Journal of Environmental Science for Sustainable Society, Vol.

1, 47-54, March 2007

RIVER BED EROSION ON THE HUNGARIAN

SECTION OF THE DANUBE
Lszl GODA 1, Bla KALOCSA 2 and Enik Anna TAMS 3
1 Dept. of Hydrology, Lower-Danube Environmental and Water Authority

(Szchenyi u. 2/c. 6500 Baja, Hungary)

E-mail: goda@adukovizig.hu
2 Dept. of Informatics, Lower-Danube Environmental and Water Authority
(Szchenyi u. 2/c. 6500 Baja, Hungary)
E-mail: kalocsa@adukovizig.hu
3 Assistant Professor, Dept. of Hydraulic engineering and Water management, EJC-TF, Baja
(Bajcsy-Zs. E. u. 14. 6500 Baja, Hungary)
E-mail: et@baja.hu

Investigations of the recent decades have revealed a significant lowering of the river bed of the Danube in
Hungary. This phenomenon can be observed on a long stretch of the Hungarian Danube and it can be
presumed that similar changes occur on other parts of the river beyond the borders. There are several
methods to reveal and to measure the lowering of the channel. Long records of the water level are available
for the trend- and homogeneity analyses. Results show that erosion processes started even at the beginning
of the 20th century. In recent years results of bathymetry are available for the more detailed investigations of
these changes. The main causes of the bed-lowering are river regulation activities, industrial dredging and
decreasing of natural sediment load. Deepening of the river bed has unfavourable effects on the natural
environment, on the navigation and also on the operation of man-made structures in the river. To avert
these injurious effects a complex solution is needed, taking into consideration not just economical but
ecological and social aspects as well.
Key Words : Danube, erosion, river regulation, navigation, sediment, bed load, dredging

approx. 6500 m3/s

Some details about the Hungarian stretch:
- length: 417 km (from 1433 to 1850 river km)
- there are just a few tributaries along the
Hungarian stretch. The growth of the mean
discharge is not more than 10-15 percent
between the entering and leaving sections.
- typical discharges at Baja station:
Qmax: 8023 m3/s, Qmin: 680 m3/s, Qavg.: 2435
m3/s.

1. BACKGROUND
Several investigations of recent decades have
revealed a significant lowering of the river bed of the
Danube in Hungary1,2,). This deepening of the
channel is a result of erosion processes that affect not
discrete sections only but long reaches of the river.
This study is devoted to the Hungarian section of the
Danube but, considering the reasons, it can be
presumed that similar symptoms could be
experienced in other parts of the Danube.

The whole catchment of the Danube is presented by

Fig.1, the Hungarian section is given in Fig.2.
Classifying the Hungarian Danube, it belongs mainly
to the middle course, partly to the lower course with a
lowland type. The river bed is alluvial. In accordance
with the lowland character the slope is mild (10-15
cm/km). The river bed is wide, with a typical width of
400-500 m. The avearage depth for the mean
discharge is 5-6 m.

Some facts about the Danube:

Total length: 2826 km.
Source: Black Forest, Germany. Estuary (delta):
The Black Sea, Ukraine/Romania
The total catchment area: 817 000 km2
18 countries share the catchment. Though 5 of the
18 countries have a very minor part.
The average annual discarge at the delta:
47

Lszl GODA et al.

Fig.1 Geomorphologic map of the Danube catchment. The Hungarian section is highlighted..

A middle course type river usually has a balanced

sediment transport and a balanced channel. The
fact that this is not so nowadays for the Danube
indicates that there were, or are, external effects
disturbing the early balance.

phenomenon,
consequences.

the

arising

problems

and

2. DISCLOSING THE DEEPENING OF

THE RIVER BED
(1) Hydrological-statistical methods
Regular hydrological observations started in the
second half of the 19th century in Hungary. It
started first with observation of the water level at
many stations along the Danube. Somewhat later,
in the beginning of the 20th century, regular
discharge measurements also began. Nowadays
there are water level records available for more
then 100 years, and discharge records for 80-90
years for many stations on the Danube.
The trend analysis performed for the annual
extremes and averages of these long records of
water levels indicates a significant falling trend of
the water levels. An example is shown in Fig.3. A
similar symptom is demonstrated by the
Kolmogorov-Smirnov type homogeneity test
(Fig.4.). These investigations are also suitable to
estimate the degree of deepening: taking the station
of the upper example (Baja) and focusing on the
annual low waters our rough estimation can be 160
cm for the analysed 105 years.

Fig.2 Hungarian section of the Danube and the main gauges

This paper intends to give an overview of the

tools to disclose and measure the deepening of the
Danube, to present the possible causes of this
48

RIVER BED EROSION ON THE HUNGARIAN SECTION OF THE DANUBE

Trend analysis

water level
cm

annual maxima

the flow regime, that is, the decrease of the natural

water resources of the river. This question can be
answered by performing the same kind of analyses
for the discharge records: the trend analysis shows
that a significant trend cannot be found in the
records of the discharges (Fig.5.).

Duna - Baja
1901 - 2006

900
800
700
600

Annual means

500

Trend analysis

discharge
m3/s

annual minima

400

annual maxima

Duna- Baja
1930 - 2005

300

7000
200

6000

100
1910

1920

1930

1940

1950

1960

1970

1980

1990

2000

5000

Time [year]

Fig.3 Falling trend of the annual extremes and means of the

water level (Baja station)

4000

The question can arise, if the observed change of

the water levels indicates really the change of the
river bed or this symptom is a result of a change in

2000

Annual means

3000
annual minima

1000
1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

Time [year]

Fig.5 Trend analysis of the discharges. There is no significant

trend in the discharge records (Baja station).

Similar investigations have been performed for

all the main stations on the Danube in Hungary and
the results were more or less the same. These
results are summarized in Fig.6 where the
tendencies of the 15-year averages of the water
levels are presented for the main stations along the
river3).
Fig.4 Szmirnov-Kolmogorov-type homogeneity test for the
annual minima of water levels (Baja station).

-50
-74

-150
-123

-200

Komrom

Esztergom

Nagymaros

Fig.6 Decrease of the 15-year averages of the water levels between 1901-2005 at the
gauging stations along the Danube in Hungary

49

-100

-250

Budapest

Dunajvros

-216
Dunafldvr

Paks

-181

-181

-99

Vc

-151

-176

Dombori

-156

Baja

Mohcs

1901-15
1916-30
1931-45
1946-60
1961-75
1976-90
1991-2005

-145

-127

Ercsi

-130

Adony

-81

-70

Deepening, cm

Lszl GODA et al.

92
91

E le v a tion a bov e the s e a le v e l, m

(2) Morphological methods

The morphological survey of the
river bed (bathymetry) is a more
direct and exact method to follow the
changes of the bottom. However,
adequate and sufficient results are
available only from the latest
decades, following the appearance of
the GPS, the echograph and GIS
applications. Nowadays bathymetry
is done every 5-6 years. Comparing
the result of more surveys, done in
different periods for the same spot,
the changing of the river bed can be
directly analyzed. These comparing
results of a typical section (near
Dunafldvr) are illustrated in Fig.7.
The longitudinal profiles of the
bottom of the channel line are shown
in Fig.8.

Changes of the river bed profile near Dunafldvr

( 1559 + 400 river km )

90
89
88

1949

87

1975

86

1981

1998
85
84

2003
83
82
-300

-200

-100

100

200

300

Distance, m

Fig.7 Changes of a profile of the river bed near station Dunafldvr.

Fig.8 Difference map of a Danube section near Dunafldvr

from the surveys in1981 and 2003

3. CAUSES OF LOWERING OF THE

RIVER BED
During medium-flow regulation strongly
meandering bends were cut, the horizontal
alignment of the river was fixed with bank
protection structures and groynes, utilizing also the
energy of the flow to form the bed. The aim of this
work was to speed up the travelling of floods and to
help the movement of drifting ice. The improved
shape of the channel was favourable for navigation
as well because the narrowed river bed resulted in
deeper water for longer duration. However this
transformation resulted in shortening of the river,
increasing the slope and therefore increasing the
sediment transportation capacity. Consequently the
original balance has been shifted towards sediment

(1) River regulation

River regulation works were started in the 19th
century and are still going on along the Hungarian
Danube4). The aim of the early activities was flood
protection and to improve conditions for
navigation. In the course of high-flow regulation
lowland areas along the river were protected from
inundations. The first dikes had a local role, later
they were followed by higher levees, stretching for
a long distance along the river.
Driven between the levees, floods are not able to
spread as they did in the natural, original state
and have a stronger effect on the bed forming
processes.
50

RIVER BED EROSION ON THE HUNGARIAN SECTION OF THE DANUBE

erosion. Narrowing of the river bed also increased

flow velocities, involving the increase of the
sediment transportation capacity again. An old
map from the the turn of the 19th and 20th centuries
illustrates the result of the river regulation work
(Fig.9).

(2) Dredging
Gravel and sand, being dredged from the river, are
valuable raw materials for construction industry.
Therefore above the requirements of low-flow
regulation huge masses of sediment were mined
from the Danube between 1960 and 1990.
According to well-founded estimations, about 70
million m3 of bed material was excavated from the
Danube during the period mentioned2,5). Taking the
affected stretch of the river and an average width of
the bed into consideration, this removed mass
would be enough to cause about 60 cm deepening
of the bed. Of course the effect of dredging can not
be projected onto the river so simply, but obviously
the dredged holes are more or less filled up by the
bed material of the upstream section, causing the
deepening of the channel. Another unfavourable
effect of dredging is that it breaks up the harder,
armoured surface of the bed. The current erodes
the upstream edge of the dredging hole and the
looser layers below will also be accessible for the
energy of the flow.
Because of the problems mentioned above it was
necessary to decrease the dredged amounts as small
as possible. Nowadays industrial dredging is not
allowed. Only river regulation and maintenance of
the navigation way can be the reasons for
dredging.
(3) Decrease of transported sediment
Numerous barrages and reservoirs were built on
the Danube and its tributaries in the second half of
the 20th century6). Regarding the Hungarian
section, first of all the barrages in Germany and
Austria have an essential importance. The barrages
on the Upper Danube are illustrated by a schematic
longitudinal profile in Fig.10.
The flow regime of the Danube was not
considerably influenced by these structures, but the
flow conditions in the channel have essentially
changed. The speed of the flow decreases
significantly in the backwater of a barrage and it
causes the bed load to stop and the 30-40 % of the
suspended load to settle. At the same time,
sediment transport capacity of the water increases
on the downstream of the barrage causing an
increased erosion of the downstream channel
(unless the backwater of the next barrage on the
downstream reaches the section in question).

Fig.9 Cut-offs on the Danube-section upstream of Baja on

the turn of the 19th-20th centuries. (Nowadays the
Danube flows in the strait cuts while the meandering
bends became narrow dead branches.)

Low-flow regulation is an indispensable auxiliary

element of medium-flow regulation. It
concentrates on the stretches of the river where
medium-flow regulation was not enough to
develop the required profile, water is shallow and
fords obstruct the navigation. Additional
regulation structures and direct dredging is applied
on these stretches to get the required measures of
the navigation profile. However, dredging helps
the deepening processes, in more respects, as it is
given below.
51

Lszl GODA et al.

Fig.8 Barrages on the Upper-Danube (Germany, Austria, Slovakia).

However the amount of the eroded material does

not reach that of the sedimented masses. So the
result of the process is the decrease of sediment
concentration, causing a higher sediment transport
capacity on the Hungarian section7).

despite of the extensive regulation dredging, the

conditions of navigation did not improve
significantly in the last decades5).
The navigational low water levels (94 %
probability, ice-free stages on the main gauges)
are calculated by the Danube Comission from time
to time. Fig.9 shows how the logitudinal profile
plotted for the navigational low water levels
follow the lowering of the river bed.

4. EFFECTS AND PROBLEMS

Deepening of the river bed has unfavourable
effects on natural environment, on navigation and
also on operation of man-made structures in the
river. Some typical examples are outlined below.

(b)
In 1983, during an extreme low water
period, there were troubles with the cooling water
of the Paks nuclear power plant, which is supplied
gravitationally from the Danube (the pumps of the
power plant hardly reached the water in the
supplying channel). Later the reach of the river
near Paks was controlled successfully with river
regulation methods and similar problems have not
arisen since then. Nevertheless, nowadays some
deepening of the channel can be observed in the
vicinity of Paks as well. Results of numerical
modelling predict a continuation of this process for
the following decades.
An other fact, calling for a special regard for this
area is that the fords downstream of Paks, acting as
bars, have a significant role keeping the low water
on the required level for the water intake. On the
other hand, these fords represent awkward
obstructions for navigation during low water
periods8).

(a)
To improve the conditions of navigation
is an important aim of river regulation work.
Medium-flow regulation activities have been more
or less finished by nowadays. Taking into
consideration that the horizontal alignment of the
Danube is practically stabilized by cutting many
bends, building bank protection structures and
groynes this work has been accomplished.
However, the bed of the river is still continuously
developing and forming. These changes are at
least partly consequences of earlier interventions.
The most frequent navigation obstructions are
fords, shallows and contractions of the navigation
channel during low water periods. Fords are
continuously building and forming in the changing,
unstable channel calling for a permanent control
of the responsible authorities. More studies,
dealing with the analysis of low water periods and
the efficiency of low water regulation prove that,
52

RIVER BED EROSION ON THE HUNGARIAN SECTION OF THE DANUBE

98

84

Paks

Dunajvros

86

Baja

88

Dunafldvr

Dombori

90

1966
1976
1984
1990
2004

82
80
1440

1490

1540

Ercsi

Adony

92

Mohcs

Elevation above sea level, m

94

Budapest

Navigation low-flow profiles on the Danube

downstream of Budapest

96

1590

1640

Distance, river km
Fig.9

Navigation low flow profiles on the Danube downstream of Budapest.

low in the river, pumping is necessary to lift the

water into the channels. In this system, developed
primarily for gravitational operation, pumping
entails high economical burden. This problem
appeared one-two times annually in the recent
years.

(c) The Gemenc Forest (part of the Danube-Drava

National Park, Hungary) is one of the substantial
floodplain forests of Europe. Regarding the many
wetlands, providing a home for rich flora and
fauna, Gemenc is irrecoverable natural value. The
water demand of the forest and the wetlands is
supplied by the Danube, partly by subsurface
feeding, partly through the net of oxbows and
channels, sometimes inundating the whole area
during floods.
Forestry experts observed certain drying
processes of the forests in the second half of the
80th-s. They noticed changes in the development of
trees and also the replacement of autochton plant
species. It was revealed by the detailed
hydrological analysis of the water levels that the
flood events reaching the oxbow lakes become
rarer and their duration decreased too. Besides the
low water periods with longer duration are also
unfavourable regarding the subsurface feeding.
Some projects started after the investigations
aiming at the improvement of water supply and
water retention in the oxbow lakes9).

5. CONSEQUENCES
The lowering of the river bed of the Danube is a
complex problem as it has been outlined above. It
has far-reaching effects which require complex
solutions, taking into consideration not just
economical but ecological and social aspects as
well. The most important aims are as follows:
- Improvement of navigation conditions regarding
European standards.
- Provision for the safe operation of water intakes.
- Provision for the water supply of nature
conservation areas and other wetlands, ecological
rehabilitation of oxbow lakes and other dead
branches.

However the effects of these interventions extend

only to small areas and, regarding the whole of the
forest, they can not be appreciated as overall and
accomplished solutions.

Further requirements while achieving

above-mentioned aims:
- Improvement of water quality
- Increasing of flood safety
- Development of recreation and tourism.

(d) The Danube has an essential role in the

agricultural water supply along the river. Large
areas are supplied by artificial, gravitationally
operated channels. When the water level is very

the

In recent years there have been several studies

dealing with the deepening of the river bed, the
53

Lszl GODA et al.

problems arising from this and the possible

solutions. Even nowadays there are several such
projects underway searching for solutions. An
essential question is whether traditional river
regulation means are suitable to reach the above
mentioned aims. Building barrages, that is to say
the canalization of the problematic sections of the
Danube, could be an evident alternative for
regulation structures and dredging. For this, as has
already been mentioned, there are a number of
examples on the Austrian and German strech of the
Danube, where the effect of these structures can
also be examined. But at present the damming of
the Danube is not supported by the Hungarian
government.
The problem of the deepening of the river bed is
closely connected with the question of sediment
transport. The studies dealing with sedimentation
bring our attention to the fact that the used data are
often incomplete and sometimes only estimates.
The reason for this is that the measurement of the
sediment (sampling and laboratory analysis) is a
complex, time consuming and costly process.
Sampling requires a high precison on the part of the
measuring staff because without this the value of
the costly results will be doubtful. During floods
sampling is difficult and sometimes inpossible.
Comparative analysis of the results of bilateral
measurements of neighbouring counries show that
the results can be significantly different even if the
measuring sections are relatively close to each
other. The reasons for this are to be found in
different methods and equipment used.
Despite of the problems we have to state that
knowledge of the sediment stransport is essential to
understand the morphological processes forming
the river bed.
Consequently the improvement of the
methodology seems to be necessary both for the
suspended and also for the bed load. Not just
measuring techniques but also data processing
methods should be improved. International
cooperation would be highly desirable to

harmonize the methodology applied in different

counties along the Danube.
ACKNOWLEDGMENT: The authors thank to
the Environmental and Water Authorities at Gyr,
Budapest and Baja and also to the VITUKI for
supplying us with the necessary data to achieve the
analyses presented in this study.

REFERENCES

1) Kalocsa, B., Zsuffa, I.: A Duna Magyar szakasznak

vzlls vltozsai (Water level changes of the Hungarian
section of the Danube), HK (Journal of the Hungarian
Hydrological Society) 1997/3-4., Budapest.
2) Goda, L. (snr.): A Duna gzli Pozsony s Mohcs kztt.
(Shallows of the River Danube between Pozsony/
Bratislava and Mohcs), VK (Hydraulic Engineering)
1995/1-2., Budapest.
3) Tams, E. A.: Navigare necesse est avagy a Duna, mint
ftvonal (Navigare necesse est or, the Danube like a
highway), in press, 2006.
4) Stancikov, A: A Duna szablyozsa (Regulation of the
River Danube), VK (Hydraulic Engineering) 2001/3.,
Budapest
5) Goda, L (snr.): A Duna gzli a folyam Szap - Mohcs
kztti szakaszn. (Shallows of the River Danube between
Szap and Mohcs) manuscript, 2006.
6) Pasoi, I., et all.: Inventory of the main hydraulic structures
in the Danube Basin. The Danube and its catchment A
hydrological monograph. Follow-up volume VIII/1.
Bucharest, 2004.
7) Rkczi, L.: A Duna hordalkjrsa (Sediment load
conditions of River Danube), VK (Hydraulic Engineering)
1993/2., Budapest
8) Laczay, I.: A paksi Duna-szakasz medervltozsnak
ellenrzse (Review of the channel changes of the Danube
reach of Paks), VK (Hydraulic Engineering) 1996/3-4.,
Budapest
9) Szlvik, L., Sziebert, J., Zellei, L., Zsuffa, I.: A
Nyki-Holt-Duna rehabilitcija (Rehabilitation of the
Nyki Oxbow-lake of the River Danube) Vzgyi
Kzlemnyek (Hydraulic Engineering), 1995/3-4.,
Budapest.
(Received February 21, 2007)

54