Ice Conditions in Eastern Europe and the Methods Employed to Influence the Formation and Break-Up of the Ice on the Dvina (Daugava) River Kanavins, E

Home , Drift ice, Ice bar, Ice bridge, Ice Ih

NRC Publications Archive Archives des publications du CNRC

Ice conditions in Eastern Europe and the methods employed to influence the formation and break-up of the ice on the Dvina (Daugava) River Kanavins, E.

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/10.4224/20331607 Technical Translation (National Research Council of Canada), 1948-12-09

NRC Publications Record / Notice d'Archives des publications de CNRC: https://nrc-publications.canada.ca/eng/view/object/?id=20f7d4a5-2c7f-4272-85cd-7d2119a58dee https://publications-cnrc.canada.ca/fra/voir/objet/?id=20f7d4a5-2c7f-4272-85cd-7d2119a58dee

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 RESGARCH LABORATORIES Ottawa, Canada

TSCHNICAL TRANSLATION

Division of Mechanical Engineering

Pages - Preface - 4, Tech, Trans, - TT-85 - Text - 46 Date - 9 December, 1948 Figures - 55 Lab, Order Noo 5424A Tables = 4 File - 12-R4-22

Title: Eisverhaeltnisse in Osteuropa und die Angewandten Massnahmen zur Beeinflussung der Eisbildung und des Eisganges in der mna (Daugava) ,

By: Edvigs Kanaving , J Reference: Report of the Directorate of Technology and Transport, Division of Hydrology, Riga, 1944,

Subject: ICE CONDITIONS IN EASTERN 3UROPE AND THZ METHODS EMPLOYZD TO INFLUTNCE THE FORMATION AND BREAK-UP OF THE ICi3 ON THE DVINA (DAUGAVA)RIVER,

Submitted by: Do Bo Nazzer, Translated by: Secreta~y, Do A, Sinclair, Subcommittee on Soil an3 Snow Mechanics,

Approved by: J, H, Parkin, Director, Page - (ii) Tech, Transo TT-85

PREFACE

The following report was prepared at the re- quest of Admiral Dr. hoco F. S iess, President of the German Naval Meteorological Servicesx$ , for presentation at the con- vention of the Naval ~,!eteorologica~Services in Hamburg, A supplementary section is added, setting forth a plan of the future programme for improving the troublesome ice conditions which mere experienced on the Dvina Hiver after construction of the electrical power plant at Keggum, This programme has been prepared on the basis of the observations described in the present report and the investigatio s of the Depart- ment of Navigable '?daterways and Harbours=f The main part of this report is taken up with a description of the investigation of the ice conditions carried out in the winters of 1941-42 and 1 42-43, and of the rnethodi employed for reducing the depth ice2) in the swift- water section of the Dvina. It should not be regarded as a complete exposition of the material or of the observations and investigations, but merely as an informative statement of the results obtained and the usefulness of the methods employed. After the material contained in the research results has been elaborated it is intended that everything should be made available in a more complete publicationo

It is a pleasant duty for the author to express his sincere appreciation to Admiral Dr, hat. Fo Spiess and to the Chief Office of Technology of Latvia for making it possible for him to publish this work.

Riga, 19440

n) Die Deutsche Seewarte.

xx) D~partementfh Wasserstrassen und Haf en. t) Translator fs Note : The explanation of this term is given on page 3, Page - (iii) Techo Trans, TT-85

TABLE OF COITTEX'TS Page- Preface (ii)

General Review of Ice Conditions on the Rivers of Eastern Europe 1 The Investigations of the Latvian Department of Marine on the Dvina up to 1940 5 Estimate of Ice Conditions on the Dvina after Co~npletionof the Keggum Power Plant 8 Investigations of the Department of Marine on the Dvina in the Winters of 1941-42 and 1942-43 15 (a) Formation of the Ice: The Various Types of Ice and their Properties 17

(b) The Freezing of the Swift-Water Section of the Dvina 22 (c) The Quantities of Ice in the Dvina and their Distribution 24 (d) Ground Ice 25

(e) The Reduction of the Drift Ice Driven under the Ice Cover (Tost) 28 The Means Employed for Reducing the Depth Ice 29

(a) Border-Ice Bridges 29

(b) Drift Ice Dams 32 (c) Artificial Production of Depth Ice 34 Possibilities of Obtaining more Favourable Grouping of the Depth Ice in the Swift-Water Section of the Dvina with the Aid of Ice Bridges 36 Page - (iv) Techo Trans TT-85

Table of Contents (contfd)

Page VII Evaluation of Research Results; Conclusions 39

VIII The Inadequacy of the Previous Methods; The Future Program for Improving the Troub.lesome Ice Conditions on the Dvina Since Completion of the Keggum Power Plant 41 (a) Inadequacy of the Maximum Program Drawn up by the Swedish Firm 41

(b) The Proposed Interim Program for Decreasing the Break-Up Difficulties on Lake Keggum 43 IX Concluding Remarks 45 Page - 1 Tech0 Trans, TT-85

ICE CONDITIONS IN ZASTSRN EUROPE AND THE METHODS EMPLOYED TO INFLUENGE THX FORMATION AND BREAK-UP OF THZ ICE ON THZ DVINA (DAUGAVA)RIVER

I GENERAL REVIEW OF ICE CONDITIONS ON THE RIVZRS OF EASTERN EUROPE Because of the troublesome ice conditions on the rivers of Eastern Europe the design, construction and operation of various hydro-electrical plants is considerably more complicated than in the countries of Western Europe, Hence, owing to the increased economic exploitation of water resources in the eastern countries, more and more attention is being paid to this problem, and special observations and investigations are being undertaken in order to discover ways and means of mitigating the destructive effects of the ice 0 The course of the phenomena is in general very complex and has thus far been the object of little research, It depends on a great many different factors of which the following are the most important:

1, Climatic factors, which determine the / length of the winter, or ice period, and the thickness of the ice cover; 4 2, Geographical factors, in which the geographical zone and compass direction of the riverqs course come under consideration; J 3. lVIorphologica1 factors, which characterize the river bed, the flow, etc,

The severity of the ice conditions depends mainly on the climate, In places where the winter is relatively long and severe the ice conditions are more troublesome, The thickness of the ice cover normally increases from south to north, Yet in the rivers of Eastern Europe, even in the far north, it does not exceed 1020 metres, It is noteworthy that the ice cover in the rivers of Siberia is considerably thicker than this, The geographical situntion of the river is closely related to the climatic conditions, The zone in which it lies and its direction of flow are of great im- Page - 2 Tech, Trans, TT-85 portance for the ice conditions, If the river flows from north to south (Volga, Don, Dnieper)t) the processes of ice formation and break-uptt ) follow a pelat ively normal course. The lower reaches of the river become free of ice before the upper reaches, Conditions are altogether different if the river flows from south to north (Northern Dvina, Pechora, etc,), In such rivers the break-up usually attains the proportions of a natural catastrophe, It begins, as a rule, in the upper reaches of the river while there is still a firm ice cover in the lower ones, Immense ice jams form which result in a sudden increase in the water level and disastrous floods, In such cases the masses of water, seeking a way around the ice jams, demolish the banks, destroy settlements, tear open new arms of the river, carry away sand from the old river bed, damage man-made structures; embankments, etc . In the rivers which flow from east to west (~vina,Memel, Neva, Svir, etc,) the ice conditions may vary widely, In cases where, due to unfavourable meteorological conditions, the break-up begins in the upper reaches of the river special difficulties may arise which demand great expenditures of effort and material on the part of the authorities and or- ganizations who must battle with the ice, The morphology of the river is a very important factor determining the nature of the ice conditions at different points, In places where the gradient changes suddenly, or where the river takes many turns and is of unequal width, or where there are a great many shallow places and islands, dangerous ice jams which are difficult to eliminate usual-ly form during the break-up, In the countries of Western Europe, where the rivers are rewlated, this circumstance is not so t) Translator's Note: Insofar as possible the spelling of the geographical names has been taken from the Encyclopaedia Britannica, Translatorfs Note: Two terms are used here, "~isaufbruch", which would refer to the spontaneous breaking up of the ice cover into floating masses, and "Eisgangn, describing the passage of these masses down-stream with the current, Subsequently the latter term appears to denote both these processes and the translator has therefore just employed the one term "break-up", Page - 3 Tech. Trans, TT-85 important. In Eastern Turope, however, almost all the rivers are still in their natural state, and for this very reason the course of the ice conditions is greatly complicated by the morphological factors, The precipitous, swif t-current places in the river do not usually freeze over. Because of the frost the water masses at the open places become supercooled and ice begins to crystallize .through the entire cross-section of the river from the surface to the bed, A completely new kind of ice is formed - a soft, sticky mass of ice with a structure and properties quite different from those of the ordinary surface ice. Walter Sperling suggests calling these masses "depth icelll) (Tief eneis) , The current carries these lumps of soft depth ice downstream and packs them into immense masses beneath the solid ice cover in layers which are several metres thick, Thus dangerous ice jams are formed, IJumerous examples, a few of which will be cited, bear witness to the devastations which have been caused by ice jams of this type on the rivers of Russia,

In the winter of 1902 an ice jam caused the water level in the Neva to rise 2075 metres above Ordnance Datum (Germany) and a part of Sto Betersburg was flooded, St, Pe tersburg experienced similar floods in the winters of 1903, 1911 and 19240

The water works in that city, also, have had to contend repeatedly with considerable depth ice difficulties Disastrous stoppages of the intake conduits occurred, for example, in 1894 (twice) and 1914, so that the entire city was without watero The water conduits of industries were also clogged, Howevsr, the normal water supply of the city is disturbed by depth ice almost every yearo Depth ice forms even in the lower course of the Vol;=a, Thus, in 1925, for example, the water intake pipes in Astrakhan were clogged,

The problem of combatting the unwelcome masses of soft depth ice became very accute in Russia when construction began on the new hydro-electric power plants on the Volkhov, Svir, Dnieper and other riverso It was discovered that the soft depth ice clogged the twbine grids so that operation of the power plants had to be entirely sus- ~ended, 1) Walter Sperling: Fragen der Xis-, bildung und des Eisganges in den Fluessen, e twaige neuere Erf ahrungen und Gesichtspunkte (Problems of Ice Formation and Break-up in Rivers, Some Recent dxperiences and Aspects), Sixth,,Balto IIydr Conf o, 1938, Report 3A, Lubeck-Berlin, 1938, Page - 4 Tech, Trans, TT-85 In December, 1928, the several hundred square metre grid structure of the turbines in the Volkhov Hydro- electric Station was covered with a soft layer of ice approximately one metre thick, The total weight of these ice masses was several hundred tons, The ice was frozen so hard to the grids that when attempts were made to dislodge it with cranes the 25 m, steel cables broke like pieces of twine, The grids could only be cleaned by divers, after which they were raised, heated with hot water, and replaced, All this was of no avail, the grids were soon clogged again, Twenty-five divers had thus worked for five days to no purpose, It was not until a rather long stretch of the Volkhov had been frozen over that the masses of soft depth ice vanished and the operation of the power plant could be resumedo Considerable depth ice difficulties beset other power plants in Russia, Many of them are obliged to suspend operation for weeks at a time, thus sustaining great losses, This happens almost every year, In general the effect of the ice on hydro- electric installations is shown in the following ways: 1, Statically - due to pressure from the expansion of the ice or due to freezing and fluctuations of the water level; 2, Dynamically - during the break-up; 3, By becoming clogged with depth ice, The static effect may be avoided fairly easily b:y breaking up the ice, The dynamic effect during the break- up can also be mitigated to a considerable extent by breaking up the ice cover ahead of time and reducing the size of the lumps, Various defences may also be erected against both of these effectso However, no effective remedy has as yet been found for the obstructions and cloggings caused by the depth ice e The following quotation from the periodical, "Elactro-~nergetics in Soviet Russia", will Illustrate the difficulties accompanying the struggle against the ice in Eastern Europe and the means required to deal with them: he official loss coefficient for each idle kilowatt hour is four rubles, The stoppage of a power plant of 100,000 kw, output means a loss to industry of 400,000 Page - 5 Tech, Trnns, TT-85 rubles per hour, A one.-day work stoppage at the Kusnetsny Metallurgical Plant due to clogging of the water intake would cost the plant 10,000,000 rublest! In view of these difficulties ad disturbances in the Russian power plants the State Hydrological Institute in Leningrad was instructed to carry out researches for several years both in the laboratory and on the .m.~lkhov,Svir, Neva, Volga, Dnieper and other rivers, The results of the observations and research appeared in the following Russian publicat ions : "Reports of the State Hydrological Institute", "~eportsof the Institute for Scientific Research in the Field of Hydraulicay "The Volkhov, Svir and Dnisper River Development Projectsn and in many periodicals, The following should be mentioned as more extensive, single works: Prof, Wi ,J, Altberg, "Depth Icew, Leningrad, 1939 ; Ing, F,J , Bydin, "Combatting the Break-up on the Svir R ver", Leningrad, 1934; BOP, Weinberg, "Ice! Moscow, 19402f , Concerning the d.egree of success in the struggle against the ice in Russia the judgmant of Ing, FoJo Bydin, given during his visit to Riga in the spring of 1940 to observe and assess the ice conditions on the Dvina, may be quoted: "Valuable observation and research data concerning the rivers of Russia have been collected, '~Vithout this informat ion new hydro- technical structures could neither be planned nor builto Suitable means of combatting the depth ice, however, have still to be found, If, in this or that instance, the means discovered and employed have proved successful, there is still no justification for regarding these as absolutely sure in all cases, as is proved in a series of concrete, practical applications, The complex depth ice processes must still be investigated carefully and followed up,"

THE DVINA UP TO 1940-.- Before beginnina a descri~tionof the in- e data will first be on the Dvina it- W, J, Altberg: Podvodny Lyod, Leningrad, 1939, F,T, Bydin: Svirstroy (Borba so lyodom) Leningrad, 1934, BOP, Wsinbsrg:Lyod, Moscow, 1940, r) Das Seedepartement, Page - 6 Tech, Trans, TT-85 self together with information about the observations and investigations made in former years,

The Dvina 1s one of the largest rivers on the Baltic coast, It rises in the Central Russian Highland (Valdai ~ighland),flows through the White Russian SoRo and Latvia, draining an area of 82,800 square kilometres, After traversing a course of 1000 kflometres it empties into the Gulf of Riga, For centuries it has served as an important east-west waterway, In the stretch from Jakobstadt to Insel Dahlen - 190 and 25 km, from the mouth respectively - the Dvina is very turbulent and has dug its bed deep into the Devonian dolomite, At many pofnts the folds and bends in the dolomite form great rapids, and this stretch is thus known as the "swift water section" of the Dvina, Over thts stretch, which is approxfnately 160 kilometres long, the mean water level of the Dvina drops 80 metres, foeo, an average of 50 centimetres per running kilometre, In certain sections, e,g,, over the 15 kflometres from Stockmannshof to Kokenhusen, the gradient actually becomes as great as one metre per kilometre, Figure 1 gives the longitudinal profile of the swift water section of the Dvfna, and Figures 2 and 3 show the most important rapids, The rapids of the Dvina may be put to advantageous use in the production of hydro-electric energy, According to the Latvian electriffcation plan, starting with the year 1933, the Dvfna is to be dfvfded into six development stages with drops rangfng from 7025 to 17,O metres, With the realization of this plan the water power of the Dvina would be fully exploited, yielding an output of 300,000 kflowatts semf-annually, At the same tfme the Dvina would become an excellent waterway for boats of three metres draught, One of these stages was completed in 19393 The dam of the Keggum power plant was built in the central part of the swift-water section of the Dvina 65 kflometres above the mouth, creating a concentrated drop of 15,75 metres. The mark of the dammed up water level is 32 metres above sea level, At this level a large dam basin, the so- called "Lake ~eggum*, is formed above the dam, It is 45 25 square kflometres, Kraftwerk Kegums (The Keggum Power Plant of Latvia), Z, Wasserkraft und Wasserwirtschaft, 1938, NO, 11/12, pp, 134-1390 Page - 7 Tech, Trans, TT-85 Since there are a number of settlements in the artificial lake region it was decided to protect them by means of embankments, The largest of these, built to protect the city of Friedrichstadt, is 3,4 kilometres in length and in places attains a height of 5 metres, The height of the embankments was determined from observations of the water level during the unusual flood in the spring of 1931, when the flow of water in the Dvina reached 9000 cubic metres per second, As in the other rivers of Eastern Europe the troublesome ice conditions in the Dvina also create considerable difficulties, In working out the Dvina development project the question of the type and quantity of ice accumu- lating in the Dvina during the winter months,and therefore to be considered during the break-up, arose, The idea was even expressed that the soft masses of depth ice might also lead to dangerous cloggings in tb newly constructed power plants, as suggested by the experiences in Russia, In this connection the Department of Marine made the first measurements of the quantities of ice in the Dvina in the winter of 1931-32, at the same time classifying the different types, This was done over a 250 kflometre stretch of the river, from Daugavpils (Dvinsk) to Insel Dahlen, The measurements led to a surprising result, It was found that in the swift-water section of the river, over the 90 kilometre stretch from Kokenhusen to Insel Dahlen, the total Ice masses came to 43,000,000 cubfc metres, or an average of 478,000 cubic metres for each running kilometre, and of this amount the soft masses of depth ice packed under the ice cover accounted fop approximately 70 per cent, In the remaining section of the river, from Jakobstadt to Daugavpils, only the hard surface ice was found, averaging approximately 100,000 cubic metres per running kilometre, To advance the research into difficult ice conditions and to learn how ice is rated abroad and what defence measures are taken fn other countries in the design and construction of hydraulics works, the Department of Marine fur ished a special report on the ice conditions on the Dvina4P to the Fourth Hydrological Conference of the Baltic States in Leningrad in 19330 Unfortunately the problem was not clarified at the conferenceo When the Dvina development was begun the

lettlaendischen Fluesse, insbesondere der Dha (~augava)in he Ice Conditions on Latvian Rivers, Especially the ~vina),4th Balt, Hydr, Conf,, Lenin- grad, 1933, Page - 8 Tech, Trans, TT-85

ice conditions in the hydrological section of the Inter- national Union of Geodesy and Geophysics Union internationale de GBodBsie et ~Qophysique)in Edinburgh5 f . The Russians also submitted a similar report to this conference on the troublesome ice conditions on their rivers, The problem again remained unclarified, In 1938 at the Sixth Hydrological Conference of the Baltic states in ~nbeckand Berlin particular attention was paid to the problem of ice conditions, Germany, Norway, Sweden and Poland submitted reports on their research results in this field, but unfortunately no conclusions were reached on this occasion, either, Regarding the problem of ice formation and break-up fn the rivers, together with cer- taln recent experiences and points of view in this connection, the following statement was all that ensued: "No formal conclusion was reached on this point of the agenda, However, great value is placed on an international exchange of obser- vat ions and experiences.' The Department of Marine complied with the suggestion implied in this statement and continued to collect data on the ice conditions from year to year, for the experiences clearly showed that the results of such inveati- gations would be very valuable in solving various problems connected with the economics of water resources,

I11 ESTIMATE OF ICE CONDITIONS ON THE DVINA AFTER COMPLETION OF THE KEGGUM POWER PLANT

When the Swedish firm, Aktiebolaget Vattenbyggnadsbyrbn (VBB), began work on the Keggum power plant, its officials carefully examined the observation and research results of the Department of Marine from previous years and made themselves thoroughly acquainted with the ice conditions on the Dvina by on-the-spot investigationso

The most complete estimate .of the ice conditions on the Dvina is furnished by the consultative memo- randa submitted by this Swedish fim (VBB), which the Depart- ment of Marine has obtained from the Power Plant Division of the Baltic Power Supply Company, Ltdo (Energieversorgung Ostland, Go m, b, H,), The author has taken the liberty of

the Rivers of Latvia, Cons, Into do Unions Sc ient ifiques, Bullet in No, 23, 1936, Page - 9 Tech, Trans, TT-85 Memorandum of 12 November 1935 (Re expected ice conditions on the Dvina after completion of the Keggum Power Plant): "The well-known studies published by Obering Po Stakle on the question of how the ice conditions on the Dvina behave in different years and at different seasons are very enlightening, It would probably be impossible to discover dependable principles for assessing the immediate problem on the basis of observations made on the Dvina alone. We value the panoramic picture afforded by these studies all the more highly because it enables us to draw certain conclusions, by comparison with other rivers, concerning the form which the ice conditions may be expected to take in the Dvina following completion of the power plant, It is be- lieved that the Russian rivers, Volkhov and Svir, may provide a suitable basis for this comparison,'' A brief comparison of the two Russian rivers with the Dvina follows, together with a more detailed description of the changes which took place in the ice conditions on the Volkhov and Svir following completion of the Volkhov and Svir power projects, At the end of the memorandum the following conclusion is stated: "It may be foreseen that the building of the dam at Keggum will bring about a considerable improvement in the ice conditions in the lower part of the Dvina. The pro- spects of an improvement downstream from the power instal- lation seem particularly good in view of the fact that the large dam basin, whose extent and volume are comparable to those of the two Russian examples which have been cited, will retain large' quantities of drift ice (~reibeis)and cause them to thaw, It should also be remembered that the presence of adjustable sluice gates will permit a certain regulation of the water level, or of the quantities of water and ice flowing downstream, "The only region where some uncertainty may still be said to exist appears to be the transitional region between the dam basin and the undammed section of the river, However, the anxiety we feel in connection with this region is in contradiction to all the observations on the Volkhov and Svirjl Memorandum of 16 February 1939 (Re expected ice conditions after complstion of the dam basin) : "During current discussions on the project in 1935 it was recommended that two VBB engineers, who were Page - 10 Tech, Trans. TT-85 planning to attend the world conference on power production in America in 1936, should make the problems in question the object of a special study, This plan was actually put into effect, a seriss of problems was formulated and the journeys were undertaken, Unfortunately it turned out that the experiences of America were of little value for the Keggum case, since the ice problems there are altogether. different, "We also tried to make use of experiences elsewhere, e.g,, on the Danube and on certain Russian rivers, One of our engineers made a special trip for this purpose to Donau-Kachler and other places to investigate the ice conditions there, The available data, however, were insufficient to provide a final solution to the problem, especially with regard to the ice conditions up-stream from the dam basin, Thus there remains only one possibility of assessing this problem constructively, namely by utilizing the experiences on the Dvina itself as described by Director P, Stakle, In the following, thersfore, a train of thought has been developed, with a view to discovering ways and means of setting up the desired maximum programme for combatting the ice. "According to Stakle the ice season is divided into three more or less distinct periods, namely:

1, Formation of drift ice (~reibeis), 2, Formation of surface ice.

3. The break-up, he break-up is of decisive importance in determining the methods to be employed in combatting the ice on the dam basin. Considerable piling up of the ice has already occurred on previous occasions under natural conditions at the place in question, ibe,, in the vicinity of Friedrich- stadt. Since during the break-up large quantities of ice must be transported in a short time from the undammed section of the river into the shallow water of the dam basin, and since this must be done at a time when the artificial lake is expected to be covered with a relatively thick layer of ice, it is not improbable that the break-up troubles after the dam has been put into operation will assume at least the same proportions in some years as formerly, unless sufficient room for the in-coming ice masses within the dam basin is provided, Page - 11 Tech, Trans, TT-85 "On the other hand, it should be noted that by suitable employment of the dam basin for the accommodation of the incoming masses of ice the possibflity is gained of reducing, if not altogether eliminating, the break-up difficulties, This is valuable, since the required measures which would be detrimental to the economic operation of the power plant, permit part of the costs to be transferred to agriculture , "The following temporary programme may be set up for decreasing the difficulties which accompany the break-up : 1, Observation of the ice regime above the dam basin throughout the entire winter; 2, Anticipation of the break-up as accurately as possible by means of weather prediction;

3, Breaking up of the ice cover over a sufficiently large section of the upper part of the dam basin soon enough before the break-up; prevention of the formation of a new ice cover;

4, Reduction of the level at the dam to one metre immediately before the break-up;

50 Release of any flood wave which may occur due to the break-up through the dam basin and the weir; accomodation of the ice flow in the dam basin, and, where necessary, its intermittent release through the weir, chiefly by employing the 80 metre sluice gates,

11 According to previous experiance the only way to ensure the creation and maintenance of a sufficiently large opening in t,he ice cover of the dam basin is to break it up with ice-breakers, These have already been employed successfully on the Dvina (Riga Harbour) as well as on other Latvian rivers. It would be still better if the ice-breakers could be combined with the tug boats which are needed in any case for the logging industry* "The above represents a maximum programme for combatting the ice break-up and may be modified, of course, in accordance with experience gained from actually putting it into effect, In favourable years, it may be considerably curtailed, In this connection the most important question is to what extent it is necessary to break up the ice cover Page - 12 Tech, Trans. TT.85 in the most unfavourable case which may be foreseen, This problem may be clarified to some extent by comparing the volwne of the dam basin with that of the stored up masses of ice in the river, The entire dm basin has a volume of from 100 to 150 million cubic metres, of which the upper 20 kilometres at most account for only 20 per cent, On the 90 kilometre stretch from the 35 to 125 kilometre marks, a stored up ice mass was measured in the winter of 1931/32 averaging no less than 500,000 cubic metres per kilometre, More detailed research results on this matter for the stretch above the dam basin are not available, Judging from StaklePs data the amount of stored up ice in thia region is expected to be considerably less than that stated above, but will be large enough to make it appear advisable to proceed with caution during the first year, so that un- fortunate experiences resulting from inadequate means may be avoided, -It will be best to have the maximum extent of the expected break-up assessed and estimated by the Latvian officials , "In most years it should be necessary to break only a relatively small area of the ice cover, Even if, for the present, an ice break-up amounting to 6,000,000 cubic metres per 30 kilometre stretch and 200,000 cubic metres of ice per kilometre be assumed, in all probability the breaking of tho ice cover over a stretch of several kilometres upstream and downstream from Friedrichstadt would be sufficient for trial purposes during the first years of operation, In any case the flooding of Friedrichstadt under such circumstances appears impossible, since the flow opening beneath the ice cover in the river is approximately six metres, "Attention must also be paid to the other winter periods, particularly to that of drift ice formation, The ice forming process above the dam basin should be similar to that in the Dvina above the Riga bridges, It is quite possible that the drift ice may cause ice jams at the upper end of the dam basin after the surface ice cover has formed on the artificial lake, It is not expected, however, that these will be greater than was formerly the case in the natural course of the river. All experiences indicate that conditions.fn this respect will be more favourable after the construction of the dam than before, "The ice breakers mentioned in connection with the programme for combatting the break-up may be used also to mitigate the ice difficulties during the period of Page 13 Tech, Trans, TT-85 formation of drift f ce, by having them cut an opening in the ice cover of the artificial lake for the incoming ice, so that an ice bar cannot form, "With regard to the section of the river below the power plant, f t is quite possible that the formation of drift ice and the local ice jams, owing to &he daily regulations, will assume somewhat larger proportions, On the other hand the fluctuations in the water level may hinder the gathering of ice, so that the difficulties due to piling up, etc., be decreased. Furthermore it can be expected (cf, above that the influx of the break-up from the upper section may be compensated, if not wholly avoided, by means of the artificial lake," In the winter of 1940 the constructional work on the power plant was nearing completion; among other things, however, the sluice gates for the transmission of ice and flood water were not yet ftnished, In this connection the Swedish firm, Svenska Entreprenad Aktiebolaget (Sentab) , which was in charge of construction on the Keggum power plant, had itself worked out a plan for the transmission of flood water and ice through the weir, and this had been recommended and accepted in consultation with the Keggum building inspection authorities and the State Electrical Power Corporation* (VEU) of Keggum, A detailed discussion of this plan would lead too far afield, It need only be mentioned that at the suggestion of the Department of Marine, to ensure that the ice masses would be conducted into the artiricial lake, dynamiting of the ice cover was carried out in the vicinity of Fried- richstadt between 23,5 and 31 kilometres up-stream from the dam, thus opening a channel of single holes 7,5 kilometres long and 50 metres wide, This was intended to replace the breaking of the ice cover by the ice breakers previously mentioned in connection with the maximum program of the VBB firm, This well considered method was successful in the year 1940, although the water level during the influx of the ice masses into the dam basin was considerably higher than expected, so that the upper surface of the newly constructed embankment was only a few decimetres above the surface of the water, The previous dynamiting had not produced the expected alleviations to the extent desired, *Das Staatliche ~lektrizitzt sunternehmen Page 14 Tech. Trans. TT-85

On the basis of the observations and experiences during the break-up of 1940 the VBB supplemented its nenorandum of the 16 February 1339 with the following:

"~ernorandumof the 25 November 1340 (He ice conditions after construction of the dam):

"In our Memorandum of the 16' February 1939 and in subsequent correspondence certain principles were laid down for the transmission of the break-up and spring flood after construction of the dam. These prin~iples~broadlyspeaking, were applied in the spring of 1940 and proved successful. The question now arises to what extent the experiences gained may be utilized for further application, or whether they may indicate any change in the method employed. This subject has been discussed in detail and the following report is submitted:

"The method employed in the spring of 1940 was based primarily on the assumption that the ice cover on the artificial lake should be broken up only at the upper end, but elsewhere should be left intact. The alternative method would be to breakup the ice cover over as much of the lake as possible, starting at the weir at Keggum. Combinations of the two methods are also possible.

k he usefLllness of any method must be judged from various points of view, as follows:

1. Piling up at Friedrichstadt;

2. Strain on the weir structures, particularly the weir defences;

3. Wear and tear on the tumbling bays, etc. ;

4. Ice conditions between Keggum and Riga harbour;

5. Operating costs and costs of ice removal*

'~Vith a view to preventing a piling up at Friedrichstadt it would indeed be advantageous to'break the ice cover over the entire artificial lake, but in our opinion this is more than is needed. It is clear, nevertheless, that the ice cover must be effectively broken at the upper end of the dam basin (Cf. our Memorandum of the 16 February 1339), Page - 15 Tech, Trans, TT-85

''The remaining points, 2 to 5, favow the retention of the ice cover, In this connection, however, point Z0 requires further discussion, If the break-up exerts its last pressure shock during its gradual progress through the darn basin (at Rembate) it is quite possible that this shock might extend, under certain circumstances, as far as the weir defenceso Since thls is undesirable, it appears advisable to free the lake of ice immediately ahead of the weir defences for a distance upstream of several hundred metres, l1

In the following spring, 1941,, the situation became very grave The huge masses of depth ice which had formed in the rapids of the Dvina during the severe winter, and which had already caused ice jams at the upper end of Lake Keggun in autumn, proved a serious obstacle to conducting the ice into the artificial lake during the break- up. The masses of ice could not be accommodated, but had to be conducted very laboriously through the weir in a series of removal operations with a rise in the water level of eight metres, This came considerably,above all embanhents, and flooded many settlements and roads in the vicinity of the lake, Several foreign experts who saw the break-up on the Dvina in the spring or' 1941 were astonished at the unusually difficult ice conditionso The Russian experts maintained that such a vast movement of ice in the eight to eighteen metre deep and five hundred to fifteen hundred metre wide dam basin of the Keggum Power Plant had no precedent either in the experience of the Russian power developments or in the literature,

The general outlook in the spring of 1941 was very gloomyo The damage caused by the water and ice could be estimated at several million Reichsmarks, The worst feature, however, was the fact that no prospect was seen for improving the situation in the future, It seemed as if the outlay for all the newly constructed embankments, roads and bridges, in Friedrichstadt as well as in many other settlements, would have to be written off as a total loss,

IV INVESTIGATIOIJS OF TII3 DEPARTMZNT OF MARINE ON TIE: DVINA IN THE WINTGRS OF 1941-42 and 1942-43 The observations and experiences of previous years had already shown very clearly that the greatest difficulties accompanying the break-up on the Dvina, especially, Page 16 Tech. Trans. TT-85 in its swift-water section, were caused by the soft masses of depth ice. In designing and building the Keggum gower Plant the Swedish firm Aktiebolagetvattenbyggnadsbyron had indeed correctly estimated the quantity of ice in the Dvina, but had not given sufficient consideration to the properties of the different kinds of ice, i.e., they had not appreciated the difficulties that may be caused by the depth ice. The course of the break-up in the spring of 1941 showed that special investigations of the ice conditions, particularly in the swift-water section would have to be begun immediately in order to find ways and means of combatting the formation of the unwanted depth ice, or at least of obtaining a more favourable grouping of the soft depth ice masses so as to prevent the dangerous ice jams in the spring. The author, after acquainting himself more closely with the research results on ice conditions in the rivers of Russia, Norway and Sweden, worked out a program of investigations and submitted it to the De'part- ment of Marine. This plan was studied in#consultation with the management of the Keggum Power Plant and the experts at Higa University, and the work in question was begun in the fall of 1941. The author was placed in charge of the research. The program comprised mainly the investigation of the thermal and dynamic conditions of the formation of depth ice, the observation of the formation of the ice cover and the associated change of water level, the measurement of the ice quantities, the investigation of ground ice formation and observation of the course of the break-up in the,Dvina. For this purpose twenty auxiliary observation posts were set up in the research region on the section of the river between Jakobstadt and Riga, in addition to the thirteen posts already in existence. For clarification of the hydrological and meteorological aspects of the problem three hydro-meteorological stations and six posts for the observation of depth ice were also erected in the same district. The formation of the depth ice was to be limited by trying to get a covering of surface ice over the rapids of the Dvina. To give a better understanding of the idea underlying the methods employed, the foraation of the ice and the ice cover in the swift-water section will be reviewed briefly. Page - 17 Techn Trans. TT-85 (a) Formation of the Ice: The Various Types of Ice and their Froperties In the autumn when the temperature of the water drops to the freezing point a border of ice is generally seen along the banks of the river. Beginning at solid crystallization centres on the shore or on the surface of the water itself, the ice crystals in the supercooled surface layer of the water grow in the form of wide, but at first very thin, crystal line branches (Fig. 4 and 5), until the entire surface of the water is covered with a thin, smooth layer of ice called surface ice (Figo 6)" This increases in thickness due to frost action. In the swift-water section of the Dvina the process of ice formation is much more complicated, With the advent of the f post only a narrow ice border forms owing to the swiftness of the current (Fig. 7); at the same time, however, ice begins to crystallize throughout the entire flow of supercooled water from the surface to the bed. Various peculiar types of ice are formed with properties entirely different from those of the surface ice, The famous Norwegian physicist, Olav Devik, very appropriately calls the process of surface ice formation a static one and that of ice formation in fast waters a dynamic one. ) The complex dynamic process of ice formation has thus far received little attentiono Several theories have been a dvanced. Opinions differ greatly, particularly with regard to the manner in which the ground ice originates, Considerable differences also prevail in the literature on tlie subject with respect to 'the terminology for the various types of ice, The researches carried out by the Department of in recent years show that the dynamic process of ice formation occurs unde.r the following conditions:

1) when the ice-free surface is supercooled;

2) when crystallization centres (eogobodies floating beneath the surface, grains of sand, snow cyrstals, etc,) are present from which crystallization can begin;

6)' Olav Devik: Thermische und dvnmische ~edin~ungender Eisbildung in" Wasser- lae&en (Thermal and Dynamic Conditions of Ice formation in Water Courses), Geophys. Publo Vol. .IX, Oslo, 1930. Page - 18 Techo Trans. TT-85 3) when there are eddies (or waves), which promote favourable crystallization conditions in the supercooled water layers This statement is confirmed by the following experiments. If a stick of wood is placed in the current during cold we at he^ (Fig- 8) icing takes place only on one side of the wood, the downstream side. If two sticks are placed in the stream in such a way that the eddy from the first stick envelopes the second, ice will begin to form only about the second stick (Fig, 9 and lo),, A lump of ice was produced artificially under favourable we at he^ conditions in twenty-four hours at a place where the river was three metres deepo The lump could be shifted about freely as desired, and within four days it had grown so large that a man was able to walk about on it (Fig. 11), The above procedure showed that it was possible to produce a depth ice formation anywhere on the river. This was a considerable advantage for future work. The observations also showed that throwing artif icial crystallization centres (snow or sand) into the open current had the effect of intensifying the depth ice formation. The depth ice is not always of the same type, but can vary very considerably with the degree of supercooling of the water, the rate of crystallization and other conditions, On the basis of the observations and experiences of this investigation the following classification of ice types is recommended: (i)by origin, and (ii)by structure and properties.

(i) In the classification by origin the term "surfa e icef7, Russian "lyod poverkhnostnytT, English "chief ice,It? is-recommended for ice formed by the static process, and the term "depth ice", corresponding o the Russian "podvodny lyod" and the English "current icerft! for that which is brought into existence by the dynamic process, t) Translatorfs note: No precedent could be found for these terms in English and the more literal translations of the German terms have therefore been retained" Page - 19 Tech. Transa TT-85 The depth ice is divisible into two groups, namely, ice which crystallizes spontaneously in the supercooled water in the form of suspended dust, called "schwebeeis" (Znglish - frazil ice), and that which forms on the river bed and adheres to objects, called "Grundeisfl, Russian "donny lyod", English "anchor icgfftt) (Fig. 12). In its turn the ground ice may take two different formsa It may appear as the characteristic oblong, cloudy type, which the local inhabitants call "jumil' (roofs) or as the muddy mass formed in deeper places where it is protected from the currento This might be called "mud iceTt (Schlammeis), At places where the ,ground ice rises above the surface the peculiar round ice formations known as ground ic e islands, Russian l'pyatrill, appearo The fine ashy crystals of the f razil ice combine gradually and form a spongy mass, which slowly rises to the surface, The local inhabitants call these masses "sogatt, corresponding to the term "gelatinou~ice", (Sulzeis ) , Russian "shouga glo~binnaya~~~ The ground ice gradually breaks away from the bed of the river, rises to the surface and floats downstream together with the gelatinous ice. It is easily recog- nizable by its dirty gray colour, which is due to frozen- in particles of the river bed and bits of decayed vegetation. The soft, sticky masses of gelatinous and other depth ice combine more and more (Figo 13), freeze together on the surface, and, under the tnfluence of the current, form characteristic oblong bodies of ice, Now the so-called "vigpu iesanaft (depth ice break-up) beginso In German this is known as "Treibeislt (drift ice), and in Russian as lfosenyiy lyedokhod" (.Figo 14 and 15) The lumps of drift ice become rounded off by the current and then appear like the "platter -ice1', (Tellereis) which is often seen at sea (Fig, 16), structure the types of ice are classified in lows: ice needles, lamelliform ice, block ice, core ice (Kerneis), granular ice, slushy ice (Bleieis), jelly ice (Gallerteis), snowy ice, frothy ice, etc., depending on the form, composition, adhesiveness, and porosity of the ice crystals, tt) Translatorts note: also "ground ice1', Page ,.20

Tech. ' Trans. TT-85

The various kinds of ice have different physical and mechanical propertie s.

The structure of the ice depends on the crystallization process. It is well known that ice crystallizes in the hexagonal system. The axes of crystallization are orthogonal and the growth of the crystals is most ral~id in the direction of the main axes. Thus in slow-flovting, supercooled water, long, thin ice needles form, chiefly on the surface. The slower the current, which carries away the heat of crystallization, the thinner and.longer are the crystal branches, and the surface of the water rapidly becomes covered with a thin, crystalline layer of ice which increases gradually in thickness due to frost action. In this way the transparent, blue core ice is formed (Fig. 17).

The eddies in the water are more or less respon- sible for the formation of the lamelliform ice. In swift water the form of the ice crystals may vary greatly, depending on the degree of supercooling as well as on the rate at which the heat of crystallization is conducted away. In addition the character of the river bottom also exerts an influence, depending on whether it is of earth, sand, gravel or large stones. Initially the ice crystals are extrenely fine. In cold weather they cause the water at turbulent places to appear very cloudy. The myriads of crystals, growing with lightning rapidity, quickly con~bine, and the water is soon charged with a spongy mass (Fig" 18)" It has been observed that the coagulations of the different lamelliform crystals then become trans- formed into lumps of a more or less granular texture. .. . The spongy, granular mass of ice is saturated with water but is imj~er~meableto it. This is a very important point. The'water retained in the ice has a different density from that of the surrounding masses of waterj it has become somewhat more dense because of the crystallization of the ice, and hence, as observations show, it may be borne along at various depths. It may rise slowly to the surface, or be carried to the bottom by the vortex current, where it will stick, particularly if ground ice formations are already present. The water retained in the soft depth ice cools slowly and freezes with difficulty. Even after long periods of intense frost the gelatinous ice masses remain permeated with water. Under the influence of the frost the soft, spongy Page '- 21 Techo Trans. TT-85

,mssas of ice gradually become firmer; it is difficult to pierce such ice with a wooden pole, Fully hardened layers of depth ice may be found in many places along the Dvinao In the "PriedulZjsn rapids (152 km, above the mouth), for example, the river bed in severe winters becomes choked with depth ice to a height of several metreso If a pole is driven into this mass of ice the shaft of the pole remians quite dryo Thus, the depth ice, although porous, even in loose form does not allow any water to pass through it, This is why depth ice causes catastropheso When the soft masses of depth ice are formed on t river bed or are driven under the solid ice cover (Tost 7)a wall is created which the current is no longer able to overcome, and this can become the chief cause of ice jams during the break-upo As already mentioned, the soft, spongy depth ice gradually hardens and forms a solid, dull, porous ice cover with a characteristically granular structure (Fig, 19), It differs entirely from the crystalline core ice in its physical and mechanical propertieso Observations show that the solid, dull, granular ice is much tougher and more resistant to external influences than the blue, transparent core ice, In spring after the break-up long lumps of solid, granular depth ice are still to be seen along the banks of the river long after the core ice has vanished (Figo 20), The measurements recently carried out by Construction Engineer No Leepinsch on the mechanical strength of ice and the investigations made in the Kaggum Power Plant afford a very interesting picture of the strength of various types of ice, and prove how unpleasant the granular ice can be for hydraulics cons truc tione Obviously no hard and fast line may be drawn between the static and the dynamic process of ice forma- tiono In most rivers they are complementaryp The ground ice brings about a step-formation of the river profile by dividing the natural gradient up into several small, concentrated falls, The static process predominates at places where the water level has been raised by ground ice structures of this type or by ice jams. At turbulent places the dynamic process plays the main part. In quieter stretches of the river the ice cover forms under inter- mediate conditions, Slow currents, which are disturbed

t) Translator f s note : The ordinary translation of this word, ntuft", might convey a false impressiono The author uses it to describe any formation of depth ice clinging to the bottom of a solid ice cover, The German term has been retained subsequently in the translationo Page 22 Tech. Trans. TT-85 by wind just as they are about to freeze, are, in fact, strong enough to prevent the top layer of water from freezing into a closed skin of ice, but cannot stop the formation of a solid ice cover. In such cases the slushy ice (Breieis) usually occurs in the top layer and the continuous ice cover does not begin to form until the masses of slushy ice span the entire width of the free current. The surface of the water is then covered with a rough, uneven layer of ice which grows downwards in thickness under the influence of the frost. The ice cover forms in this manner on the Dvina above Jakobstadt. A completely smooth, crystalline ice cover occurs only in rare instances.

(b) The Freezing of the Swift-Water Section of the Dvina

At the beginning of the winter, when under normal conditions the water temperature usually falls to the freezing point, slow places on the Dvina above Jakobstadt gradually become covered with a relatively smooth covering of ice, while downstream in the swift water sectfon a more intense dynamic process of ice formation sets in. With steady frost the depth ice forms in ever greater quantities until it fills up the entire width of the river surface with masses which are so dense that they cannot move (Fig* 21). This might be designated by the German term "Eisstand", Russian "lyedostavl' (ice stand). The drift ice may be halted by a stretch of the river which is covered with surface ice, a narrowing of the river bed, a sharp curve, an island or any other obstacle. Usually the halting of the drift ice in the swift water section of the Dvina begins at Riga above the iron bridges, where, under the influence of the frost, drift ice which has come down into the broad, deep river channel soon freezes and forms a solid ice cover. Since the building of the power plant at Keggum a similar coherent ice cover also forms in the artificial lake of the power plant, and the drift ice in the swift water section is thus halted simultaneously at two places, namely the Riga iron bridges, and Friedrichstadt (95 km. upstream from the mouth). New masses of drift ice are continual1 coming down the river and striking against the "ice standi: and thus the "ice stand head" (~isstandspitze) as the natives call it (Fig, 22) moves upstream at an average rate, depending on the gradient and the frost, of 3 km. every 24 hours. The halted drift ice freezes and forms a solid, rough ice cover. In more turbulent places the drift ice cover may even slide downstream several times before freezing and pack itself into high ice mounds Page 23 Tech. Trans, TT-85 the so-called "~orossen"(Fig. 25) . In normal winters the head of the ice stand reaches the first of the larger rapids on the Dvina, the Stutschka Canyon (135 km, from the mouth) and stops there. It moves several kilometres further upstream only during very severe winters, when the depth ice packs itself into such large masses that the water level is raised more than two metres, thus covering the rapids. The rapids above the Stutschka Canyon - Kahklis, Olingrahze, Bebruleia, stock-nannh~fsgrube Priedula js - never freeze over. There the depth ice forrns in immense masses during the whole winter, Here and there the rocky bed of the river becomes covered witn characteristic circular, pulvinate ground ice structures which even in a single night may attain a thickness of one to two metres, Gradually the ground ice detaches itself from the bed and floats downstream, together wlth the soft masses of gelatinous ice. Thus, the whole winter long this approximately twenty kilometre stretch of the swift water section of the river produces masses of depth ice and the current carries them downstreamo The drift ice masses are driven under the solid ice cover, filling the bed of the river with Tost at times up to 80 per cent of the cross-section, and creating high mounds of Toro~sen~The relationship between the water level and the discharge quantities is completely destroyed, In the winter months this circumstance creates serious difficulties for the calculation of the discharge,

Long sections of the unfrozen rapids of the Dvina are covered with ground ice islands which are several metres thick and which remain at the same place throughout the wicter (Fig. 24)0 In the open stretches of the rapfds the bed of the river becomes covered with a thick layer of ground ice, and the water flows continually over the ice base (Figo 25). With the increase in the masses of ground ice the water level rises higher and higher and the water passes over the boundary ice, There it freezes and forms a long wall of ice (Fig" 261, At places where the ground ice remains on the river bed for a relatively long time it becomes hard and even encloses stones. Then, when it breaks away from the river bed the stones are carried along with ito Stones of this sort, frozen into the ice, are often seen in the swift water section of the Dvina, Some of them may weigh tens of kilograms. The ground ice, indeed, often brings about considerable changes in the rapids by raising loose stones from the bed and Page - 24 Tech. Trans. TT-85 transporting them elsewhere; for example, as shown in the cross-section (Fig. 25), the ground ice in the Priedulajs rapids has dug a channel in the bed of the river, completely clearing it of stones. On the stretch above Jakobstadt, where the currsnt is slower, an uninterrupted ice cover is formed mainly by the static process. At the beginning of winter ground ice may also be found there in places. It develops poorly, due to the relatively slight supercooling of the water surface or because of weak turbulence of the current, The ground ice structures are very thin and are found only here and there on the larger protrusions of the river bed or on the dark sub-base. (c) The Quantities of Ice in the Dvina and Their Distribution In order to determine the extent, the variations and the distribution of the ice masses at various places in the river the Departmsnt of Marine has for several years been conducting systematic measurements of the quantities of ice, The measurements are carried out in the following mannero On the frozen-over stretuhes of the river cross- sections of the ice cover are taken at intervals of approximately one kilometre, by cutting holes in the ice twenty metres apart (Fig. 27). Measurements are then taken of the thickness of the surface ice, the thickness of the tost and the packed ice (Torossen), and of the layer of snow on top of the ice cover, At places where the quantities of packed ice are great, or where the ice conditions change suddenly, the cross-sections are chosen closer together. To determine more accurately the quantity of tost driven under the ice cover, supplementary holes are cut in the centre of the river every 200 metres between the cross-seutions. The measuring places are marked on the plan of the Dvina in the scale of 1:5000. An example of such a calculation of the ice quantity is shown in Figure 28. As shown by the figure, the tost gathers, chiefly below the rapids in the depression of the river bed, in such large quantities that in places it fills up the greater part of the river cross-section. Page 25 Tech* Trans. TT-85

Table 1 gives a sunmary of the res~ltsof the ice quantity measurements obtained in the year 1932, before construction of the power plant at Keggum, and in the last three winters, following its completion.

It may be seen from these results that the quantity of ice which collects in the swift water section of the Dvina is from two to four times that found in the quieter places of the upper reacheso Most of this consists of the soft, sticky masses of depth ice driven under the ice cover.

Comparing the measuring results for individual sections of the river before and after construction of the power plant at Keggum, the favourable effect of the latter on the ice conditions downstream may be clearly seen. The quantities of ice on the stretch from Keggum to Insel Dahlen were considerably smaller in the last three winters than in the winter of 1932. The depth ice formerly packed beneath the ice cover has completely vanished,

(d) Ground Ice

At turbulent stretches of the river which do not freeze over the surface of the water in winter is in constant contact with the cold air and is thus subject to continuous coollng (Fig. 30), As observations show, the loss of heat from the surface of the water is mainly due to radiation, evaporation and conduction. In lower layers it is due to mixing of the water as a result of turbulence, Heat conduction during this process is important only for the further temperature equalization of the masses of watero

After the water has been cooled to 0' C the crystallization of the depth ice begins. If a sudden supercooling of the surface of the water occurs immense quantities of depth ice form at places where there is an abundance of crystallization centres and where the removal of the heat of crystallization is assured in the entire cross-section of the river, In such cases the water in turbulent places appears completely clouded by small, ash-like crystals of ice, Where the current is slower some of these fine crystals begin to stick together and form the gelatinous ice which has already been described. Others fasten onto objects on the bottom of the river or onto the hard ground ice lying there, In shallower places, Page 26 Tech. Trans. TT-85

where +he supercooled masses of water are borne to the bottom by the current vortices, various structures of solid ground ice are formed on the irregularities of the bottom in the places which are more or less protected from the current.

It is difficult to determine how far advanced the supercooling of the surface of the water must be before the depth ice begins to crystallize. Measurement with the ordinary nercury thermometer is impossible since in the first place the supercooling occurs only in a very thin top layer of the water, and secondly, when the thermometer is immersed in the supercooled water it quickly becomes coated with ice and the determination of the temperature difference is thereby disturbed. The Russian experts are of the opinion that normally only a few thou- sands of a degree are needed, less frequently as much as a hundredth, and quite rarely up to a tenth.

For measuring the quantities of ground ice in the Dvina sectional sieves (segmentartige ~iebe)25 cm. in diameter with a surface of 1330, sq. cm,, were employed (Fig, 31), Under conditions favourable to the formation of ice 90 cu. cm. of ground ice per sq. cm. were measured on such a sieve for a twelve hour period (Fig, 32). When measurements were carried out in various river profiles after a period of intense ice formation it was found that ground ice layers up to a thickness of two metres had formed in a period of 24 hours.

Ground ice formations were noted in considerable quantities everywhere at the open places in the swift water section of the Dvina, The ground ice grows in particularly large masses at the lower ends of the gradients, at the curves in the river and directly below them. It is less prevalent at the beginnings of rapids and is only rarely found in those stretches where a moderate, uniform current prevails,

As the observations show, the character of the bottom (stones, rocks, sand, loam, mud, etc,) does not play a large part in the production of the ground ice. With respect to the quantities of ground ice retained, however, the different types of bottom have varied effects. The more turbulent the vortex currents which carry the supercooled water layers to the bottom, and the rougher the bed of the river, the greater the ground ice masses will be, Immense masses of ground ice may be Page 27 Tech. Transo TT-85 deposited on large stones, and these later grcw into ground ice islands which are several metres thick and rise above the surface of the water (Fig- 33)" These ground ice islands have a characteristic profile, Lighter objects are surrounded by the ground ice, raised and carried away.

The approximate masses of ground ice which might collect in a section of the river at various times and under various conditions have been calculated by means of the observation sieves. Thp: resuJts of the measurement are given in Table 20 I.

As the table indicates, the largest masses of ground ice form in autumn when the open water surfaces are greatest and the loss of heat from the surface is at a maximum, During the winter the open places become covered to an increasing extent with border ice and are also continually being reduced by the ground ice struc- tupes. On the other hand, in the second half of the winter, after the solar radiation has increased, the loss of heat at the surface of the water becomes less and this has an appreciable effect on the formation of the ground ice.

The average arnount of ground ice in the swift water section of the Dvina may be estimated at 0,060 cu m,, in autumn, to 0,025 cu, m, in spring, for each square metre of free water surface per twenty-four hour period.

It should be noted that the depth ice, or ground ice, produced in the early autumn season has very great adhesive power and equally great toughness. During the second half of the winter, and in the spring, the depth ice becomes less and less adhesive8 hence the depth ice which is produced in the spring is not nearly so dangerous from the point of view of ice jams as that which is formed in autumn,

As already mentioned above, the ground ice frees itself from the bottom by degreeso The conditions necessary for this separation in general depend on a whole complex of meteorological and hydrological factors, Observations show that the two processes of formation and separation may even occur simultaneously. The ground ice clings to the bottom as long as the supercooled water masses continue to flow down to it, and the effect of the ground heat is neutralized, provided the object to which Page 28 Tech. Trans. .TT-85

it c1ip.g~is capable of holding it. As soon as a stretch of the river becomes subject to the effect of the obstruction, then, due to the action of the ground heat or the underground water, the ground ice frees itself and rises. Another important factor involved in the freeing of the ground ice is found in the weather conditions, especially solar radiation, increased air temperature, etc. (Fig. 34).

Altogether the process of depth ice formation is very complex and many scientists have investigated it during the past several centuries. The theories which have been advanced are many and various, but none of them can claim to be complete. Opinions differ particularly with respect to the manner in which the ground ice originates. In this field careful observations and researches are still needed, and physicists and hyd~ologicalengineers should work closely together for this purpose.

(e) The Reduction of the Drift Ice Driven under the Ice Cover (Tost)

The quantities of ice driven under the ice cover also vary greatly in rivers, depending on the winter season generally and on the hydrological conditions in different winters and at different times, with more or less appreciable effect on the discharge. Particularly severe ice barriers occur i$ the swift water section of the Dvina and cornpletely upset the normal relationship between the water levels and the discharge quantities.

The extent and variations of the ice barriers are difficult to determine. For this purpose attempts were made in the winters of 1941-4.2 and 1942-43 to determine the quantities of ice driven under the ice cover and the variations in these quantities. Direct measurement was employed, as well as the procedure which Prof. Kolupaila has recommended, namely, observation of the fluctuations in the water level and discharge quantities. The diagrams of Figure 35 show the water levels and discharge quantities for these two winters. -The abnormal increase in the water level in winter, due to the ice barriers, is characterized by the coefficient. Q where Q is the measured flow quantity of the water and k=~ Go the corresponding discharge quantity in the ice-free river. As seen from the diagram the indreases in the water level, H, are greatest and most variable in autumn. Depending on the amount of oncoming drift ice and on the current, the drift ice masses are also regrouped beneath the ice cover, If the river freezes over entirely the masses of drift ice become stabilized and begin to reduce Page 29 Tech. ;ranso TT-85

in quantity insofar as they are melted by the r3latively warmer ground water. In spring, as soon as the melting process begins, a greater amount of water flows through the ice barriers. This is difficult to observe by direct measurement, but measurements of the discharge give a good indication of it, for the water level always remains unchanged for a certain time before an appreciable change in the discharge occurs; indeed, it may even fall. The same holds true for the beginning of the thaw. To determine the variation in the ice barriers direct measurements were also made at various times on individual profiles. The results are set down in Table 3. The groupings and variations of the drift ice on the Stutschka Canyon section in the winter of 1941-42 are represented in the drawings of Figure 36.

- The quantities of ice driven under the ice cover and their variations at various times and places from Riga to Daugavpils (~vinsk)are shown in Table 3.

V THE MEANS EMPLOYED FOR REDUCING THE DEPTH ICE The formation of the ice cover in the swift water section of the Dvina has been closely observed since 1928. These observations led to the idea of creating ice heads, such as those which normally form at the beginning of winter in the Riga Harbour district and the Keggum artificial lake above F'riedrichstadt, at the lower end of the larger rapids which did not freeze over. The purpose behind this idea was to halt the drift ice masses coming down from the rapids, to employ tRem to dam up the water, and thus partially or completely to cover the rapids themselves with the obstruction, and promote the formation of an ice cover in the stretches of the river which were influenced by the obsteuction. In this manner the further formation of the depth ice in the rapids would be reduced considerably, or might possibly be eliminated altogether, even in autumn, due to the artificially produced freezing. For this purpose the following means were employed in the winters of 1941-42 and 1942-43: 1) border-ice bridges; 2) drift ice dams;3) artificial production of depth ice.

a) Border-ice Bridges

A border-ice bridge consists of a section sawed out of the border ice and placed diagonally across the river (Fig. 37 and 38). The masses of drift;.ice coming Page - 30 Tech. Trans. TT-85 down from the upper reaches of the river are halted by this artificial barrier and freeze together under severe frost conditions. Since the process of depth ice formation is normally most intense during autumn the ice bridges were begun immediately after the frost set ino The locations chosen for these operations were points immediately below the rapids where the flow of the river is quiet and the bed is deep, At these points the required building material, the border ice, is at its widest and firmest. It is also desirable to have a bend in the river at the location of the bridge, since the possibility of obtaining a favourable transmission of the pressure of the current from the bridge to the shore is thus provided. The method of constructing an ice bridge is shown in Figure 39. The strip of border ice should be chosen at the place where the flow of open water is narrowest and on the side from which it is easiest to get the strip into place across the river with the aid of the current. Especially important is the forming of the foot of the ice bridge. The angle of inclination of the foot, a , should be chosen such that the bridge may not slide off. Friction in the direction of midstream must not be allowed, and it is also undesirable in the shoreward direction. The length, L, of the bridge should be chosen between the limits 1 and 1.4 b, where b is the span of the open water in the river, Normally L will be equal to 1.2 b for the slope of the supports tana = 2.5. In practice the ice bridge strip may be as long as 250 metres, i.e., the space to be bridged may be as much as 200 metres wide, Longer sections than this will not withstand the bending stress and will break. The width chosen will depend on the thickness and strength of the border ice, It is undesirable to have the strip too wide, Often it proved necessary to strengthen the foot of the bridge or the ice strip itself, This can be done by freezing wires into the ice, a process which results in something akin to reinforced concrete, As already mentioned, the ice bridges should be erected immediately after the first frost and after the formation of the first depth ice, so that the soft, sticky ice cannot collect in places where it is not desired. Page 31 Lech. Trans. TT-85

During early autumn, however, the border ice is still too narrow and weak. Before the bridge is erected, therefore, it should be artificially expanded and strengthened. The method by which the border ice is artificially expanded is shown in Figure 40,

By means of such bridges one of the largest rapids of the Dvina, the so-called Priedulajs Rapids at Stockmannshof, between 151 and 154 km. from the mouth, was covered with surface ice in the autumn of 1941. The results were good. In a few days the bed of the river beneath the. solidly frozen top layer at the bridge was completely clogged with depth ice to a depth of 10 metres. When the quantity of ice was measured on the 15 December it was found that approximately 150,000 cu. m, of sticky depth ice had already collected in the 200 metre stretch of the river. The water of the Dvina continued to flow only through single, hollowed-,out channels near the shore. The water level had risen two metres and the entire lower end of the Priedulajs Rapids was already under the influence of the dam, Near the dam the border ice had expanded rapidly and the necessary conditions for freezing over became more and more favourable, until practically the whole rapids were covered with a solid ice cover by the 16 January (Fig. 41).

None of the local inhabitants ever before remembers having seen the Priedulajs Rapids completely frozen over. This year, for the first time, that has been accomplished with the aid of the ice bridges.

Later, similar ice bridges were constructed down- strearn on stretches of the river which were not frozen over, and by February the entire swift water section had a solid ice cover. 8

From the sketches (Supplement 1) t, which were made at regular intervals of five to ten days, the course of the gradual freezing over of the Dvina rapids, accomplished with the aid of the ice bridges, may be seen.

- - t) Translator 1s note : Not received with document, Page 32 Tech. Trans. TT-85 b) D~iftIce Dams

In the winter of 1941-42 it did not prove possible, by means of border-ice bridges, to limit the formation of depth ice on the second largest rapid of the Dvina, the Stutschka Canyon, nor on the one above it, the three kilometer long Kahklis Rapids (from 135 to 138 krn.'above the mouth). On this stretch of the river the border ice, because of the swift current, was too thin, and its artificial expansibn was prevented by the steepness of the banks. Another method of bringing the masses of drift ice. to a halt was employed for the first time in the autumn of 1942, the so-called method of drift-ice damming. + This method is based on the following scheme (Fig. 42): Use is made of the fact that the main current o'f the river, bearing the larger masses of drift ice, proceeds in zig-zag fashion down the bed of the river, now approaching one bank, now the other. A rake, fifty metres in length, constructed of poles, is placed diagonally in the current at a place where it approaches the shore and is anchored to the shore by means of wires (Fig. 43). The rake retains part of the drift ice, which due to the effect of the frost soon freezes and produces an arc-shaped border of ice. The pressure of the current is soon transferred from the rake to the shore, and when this occurs the rake has served its purpose. The arc- shaped border of ice, in its turn, deflects the cur~ent sharply towards the opposite shore, When similar rakes of poles are introduced into the river from the oppds'ite shore it is found possible, under favourable conditions, to create a funnel shaped narrows at a place approximately 280 m. wide, at which the masses of drift ice soon pile up and come to a halt. In this manner an artificial obstruction of depth ice is formed which serves the same purpose as the border-ice bridges described above.

The rakes were constructed of several poles of flr and pine, each 10 to 15 cm, thick. A two millimetre iron cable was used to anchor them to the shore.

Using this method of obstruction all drift ice- masses coming from the rapids were halted within four days at the Purnin Bend (135 km, above the mouth) in the early autumn under moderate frost conditions down to Page 33 Tech. Trans. TT-85

0 -10 C. This created a dam of ice which exte lded for approximately 1.5 km. The changes in the river gradient at various times during the winter of 1942-43 are represented in Figure 44. As the sketch shows, within approximately one month one of the largest rapids of the Dvina was covered with ice, step by step, following fre- quent changes in the position of the ice masses with increases in the water level of from 1,O to 1,6 metres. This is a section of the river which otherwise never freezes over, even during the severest winters.

Considering an ice darn of this kind in the river, the following forces may be noted: on the one hand the resisting forces of the masses of ice, which depend on the toughness of the depth or drift ice, the effect of the frost, the quantity of the ice masses, and the effect of the river bed at the place of the obstruction; on the other hand the destroying force, which is a function of the river gradient and the discharge, Q. As soon as the ice is halted at any point the resisting force begins to increase. This in turn immediately results in an increase in the destroying force at the expense of the gradient, i,e,, the water level above the obstruction begins to rise. Thus the water level above the obstruction serves as a manometer which determines the equilibrium between these two forces. If the water level rises so high that its hydrostatic force is able to lift the ice masses, then the destroying force takes the upper hand and the masses of ice are compelled to change position, i.e., the gradient of the river over a certain section changes.

As the observations show, enormous masses of depth ice can be retained by artificial ice obstructions of this type, On the 17 January, 1943, 700,000 cubic metres of ice were measured over a 2.6 kilometre seption of the river. These masses, in places, piled up in layers several metres thick, which extended right to the bottom on either side of the river, especially beneath the artificially produced border ice. These are the so- called ice dam shoulders. On the section of the river below the ice darn a relatively smooth cover of surface ice was formed. or' purpopes of comparison the ice cover above Kokenhusen is shown in the winters of 1941-42 and 1942-43 respectively in Figure 45.

To reduce the masses of depth ice in the Dvina rapids further upstream the border-ice bridges were also employed in the winter of 1942-43. Since the Dvina froze Page 34 Teeh. Trans. TT-85

in the sutumn of 1942 at a much lower water level and rate of discharge than in the previous year, the construction of ice bridges presented no special difficulties, Immediately after the first frost, in the middle of November, the border ice was so broad and strong that no special pr.o'duction of border ice as in the previous year was required. The sawing out of strips of border ice, and.%%- placing of these diagonally across the river, was nearly always successful. Five ice bridges were constructed from the 12 November to the 9 December, and by the second half of January practically the entire swift water section of the Dvina was covered with ice.

c) Artificial Production of Depth Ice

In addition to these border-ice bridges and drift-ice dams various methods of artificial depth-ice production were also used during the winter as a means of reducing the depth ice or forming an ice cover on the rapids.

As already mentioned, from the research into the thermal and dynamic conditions underlying the formation of depth ice ways and means were found of artificially promoting the formation of depth ice at any desired place in the rapids. This method was now used in order to promote the freezing over of narrow but very turbulent places. Experiments of this type were carried out at the Stockrnannshof Canyon (150 km. above the mouth). Almost the entire boiling canyon rapid was covered with formations of depth ice, leaving only narrow cracks through which the water bubbled down. Light wooden poles were thrust at close intervals into the five metre per second current so that they reached the bottom, which was already covered with ground ice. Because of the artificially produced vortices the wooden poles were soon covered with soft, sticky depth ice down to a depth of approximately 50'cm. Under favourable conditions the test surface froze over artificially within a few days. The current flowed over the lower part of the ice cover. The work is shown in the photographs, Figures 46 and 47. Another method ww employed in places which are similar, but where the bed of the river is deeper. A depth ice floe was produced artificially on an open stretch of the river above the place where the stream narrows (Figure 48), and was brought down into this narrow place. Page 35, Tech. Trans. TT-85

In this manner it is comparatively easy to produce a covering of ice over places which normally hardly ever freeze over* Experiments of this nature were carried out in the winter of 1941-42 in the Priedulajs Rapids district with the result that the formation of the ice cover on the unfrozen stretches of the river was considerably enhanced.

It is difficult to lay down general principles regarding the method which may be most advantageously employed at any given place, The choice 01' the method must be determined by the conditions at the place in question. It is necessary first to investigate thoroughly the character of the river bed and the current, to determine the weather conditions and the character of the border ice, to discover the location of underwater springs, etc , , before any conclusion may be reached regarding the employment of this or that method. Very advantageous combinations of the different methods are also possible.

Corresponding variants of each of the methods described may also be adapted to the given circumstances. Border-ice bridges, for example, may be constructed by letting the border ice into the river from both sideso This is an advantage if the ice bridges are to be erected at a relatively straight section of the river. With these ice bridges the pressure of the current is distributed over two foot-supports. The border ice strips themselves in this case are much shorter and therefore much more able to resist damage.

The drift ice dams have considerable advantages over the border-ice bridges. In the first place it is possible to put the wooden rakes in the river during the first frosty days so that the formation of depth'ice is interrupted from the very beginning, whereas for the ice bridges the border ice must first be produced artificially. Secondly, the drift ice dams are more advantageous with respect to position. They may be erected under much less favourable conditions than the border-ice bridges, since even the rapids themselves are not ruled out, To be sure the work must be approached with great understanding, Considerable knowledge is especially required in order to discover the possibility of transferring the pressure of the current from the rake to the shore, or for the formation of the ice dam shoulderso Page 36 Tech. Trans" TT-85

VI PO SSIBI' ITIES OF OBTAINING MORE FAVOUHABLE GROUPING OF THE DEPTH ICE IN THE SWIFT WATER SECTION OF THE DVINA WITH THE AID OF ICE BRIDGES

As already mentioned, since the completion of the power plant at Keggum the main masses of depth ice coming from the rapids of the Dvina usually gather at the upper end of the artificial lake and further upstream on the section from Rikas to Aiselkschni (128 km, from the mouth). Packed together in dense masses beneath the ice cover, and here and there stopping up as much as eighty per cent of the river bed cross-section, these masses, because of their toughness and adhesive qualities, form an actual wall which the river is no longer able to overcome. Every year, during the break- up these masses create serious difficulties. It is known from observation and experience that the larger and softer the masses of depth ice are, i,e., the more water-tight, the more slowly they will disintegrate during the break-up, and the greater the flood will be. Other factors, chiefly the quantities of water and the morphological elements, play a secondary part by setting the ice in motion at an earlier or later date and bringing the masses of ice to a halt at this or that place. Usually at the very beginning of the break-up an ice jam forms at the neck of the Keggum artificial lake and advances with a wave of water which may be as high as 10 metres, and drives into the solid ice cover (Fig, 49), The ice jam grows constantly in size as additional masses of ice come down from the upper waters of the Dvina, and the danger of flooding increases.

The procedure recommended by the Swedish firm which planned the Keggum power plant, whereby the ice was to be brought into the lake by artificial "pumping1', i,e., by raising and lowering the water level at the dam, was a complete failure. The soft, water-tight ice masses make it impossible to influence the water level above the ice jam by this means. It appears, indeed, that the breaking of the ice cover over the entire lake would not alleviate the situation, for the following was learned fr6m last year's experience: only when the immense "stoppern of depth ice has either been broken up by the action of the water, or has been moved over the weirs of the power plant, as was done in the spring of 1941, will the water level in the lake become equalized, leaving, as was the case in the spring of 1942, an unbroken ice cover approximately 200 metres wide. Page 27 Tech, Trans. TT-85 To improve the situation in the sprin~during the break-up period, means were sought of obtaining a more favourable grouping of the depth ice in the swift-water section of the Dvina. This was accomplished by means of the ice bridges already described. The measurements of ice quantities and the observations of the course of the break-up in the spring of 1943 demonstrate the success of this methodo When the measurements of ice quantities were made in the swift-water section of the Dvina it was found, sur- prisingly, that the masses of depth ice had grouped themselves principally at the five artif icially erected ice bridges. The locations of the ice bridges, or dams, are shown schematically in the sketch diagram, Figure 50, together with the quantities of ice, their distribution and the groupings of the masses of ice. The measurements carried out during the previous winter in the Stockmannshof district showed a similar arrangement* Thus the first means of obtaining a favourable grouping of the ice masses was found. It remained merely to discover from observation to what extent the new grouping of the depth ice masses would exert a favourable influence on the break-up in the spring. In this connection the author will briefly review the course of the 1943 spring break-up by which the strength of the ice bridges and the nature of the depth ice is characterized. Following the exceptionally severe winter of 1941-42, that of 1942-43 was very mild. Over the entire winter periods of moderate frost alternated with sudden thaws. In mid-December, directly after the erection of the ice bridges, a steady period of thaw with rain set in, leading to a general ice break-up on the Windau, and causing an increased flow of water into the Dvina of approximately 230 cu. me per sec. Considerable anxiety was felt for the safety of the ice bridges. However, this fear proved unfounded. The ice drifted down only over a 20 kilometre section of the river in the neighborhood of Kokenhusen below the ice bridges and jammed in the sa- callad Riku lihkums (Rika Bend) in the upper part of Lake Keggum (110 kmo above the mouth). The rapids section, ioee, the ice cover which had been created due to the influence of the ice bridges, was not affected. Figure 51 shows one of the ice bridges at Stockmannshof after the thaw.

At the beginning of January, when severe frost had again set in, the ice cover in the swift water section of the Dvina again expanded rapidly due to the influence of the ice bridges, and by the end of the month practically all the Dvina rapids were frozen overo Page 38 Tech. T,rans. TT-85 At the beginning of February a period of unusually warm weather began and lasted for some time, Under the influence of thaw and rain the ice in the lilindau again broke for the second time and the break-up also began on the Kurl Aa. No peat changes occurred in the ice cover on the Dvina Rapids. It was not until the twenty-third of February, when the discharge of the Dvina reached 285 cu. m, per sec., i ,e., twice what it was at the tine of the freeze-up in autumn, that the first local movement of ice took place on the turbulent Stocbannshof Canyon Rapids and several hundred metres downstream, The next day the ice crumbled in the Stutschka- Canyon and on other rapids, The following picture presented itself on the twenty-eighth. The masses of ice began moving away from the artificial dam at Purnin Bend and piled up as usual at the upper end of the Keggum Lake. The masses of ice below Stockmannshof, from the 17 km. stretch of rapids, were borne over the Stutschka-Canyon and jammed at the Purnin Bend artificial ice dam. The ice came to a halt on a stretch of the river approximately 3 km. long, piling up into bars which were several metres high in places, and the water level at the upper end rose 4.83 metres above the normal level (Fig. 52). At the same time the masses of ice on the Priedulajs Rapids jammed against the artificial ice bridge at Stockmannshof. At both places the ice jams held for forty-four hours and did not begin to move until the night of the 2 March. Meanwhile the discharge of the Dvina had increased to 500 cu. metres per second. The pieces of hard, crystalline surface ice averaged 40 cm. The important point is, however, that only a part of the dammed up masses of ice moved, the ice of the artificial bridge at Stockmannshof and of the obstruction at Purnin Bend opening a crack in the centre of the river only 50 m. across" The rest of the ice remained in the places where it had been packed together in high bars in shallows and on the shore, and disintegrated gradually on the spot. Figure 53 shows the situation after the break- up at the Purnin Bend. As a rough estimate, only about half the masses of ice from the swift water section of the Dvina reached Lake Keggum and even these were considerably ground up by dynamic forceb during their progress through the rapids. The course of the break-up is schematically represented in the sketch, Figure 54,

With these results the success of the ice bridges has exceeded all expectations, Page 39 TechD Trans, TT-85

The observations made on the morning on 4 March at the Rika station at the upper end of Lake Keggum were very significant, The air temperature had dropped during the previous night to -loOc and the process of depth ice formation had begun anew in the Dvina rapids, Within a few hours the water level at Rika Station had risen 95 cm, It was thus shown that the ice formation of a single night has a considerable effect on the discharge of the Dvina, and the ability of the depth ice to resist the passage of water is well characterized by this effect,

VII EVALUATION OF RESEARCH. RESULTS: CONCLUSIONS In assessing the research results two aspects may be considered, It may be asked (a) what significance the investigations have from the point of view of practice, and (b) what they mean to the study of ice,

From the practical standpoint the artificial bridges and the artificial production of ice provide the hydraulics engineer with a powerful weapon, These afford a very simple means, namely the use of the ice itself as a building material, of accomplishing the following under all circumstances :

(1) The cessation of the winter transfer of heat from the open surfaces of flowing streams;

(2) The regrouping and redistribution, as required, of the masses of ice in the riverso The loss of heat from the open water surface is interrupted during the winter itself by the formation of the surface ice cover, At turbulent places where such an ice cover cannot form in the natural way it must be produced artificiallyo Using these artificial means, the following, in turn, can be accomplished: (a) Considerable reduction, or even complete cessation, of the formation of the undesirable depth ice;

(b ) Noticeable improvement in the conservation of heat during the winter, Page 40 Teah. Trans, TT-85 By covering the rapids with ice the most important cause of the formation of depth ice, namely the supercooling of the water surface, is removed. The further formation of depth ice is thus prevented. The success of the method depends on how quickly and how completely these measures may be carried out. The improvement of the thermal conditions on the rivers is of great importance, especially at the time of the break-up. It has been found from observations', that an increase in the temperature of the water, even tq a few hundredths of a degree above 6', will largely destroy the adhesiveness of the soft, sticky depth ice, so that the water may pass through it much more easilyo Thus the increase in the thickness of the ice cover is halted, the under side of the ice remains smooth and the breaking up process begins. The melting process is considerably accelerated by the improvement in the thermal conditions, and under favourable circumstances the ice often disinte- grates without shifting. The second possibility, that of obtaining a favourable grouping of the masses of ice at different sections of the river, plays a particularly important part, especially in unregulated rivers where the ice break-up is usually accompanied by ice jams and floods. By appropriate use of this possibility the extent of the ice jams can be considerably reduced. As to the course of the complicated ice phenomena, which have as yet received little investigation, it is the writer 1s opinion that certain problems, particularly those connected with the thermal and dynamic causes of depth ice formation, have been introduced and may even have been solved. The investigations had a more practical nature, however, and the scientific aspects could be taken into account only insofar as the war time conditions and the available instruments would permit. Special difficulties arose during the investigations due to the lack of proper instruments, The restricting effect of this was felt particularly in the investigation of the thermal conditions and in the measurement of solar radiation. Nevertheless both these phenomena could be followed very well. It was possible, for example, to obtain important observation material and research results on the formation and growth of the surface ice cover, the quantity and grouping of the masses of depth ice, the breaking up of the ice cover, the break-up, etc, These results are now being classified, studied and evaluated, and the relationships between the Page' 91 Teeho "Trans TT-85 meteorological and hydrological factors are being investi- gatedo When these relationships are known they will in turn provide important suggestions for future investigationso

VIII THE INADEQUACY OF THE PREVIOUS METHODS; THE FUTURE PROGRAM FOR IMPROVING THE TROUBLESOW ICE CONDITIONS ON THE DVINA SINCE COMPLETION OF THE KEGGUM POWER PLANT (a) Swedish Firm As the observations and experiences of recent years have shown, .ram worked out by the Swedish firm Vattenbyggnadsbyrthe troL n, designers of the 'Keggum Power plant, for dealing with the break-up difficulties after completion of the power plant in general has not justified itself. The main features of this firrnqs maximum program (cf , Memorandum of the 16 February, 1939 ) , apart from points 1 and 2 dealing with the observation of the ice regime above the dam during the entire winter and the prediction of the break-up, must be rejected, The reasons for this are as follows:

(i) The breaking of the ice cover in the artificial lake before the start of the natural break-up, as recommended in point 3, is technically difficult, if not impossible, to carry out, As the Swedish firm itself noted in its memorandum, the only sure method of breaking the ice cover is by means of ice breakers, However, a sufficiently large open surface can be obtained with the ice bhakers at the upper end of the lake only if the ice cover is also removed at the lower end, or if the entire surface of the lake' is broken, Yet even in this case the masses of soft depth ice packed underneath the ice cover at the upper end would create insurmountable difficulties for the ice breakers, Moreover this procedure would react noticeably on the structures of the power plant itself, especially on the weirs and the energy destroying apparatusD This was emphasized by he Swedish firm (cf, Memorandum of 21 November, 1930tj) after the experiences in the spring of 1940, t) TranslatorPs note: this may refer to the previously quoted Memorandum of 25 November, 1940, or to another memorandum of 21 November of the same yeay. In any case it is apparent that 'l1930" is a misprint, Page 42 Techo Trans, TT-85 The suggestion advanced by some experts, to break up the ice in the lake at the beginning of the winter and again in spring, or to keep the surface of the lake completely or partially open for a shorter or longer period, is fraught with considerable dangerso The formation of the depth ice would begin throughout the lake (due to supercooling of the water masses) and considerable disturbance to the operation of the power plant might result, or its activity might be interrupted altogether, as has often been the experience in other countries,

Neither is there any sound basis for the suggestion put forward by the Swedish firm on 16 February, 1939, of employing the ice breakers provided for in the program for combatting the break-up to open a channel in the solid ice cover of the lake for the on-coming masses of depth ice, thus decreasing the ice difficulties at the beginning of the winter during the drift ice period. As soon as the work began the ice breakers would become mired in the depth ice masses and could not be set free, From these considerations the breaking up of the ice cover in Lake Keggum by means of the ice breakers was suspended, although two such craft had been built for this purpose and launched on the lake, As already mentioned, an attempt had previously been made to promote the breaking of the ice in the neighborhood of Friedrichstadt by dynamiting, but generally speaking this was not successful,

(ii) The lowering of the level at the dam by one metre, as recommended in point 4, is undesirable. The management of the Keggum power plant con- curred in this, and during the last two springs the opposite procedure was employed with good success, ioee, insofar as possible the water level was maintained constant during the whole period, so that the ice cover at the lower end of the lake would not crumble unnecessarily and thus permit the ice from the break-up, in its step-by-step advance, to force the solid ice cover prematurely against the weirs.

(iii) The procedure recommended in point 5, of guiding any flood wave which might be induced by the break-up through the artificial lake and over the weirs, while retaining the ice itself in the lake, cannot be carried out, Page 4.3 Tech, Trans, TT-85

It is just these high waves of water s;-.~duced by the ice jams which have been the aowce of the greatest difficulties since completion of the Keggum Power Plant, As experience has taught, they cannot be influenced in any way by the power plant Contrary to the expectations of the Swedish engineers they assume even greater proportions than in the years before the power plant was completedo In this respect the maximum program of the Swedish firm fails completely. Neither the breaking of the ice cover nor the variation of the water level brings about a re- duotion in the ice jams produced by the masses of depth ice, For this purpose an entirely different and much more ra- tional program must be adopted, The observations and research results obtained by the Department of Marine during the last two winters, and described in the present report, open the way to the inaugwation of suEh a program-

(b) The Proposed Interim Program for Decreasing the Break- UQ Difficulties on Lake Kewaum

On the basis of the investigations and experiments carried out in thz winters of 1941-43 the adoption of the following program 'is recommended, The basis of the scheme is not to fight against nature by breaking up the ice cover or by artificially raising and lowering the water level, as suggested by the Swedish firm, but to help nature draw an ice cover as quickly as possible over the swift water regions of the Dvina, which normally do not freeze over, This method makes possible s 1, Restrictlsn of the main cause of the difficulties, namely the formation of the depth ice;

2, A more f avourable grouping of the masses of ice in ths swift water regions, thus preventing the depth ice from entering the uppela end of Lake Keggumg 30 An appreciable improvement in the thermal conditions in the water during the wintero

To achieve this aim the following procedure will be adapted:

lo In autumn, immediately after the first frost, when the temperature drops to 0' C and the process of depth ice formation becomes intensified, three main ice bridges or dams should be erected simultaneously in order to restrict the formation of the depth ice in the larger rapids, so that the depth ice masses which do form will remain at the same place, The places where the ice bridges Page 44 Tech. Trans. TT-85 (or damr' are to be erected may be seen in the plan, Figure 55. They should be built according to the methods described in the report, taking into account the local terrain and the weather conditions.

2. After the main bridges have been built additional bridges should be set up. Depending on the weather conditions, these may serve either to afford relief from suddenly appearing depth ice masses or to promote freezing of the rapids, should the effect of the main bridges be insufficiento The locations of the auxiliary bridges are also marked on the plan.

3. In the turbulent rapids, where the ice bridges are no longer effective and where the strong current opposes the formation of an uninterrupted ice cover these aims may be achieved by artificial production of depth ice. The location and method of depth ice formation should be chosen according to the conditions at the time and in consideration of the experiences set forth in this report and those gained during future work. 4. To ensure success instruments for the careful recording of the hydrological and meteorological processes and their relationships should be set up well in advance. Particular heed should be paid to the timely prediction of ice formation and the effect of the thaw on the course of the break-up. For this purpose careful, continuous observations and investigations are to be carried on at the newly constructed hydro-meteorological stations at Stockmannshof and Keggum* In spring, however, during the break-up period, the required hydro-me teorological inf or- mation must be obtained from the entire drainage area of the Dvina, 5. Preparations for the production of an ice cover before the break-up in Lake Keggum are not necessary. Due to the decrease in the masses of depth ice the increases in the water level during the break-up period are much less than previously assumed,

6, At the power plant itself all preparations necessary for elimination of the static effect of the ice and for the necessary security in case of a possible movement of ice are to be carried out. Page - 45 Tech, Trans, TT-85

As for the stretch of the river below Keggum, the ice conditions here, as shown by the observations and measurements, have improved considerably as a result of the completion of the dam, On this stretch of the river the city of Riga in future has to contend only with the local ice. The passage of ice from the upper reaches of the river is no longer possible, thanks to the artificial lake of the power plant; rather, the artificial regulation of the water level in the dam basin makes it possible, to promote somewhat the freeing of this stretch of the Dvina and the Riga harbour from ice, This program should not be regarded as final and unalterableo The experiences and results of future work must be used continually to modify the program and to supplement it, so that it will gradually become perfected and, insofar a# possible, cover all contingencies arising from nature,

CONCLUDING REMARKS Before closing it should be pointed out that the investigations into the ice conditions on the Dvina carried out during the last two years by the Department of Navigable Waterways and Xarbours should by no means be regarded as a finished task, but merely as a guide to future work, Though uniform in its composition, ice varies greatly in structure and qualityo In contrast to all other natural bodies it forms, exists, and vanishes in dependence on the seasons, the geographical and topographical conditions of its location, and a great many other factors, These may differ so widely from place to place and from period to period that those treated in the present report can be considered valid only for one place, namely the lower course of the Dvina River

Moreover, nature has endowed water itself with many different properties which ape not present in other substanceso It is the only material in nature whose transition from tha solid to the fluid state of aggregation occurs at a temperature differing only slightly from the mean temperature of the earth's surface, This transition, in turn, releases relatively strong effects of internal energy which, in view of the immensity of the hydrosphere, is of foremost importance both for various geographical processes, as well as for human life, Page 46 Tech. Trans. TT-85 Thus, owing to the various physical properties of ice and water, as well as to a great many external factors, ice is a very complex object of investigation, but one to which as yet relatively little attention has been paid. As the economics of water resources continues to be developed the founding of special institutes for the study of ice and snow will become imperative. This opinion was represented with good reason at the hydrological conferences of the former Baltic States. Institutes of this nature would be of particular value to Eastern 3urope. Fig. 1 Tech.Trana.TT-

Bigh rater maru-"-- 2s3 5 a s J P I for 1931 : B a Fig. 2, 3 Tech. Trans, TT-86.

Figure 2 The embe %pide at s$ock n&of (149 km. from mou%h), Fig, 4 - 6 Tech,Trane.FT-85

Figure 4 Thin crys%dline branches of ice a$ the shore,

Figure 5 Grodh of %ha cryetalbfne branches on the supercss%~%surface ~rfthe water, Fig, 7, 8 Teeb,Traws ,TT-85

Figure 7 The Dvina a% Bliarberg (3141 km. from rnrauth)s border ice only in a narrow strip along the bsnk, &if% ice i~ the een%re sf the river.

Figuse 8 Iciwg only sn the downatream aide of the wooden pale, Figure 1% The Dvina a% S$ockmannsho%. The entire bed sf the river ie covered with ground ice.

Figure 13 Gelatinous ice and ground ice 4n $he Dvina,

Figure ll

Figures 9 to 11 Construction 0% am artificial depth-ice clog, The fin& stick produeea eddies; crys$allization sf ice occurs only on the second stick. An artificial depth ice flaw is produced in three $aye at -20" 6. Fig, 14 - 17 Tec$,Trana,TT-85

Figure 19 Core ioes 48 om,$ anow icep 3.5 em, Pig, 18 - 21 Tech ,%raws, TT-85

Figure 18 Depth ies. Pi gpre 19 Granular ice,

Figure 28 Lumps aC grm~Tt<: ice 'on the share sf ths DvkL-L; Figure %P An ics stan$., t.qo aaeks after the brs&.-l6~r-. Fig. 22 - 24 %ecA.Trans,TT-85

Figure 22 An ice stand head above Kakenhusen,

Figure 23 Mounds sf ice Figure 24 Forfiiation sf ground (~srgisse). ice in %ha Dvina 8% Stsckmannshof, (148 b, from the mouth) Lateral profile at bfedulajs sn the 29 January 1943,

l,Ydfdth sf rfves. 2,Distances;3,Bepths, 4,\'dater lave1 in metres pebove sea -beveP(Baltlc), 5,Beight of ice surf ace above Baltic Sea, Figure 26 Dm sf border ice 8% %he Prisdulaja Rapids (151 b, above the mouth),

Fig~~rs27 Msssusenera-t of ice quantikies in the Dv%na, Pig, 28 Teab,%awe ,TT-85 to~Ptudr1Profile of' %he bfaa and 104 titles $n tihe Wgnter of 1011-42, 140 to .170.h, Pig, 28 - 32, 34 Tsch,T~ana,TT-85 Ffg. 33 %ech,Trana,W-85

LOkBGITmSNfi SECTION OF THE GROWB ICE PSMD Isenswed an 87 Jmumy 1842,

Scale

ICE EMUREMENTON THZ DVEWAIN THE STUTSC MPIm SECTfOfl (135 ha, from the mouth) in the Vdinter of 1941-42

Plan sf the r%+er bed Stutachka Canyon Rapids Longitudinalsections I I\

leaswements fn the plan Figure 3$ Strip aQ border ice n - ready for Pcs bridge, r agure 38 Comarpletec! ice bridge.

Figure 3% TechnicaL p%aeemsn$ sf an ice bridge, h$ifitsiaH extensf~rrsf the barde~fee at Qtaeb in the Winter of 1941-42,

Figure 41 The Wins at S$ockmm4nahof, The u'6anyos* Rapids are covered with ice as a result of %he b ridg%wg,

Figure 40

Figure 43 %Tooden re&@ for prsduetisn sf bar&@$5~8. Fig, 42 TEE DVPNA AT SSBLEmEKI %ech,T%ana,TT-86 Artflieio%Ice d

1 Longitudinal proflile I

Legend i Border - - ice:- %- Brift ice dm - , 4-==-- ~rlftice shoring direction of flow Fig* 44 Tech,Trana,TT-86 Figure 45 The Dvina a% Kokenhusea, Above: in the wi~%er sf 1941-42, Belowars in $hs winter of 1942-43, Fig, 46 - 48 Tech,Trawa.TT-85

Figure 46

Figures 46 and 47 Pro&ua$%on of depth ies, Left: wooden poles ~o~erbdwith depth ice, Right: method of coverlng the rapids with depth ice,

Figure 48 Depth of 2ce floes for eoverfng a turbulent plaes, Pig, 49 - 61 Te sheTrans, TT- 86

Figure 49 Head of the ice jam in kke K8gv9 springg 194%.

Fisuse 51 Ice bridge at Stskmannshsf after the thaw December 1941.

r: ...... % ;=. ~ ~ -.... .>,.;.+ .. >,..... :,..... *xT.rcT...... > .'...... 7.~.~.m.5~..r.:.5..:.:...T...T...... >...... -<.....-. -...... --i-- >...... -...7r.T-ZZ. :.:,:.:.:.: .-.T.x. 7 <..:,. +n: -... v.* ...-?...... 7.~z.~.-.~.-...... ~.7~. ?.:...:.-., ",.* ..-..: ..T....,.x..7.-.-.... ?.-..... Pig, 54 Tech. TPEIna ,TT-85

CO~SEOF ~m BREAK UP PH SWPF~ATEWSECTEOB OF T= DVPKA OM Tm JACOBSTA~T-ASC~R~ENSTmTCW 1% THE SPRING OF 1943

'7-s Aam

a Open water Ice bridge or dam Pig, 52, 53, 55 Tach ,Trans ,TT-85

Ff @re 53 Site sf ice bridge after %he break-up, A channel a~~soxima%ePv40 me%reawide his been broken open fn th centre of the river, QUANTITIES OF ICE IH ME WINA blM MILLIONS OF CU.M.9

Before eowatmetibn of the lregg$wa Power Plant. BPtts~completlen 0% the I(sggwa Power Plant e-4 BID 0 CT .Ye O

ApprorLnrately 1.070 will, eu. n. of this mount is cars Ice which WiPtrd bra at(;op the break-up of the 15-17 November 1842.

"x'~ha qumblty of Ic~m.8 not measurmd dimctly but was determined trm the depth 06 snow and Pea masumd at the pours end by comparisonwith tho iee eonditI~mln the pP88iWnB winter.

Table 11% TeoPa. Trans. T-88