erosion and sedimentation problems in serbia

Hydrological Sciences—Journal—des Sciences Hydrologiques, 44(1) February 1999 ,--,

Erosion and sedimentation problems in Serbia

SLOBODAN PETKOVIC, NADA DRAGOVIC & SLADJANA MARKOVIC Faculty of Forestry, University of Belgrade, Belgrade, Yugoslavia

e-mail: [email protected]

Abstract Erosion processes and sediment yield over Serbian territory are reviewed, as well as the distribution of sediment load in the hydrological network in Serbia. The overview of erosion problems consists of an analysis of the main erosion factors over the Serbian territory, a review of field investigations of soil loss and a description of the erosion control measures implemented in the region. The review of sedimentation problems includes a presentation of the sediment budget of the hydrographie network of Serbia, an analysis of bed load transport in the main Serbian rivers and consideration of issues related to the reservoir siltation.

Problèmes d'érosion et de sédimentation en Serbie Résumé L'objet de cet article est d'exposer les problèmes d'érosion et de transport solide en Serbie (République Fédérale de Yougoslavie). Cet exposé comprend une analyse des processus érosifs et sédimentaires dans le teritoire serbe et une présentation du transport solide dans le réseau hydrographique. En ce qui concerne les problèmes d'érosion on présentera les principaux paramètres régissant ce phénomène en Serbie, l'interprétation des observations des quantités érodées dans le milieu naturel et une description des ouvrages de protection et de contrôle de l'érosion. Pour ce qui est des problèmes sédimentaires nous présenterons un bilan du transport solide dans le réseau hydrographique serbe, une analyse du transport par charriage dans les principales rivières et nous aborderons les problèmes liés à l'envasement des retenues.

INTRODUCTION

The significance of soil erosion and related sediment problems is broadly recognized throughout the world. Erosion and sedimentation are part of the natural evolution of the landscape, but they constitute some of the most fundamental problems for the development of agriculture and forestry and for utilization of natural resources. Erosion and sedimentation processes are related to conditions in a river basin-topography, geology, climate, hydrology, vegetation and sediment characteristics.

Erosion and sediment issues are often country-specific. Erosion problems in Serbia refer to both the on-site and off-site effects. The on-site effects are significant in the hilly and mountainous area, which represents about 75% of the Serbian territory. The soil loss of arable land in this area is considerable. On the other hand, the off-site effects of erosion are related to the excessive sediment transport in the streams downstream of an eroded area. In terms of sedimentation problems, reservoir siltation is of primary importance.

Open for discussion until I August 1999

mailto:[email protected]

64 Slobodan Petkovic et al.

COUNTRYWIDE OVERVIEW OF EROSION PROCESSES

The Serbian territory is in the central part of the Balkan Peninsula and extends over an area of 88 000 km2. This territory consists of two different regions—the large Vojvodina plain to the north and the hilly and mountainous area to the south with the Danube and Sava rivers representing the border between them. Geomorphological features of the Serbian territory are strongly related to the soil erosion problems: wind erosion predominating in the northern plain, and water erosion in the southern region. It should be stated that the wind erosion is not considered in this paper and the further review of erosion problems relates to the erosion of soil by water.

The natural conditions in the southern part of the Serbian territory are favourable for the development of erosion processes. All three groups of erosion factors—energy, resistance and protection (Morgan, 1986)—promote the soil erosion. The energy group includes the ability of rainfall and runoff to cause erosion. The relief in the southern part of Serbia is characterized by relatively steep slopes, which directly influence the power of the erosive agents. Considering the resistance factors, the significant erodibility of the soil and geological substrata should be mentioned. The geological structure of the major part of the considered area consists of rocks of high erodibility (conglomerates, schists, etc.) which contribute to the denudation processes. Resistant rocks (granites, andezites, etc.) are present in a smaller area of this region.

The protection group of erosion factors is related to the population density and land use. The average population density in Serbia is moderate—about 100 inhabitants per km2. The urban and rural populations are almost equal. Considering the land use in the southern part of Serbia, it should be stated that forest areas cover 27% of the territory, while cultivated land represents about 60% of this region. The cultivated land comprises about 50% arable land, 40% grasslands and pastures and 10% other uses.

The meteorological and hydrological conditions in the southern part of Serbia are heterogeneous. The mean annual precipitation ranges from 600 to 850 mm and the runoff from 5 to 15 1 s4 km2. The rainfall intensity is variable and can be very high: the maximum daily precipitation can exceed 100 mm, while the maximum intensity is about 80 mm h' . The values of the rainfall erosivity index (r) in the USLE equation ranges from 50 to 150 within the Serbian territory.

In order to alleviate the erosion problems in Serbia, soil conservation and erosion control measures were implemented in the most affected areas. These measures included the standard agronomic methods of soil and crop management (contour tillage, mulching, etc.) and mechanical methods (terracing, grass waterways and structures). In addition, torrent control measures have been implemented, in order to prevent excessive sediment transport and reservoir siltation. From this point of view, check dams are the most effective structures.

Soil erosion surveys in Serbia consist of mapping and aerial photographs. This method permits the countrywide erosion hazard assessment. The general erosion map of the Serbian territory is shown in Fig. 1. The whole territory is divided into three basic units corresponding to the high, medium and low erosion intensities. The northern region of Vojvodina plain belongs to the low erosion area (except the small

Erosion and sedimentation problems in Serbia 65

HUNGARY

ROMANIA

CROATIA»!

1 ^ K v-f fi (

U3 LOW

Sm MEDIUM

• i HIGH

0 50 100!

BOSNIA

MONTENEGRO

BULGARIA

ALBANIA MACEDONIA

Fig. 1 Erosion map of Serbian territory.


hilly region). The river valleys in southern part of Serbia are also characterized by low erosion. The total area of low erosion intensity represents about 60% of Serbian territory.

A medium erosion rate mostly corresponds to the hilly regions, extending to about 25% of Serbia. Finally, the area of high erosion corresponds to the mountainous region, with the steep relief, extending to 15% of the territory. The highest erosion intensity occurs on bare soil and in areas of deforestation. The design of erosion control measures is based on detailed surveys and mapping.

Estimates of erosion and soil loss rates for Serbian territory are based on empirical modelling and field measurements. The empirical formula for estimating the sediment yield is derived from field measurements of sediment transport in rivers and siltation rates in reservoirs (Petkovic, 1993). This formula is based on the relationship between sediment yield and the meteorological, hydrological and géomorphologie factors for this area, in the following expression:

.i-O.I rO.6 Tj-0.5 770,7 , , . .

g = a- A • 1 • H • E (1) where g is the annual sediment yield of the river (m3 km"2 year1), A is the watershed area (km2), / is the géomorphologie factor, H is the hydrological factor and E is the erosion factor. The géomorphologie factor, / is expressed in terms of mean slope of watershed, Sw and river channel slope, SR:

I - yJ^w^R (2)

The hydrological factor, H is expressed in terms of unit area runoff in the watershed, q (m3 km"2 s"1), and the ratio (ex) of the amount of daily precipitation over 20 mm and total annual rainfall, PA:

I>>20) H = qa = q^~-p (3)

"A

The erosion factor, E depends on the state of erosion processes in the catchment area, expressed by the index E0 (ranging from 0.03, for very low erosion to 0.3, for excessive erosion) and the ratio e = AEIA (where AE is the area of high erosion):

E = E0+e (4)

The value of the constant (a) should be determined on the basis of regression analysis of available data on g, A, I, H and E. For Serbia, the value of a is 1.3 x 105. Part of the database, concerning reservoir siltation, is presented in Table 1. The calculation of unit-area sediment yield of some Serbian rivers, for given parameter values, using the described empirical formula, is shown in Table 2. The comparison between measured and calculated values indicates reasonable agreement.

Field measurements of soil erosion in Serbia are carried out at two experimental stations. The work at experimental stations is based on standard erosion plots, of known area, slope steepness and length, soil type and vegetation cover, from which both runoff and soil losses are monitored. The results of measurements of soil loss at one experimental station in Serbia are shown in Figs 2 and 3. These Figures show


15 m

ent

x km

-am

as in oo ro ,—t

cs <n o •** •n as o —« CS r-1 OS CS

O T3 w

Ilè§.

I K

O eO O O O O

o oo so o o

O ^ SO CO O VO oo y ) h \o i n h OS OS OS OS C7\ OS

»-H u-) r - co oo so tr~~ "ri MD in co o o\ ON o\ C\ o\ U\

i-H oo so r~- o oo o — —* t-H co o

o o o o cs q o o © o o o

o 00 CS O

O O O O OS SD

o o es o

o o •*t

IT)

\~~~ IT)

O OO T—l

o CO

O oo co co

Is-Jâ-J J

« • = " >

I § S. •§ g « aj > o 43 -p -§ G N P- 2 ,E g

| B M W W S > S SS X X O O

5 "3 >

• a

a, S o U

S R

ïï bj

ft c « o

o

O o o

oo o

(M O

u-i

*3-

o

o

oo

o

Cl

-3-o o

o

oo o o

o

•st"

o

o

o o o

o M faj

S ^

X &i

c o

xs

.O T î

^ G

a a,)

<+-« O (3

amet

& « o o

J3 ft 010

o Cl)

a

i o

* ^

^

O O O O O O OO O --H o s co "ri

o o «n •<*•

" * CS ^ - H t-f CO O

o o o o o o o

o o o o o o o o

m (N r i - N M

o o o o o o

r - r - co o es C S —< "~~< C O -sf

o o o o o o o o o o o o o o o O Tt CS O CS ro es es co co o o o o o

i n o o o o s r ^ s o o ^ O O ^ O O O *—' t-H q q q o o o o o o o o o o o o o

o -* co tn o

o o o o o o

o o o o o o o o o o o o

O'* o oc o

o o o o o o o o

SO O Os CS —< Os O —< co *3- ^ ^ i> oo

co o s e s oo —<

# U >M 'J5

O « O —— J) S B ^


the relationship between soil loss and precipitation amount (for a single event), for different types of vegetation cover. This relationship for four different types of vegetation cover (bare soil, arable land, grassland and forest) and for given plot steepness (24%) is presented in Fig, 2. It is evident that the considered range of rainfall depth, Hr is 10-80 mm, as the soil loss was not observed for rainfall events with H < 10 mm. The measured values of soil loss, E, covered a very wide range from 10"4 to 10 kg nï2. Considering the effect of vegetation cover, the minimum soil loss, ranging from 0.0001 to 0.035 kg m~2, can be related to the forest area. The observed range of E for the grassland is from 0.0005 to 0.28 kg m"2. The monitored values of soil loss for arable land and bare soil are much larger: for arable land, .Evalues range from 0.01 to 4.80 kg m2, while for the bare soil they are 0.04-7.40 kg va2.

The effect of plot steepness on rates of soil loss is shown in Fig. 3. The relationship between Hr and E is given for two different slopes (8 and 24°) and two different types of vegetation cover (grassland and bare soil). It can be seen that the

1

"

J,

1

I

I

_____

1

_

T

I

i

1

-I

i

T

x

10 100 Hr (mm)

a bare soil 4 arable land j grassland j forest

Fig. 2 Relationship between soil loss, E and precipitation height, Hr for different types of vegetation cover.


effect of plot steepness is much greater for grassland—the E values are roughly ten times larger for the 24° slope than for the 8° slope. However, the ratio for the bare soil is only 2:1. Consequently, it can be seen that the relationship between runoff and infiltration and processes of erosion and sediment transport by overland flow are significantly influenced by the presence of vegetation cover. It should be stated that the dispersion of experimental points in Figs 2 and 3 reflects the complexity of phenomena. The relationship between H and E is influenced by the rainfall intensity, which has not been measured.

The results of erosion measurements at experimental stations in Serbia can be expressed in terms of average annual soil loss EAN (in t km"2 year4). The values of EAN are 2500-6500 t km"2 year1 for bare soil, 1200-3300 t km"2 year"1 for arable land, 50-220 t km"2 year"1 for grassland, and 15-120 t km"2 year"1 for forest. Published data (Morgan, 1986) show that the annual soil loss for bare soil in the UK and Belgium are 1000-4500, and 700-8200 t km"2 year"\ land are 10-300 and 300-3000 t km"2 year"1,

respectively, and those for cultivated respectively. It can be concluded that

the observed EAN values at the experimental stations in Serbia generally correspond to the range of erosion rates elsewhere in Europe. Data published in Zachar (1982)

10

0.1

0.01

0.001

•

r 1—I

I 11

- — I

— — — —i -i — H • y - I — 1

i

e • ! | t ;

- « = ^ = n = — = = : • ~ '~ I « i 1 1 ~T

e 1 L, !

»

f

f i

? — ^ ~- ^—^ :

? j

j

i 1 1 1 1 1 , 0.0001 10 Hr(mm)

B bare soil 24 T grassland 24 f b a r e s o y g

Fig. 3 Relationship between soil loss, E and precipitation height, H, for different types of vegetation cover and different plot slope.

100

grassland 8


indicate a similar range of EAN. For arable land in the European part of Russia, observed EAN values are 1000-10000 t km2 year1, while in France they are 1500-4500 t km"2 year1.

Estimation of total annual erosion rate for the Serbian territory is very difficult, because of the complexity of erosion processes and the diversity of natural conditions. However, it might be possible to estimate this value on the basis of erosion mapping, field investigations and empirical modelling. An evaluation of total annual soil loss from the Serbian territory indicates a range of 40-50 x 1061 km"2.

SEDIMENT TRANSPORT IN SERBIAN RIVERS

The river network in the region of Serbia is relatively dense and includes a large number of water courses of different sizes, ranging from small creeks to very big rivers such as the Danube. The greatest part of the Serbian territory belongs to the Danube drainage basin. The most important tributaries of the Danube in Serbia are the Tisza, Sava and Velika Morava rivers. Figure 4 shows the basic river systems in Serbia.

Investigation of sediment transport in Serbian rivers is of increasing significance in the recent period, because of the serious sedimentation problems. Consequently, sediment measurements are carried out on all rivers which are important from the point of view of water engineering. However, the majority of measurements are related to suspended sediment discharge. Bed load measurements, because of their complexity, are carried out only on few rivers in Serbia.

Sediment monitoring methodology is standard and consists of daily sampling of water and sediment, with laboratory determination of the sediment concentration. The technique of sampling depends on the stream size—for large rivers, a special type of pumping sampler (with a 40 1 bottle) has been developed; the sampling of suspended sediment in smaller rivers is based on a simple bottle (1-5 1) sampler.

Daily measurements of suspended sediment concentration and daily recording of water level permit determination of the suspended sediment discharge for a given time period (using the water discharge rating curve). Thus, the annual suspended sediment discharge is known for a large number of Serbian rivers. Considering a longer time period (which depends on available time series for each station), the average value of annual suspended sediment transport can be determined. The distribution of annual suspended sediment transport over the Serbian territory is shown in Fig. 4. It can be stated that the greater part of the total sediment transport originates from outside the country. The sediment input of the Danube and Tisza rivers (from Hungarian territory) is about 11 x 106 t, while the input of the River Sava (from Croatia) is about 3 x 106 t. Annual suspended sediment transport of the internal rivers is about 9 x 1061.

The influence of the Iron Gate I reservoir on the River Danube on sedimentation processes in Serbia should be emphasized. This reservoir traps about 80% of the total sediment transport of the River Danube and it is probably the greatest sediment storage reservoir in Europe.


Fig. 4 Annual suspended sediment transport in the hydrographie network of Serbia.

According to the sediment management concept, the interdependence between erosion and sedimentation phenomena should be investigated. It is well known that sediment transport in rivers is related to the sediment yield from corresponding watersheds. The sediment transport rate in a given river can be expressed in terms of unit-area sediment yield. On the basis of field measurements of sediment transport in rivers and of the siltation of reservoirs, the correlation between the annual sediment yield, g (m3 km2 year*1) and watershed area, A (km2) can be established (Fig. 5).

The correlation between g and A in Fig. 5 shows the inverse relationship between these parameters: an increase of A corresponds to a decrease of g. Lower values of


g (m3-km"2-yeaï'1)

10000

1000

100

10

10 100 1000 10000 100000 A (km2)

Fig . 5 Relationship between sediment yield and watershed area.

iO2

io°-

,U ) 0 - i ]Q-3 tO"2 IO' ' 10° 10' 0 102

Fig . 6 Theoretical interpretation of bed load measurements .

sediment yield are expected in larger river basins, due to the decrease of sediment delivery ratio with an increase of A. The correlation between g and A is characterized by a significant dispersion of points, caused by the complexity of erosion and sedimentation phenomena. Consequently, two envelope curves have been drawn on Fig. 5: the upper envelope corresponds to the river basins in the area of high erosion, while the lower envelope corresponds to the catchments of low erosion. For small watersheds (A = 100 km2), g ranges from 100 to 1000 m3knr2 year"1, while for large river basins {A = 10 000 km2) it ranges from 40 to 400 m3km~2 year1.

From field measurements on some rivers, the total annual bed load transport is estimated to be about 10% of suspended sediment transport. The results of bed load


measurements in two Serbian rivers are presented in Table 3, These results are also presented in Fig. 6, in the standard hydraulic relationship between the dimensionless bed load transport parameters (§ and \\i). Given a value of 9 x 106 t of annual suspended load transport in internal rivers in Serbia, the corresponding value of bed load transport is about 1 x 106 t. Regardless of the smaller quantity of the bed load, its role is very important in river engineering, from the point of view of morphological evolution of rivers and siltation of reservoirs. Furthermore, the bed load is very significant from the sediment management viewpoint, particularly in dredging activities for civil engineering purposes.

Table 3 Hydraulic and bed load transport summary.

Measured parameters: Q S H Dm

Computed parameters:

River V 53

298 342 424 438 528 950

1090

'elika Mor, 0.03 0.05 0.02 0.06 0.06 0.02 0.06 0.05

ava, location 1.49 2.58 5.38 2.68 2.39 5.84 6.79 6.46

.1(1 km 40 73 63

107 116 100 90 84

from the m 0.32 1.23 0.35 1.91 1.70 0.37 2.17 2.02

River Velika Morava, location II (22 km from the

310 530 709

0.15 0.16 0.17

2.60 3.50 4.13

70 90

100

River Drina (5 km from the mouth):

199 206 266 268 446 542 707 773 900

0.13 0.13 0.19 0.19 0.28 0.41 0.47 0.53 0.69

2.04 1.97 3.01 2.77 2.91 2.31 2.52 3.06 2.70

100 100 120 120 140 150 160 170 180

2.50 3.10 3.50

16.10 16.50 12.00 4.30

25.20 19.90 24.40 22.30 21.70

outh): 0.001 0.033 0.003 0.014 0.014 0.010 0.21 0.023

mouth): 0.140 0.189 0.340

0.001 0.001 0.001 0.002 0.001 0.024 0.036 0.175 0.248

0.03 2.45 0.18 1.56 1.60 0.99 1.94 1.95

9.80 17.20 34.20

0.06 0.02 0.01 0.20 0.10 3.65 5.80

29.80 44.80

0.44 1.27 1.05 1.59 1.42 1.06 3.41 3.16

3.82 5.60 6.80

2.60 2.60 5.59 5.15 74.95 9.50

11.60 16.10 18.36

12.1 16.2 5.6

20.4 20.3 5.80

10.6 10.7

11.0 9.20 8.6

97.0 95.0 34.0 14.2 52.0 34.9 35.1 22.9 19.7

0.015 0.068 0.041 0.015 0.018 0.125 0.018 0.022

0.1 0.097 0.145

0.0001 0.0001 0.0001 0.0002 0.0001 0.0003 0.0003 0.0046 0.0024

Q: water discharge (m s ); S: water surface slope (m km"1); H: mean depth (m); BA: active channel width (m); Dm: mean grain diameter (mm); gB: unit bed load discharge (kg s"1

GB. bed load discharge (kg s"1).

pgHS is the shear stress (N m"~)

8\P, ~ P D„, - is the shear intensity parameter

m'1);

[(ps/ p - l)g - is the sediment transport parameter

D'

As far as sediment budget in Serbian rivers is concerned, there is a significant difference between the contribution of the bed load and suspended load. The suspended sediment budget is characterized by the continuity of sediment transport in the hydrographie network (from the eroded area, through the small creeks and


greater streams, to the River Danube), However, the situation with bed load is quite different—there is no continuity of transport, because of varying hydraulic conditions and transport capacities of rivers. This discontinuity of the bed load transport occurs at the microscale (in several river basins), but also at the macroscale of the Serbian hydrographie network.

The largest internal rivers in Serbia (Fig. 4) are the River Velika Morava (beginning below the confluence of the Juzna Morava and the Zapadna Morava rivers) and the River Drina (on the border between Serbia and Bosnia). These rivers are the final recipients of bed load transport from tributaries over the largest part of the Serbian territory. The Velika Morava and Drina are typical gravel bed rivers, their river beds consisting of about 80% gravel (Fig. 7). The bed load measurements in these rivers have been carried out over a period of 10 years. From the point of view of continuity of bed load transport, the coarser load (corresponding mostly to the gravel) is of particular interest. The average annual coarser bed load transport in the River Velika Morava is about 350 x 103 t. The inputs of the Juzna and Zapadna Morava are 210 x 103 and 120 x 103 t respectively (and the rest is the input of smaller tributaries).

The continuity of sediment transport along the River Drina is strongly affected by the presence of two dams and reservoirs (Zvornik and Bajina Basta). These structures stop bed load transport. Downstream of these dams, the bed load transport is restored, because of fluvial erosion. It should be emphasized that the bank erosion is very intensive on the River Drina, particularly downstream of the Zvornik Dam. According to the bed load measurements, the average annual coarser load (gravel) transport upstream of the Bajina Basta Dam is 370 x 103 t, on the reach between

GRAVEL

\ \ \ \

\ \

\ \ \ \ \ \ \ \

DRINA \

1

" " - ^

SAVA '

vy VELIKA N

\ , \ \ \

\ \ \ \

SAND

^

IORA

DANUBE \

\ \ \ \ \

/A ,

\ \ \ \ \ \

V

\

200 100 10 1 0 1 GRAIN SIZE (mm)

Fig. 7 Grain size distribution of bed material.


SAVA 50 ! SAVA;,

"IRON GATE I" DAM

DANUBE

A

v..

"ZVORNIK" DAM

180

BORDERS OF BACKWATER ZONE OF "IRON GATE I" DAM

BAJINA BASTA" DAM

370

î"\ Fig. 8 Sketch of annual budget (1031 year1) of coarser bed load (gravel) in the main river system in Serbia.

Bajina Basta and Zvornik dams it is 180 x 1031 and downstream of the Zvornik Dam it is 270 x 1031 (Fig. 8).

The Danube is a large alluvial river, with a sandy bed (Fig. 7). After the construction of the Iron Gate I Dam, a large backwater zone was created, extending 300 km along the River Danube. The tributaries are also affected—the backwater extends 100 km up the River Sava and 25 km up the River Velika Morava. The hydraulic effect of the Iron Gate I backwater, from the point of view of decrease of shear stress and bed load transport capacity, is much greater on the Velika Morava than on the Sava. Consequently, since the construction of the Iron Gate I Dam, hydraulic conditions no longer permit transport of the coarser bed load along the backwater zone of the Velika Morava. Thus, the backwater zone on the Velika Morava represents a deposition area for the gravel, transported from the upper reach of the river.

The input of coarser bed load from the Drina into the Sava exceeds the transport capacity of this river, because of the low channel slope and reduced stream power. Consequently, the reach of the Sava downstream of the mouth of the Drina represents a deposition zone for gravel from the Drina. The annual coarser bed load transport along the River Sava is limited to about 50 000 t. The final deposition of the gravel from the River Drina occurs in the backwater zone (from the Iron Gate I Dam) on the River Sava. Thus, the gravel transported by the Drina and Sava rivers cannot reach the River Danube. The grain size distribution of the Sava bed is influenced by the sediment transport conditions. The major part of the bed consist of sand, while the smaller part is gravel (Fig. 7).

A sketch of the annual budget of coarser bed load in the main river system in Serbia is shown in Fig. 8. It can be concluded that the Velika Morava and Drina rivers are the main recipients and transporters of gravel produced within the Serbian territory. However, all of this gravel is deposited in the backwater zone of the River


Velika Morava and in the River Sava, so the gravel transported by the internal Serbian rivers cannot reach the River Danube.

From the engineering point of view, the sedimentation problems in Serbia concern principally river channel stability and reservoir siltation. In terms of channel stability, the main purpose of the river engineering works is related to the adjustment between sediment discharge and transport capacity. Sediment transport equilibrium is necessary, in order to prevent fluvial erosion and sediment storage in the river bed.

Reservoir siltation is a very serious problem in Serbia. Several reservoirs within the Serbian territory, such as, for example, the Zvornik reservoir on the River Drina, have lost a significant part of their storage volume because of sedimentation. The Iron Gate I reservoir traps about 15 x 106 t of River Danube sediment every year. However, because of the very large volume of this reservoir (about 2 x 109 m3), sedimentation has not yet significantly affected its water storage capacity. It should be emphasized that in recent years extensive erosion control measures have been implemented over the Serbian territory, in order to reduce reservoir siltation.

CONCLUSIONS

Erosion and sedimentation problems in Serbia are very complex. Erosion processes extend over a large region, because natural factors, affecting soil erosion are favourable. However, the extensive erosion control measures have reduced the soil erosion intensity. Consequently, the mean erosion rate over the whole area is moderate.

The most serious erosion problems are related to soil loss from the cultivated land in the hilly and mountainous region of Serbia. The greatest soil loss occurs from the arable land on steep slopes. A possible solution is a change of land use, since the soil loss values of pastures and grasslands are much smaller than on arable land at the same slope. Field investigations of soil loss at erosion plots indicate that the erosion rates in the Serbian territory are similar to those of other European regions.

The greater part of total sediment transport in Serbia originates from outside the country. The mean annual sediment input from the Danube, Tisza and Sava rivers is about 14 x 1061 compared to that from internal rivers of about 9 x 106 t year1. The total annual sediment transport represents about 20-25% of total annual soil loss within the Serbian territory.

Total annual bed load transport in Serbian rivers is estimated to be about 10% of suspended sediment transport in internal rivers, that is about 1 x 1061 year"1. The bed load transport in the main fluvial system in Serbia is characterized by the discontinuity of transport of coarser material (gravel). The Velika Morava and Drina rivers are the principal recipients and transporters of gravel produced within the Serbian territory. All the transported gravel is deposited in the backwater zone of the Iron Gate I Dam, which extends to the Velika Morava and Sava rivers. As a result, the gravel transported by Serbian internal rivers cannot reach the Danube. The grain size distribution of bed material of the Danube, Sava, Drina and Velika Morava rivers closely reflects this discontinuity of coarser bed load transport.


REFERENCES

Morgan, R. P. C. (1986) Soil Erosion and Conservation. Longman, Essex, UK. Petkovic, S. (1993) Analysis of sediment transport in Serbia. In: Cause and Effects of Soil Erosion and Strategies for

Erosion Control. Monograph, Faculty of Forestry, Belgrade, Serbia. Zachar, D. (1982) Soil Erosion, Elsevier, Amsterdam, The Netherlands.

Received 11 April 1997; accepted 11 August 1998

erosion and sedimentation problems in serbia

Documents