\nJCv3~=T\Da rlrlrl rl rl rl CU rl I+ rl II Table 8 -- - Sample Frost- Height of Amount of Amount of Strata Total No. heaving frozen soil frost- frost- divisions frost- ratio for each heaving, heaving, heaving % congelation for each for each (cm > structure congelation layer (cm) (cm> structure (cm)

2 6.2 5 .g 0 .3 0 .3 Sandy loam ___,._.I,..L ._.,._Il.~..,-.,..~.,- *-----*.- ...... --*--.------..- _ layer 3 10 .O 11 .O 1.1 upper blue -4 -,.--.-...., 24 a9...... ""...... ? ,4 .., ,,.,., ,., ,.-.. ,.,.,9*6 -.w..._.,,,,,,,.l.,~L clay layer (4)* 2b.9 13.O 3.2 Lower blue 6 47 .6 18.O 6.6 clay layer 7 66.2 14.O 9 .1 ...... &&--**.8 672,..,,.,.*m*pv-. 14.*!?*..-, ,-----.. >.. 2 *4.n:-3K21 9 ..,..-...... -.--.~-..-..-.-...... ,..,..,.,...... 24.2 10 .O . 2.*4..-...... ,.,_...... ,.. 2:k- Brown loam 34.7 layer

Sandy loam layer Uppor blue clay layer %3-&----. 2? 02 -.------10-0- -...----.--. ? .3" -- 2 .L 15 5 -6 14.Q 0 .Y 16 40.3 10.O 4 .O Lower blue 17 56.2 22 .o 12 .3 clay layer Pure ice 100 .O 7 .O 7 .O layer Pure ice 100 .O 2 .O 2 .O layer 19 34 .O 6 .O 2 .O Brown loam 50.2 layer

9 No soil samples taken here; the amount of frost-heaving shown has been calculated from the thiclonesses of the structures in the layer, using the f rost-heaving ratio found for the @eighborind point of similar congelat ion- structure. CONCLUSIONS

An excavatory survey was carried out in the permafrost belt of North Manchuria during the season of greatest thawing of the active layer. It was found that the thav penetrates to the vicinity of the 2 m level; below this is the permafrost formation. Soil samples were taken both from the active layer and from the permafrost layer, and tho water-contents, congelation-structures, etc., were determined. The study showed that the frozen soil layer has congelation-structures similar to those observed in ordinary frost-heaving near the ground surface; in fact the presence of strong f rost-heaving vras evident.

After collsting these results, the permafrost layer was again studied from the points of view of stratification and congelation-structure. It was concluded that the frozen soil under the stratum of peat in the permafrost layer had once been at the ground surface and had frozen and frost-heaved, after which it got buried to its present depth by earth which, several times in succession during the spring thaws, was washed down from the slope above; consequently it no longer thawed, and thus became part of the now-existing permafrost bed. We found that we were able to confirm this origin qualitatively by our study of stratification and congelation-structure in the permafrost layer.

The above survey was carried out under the direction of Professor Ukichir6 Nakaya, as part of a research program of the Manchurian Railways1 committee on Measures against Extreme Cold. During the field work, assistance was received from Vice-I)irector Takano of the Construction and hlaintenance Bureau, and from the personnel of the [Manchurian Railway] Construction Office at Tsitsihar. The author takes this opportunity to express his gratitude.

Fig. 2

NORTHEAST C------Y POINT WHERE GROUND TEMPERATURES TAKEN LlNE 4

n 0.5M ?' LIN~3 - LINE 6 LINE7 /' / - + + /@- / 7 0 0 W / ' f LINE 6 0 0 0 A 0 cnJ LINE 2 9 0 ' J 0 /' C cn 0 W / C LlNE Q 5 w t /' f C / C 0' 0' Z 0 a 0 3 3 LINE t 0 + 0 f LINE 10 / 5: / D P .,/' + I' f cp L ,/ LINE I4 LINE 13 LINE It LINE II k' 1.0 M LlNE OF . MAXIMUM SLOPE te.sO)

Fig. 3 o THAW LAYER 0 0 -a 0 a"- PERMAFROST .E 0 (ACTIVE LAYER) LL LOO 0 ------

o n 6 Fig. 5 Fig. 6 PERMAFROST PROFILE, WALL C PERMAFROST PROFILE, WALL B

T - Permafrost table (not T - Pertrafrost table clear1y defined) Fig. 7 Fig. 8

Fig. 9

Fig. II Fig. I2 - - -.----- ",.-"--.-. .. .. - .. - - Fig. 13 Fig. 14

Fig. 15 Fig. 16