Time-focused investigation of river channel morphological changes due to extreme floods

Abstract

The typical morphological response of the meandering rivers to large floods is the lateral shift of their channel which triggers the formation of a new morphological structure from the initial destruction by erosion over deposition of new sediments and stabilization of vegetation. The article deals with the effect of extreme flood events on lateral channel shift and bar pattern with relation to changes of the riparian zone land cover structure by using multitemporal analyzes of aerial photographs (three time horizons – 1987, 2002 and 2009) in the GIS environment on the example of the 13.2 km long less regulated and laterally partly-confined meandering reach of the Ondava River (Eastern Slovakia). The photographs were chosen in a way to capture the morphological changes that occurred after floods. The average lateral channel shift per year was 1.17 m in 1987–2009, maximum 217 m. The river has eroded in total 35.6 ha and deposited 31.6 ha. Gravel bars in 1987, 2002 and 2009 spread a channel total area of 21.1 ha, 17.8 ha and 19.7 ha. The most eroded category is that of arable land, followed by grasslands and pastures and shrubs. We conclude that in case of the Ondava River, low magnitude high frequency floods, instead of causing destruction of the system, led to the stabilization of the channel, erosion of the concave bank and to the formation of the meandering planform. In contrast, the short recurrence interval of extreme floods led to an increased in tensity of erosion processes, a change of the meandering planform to slightly braided one, straightening of the channel and formation of gravel bars.

Full text

Time-focused investigation of river channel morphological
changes due to extreme floods
MiloáRusnák and Milan Lehotské
with 9 figures and 4 tables
Summary. The typical morphological response of the meandering rivers to large floods is the
lateral shift of their channel which triggers the formation of a new morphological structure
from the initial destruction by erosion over deposition of new sediments and stabilization of
vegetation. The article deals with the effect of extreme flood events on lateral channel shift and
bar pattern with relation to changes of the riparian zone land cover structure by using multi-
temporal analyzes of aerial photographs (three time horizons – 1987, 2002 and 2009) in the GIS
environment on the example of the 13.2 km long less regulated and laterally partly-confined
meandering reach of the Ondava River (Eastern Slovakia). The photographs were chosen in
a way to capture the morphological changes that occurred after floods. The average lateral
channel shift per year was 1.17 m in 1987–2009, maximum 217 m. The river has eroded in total
35.6 ha and deposited 31.6 ha. Gravel bars in 1987, 2002 and 2009 spread a channel total area
of 21.1 ha, 17.8 ha and 19.7 ha. The most eroded category is that of arable land, followed by
grasslands and pastures and shrubs. We conclude that in case of the Ondava River, low mag-
nitude high frequency floods, instead of causing destruction of the system, led to the stabi-
lization of the channel, erosion of the concave bank and to the formation of the meandering
planform. In contrast, the short recurrence interval of extreme floods led to an increased in -
tensity of erosion processes, a change of the meandering planform to slightly braided one,
straightening of the channel and formation of gravel bars.
Key words: bank shift, flood, multitemporal, GIS
1Introduction
One of the key factors affecting the behaviour of a river channel is the extreme dis-
charge, which increases the energy of the stream. It leads to the intensification of
the erosion and deposition processes in channel and floodplain. Geomorphological
effects of a flood also depend on its magnitude, frequency and timing, as well as phys-
ical properties of the channel and banks (floodplain) (Miller 1990). Ward (1978)
pointed to the significance of flood frequency. As the interval between two flood
events shortens, the effect of the second flood becomes more significant due to lack
of time for the channel to recover (for instance, by stabilization of vegetation) after
the first flood. Michalková et al. (2011) associate the bank erosion first of all with a
more frequent reaching of critical shear stress under the effect of more frequent and
longer floods. Meanwhile, it is important to distinguish the effect of floods in terms
of the flooding (inundation) and the effect caused by bank erosion (Fuller 2005).
Hickin & Sichingabula (1988) report that in general, the greater geomorphologi-
Zeitschrift für Geomorphologie Vol. 58,2 (2014), 251–266 Article
published online November 2013; published in print June 2014

©2013 Gebr. Borntraeger Verlagsbuchhandlung, Stuttgart, Germany www.borntraeger-cramer.de
DOI: 10.1127/0372-8854/2013/0124 0372-8854/13/0124 $ 4.00
eschweizerbart_xxx

cal effect is connected with relatively smaller but longer duration floods and empha-
size the significance of the duration of the extreme event. Bertoldi et al. (2009)
researching on the River Tagliamento, (Italy) found out that the discharge higher than
bankfull discharge is responsible for significant erosion of the floodplain and edges
of river islands; it leads to the formation of new avulsion channels. Smaller discharges
below bankfull levels form the channel and the rare larger discharges initiate the
transport of coarse-grained sediment, bank erosion and channel destruction (Phil -
lips 2002). While high magnitude floods are rather destructive causing lateral shift-
ing of channels and remodeling of the floodplain, effects of low magnitude and high
frequency floods are constructive as they contribute to lateral accretion of sediments
and stabilization of the fluvial system (Corenblit et al. 2007a).
The main response of the river system to a flood, typical for the meandering
rivers is the lateral shifting of channels. This phenomenon reflects the relationship
between the bank erosion processes and the formation of the landforms attached to
the bank (Hickin 1974), i. e. it is the result of the interaction between the aggradation
and degradation and bank erosion/deposition processes. Lateral shift of the mean-
dering channel is generally seen as natural hazard because it threatens artificial man-
made landscape structures (i. a. roads, arable land, and settlement). Russell et al.
(2004) report that lateral dimension of the stream is subject to interaction between
floods (magnitude, frequency, duration and timing) and other factors influencing its
erodibility: bank strati graphy, type and density of vegetation, wood debris, bank
drainage and geo-technical characteristics. Formation of new (deposition) and geo-
morphological changes (erosion) of the original channel and bank landforms leads to
succession and rejuvenation as being the dominant processes in the changing vegeta-
tion of the riparian zone (Geerling et al. 2006). The vegetation cover during low
magnitude floods prevents bank erosion, channel degradation and in the period
between floods it stabilizes the individual geomorphological landforms (mostly
gravel bars) thus fulfilling one of the key functions in the formation of the river
ecosystem structure (Corenblit et al. 2007b, Ward et al. 2002).
Remote sensing data are a perfect tool for monitoring the changes taking place
in the riverine landscape, identification of bank erosion (Michalková et al. 2011),
effects of vegetation on erosion (Micheli et al. 2004) and development of gravel
bars (Hooke & Yorke 2010). They capture the dynamics of vegetation succession
(Gilvear & Willby 2006, Corenblit et al. 2010) and spatial changes in the struc-
ture of the riparian zone with high accuracy (Gilvear et al. 2004, Bryant & Gilvear
1999). They offer the possibilities of identifying changes in-channel landforms, chan-
nel lateral migration, deposition on bars and dynamics in the development of the
channel and floodplain.
The aim of this article is to analyze the lateral channel shift and the associated
changes in in- channel morphological structures and riparian land cover ones by
multi-temporal analyzes of aerial photographs in the GIS environment on an exam-
ple of less regulated and laterally partly-confined meandering channel.
2Study area
Research focused on a selected untrained reach of the gravel-bed meandering the
Ondava River in the middle part of the basin. The total length of the reach was
M. Rusnák and M. Lehotské
252
eschweizerbart_xxx

13.2 km (in 2009) (fig. 1). The basin of the Ondava is situated in the central part of
the Ondavská vrchovina Upland built of flysch sediments. The basin relief steadily
declines from the north to the south and the main valley is mostly N-S oriented par-
tially copying old geological structures. The dendritic character of the river network
has gained that of a fan-like shape in the headwater area under the effect of river back-
erosion and river piracy.
The Ondava springs 546 m a.s.l. in the Slovak-Polish border area and the length
of its flow is 144.4 km as far as the confluence with the River Latorica. The area of
the basin amounts to 3,354.7 km2and the long term average discharge taken in the
gauging station of Stropkov is Qa= 5,730 m3.s–1 (Qmax = 550 m3.s–1 /19.7. 1974/). The
Ondava in the study reach is a 6th order stream (Strahler order). It is characterized by
the high variability of discharges as proved by the fact that discharge with recurrence
interval of 1 year (N1) represents a 25–30 multiple of the long-term mean annual dis-
charge and the flood discharges reach even the 95-multiple of the long-term mean dis-
charge (for the max. culmination discharge 550 m3.s–1 from 1974). The top water lev-
els due to snow thaw are reached at the beginning of spring (March).
The daily discharge data registrated at the gauging station Stropkov was the
source for the analysis of the frequency and magnitude of flood events (source
SHMÚ – Slovak hydro-meteorological institute). The graph (fig. 2) makes it possible
to determine two time periods when extreme floods occurred recently. The first one
(1987– 1992) when the greatest large flood events occurred in 1987, 1989, and 1992 with
the maximum culmination discharge of 350 m3.s–1 (15th June, 1987, N10–20; 1989
/300 m3.s–1/ and 1992 /285 m3.s–1/ at the discharge of N5–10). After 1992, a period of de-
creased flood activity characterized by discharge with recurrence interval of 1–2 years
(9 years to the most significant event in 2001) followed. During the second time pe-
riod (2001–2008) extreme floods also occurred in 2001 (235 m3.s–1, N5), 2004 (Q =
230 m3.s–1, N5) and first of all in 2006 (culmination discharge Q = 300m3.s–1 /N5–10/).
3Data and methods
Morphological changes of river channel and riparian zone as the response to flood
events were investigated in forth time horizons (table 1). The first one in 1987 cap-
turing the morphological response to flood in 1987 and serving as the reference hori-
zon. The second horizon (2002) studying responses to floods in 1989 and 1992 fol-
lowed by the phase of reduced flood activity and third one (2009) investigating mor-
phological effect of floods in the 2004 and 2006 and the same one as forth horizon for
the study morphological response to all floods occurs during the hole time span
1987–2009. Black-and-white aerial photos (taken on July 7, 1987 and August 23,
1987, pixel resolution 0.5 m) and colour orthophotos (second half of June 2002 with
pixel resolution 1 m and from April 15, 2009, pixel size of 0.89 m), were the source of
spatial data. The photographs were chosen in a way to capture the morphological
changes that occurred in the study period after floods and development of the chan-
nel after the 1987 (Q = 350 m3.s–1) extreme event.
Lines of the right and left banks of the channel, areas of four types of in-chan-
nel landforms (1. lateral bars, 2. point bars, 3. central bars, 4. islands) and areas of five
categories of land cover in the riparian zone (1. Riparian forest – tree cover
⬎
70%,
2. Shrubs, 3. Grassland and pastures, 4. Arable land, 5. Urbanized areas) were
River channel morphological changes due to extreme floods 253
eschweizerbart_xxx

M. Rusnák and M. Lehotské
254
Fig. 1. Localiza-
tion the Ondava
River catchment
and study reach
of river.
eschweizerbart_xxx

obtained by digitization of raster sources. The bank line was identified on images
either as a distinct channel edge or the contact line of the channel with the riparian
vegetation or the contact line of the gravel bar attached to the bank with riparian veg-
etation. The area of this delimited channel is considered the ground plan projection
of the bankfull channel.
The lateral channel shift in the ArcGIS has been established for spatial segments
generation based on the perpendicular to the central line with 250 m separation (after
methodology Lehotskéet al. 2013). The direction and the size of the shift of the left
and right banks in meters were identified (fig. 3). Areas of in-channel forms, areas of
erosion and deposition as well as areas of the reworked floodplain were generated by
an overlay of the vector source data obtained by digitisation in the ArcGIS. Changes
in land cover categories in the riparian zone due to the lateral channel shift were
analysed in the ArcGIS 20-m wide generated buffer zone stretching along the right
River channel morphological changes due to extreme floods 255
Table 1. Four time horizons, years of flood events occurrences and their recurrence inter-
val.
Time N10–20 N5–10 N5N1–2
horizon Q (350 Q (from 285 to Q (235 and Q (from 141 to 210 m3.s–1)
m3.s–1) 300 m3.s–1) 230 m3.s–1)
1st (1987) 1987
2nd (2002) 1989, 1992 2001 1993, 1995, 1996, 1998,
1999, 2000
3rd (2009) 2006 2004 2005, 2008
4rd (2009) 1987 1989, 1992, 2006 2001, 2004 1993, 1995, 1996, 1998,
1999, 2000, 2005, 2008
Fig. 2. Variability of daily discharge in period 1987–2009 on the Ondava River recorded in
Stropkov gauging station and discharge of flood events in this time period.
eschweizerbart_xxx

bank and left bank areas of the channel delimited by the unification of its area in 1987,
2002, and 2009. All data were further specified during field campaigns in 2011 (April–
May) and 2012 (May–June).
4Results
4.1 Lateral shift of the channel and changes in in-channel landforms
The overall average shift of the Ondava channel oscillated between 21.7 m (1987–
2002) and 8.1 m (2002–2009) (table 2). The maximal bank shift is 217 which is the
result of channel avulsion. The mean channel shift per year (1987–2009) is 1.17 m/
year and the mean movements of individual banks are 1.07 m/year (left bank), and
1.27 m/year (right bank).
The river has eroded in total 35.6 ha and deposited 31.6 ha since 1987 (table3).
The other 5.3 ha represent the territory of the reworked floodplain, which was eroded
M. Rusnák and M. Lehotské
256
Fig. 3. Diagram illustrates a process of evaluation of lateral shift, areas of deposition, erosion,
reworked floodplain and method of identification the direction of bank’s movement (A); and
delineation of spatial segment 250 meters long (B).
eschweizerbart_xxx

River channel morphological changes due to extreme floods 257
Table 2. Lateral shift of the Ondava River channel in 1987–2002. 2002–2009 and 1987–2009.
Time period lateral shift (m) lateral shift per year (m/year)
total*left right channel channel max. total*left right channel channel
bank bank left** right*** shift bank bank left** right***
1) 1987–2002 21.7 18.9 24.4 12.9 8.7 158 1.45 1.26 1.63 0.86 0.58
2) 2002–2009 8.1 7.9 8.2 4.4 3.7 92 1.15 1.13 1.17 0.63 0.52
3) 1987–2009 25.7 23.6 27.8 15.3 10.4 217 1.17 1.07 1.27 0.70 0.47
*average lateral shift of channel; ** average shift of channel to left side; *** average shift of channel to right side
Table 3. Erosion and deposition area and rate of erosion/deposition per year between years 1987–2002. 2002–2009 and 1987–2009.
Time period Erosion area Deposition area reworked m2/year
floodplain
total (m2) proportion total (m2) proportion (m2) depos./year erosion/year
(%)*(%)**
1) 1987–2002 204,281 43.3 260,272 49.4 32,412.6 17,351 13,619
2) 2002–2009 152,178 26.8 55,661 11.8 0.0 7,952 21,740
3) 1987–2009 356,459 62.8 315,933 59.9 53,220.6 14,361 16,203
*percentage proportion of erosion from new channel; ** percentage proportion of deposition from old channel
eschweizerbart_xxx

and again deposited and stabilized by vegetation in the same period. In the first
period, deposition area equaled a 1.27 multiple of erosion one while in the following
period it was only 0.37 (in total, in 1987–2009 the deposition area was a 0.86 multi-
ple of erosion area). In the second period, the area of deposition dramatically dropped
(from 17,351 m2/year between 1987–2002 to 7,952 m2/year between 2002–2009) and
erosion increased (from 13,619 m2/year to 21,740m2/year).
Lateral shift also led to changes of the in-channel structure (fig. 4). Gravel bars
in 1987, 2002 and 2009 spread a total area of 21.1 ha, 17.8 ha and 19.7 ha respectively,
which represents 39.9%, 37.7%, and 34.7% of the channel area. The greatest area
of the four studied types of bars is occupied by the lateral bars (12.4 ha, 5.8 ha and
10.0 ha (in 1987, 2002, and 2009 respectively). The second most spread type was
the point bar on an area of 5.9 ha, 11.4 ha, and 5.1 ha. Central bars were identified
on smaller areas of 2.2 ha/0.4 ha/4.5 ha in 1987/2002/2009 respectively and so were
islands (on 0.6 ha/0.2ha and 0.2 ha).
4.2 Lateral channel shift related to riparian zone land cover changes
Riparian forest category is most abundantly represented in the land cover of the ripar-
ian zone (defined by the 20 m buffer). In contrast, the areas of arable land and grass-
land occupy distinctly smaller area (table 4).
Lateral channel movement eroded most the areas of riparian forest (fig. 5a). Over
the total period of study, 20.5 ha of the above mentioned land cover category eroded
(11.2 ha in 1987–2002a 9.3 ha between 2002 and 2009, i. e. 47.2% and 61 % of the total
eroded area respectively). The second most eroded land cover category is arable land
(in total 13.4 ha (8.6 ha in the first/36.4%/a 4.8 ha in the second time period/31.3 %)).
Moreover, 4.3ha of grasslands (3.3 ha between 1987 and 2002 and 1 ha in 2002–2009)
M. Rusnák and M. Lehotské
258
Fig. 4. Changes in percentage proportion of lateral bars, point bars, central bars and islands
in years 1987, 2002 and 2009.
eschweizerbart_xxx

River channel morphological changes due to extreme floods 259
Table 4. Riparian land cover structure in years 1987, 2002 and 2009.
year/land riparian shrubs grassland arable urbanized channel
cover forest and land area
category(ha) pastures
1987 44.6 3.7 12.5 23.2 0.8 52.7
2002 69.4 0.6 6.7 13.0 0.8 47.1
2009 66.5 0.1 5.3 8.1 0.8 56.8
Fig. 5. Erosion of land cover categories (in percentage) in the riparian zone (20 m buffer zone
from channel) (a) and from the area of the individual land cover type (b) between years 1987–
2002 and 2002–2009. (land cover categories: 1) riparian forest, 2) shrubs, 3) grassland and pas-
ture, 4) arable land, 5) urbanized area).
eschweizerbart_xxx

and 0.8 ha of shrubs (0.6 ha in the first time period and 0.2 ha in the second one) were
eroded.
The most eroded landscape category is that of riparian forest. This category is
also the most abundantly represented one in the riparian zone. Its proportion in indi-
vidual time horizons was 32.4% (1987), 50.4% (2002), and 48.3% (2009) of the ripar-
ian zone (fig. 6a). If only the proportions of individual land cover categories are taken
into account (minus the channel areas) then the representation of riparian forest
increased from 52.5% (1987), over 76.7% (2002) to 82.3% in 2009 (fig. 6b).
Comparison of eroded proportions of land cover categories in the overall area
of individual land cover categories in riparian zone (fig. 5b), points to the propensity
of given categories to bank erosion. The most eroded category is that of arable land
M. Rusnák and M. Lehotské
260
Fig. 6. Changes in proportion of land cover (in percentage) from the area of riparian zone in
years 1987, 2002 and 2009 (a – with channel, b – without channel). (Land cover categories:
1) riparian forest, 2) shrubs, 3) grassland and pasture, 4) arable land, 5) urbanized area, 6) chan-
nel).
eschweizerbart_xxx

(37.2% in 1987–2002 and 36.8% in 2002–2009 of arable land area in riparian zone).
It is followed by grasslands and pastures (26.3% and 15.0% of area of the given type),
and shrubs (15.6% and 27.2%). Erosion of forest areas is only 25.1% and 13.4% of
the area of their overall area and there is no erosion of the urbanized category.
Individual deposition areas which sprang in the former position of the channel
after its shifting are covered first of all by forest in 99.4% and 98.5% proportions of
the overall deposition area in the relevant time periods 1987–2002, and 2002–2009.
The overall area of forest on this new created areas (deposition areas) is 29.1 ha in
the first and 5.5 ha in the second period. The consequence of such spatial land cover
changes in the riparian zone is the progressive increase of the proportion of forest and
decrease of other land cover categories which are gradually eroded and replaced by
the riparian forest (predominantly consisting of Salix sp. and Populus sp.).
Finally, two different river reaches in terms of behaviour were identified (fig.7).
The first, a northerly one, represents a meandering stream in the central part of a wide
valley (about 1 to 1.5 km) free from lateral limitations with a high sinuosity index
(SI ⬎1.5), characteristic for distinct lateral shifts between 7 and 12 km in fig. 7. The
stream creates a system of meanders (7 bends) with noticeable bank erosion (about
4 to 8 m a year). The southern reach of the study area is characteristic of alternation
of valley-transversal (free) reaches and reaches abutted to valley slopes alternatively
on the left and right sides. Bank erosion and channel shift take place precisely in spots
where the stream passes from the abutted to the free reaches. Lateral movement of
the bank is approximately less than 2 m/year.
5Discussion
The large flood triggers the formation of a new channel and riparian zone pattern
from the initial destruction by erosion over deposition of new sediments and stabi-
lization of vegetation, and finally the succession stage of the riparian forest. In case
of the Ondava River, high frequency, low magnitude floods at the end of the first
time period, (in total six flood from 1993 with recurrence interval 1–2 years) instead
of causing destruction of the channel planform led to its stabilization by erosion of
the concave bank and the meander growth. Corenblit et al. (2010) also reports that
River channel morphological changes due to extreme floods 261
Fig. 7. Downstrem variability of the banks lateral shift (m/y) of the Ondava River in time
periods: (a) 1987–2002, (b) 2002–2009 – left bank, (c) 1987–2002, (d) 2002–2009 – right bank.
eschweizerbart_xxx

low magnitude floods on the River Tech (France) (N2–3) are not destructive as far as
channel pattern. They lead to bio-geomorphological interaction between vegetation
and geomorphological landforms stabilizing the river system. In contrast, the recur-
rent large floods in the second time period led to an increased intensity of erosion
processes, change of the planform, destruction of the meandering pattern, straight-
ening of the channel and formation of gravel bars (remodeled geomorphological land-
forms) (Lehotskéet al. 2013). The short spacing between relatively large floods in
2004 (N5) and particularly in 2006 (N10) did not favour stabilization of the channel in
accordance with Ward (1978). It manifested itself by cutting off the meander loop
during the 2008 flood although it was less intensive (N1–2). Likewise, Surian et al.
(2009) stress the effect of low magnitude floods (N1–2 and less) on partial morpho-
logical changes of the channel and its shape and that of large floods on the total
remodeling of the channel.
The same morphological development resulting in the planform change and
rearrangement of gravel bars is confirmed also by our study river reach. Decreased
representation of lateral bars and increased occurrence of point bars in 2002 is espe-
cially distinct. After floods at the beginning of the first period (end of the 1980s and
beginning of the 1990s) the destructed and straightened channel of the Ondava River
stabilized and meandered (increase of the sinuosity from 1.34 to 1.57) – meander
bends and point bars were formed. After floods in the second period (especially after
2004, 2006 and 2008 floods) the channel was straightened (decrease of sinuosity to
1.41), point bars were cut off, the original channel was silted and bars joined to the
floodplain and the structure of in-channel landforms was changed. Cutting off the
point bars has caused increased occurrence of central bars in Ondava River. The
sporadic appearance of islands is the result of cutting off the point bars covered by
vegetation in a meander bend, or by stabilization of the central bar by succession of
vegetation.
Succession of vegetation on point bars affected also by accumulation of large
woody debris and local topography is characteristic in the development of the study
river reach. Bars become stands for vegetation, which stabilizes them by the pioneer
wood species almost exclusively represented by Salix sp. The succession process
simultaneously takes place in cut off meanders that are gradually filled up by deposits.
The meander cut off in 2008 is the example (fig. 8). The central part of its former point
bar with medium to coarse gravel (φ–2 to –6) is by 0.5 m higher than the water table
in the cutoff. Regarding this elevated vertical position in relation to the new incised
channel, the point bar is attached into the lowest floodplain level invaded by herb
associations. Succession proceeds from the margin to its centre. Vegetation progres-
sively expands on the one hand from the remnant of the original forest cut off dur-
ing the flood in 2008 and on the other hand from the original channel represented by
the cutoff nowadays (more progressive succession). Local topography and grain size
of deposition determine the expansion of succession of willows (high 1–1.5 m in
2011, 4–5 m in 2012) from the oxbow lake (fig.9).
After Corenblit et al. (2010) for greater rejuvenation (river corridor) of the
mediterranean river Tech a discharge bigger than N100 is necessary. Micheli et al.
(2004) report that the lateral migration of a stream is approximately bigger by a half
if the bank is used as arable land compared to alluvial forest. This fact is also evident
in the study area and so is the increase of areas and proportions of the alluvial forest
M. Rusnák and M. Lehotské
262
eschweizerbart_xxx

(table 2), as proved by the study of Cebecauerová & Lehotské(2012) using an
example of another eastern Slovak River. The discharge of approximate size of N10
can be then considered a threshold, which initiates morphological changes of the
channel and local rejuvenation in its meandering reaches in the study area.
6Conclusions
The Ondava represents a recently very dynamic system with distinct lateral instabil-
ity and bank erosion. The trajectory of its changes reflects the present dynamics of
the system with variability of discharges during flood events. Floods influence the
shifting of banks and the change of the planform, which remodels the character of
the riparian zone and the floodplain. Erosion and deposition processes lead to the
destruction of the floodplain on the one hand and formation of new conditions for
plant and animal species (rejuvenation) on the other hand, as well as progressive suc-
River channel morphological changes due to extreme floods 263
Fig. 8. Development of meander bends and meandering planform between 1987–2009. Chute
cutoff of meanders led to straightening of course and to the succession of vegetation on new
created surface/level (former point bars). The cross-section refers to the spread of vegetation
on this new surface from their edges to the center.
eschweizerbart_xxx

cession of vegetation. Vegetation blocks erosion during low magnitude floods and in
the time between floods it stabilizes the individual landforms. The occurrence of fre-
quently repeated low magnitude floods at the end of the 1990s (6 floods sized N1–2
after 1993) in the study area did not change the morphological system, instead it led
to stabilization, erosion of the concave bank and the formation of a meandering plan-
form pattern.
Acknowledgements
The research was supported by Science Grant Agency (VEGA) of the Ministry of Education
of the Slovak Republic and the Slovak Academy of Sciences; 02/0,106/12. The Slovak Hydrom-
eteorological Institute provided the hydrological data.
M. Rusnák and M. Lehotské
264
Fig. 9. Progressing succession of vegetation on former point bar after their cut-off by a flood
event in 2008. First image from date 1 April 2011 and second from 15 June 2012.
eschweizerbart_xxx

References
Bertoldi, W., Gurnell, A. M., Surian, N., Tockner, K., Zanoni, L., Ziliani, L. &
Zolezzi, G. (2009): Understanding reference processes: linkages between river flows,
sediment dynamics and vegetated landforms along the Tagliamento River, Italy. – River
Res. Appl. 25: 501–516.
Bryant, R.G. & Gilvear, D. J. (1999): Quantifying geomorphic and riparian land cover
changes either side of a large flood event using airborne remote sensing: River Tay, Scot-
land. – Geomorphol. 29: 307–321.
Cebecauerová, M. & Lehotské, M. (2012): Komplexita ripariálnej zóny – príklad rurálneho
segment vodného toku Torysa [Complexity of riparian zone – case study of rural seg-
ment of the River Torysa]. – Geografickýčasopis 64(2): 133–154.
Corenblit, D., Steiger, J., Gurnell, A. M., Tabacchi, E. & Roques, L. (2007a): Control of
sediment dynamics by vegetation as a key function driving biogeomorphic succession
within fluvial corridors. – Earth Surf. Process. Landf. 34: 1790–1810.
Corenblit, D., Steiger, J. & Tabacchi, E. (2010): Biogeomorphologic succession dynamics
in a Mediterranean river system. – Ecography 33: 1136–1148.
Corenblit, D., Tabacchi, E., Steiger, J. & Gurnell, A. M. (2007b): Reciprocal interactions
and adjustments between fluvial landforms and vegetation dynamics in river corridors:
A review of complementary approaches. – Earth-Sci. Rev. 84: 56–86.
Fuller, I. C. (2005): February floods in the lower North Island, 2004: Catastrophe – causes
and consequences. – New Zealand Geographer 61: 40–50.
Geerling, G. W., Ragas, A. M. J., Leuven, R. S. E. W., Bergvande, J.H., Breedveld, M.,
Liefhebber, D. & Smiths, A. J. M. (2006): Succession and rejuvenation in floodplains
along the river Allier (France). – Hydrobiologia 565: 71–86.
Gilvear, D., Davids, C. & Tyler, A. N. (2004): The use of remotely sensed data to detect
channel hydromorphology; River Tummel, Scotland. – River Res. Appl. 20: 795–811.
Gilvear, D. & Willby, N. (2006): Channel dynamics and geomorphic variability as controls
on gravel bar vegetation, River Tummel, Scotland. – River Res. Appl. 22: 457–474.
Hickin, E. J. (1974): The development of meanders in natural river-channels. – Am. J. Sci. 274:
414–442.
Hickin, E. J. & Sichingabula, H.M. (1988): The geomorphic impact of the catastrophic
October 1984 flood on the planform of Squamish River, southwestern British Colum-
bia. – Can. J. Earth Sci. 25: 1078–1087.
Hooke, J. M. & Yorke, L. (2010): Rates, distributions and mechanisms of change in meander
morphology over decadal timescales, River Dane, UK. – Earth Surf. Process. Landf. 35:
1601–1614.
Lehotské, M., Frandofer, M., Novotné, J., Rusnák, M. & Szmańda, J.B. (2013): Geomor-
phic/Sedimentary responses of rivers to floods: Case studies from Slovakia. – In: Lóczy,
D. (eds.): Geomorphological impacts of extreme weather: Case studies from central and
eastern Europe: 37–52; Springer.
Michalková, M., Piégay, H., Kondolf, G. M. & Greco, S.E. (2011): Lateral erosion of the
Sacramento River, California (1942–1999), and responses of channel and floodplain lake
to human influences. – Earth Surf. Process. Landf. 36: 257–272.
Micheli, E. R., Kirchner, J. W. & Larsen, E. W. (2004): Quantifying the effect of riparian
forest versus agricultural vegetation on river meander migration rates, Central Sacra-
mento River, California, USA. – River Res. Appl. 20: 537–548.
Miller, A. J. (1990): Flood hydrology and geomorphic effectiveness in the Central Appalachi-
ans. – Earth Surf. Process. Landf. 15: 119–134.
Phillips, J. D. (2002): Geomorphic impacts of flash flooding in a forested headwater basin. –
J. Hydrol. 269: 236–250.
River channel morphological changes due to extreme floods 265
eschweizerbart_xxx

Russell, J. A., Bledsoe, B.P. & Hession, W.C. (2004): Width of streams and rivers in response
to vegetation, bank material, and other factors. – J. Am. Water Resour. As. 40: 1159–1172.
Surian, N., Mao, L., Giacomin, M. & Ziliani, L. (2009): Morphological effects of different
channel-forming discharges in a gravel-bed river. – Earth Surf. Process. Landf. 34: 1093–
1107.
Ward, J.V., Tockner, K., Arscott, D. B. & Claret, C. (2002): Riverine landscape diversity. –
Freshwater Biol. 47: 517–539.
Ward, R. (1978): Floods – A Geographical Perspective. – MacMillan Press, London, 244 p.
Manuscript received: July 2013; accepted: August 2013.
Address of the authors: MilošRusnák, Milan Lehotský, Institute of Geography Slovak Acad-
emy of Sciences, Department of Physical Geography, Geomorphology and Natural Hazards,
Štefánikova 49, 814 73, Bratislava, Slovakia. E-Mail: milos.rusnak@savba.sk
M. Rusnák and M. Lehotské
266
eschweizerbart_xxx