The Danube: Morphology, Evolution and

67 downloads 0 Views 2MB Size Report
6279(14)60006-8. Bachmeier, U. (2017) 25 Jahre Rhein-Main-Donau-Kanal: Ein gescheitertes Jahrhundertprojekt. Augsburger Allgemeine, 08.09.2017.
The Danube: Morphology, Evolution and Environmental Issues 12 Dénes Lóczy Department of Physical and Environmental Geography, Institute of Geography, University of Pécs, Ifjúság útja 6, H-7624 Pécs, Hungary

12.1 INTRODUCTION The Danube, the second longest river (2,860 km) in Europe, a truly international river, is known by various names: Donau (German), Dunaj (Slovakian), Duna (Hungarian), Dunav (Southern Slavic), Dunarea (Romanian), Dunaj (Ukrainian), Danubius (Latin), Ister (Greek). The name Danubius is of Celtic origin (‘danu’ meaning ‘fast-flowing river’) and probably refers to a river gorge cutting through the Swabian or the Franconian Jura. The Danube is the only major European river that follows a general west to east course and, thus, transports water from high mountains with snow cover in winter and basins of abundant rainfall towards the drier parts of the continent’s humid temperate zone. Its basin (Figure 12.1) extends to 801,463 km2 (ICPDR 2015b), previously estimated at 817,028 km2 area. With 475 m average elevation, the drainage area covers most of the Eastern Alps, the eastern slopes of the Black Forest, the southern slopes of the SwabianFranconian Jura (Alb), the Swabian-Bavarian Basin (Alpine Foreland), the southwestern section of the Bohemian Forest, the minor basins of the Austrian Alpine Foreland, the Vienna Basin, the eastern half of the Czecho-Moravian Hills, the Carpathians, the Carpathian Basin, the Dinaric Range, low mountains in Serbia, the northern slopes of the Balkan Range, the Romanian Plain and the western and northern slopes of the Dobrogea Hills (Sommerwerk et al., 2009). The highest point in the watershed is Piz Bernina (4,049 m) in the Eastern Alps. More than 1,000 km2 of the catchment in the Eastern Alps is glaciated, and glacial meltwater is an important source of discharge for the 15 major right-bank tributaries. There are 20 major left-bank tributaries to the Danube, mostly fed by rainfall and snowmelt. The Danube functions as a reserve for drinking, industrial and irrigation water; a source for generating electricity; an artery for navigation, an area for recreation. It is also an embodiment of scenic beauty and a source of inspiration for writers (Jules Verne’s novel ‘Le pilote du Danube’), poets (the Hungarian Attila József’s philosophical poem ‘By the Danube’), and composers (Johann Strauss’ waltz ‘The Blue Danube’). Most recently the Serbian Zoran Živković has become famous with his tales ‘The Five Wonders of the Danube’. 12.2 WATER AND SEDIMENT The highest mean annual water discharge of the Danube (6,460 m3 s-1) is measured at Galaţi, Romania. The Danube transports 191 km3 of water annually to the Black Sea (6,047 m3 s-1; Malciu, 2000), ranking twenty-second among the rivers of the world

Large Rivers: Geomorphology and Management, Edited by A. Gupta © 2007 John Wiley & Sons, Ltd

(Czaya, 1981). The absolute maximum recorded discharge to the Black Sea was more than 35,000 m3 s-1 in 1897 (DDBRA, 2000). There is a major increase in mean discharge along the Austrian section between the confluence with the Inn (river km 2,230, 625 m3 s1 ) and at Vienna (number 8 in Figure 12.1, 1,920 m3 s-1). Maximum grain size of bedload (150 mm) is also observed at Passau, the confluence with the Inn. An only moderate water input derives from the northern sector of the Carpathian Basin (at Bratislava, number 9, 2,080 m3 s−1, at Budapest, number 12, 2,340 m3 s-1). The discharge of the Danube increases downstream from contributions by the Drava, the Tisza and, particularly, the Sava (at Orşova, number 20, 5,600 m3 s-1). While annual discharges at stations of the upper section are increasing, they are dropping along the Serbian and Romanian reaches. Thus, the stark contrast between the ‘humid’ and ‘arid’ portions of the basin is becoming even more marked (Pistocchi et al., 2015).

Figure 12.1 The Danube catchment. All geographical names mentioned in the text (major tributaries, gauging stations [1–26 as numbered in Table 12.1], the Danubian countries and macroregions) are indicated on this map. Capital letters within circles show countries

Taken as a whole, the Danube cannot be considered a mountain river unlike some of its tributaries. Almost 80 % of its gradient falls to the Upper Danube. Except for the irregularities caused by gorges, the channel has a graded longitudinal profile (Figure 12.2). The pre-dam average bed slope was 0.00025 in the Iron Gate Gorge (Lászlóffy, 1965). However, it drops below 0.00004 along the lowland reaches of the Romanian– Bulgarian border, and 0.00001 for the last 500 km.

Figure 12.2 The longitudinal profile of the Danube with impounded reaches with the elevations of sources of major tributaries. 1,

The river regime mostly reflects the temperate climate of the catchment. Two periods of flooding are usually observed: early spring and early summer (Lóczy and Juhász, 1997). Sudden snowmelt along the Upper Danube may cause ice-jam floods in early spring particularly along the Austrian and Hungarian sections (like in Buda in 1838). Early summer ‘green’ floods (so-called because of the green twigs and branches that are carried by the river) are associated with the precipitation maximum on the Upper Danube catchment, often coinciding in time with higher meltwater outflow from Alpine glaciers. The lowest water stages are usually recorded in autumn and winter, but climate change has recently generated winter floods, too (Caspary, 1995). The channel conditions along the Danube are variable. The hydraulic radius is very large in gorges with narrow channels (particularly in the Iron Gate at Djerdap, where the channel width is only 170 m). It is low in the broad channels of the lowlands (e.g. at the new Romanian-Bulgarian New Europe Bridge of Vidin, where the channel is about 1,300 m wide). At Belgrade, number 17 in Figure 12.1, the channel is almost 1 km wide and before entering gorges upstream at Djerdap (number 20), its width is more than 2 km. The maximum measured river depth is at the damsite of Djerdap (90 m – Radović and Stevanović, 1996). The channel is highly variable along reaches where flow regulations and structures like bridges have strongly modified natural channel shape. At Budapest the river is only 3 m deep at some locations, while at Liberty Bridge it is >10 m. Maximum current velocity is 5–6 m s-1 during floods along the upper section. At Budapest mean current velocity is 0.5 m s-1, amounting to 2.5 m s−1 at exceptionally high water levels. Various estimates exist for the sediment transported in the Danube. The data on sediment load confirm the general lowland character of the river (Table 2.1, Figure 12.3). Suspended load prevails over bedload. The amounts have been reduced by hydroelectric dams constructed on the Danube and its tributaries (Kraus-Kalveit and Pannonhalmi, 2001). Bedload has been reduced to a very small fraction (often a tenth or twentieth) of the values during the 1960s in the upper section (Figure 12.3). Today, suspended solids do not occur in concentrations above 50 mg l−1 (ICPDR, 2002; Babić-Mladenović et al.,

2013). At present, average total load at the apex of the delta is estimated at 67–80 million tonnes (Domokos et al., 2000) or at 51.7 million t year−1, averaged over 1858–1988 (Malciu, 2000). Maximum suspended load transport at the delta apex is 2,048 kg s-1 (i.e. ca. 65 million t y-1 over 1840–2010 – Bondar and Iordache, 2016).The extreme values recorded are 162 million t in 1941 and 19.8 million t in 1921 (DDBRA, 2000). Along the Lower Danube multiannual average bedload transport (1840–2012) decreases from 14.6 kg s-1 (at Vidin) to 5.6 kg s-1 (at the apex of the delta), while suspended load reduces downstream from 54.9 kg s-1 (near Giurgiu) to 130.1 kg s-1 (at the apex, Bondar and Iordache, 2016). Intensive deposition occurs along the Kilia (Chilia) distributary at the delta, where large shoals build up (Brânduş and Canciu, 2011).

Table 12.1 Proportions of the estimated bedload and suspended load transport along the Danube after lowflow regulations (after Lászlóffy, 1965) No.a

1 2 3 4 5 6 7

Locality

River km from mouth

Annual average (million tonnes) 0.03/ 0.15/0.03b 0.04 0.06 0.70 0.54 0.72

Bedload Average grain size (mm)

Maximum grain size (mm)

Ulm 2,586 14 65 Ingolstadt 2,458 17 65 Regensburg 2,376 7 120 Hofkirchen 2,257 10 100 Passau 2,225 30 150 Linz 2,135 20 100 Stein– 2,004 18 110 Krems 8 Vienna 1,934 1.07/ca. 0.4b 11 80 9 Bratislava 1,869 0.90 10 70 10 Gönyü 1,791 0.05 3 35 11 Nagymaros 1,695 0.02 2 30 12 Budapest 1,647 0.01/−b 2/−b 30/−b b b 13 Mohács 1,447 0.04/− 1/− 2/−b b 14 Bogojevo 1,367 0.05 1/− 2/−b b 15 Novi Sad 1,255 0.01 1/− 1/−b 16 Zimun 1,173 – – – 17 Belgrade 1,171 0.05 1 2 18 Baziaş 1,072 0.10 5 60 19 Drencova 1,015 0.03 2 30 20 Orşova 955 – – – 21 Calafat 795 0.0046c – – 22 Corabia 630 – – – 23 Giurgiu 493 – – – 24 Cernavodǎ 300 – – – 25 Brâila 170 0.0018c – – 26 Sulina 0 – – – a Numbers listed are the locations shown in Figure 12.1. b Updated with new data (from various sources) after barrage constructions. c Long-term (1840-2012) averages (Bondar and Iordache, 2016).

Suspended load Annual Specific average load (mg (million L−1) tonnes) 0.26 72 0.96/0.40b 100/42b 1.0 74 1.1/0.055b 55/8b b 3.9/0.35 88/6b b 4.0/0.34 87/10b b 4.5/0.46 87/11b 4.7/0.048b 7.3/1.21b 6.5/0.37b 10/0.91b 11/1.46b 15/1.3b 18/1.68b 18 33 40 44 45 45 46 52 66/1.5c 70 70/3.5c 76

79/10b 113/13b 99 135 149/18b 199/38b 191/42b 191 271 239 250 255 255 261 292 357 375 375 380

Figure 12.3 Sediment transport along the Danube (Habersack et al., 2016). A. Suspended load, 1, around 1960, 2, 1985–2000. B. Bedload (gravel fraction only), 3, around 1960, 4, 1995–2010

The Danube freezes in cold winters. On the long-term average (1956–1985) river ice occurred at Bratislava in 94 % and ice-jams in 32 % of years. The average duration of drift ice is 24 days, usually between mid-December and early February (Stančíkova, 2010). On the regulated Danube annual river ice duration at Budapest reduced from 27 days (in the 1817–1875 period) to 7 days (1940–2004) (Takács et al., 2013). No ice-jam flood occurred over the past 50 years. In January 2012 and 2017, however, the river was frozen over after a quarter of a century ice-free period. 12.3 HEADWATERS OF THE DANUBE In the time of the great explorations the location of the sources of major rivers in remote parts of the Earth presented a real challenge for scientists. Although it is located in the heart of Europe, the true source of the Danube remained a mystery for a long time. The spring in the park of the Fürstenberg castle of Donaueschingen regarded as the source (‘Donauquelle’) since the Romans, is in reality just a place where a minor headwater issues. The Danube rises from two streams (Figure 12.4): the Brigach (source at 1,125 m above sea level) and the Breg (from a swamp meadow at 1,078 m). Longer than the Brigach, the Breg is rightfully considered the uppermost reach of the Danube. However, the name Danube only applies to the river from the confluence at Donaueschingen, ca. 45 km downstream. The Danube headwaters cut across the Swabian-Franconian escarpments, produced in the Pliocene by rapid uplift and differential erosion. The streams transport granitic debris.

Figure 12.4 The channelized Breg and the braided Brigach join to form the Danube near Donaueschingen, Germany (photograph by Lóczy)

A large part of the Danube water does not originate from the uppermost Danube, but from the Inn, a major tributary with the highest discharge (length 510 km; catchment area 26,100 km2, mean discharge 730.8 m3 s−1 at its confluence with the Danube at Passau – number 5, Figure 12.1). The Inn rises from a tarn (Lake Lunghino) in the Rhaetian Alps, Switzerland. The whitish water colour indicates that it also transports more suspended sediment (clay particles from Tertiary sediments and silt from Quaternary till) than the Danube. 12.4 HYDROMORPHOLOGY OF THE DANUBE SECTIONS The river is commonly divided into three sections. The Upper Danube ends at the narrow passage of the Devín Gate, the Middle Danube stretches downstream to the Iron Gate, and the Lower Danube flows through the lowland beyond and the delta (Sommerhäuser et al., 2003). The headwaters and the upper section are traditionally identified with the occurrence of salmonid fishes (e.g. Salmo trutta), the middle section with barbel (Barbus barbus), and the lower lowland reaches with bream (Abramis brama). The delta is characterized by a brackish water biota. Pristine or rather semi-natural conditions are restricted to very short reaches of the Danube and its tributaries. 12.4.1 The Upper Danube in Germany and Austria An extraordinary phenomenon occurs at a short distance below the confluence of the headwaters. The Danube regularly loses water to the Rhine system. This happens downstream of Geisingen, where the river breaks through the Swabian Jura in a picturesque gorge. The Palaeo-Danube formed large meanders and, following the uplift of the Swabian Jura, incised into the Jurassic limestone. The Swabian Jura, the largest karst region (ca. 5,000 km2) in Europe (Geyer and Gwinner, 1984), is a source of the Danube’s coarse gravel (Fig. 12.5).

Figure 12.5 Percolation of water underground at Immendingen. The riverbed is dry for an average of 55 days in a year (photograph by Lóczy)

The extensive caves and underground passages (estimated total volume 50 million m3), opening to the surface in ponors, are responsible for the percolation of Danube water deep underground and eventually into the Rhine system (Hötzl, 1996). Thus, the Rhine extends its subsurface drainage area, capturing more and more of the Danube catchment. The Danube channel may dry up entirely for months as happened at Immendingen in 1874, 1911, 1921, 1928 and 1943, and recently almost every year (Figure 12.5). Between Immendingen and Fridingen, loss of water underground is calculated at 4 m3 s−1 as the total for several sites of percolation. The water follows underground passages for 12–20 km and reappears in abundant springs. The Aachtopf at Aach (Figure 12.6) is the largest spring in Germany (medium yield 8.59 m3 s−1; maximum 24.10 m3 s−1; minimum 1.31 m3 s−1).

Figure 12.6 The Aachquelle, a spring of abundant discharge in Germany (photograph by Lóczy)

Tracer experiments (Werner et al., 1997) – in 1877 with 20 t of rock salt – proved that the catchment of the Danube is linked to that of the Rhine through underground karst passages. About two-thirds of the water yield of the Achtopf derives from the Danube. The springwater feeds a stream (Radolfzeller Ach) which flows into Lake Constance, part of the Rhine catchment. Along the uppermost section (until Linz in Austria, number 6, Figure 12.1) downcutting channels are typical. The entrenched meanders between Geisingen and Sigmaringen, however, indicate that earlier surface slope was lower. In the Swabian-Bavarian Basin the first two significant right-bank tributaries are the Iller (68 m3 s−1 discharge at confluence) and the Lech (118 m3 s−1). Both the coarse glacial deposits and the underlying upper freshwater molasse beds are excellent aquifers and add to coarse bedload. At Regensburg (number 3, Figure 12.1), the northernmost point of its valley, the Danube meanders broadly following the moderate subsidence (maximum 16 m over 10 ka) of the Alpine Foreland (Buch, 1989). Downstream of Regensburg the tributaries arriving from the Alps (the Isar with 176 m3 s−1 discharge and the Inn) and carrying huge amounts of bedload build large alluvial fans and broad terraces and force the river channel towards the northern margin of the basin.

After receiving its largest tributary, the Inn, at Passau, the Danube enters the Linz Basin and follows the northern margin of the Austrian Alpine Foreland. Alpine tributaries of high discharge and sediment load, the Traun (150 m3 s−1), the Enns (230 m3 s−1), and the Ybbs (42 m3 s−1), maintain the asymmetry of the catchment mentioned above (Figure 12.1). Crossing the Linz and Machland basins, the Danube enters one of the most spectacular gorges of the Danube, the Wachau. The largest depression of the Austrian section is the Vienna Basin, where a 220-m-thick middle Pliocene to Holocene alluvial sequence overlies Lower Miocene to Lower Pliocene marine sediments (Jungwirth et al., 2014). 12.4.2 The Middle Danube (Slovakia, Hungary and Serbia) In the Devín Gate (Porta Hungarica) between the Leitha Mountains and the Little Carpathians, where eight terraces have been identified (Mahr and Šajgalik, 1979), the river enters the Pannonian Basin, the largest of all the Danubian basins. The Little Hungarian Plain is a focus of drainage network (Pécsi and Schweitzer, 1995). The Palaeo-Danube built a double alluvial fan (Pécsi, 1959) downstream the mouth of the Morava (105 m3 s−1 mean discharge and large amounts of suspended load, Table 12.1) at Bratislava. On this alluvial fan two meandering anabranching channels have formed two large islands: the Szigetköz (375 km2) bordered by the Moson Danube on the Hungarian side and the Žitný ostrov (in Hungarian: Csallóköz, 1,600 km2) bordered by the 128-kmlong Lesser Danube on the Slovak side. The Old Danube has a braided channel. River competence decreases rapidly along the present-day gravel bed. Maximum grain size is 15–20 mm at the upper tip of the Szigetköz and 5 mm at the lower end. The environment along this section has been fundamentally altered by the construction of the Gabčíkovo Barrage (see below). Before barrages were constructed on the Danube, coarse bedload arrived via major left-bank tributaries, the Váh (190 m3 s−1 mean discharge) and the Hron (58 m3 s−1). In contrast, the largest right-bank tributary, the Rába (80 m3 s−1) at Győr (near number 10, Figure 12.1) only transports suspended load. In the terraced valley east of Győr the Danube flows in a braided channel with numerous mid-channel bars built of the Váh and Hron deposits. A dramatic change takes place when the Danube cuts through the andesitic Visegrád and Börzsöny Mountains range in the Visegrád Gorge, where the river is 300 m wide and at least 5 m deep. Seven terraces are identified in this horseshoe-shaped bend (Karátson 2015). At the town of Vác (river km 1,680) the north–south stretch begins in the Great Hungarian Plain, the largest and driest of the Danubian basins. Pleistocene anabranching created extensive islands, the largest being the Szentendre (gravel) and the Csepel Islands (coarse sand). The present meandering river follows the tectonic Buda Thermal Line, which gives rise to hot springs (even on the channel bed) and the famous baths of Budapest. Even before the construction of hydroelectric plants in Austria, gravel transport in the Danube did not reach much downstream Budapest (Table 12.1). South of Budapest the Danube receives no major tributary for more than 250 km. The striking valley asymmetry continues. On the right bank the high undercut loess bluffs of the Mezőföld occur in contrast to the flat alluvial left bank. South of Mohács the Danube leaves Hungary marking the boundary between Croatia and Serbia (Jankovič and Jovičič,

1994). The three largest tributaries along this section differ in their water and sediment discharges. The Drava (725 km length, 670 m3 s−1 mean discharge) flows into the Danube east of Osijek (upstream of number 14, Fig. 12.1). Along its upper reaches it transports a large amount of bedload, but only some sand and suspended silt and clay reach the Danube (Schwarz, 2018; Tamás, 2018). The longest tributary, the Tisza (962 km channelised length), discharges 995 m3 s−1 of water at the confluence of Titel (near number 15, Figure 12.1), while the Sava of almost the same length (940 km) yields 1,800 m3 s−1 at Belgrade. As opposed to the suspended fine silt of the ‘blonde’ (light-coloured) Tisza, the Sava brings walnut-sized limestone and ophiolite boulders from the Dinaric Mountains. An even greater contrast is found between the sediment loads of the next leftbank tributary, the Timiş (104 m3 s−1), and a right-bank stream, the Velika Morava (the second river of this name, 244 m3 s−1 – Gavrilović and Dukić, 2014). The former transports very fine sand and silt, while >10 cm diameter volcanic gravels travel down the Velika Morava (Bendeffy, 1979). Here the discharge of the Danube rises to 5,840 m3 s−1. In the southernmost Great Hungarian Plain the river is forced to turn east first at Vukovar by the Palaeozoic metamorphics of Fruška Gora (539 m) and again near Belgrade by the Avala laccolith (511 m). Leaving the Pannonian Basin, the river crosses the arc-shaped Southern Carpathians– Serbian Ore Mountains in the 145-km-long narrow reaches between Baziaş (number 18, Figure 12.1) and Turnu Severin. In the gorge of Djerdap Jurassic limestone is exposed on 250–300 m high walls on both sides. The Danube is dammed for hydropower (Stevović et al., 2017). The deepest points of large potholes in the rocky river bed are only about 7– 8 m above sea level. Leaving the Orşova Basin and merging with the Cerna, the Danube enters the spectacular Iron Gate (Porţile de Fier) Gorge (Orghidan, 1966; Figure 12.7).

Figure 12.7 The deepest gorge on the Danube: the Iron Gate on the Romanian–Serbian border. (Photograph by Lóczy)

12.4.3 The Lower Danube (Romania, Bulgaria and Ukraine) Downstream the Iron Gate large-scale channel aggradation prevails. The river proceeds in large bends towards the south and turns east again south of Vidin. Only 3.3 % of the Danube catchment lies in Bulgaria. Along the right bank stretches the 130 m high steep margin of the Danubian Tableland of Bulgaria (502 m maximum altitude) (Dinev and

Mishev, 1980), affected by extensive landslides (Dobrev et al., 2014). A relatively significant tributary along this section is the Iskâr (368 km long, 8,646 km2 catchment area, and 55 m3 s−1 mean discharge). Like other tributaries, it crosses the ranges of the Balkan Mountains in two gorges but reaches the Danube in braided channels. The tributaries from the Romanian Plain are larger (the Olt having the highest discharge, 215 m3 s−1) and show both meandering and braided patterns. Five major terraces were identified on the Romanian side (Popp, 1974). Moving downstream, the number of terraces is reduced to only one at Câlâraşi (river km 300). The marshy Holocene floodplain (called ‘lunca’ in Romanian) is 8–12 km wide along the western section and broadens to 20 km at Câlâraşi. It is dotted by numerous oxbow lakes (Grecu et al., 2009, 2011). The river terraces in Bulgaria are composed of more than 20 m thick sand and gravel above the typical floodplain features of long parallel sand ridges locally called ‘gredove’ (meaning ‘beams’) and blown sand dunes of 8–10 m (maximum 15 m) height and 0.5 km length. Both are valuable habitats (Tzonev 2015). Today the Danube flows along the southern margin of the Romanian Plain in an asymmetric valley as a braided sand channel 300–1,600 m wide at low water (Dinev and Mishev, 1980). Navigation is hindered by several fords, the shallowest being 0.9 m deep. There are 97 islands of various size along this anabranching section where sediment accumulates, followed downstream by a meandering Danube. In the eastern half of the Romanian Plain, in Muntenia, yazoo channels are typical at 10–20 km distance from the Danube. The present river system evolved following the gradual subsidence of the Siret confluence area as well as the building of natural levees of 5–10 m height. At the town of Cernavodǎ (number 24, Figure 12.1) the river approaches the Black Sea but the eroded mass of Dobrogea Hills forces it to turn north at Câlâraşi only 40 km from the sea. The Dobrogea Hills (467 m, Palaeozoic granites and phyllites and Mesozoic carbonates) present a barrier to river flow since their uplift during Neogene times. West of the Dobrogea Hills the Danube has large marshy islands (Balta Ialomiţei, Balta Brǎilei) at the confluences of major tributaries. At the major river port of Galaţi it turns again in an easterly direction and receives the last major tributaries, the Siret (255 m3 s−1; Romanescu and Stoleriu, 2013) and the Prut (76 m3 s−1; Romanescu and Stoleriu, 2017) from the eastern and northeastern Carpathians. The arid hills of Romania and Moldova (specific runoff below 50 mm y−1 or 1.5 L s−1 km−2) along the middle and lower reaches only provide a very small amount of suspended load to the Danube (Table 12.1). 12.5 THE DANUBE DELTA The town of Tulcea is at the head of a typical arcuate delta of 4,345 km2 area. Modern maps (Figure 12.8) only show three main distributary channels (Chilia [Kilia], Sfântu Gheorghe [St George] and Sulina) but five to seven of them have been reported in Greek antiquity. The large-scale mapping project of the Danube by Marsigli (1726) shows five branches. The Delta is a mosaic of shallow lakes and channels. Recent measurements confirm that the total length of natural watercourses (1,743 km) roughly equals that of human-made canals (1,753 km) in the Delta. Lakes of maximum 4 m depth make up 258 km2 (8 % of delta area), the largest being the Dranov (20 km2) and Red Lakes (Lacul

Roşu, 14.5 km2; DDBRA, 2000). Several elongated freshwater lakes outside the Delta resulted from rapid deposition that blocked north to south running watercourses.

Figure 12.8 The Danube Delta. (a) Earlier stage (after Hartley, 1887) and (b) current stage (after Romanescu, 1996). 1, lagoons, lakes; 2, sand dunes on barrier islands, beaches; 3, settlement. [(b) Reprinted from Zeitschrift für Geomorphologie Supplement-Band 106, Romanescu, G., 1996, with permission from Gebr. Borntraeger Verlagsbuchhandlung]

A truly amphibian landscape, the Delta has only 9 % of its area permanently above water. Only sand dunes are higher than 10 m. Natural levees, up to 4.2 m height and 100– 250 m wide along the three main distributary channels make up 6 % of the Delta area (DDBRA, 2000). Crevasse splays are common. About 20.5 % is below the Black Sea mean level and more than half (54.5 %) is between 0 and 1 m altitude (Panin, 2003). About 75 % of the surface deposits are fine to very fine sand but the low slope of the prograding delta is muddy (Coleman et al., 1986). 12.5.1 Delta Habitats and Environmental Problems

Biodiversity is remarkable in the Danube Delta but threats to it are also high. The most common habitat types (DDBRA 2000) are: • aquatic habitats (freshwater watercourses and lakes, brackish and saltwater coastal lagoons);

Figure 12.8 Continued

• • • • • • •

regularly flooded small islands (‘plaur’); other wetlands (reed beds and willow stands); riparian (willow and poplar) forests; wet and dry meadows; sand and mud beaches; shrubs; barren land and man-made habitats (human settlements; agricultural and forested polders; poplar plantations and fish ponds).

The filling of lakes with organic-rich substances leads to the formation of extensive floating organic mats (Figure 12.9), a special feature of the Delta. The Danube Delta Biosphere Reserve (6,792 km2) was established in 1991 and approved as a Ramsar site and a Natural World Heritage Site by UNESCO (DDBRA, 2000). Protection was extended over the Ukrainian part of the delta in 1998 (Vadineanu

and Voloshyn, 1999). There are 2,840 km2 of reed beds in Romania alone. In the ca. 460 km2 wetlands of the Ukrainian delta, 240 of the total 320 bird species (cormorants, pelicans, geese etc.) live. It is also one of the last refuges of the European mink (Mustela lutreola), the wildcat, the freshwater otter, the globally threatened monk seal (Monachus monachus), and certain reptiles. The Delta is an extremely valuable buffer zone between the Danube Basin and the Black Sea (UNEP, 1991). Environmental pressure is very high. There are four primary causes of the decline in delta ecosystems (Pringle et al., 1993): • •

the high nutrient yields of the Danube leading to eutrophication (algal blooms); channelization, floodplain drainage, and the impacts of engineering works (damming, dredging): reduction or elimination of natural freshwater and sediment recharge leading to coastal erosion (locally 17 m y-1 shoreline retreat);

Figure 12.9 Floating vegetation on one of the lakes in the Danube Delta. (Copyright Danube Delta Biosphere Reserve Authority, Romania)





intensive agriculture (35,000 km2 agricultural land) next to the Danube channel and in the delta itself and aquaculture (2,440 km2 of fish farms) causing pollution (pesticides, herbicides and fertilizers; mercury and heavy metals from industrial sources); groundwater salinization (salt content locally reaching 800 mg L−1).

The rising sea level (7.6 ± 0.3 mm y−1 between 1992 and 2005 – Kubryakov and Stanichnyi, 2012) also contributes to salt water intrusion and coastal erosion (Malciu, 2000). Land loss is estimated as high as 11.5 km2 y−1 (DDBRA, 2000). As the Sulina (number 26, Figure 12.1) branch was developed into a waterway (straightened and

dredged), it lost its links with wetlands of high filtering capacity. Now it delivers pollutants in a concentrated form to the Black Sea. The environmental deterioration of the Danube Delta has a deleterious influence also on the Black Sea ecosystem. In the 1990s the chemocline (the upper boundary of oxygenpoor deep water) of the Black Sea rose from 170 to 110 m below the surface within a few years (Pringle et al., 1993). The impact of climate change is also observed in the increased frequency of drought periods and the expansion of invasive plants in the Delta (Doroftei and Mierlă, 2012). 12.6 THE EVOLUTION OF THE VALLEY OF THE DANUBE 12.6.1 The Upper Section A striking feature of the Upper Danube section was the drainage modification over several million years during the Neogene. The catchments of three major European rivers, the Danube, the Rhine and the Rhône, developed simultaneously after the last marine inundation (16 Ma ago). The controlling tectonic processes were associated with the uplift of the Alps and the gradual subsidence of the Upper Rhine Rift Valley. The uplift of the Alps involved the accumulation of a marine and freshwater molasse sequence (6,500 m deep to the south and thinning out towards the north) in the Northern Alpine Foreland Basin (Habbe, 1994). The Vosges and Black Forest mountains rose bounding the southward extending Upper Rhine Rift Valley resulting in the retreat of the Molasse Sea. The Palaeo-Danube appeared in the Middle Miocene (15 Ma ago), draining the Alpine Foreland towards the east. Testimony to this oldest course of the river is provided by the highest (and thus oldest) of the nine terraces, ca. 245 m above the present-day mean water level (Fischer, 1989). Later this indeterminate east-flowing drainage suddenly changed towards the west or south-west due to tilted uplift and disturbances caused by the meteorite impact that formed the Ries crater at Nördlingen 14.8 ± 0.6 Ma ago (Kavasch, 1992). The ejecta reshaped regional relief and drainage pattern, filling up valleys, blocking watercourses into extensive lakes, and adding certain rivers, like the Palaeo-Main to the Danube system. It is not known for how long the west-flowing Palaeo-Danube survived, but by the late Badenian (ca. 8 Ma ago) an eastward-flowing draining system developed again. The headwaters to the west were probably the Aare and the Valais Rhône (the Aare-Danube – Domokos et al., 2000). In the Sarmatian the Aare-Danube was primarily fed by watercourses from the southern Vosges and the Black Forest as well as by those flowing southeast from the present-day Neckar catchment. Most of its discharge, however, was contributed by the Alpine Rhine, a major right-bank tributary (Villinger, 1986). The Feldberg-Danube drained the central and southern Black Forest. The predecessors of the present-day Danube headwaters, the Breg and the Brigach, probably also existed as early as the Late Miocene (Villinger, 1998). By the late Pliocene (ca. 3 Ma ago) intensive uplifting of the Black Forest region blocked the Aare-Danube, which turned to the northeast, more or less occupying its present course. The headwaters of the Rhine extended southwards rapidly through headward erosion. The Aare-Danube was unable to keep pace with the rapid uplift of the Swiss Jura and had to find a new drainage route to the west. The Rhône Basin gained substantially in area at the expense of the Danube catchment.

The regression of Rhine headwaters reached the Swiss Jura by the very end of the Pliocene and captured the Upper Aare, which became a north-flowing tributary of the Rhine system (Rutte, 1987). The Alpine Rhine, flowing through Lake Constance, remained in the Danube system, but later it was also diverted into the Rhine system. Advancing Alpine glaciers and gravel-depositing southern tributaries caused the Danube to shift its channel to the north in the Pleistocene. The stream deepened its valley floor 150–200 m and resulted in large-scale river beheadings in Europe (Hebestreit et al., 1993). By the end of the Pleistocene the present drainage pattern in the source area of the Danube had taken shape. During postglacial times higher discharges led to the stabilization of drainage lines by incision. The complicated story, however, has not yet come to an end. The relief between the valley floor of the Danube (678 m above sea level at Donaueschingen) and that of the Rhine (345 m at the confluence with the Aare) clearly indicates that the capture of the uppermost Danube section is foreseeable along the valley of the one-time Feldberg-Danube in the geologically near future. One of the most significant shifts of the river course happened at Ehingen (515 m above sea level), where in the Middle Pleistocene the Danube abandoned its previous channel along the Swabian Jura cliff line. Another instance is attested by one of the longest abandoned Danube valley sections, the present Wellheim and Altmühl valley (Fischer, 1991), first occupied immediately after the Ries meteorite impact and again by the incised meanders of the Alb-Danube ca. 4 Ma ago. Ca. 200 ka ago the Danube changed course again and occupied the floor of a collapsed karst polje with westretreating springs for several millennia. At present the channel has a more southerly location. Major abandoned Danube valley sections have been described by German geomorphologists in the eastern Bavarian Basin (Jerz, 1995). Further channel changes during the Plio-Pleistocene have been reconstructed for the Danube section along the Austrian Alpine Foreland, e.g. the spectacular Strudengau Gorge, occupied by the Danube in the Late Pliocene (Fink, 1966). Gradual uplift of the mountain margin and the simultaneous downcutting of the river and its tributaries produced a spectacular series of terraces (Figure 12.10). In the Vienna Basin subsidence took place in a complicated pattern, adjusted to several foci of limited extension. This explains why the river turned north, then east, and occupied the present-day Vienna Gate only at a later date in the Pleistocene. Along the Austrian section natural flow and sediment transport patterns have been fundamentally changed by the recent impoundment of the river (Hohensinner et al., 2004, 2005).

Figure 12.10 Terraces in the Vienna Basin (a, after Fink, 1972) and in the Danube Bend (b, after Pécsi et al., 1988). tI–tVII, Danube terraces covered by travertine horizons (T 1–T7) and loess; PI–PII, Pliocene pediment surfaces covered by travertines (T 8–T9); mI–mII, Upper Miocene raised beaches covered by travertines (T10–T12); T0, older surface of planation. [(b) From Neogene and Quaternary geomorphological surfaces and lithostratigraphical units in the Transdanubian Maintains, Pécsi et al., 1988, with permission from F. Schweitzer]

12.6.2 The Middle Section Recently, it has been proposed that the development of modern drainage network in the spacious Carpathian (Middle Danubian) Basin was interrupted by a marked spell of arid climate at the beginning of the Pliocene (ca. 5 Ma ago; Schweitzer, 1997), coincident

with the Messinian Salinity Crisis. Thus, the evolution of the present drainage could only resume during times of red clay formation under humid climate (4.5–3.3 Ma). From that time on the ancient Danube – along with the Tisza – became a main hydrographic axis in the Basin. There are three possible locations where the Palaeo-Danube may have entered the Carpathian Basin at various times (Figure 12.11): the Ebenfurth or Sopron Gate (southwest of the Leitha Mountains), the Bruck Gate (northeast of the same mountains, now occupied by the Leitha River) and the Porta Hungarica (Devín Gate) between the Alps (Hainburg Block) and the (Little) Carpathians. The river was relocated from one gate to the other in this sequence, producing an older (Figure 12.11b) alluvial fan with its apex in the Bruck Gate and a younger (Figure 12.11c) in the Devín Gate (Lehotský et al., 2009). According to conservative estimates the Danube had deposited almost 90 % of its annual 400,000 to 600,000 million m3 of bedload in the fan before channelization (Pécsi, 1959). Under natural conditions (studied in detail for the period 1712–1886 by Pišút, 2002) the high-energy river showed intensive lateral erosion. The key mechanisms of channel planform change were avulsions and meander migration leading to chute and neck cutoffs. The adjustment of the river to human interventions and major floods resulted first in large-scale meandering and then in the development of an anabranching–meandering pattern. When the present drainage network began to take shape, the floor of the Carpathian Basin was 2,000 m deeper than that of the Vienna Basin. This enormous drop in slope resulted in massive gravel accumulation on the western margin of the Little Hungarian Plain at Parndorf.

Figure 12.11 The reconstructed ancient drainage of the Carpathian Basin. (a) End of Pliocene (after Sümeghy, 1953); (b) Early Pleistocene; (c) middle of last glaciation (both after Borsy and Félegyházi, 1983). [(a) Reprinted from Annual Report of the Hungarian Geological Institute for 1951, Sümeghy, J., Issues concerning the Pliocene and Pleistocene stratigraphy in Hungarian basins, 1953, with permission of the Geological Institute of Hungary. (b, c) Reprinted from Reconstructed ancient drainage of the Carpathian Basin, Borsy, Félegyházi, Copyright (1983) with permission from E. Félegyházi]

The origin of the picturesque Visegrád Gorge or Danube Bend has been the subject of debates among Hungarian geomorphologists for more than a century (Pécsi, 1999). The appearance of the Danube here is now placed in the late Miocene or early Pliocene (Korpás, 1998). The gorge has been regarded as antecedent, superimposed or both by various scientists (Karátson, 2013). Its formation was initiated by river incision into the post-volcanic sedimentary cover in the Middle or Late Pleistocene, triggered by (200–300 m) mountain uplift and climate change. Intense relief dissection has resulted in the exhumation of the high volcanic relief. The U-planform of the Bend is now explained by a volcanic reconstruction with the river occupying the interior of an eroded double caldera (Karátson, 2013). Some late Miocene rivers coming from the Northwestern Carpathians were probably depositing along the shore zone of the retreating Lake Pannon. The Visegrád Gorge functioned as a strait in the centre of a volcanic archipelago between the inundated Little and Great Hungarian Plains over a longer time span. The marked mineral composition of Danubian deposits permitted the tracing of the ancient courses of the river in the Great Hungarian Plain (Borsy and Félegyházi, 1983; Borsy, 1990). During the Pleistocene the Danube built a vast alluvial fan covering most of the present-day Danube–Tisza Interfluve (Figure 12.11). The sands of the alluvial fan were partially redeposited by aeolian processes during drier periods. In the early Pleistocene the Palaeo-Danube followed a northwest–southeast course probably occupying the late Pliocene channel of the river Ipel’/Ipoly (Figure 12.11a). Carrying a heavy and coarse sediment load from the rapidly uplifting mountains and accommodating it in a channel of high width/depth ratio, the river formed an anabranching system (Figure 12.11b). Over the early and middle Pleistocene a gradual westward shift of this drainage system, rotation by almost 20° to the west, took place. In postglacial times intensified incision followed (Gábris and Nádor, 2007) and the anabranching system was considerably reduced in size (Figure 12.11c). However, its survival to historical times is attested by two residual hills with alluvial and loess mantles on the Danube–Tisza Interfluve (Solti-halom 124 m; Tétel-halom 114 m). The floodplain features and deposits of the Sárköz embayment provide further evidence that the river once followed even more westerly courses than today. Reconstructed from archaeological evidence the rate of bluff retreat south of Budapest could be between 2 m and 15 m per 100 years (Lóczy et al., 1989; Kovács et al., 2015). Braiding was common before flow regulation, now replaced by a clear meandering tendency. 12.6.3 The Lower Section The Iron Gate, like the Danube Bend in Hungary, was previously assumed to have originated from a river capture. Now it is recognized as an antecedent valley. During the various stages of the Pliocene it was a strait between the Pannonian and Moesian basins of the Paratethys Sea. The floor of potholes (75 m deep) reach below present sea level here. The evolution of the Romanian (Walachian) Lowlands was controlled by tectonic movements and marine transgressions (Grecu et al. 2009, 2011). Subsidence in the younger Rhodanian orogenic phase took place along north-to-south fault-lines following the Siret River and also at right angles to that, meeting at the town of Galaţi, a seismically still active area. The uneven depression was filled by Miocene–Pliocene marine and

lacustrine sediments and with a 350 m thick Quaternary series mostly carried by the south-flowing Carpathian rivers. The Danube gradually occupied this area following the west to east marine regression (Popp, 1974). The history of geomorphic evolution of the Plain, the Danube Delta and the Black Sea level are closely intertwined (Aksu et al., 2002). 12.6.4 The Delta The level of the Black Sea was 60–130 m lower during the glacial times. The entire delta sequence of the present time accumulated during the Holocene. Delta formation started in the Early Holocene, 9,000–8,500 years ago (Panin et al., 1983; Vespremeanu-Stroe et al., 2014), immediately following the assumed catastrophically rapid infilling of the Black Sea Basin (Ryan et al., 1997). It is claimed that the Black Sea-Lake remained fresh during deglaciation until the marine transgression at 9,300 calendar years BP (Yanchilina et al., 2017). The transgression was fast and took no longer than a couple of decades. Findings from optical luminescence analyses, however, indicate a more recent date, ca. 5,200 years (Giosan et al., 2006). Probably this younger age also implies a higher rate of erosion in the Danube Basin during the late Holocene. It is suggested that Late Holocene sea levels remained relatively stable (fluctuations limited to a range of +1.5 m and −2 m) in the region. Local submergences, which influenced delta evolution, have been attributed to local subsidences (Giosan et al., 2005, 2006). Southward directed longshore drift (Giosan et al., 1998) built the first barrier of the gulf ca. 5,200 years ago (Giosan et al., 2006). The line of the past barrier islands matches the current Letea-Caraorman-Crasnicol sand dune complex (grinduri) of 45–55 m thickness and a maximum altitude of 12.4 m (Figure 12.8). The first delta was built by the present St George Branch. Later it became inactive and a nothern distributary (Sulina) and delta lobe began to develop 3,600 years ago, and at the latest, expanded until 1,800 years before present (Giosan et al., 2006). Subsequently, new distributaries and young delta lobes were created around the delta periphery from ca. 2,000 years ago to historical times: the second St George Branch to the south and the northernmost distributary at Chilia (Preoteasa et al., 2016). In delta evolution the following stages recurred (Filip and Giosan, 2014): • subaqueous accumulation of a mouth bar; • emergence of a barrier island; • elongation of a spit in the direction of longshore drift and backward migration, often resulting in a barrier. Wave action eroded and reworked the older delta lobes, but fluvial-dominated evolution temporarily led to marked progradation (at 100 m y-1 rate for the Kilia Branch in the late 19th–early 20th centuries – Vespremeanu-Stroe et al., 2014). Between 1830 and 1922 the Kilia secondary delta grew by about 205 km2 in area, and in the period 1922– 2006 by 83 km2 (Brânduş and Canciu, 2011). Everywhere south of the Kilia lobe, however, coastal erosion is the characteristic process. 12.7 HUMAN IMPACTS

In this long and densely inhabited region, human activities have inevitably left their mark on the Danube and its riparian landscape. Now the Danube is regulated along over 80 % of its length. 12.7.1 A Brief History of Channelization Navigation on the Danube has always been a serious problem. Although the first towing paths were carved into the cliffs of the Iron Gate Gorge under Emperor Trajan around 100 AD, regular traffic was only launched by the Danube Steamboat Navigation Company (DDSG) in 1837. When the Turks controlled the Sulina channel of the Delta, they compelled every outgoing vessel to tow a large rake to stir up silt to enable the current to carry it out to the open sea (Hadfield, 1986). Later, after its foundation in 1856, the European Commission of the Danube encouraged the removal of sandbanks from the delta distributaries. In those days the river was described as: ‘Fast and difficult down to Vienna, with rapids and sandbanks, it remained fast to Komárom in Hungary, though because of shoals below Pressburg (Bratislava) this part of the river was in 1879 said to be fully navigable only for 202 days a year. Beyond Komárom it is navigable to the Carpathian Gorge, after which it spreads out on the flatlands towards the delta’ (Hadfield, 1986: 167–168). Today the Danube is navigable with small vessels to Ulm (2,586 km from the mouth), with ships above 1,300 t volume to Regensburg (river km 2,379). Flood control, however, has been an even stronger motivation for human intervention (Somogyi, 2000). In the 13th century dykes, unconnected as yet, were built along several reaches of the river (Stancíkova, 2001). The late 18th–early 19th century saw the first planned regulation measures. At first only settlements, later agricultural fields, were protected from flooding. A major stage was medium-flow regulation (Table 12.2), later supplemented by low-flow regulation indispensable for improving navigation or the undisturbed release of floating ice. The Fertő–Hanság water system in northwestern Hungary used to be linked to the Danube but it was detached and drained during the 1950–1960 channelization (Erdélyi, 1994). A natural disaster followed: peat resources desiccated into muck, were eroded and locally burnt spontaneously. The Ferenc Canal was built in the south of the Great Plain in 1795–1802 to connect the Tisza with the Danube (Lóczy et al., 2014). Due to the westward channel shifts of the Danube, however, it had to be extended in 1851–1868 to reach the river. The lock at the town of Bezdan is claimed to be the first concrete lock in Europe. Along the Upper Danube the navigable Vienna flood canal (3 m deep, 305 m wide and 16 km long) was completed by 1866. Table 12.2 A summary of medium-flow regulation of the Danube Section

Date of main activities 1820–1890

Reduction of river length (%) −73

1826–1867

−15

In Austria

1850–1914

ca. −15

Middle Danube in Hungary

1871–1914

−18

Upper Danube in Baden-Württemberg In Bavaria

Measures Channel straightening, cut-offs, flood-control dykes, bank stabilization Channel straightening, cut-offs, flood-control dykes, bank stabilization, channel deepening by dredging and explosions By-channel closures, bank stabilization, flood-control dykes, training walls Flood-control dykes, by-channel closures, cut-offs, bank stabilization, groynes,

In Serbia

1894–1977

ca. −10

Lower Danube in Romania/Bulgaria Danube Delta (Sulina branch)

No regulation 1860–1901

– −30

confluence relocations, protecting walls along urban sections Flood-control dykes, bank stabilization, groynes Only dredging and some bank stabilization, flood-control dykes Straightening, dredging, longitudinal structures, canal building

Based on UNESCO, 1999.

The Iron Gate remained the most dangerous gorge to pass. Between 1833 and 1837 Hungary built the so-called Széchenyi Road with a 2 km long and 3 m deep shipping canal and between 1895 and 1898 the 2.5 km long Sip Canal, stone-pitched on both sides. On the Serbian bank rails were laid to allow a heavy steam locomotive to haul barges up (Hadfield, 1986). It could not solve all problems. For instance, infamous rocks like Ada Kaleh, rapids, and swift currents continued to present a threat to shipping. Some rocks were blown up but a final solution was only reached when the river was dammed in the Iron Gate between 1964 and 1972 to generate 2,136 MW of electricity. Iron Gate II dam followed 80 km downstream in 1984 (Hadfield, 1986). As the barrage at Nagymaros has not been built (see below), the rock sill at Gönyü (Little Hungarian Plain section, no 10, Figure 12.1) at 0.9 m depth survives as the last major single natural obstacle to navigation along the Danube. It had been long planned to connect the Danube with the Black Sea port, Constanţa, saving 380 km en route. The 64.2 km long, 90 m wide and 7 m deep canal completed by 1984 leaves the Danube at Cernavodǎ and shortens the distance to the sea by 400 km (Dragomirescu, 1993). Regulating the Danube has resulted in increased channel incision along the whole length of the river. The dredging just above Budapest (with the removal of 25 million m3 of gravels and sand) led to the thinning of gravel beds. These beds, which were 4–7 m thick in the 1960s, were eroded to 1–4 m thickness in the 1990s. The channel incision of 1.6 m was beneficial for navigation but was undesirable from other aspects (loss of the filtering capacity of gravel beds; uneven riverbed; depressions in bed filled with toxic deposits – Erdélyi, 1994). Contradictory opinions exist regarding impacts of river flow regulations and drainage measures. With an increasing emphasis on the presevation of wetland habitats in natural conditions and a decreasing demand for agricultural land, their benefits are often underestimated against their negative consequences. For example, losses to fishery (as valuable fish species would require a higher range of water level fluctuation) or secondary alkalization of soils in the floodplains (Somogyi, 2000). Regulation measures still continue but with a changed emphasis. Today the strongest motivation is river restoration for preservation of wildlife (Gewässerdirektion Donau/Bodensee, 2001). 12.7.2 The Rhine–Main–Danube Canal The construction of the 171-km-long Main–Danube Canal began in 1959 and completed by 1992. It is the key section of the main European waterway, the Rhine–Main–Danube navigation route of 3,500 km total length. It mostly follows the old Ludwig Canal, established at the order of King Ludwig I of Bavaria between 1836 and 1845 (Hadfield,

1986), which has 100 locks and can accommodate 100-tonne barges. In order to overcome the 175 m rise above the Main and the 68 m drop to the Danube, 15 lock systems had to be built along the Main–Danube Canal. The required depth is 3.5–4 m and the width is 40 m; for comparison, the corresponding data along the Danube between Kelheim and Passau are 2.8–3 m and 50–100 m, respectively. The Canal is navigable by vessels of up to 110 m length, 11.4 m width, 2.8 m load draught and 2,800 tonnes capacity. Each kilometre of the canal cost 27.5 million DM (at 1992 reckoning). It was cheaper than railway construction (ca. 36 million DM km−1) but somewhat more expensive than building a motorway (ca. 20 million DM km−1) (Die Rhein–Main–Donau–Wasserstraße, 1992). On the short term the investment proved to be profitable as freight transport grew to 6.77 million t by 1998, but in 1999 the trend was interrupted by the bombing of Yugoslavia. Although navigation had been restored along the Serbian section by 2005 and gained an increasing importance in the Serbian economy too (Radmilović and Maraš, 2011), long-term average freight transport on the Canal was only 6.2 million t – in contrast to the expected 15–20 million t (Bachmeier, 2017). Tourism (river cruises), however, experienced a boom: 1,272 floating hotels crossed the Canal in 2016. Environmentalists, however, have a dire picture about the project. Apart from the loss of habitats (around 600 hectares), an ecological hazard of the Canal is that it allows aquatic animals to spread from Western to Eastern Europe or in the opposite direction. Invasive species (such as the Chinese mitten crab [Eriocheir sinensis] in 2002) successfully compete with native species and lack predators (Leuven et al., 2009). On the other hand, the introduction of some species (including asp [Leuciscus aspius]) may eventually enrich wildlife in the Danube. 12.7.3 An Example of Damming the Danube: the Gabčíkovo Barrage in Slovakia Since the completion of the Kachlet dam in 1927, the 69 reservoirs (total volume: 7,300 106 m3) and hydroelectric plants established in the Danube Basin turned the Danube a highly engineered river (Habersack et al., 2016). Some of the structures generated heated debates. An environmentally controversial engineering structure resulted from a Czechoslovak–Hungarian agreement signed in 1977 (Bravard, 1986; Nemerkényi, 1990). Now it comprises a barrage at Gabčíkovo (river km 1,819) in Slovakia with a hydroelectric plant (720 MW capacity) and two navigation locks, the Čunovo weir and reservoir of 40 km2 area and 3.5 m depth and a 29-km-long navigation canal. The implementation of the large-scale barrage system is only partial. The planned lower barrage at Nagymaros has not been built and the large reservoir at Dunakiliti remains unused because in October 1992 Slovakia diverted the main Danube discharge into the navigation canal on their own territory. In 1997 the International Court in The Hague rejected Hungary’s assertion that under the changed circumstances the enforcement of the Czechoslovak–Hungarian treaty were not possible (Schwabach, 1996; ICJ, 1997). It was decided that successor states are bound to a bilateral treaty of territorial character. The judgement did not regulate the sharing of water between the two countries. According to an agreement made in 1995, 250–350 m3 s−1 water is released into the Old Danube channel – as opposed to 2,100 m3 s−1 before construction – and 20 m3 s−1 (40–50 m3 s−1 before damming) into the Moson Danube.

A strong argument against the barrage had been that it would damage the valuable groundwater reservoir of the Szigetköz (Erdélyi, 1994). The Danube is estimated to recharge the groundwater for an area of 910 km2 in Hungary above Győr. Groundwater reserves amount to 5.4 km3 in Hungary and 13 km3 in Slovakia, primarily not recharged through vertical infiltration but laterally as groundwater flow (Lehoczky, 1979). While groundwater flow in alluvial gravels used to be rapid, on average 3.8 to 10.3 m 3 s−1 (particularly in the northwest), it is assumed that river dams may cause siltation and the clogging of this flow in the long run. Through a possible connection with the karst water system of the Transdanubian Mountains, huge freshwater reserves are affected by the project (Erdélyi, 1994). When the Danube enters the Carpathian Basin from Austria, it has First Class water quality by the Hungarian standards. The water quality of Slovakian rivers, the Váh and Hron is still worse than that of the Danube (Poórová et al., 2017 – Figure 12.12). The diversion of most of the discharge has reduced the opportunity of filtering out contaminants through groundwater flow.

Fig. 12.12 Biological (BOD5) and chemical oxygen demands (COD) and total amount of pollutants (Ntotal and Ptotal) in the Danube and its Slovakian tributaries in 2016 (Poórová et al., 2017)

The pattern of the fluvial landscape reflects a highly uneven distribution of fine sediment cover (floodplain deposits) above sand-and-gravel point bars and, thus, great variations in the groundwater table depth and the availability of capillary water for crops (Góczán and Lóczy, 1990; Lóczy and Balogh, 1990). The diversion of most of the discharge into the navigation canal resulted in an increased range of groundwater level

fluctuation in a zone next to the Danube, and dropping levels (average 2–3 m; maximum −6 m) away from it. In order to prevent the lowering of groundwater to the gravel beds, an engineering structure, an underwater weir with fish ramp, was built across the channel in 1995. Experience shows that it successfully improves water recharge for the floodplain (Schiemer et al., 1999). A monitoring system to follow changes in groundwater level and quality is in operation. The hydromorphological and ecological impacts of the project are considerable (Guti, 2002; Farkas-Iványi and Guti, 2014). Lateral river-floodplain connectivity, the temporal flushing of side arms during floods, ceased. Narrowing of side arms and shrinking of aquatic habitats ensued. At the same time, the deposition of fines under lentic conditions provided good substrate for aquatic and semi-aquatic macrophytes and promoted spontaneous riparian forest growth. 12.7.4 Water temperature rise Global climate change also affects the Danube Basin. Mean annual air and water temperatures in the Danube Basin started to rise in 1988 (Bonacci et al., 2008). Since then both annual and seasonal maximum and minimum temperatures tend to increase along the Danube, linked to the atmospheric pattern governed by the North Atlantic Oscillation (Rîmbu et al., 2002). The maximum warming trend of Danube water was observed in the middle section, at the Bogojevo gauging station, Serbia (no 14, Fig. 12.1; +0.05 °C y-1) (Basarin et al., 2016). Research in the Bratislava reach indicates that the heat load of the Danube water remained virtually unchanged over the study period (1926–2005) (Pekarova et al., 2008). Fluctuations are closely bound to changes in discharge. The impact of the inflow of cooling water from nuclear plants could not be proven for the Hungarian section (Lovász, 2012). According to the water use license of the Paks Nuclear Plant, the following restrictions of heat load apply: water temperature 500 m downstream from cooling water inflow should not exceed 30°C; if the temperature of the Danube water is higher than 4°C, the maximum temperature of inflow should remain below 11°C, if lower than 4°C, below 14°C (EIS, 2015). The Report acknowledges that when the Paks II Nuclear Plant will be in operation, the first restriction cannot be met throughout the year. The water temperature of the Danube in the future, however, will depend more on climate change than on heat load from the nuclear plants. In the case of the Cernavodă nuclear plant, Romania, the fluctuation observed in the amount of chlorophyll-a seems to be natural (Török, 2012). Remote sensing studies show that in winter, the water temperature of the localized thermal plume is about 1.5°C higher, while in summer and autumn, the thermal plume extends 5–6 km far along Danube–Black Sea Canal, but with only 1°C higher temperature (Zoran et al., 2005). Also at Cernavodă, a second plant (units 3 and 4) adds to this heat load and causes up to 3°C increase (ICIM, 2006). 12.7.5 Pollution The Danube catchment lies across 14 countries, a distribution that underlines the importance of its transboundary influence (Literáthy, 2001). The water quality of the Danube has been investigated recently on several occasions (ICPDR, 2002, 2008). As far

as nutrients are concerned, organic nitrogen contents remain constant all along the river. Phosphorus (ortho-phosphate-P and total P) shows only a moderate increase in Danube water but not in bottom sediments downstream. Heavy metals (particularly arsenic, copper and nickel) are observed in high concentration at the confluences of Bulgarian tributaries (Lom, Iskâr and Timok), particularly during low water stages. High arsenic, chromium, mercury, lead and zinc concentrations were found even in the sediments of the Iron Gate Reservoir. The most dangerous pollutants are organic compounds of industrial, agricultural and municipal origin (ICPDR, 2015b). Even earlier surveys, such as the one led by J.-Y. Cousteau in 1991–92 (Cousteau Equipe, 1993) demonstrated high oil pollution in the river. The concentrations of polyaromatic hydrocarbons (PAH) were particularly high in mussels collected from Middle Danube tributaries. Among pesticides the main source of atrazine to the Danube was found to be the Sava River (0.78 µg l−1 maximum concentration – ICPDR, 2008). The findings of the Third Joint Danube Survey can be summarized as follows (ICPDR, 2015a). Since the previous surveys total nitrogen and total phosphorus concentrations has dropped. This reflect better municipal wastewater management in the basin. No significant change has been observed in metals and suspended particulate matter, but mercury occurred in fish samples raises concern. Most of the substances which have priority in the European Union Water Framework Directive (WFD) were found in concentrations below the new environmental quality standards (ICPDR, 2015a). The Survey also sought answer to the question to what extent the goals of the WFD have been achieved (ICPDR, 2015a). Nitrogen inputs to surface waters on the catchment dropped by 12 % from the value for 2009 to 602,000 t y-1 in 2015. The same figures for phosphorus are 46,000 t y-1 in 2015 (21 % lower than in 2009). The Danube, however, transports 40 % more nitrogen and 15 % more phosphorus into the Black Sea than in the 1960s. The environmental objectives of the WFD for the Danube river basin have not been met by 2015 (Liška, 2015). The survey of macroinvertebrates revealed that morphologically highly degraded reaches of the Upper Danube indicate moderate status, some more natural sites of the Upper and Middle Danube generally have a good status (ICPDR, 2015a). Poor saprobic conditions occur on some Danube sections in Serbia and Romania–Bulgaria (Pešić et al., 2010). Within the frame of the Third Joint Danube Survey, the overall hydromorphological assessment shows that about 60 % of the Danube falls below class 3 (21 % in the second class ‘slightly modified’ and 39 % in the third ‘moderately modified’), while 40 % into the classes four (26 %) and five (14 %) (Fig. 12.13).

Figure 12.13 Hydromorphological assessment of the Danube sections (ICPDR, 2015b)

12.7.6 How Much is the Danube Worth? Evaluations have been made earlier for the ecological, nutrient load reduction, and rehabilitation potentials of the channel and floodplain of the Danube. A major trend in the development of land evaluation, however, is to express the value of natural potentials (including water and riparian resources) in monetary terms (Avis et al., 2000). Although the economics of natural resources is difficult to establish, the first attempt has already been carried out (Andreasson-Gren and Groth, 1995). Based primarily on ecosystem productivity and substitutability, the average economic value of the Danube floodplains is estimated at £383 ha−1 year−1. Since the total floodplain area is ca. 1.7 million ha, its value adds up to £650 million year−1. Internationally coordinated human utilization, carefully designed channel, floodplain and ecosystem restoration measures and the appropriate operation of the Trans-National Monitoring Network (TNMN) are needed to preserve or improve environmental quality and natural wealth along this major cultural river of Europe. REFERENCES Aksu, A.E., Mudie, P.J., Rochon, A., Kaminski, M.A., Abrajano, T. and Yasar, D. (2002) Persistent Holocene Outflow from the Black Sea to the Eastern Mediterranean Contradicts Noah’s Flood Hypothesis. GSA Today, May, 4–10. Andreasson-Gren, M. and Groth, K.H. (1995) Economic Evaluation of Danube Floodplains. WWF International Discussion Paper, Gland, Switzerland. Avis, Ch., Tydeman, Ch. and Royo Gelabert, E. (2000) What role for water pricing? Ten actions for internalising sustainability. World Wide Fund for Nature, Brussels, 18 p. Babić-Mladenović, M., Kolarov, V. and Damjanović, V. (2013) Sediment regime of the Danube River in Serbia. International Journal of Sediment Research 28, pp. 470–485. doi: 10.1016/S10016279(14)60006-8 Bachmeier, U. (2017) 25 Jahre Rhein-Main-Donau-Kanal: Ein gescheitertes Jahrhundertprojekt. Augsburger Allgemeine, 08.09.2017. https://www.augsburger-allgemeine.de/bayern/25-Jahre-RheinMain-Donau-Kanal-Ein-gescheitertes-Jahrhundertprojekt-id42604111.html Basarin, B., Lukić, T., Pavić, D. and Wilby, R.L. (2016) Trends and multi‐annual variability of water temperatures in the river Danube, Serbia. Hydrological Processes 30, pp. 3315–3329. doi: 10.1002/hyp.10863

Bendeffy, L. (1979) A Duna medrében görgetett hordalék eredete és kőzetminősége (Origin and lithology of Danubian bedload). Földrajzi Értesítő 28(1–2), pp. 73–89. Bonacci, O., Trninić, D. and Roje-Bonacci, T. (2008) Analysis of the water temperature regime of the Danube and its tributaries in Croatia. Hydrological Processes 22, 1014–1021. doi: 10.1002/hyp.6975 Bondar, C. and Iordache, G. (2016) Sediment transport on the Romanian section of the Danube River. GeoEco-Marina 22, pp. 29–49. Borsy, Z. (1990) Evolution of the alluvial fan of the Alföld. In: Alluvial Fans: a Field Approach (A.H. Rachocki and M. Church, Eds). John Wiley & Sons, Ltd, Chichester, pp. 229–248. Borsy, Z. and Félegyházi, E. (1983) Evolution of the network of watercourses in the end of the Pleistocene to our days. Quaternary Studies in Poland 4, pp. 115–125. Brânduş, C. and Canciu, C. (2011) On the geomorphologic evolution of the Chilia secondary delta. Present Environment and Sustainable Development, 5(2), 101–110. Bravard, J.P. (1986) Gabčíkovo: un grand project et une controverse. Revue de Géographie de Lyon 74(1), pp. 19–41. Buch, M.W. (1989) Late Pleistocene and Holocene development of the Danube Valley east of Regensburg. In: Landforms and Landform Evolution in West Germany (F. Ahnert, Ed.). Catena, CremlingenDestedt, pp. 279–287 (CATENA Supplement 15). Caspary, H.J. (1995) Recent winter floods in Germany caused by changes in the atmospheric circulation across Europe. Physics and Chemistry of the Earth, 20, pp. 459–462. Coleman, J.M., Roberts, H.H. and Huh, O.K. (1986) Chapter 5. Deltaic Landforms. Plate D-2: Danube River Delta, Romania. In: Internet vesion of Geomorphology from Space (N.M. Short, Ed.). NASA, Distributed Active Archive Center, Goddard Space Flight Center, Greenbelt, MD. 2 p. http://daac.gsfc.nasa.gov/DAAC_DOCS/geomorphology/GEO_5/GEO_PLATE_D-2.html Cousteau Equipe. (1993) The Danube . . . for whom and for what? Cousteau Equipe’s Final Report, Paris. Czaya, E. (1981) Ströme der Erde. – Verlag für Kunst und Wissenschaft, Leipzig. 212 p. DDBRA. (2000) Danube Delta Biosphere Reserve. Danube Delta Biosphere Reserve Authority, Tulcea, Romania, 25 p. http://www.ddbra.ro Die Rhein–Main–Donau–Wasserstraße. (1992) Schiffahrtsverein Deutschland, 2 p. http://www.schiffahrtsverein.de/wasserstr.htm Dinev, L. and Mishev, K. (1980) B’lgariya kratka geografiya (A Short Geography of Bulgaria). D’rzhavno izdatelstvo Nauka i izkustvo, Sofiya, 351 p. Dobrev, N., Berov, B., Ivanov, P., Krastanov, M., Benderev, A. and Mihalkova, B. (2014) Overview on landslide distribution along Bulgarian river bank between Vidin and Nikopol. Problems of Geography, Bulgarian Academy of Sciences, Sofia, 1–2, pp. 64–75. Domokos, M., Neppel, F. and Somogyi, S. (2000) Paläogeographische Geschichte der Donau und ihres Einzugsgebietes. Hydrologie und Wasserbewirtschaftung 44(4), pp. 172–183. Doroftei, M. and Mierlă, M. (2012) Danube Delta habitats versus climate change. BALWOIS 2012, Ohrid, Macedonia, 28 May – 2 June 2012. Dragomirescu, S. (1993) Hydroelectricity in the Romanian Carpathians. GeoJournal 29(1), pp. 31–39. EIS. (2015) Modelling of the Danube River morphology and Danube heat load. Environmental Impact Study. PaksII Kht., 250 p. Erdélyi, M. (1994) Hydrogeology of the Hungarian upper Danube section (before and after damming the river). Hungarian Natural History Museum, Budapest, 115 p. Farkas-Iványi, K. and Guti, G. (2014) The effect of hydromorphological changes on habitat composition of the Szigetköz floodplain. Acta zoologica bulgarica Suppl. 7, pp. 117–121. Filip, F. and Giosan, L. (2014) Evolution of Chilia lobes of the Danube Delta: reorganization of deltaic processes under cultural pressure. Anthropocene, 5, 65–70. doi: 10.1016/j.ancene.2014.07.003 Fink, J. (1966) Die Paläogeographie der Donau. In: Limnologie der Donau (Liepolt, R. Ed.) 2. Schweitzerbart’sche Buchhandlungverlag, Stuttgart, pp. 1–50. Fink, J. (1972) Die Terrassen im Wiener Stadtgebiet. Blätter zur Physische Geographie. Manuskript. Vienna, 87/1972. Fischer, K. (1989) The landforms of the German Alps and the Alpine Foreland. In: Landforms and Landform Evolution in West Germany (F. Ahnert, Ed.). Catena, Cremlingen-Destedt, pp. 69–83 (CATENA Supplement 15). Fischer, K. (1991) Flußgeschichte und Reliefgenese im Norden und Osten des Nördlinger Rieses. In: Rieser Kulturtage Dokumentation 1990, Nördlingen, Vol. VIII, pp. 60–82.

Gábris, Gy. and Nádor, A. (2007) Long-term fluvial archives in Hungary: response of the Danube and Tisza rivers to tectonic movements and climatic changes during the Quaternary: a review. Quaternary Science Reviews, 26(22-24), pp. 2758–2782. Gasparotti, C. (2014) The main factors of water pollution in Danube River basin. EuroEconomica 33, 75– 88. Gavrilović, L. and Dukić, D. (2014) Velika Morava. In: Reke Srbije (Rivers of Serbia). Zavod za ucbenike, Beograd. pp. 67–97. Gewässerdirektion Donau/Bodensee. (2001) Aktionsplan Durchgängigkeit baden-württembergische Donau und Breg. Riedlingen, 15 p. + 3 Anlagen (Lebensraum Donau Erhalten– Entwickeln). Geyer, O. and Gwinner, M. (1984) Die Schwäbische Alb und ihr Vorland. Sammlung Geologischer Führer, Vol. 67. 3. Auflage. Gebrüder Bornträger, Berlin, 275 p. Giosan, L., Bokuniewicz, H.J., Panin, N. and Postalache, I. (1998) Longshore sediment transport pattern along the Romanian Danube Delta Coast. Journal of Coastal Research 15, pp. 859–871. Giosan, L., Donnelly, J.P., Vespremeanu, E.I. and Buonaiutu, F.S. (2005) River delta morphodynamics: Examples from the Danube delta. In: River Deltas – Concepts, Models and Examples (L. Giosan, and J.P. Bhattacharya, Eds). Society for Sedimentary Geology Special Publication 83, pp. 87– 132. Giosan, L., Donnelly, J.P., Constantinescu, S., Filip, F., Ovejanu, I., Vespremeanu-Stroe, A., Vespremeanu, E.I. and Duller, G.A.T. (2006) Young Danube delta documents stable Black Sea level since the middle Holocene: Morphodynamics, paleogeographic and archaeological implications. Geology 34(9), pp. 757–761. doi: 10.1130/G22587.1. Góczán, L. and Lóczy, D. (1990) The Slovak–Hungarian Barrage System on the Danube river and its environmental implications. Geographia Polonica 58, pp. 89–98. Grecu, F., Săcrieru, R., Ghiță, C. and Vacaru, L. (2009) Geomorphological landmarks of the eastern Romanian Plain Holocene evolution. Zeitschrift für Geomorphologie 53, Suppl. 1, pp. 101–110. Grecu, F., Ghiță, C., Albu, M. and Circiumaru, E. (2011) Geomorphometric analysis on the some riverbeds in the Romanian plain. International Journal of the Physical Sciences, 6(30), pp. 7055–7064. doi: 10.5897/IJPS11.913 Guti, G. (2002) Changes in the Szigetköz floodplain of the Danube and its fish communities after river diversion by the Gabčikovo Dam. Verhandlungen, Internationale Verein der Limnologie, 28, pp. 840– 844. Habbe, K.A. (1994) Das deutsche Alpenvorland. In: Physische Geographie Deutschlands (H. Liedtke, and J. Marcinek, Eds). Perthes, Gotha, pp. 439–475. Habersack, H., Hein, T., Stanica A., Liska, I., Mair, R., Jäger, E., Hauer, C. and Bradley, C. (2016) Challenges of river basin management: Current status of, and prospects for, the River Danube from a river engineering perspective. Science of the Total Environment 543, Part A, pp. 828–845. doi: 10.1016/j.scitotenv.2015.10.123 Hadfield, C. (1986) World Canals: Inland Navigation Past and Present. Facts on File Publications, New York, 432 p. Hartley, C.A. (1887) Cartes du Delta du Danube et plan comparatifs de l’embouchure et des sections fluviales du bras de Sulina. Commission Européenne du Danube, Leipzig, Plats I–IV. Hebestreit, C., Schiedek, T., Bauer, M. and Pfaffenberger, C. (1993) Zeitmarken der Wutacheintiefung – Terassenkorrelation, Terrassenstratigraphie und Kalktuffe. Jahresbericht und Mitteilungen Oberrheinischer Geologischer Verein, Neue Folge 75, Stuttgart, pp. 291–312. Hohensinner, S., Habersack, H., Jungwirth, M. and Zauner, G. (2004). Reconstruction of the characteristics of a natural alluvial river-floodplain system and hydromorphological changes following human modifications: The Danube River (1812–1991). River Research and Applications, 20, 25–41. Hohensinner S., Jungwirth, M., Muhar, S. and Habersack, H. (2005) Historical analyses: a foundation for developing and evaluating river-type specific restoration programs. International Journal of River Basin Management 3(2), pp. 87–96. Hötzl, H. (1996) Origin of the Danube-Aach systems. Environmental Geology 27, pp. 87–96. ICIM. (2006) Environmental Impact Assessment Report for Cernavoda NPP Units 3 and 4. Chapter 9. Non-technical summary. National Institute of Research and Development for Environmental Protection (ICIM), Bucharest. 9 p. ICJ. (1997) Reports of Judgments, Advisory Opinions and Orders: Case Concerning the GabčíkovoNagymaros Project. International Court of Justice, The Hague. 81 p. http://www.icj-cij.org/files/case-related/92/092-19970925-JUD-01-00-EN.pdf

ICPDR. (2002) Joint Danube Survey: 2581 km along the Danube. Technical Report. English summary. International Convention for the Protection of the Danube River Basin, 26 p. ICPDR. (2008) Joint Danube Survey 2. Final Scientific Report. Liška, I., Wagner, F. and Slobodník, J. (Eds). International Commission for the Protection of the Danube River, Vienna. 242 p. ICPDR. (2015a) Joint Danube Survey 3. Final Scientific Report. Liška, I., Wagner, F., Sengl, M., Deutsch, K. and Slobodník, J. (Eds). International Commission for the Protection of the Danube River, Vienna. 369 p. ICPDR. (2015b) The Danube River Basin District Management Plan. Update 2015. International Convention for the Protection of the Danube River Basin, Vienna. 192 p. https://www.icpdr.org/flowpaper/viewer/default/files/nodes/documents/drbmp-update2015.pdf Jankovič, D. and Jovičič, M. (eds) 1994. The Danube in Yugoslavia: contamination, protection and exploitation. Belgrade, 219 p. Jerz, H. (1995) Bayern. In: Das Quartär Deutschlands (L. Benda, Ed.). Gebrüder Borntraeger, Berlin, pp. 296–326. Jungwirth, M., Haidvogl, G., Hohensinner, S., Waldbacher, H. and Zauner, G. (2014) Österreichs Donau – Landschaft – Fisch – Geschichte. Institut für Hydrobiologie und Gewässermanagement, BOKU, Wien. 420 p. Karátson, D. (2015) Danube Bend: Miocene Half-Caldera and Pleistocene Gorge. In: Landscapes and Landforms of Hungary (D. Lóczy Ed.). Springer International Publishing, Cham, Switzerland. pp. 129– 137. Kavasch, J. (1992) Meteoritenkrater Ries. Ein geologischer Führer. 10. Mehrfach überarbeitete Auflage. Ludwig Auer Verlag, Donauwörth, 112 p. Korpás, L. (Ed.) (1998) Magyarázó a Börzsöny és a Visegrádi-hegység földtani térképéhez, 1 : 50 000 (Memoir to the 1 to 50 000 Geological Map ‘Börzsöny and Visegrád Mountains’), Budapest, 20 p. Kovács, I.P., Fábián, S.Á., Radvánszky, B. and Varga, G. (2015) Dunaszekcső Castle Hill: Landslides along the Danubian loess bluff. In: Landscapes and Landforms of Hungary (D. Lóczy Ed.). Springer International Publishing, Cham, Switzerland. pp. 113–120. doi: 10.1007/978-3-319-08997-3_14 Kraus-Kalveit, I. and Pannonhalmi, M. (2001) Vom Rhein zur ungarischen Donau. I–II. Ministry of the Environment and Forestry, Rhineland–Palatinate, Mainz. Kubryakov, A.A. and Stanichnyi, S.V. (2012) The Black Sea level trends from tide gages and satellite altimetry. Russian Meteorology and Hydrology 38(5), pp. 329–333. Lászlóffy, W. (1965) Die Hydrographie der Donau (Der Fluß als Lebensraum). In: Limnologie der Donau 1 (R. Liepolt, Ed.). Schweitzerbart’sche Buchhandlungverlag, Stuttgart, pp. 16– 57. Lehoczky, J. (1979) The Danube river quality alterations in the course of natural infiltration. Proceedings, Third Conference on Water Quality and Technology, Technical University, Budapest, pp. 1–14. Lehotský, M., Novotný and J. Szmańda 2009. Flooding and geomorphic evolution of modern floodplain: Case study of the Danube River downstream of the Devíńska Gate. Revue Roumaine de Géographie, 53(1), pp. 81–90. Leuven, R.S.E.W., van der Velde, G., Baijens, I., Snijders, J., van der Zwart, C., Lenders, H.J.R. and bij de Vaate, A. (2009). The river Rhine: a global highway for dispersal of aquatic invasive species. Biological Invasions, 11, 1989–2008. doi: 10.1007/s10530-009-9491-7 Liška, I. (2015) Managing and international river basin towards water quality protection: the Danube case. In: (Liška, I. Ed.) The Danube River Basin. Springer-Verlag, Heidelberg – Berlin. pp. 1–21. Literáthy, P. (2001) Laboratory Quality Management for Improving Water Pollution Monitoring of Transboundary Rivers – the Case of the Danube River Basin. Water Resources Research Centre (VITUKI), Budapest, 13 p. Lóczy, D. and Balogh, J. (1990) Geo-ecological mapping in Hungary from satellite imagery. Zeitschrift für Geomorphologie, Suppl. 80, pp. 7–16. Lóczy, D., Balogh, J. and Ringer, Á. (1989) Landslide hazard induced by river undercutting along the Danube. Geografia Física e Dinamica Quaternaria, Suppl. II, 5–11. Lóczy, D. and Juhász, Á. (1997) Hungary. In: Geomorphological Hazards of Europe (C. Embleton, and C. Embleton-Hamann, Eds). Elsevier, Amsterdam, pp. 243–262. Lóczy, D., Mátrai, I., Fehér, G. and Váradi, Z. (2014) Ecological Evaluation of the Baja-Bezdan Canal (Hungary-Serbia) for Reconstruction Planning. Water Resources Management, 28(3) pp. 815–831. doi: 10.1007/s11269-014-0517-2

Lovász, Gy. (2012) Water temperatures of the Danube and Tisza Rivers in Hungary. Hungarian Geographical Bulletin, 61(4), 317–325. Mahr, T. and Šajgalik, J. (1979) Vývoj, zloženie a vlasnosti štvrtohorných sedimentov zapadnej casti Bratislavy (Development, composition and properties of Quaternary sediments in the W of Bratislava). Geologické práce, Správy 73(1), 161–172. (English summary) Malciu, V. (2000) Implications of accelerated sea-level rise (ASLR) for Romania. Proceedings of SURVAS Expert Workshop on European Vulnerability and Adaptation to Impacts of Accelerated Sea-Level Rise (ASLR), Hamburg, 19–21 June 2000, pp. 23–24. Marsigli, L.F. (1726) Pannonico-mysicus observationibus, Vols I–VII. The Hague. Mladenović, M.B., Kolarov, V. and Damjanović, V. (2013) Sediment regime of the Danube River in Serbia. International Journal of Sediment Research 28(4), pp. 470–485. Nemerkényi, A. (1990) Die Staustufe Gabčíkovo-Nagymaros an der Donau. Geographische Rundschau 42(6), pp. 346–350. Orghidan, N. (1966) Die Donau und das Eiserne Tor. Revue Roumaine, Serie Géographie 10(1), pp. 29–37. Panin, N. (1989) Danube delta. Genesis, evolution and sedimentology. Revue Roumaine de Géologie, Géophysique, Géographie 33, pp. 25–36. Panin, N., Panin, S., Herz, N. and Noakes, J.E. (1983) Radiocarbon dating of Danube delta deposits. Quaternary Research 19, pp. 249–255. Panin, N. (2003) The Danube Delta. Geomorphology and Holocene Evolution: a Synthesis. Géomorphologie : relief, processus, environnement, 9(4), pp. 247–262. Pécsi, M. (1959) A magyarországi Duna-völgy kialakulása és felszínalaktana (Zusammenfassung: Entwicklung und Morphologie des Donautales in Ungarn). Akadémiai Kiadó, Budapest, 345 p. Pécsi, M. (1999) Fluvial landforms. In: Landform evolution studies in Hungary (M. Pécsi, Ed.). Akadémiai Kiadó, Budapest, pp. 67–72. Pécsi, M. and Schweitzer, F. (1995) The development of the Hungarian lowland and upland section of the Danube valley. In: Schirmer, W. (ed.) Quaternary Field Trips in Central Europe. INQUA 14th International Congress, August 3–10, 1995, Berlin, Vol. 1. Regional field trips. Verlag Friedrich Pfeil, Munich, pp. 305–310, 316–317. Pécsi, M., Scheuer, Gy. and Schweitzer, F. (1988) Neogene and Quaternary geomorphological surfaces and lithostratigraphical units in the Transdanubian Mountains. In: Paleogeography of Carpathian Regions (M. Pécsi, and L. Starkel, Eds). Geographical Research Institute Hungarian Academy of Sciences, Budapest, pp. 11–42. Pekarova, P., Halmova, D., Miklanek, P., Onderka, M., Pekar, J. and Skoda, P. (2008) Is the water temperature of the Danube River at Bratislava, Slovakia, rising? Journal of Hydrometeorology, 9, 1115–1122. doi: 10.1175/2008JHM948.1 Pešić, A.M., Kovačević-Majkić, J., Milivojević, M. (2010) Water quality analysis of Danube river in Serbia: Pollution and protection problems. Bulletin of the Serbian Geographical Society 90(2), pp. 47– 68. doi: 10.2298/GSGD1002047M Pistocchi, A., Beck, H., Bisselink, B., Gelati, E., Lavalle, C. and Fehér, J. (2015) Water scenarios for the Danube River Basin: Elements for the assessment of the Danube agriculture-energy-water nexus. EUR 27700 EN. European Union Joint Research Centre, Ispra, Italy. 51 p. doi:10.2788/375680 Pišút, P. (2002) Channel evolution of the pre-channelized Danube River in Bratislava, Slovakia (1712– 1886). Earth Surface Processes and Landforms 27, pp. 369–390. Poórová, J., Döményová, J. and Ďurkovičová, D. (2017) Assessment of the balance of waste water discharged into surface flows in Slovakia. 27th Conference of the Danubian Countries on Hydrological Forecasting and Hydrological Bases of Water Management, 26–28 September 2017, Golden Sands, Bulgaria. pp. 595–599. Popp, N. (1974) A Duna negyedkori fejlődéséről alkotott szintézis általános eredményei Romániában (A Synthesis of Research on the Quaternary Evolution of the Danube in Romania). Földrajzi Értesítő, 23(1), pp. 19–25. Preoteasa, L., Vespremeanu-Stroe, A., Tătui, F., Zăinescu, F., Timar-Gabor, A. and Cîrdan, I. (2016) The evolution of an asymmetric deltaic lobe (Sf. Gheorghe, Danube) in association with cyclic development of the river-mouth bar: Long-term pattern and present adaptations to human-induced sediment depletion. Geomorphology, 253, pp. 59–73. doi: 10.1016/j.geomorph.2015.09.023 Pringle, C., Vellidis, G., Heliotis, F., Bandacu, D. and Cristofor, S. (1993) Environmental problems of the Danube Delta. American Scientist 81, p. 350.

Radmilović, Z. and Maraš, V. (2011) Role of Danube inland navigation in Europe. International Journal for Traffic and Transport Engineering, 1(1), pp. 28–40. Radović, I. and Stevanović, V. (1996) Biodiversity in the Yugoslav sector of the Danube. Proceedings of International Senckenberg Conference “Global biodiversity research in Europe”, 9–13 December 1996, Frankfurt/Main, Germany Rîmbu, N., Boronean, C., Buţa, C. and Dima, M. (2002) Decadal variability of the Danube River flow in the lower basin and its relation with the North Atlantic Oscillation. International Journal of Climatology 22, 1169–1179. doi: 10.1002/joc.788 Romanescu, G. (1996) L’évolution hydrogéomorphologique du delta du Danube. Etape Pleistocene– Holocene inférieur. Zeitschrift für Geomorphologie Suppl. 106, 267–295. Romanescu, G. and Stoleriu, C. (2013) Causes and effects of the catastrophic flooding on the Siret River (Romania) in July–August 2008. Natural Hazards, 69(3), pp. 1351–1367. doi: 10.1007-11069-0120525-6 Romanescu, G. and Stoleriu, C. (2017) Exceptional floods in the Prut basin, Romania, in the context of heavy rains in the summer of 2010. Natural Hazards and Earth System Sciences, 17, pp. 381–396. doi: 10.5194/nhess-17-381-2017 Rutte, E. (1987) Rhein, Main, Donau: Wie – wann – warum sie wurden. Eine geologische Geschichte. Jan Thorbecke Verlag, Sigmaringen, 154 p. Ryan, W.B.F., Pitman III, W.C., Major, C.O., Shimkus, K., Moskalenko, V., Jones, G.A., Dimitrov, P., Görür, N., Sakinç, M. and Yuce, H. (1997) An abrupt drowning of the Black Sea shelf. Marine Geology 138, pp. 119–126. Schiemer, F., Baumgartner, C., Tockner, K. (1999) Restoration of floodplain rivers: The 'Danube Restoration Project'. Regulated Rivers: Research and Management, 15, pp. 231–244. Schwabach, A. (1996) Diverting the Danube: The Gabcikovo-Nagymaros dispute and international freshwater law. Berkeley Journal of International Law, 14(2), 290–343. doi: 10.15779/Z38RM0F Schwarz, U. (2015) Hydromorphology of the Danube. In: The Danube River Basin. (Liška, I. Ed.). Springer-Verlag, Heidelberg – Berlin. pp. 469–480. Schwarz, U. (2018) Hydromorphology of the Lower Drava. In: The Drava River: Environmental Problems and Solutions (D. Lóczy Ed.). Springer International Publishing, Cham, Switzerland Schweitzer, F. (1997) On late Miocene–early Pliocene desert climate in the Carpathian Basin. Zeitschrift für Geomorphologie, Suppl. 110, pp. 37–43. Sommerhäuser, M., Robert, S., Birk, S., Hering, D., Moog, O., Stubauer, I. and Ofenböck, T. (2003) Developing the typology of surface waters and defining the relevant reference conditions. Final Report. UNDP/GEF Danube Regional Project “Strengthening the implementation capacities for nutrient reduction and transboundary cooperation in the Danube River Basin.” 97 p. Sommerwerk N., Baumgartner C., Bloesch J., Hein T., Ostojić A., Paunovic M., Bloesch J., Siber R. and Tockner K. (2009) The Danube River Basin. In: Rivers of Europe (K. Tockner, U. Uehlinger and C.T. Robinson Eds). Elsevier/Academic Press, Amsterdam. pp. 59–112. doi: 10.1016/B978-0-12-369449-2.00003-5 Somogyi, S. (ed.) (2000) A XIX. századi folyószabályozások és ármentesítések földrajzi és ökológiai hatásai (Geographical and ecological impacts of flow regulation and drainage in the 19th century). Geographical Research Institute, Hungarian Academy of Sciences, Budapest, 302 p. Stančíkova, A. (2001) A Duna szabályozása (Regulation of the Danube). Vízügyi Közlemények 83(3), pp. 451–472. Stančíkova, A. (2010) Thermal and ice regimes of the Danube River and its tributaries. In: Hydrological Processes of the Danube River Basin: Perspectives from the Danubian Countries (M. Brilly, Ed.). Springer Sciences + Business, Dordrecht. pp. 259–292. Stevović, I., Jovanović, J., Stevović, S. (2017) Sustainable management of Danube renewable resources in the region of Iron Gate: Djerdap 1, 2 and 3 case study. Management of Environmental Quality, 28(5), pp. 664–680. doi: 10.1108/MEQ-07-2014-0114 Sümeghy, J. (1953) Medencéink pliocén és pleisztocén rétegtani kérdései (Issues concerning the Pliocene and Pleistocene stratigraphy in Hungarian basins). Annual Report of the Hungarian Geological Institute for 1951, Budapest, pp. 83–109. Takács, K., Kern, Z. and Nagy, B. (2013) Impacts of anthropogenic effects on river ice regime: Examples from Eastern Central Europe. Quaternary International, 293, pp. 275–282.

Tamás E.A. (2018) Water and Sediment Discharge along the Hungarian Section. In: The Drava River: Environmental Problems and Solutions (D. Lóczy Ed.). Springer International Publishing, Cham, Switzerland Török, L. (2012) Data on influence of cooling water discharge from Cernavodă Nuclear Power Plant on Danube River phytoplankton development. Scientific Annals of the Danube Delta Institute, Tulcea, Romania 18, 249–260. Tzonev, R. (2015) Danubian sand dunes and inland dunes (gredove) from sandy loess. In: Red Data Book of the Republic of Bulgaria. Volume 3. Natural habitats (V. Biserkov Editor-in-Chief). Bulgarian Academy of Sciences – Ministry of Environment and Waters, Sofia. pp. 161–162. UNEP. (1991) World Heritage Sites. Danube Delta Biosphere Reserve. Protected Areas Programme, UNEP World Conservation Monitoring Centre, Cambridge, 11 p. UNESCO. (1999) The Danube and its Basin – a Hydrologic Monograph. Followup Volume 5. The Danube River Bed Conditions. Regional Cooperation of the Danube Countries in the frame of the International Hydrological Programme of UNESCO. VÚVH, Bratislava, 186 p. Vadineanu, A. and Voloshyn, V. (eds) (1999) The Danube Delta Biosphere Reserve Romania/Ukraine. National UNESCO–MAB Committee of Romania, Bucharest, and National UNESCO–MAB Committee of Ukraine, Kyiv, 46 p. Vespremeanu-Stroe, A., Zăinescu, F., Preoteasa, L., Tătui, F., Rotaru, S., Morhange, C., Stoica, M., Hanganu, J., Timar-Gabor, A., Cârdan, I. and Piotrowska, N. (2014) Holocene evolution of the Danube delta: An integral reconstruction and a revised chronology. Marine Geology, 388, 38–61. doi: 10.1016/j.margeo.2017.04.002 Villinger, E. (1986) Untersuchungen zur Flussgeschichte von Aare-Donau/Alpenrhein und zur Entwicklung des Malm-Karsts in Südwestdeutschland. Jahresheft des geologischen Landesamts BadenWürttemberg, Freiburg im Breisgau 28, 297–362. Villinger, E. (1998) Zur Flussgeschichte von Rhein und Donau in Südwestdeutschland. Jahresbericht und Mitteilungen Oberrheinischer Geologischer Verein, Neue Folge 75, Stuttgart, pp. 291–312. Werner, A., Hötzl, H., Käss, W. and Maloszewski, P. (1997) Interpretations of Tracer Experiments in the Danube–Aach System, Western Swabian Alb, Germany, with analytical models. In: Karst Waters and Environmental Impacts (Y. Günay, and J. Johnson, Eds). Balkema, Rotterdam, pp. 153–160. Yanchilina, A., Ryan, W.B.F., McManus, J.F., Dimitrov, P., Dimitrov, D., Slavov, K., Filipova-Marinovac, M. (2017) Compilation of geophysical, geochronological, and geochemical evidence indicates a rapid Mediterranean-derived submergence of the Black Sea's shelf and subsequent substantial salinification in the early Holocene. Marine Geology, 383, pp. 14–34. doi: 10.1016/j.margeo.2016.11.001 Zoran, M., Nicolae, D., Talianu, C., Ciobanu, M. and Ciuciu, J.G. (2005) Analyses of thermal plume of Cernavoda nuclear power plant by satellite remote sensing data. Proceedings of SPIE - The International Society for Optical Engineering 5983, 115–122. doi: 10.1117/12.627709