Dams, sediment sources and reservoir silting in Romania [609289]

Dams, sediment sources and reservoir silting in Romania
Maria Ra ˜doaneT, Nicolae Ra ˜doane
Department of Geography, University bStefan cel Mare Q, Suceava, 5800, Romania
Received 15 December 2002; received in revised form 8 April 2004; accepted 9 April 2004
Available online 22 April 2005
Abstract
Romania ranks among countries with the greatest achievements in the field of dams in the world. Among the 80 membership
countries of the ICOLD, Romania ranks 19th in blarge dams Qand the 9th in Europe. The reservoirs arranged behind dams are
characterised by small capacities, generally under 200 million mc. The total number of big dams is 246, among which almost
half are dams under 40 m in height. The highest dam is Gura Apelor, on Ra ˆul Mare, in the Retezat Mountains and it is 168 m.
We can add to these other 1500 dams, under 15 m in height, with the reservoirs having capacities under 1 million m3. The
anthropic intervention through the arranging of dams and reservoirs in the river systems of Romania’s territory is significant and
justifies the concern of geomorphologists to know the relations between the dynamics of the landscape and the behaviour ofthese anthropic structures.
This work refers only to one of the processes that reservoirs undergo once they are placed in a river system—the silting.
More precisely, we try to make a synthesis of the knowledge stage of the reservoirs silting in Romania that we want to approach
on the basis of the relations with our territory morphodynamics and considering the substantial accumulations of new data. Thetotal erosion rate from Romania’s territory is, on the average, 125 million tons/year out of which 45–50 million tons/year are
transferred by rivers.
We analyzed 138 reservoirs with initial volumes between 1 /C210
6m3and 1230 /C2106m3for which there is a determination of
the silting time. The situation of the reservoirs silting from Romania is as follows: very serious for 15 dam reservoirs with
average dimensions of 8 million m3, all of them situated in the Sub-Carpathian area, one of the important sediment production
areas (over 500 tons/km2/year), with the silting time T50 of these reservoirs having values between 2 and 10 years; serious for
30 reservoirs, with average capacities of 35 million mc, and the silting time T50 between 10 and 50 years. In this case, thereservoirs are also situated in the area of important specific sediment production of over 250 tons/km
2/year (the case of the
rivers Olt, Arge Y, Buza ˘u, and Bistri _a but also of the reservoirs in the Ba ˆrlad basin), many of them being arranged as a cascade
of small reservoirs on the main rivers; difficult for 13 reservoirs, with a silting time b100 year and which are usually situated in
the area at about 200 tons/km2/year.
D2005 Elsevier B.V. All rights reserved.
Keywords: Construction of dams; Sediment sources; Reservoir sedimentation; Morphodynamic features of Romania’s territory
0169-555X/$ – see front matter D2005 Elsevier B.V . All rights reserved.
doi:10.1016/j.geomorph.2004.04.010TCorresponding author. Fax: +40 230 520080.
E-mail address: [anonimizat] (M. Ra ˜doane).
URL: http://www.usv.ro/istgeo (M. Ra ˜doane).Geomorphology 71 (2005) 112– 125
www.elsevier.com/locate/geomorph

1. Introduction
Dams and the reservoirs created behind them have
represented a domain of interest for geomorphology,especially for its dynamic branch, because it has beenstated that such anthropic structures cause irreversiblechanges in the dynamics of the fluvial systems. Theissue has been debated in a series of works and PhD
theses, especially by the research team from
bStejarul QResearch Station from Piatra Neam _, but
also by researchers from other domains of activity.The arguments brought forward deal especially withthe fact that nowadays the large dams of the worldwith their hydrographic systems have been (in mostpart and some of them totally) controlled by damswith their reservoirs, with a water volume of 5–6
times the average discharge of all the rivers in the
world, estimated at almost 1250 km
3/s (Ichim and
Ra˘doane, 1986 ). The arrangement of transversal dam
work introduced great discontinuities in the trans-portation of sediments, in the evolution of riverbeds,and in the adjoining slopes, which in geological timeare controlled with a very reduced rate of manifes-tation by the tectonic movements and the variations
of the general base level. As for the development in
space and the duration of manifestation of theinfluence of such anthropic structures, Williams and
Wolman (1984) estimated, on the basis of the
analysis of an important number of cases, that thecourse of big rivers may be hundreds of kilometersand the duration thousands of years. Or, as we shallsee, such structures are to be found also in Romania
having a total volume of almost 13 billion m
3(one
third of the entire volume of water carried in 1 yearby the interior rivers). Moreover, they are accom-panied by the dislocation of important amounts ofrocks, by terrigenous materials, which, only between1950 and 1990 in the context of hydropowerarrangement, have totaled 500 million m
3of embank-
ment, 771 km of dams, 33 million m3of surface
concrete and, 12 million m3of underground excava-
tions on 669 km of drifts.
This work refers only to one of the processes that
reservoirs undergo once they are placed in a riversystem—the silting. More precisely, we try to make asynthesis of the knowledge stage of the dam lakessilting in Romania that we want to approach on thebasis of the relations with our territory morphody-namics and considering the substantial accumulations
of new data. The factual material that we have isstructured as follows: (i) the construction of dams andthe arrangement of reservoirs in Romania, (ii) theproblem of sediment sources and, (iii) the silting ofreservoirs.
2. The construction of dams and the arrangement
of reservoirs in Romania
Romania is known as a country where the tradition
of dam construction and the arrangement of lakes isvery old ( Fig. 1 ). The Saard and Cristurul Pools near
Turda are from the twelfth century. The oldestreservoir dates from as early as 1780, whose dam of
23 m and after several repairs is still functional—
Ta˘utu Mare Reservoir from Metaliferi Mountains was
built for the gold mines. Since the fifteenth century inRomania much interest has been shown for thearrangement of rivers with small accretion andwaterfalls. Documents are available that certify poolsas early as 1448; and the Bra Yov area, according to
some historical documents, between 1503 and 1550
there were 28 pools. In another old document,
Moldova was described as rich in pools, some ofthem probably existing at least from the period ofStefan cel Mare (the reservoirs Ha ˆrla˘u, Belce Yti,
Sipote, Dinischean); and others were arranged later,especially during Alexandru La ˘puYneanu’s and Vasile
Lupu’s reigns. Moreover, in Vasile Lupu’s period,DracYani Pool was enlarged; and it is still considered
to be one of the largest pools in Romania (surface
area =486 ha and capacity =5.5 million m
3).
The modern and contemporary period marked in
Romania an increasing interest in the arrangement ofwaterfalls for hydropower purposes. So that at the endof the nineteenth century the first hydroelectric powerstations were built on Da ˆmbovi _a in Bucharest (1890)
and on Sadu near Sibiu (1896) without having too
much water accumulation. The ample study about the
hydropower reserves by the brilliant scientist Pavel
(1933) may be considered the first synthesis on
arrangement conditions of dams and on dam reser-voirs in Romania. Until 1940, only 128 hydropowerplants had been built, but the water accumulationswere not so important. Starting from the 1960, thepace of arranging reservoirs became faster, culminat-M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 113

ing between 1980 and 1990 when 78 reservoirs were
put into operation ( Figs. 2 and 3 ).
After this period, a severe decline of dam
constructions was registered between 1991 and2000, when only 17 dams from those which hadalready been started were finished. This dynamic isalso graphically illustrated in Fig. 4 , which indicates
the pace of dam construction in Romania in the
twentieth century, after the official data published by
theRomanian Committee for the Big Dams (2000) .A t
present, the data indicate that Romania is among thecountries with the greatest achievements in the world
Fig. 2. Izvoru Muntelui Dam, no. 27 in Fig. 8 (height = 127 m; length = 430 m; reservoir volume = 1230 million m3).
Fig. 1. Locations of dams built in Romania in ancient times and the Middle Ages ( Dascalescu, 2000 ).M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 114

regarding dams, which also allowed technology
export (Algiers, Iran, Turkey). Among the 80 country
members of the International Committee of the BigDams, Romania occupies the 19th place regarding thenumber of bbig dams Q(considered over 15 m height)
and the 9th place in Europe. The total number of bigdams is 246, among which almost half are dams under
40 m height. The highest dam is Gura Apelor on
Raˆul Mare, in the Retezat Mountains, and it is 168 m.
We can add to these another 1500 dams under 15 mheight, with reservoirs having capacities under1 million m
3.
0301113337678
17
0102030405060708090
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000Time in yearsNumber of dams
Fig. 4. The rate of dam construction in Romania in the twentieth century (data offered by the Romanian Committee of the Big Dams, 2000 ).
Fig. 3. Vidraru Dam, no. 1 in Fig. 8 (height = 166 m; length = 480 m; reservoir volume = 465 million m3).M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 115

As a conclusion, we can estimate that the anthropic
intervention through the arranging of dams andreservoirs in the river systems of Romania’s territoryis significant and justifies the concern of geomorphol-ogists to know the relationship between the dynamicsof the landscape and the behaviour of these anthropicstructures.
3. The sediment sources
The position in a temperate-continental climate and
the presence of the Carpathians is defining for thedistribution and the system of the geomorphologicprocess, generating sediments, and expressing, finally,the morphodynamic specific of a territory. This is also
the reason why a great importance was given to the
problem of sediments flowing on Romania’s rivers, asan indirect expression of the dynamic state of thelandscape. Beginning with the main work of Diaconu
(1971) , we had, for the first time, a global image of
the susceptibility to erosion of our country’s territoryand of an overwhelming importance for time pre-dictions about dam reservoir silting. Two theses
concerning a general perspective over the dynamic
of the Romanian relief have been subsequentlyelaborated. The first, written by Mo_oc (1984) ,
proposes a general image over the whole territory ofRomania, referring to sediment delivery in compar-ison with the major types of morphogenetic processesand the main unities of our country’s landscape. It is apioneer work as far as Romania is concerned. The
other synthesis was published by Mociorni _a and
Brate Y(1987) , who updated the outflow map after
1970–1980 when most of our rivers reached themaximum discharges with a recurrence interval of 100years. It is a work that relies on the entire databaseresulting from the national measurement network overa more than 35 years period. Other updated syntheseshave not been known after this date, except for re-
editions and interpretations of the above mentioned
works.
As far as we are concerned, on the general context
concerning the sediment source problems, a newsynthesis has been given (unfortunately it has notbeen realised for the entire territory of Romania, but alarge part has been treated) on the basis of an updateddatabase. Our approach focuses on the definition ofthe sediment sources such as: (i) source area related to
the slope basin or riverbeds and with land use(agricultural, forest, buildings, mining, etc.) and (ii)in comparison to the generating processes, namely,those that make transition to and into the riverbeds ofsediments.
The processing of a large amount of data obtained
from various sources (measurements in hydrometric
cross-sections from the national network ensured by
the Romanian Waters Administration, indirect estima-tions because of sediment stock from some reservoirs,personal measurements on small basins) has led us tothe selection of two control factors as criteria ofsediment source analysis for a large territory such asRomania. They are: (i) the lithological composition ofthe rock generator sublayer and (ii) the size of the
drainage basins that provide a selection of the amount
of the sediments conveyed from the origin area to thedischarging area. The choice of these two factors isalso motivated by arguments acquired from thespecialty literature analysis (that is authors who havesuggested prognosis models of the sediment yield,such as: Gregory and Walling, 1976; Dietrich and
Dunne, 1978; Jansson, 1982; Griffiths, 1981; De
Villiers, 1985 , etc.), but also from personal experi-
ences (the model of multiple regression for theestimation of the sediment yield in drainage basinswith 400 km
2;Ichim et al., 1987 ).
Our proposal focused on acquiring some predictive
equations of sediment yield for Romania in which thetwo controlling factors (the lithologic substratum andthe size of the drainage basins) should be considered
independent variables. The database that we had at
our disposal refers to 212 cross-sections controlled bybasins varying from 0.17 km
2to over 10 000 km2
from 13 areas in our country and were lithologicallyand geomorphologically different. The data process-ing consisted of many stages that have ultimately ledto the equations listed in Table 1 . These relations are
rendered in Fig. 5 , from which we can easily infer that
for Romania’s territory there is a considerable
variability of sediments generating in different areasof the country. These equations are power functions,and their parameters (mainly aand bregression
coefficients) can be used in subsequent classificationanalysis.
Regression coefficient ahas values from 42.861 to
10 006.4. Its meaning in the relation is that it is closelyM. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 116

connected to the actual conditions for which the
function was created (mainly morpholithological inour case). From this point of view, aregression
coefficient can be used in cluster classification analysis.Regression coefficient bdenotes the inclination
degree of the slope regression line. In our case, itvaries between /C00.0072 when the regression line is
almost horizontal and /C00.4316 when the line is more
Fig. 5. Relationships between the sediment yields and the drainage basin areas for different morpholithological conditions of the Romania
(numbers correspond with Table 1 ).Table 1
The centralization of the relations between the specific sediments production and the drainage basin area for various morphological conditions inRomania (SY=sediment yield; A=drainage basin area)
Description of the area Statistic parameters of SY= aA
brelations No. of
observations ( n)Ab r R2
(1) The flysch mountain area (the Eastern Carpathians) 738.48 /C00.167 0.799 0.639 49
(2) The Neogene molasse area (the Eastern Sub-Carpathians) 5677.47 0.220 0.904 0.817 35(3) The Neogene molasse and Quaternary deposits area
(Sub-Carpathians and Getic Piedmont)9367.43 /C00.277 0.647 0.419 11
(4) The crystalline mountains area (Jiu)-mining influences 320.40 /C00.103 0.364 0.133 12
(5) The Neogene molasse area (Jiu-Olte _) 2094.93 /C00.175 0.879 0.772 18
(6) The Getic Piemont (on the basis of lake sediments)
a10006.40 /C00.194 0.336 0.113 6
(7) The crystalline mountains area (on the basis of lake sediments)a1450.82 /C00.124 0.746 0.557 4
(8) Small basins in the crystalline mountains areaa42.86 /C00.007 0.451 0.203 8
(9) Moldavian Plateau (Ba ˆrlad) 2203.00 /C00.318 0.670 0.449 15
(10) Moldavian Plateau, Moldova Plain (Jijia) 3217.98 /C00.432 0.362 0.131 13
(11) Oltenia Plain 268.27 /C00.284 0.511 0.261 12
(12) The crystalline and volcanic mountain area (Some Y-ViYeu) 361.16 /C00.224 0.491 0.241 12
(13) The region of the internal flysch (Some Y-ViYeu) 435.25 /C00.073 0.212 0.045 17
aThe bed load was taken into calculation.M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 117

inclined; therefore, sensitivity is a higher. The other
statistical tests have helped us to accept or refuse thepredicative significance of the equations, some ofthem being less sensitive then the others.
Cluster analysis was applied in order to find the
best groups among the great number of studied data,on the basis of which we should realise a territoryregionalization with a similar power of generating
sediments. With that end in view, we used the
regression coefficient a, whose variation synthesizes
the complexity of the morphological and morphody-namic conditions in a certain area. The clusters werefixed by calculating the variance-within-the-groupsand the variance-among-groups, according to themethodology described by Johnston (1986) .T h e
graphic result is rendered in Fig. 6 where it is
observed that the coefficients tend to group into three
clusters.
Cluster I brings together most of the coefficients
(in which four groups have been concentrated) andcharacterizes the zones with sediment yield of under700 tons/km
2/year. These zones include most of the
regions in our country: Moldova Plain, Oltenia Plain,the area of the internal and external flysch, the
crystalline zone, and the volcanic mountains.
Cluster II is made up of two subgroups and
characterizes the zones with sediment production ofabout 2000 tons/km
2/year, mainly the Moldavian
Plateau and the Getic Plateau.
Cluster III characterizes the exceptional sediment
productions registered by some small basins in the
Bend Subcarpathians and the Getic Piedmont.
On the basis of this classification, we have
obtained generalized relations Sy=f(A)(Fig. 7 ) for
the mentioned groups, which may be compared toWalling’s generalized tendencies (1983) for different
regions of the world. The general observations to bekept in mind from this analysis are that
(i) In Romania’s case, the sediment transit is
bdelayed Qfrom the source area to the delivery
area, expressed by the negative relation of thespecific sediment production once the size ofthe drainage basin has increased. The phenom-enon is due to the selective transportation of thesediments inside a drainage system. This bloss
of sediments Qtakes place on the slope of
/C00.190 for the areas in Clusters I and II and
on a higher slope of b= 0.328 for the areas in
Cluster III.
(ii) Most of Romania’s territory enters the global
centralized tendency belonging to Walling
(1983) , with a regression slope of /C00.125.
We may estimate that the areas in Cluster I,which characterizes most our country’s mor-
pholithological areas are placed in the regres-
sion line of maximum intensity areas, asituation that may be assimilated to a mediumcondition of specific sediment production forthe most part of the globe. On the contrary, thesecond and third group, although characteriz-ing areas with little extinction (the bend sub-Carpathians and Getic sub-Carpathians), are
registered as some of the most productive
alluvial suppliers in the world.
4. Reservoir silting
Interest in the reservoir silting study decreased
immediately after the dam construction slowed down
Fig. 6. Cluster analysis on the aregression coefficients from the
relationship Sy = f(A) for the morpholithological conditions of
Romania.M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 118

in Romania, although the present reservoirs need
attentive supervision from this point of view. But we
must not neglect the fact that a study on the silting
of dam reservoirs needs an expensive price, the mainreason for which the research in this domain hasdecreased lately. As far as we are concerned, wehave a rich experience in the study of the phenom-enon and an important database that helps usapproach the problems of better knowing the damreservoirs silting in Romania (the way it is now), and
in trying to relate it to the morphodynamic character-
istics of our country’s territory to highlight theextreme situations and the possible causes of thisprocess.
The database on which our observations are
based, includes, on the one hand, a situation on thedimensions of the lacustrine basins (capacity, area,position inside the hydrographic basin); on the other
hand, a situation on the silted volume of the basin
and an evaluation of the silting time of 50% of theinitial storage capacity of the lake. All these datahave been obtained from extremely different sources,
from our own research on some lakes in Bistri _a
Valley, Siret Valley, and Buza ˘u Valley to attentive
research of scientific production appearing in the lastdecades: the archives of The Institute of HydropowerStudies and Projections, The Romanian Committeefor Great Dams, National Institute of Meteorologyand Hydrology, Aquaproject, and others. Because ofthe small area that our work covers, we will presenta part of this statistical data in the form of a table
(Table 2 ).
The estimation of the stage of the reservoir silting
requires some knowledge about the initial capacity ofthe reservoirs: of a parameter that is called accumu-lations coefficient ( a) and that is defined as the
relation between the volume of the drainage basin’sflow drain and the reservoir’s initial capacity; and ofanother parameter that is called silting time, of 50%
from the reservoir’s volume (T50). The analysis of
these parameters has been done on the basis of acorrelation with the morphodynamic features of
Fig. 7. Generalized relationships between the sediment yield and drainage basin area for the Romania Ts territory (A) and obtained by Walling
(1983) for different conditions of the world (B).M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 119

Table 2
The silting situations of some reservoirs from Romania (their position is showed in Fig. 6 )
No. of
reservoirDrainage
basinReservoir Drainage basin
area, A,k m2Initial
volume10
6m3Siltedvolume%Year of
constructionPeriod of
investigation(years)Source
1A r g e Y Vidraru 286 465.00 0.43 1967 11 Ionescu (1980)
2 Oie Yti 441 1.80 74.00 1967 23 Ichim et al. (1994)
3 Cerbureni 480 1.62 68.00 1966 11 b
4 Curtea de Arge Y 570 0.89 88.00 1972 13 b
5 Zigoneni 625 13.30 23.70 1973 13 b
6V ı ˆlcele 696 41.50 8.42 1977 11 b
7 Budeasa 1100 30.80 8.30 1980 8 b
8 Bascov 1162 5.40 93.00 1971 7 b
9 Pite Yti 3085 4.80 85.00 1970 6 b
10 Golesti 3100 107.90 1.95 1983 6 b
11 Doftana Paltinu 334 56.00 4.00 1972 17 RoYca (1987)
12 Ialomi _a Pucioasa 448 10.60 33.50 1974 13 Ionescu (1980)
13 Sa ˘cele 170 18.30 6.02 1976 12 b
14 R. Va ˘lcea 15285 21.40 73.20 1974 13 b
15 Da ˘eYti 15143 11.70 39.10 1976 8 b
16 Olt Rı ˆureni 15536 7.30 14.00 1977 8 b
17 Govora 15727 18.50 27.00 1975 3 b
18 Ba ˘beni 16847 59.65 8.30 1977 5 b
19 Stra ˘jeYti 18229 202.70 3.20 1978 5 RoYca (1987)
20 Ione Yti 17222 24.90 2.50 1980 6 b
21 Zavideni 17480 51.20 2.18 1979 6 Ionescu (1980)
22 Dragasani 17691 66.60 1.28 1980 6 b
23 Mure Y Cinci Y 301 43.00 3.50 1969 21 b
24 Bucecea 1983 14.40 32.30 1978 18 Olariu (1992)
25 Galbeni 19445 40.00 59.60 1983 12 b
26 Poiana Uzului 420 170.00 5.11 1975 20 b
27 Iz. Muntelui 4025 1230.00 1.30 1962 27 Ra˘doane (2002)
28 Pı ˆnga˘ra_i 5144 6.70 48.00 1964 23 Ciaglic et al. (1973)
29 Vaduri 5220 5.60 11.40 1966 16 Ra˘doane (2002)
30 Bı ˆtca Doamnei 5229 10.00 27.20 1966 24 b
31 Piatra N. 5232 12.00 3.40 1966 8 b
32 Racova 6566 8.60 50.40 1964 21 Olariu (1992)
33 Gı ˆrleni 6633 5.10 25.30 1965 17 b
34 Siret Lilieci 6727 7.40 10.70 1966 17 b
35 Baca ˘u 6763 7.40 18.40 1966 20 b
36 Belci 1093 12.00 50.00 1964 27 b
37 Pu YcaYi 336 17.20 62.30 1973 25 Purnavel (1999)
38 Antohe Yti 40 0.22 40.91 1984 11 b
39 Ga ˘iceana 47 0.41 41.46 1984 11 b
40 Cuibul Vulturilor 542 9.50 32.63 1978 14 b
41 Rı ˆpa Albastra ˘ 253 10.60 21.13 1979 14 b
42 Solesti 414 52.70 1.14 1974 11 b
43 Fitiche Yti 163 5.50 52.60 1977 16 b
44 Prut Pod Iloaiei 525 37.00 32.30 1964 11 Zavati and Giurma
(1982)
45 Cucuteni 122 14.00 5.43 1964 10 b
46 Eza ˘reni 41 3.50 13.60 1963 12 b
47 Ciube Yti 81 12.30 5.20 1963 12 b
48 Aroneanu 47 8.30 19.98 1964 11 Pricop et al. (1988)
49 Stanca 12000 1400.00 2.50 1978 8 Ra˘doane (2002)
50 Iad Lesu 89 28.30 0.06 1973 1 b
51 Gladna Surduc 135 50.00 2.40 1976 9 bM. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 120

Romania Ts territory. Moreover, the table was com-
pleted with a map that indicates the repartition of thereservoirs following the latest data supplied by theRomanian Committee for the Big Dams, superim-posed over the map of the specific production ofsediments on Romania’s territory (map brought up todate as well; Fig 8 ).
4.1. Concerning the capacity of the lacustrine basins
As one can see from the graphic representations
belonging to Fig. 9 A, the reservoirs in Romania are
characterised by relatively small storage capacities.Almost 90% of the existent reservoirs have capacitiesunder 200 million m
3and among these, half of them
have capacities under 20 million m3. The relief
conditions and the conditions concerning Romania’sriver flow offered fewer changes for the arrangement
of big dams and, implicitly, of big lakes. The onlyexceptions are few: the Izvoru Muntelui Reservoir(the biggest among the interior rivers of our country),Vidraru on Arge Y, Vidra on Lotru, Siriu on Buza ˘u,
Gura Apelor on Ra ˆu Mare, etc. A lot of the existing
rivers are arranged with waterfalls (Bistri _a, Siret,
Buza˘u, Arge Y, and Olt) with specific exploitation
conditions, which reflects directly over a certain
silting rate. This is how the large number of lakeson the rivers Olt, Arge Yand Siret (illustrated on a
graphic in Fig. 9 B) are to be explained. The reservoir
capacity and the exploitation conditions are importantelements, which control the sediment restrainingdegree and sediments from the source area.
The capacity of lakes is decisive for the evaluation
of the rhythm and of the silting time due to very
Fig. 8. Position of reservoirs in relation with the specific production of sediments. Numbers have correspondence in Table 2 .M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 121

simple reasoning: the larger a lacustrine basin, the
more it can stock a large enough volume of sedimentswithout affecting its functionality—and there aremany examples in this direction. On the contrary, alacustrine basin with a reduced capacity can become
silted in a relatively short period, a few years or a few
tens of years, even at relatively modest rates ofsedimentation. The study of Dendy et al. (1973) for
1100 dam reservoirs in the USA indicated that thegreat majority of small lakes are becoming silted inb30 years.4.2. The silting rate
Some reservoirs in Romania have been functional
for centuries (like those in the Banat or MetaliferiMountains), but there are also lakes that became silted
in a period of a few years. From the data we have at
our disposal, we keep in mind a few observations ofgeneral character:
(i) in the whole country in an average period of 15
years, the reservoirs from the interior riversDistribution of reservoirs by
hydrographical systems
051015202530354045
Somes-TisaCrisuriMuresBanat
Jiu-CernaOlt
VedeaArges
lalomita-Buzau-MostisteaSiret
Prut-BarladNumber of reservoirsBReservoirs capacity (m3)
0106
45
20
101314
4422 232
020406080100120
0204060 80100120
Classes of capacity (mil. m3)Number of reservoirsA
Fig. 9. The repartition of the dam reservoirs in Romania. (A) The histogram of the dam reservoir’s capacity, under 140 millions mc. (B) The
repartition of dam reservoirs on hydrographical systems.M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 122

have had depositions about 200 million m3of
sediments (from which almost a half are onlythe reservoirs on the rivers Arge Yand Olt), with
a yearly installment of 13.4 million m
3, which
represents 27% from the total supply of sedi-ments, averaged and multiplied yearly;
(ii) the most important yearly rates of silting have
been on the lakes from the sub-Carpathian area
with easily erodible rocks, on the Arge Yriver:
PiteYti 15.7%, Bascov 11.7%, Oie Yti 9.5%,
Cerbureni 7.3%, and Curtea de Arge Y5.3%;
also Lake Galbeni on the Siret River, 10.6%;
(iii) average yearly rates of quick silting have been
recorded also at the first lakes built on the OltRiver: Govora 8,.7%, Rm. Va ˆlcea 5.63% and
Da˘eYti 4.90%; in the same category are
included the lakes Pa ˆnga˘ra_i on the Bistri _a
River, 3.45%, or Pucioasa on the Ialomi _a
River, 2.58%; and
(iv) a low rate of silting has been registered at the
big reservoirs: Izvoru Muntelui at 0.03% andVidraru at 0.04%, which ensure them with amillenary running, unless some incalculablesituations occur.
Retaining as a basis of interpretation the necessary
time for silting of 50% of the initial volume of everyreservoir, 138 reservoirs from Romania, with an initialstorage capacity between 1 /C210
6m3and 1230 /C2106
m3, have been analysed. For these reservoirs, deter-
minations have been made regarding the silting rateusing various methods and different experts. Their
repartition (depending on the major units of relief;Fig. 10 ) indicates that from the total number of
analysed reservoirs only 44 are found in the mountainareas of the country, the region with the smallest rateof sediment production. The other lakes are placed inthe regions of plateaus and hills, the Sub-Carpathians,piedmont and plain, all these being characterisedthrough an accelerated rate of producing the sedi-ments, except the plains.
In this general situation, the silting time of 50% of
the reservoir’s volume reflects the means of reply,through silting, of the drainage basins in comparisonwith the main morphodynamic regions of theanalysed territory: it is reduced to b100 years for
the reservoirs found in the regions with the greatestproduction of sediments (sub-Carpathians, plateauand piedmont) and it is hundreds of years for the
reservoirs situated in mountain and plain areas. In
other words, only 57 reservoirs have enough siltingtime to justify the investment and important pertur-bation on the environment.
As a conclusion, the situation of the reservoirs
silting from Romania is as follows:
(i)very serious for 15 reservoirs with average
dimensions of 8 million m
3, all of them situated
in the Sub-Carpathian area, one of the importantsediment yields (over 500 tons/km
2/year); the
silting time T50 of these reservoirs have valuesbetween 2 and 10 years;
(ii)serious for 30 reservoirs, with average capaci-
ties of 35 million mc, and the silting time T50varies between 10 and 50 years. In this case, the
reservoirs are also situated in the area of
44
1533
1613898
82120
50404
1101001000
Mountains Sub-Carpathians Plateaus, Hills Piedmonts PlainsNo reservoirs
T50
Fig. 10. Distribution of reservoir number and time of silting 50% of initial storage capacity in relation with main units of landforms.M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 123

important specific sediment yield of over 250
tons/km2/year, the case of the Olt, Arge Y, Buza ˘u
and Bistri _a Rivers ( Fig. 11 ), but also of the
reservoirs in the Ba ˆrlad basin.
(iii)difficult for 13 reservoirs, with a silting time
b100 years and that are usually situated at
about 200 tons/km2/year (e.g., Rogoje Yti on the
Siret River, Izbiceni on the Olt River, Baca ˘uo n
the Bistri _a River, Va ˘liug on the Ba ˆrzava River).
Following this general image of the silting
phenomenon in the reservoirs in Romania, we haveto take into consideration the fact that in somearrangement projects priority it was given to the
strict economic aspect and it was seriously eluded
the knowledge the relief potential to reply in such anaccelerated rate to sediment release and transportthrough the collector net. The reservoirs on theArgeY,O l to rB i s t r i _a Rivers are arranged in
cascades. Important sums of money are spent inefforts made to desilt some important reservoirs forthe functioning of the hydropower system, such as
the Oie Yti reservoir on the Arge YRiver and Pa ˆnga˘ra_i
on the Bistri _a River. On the other hand, it is
admitted the fact that there haven’t been donesediments keeping works in the source areas firstand only after that the proper execution of thereservoir; in fact so many time the right order hasbeen totally changed. The most notorious examples
of this situation are the Bascov and Pite Yti reservoirs:
entirely silted in 2 years.
Acknowledgements
We would like to thank to Dr. P. Beyer and to Dr.
R.A. Marston for substantial and constructive com-ments on preliminary draft of this paper. Also, specialthanks to an anonymous referee for his helpfulreviews.
References
Ciaglic, V., V ornicu, P., Stefan, A., Rudnic, I., Micu, I., 1973.
Contribu _ii la cunoa Yterea fenomenului de colmatare a lacului de
acumulare Pa ˆngara _i. Hidrobiologia, 14.
Dascalescu, N., 2000. History of Dam Construction in Romania.
Romanian National Committee on Large Dams, Bucharest,pp. 16– 26.
Dendy, F.E., Champion, W.A., Wilson, R.B., 1973. Sedimenta-
tion survey in the United States. Man-made Lakes: Their
Problem and Environmental Effects, Am. Geoph. Union, vol. 17,pp. 347– 359 (Washington D.C.).
De Villiers, B.A., 1985. A multivariate statistical evaluation of a
group of drainage basin variables: a South-Africa case study.
First International Conference on Geomorphology, Manchester,UK. 134 pp.
Fig. 11. Izvoru Muntelei Reservoir. View on upstream sedimentation area after a period of emersion.M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 124

Diaconu, C., 1971. Probleme ale scurgerii aluviunilor pe ra ˆurile din
Roma ˆnia. Studii de Hidrologie, vol. 30. INMH. 307 pp.
Dietrich, W.E., Dunne, T., 1978. Sediment budget for a small
catchment in mountainous terrain. Z. Geomorphol. Suppl. (29),
191–206.
Gregory, K.J., Walling, D., 1976. Drainage basins. Forms and
Processes. Ed. Arnold, London. 458 pp.
Griffiths, A.G., 1981. Some suspended sediment yields from
South Island catchments, New Zealand. Water Resour. Bull. 17,662–671.
Ichim, I., Ra ˘doane, M., 1986. Efectul Barajelor ı ˆn Dinamica
Reliefului. Edit. Academiei, Bucure Yti. 157 pp.
Ichim, I., Ursu, C., Ra ˘doane, M., Dumitrescu, G., 1987. Cercetarea
asistata ˘de calculator a ierarhiza ˘rii factorilor de control ai
produc _iei de aluviuni din bazine hidrografice mici. SC GGG,
Ser. Geogr. 34, 17– 28.
Ichim, I., Ra ˘doane, M., Ra ˘doane, N., Grasu, C., Cochior, C.,
1994. Bugetul de aluviuni al bazinului ra ˆului Olte _. Lucra ˘rile
Sesiunii Stiin _ifice Anuale. Institutul de Geografie, Bucure Yti,
pp. 17–29.
Ionescu, F., 1980. Considera _ii privind colmatarea acumula ˘rilor.
Hidrotehnica 25 (12), 272– 277.
Jansson, B.M., 1982. Land erosion by water in different climates.
UNGI Rapport, vol. 57. Uppsala Univ., Sweden. 141 pp.
Johnston, R.J., 1986. Multivariate Statistical Analysis in Geography.
Longman, Londra. 276 pp.
Mociorni _a, C., Brate Y, E., 1987. Unele aspecte privind scurgerea de
aluviuni ı ˆn suspensie ı ˆn Roma ˆnia. Hidrotehnica 32 (7), 11 – 19.
Mo_oc, M., 1984. Participarea proceselor de eroziune Yia
folosin _elor terenului la diferen _ierea transportului de aluviuni
ıˆn suspensie pe ra ˆurile din Roma ˆnia. Bul. Inf. ASAS 13,
7 –16.Olariu, P., 1992. Impactul antropic asupra regimului scurgerii apei Yi
aluviunilor ı ˆn bazinul hidrografic Siret. Lucr. IV, Simpozion
PEA, pp. 121– 130 (Piatra Neam _).
Pavel, D., 1933. Plan Ge ´ne´ral d’Ame ´nagement des Forces
Hydrauliques en Roumanie. Nat. Roum. Pour l’e ´tude de
l’ame´nagement et de l’outil, des sources d’e ´ne´rgie, vol. 58.
382 pp.
Pricop, A., Nicolau, A., Leu, D., 1988. Studiu privind influen _a unor
factori cauzali asupra colmata ˘rii lacurilor de acumulare din
bazinul hidrografic Bahlui. Lucr. Celui de al II-lea SimpozionPEA, pp. 114–120 (Piatra Neam _).
Purnavel, Gh., 1999. Cerceta ˘ri privind efectul lucra ˘rilor de
amenajare a forma _iunilor toren _iale, aflate ı ˆn zona de influen _a˘
a˘excesiva ˘a lacurilor de acumulare, asupra procesului de
colmatare a acestora; cu referire la Podi Yul Central Moldove-
nesc. Rez. Tezei de doctorat, Universitatea Tehnica ˘bGh.
Asachi QIaYi.
Ra˘doane, N., 2002. Geomorfologia Bazinelor Hidrografice Mici.
Editura Universita ˘_ii Suceava. 250 pp.
Romanian Committee for the Big Dams, 2000. Romanian National
Committee on Large Dams. Bucharest. 351 pp.
RoYca, D., 1987. Cerceta ˘ri complexe asupra colmata ˘rii lacurilor de
acumulare. Lucr. Primului Simpozion P.E.A., pp. 63– 70 (Piatra
Neam _).
Walling, D.E., 1983. The sediment delivery problem. J. Hydrol. 65,
209– 237.
Williams, G.P., Wolman, M.G., 1984. Downstream Effects of Dams
on Alluvial Rivers, U.S. Geol. Survey Professional Paper, vol. 1,286. Washington, DC, 83 pp.
Zavati, V., Giurma, I., 1982. Cerceta ˘ri privind colmatarea unor
lacuri de acumulare din bazinul hidrografic Bahlui. Hidrotehnica
27 (2), 21–29.M. Ra ˜doane, N. Ra ˜doane / Geomorphology 71 (2005) 112–125 125