Available via license: CC BY-NC 4.0
Content may be subject to copyright.
Bulletin of the Geological Society of Greece
Vol. 43, 2010
FLOOD HISTORY ANALYSIS AND ITS
CONTRIBUTION TO FLOOD HAZARD
ASSESSMENT. THE CASE OF MARATHONAS,
GREECE
Diakakis M. Department of Geology and
Geoenvironment, National
and Kapodistrian University
of Athens
10.12681/bgsg.11308
Copyright © 2010 M. Diakakis
To cite this article:
Diakakis (2010). FLOOD HISTORY ANALYSIS AND ITS CONTRIBUTION TO FLOOD HAZARD ASSESSMENT. THE
CASE OF MARATHONAS, GREECE. Bulletin of the Geological Society of Greece, 43, 1323-1334.
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:27 |
XLIII, No 3 – 1323
FLOOD HISTORY ANALYSIS AND ITS CONTRIBUTION TO FLOOD
HAZARD ASSESSMENT. THE CASE OF MARATHONAS, GREECE
Diakakis M.1
1 Department of Geology and Geoenvironment, National and Kapodistrian University of Athens 15784
Athens, Greece, diakakism@geol.uoa.gr
Abstract
Flood history analysis contributes decisively to a more realistic assessment of flood hazard. In this
work, systematic data collection on past flood events in Marathonas area (Attica, Greece) together
with the development of a database, allowed the thorough study of flooding phenomena and their
evolution over time. The study area consists of three dominant catchments with rich flooding history,
namely Rapentosa, Charadros and Kato Souli. Information gathered from governmental and insur-
ance organizations, emergency agencies, the press, field interviews and other documentary sources,
along with geomorphologic and geologic evidence, were stored, structured and analyzed in a GIS
platform and were used to reconstruct flood events with detail. Valuable results were produced con-
cerning the causes, the characteristics, the spatial distribution of damages and the extent of inun-
dation for each event. Moreover, the rate of recurrence of flooding phenomena was calculated across
the floodplain, so that the areas of higher risk were identified and delineated. The active part of the
floodplain was outlined and its migration overtime was studied. Furthermore, the methodology un-
derlined the imperfections of the existing risk mitigation strategy and the past emergency experiences
were appraised in a way that they highlight the priorities and will help improve management of fu-
ture situations of risk.
Key words: flood hazard, flood history, Marathonas, flood frequency, Marathon, Attica, Greece.
1. Introduction
In many parts of Greece flooding occurs in small, flash flood prone watersheds drained by ephemeral
water courses with little or no water at all for most of the year. In this context and given the scarcity
of instrumental hydrological records, classic hydraulic modeling may not be adequate in assessing
flood hazard. Thus, the use of alternate methods becomes necessary for a better understanding of the
flooding processes and the mitigation of the associated risk. Analysis of flooding history is a tech-
nique that can produce results within these limitations. Historical flood records have been used in
the past in the Mediterranean region (e.g. Garcia & Garcia 2003) and around the world (Benito et
al 2004, Hergert & Meurs 2007) to improve the knowledge in the field of hydrological extremes. His-
torical flood analysis focuses on identifying hazard areas based on careful reconstruction and ex-
amination of past flood events. The objective of this work is to outline a methodology that
contributes to an in depth appraisal of the flooding problem in a specific area, in this case Marathonas
in Greece, regardless the availability of instrumental records.
Δελτίο της Ελληνικής Γεωλογικής Εταιρίας, 2010 Bulletin of the Geological Society of Greece, 2010
Πρακτικά 12ου Διεθνούς Συνεδρίου Proceedings of the 12th International Congress
Πάτρα, Μάιος 2010 Patras, May, 2010
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:27 |
2. Background
2.1 Setting
The study area lies in the north-east part of Attica in Greece and consists of three relatively small
catchments, namely Rapentosa, Charadros and Myrtia (Kato Souli). In terms of geomorphology, it
comprises of a hilly area, with steep slopes and the famous for its battle plain of Marathonas, through
which the drainage networks flow into the sea. The area is inhabited by approximately 8000 people
with Marathonas and Kato Souli being the most important settlements.
The southernmost catchment (37,5 km2) is drained by Rapentosa torrent and borders with Nea Makri
from the south and with Charadros catchment from the north. The drainage network reaches the
north side of Penteli and Dionysos and runs off, forming a steep gorge, into Marathonas plain and
through Vranas settlement. At present the torrent is drained at the last part of its course by an artifi-
cial underground waterway. Bordering from the north with Rapentosa, Charadros is the largest of the
3 catchments, even though its drainage area was reduced to 60,2 km2after the construction of
Marathonas’ dam in 1929. It reaches the sea at the central part of the plain, passing between Kotroni
and Stavrokoraki hills and through the modern village of Marathonas. The third and smallest catch-
ment (15,2 km2) is drained by Myrtia torrent, a poorly developed drainage network which flows
into a marsh, lying on the north-east part of the plain. The water, leaving the hilly area drains into
artificial channels until it reaches the swamp and subsequently the sea through a manmade canal. All
three catchments are dry for most of the year.
XLIII, No 3 – 1324
Fig. 1: Map showing the three catchments, the drainage network and the main settlements and landmarks. At
the north east part of the map Myrtia borders with Charadros, which in turn borders with Rapentosa catchment
from the south.
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:27 |
In terms of geology, the study area consists of a series of Mesozoic formations, accompanied by
postalpine Cenozoic sediments. From base to top, one can distinguish lower-middle Triassic mar-
bles followed by Northeastern Attiki Schist formations with marbles intercalations of Jurassic age
and Northeastern Attiki Marbles of Cretaceous age. Upper Miocene breccioconglomerates and flu-
violacustrine formations can be found on top of the Mesozoic series. Fine grained fluvial deposits
of Pleistocene age are abundant Charadros banks. The plain is dominated by Holocene alluvial de-
posits accompanied by some talus cones.
2.2 Previous works
Historical flood records have been used in the past for the study of flood risk. Potter (1978) first sug-
gested that the past behaviour of a river is a very important ingredient in all hydrological investiga-
tions and can be exploited thoroughly with the use of information contained in historical records.
Bayliss and Reed (2001) proposed flood history study as an alternative practice to improve our
knowledge on hydrologic extremes. Glaser and Stangl (2003) suggest that historical records are an
important source for the evaluation of frequency of such events. Both Potter (1978) and Bayliss and
Reed (2001) recommended periodical journals, newspapers, chronicles, specialists reports, govern-
ment databases, pictures, the world wide web and other various sources as a basis for information
mining. Barriendos et al (2003) suggested local and central government, church and private collec-
tions and notarial archives as good sources of information.
Previous works exploited documentary data to provide further understanding on the flooding problem
of several places. Glaser and Stangl (2003) studied extreme inundation events in central Europe since
1300 using documents dating back to the middle ages and other anecdotal evidence. In their study they
connected flood occurrence with historic climate and atmospheric patterns. Macdonald et al (2003) used
documentary data, newspapers, epigraphic records and eyewitness observations along with instru-
mental records in order to examine the flood-generating mechanisms of river Ouse in UK. In their
study they reconstructed flooding history since 1600AD and suggested that its analysis can improve
understanding of flood risk considerably. Herget and Meurs (2007) studied past extreme events in the
city of Cologne in Germany with the help of written descriptions, paintings and other evidence and
recreated peak river discharges for historical floods. Past flood data were also utilized to project the evo-
lution of flooding in time and space. Conesa Garcia and Garcia Garcia (2003) used anecdotal evidence
to reconstruct flooding history of Cartagena, in Spain, identifying high and low risk locations.
Documentary hydrological data have been correlated in the past with flood event frequencies to im-
prove risk prevention and readiness. Bayliss and Reed (2001) studied the case of river Avon at Worces-
tershire and suggested that it is essential in flood risk studies to review flood frequency curves by
augmenting existing data with the use of historical records. Agasse (2003) collected data from the
press, official telegrams, unpublished reports and interviews with local people and reconstructed Nor-
mandy’s flooding record during 17th to 20th centuries. In this study Agasse examined the trends of
flood frequency and intensity. Williams and Archer (2002) collected data from unpublished reports,
meteorological magazines and websites to reconstruct floods in the English Midlands. In their study,
they demonstrated how flood history improves risk assessment as it is carried out through the use of
conventional hydrometric record. Benito et al (2004) suggested that systematic and non-systematic
data can be combined to improve flood frequency analysis with direct impact on flood risk assessment.
3. Data
The basic concept of examining flooding history is the collection of information about every single
event recorded, followed by data structuring and analysis with specific techniques. Therefore, the
starting point of this investigation was the identification of data sources.
XLIII, No 3 – 1325
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:27 |
3.1. Data types and sources
As far as local government organizations are concerned, the Prefectural Administration of East At-
tica provided detailed reports about flood damages on infrastructure, including damaged property
owners’ accounts, description of locations, of water height and extent. The municipality of
Marathonas provided documents and verbal descriptions of past events. Information was also gath-
ered from central government organizations, i.e. the General Secretariat of Civil Protection and the
Earthquake Rehabilitation Service, in the form of maps demarcating flood zones and districts that
were hit in the past. Greek Fire Department provided a database of emergency incidents concerning
flooding in the area, reporting locations, time and type of event. Data on damages on agriculture were
collected from the Hellenic National Agricultural Insurance Organization along with damage type
and time specifications. National and local newspapers and were researched in depth for articles de-
scribing flood damages and conditions. Their archive was easily accessible and of very good qual-
ity and coverage through the years. World Wide Web was proved an excellent guide to locate
information sources and damage compensation mandates. Pictures and videos obtained from na-
tional channels, newspapers, websites and local residents were exceptionally revealing of the con-
ditions of flooding. Details on damages and observation on water stage and extent were collected by
interviewing local people. Data were also collected from the Archaeological Society at Athens which
contained a considerable amount of information in the form of diaries and written descriptions due
to the continuous presence of researchers since the 19th century.
3.2. Data compilation
Since the ancient years the plain was notorious for its floods. Charadros became proverbial for its
fierce floodwaters in ancient local societies (Hammond 1968). Locally, the river god was worshiped
as the lord of the violent force of Charadros (Sekunda 2002), showing a form of primitive knowl-
edge concerning the local flooding phenomena. However, the first detailed description of a flood
event comes from Leake (1841). In his work, he describes a violent flood that washed away many
houses in the settlements of Marathonas and Bei in the autumn of 1805, changing the landscape
completely. The events occurred after Rapentosa and Charadros torrents inundated large areas of
the plain. Leake (1841) also mentions that Charadros and Rapentosa are noted for their ‘occasional
impetuosity’ and that the plain is usually subject to inundations from the two torrents (‘particularly
Charadros’). Davidson (1880) describes a flash flood in 1879 that carried away a lot of material
from the banks of Charadros from the inner parts of the plain to the sea. Hughes (1901) also notes
the strong effect that flooding from Rapentosa and Charadros has on the plain morphology and the
amount of material deposited there during extreme events. The first captured picture of a flood in
Marathonas comes from Charadros (in 1926) showing high velocity waters, damaging structures
and equipment that were used at the time for the preparations of the construction of Marathonas
dam (EYDAP 2003). Sotiriadis during his archaeological investigations (1925-1937) noted that al-
though the two main torrents are dry even in the winter, the plain is “occasionally inundated” (Sotiri-
adis 1933). In 1959 (July the 3rd) overflowing of Rapentosa torrent hit the plain damaging crops and
drowning animals (Empros 1959). Almost a decade later Hammond (1968) notices that Charadros
and Vranas overflow occasionally transporting large amounts of materials on the plain and to the sea.
In 21st of October 1979, Kato Souli and Agios Georgios flooded after a high intensity rainfall, leading
to approximately 1600m2of damaged crops. The event was accredited then to a debris-clogged canal in
the Kato Souli area. In 1980 (October 27th) Rapentosa inundated its banks, leading to one casualty, two
injured people, many damaged cars, houses and crops in Vranas village. Unfortunately, although these
two events were covered extensively by the press at the time and are very well known to the local com-
munity there are no official government documents survive today. However, numerous detailed reports
can be found in 1979 and 1980 newspapers articles database (Nea 1979, Apogevmatini 1980).
XLIII, No 3 – 1326
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:28 |
After 1987 the data and the descriptions become more detailed. The archive of the Prefectural Ad-
ministration of East Attica (2007) contained ten floods events accompanied by detailed descriptions
of damages and locations. The database of the Hellenic National Agricultural Insurance Organiza-
tion (2007) for the study area, contained eight flood events that damaged agricultural land or equip-
ment and detailed information on extent, location, cost and time of damages recorded. The events
XLIII, No 3 – 1327
Table 1. Flood events in the area of study based on the records of the Prefectural Administration of East
Attica (2007) and the Hellenic National Agricultural Insurance Organization (2007).
Recorded flood events
Event Date Locations that suffered damages
(shown in fig. 1)
Damages in buildings and in-
frastructure
Damages in
agriculture
12th Nov. 1987 Patitiria, Tsepi, Vranas, Tymvos,
Valaria and Paralia
No detailed record No record
26t Feb. 1988 Vranas, Patitiria, Tsepi, Valaria,
Paralia and Tymvos
No detailed record. Approxi-
mately 30 houses, 20 vehicles
No record
27th Jan. 1996 Paralia. Valaria and Patitira 2 businesses, fencing in some
houses, road network,
public utilities
No record
12th Nov. 1998 Patitira, Tsepi and Valaria 10 buildings and the
road network
No record
20th Nov. 1998Patitiria, Tsepi, Tymvos and Valaria 2 buildings and the road network No detailed record
27th Mar. 1999 Patitiria, Vreksiza, Tymvos and
Valaria
No record 11 businesses,
truck farms
14th Jan. 2001 Patitiria, Vranas, Tsepi, Paralia,
Bei, Kato Souli, Rizari, Tymvos
9 buildings (household
utensils and parts of the
structure), many vehicles
42 businesses
greenhouses, olive
groves, truck farms
3rd Nov. 2001 Vranas, Tsepi, Valaria, Bei, Paralia,
Marathonas and Tymvos
18 buildings, some vehicles
and the road network
33 businesses,
greenhouses, olive
groves, truck farms
14th Dec. 2002 Valaria, Tsepi and Patitiria No record 9 businesses,
mainly greenhouses
26th Jan. 2003 Vranas, Tsepi, Rizari, Bei, Pati-
tiria, Agios Georgios
6 buildings, road network
and the sewerage system
21 businesses,
mainly truck farms
16 Sep. 2005 Kato Souli, Agios Georgios,
Marathonas village, Patitiria, Tsepi,
Paralia, Rizari and Bei
Only minor damages to some
buildings and fencing
21 businesses,
mainly greenhouses
23 Nov. 2005 Patitiria, Tymvos, Tsepi, Valaria,
Marathonas village, Plasi, Paralia,
Kato Souli and Agios Georgios
10 buildings (houses and
businesses), road network,
public utilities
29 businesses,
greenhouses and
fields, truck farms
18 (3)
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:28 |
from these two sources are summarized in table 1.
These twelve flood events in total were confirmed by cross-referencing with the record of the Greek
Fire Department (2007), the archive of the General Secretariat of Civil Protection (GSCP 2006), the
record of Earthquake Rehabilitation Service (2006), articles from the Greek Parliament Newspaper
Archive and television broadcasts obtained from the archive of national television stations. In some
cases complementary information was collected from these sources.
4. Methods
4.1 Data analysis
Initially , upon collection exploitable and good quality information was selected from raw data and
stored in a database, based on categorizing different types of evidence like water stage, extent, speed,
severity and type of damage, time and duration of flooding. This step was followed by the compi-
lation of an overall list of past flood events in the area. In a second phase, damages on structures and
agriculture were plotted in a GIS environment along with evidence provided from pictures and
videos to form an accurate reconstruction of every flood event between 1979 and 2008. Each one
of these polygon shapes symbolizing reconstructed inundated areas, were converted to raster data
with cell value of 1 for inundated locations and 0 for non-inundated locations. These raster data
were subsequently mathematically added with the use of Weighted Sum tool of computer software
ArcMAP 9.2 (ESRI 2008). Thus, raster cells were added with same-location cells of each flood
event. In this way, the resulting raster symbolized the spatial extent of all the flood events and each
cell contained information on how many times it was inundated. This very attribute provided a clear
view of the flood recurrence rate for each location for the last 30 years, which was the time period
with adequate data. It also allowed interesting observations on the evolution of flooding. Finally de-
scriptions, reports and other evidence were examined to determine the main causes of the flooding
problem in the area.
4.2 Geologic and geomorphologic observations
Assessment of the influence of geologic and geomorphologic features in extreme hydrological events
was considered important. Therefore, factors of geology and geomorphology of the three basins
were calculated in an effort to estimate in what extent they are linked to flooding phenomena.
Namely mean slope and basin elongation ratio were computed due to their influence on the hydro-
graph form (Sith & Ward 1998), whereas basin area and geologic formations permeability due to their
importance in water volume drained. In this work, geologic formations were classified in three
groups based on their hydrogeological behaviour. The first group consisted of higher permeability
carbonate rocks, the second of impermeable formations, like schist, and the third from post alpine
porous sediments of intermediate permeability.Additionally, the channel morphometry of the three
torrents was studied in order to examine the degree to which the flooding phenomena are enhanced
by parameters of the drainage network like channel’s morphomology and dimensions. This was car-
ried out by field observations and by studying geologic, topographic and historical maps of the area.
5. Results
5.1 Flood history summary
Examination of the collected evidence showed a rich history of flooding in the plain of Marathonas.
From ancient years until modern days the area suffered numerous incidents. Apart from the one known
XLIII, No 3 – 1328
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:28 |
XLIII, No 3 – 1329
casualty and some injuries, damages included buildings (houses and businesses), agricultural land
and equipment, vehicles, the road network and important infrastructure. Flooding continued through
the years even though significant flood defenses were developed. Based on the compiled data, one can
identify 18 distinct flood events (shown in table 2) extending back to 1805. The record is considered
to be complete only after 1979, in contrast with period before this year when data coverage was not
adequate. Table 2 shows all the recorded flooding incidents separately for each torrent.
5.2 Characteristics and evolution of flooding
Careful study of the events’ descriptions showed a series of common characteristics like the abrupt
rise of flood waters and the relatively short duration of inundation. It also showed higher water ve-
locities around Rapentosa torrent (Vranas and Patitiria) than at Charadros (Bei and Paralia) and even
lower speeds at Vreksiza, Valaria, Agios Georgios and Kato Souli. In some cases, vegetation debris
and sediment content was abundant in floodwaters (Prefectural Administration of East Attica 2007),
a phenomenon that can be attributed to the forest fires of 1995 and 1998 in the area. Concerning the
evolution of affected areas, it is evident that before the construction of Marathonas Dam, flooding
in Charadros was more frequent. Thus, after 1929 flood areas around Charadros were considerably
reduced along with associated damages. It is also apparent that flooding problems in Myrtia torrent
have been intensified in the last six years in the record, most probably due to the recent development
of housing, road networks and other public works without appropriate sewerage.
Table 2. Table showing recorded flood events in the study area
Flood events in Marathonas area
Year Date Basin Year Date Basin
Charadros Rapentosa Myrtia Charadros Rapentosa Myrtia
1805 Autumn × × 1998 November
12th
×
1879 Autumn ×1998 November
20th
×
1926 Unknown ×1999 March 27th ×
1959 July 3rd ×2001 January 14th × × ×
1979 October
21st
×2001 November
4th
× ×
1980 October
27th
×2002 December
14th
× ×
1987 November
12th
× × 2003 January 29th × × ×
1988 February
26th
×2005 September
16th
× × ×
1996 January
26th
× × 2005 November
23rd
× × ×
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:28 |
Human interference with hydrologic processes is considered to have intensified flooding phenom-
ena through blocking drainage routes towards the sea and reducing the flow capacity of channels.
Such effects are evident at Patitiria and Tymvos, Paralia and Agios Georgios. Changes due to the
recent development of flood defenses have to be evaluated in the long term.
5.3. Damage type and distribution
Analysis of the spatial distribution of damages, illustrates that Rapentosa (mainly Tsepi and Patitiria)
are the most affected areas. In most cases damages included agricultural land and equipment, green-
houses, vehicles, buildings fencing and masonry, household utensils and loss of some domestic and
livestock animals.
5.4 Flood recurrence rate
Flooding has a higher repetition rate in Rapentosa torrent (13 events in 30 years) than in Charadros (7
events in 30 years) and Myrtia catchment (6 events in 30 years). Data analysis, as described in 4.1,
showed explicitly the areas with higher recurrence rate. Figure 3 illustrates these areas with a ranking
based on the number of events suffered in 30 years. Thus, high recurrence rate represents areas that
flooded 10 to 14 times during this period, whereas medium recurrence rate locations that flooded 5 to
9 times and low recurrence rate areas 1 to 4 times. According to this process high recurrence rate lo-
cations can be identified at Patitiria and Tymvos and medium rate areas at Valaria and Agios Georgios.
5.5 Correlation with geologic and geomorphologic data
Examination of calculated basin properties showed that higher recurrence flood rate in Rapentosa
in comparison with the other two torrents is partly connected to its increased percentage of imper-
XLIII, No 3 – 1330
Fig. 2: Chart showing the fluctuation of affected areas in square metres through the years, for each flood event.
In horizontal axis, events are expressed in year of occurrence.
Table 3. Recorded damages for compensation purposes (Prefectural Administration of East Attica
2007, Hellenic National Agricultural Insurance Organization 2007).
Officially recorded damages Rapentosa Charadros Myrtia
Buildings damaged (1996-2009) 38 910
Damaged agricultural land (1999-2009) 802.000 m278.000 m2287.000 m2
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:28 |
XLIII, No 3 – 1331
meable formations and its higher mean slope value. The first attribute augments the total volume
of runoff and the second increases peak flow rates, reduces the watershed’s time of concentration and
raises the stream power of storm runoff as it enters the plain. On the other hand, basin area and elon-
gation ratio do not show good correlation with historical analysis results, denoting a less critical
role for these parameters.
Field and map observations on the torrents’ morphology suggest channel morphometry plays a very
important role. As far as Rapentosa is concerned, the torrent’s morphological characteristics degen-
erate and it eventually disappears morphologically at about 2km before reaching the sea. A careful
study of older published maps and sketches shows the same regime for the last 180 years. Namely,
the same setting as today appears in 1829 Leake’s sketch-map (in Hammond 1968), in 1880 Kau-
per’s map (Kauper 1889), in Hughes’ sketch of 1901 (Hughes 1901), in Caspari’s (1911) map, in Hel-
lenic Military Geographical Service’s (HMGS) sketch-map of 1928 (in Hammond 1968), in
Fig. 3: Map showing the areas that flooded during last 30 years. High recurrence rate represents areas that
flooded 10 to 14 times in 30 years. Medium recurrence rate stand for locations that flooded 5 to 9 times and
low recurrence rate areas that flooded 1 to 4 times in the same period.
Table 4. Table showing calculated properties for the three drainage basins.
Calculated basin properties
Properties Rapentosa Charadros Myrtia
Mean Slope (%) 33.3 21.3 22
Area (Km2)37.5 60.2 11.7
Elongation Ratio 0.46 0.56 0.70
High permeability formations (% of total area) 45.4% 50.3% 58.4%
Medium permeability formations (% of total area) 13.2% 19.8% 29.3%
Impermeable formations (% of total area) 41.4% 29.9% 12.3%
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:28 |
Vanderpool’s 1966 sketch-map (Vanderpool 1966), and in HMGS topographical maps of 1970’s.
This shows a natural diffusion of the channel in its last section. The point where the torrents char-
acteristics disappear in all the maps, coincides with the higher flood recurrence rate areas (i.e. Pati-
tiria). This regime together with the high percentage of impermeable formations and its high mean
slope increases Rapentosa’s natural potential for flooding in comparison with the other two water-
sheds. In relation with channel dimensions, field observations show that human activities have
greatly affected the torrents’ flow capacity. In case of Rapentosa in Vranas and Patitiria and in case
of Charadros in Bei and along the coast.
6. Conclusions
Past flooding in Marathonas is a well known phenomenon in an anecdotal sense, but it has not been
studied systematically. This analysis proves that Rapentosa is the most hazardous between the three
torrents studied. It also shows that Kato Souli (Myrtia torrent) suffered more damages in the last
decade compared to the past. The locations where more damages are concentrated are Patitiria,
Tsepi,Tymvos and Agios Georgios. These areas show higher possibility to be also affected by flood-
ing in the future. In the case of Rapentosa, human development in the course of the river, mainly in
Patitira and Tsepi, augments exposure and vulnerability and obstructs local hydrologic processes in
a degree to which the possibility of overflowing is increased. Hazard is further augmented due to nat-
ural basin characteristics and the drainage network’s natural diffusion. Extensive development along
the coast and around the marsh interferes also with drainage processes obstructing surface runoff in
Charadros and Myrtia.
In this context and taking into account that high flow rates are part of a river’s natural process, a sig-
nificant step towards the problem’s solution would be a long term policy of reducing intervention
to hydrological processes. This involves of course gradual removal of human activities from the ac-
tive floodplain. Meanwhile, developing flood defenses is valuable but in many cases, including the
event of November 2005 in Marathonas is proved to be inadequate.
It should be also kept in mind that the methodology has a number of limitations. For example flood-
ing phenomena are subject to natural and human induced change and therefore the method’s results
must not be perceived as stationary and have to be reviewed every time new knowledge is available.
Moreover, quiet periods in the flood record, in some cases, may not correspond to periods lacking
flood events.
This method can be used either as a standalone process or to verify results of other methodologies.
Generally, examining high temporal extent historical records are a special benefit when studying
rare natural disasters, especially when hydrological data are scarce. The use of such methods for re-
construction of flood history has a great potential in Greece due to the early and continuous habita-
tion. In this work, past flooding is perceived as the result of an evolving natural experiment. In this
context, flooding history analysis offers and in depth appreciation of the flooding problem and its
evolution over time.
7. References
Agasse, E., 2003, Flooding from the 17th to 20th centuries in Normandy (western France) methodology
and use of historical data. In Thorndycraft, V.R., Benito G., Barriendos M. and Llasat M.C., (eds.):
Palaeofloods, Historical Floods & Climatic Variability: Applications in Flood Risk Assessment, Pro-
ceedings of the PHEFRA Workshop, Barcelona, Spain, 16-19th October, 2002, CSIC, Madrid, pp. 99-
106.
XLIII, No 3 – 1332
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:28 |
Apogevmantini 1980. Ghild drowned in Marathonas (in Greek). Newspaper “Apogevmatini”, October
29 1980, p.1, Greek Parliament Newspaper Archive, Athens, Greece.
Barriendos, M., Llasat, C., Barrera, A., Rigo, T. 2003. The study of flood events from documentary
sources: Methodological guidelines for historical source identification and flood characterization in
the Iberian Peninsula. In Thorndycraft, V.R., Benito G., Barriendos M. and Llasat M.C., (eds.): Palae-
ofloods, Historical Floods & Climatic Variability: Applications in Flood Risk Assessment, Proceed-
ings of the PHEFRA Workshop, Barcelona, Spain, 16-19th October, 2002, CSIC, Madrid, pp. 87-92.
Bayliss, A.C., Reed, D.W. 2001. The use of historical data in flood frequency estimation, Report to MAFF,
Centre of Ecology and Hydrology, Wallingford.
Benito, G., Lang, M., Barriendos, M., Llasat M.C., Frances, F., Ouarda, T., Thorndycraft, V.R., Enzel Y.,
Bardossy A., Coeur, D., Bobee B. 2004. Use of Systematic, Palaeoflood and Historical Data for the
Improvement of Flood Risk Estimation. Review of Scientific Methods. Natural Hazards 31: 623–643.
Caspari, M. O. B., 1911. Stray notes on the Persian Wars, The Journal of Hellenic Studies, Vol. 31, pp.
100-109.
Conesa-Garcia C., Garcia-Garcia E. 2003. Historical evolution of the flood areas in Cartagena (South-East
Spain). In Thorndycraft, V.R., Benito G., Barriendos M. and Llasat M.C., (eds.): Palaeofloods, His-
torical Floods & Climatic Variability: Applications in Flood Risk Assessment, Proceedings of the
PHEFRA Workshop, Barcelona, Spain, 16-19th October, 2002, CSIC, Madrid, pp. 149-154.
Davidson, T. 1880. The Dionysion at Marathon, The American Journal of Philology, Vol. 1, pp. 58-59.
Earthquake Rehabilitation Service 2006. Statistics: Emergency incidents and correlation with YAS archive.
Earthquake Rehabiliation Service (YAS), Ministry of Ministry of the Environment, Physical Planning
and Public Works, Athens, Greece.
Empros 1959. ‘Damages caused by the storm in Attica region’(in Greek). Newspaper Empros, 4th July
1959, p.10, Digital Newspapers Collection, Greek National Library, Athens, Greece.
ESRI 2008. ArcGIS 9.2 Geographical Information System Software. Economic and Social Research In-
stitute. Geographic Information System, Redlands, California.
EYDAP 2003. Water cycle in Attica, Athens Water Supply and Sewerage Company (EYDAP SA). [On-
line], http://www.eydap.gr/media/stagonoulis/KuklosNerou/gr/024.htm [Last Accessed: 3/9/08].
Glaser, R., Stangl, H., 2003. Floods in Central Europe since 1300. In Thorndycraft, V.R., Benito G., Bar-
riendos M. and Llasat M.C., (eds.): Palaeofloods, Historical Floods & Climatic Variability: Appli-
cations in Flood Risk Assessment, Proceedings of the PHEFRA Workshop, Barcelona, Spain, 16-19th
October, 2002, CSIC, Madrid, pp. 93-98.
Greek Fire Department 2007. Emergency incidents in the Municipality of Marathonas, Greece. Depart-
ment of Informatics and Communication, Greek Fire Department, Athens, Greece.
GSCP 2006. Flooding in East Attica Prefecture. General Secretariat of Civil Protection, Ministry of the
Interior, Public Administration & Decentralization, Athens, Greece.
Hammond, N. G. L., 1968. The Campaign and the Battle of Marathon, The Journal of Hellenic Studies,
88, pp. 13-57.
Hellenic National Agricultural Insurance Organization 2007. Flood damages in municipality of
Marathonas, Greece (1999-2008), Hellenic National Agricultural Insurance Organization (ELGA),
Ministry of Rural Development and Food.
Hergert J., Meurs H., 2007. Reconstructing peak discharges of historic flood levels in the city of Cologne,
Germany, 4th International Palaeoflood Workshop, Crete, Greece, 24th-30th June 2007
Hughes, M. 1901 Marathon. The Classical Review, 15 (2),pp. 131-136.
Kauper J. A. 1889. Karten Von Attika (Marathon). Berlin, Deitrich Reimer
XLIII, No 3 – 1333
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:28 |
Leake W. M. (1841). The topography of Athens and the Demi, Volume 2, The Demi of Attica. [online]
http://books.google.co.uk/books?id=sB0GAAAAQAAJ&printsec=titlepage [Last accessed: 3/9/08]
Macdonald, N., Black, A.R., Werritty, A., 2003. Historical and pooled flood frequency analysis for the
River Ouse, York, UK. In Thorndycraft, V.R., Benito G., Barriendos M. and Llasat M.C., (eds.):
Palaeofloods, Historical Floods & Climatic Variability: Applications in Flood Risk Assessment, Pro-
ceedings of the PHEFRA Workshop, Barcelona, Spain, 16-19th October, 2002, CSIC, Madrid, pp.
217-222.
Nea 1979. ‘Woman in peril in Kato Souli’ (in Greek). Newspaper “Ta Nea” 22 October 1979, p.1, Greek
Parliament Newspaper Archive, Athens, Greece
Potter, H. R. 1978. The use of historic records for the augmentation of hydrological data. Wallingford,
Institute of Hydrology, (IH Report No.46).
Prefectural Administration of East Attica 2007. Technical Services Archive, Natural Disaster Damages
(1987-2005). Technical Services Department, Prefectural Administration of East Attica.
Robinson, D. M., Blegen, E. P., 1936. Archaeological News and Discussions. American Journal of Ar-
chaeology, Vol. 40, No. 2, (Apr. - Jun., 1936), pp. 247-270.
Sekunda, N., 2002. Marathon 490 BC: The first Persian War. Oxford: Osprey Publishing
Sith, K., Ward, R., 1998. Floods : physical processes and human impacts. Chichester : John Willey & Sons
Sotiriadis, G., 1933. Excavations at Marathon. Proceedings of the Archaeological Society (Praktika) pp.
31-46.
Vanderpool E., 1966. The Deme of Marathon and the Herakleion. American Journal of Archaeology, Vol.
70, No. 4, (Oct., 1966), pp. 319-323
Williams, A., Archer, D., 2002. The use of historical flood information in the English Midlands to improve
risk assessment. Hydrological Sciences Journal, Vol. 47(1), pp. 67-76.
XLIII, No 3 – 1334
Powered by TCPDF (www.tcpdf.org)
http://epublishing.ekt.gr | e-Publisher: EKT | Downloaded at 02/02/2017 21:04:28 |
