Historic fluvial development of the Alpine-foreland Tagliamento
River, Italy, and consequences for floodplain management
Mathias Spaliviero*
Department of Water Resources, International Institute for Geo-Information Science and Earth Observation (ITC), PO Box 6,
7500 AA Enschede, The Netherlands
Received 25 March 2002; received in revised form 16 August 2002; accepted 10 September 2002
Abstract
The fluvial geomorphological development of the Tagliamento River and its flooding history is analysed using historical
documents and maps, remote-sensed data and hydrological information. The river has been building a complex alluvial fan
starting from the middle part of its alluvial course in the Venetia­Friuli alluvial plain. The riverbed is aggrading over its entire
braided length. The transition from braiding to meandering near Madrisio has shifted downstream where the river width
determined by the dikes becomes narrower, causing major problems. The flood hazard concentrates at those places and zones
where flooding occurred during historical times. Prior to the agrarian and industrial revolution, land use was adjusted to the
flooding regime of the river. Subsequent land-use pressure led to a confinement of the river by dikes to such an extent that the
flood risk in the floodplain downstream of Madrisio has increased consistently, and represents nowadays a major territorial
planning issue. The planned retention basins upstream of the middle Tagliamento will alleviate the problem, but not solve it in
the medium and long term. Therefore, fluvial corridors in the lower-middle parts (from Pinzano to the sea) have been identified
on the basis of the flooding history in relation to fluvial development during historical times. The result should be used for
hydraulic simulation studies and land-use planning.
D 2002 Elsevier Science B.V. All rights reserved.
Keywords: Flood events; Historical analysis; Fluvial dynamics; Fluvial corridors
1. Introduction
Flooding damage caused by the Tagliamento River
has been reported since historical times. Meteorology, a
large area of the upper catchment and steep slopes
confer to the river a torrential regime that generates
flash flood events of considerable size in the alluvial
plain. Alluvial fan deposition by braided rivers takes
place due to large sediment loads entering the foreland.
This situation is typical of the Po foreland, but a parallel
situation is found, for example, in the Himalayan range
and the Gangetic foreland. Furthermore, the anthropogenic
interventions of the last centuries, such as the
construction of dikes and draining of existing wetlands
in the lower lands, have severely reduced the river's
space, causing highly critical hydraulic conditions in
the lower part of the floodplain where the river changes
from braiding to meandering (Fig. 1).
0169-555X/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0169-555X(02)00264-7
* Tel.: +31-53-487-4311; fax: +31-53-487-4336.
E-mail address: spaliviero@itc.nl (M. Spaliviero).
www.elsevier.com/locate/geomorph
Geomorphology 52 (2003) 317­333
In this context, a number of questions arise. First,
how does the geomorphological setting influence the
overall hydraulic morphology of the river? Second,
what evidence is available to analyse and map
changes of the river since historical times? Third,
what has been the role of human activity on the river?
Finally, can likely near future developments of the
river be described and taken into account to work out
measures for flood control?
To understand the existing flood hazard, it is useful
to study the behaviour of the river in the past. For this
purpose, different approaches were adopted that produced
new findings concerning the dynamics of the
Tagliamento River. Historical descriptions of flooding
were positioned on a map base. A land-use map of
1833, which forms the first reliable cartographic data,
was compared with topographic maps of 1899, 1925
and other recent maps to relate land-use changes with
channel changes. Aerial photos from different periods
(1953­1955, 1966 and 1997) as well as recent satellite
images were interpreted to retrieve geomorphological
observations concerning the river's evolution,
which were then related to existing hydrological data.
2. Description of the present situation
The Tagliamento River originates from Mauria
Mountain at 1195 m and is 175 km long. It has a
catchment area of 2871 km2
(Basin Authority for the
North Adriatic Rivers, 1997), which shows a funnel
shape with a wide and mountainous upper part, a
Fig. 1. The Tagliamento catchment. Source: Basin Authority for the North Adriatic Rivers.
M. Spaliviero / Geomorphology 52 (2003) 317­333318
smaller and hilly central part and a long and narrow
band in the floodplain (Fig. 1). The catchment
includes 60 towns with a total population of approximately
180,000 inhabitants.
The principal tributaries of the upper catchment,
the Lumiei, the Degano, the But and the Fella,
converge and join the Tagliamento River forming a
palmate pattern. Their basins are characterised by
steep slopes and lie in one of the wettest regions of
Italy, where annual precipitation can reach 3000 mm
(Fig. 2). Rainfall is concentrated mainly in heavy and
erosive showers determining the torrential regime of
the river. Furthermore, the mountain basin is seismically
active and has a dense distribution of landslides,
resulting in much bed load and a braided nature of the
river downstream.
Although this paper focuses on the relation between
historic fluvial changes and flooding risk, it is valuable
to provide the main hydrological data. The rainfall
characteristics cause high peak flows in the central
Fig. 2. Average annual precipitation for the period 1961­1990. Source: Basin Authority for the North Adriatic Rivers.
M. Spaliviero / Geomorphology 52 (2003) 317­333 319
and lower part of the basin. The crest of the peak flow
propagates rapidly downstream. The difference in
arrival times of the peak between the upstream Venzone
gauging station and the downstream Latisana gauging
station (Fig. 1) is generally 12­16 h. Since the distance
between the two gauging stations is 75 km, the crest
propagates with a speed of 1.75 m/s. The annual
average discharge is approximately 100 m3
/s, while
the 50-year return-period (RP) peak discharge is 3500
m3
/s. For 100 years RP, 4300 m3
/s, and for 500 years
RP, over 6000 m3
/s (Maione and Machne, 1982). Fig. 3
reports the hydrograph of the exceptional 1966 flood
Fig. 3. Recorded and estimated hydrographs of major flood events of the Tagliamento River. Source: Basin Authority for the North Adriatic
Rivers.
Fig. 4. Maximum annual water levels recorded at Venzone gauge station for the period 1886­1996. Source: Basin Authority for the North
Adriatic Rivers.
M. Spaliviero / Geomorphology 52 (2003) 317­333320
recorded at Venzone and the estimated hydrographs at
Venzone and Pinzano of flood events of 100 years RP
for different rainfall durations, according to the Basin
Authority for the North Adriatic Rivers (1997). A linear
trend line fitted on a plot with the annual maximum
water levels recorded at Venzone gauging station
between 1886 and 1996 shows a slight long-term
increase (Fig. 4). It is difficult to identify how much
this increase is due to aggradation of the riverbed or to
the increases of discharge; furthermore, in the latter
Fig. 5. Alluvial fans of the rivers Tagliamento, Corno and Meduna.
M. Spaliviero / Geomorphology 52 (2003) 317­333 321
case, it has to be considered that flood discharges are
based on extrapolation of rating curves and not on
direct measurements.
The study area, which starts from Pinzano and
reaches the outlet in the Adriatic Sea, is represented
by the middle and lower Tagliamento (Fig. 1). From the
Pinzano gorge to the confluence with the Cosa tributary,
the river is confined between two high Quaternary
terraces and can reach a maximum width of 3 km near
Spilimbergo during periods of high water. There starts
the large S. Vito alluvial fan of the river (Figs. 5 and 7),
which has an area of 500 km2
.
The riverbed in the middle part is fully braided, as
would be expected under conditions of irregular discharge,
high bed load and associated relatively steep
gradient of the alluvial fan. Because the braided reach
is wide and shallow and the channel banks are
unstable, the rate of sediment transport per unit width
of channel may be relatively low (Leopold et al.,
1964). In this river stretch, the average width of the
active riverbed is about 1 km and is aggrading, as
discussed below. The cross-section in Fig. 6 shows the
riverbed at the highest position on the San Vito cone.
Dry season flows infiltrate fully.
The lower Tagliamento starts near Madrisio, where
the river changes to a single channel showing a semimeandering
pattern. This change is accompanied by a
less steep gradient and finer grained sediments in the
riverbed. The river width reduces drastically and
measures 175 m at Latisana. At this location, as an
average, the river bottom is 14 m deeper than the
immediate surroundings, while the dikes are on average
6 m higher. Latisana, located 25 km from the
coast, was an estuarine settlement in the past; at this
distance, the river is under a tidal backwater effect.
This factor caused overflow in front of the town
during the last major flood event, in 1966, when a
peak discharge greater than 4000 m3
/s was estimated
at Latisana gauging station. During the second half of
the 19th century, the Cavrato channel was built 10 km
downstream of Latisana to dissipate the high water. Its
capacity has been recently enlarged to 2500 m3
/s and
it starts to function once the Tagliamento reaches a
discharge of 2000 m3
/s (Foramitti, 1990).
3. Historical evolution of the fluvial dynamics
Historical documents were used to reconstruct and
map the evolution of the Tagliamento during the past
centuries. Of particular interest is to analyse the
location and area of the floodplain, the flood and
sediment regime and to define fluvial corridors.
The overall setting of the lower and middle Tagliamento
catchment is shown on Landsat TM false
colour composite of Fig. 7. The band combination
used for the false colour image is a result of experimentation
to obtain an image with most hydrological
and geomorphological information. The older Quaternary
deposits can be recognized by the large alluvial
fan of the ancestors of the Tagliamento River: the
Udine fan. It stretches from the city of Udine in the
east to Pordenone in the west, where the wide Meduna
and Cellina braided riverbeds are deflected along the
Fig. 6. Transversel profile Casarsa­Codroipo extracted from topographic map 1:10,000.
M. Spaliviero / Geomorphology 52 (2003) 317­333322
Fig. 7. Interpretation of the study area using a LANDSAT TM5 image.
M. Spaliviero / Geomorphology 52 (2003) 317­333 323
old fan to the southwest. As mentioned, the upstream
part of the middle Tagliamento River is incised in the
older fan deposits. The zonation of decreasing grain
sizes of the Udine fan is reflected in the land use
pattern, as indicated by four zones in the image. At the
start of the Holocene, the Tagliamento River occupied
the western part of the old fan and formed during the
period 18000­6000 BP the San Vito alluvial fan
(Fantin, 1990), shown also in Fig. 5. The western
position is probably due to the complex configuration
of the glacial deposits within the Osoppo amphitheatre
that obliged the river to pass through the Pinzano
gorge. It is also possible to delineate the fluvial deposits
from the coastal plain sediments on the image,
as shown. The partial on-lap of the Tagliamento
fluvial deposits on the coastal plain sediments south
of Latisana can be noted, as well as the redistribution
of the sands by long shore transport in the small and
young delta, which has grown beyond the coastal
plain. Before the 10th century, there was an uninterrupted
lagoon between Ravenna in the south and
Grado in the north (Fantin, 1990). This important
water body was divided in the lagoons of Comacchio,
Venice, Caorle, Marano and Grado between the 10th
and the 14th centuries by intense alluvial depositions
of the north Adriatic rivers, in particular the Tagliamento
River, due to deforestation in the Alpine region
(Fantin, 1990).
During the Roman Empire, the Tagliamento River
flowed in two main branches: Tiliaventum Maius,
which had approximately the present position, and
Tiliaventum Minus (TM in Fig. 7) that most probably
was more in the east, where the rivers Corno and Stella
are located (Averone, 1911). The rivers Le`mene and
Stella and the Lugugnana channel were all Tagliamento's
riverbeds (A, B, C and D in Fig. 7) during the past
centuries (Averone, 1911); at present, only groundwater
is discharged in these beds (Stefanini and Cucchi,
1977, 1978). In fact, the diverging branches mark
active fan formation downstream of the incised part
of the Tagliamento River. These ancient branches were
often reused during the main flood events (Scimone,
1927; Ciconi, 1855; Rinaldi, 1870; Castellarin, 1990)
and urban settlements like Portogruaro, Palazzolo or
Precenicco, which are located relatively far from the
Tagliamento, were flooded repeatedly (Figs. 8 and 9)
until the dikes for the middle Tagliamento were completed,
after 1850 (Fig. 10).
During the period 1400­1599 (Fig. 8), the middle
Tagliamento flowed more to the west compared to its
present position (Rinaldi, 1870), i.e., in alignment
with its ancient riverbeds A, B and C (Fig. 7).
Therefore, the towns of Cordovado and Portogruaro
were flooded several times.
From 1600 to 1799 (Fig. 9), an eastwards migration
took place and, consequently, the ancient riverbed
D was reoccupied several times during the inundations
(Castellarin, 1990). Even the Tiliaventum Minus
was used twice (Fig. 10).
In particular, from 1596 to 1692, the river bifurcated
at the transition between the medium and the
lower Tagliamento, forming an island where the
settlements of San Paolo, Bolzano and Mussons were
located (Castellarin, 1990). After an important flood
in 1692, the river occupied the eastern branch, its
present position. However, since that time, the abandoned
western branch was used several times during
the flood events (even during the last one in 1966),
causing important damage in particular to the city of
San Paolo (Fig. 10).
During the flood events of the first half of the 19th
century, when the middle Tagliamento was still free to
migrate eastwards or westwards at the top of its alluvial
fan, the old riverbeds of both sides were reused. With
the growth of the population after the industrial and
agrarian revolution, the pressure on the riparian lands
increased and agricultural practices changed. The
information reproduced in Fig. 11 was derived from
the land-use map of 1833 made by official surveyors of
the Austrian Empire, which shows the situation before
the major pressure developed. Areas prone to inundation
were grasslands, wetlands or were used for pastures.
In the middle Tagliamento, these uncultivated
areas clearly indicate that the old river courses (A, B, C
and D) and the bifurcation close to San Paolo were left
as they were for centuries; however, patches of cultivated
areas along the edges of the floodplain witness
the attempt to occupy what was fluvial space in favour
of agriculture. The town of Latisana, being a major
stakeholder and at the most vulnerable position, played
a role in maintaining the status quo flood protection in a
natural manner. The uncultivated lands of the lower
Tagliamento served as overflows during major flood
events through openings in the dikes. These areas were
part of a large lagoon (Fig. 7). This technique of
dissipation was applied to protect the important settleM.
Spaliviero / Geomorphology 52 (2003) 317­333324
Fig. 8. Fluvial dynamics for the period 1400­1599.
M. Spaliviero / Geomorphology 52 (2003) 317­333 325
Fig. 9. Fluvial dynamics for the period 1600­1799.
M. Spaliviero / Geomorphology 52 (2003) 317­333326
Fig. 10. Synthesis of the historical dynamics of the Tagliamento River.
M. Spaliviero / Geomorphology 52 (2003) 317­333 327
Fig. 11. Interpretation of the land cover map of 1833. Source: land-use map made by the official surveyors of the Austrian Empire, 1833.
M. Spaliviero / Geomorphology 52 (2003) 317­333328
ments and the cultivated fields located in the riparian
zones of the lower floodplain (Scimone, 1927; Castellarin,
1990).
After 1850, dikes were completed for both the
middle and the lower parts of the river (Fig. 10).
Meanwhile, the openings in the dikes were closed
(Croci, 1899), the wetlands adjacent to the lower
Tagliamento were drained for cultivation purposes,
as well as significant parts of the lagoons of Marano
and Caorle, and the grasslands and pastures were
progressively converted into cultivated fields. Thus,
during the second half of the 19th century, the
hydraulic regime of the Tagliamento River was drastically
modified by man. Instead of the former situation
where the floodwaters had space to spread out
over a wide floodplain and over several branches and
overflows, they were now contained by dikes in a
narrowed floodplain. Consequently the frequency and
the magnitude of overflows increased in the lower
Tagliamento and, in particular, in the stretch where the
river changes from braiding to meandering, upstream
of Latisana town (Fig. 10). Even if the reliability of
historical records depends on the potential for channel
changes that could alter stage-discharge relations at
the site (Cook, 1987), Table 1 seems to confirm that in
Latisana, significantly higher water levels were recorded
after 1850. It can be expected that the situation
will get worse, since the downstream part is the narrowest,
as it is still adjusted to the earlier conditions
where the flood peaks were attenuated by upstream
storage over a wide floodplain and branches, and the
coarse bed load was laid down in the upstream parts.
The channel patterns of the area upstream Latisana
were interpreted using aerial photos of 1953 and 1997,
and it was found that traces of meanders are present
in stretches where the river is now braiding. Therefore,
we can conclude that sediment wedge associated
with the aggradation of the middle section was curtailed
by the dikes and has moved downstream; this
process may have contributed to the trend shown in
Table 1.
There is no reason to expect that either the floodproducing
rainfall or the sediment yield of the upper
mountain catchment will reduce in the near future, so
the sediment wedge will continue to migrate downstream.
From a geomorphological point of view, the
river will tend to maintain a steep gradient as an
adjustment to the high sediment load and irregular
discharges (Schumm et al., 1987), which it can
achieve by extending the braided course ultimately
to the sea. Similar narrowing of braided rivers (e.g.
the Agri River in South Italy) by dikes has led to
substantial increases of the absolute level of the bed,
with the consequence that flood discharges cannot be
contained by the cross-sectional area between the
dikes (Meijerink, 1984). Breakthrough of the dikes
can then be expected and in the worst case, the river
finds a new course in the adjoining, topographically
lower areas.
The map in Fig. 10, which synthesizes the fluvial
dynamics resulting from the historical analysis,
confirms this tendency. It can be observed that the
river overflowed mainly at the same locations, in
particular, close to San Paolo where, as mentioned
above, the river bifurcated for almost one century,
and downstream where the transition from braided
streams to a single meandering channel engenders a
much narrower river section.
This paper does not provide a full understanding
of the fluvial dynamics, which would require further
investigations. Other natural factors could be considered,
such as tectonic movements determining
uplift or subsidence processes, changes in sea level
and changes in rainfall regime. For example, the
possible effects of the little Ice Age on the bed load
are unknown. Little information exists on the frequency
of landslide contribution to the sediment
load, a part from some reports by Ciconi (1855)
concerning the flood events of 1851, which caused
the bifurcation near San Paolo as did that of 1962.
Detailed topographic surveys, such as longitudinal
profile and cross-sections, would be needed to
document the process of river aggradation since
the 19th century.
Table 1
High water levels recorded since 1800 in Venzone gauging station
and in Latisana gauging station during the major flood events
Major flood events
after 1800
High water level
in Venzone (m)
High water level
in Latisana (m)
15/10/1823 8.1
02/11/1851 8.2
28/10/1882 3.9 8.76
21/10/1896 3.7 9.88
02/09/1965 4.37 10.5
04/11/1966 4.85 10.88
M. Spaliviero / Geomorphology 52 (2003) 317­333 329
4. Planned measures for flood mitigation
The only remediation so far for the problems
created by the interventions of the 19th century was
to increase the height of the dikes and to maintain the
Cavrato channel as the only diversion for floodwaters.
At present, a plan to construct three interlinked
retention basins between Pinzano and Spilimbergo is
under discussion for approval. The attenuation of the
flood peaks by channel, floodplain storage and transmission
losses in the middle Tagliamento would be
modified by the basins; they are designed for a capacity
of 30 million m3
that should reduce the peak discharge
from 4600­5000 to 4000 m3
/s at Latisana (Foramitti,
1990). It is planned to construct the basins in the
western part of the post-glacial valley, nowadays
intensely cultivated. Their construction would cause
severe reduction of the riverbed width with consequences
for the erosion­sedimentation regime of the river,
as well as rapid groundwater seepage because of the
coarse gravels below the retention basins, with dikes of
10-m height (Foramitti, 1990). Furthermore, in light of
this research, that project does not solve the risk of
aggradation in the downstream stretch between San
Paolo and Latisana in the medium to long term.
Concerning the area where it is planned to construct
the basins, a progressive expansion of Spilimbergo
City close to the western terrace of the postglacial
valley has occurred since the beginning of the
20th century, while the Tagliamento River was shifting
eastwards. New settlements were built in an area
that had been actively occupied by the riverbed a few
decades earlier. Even if at present the river is flowing
more than 2 km away from these settlements, close to
the eastern terrace, the valley was entirely occupied by
the riverbed less than two centuries ago and has been
regularly flooded during important flood events (the
last time was in 1998).
5. The option of fluvial corridors
The medium to long term prospects necessitate to
develop options other than solely retention basins.
First of all, floods larger than the 1966 event may be
expected, not only on statistical grounds, but also
because most climatic change models agree that rainfall
extremes will increase. Even with an attenuation
of the flood peak by the planned retention basins by
some 15­20%, the effects of the moving sediment
wedge towards Latisana will still cause a dangerous
situation.
One option was building a dam in the narrow gorge
of Pinzano in order to create a reservoir upstream of
50 million m3
, but it met strong opposition from the
local communities (Foramitti, 1990).
Another option left is to consider fluvial corridors
(Gardiner, 1990; Govi and Turitto, 1994) in the
middle to lower Tagliamento. The corridors should
reflect the historical fluvial geomorphological developments
and the areas where floodwaters found their
way. In that perspective, a fluvial corridor is described
here as a zone that belonged in historical times to the
active part of the fluvial ecosystem and that appears to
be still under the rivers' influence when analysing
flood dynamics. The map of Fig. 12 shows fluvial
corridors by taking into account the past dynamics of
river shifts and locations of frequent overflows. The
map provides the background information as a starting
point for further working out expansion of the present
Tagliamento corridor, which has too limited capacity.
Widening the river by shifting the dikes downstream
of the transition between middle and lower Tagliamento
would provide an obvious corridor, but the
presence of significant urban settlements at both river
edges, such as Latisana and S. Michele, causes problems.
In this sense, the map can be used for the
planning of urban expansion and infrastructure to
avoid developments that may form additional burdens
in the near future.
The middle Tagliamento has two categories of
corridors corresponding to different stages of the
river's evolution. The fluvial corridor A1 includes
more recent changes in the riverbed width, approximately
between 1700 and 1850. During flood events,
the dikes are subject to the lateral erosion of the river
trying to reoccupy its former corridor. Particularly
dangerous is the section narrowed by the dikes of
the two Delizia bridges. The fluvial corridor A2
contains the riverbed migrations approximately during
the period 1500­1700. Since the construction of the
dikes, this area has been relatively stable, except at
places of the former courses indicated by arrows in
Fig. 12. During the last major flood event of 1966,
near San Paolo, the right bank dike was overtopped
and damaged.
M. Spaliviero / Geomorphology 52 (2003) 317­333330
Fig. 12. Definition of the fluvial corridors.
M. Spaliviero / Geomorphology 52 (2003) 317­333 331
The abandoned meanders of the lower Tagliamento
were detected by aerial photo interpretation. Their
spatial pattern and distribution density allowed the
definition of the fluvial corridor B1. It is interesting to
notice that the old meanders are more concentrated
where the river overflowed frequently in the past, e.g.
in front of Latisana. Several abandoned meanders
were found upstream of the place where the river
starts to meander nowadays. Thus, as mentioned
above, the sediment wedge associated with the braiding­meandering
transition has moved downstream,
but the floodplain confined by the dikes is narrowing,
which creates hydraulic instability.
For the zone B2, the limit of the depressed areas
used in the past for the dissipation of the flood is taken
as corridor receiving the floodwaters from zone B1.
Because of the downstream shift of the braiding­
meandering transition and the confined narrow width
of the lower Tagliamento, this zone will remain a
vulnerable zone in the medium to long term. The map
shows also the limit of the inundation of 1966, when
the river overflowed at the same locations as the
historical events.
The number, trace and dimensions of the corridor(s)
should be based on hydraulic modelling,
assuming a further rise of the braided riverbed,
detailed topography and by the land use and land
ownership within potential traces.
Finally, concerning the future evolution of the
riverbeďs position, it seems that the lower Tagliamento
tends to shift eastwards since a greater frequency of
damage was suffered by the eastern riparian territories
during the last major inundations. On the other hand,
the middle Tagliamento pushes westwards to reoccupy
its former riverbeds A, B and C (see Fig. 7).
6. Conclusion
The fluvial geomorphological development of the
Tagliamento River and flooding history is described
and analysed using historical documents and interpretation
of satellite images and aerial photographs. The
geologic and climatic conditions of the upper catchment
determine a hydraulic regime with high sediment
load and irregular discharges, which led to the
formation of a complex geomorphology of the alluvial
plain. The riverbed of the middle course is aggrading
and the transition from braiding to meandering is
shifting downstream. These natural processes favour
the flood risk, which was increased by the confinement
of the river by dikes after 1850. The historical
analysis confirmed that the flood hazards tend to
concentrate in the same locations where historic
fluvial changes occurred, allowing the delineation of
the areas under major risk. Therefore, since the
planned retention basins will not solve the problem
in the medium-long term, fluvial corridors were
defined. The results of the approach adopted in this
study could be integrated with hydraulic studies that
are commonly carried out, in order to obtain more
information on the complexity of fluvial dynamics.
For example, modelling the propagation of flash
floods should be extended to include the use of
corridors, to develop a plan for flood containment.
The effects of further aggradation of the riverbed
should be included in the modelling. It will be wise
policy to start considering restrictions on land use and
infrastructural development in the zones of anticipated
future fluvial corridors.
Acknowledgements
I would like to especially thank Prof. Mario Govi
for the important scientific contribution to this
research, Prof. A.M.J. Meijerink for the precious
support in writing this article and improving its
scientific contents, Prof. E. Kosters for the pertinent
advice he gave me after reading it, Prof. A.M. Harvey
and Prof. W.A. Marcus who carefully reviewed the
paper, resulting in significant improvements, and Ir.
Lluis Barcons for the technical support in map
digitising and map editing.
References
Averone, A., 1911. Saggio sulľantica idrografia veneta. Arnoldo
Forni Editore, Bologna.
Basin Authority for the North Adriatic Rivers (in Italian: Autorita` di
Bacino Nazionale dei Fiumi Isonzo, Tagliamento, Livenza,
Piave, Brenta-Bacchiglione), 1997. Piano di bacino del fiume
Tagliamento-Piano Stralcio per la sicurezza idraulica del medio
e basso corso, Venice.
Castellarin, B., 1990. Le inondazioni del Tagliamento. Latisana,
Udine.
M. Spaliviero / Geomorphology 52 (2003) 317­333332
Ciconi, G., 1855. Dalle principali inondazioni friulane. Edizioni
Trombetti-Murero, Udine.
Cook, J.L., 1987. Quantifying peak discharges for historical floods.
Journal of Hydrology 37, 29­40.
Croci, A., 1899. Il fiume Tagliamento e la sistemazione dei suoi
tronchi inferiori. Genio civile di Udine, Udine.
Fantin, E., 1990. Morfologia del fiume Tagliamento. Cenni storici
sulľevoluzione del fiume. Latisana, Udine.
Foramitti, R., 1990. Il Tagliamento: studi e progetti-Un fiume che
deve unire. Proceedings from ``Le alluvioni del Tagliamento a
Latisana'', Latisana, Udine, pp. 91­107.
Gardiner, J.L., 1990. River Projects and Conservation, A Manual
for Holistic Appraisal. Wiley, New York.
Govi, M., Turitto, O., 1994. Problemi di riconoscimento delle fasce
di pertinenza fluviale. Proceedings from ``IV Convegno Internazionale
di Geoingegneria Difesa e Valorizzazione del suolo e
degli Acquiferi'', Turin, 10­11 March, pp. 261­272.
Leopold, L.B., Wolman, M.G., Miller, J.P., 1964. Fluvial Processes
in Geomorphology. Freeman, San Francisco.
Maione, U., Machne, G., 1982. Studio sulla formazione e sulla
propagazione delle piene del fiume Tagliamento-Relazione tecnica.
Regione Friuli-Venetia Giulia, Udine.
Meijerink, A.M.J., 1984. Field work observations in the Agri valley,
Basilicata, South Italy ITC, Enschede.
Rinaldi, G., 1870. Intorno alle condizioni attuali del Torrente Tagliamento
sui gravi pericoli ďinondazione in cui versano i territori
adiacenti al tronco tra Spilimbergo e S. Paolo di Morsano,
nonche` sui lavori occorenti alla difesa dei medesimi, colľindicazione
sulla competenza passiva delle spese, e sulle modalita` di
procedere onde assicurare i mezzi a cio` necessari. Genio Civile
di Udine, Udine.
Schumm, S.A., Mosley, M.P., Weaver, W.E., 1987. Experimental
Fluvial Geomorphology. Wiley, New York.
Scimone, S., 1927. Breve monografia sul fiume Tagliamento. Magistrato
alle Acque, Udine.
Stefanini, S., Cucchi, F., 1977. Gli acquiferi nel sottosuolo della
provincia di Udine (Friuli-Venezia Giulia). Quaderni delľIstituto
di Ricerca sulle Acque, P/368, Consiglio nazionale delle
Ricerche.
Stefanini, S., Cucchi, F., 1978. Gli acquiferi nel sottosuolo della
pianura veneta fra i fiumi Piave e Tagliamento. Quaderni delľIstituto
di Ricerca sulle Acque, P/514, Consiglio nazionale delle
Ricerche.
M. Spaliviero / Geomorphology 52 (2003) 317­333 333