GeolCarp_Vol62_No1_55

www.geologicacarpathica.sk

GEOLOGICA CARPATHICA

, FEBRUARY 2011, 62, 1, 55—63 doi: 10.2478/v10096-011-0005-z

Gravity instabilities in the Dohrn Canyon (Bay of Naples,

Southern Tyrrhenian Sea): potential wave and run-up

(tsunami) reconstruction from a fossil submarine landslide

VINCENZO DI FIORE, GEMMA AIELLO

and BRUNO D’ARGENIO

Institute for Marine Coastal Environment (IAMC), National Research Council of Italy (CNR), Calata Porta di Massa, Porto di Napoli,

80133 Napoli, Italy; *gemma.aiello@iamc.cnr.it

(Manuscript received June 30, 2010; accepted in revised form December 6, 2010)

Abstract: We discuss a mathematical model for wave and run-up generated submarine landslides in the canyons of the Bay
of Naples (Magnaghi-Dohrn canyon system). The morpho-bathymetry and submarine gravity instabilities of such incisions
have been investigated through the interpretation of a high resolution DEM. The canyons are located in a sector of the bay
where there is a variable interaction of volcanic activity (Phlegrean Fields and Ischia and Procida Islands) with sedimentary
processes due to the Sarno-Sebeto rivers. At present the Naples canyon-system is inactive, as is shown by the Holocene
sedimentary drapes deposited during the present sea-level highstand, but gravity instabilities occurred in the recent past at the
canyons’ heads. In particular the Dohrn Canyon is characterized by a double regressive head, while the Magnaghi Canyon
shows a trilobate head, formed by the junction of three main tributary channels and coincident with the retreat of the shelf
break around the 140 m isobath. The results of a simulation of failures in the above source areas show that the amplitude of
wave run-up, expressed in terms of the sea floor depth percentage, may range up to 2.5 % of the water depth at the sea bottom.

Key words: Bay of Naples, tsunamigenic potential, run-up landslide, numerical modelling.

Introduction

The aim of this paper is to study potential wave and run-up
events caused by submarine landslides located in the can-
yons of the Bay of Naples.

Tsunami waves, often caused by gravitative failures, may

be generated by earthquakes and less frequently by volcanic
eruptions. In the Bay of Naples all the above trigger factors
are present. As a consequence, the continental slope off the
bay represents an appropriate natural laboratory to study
geological events potentially leading to submarine slides
with their tsunamigenic potential.

The geological setting of the bay has been studied in detail

in the framework of research programmes for submarine geo-
logical  cartography  (Aiello  et  al.  2001;  Bruno  et  al.  2003;
D’Argenio et al. 2004). In this gulf the continental slope and
the outer shelf are deeply incised by two submarine canyons of
kilometric  extent,  namely  the  Dohrn  and  Magnaghi  Canyons,
representing  the  drainage  system  of  this  active  volcanic  area
during  the  Late  Quaternary.  Detailed  mapping  of  the  outer
shelf and slope morphology contributed to the understanding
of erosional and depositional processes related to continental
slope settings and allowed their geological interpretation to be
proposed (Aiello et al. 2001; D’Argenio et al. 2004).

Several studies about tsunamis have recently been carried

out in the Bay of Naples. Landslide-generated tsunamis in the
offshore of Ischia Island have been featured by Zaniboni et al.
(2007)  in  the  framework  of  a  project  on  the  volcanic  hazard
and  risk  assessment  of  the  island  of  Ischia,  by  the  National
Institute  for  Geophysics  and  Volcanology  and  the  Italian
Department for Civil Protection. The above authors focussed

on the study of tsunamis generated by landslides from Ischia’s
slopes. The catastrophic collapse that formed the large scar in
the  southern  flank  of  Ischia  (the  Ischia  Debris  Avalanche  of
Chiocci & de Alteriis 2006) may be considered as the upper
limit  for  tsunamigenic  failures  in  the  slopes  of  Ischia,  even
though the duplication of such an event does not appear prob-
able  at  the  moment.  In  this  study  we  selected  an  area  of  the
Dohrn Canyon showing evidence of paleo-landslides and we
applied a model to evaluate the wave run-up generated by an
estimated flow, using a recent theoretical model (Lynett & Liu
2002; Di Fiore et al. 2008).

General setting of the area

The eastern margin of the Tyrrhenian Sea is characterized by

a  number  of  basins  formed  during  the  latest  Neogene-Quater-
nary across the structural boundary between the Apennine fold
and  thrust  belt  and  the  Tyrrhenian  back-arc  extensional  area
(Fig. 1).  These  basins,  including  the  Bay  of  Naples  and  the
Bay of Salerno (Fig. 1) evolved as a consequence of the large-
scale orogen-parallel extension and related transtensional tec-
tonics  that  accompanied  the  anti-clockwise  rotation  of  the
Apennine  belt  and  lithospheric  stretching  in  the  central
Tyrrhenian  Basin  (Sacchi  et  al.  1994;  Ferranti  et  al.  1996;
D’Argenio et al. 2004).

As a consequence, the Campania segment of the peri-

Tyrrhenian structural belt displays the characteristics of a pas-
sive  continental  margin,  where  Quaternary  orogen-parallel
extension  caused  the  formation  of  half-graben  systems  (Gulf
of  Gaeta,  Bay  of  Naples,  Bay  of  Salerno  and  intervening

DI FIORE, AIELLO and D’ARGENIO

GEOLOGICA CARPATHICA

GEOLOGICA CARPATHICA, 2011, 62, 1, 55—63

structural highs (e.g. Sorrento Peninsula, Mt Massico),
trending perpendicularly to the main axis of the Apennine
thrust belt (Mariani & Prato 1988; Sacchi et al. 1994; Milia
& Torrente 1999; Aiello et al. 2000).

Off the Campania coasts the peri-Tyrrhenian basins often

form the seaward extension of the coastal plains, whose for-
mation  was  controlled  by  extensional  tectonics  during  the
Plio-Quaternary  (Mariani  &  Prato  1988;  Brancaccio  et  al.
1991).  Their  tectono-sedimentary  evolution  is  connected
with the Neogene evolution of the Apenninic chain (Royden
et al. 1987; Patacca & Scandone 1989). In particular, the de-
formational  history  of  the  peri-Tyrrhenian  basins  is  charac-
terized by alternating compressional and extensional tectonic
phases during Plio-Quaternary times (Bartole 1984; Argnani
& Trincardi 1990; Agate et al. 1993; Sacchi et al. 1994).

The Neogene evolution of the south-eastern peri-Tyrrhe-

nian basins was controlled by large-scale extensional tecton-
ics, responsible for the thinning of the western sectors of the
southern Apenninic chain and dating back to the Late Mio-
cene.  Thinning  due  to  extension  was  not  homogeneously
distributed  along  the  overthrust  belt,  but  rather  localized  in
discrete  hyper-extension  domains  (Ferranti  et  al.  1996),
where the thrust pile was locally reduced to about one half of
its original thickness.

Modes of extension along the peri-Tyrrhenian basins are

often characterized by listric normal faults and associated
antithetic faults, which generated SW-NE trending half-gra-
ben systems along the Tyrrhenian Basin-southern Apennines
system. Most of them extend landward to the E-NE. This
causes a typical coastal landscape made of alternating trans-
versal mountain ridges and intervening coastal plains.

In the Campania Region Quaternary basin fillings overlie

submerged “internal” (western) tectonic structures of the Apen-
ninic chain, resulting from the seawards extension of the tecton-
ic units cropping out in the coastal belt of the southern
Apennines (D’Argenio et al. 1973: fig. 1). These units form the

acoustic basement of the coastal basins and are composed either
of terrigenous-shaly basinal sequences or of thick platform and
basinal Mesozoic-Cenozoic carbonates (Fig. 1). Extensional
tectonics accompanying the uplift of the southern Apennines
begin in the Early Pliocene and continue up to the Middle—Late
Pleistocene, playing a major role in controlling the present-day
physiography of the Campania Region. Indeed, Quaternary ma-
rine and continental sediments of the Campania coastal plains
reach a thickness of up to 3000 m in the Volturno Plain and of
1500 m in the Sele Plain (Ortolani & Torre 1981).

NW-SE, NE-SW and E trending post-orogenic structures

(mainly  extensional  faults)  have  been  previously  recognized
under the sea near the Campania Region (Bartole et al. 1983).
While  the  Apenninic  (NW-SE) trending structures character-
ize the continental slope areas between the Pontine Islands and
the  Cilento  Promontory,  the  Anti-Apenninic  (NE-SW)  ones
often  occur  under  the  sea  near  the  Bay  of  Salerno  and  the
structural high of the Sorrento Peninsula.

In turn, the Quaternary extension along the Campania seg-

ment  of  the  Southern  Apennine—Eastern  Tyrrhenian  hinge
zone caused the onset of intense volcanism. It was responsi-
ble  for  the  creation  of  both  single  large  volcanoes  (Roc-
camonfina  and  Somma-Vesuvius – Principe  et  al.  1987:
fig. 1) and volcanic complexes (Ischia, Procida and Phlegrean
Fields – Rosi & Sbrana 1987: fig. 1). In the Phlegrean area
a thermo-metamorphic basement about 1500 m deep has been
inferred (Rosi & Sbrana 1987), whereas in the Volturno Plain
the  “Villa-Literno”  1  and  “Parete”  1  wells  drilled  into  thick
basaltic and andesitic lavas (Ortolani & Aprile 1978).

Morpho-bathymetry and sea bottom instability of

the Naples bay canyons

An extensive, high resolution bathymetric survey of the

continental shelf/slope system of Campania, Southern Italy

Fig. 1.  Tectonic  sketch  map  of  western
Campania  Apennines  (modified  after
D’Argenio  et  al.  2004).  1  –  Shallow-
water  carbonate  and  deep  basinal  units
(Mesozoic); 2 – Piggy-back siliciclastic
units (Tertiary); 3 – Pyroclastic depos-
its  and  lavas  (Quaternary);  4  –  Conti-
nental and marine deposits (Quaternary);
5  –  Normal  fault;  6  –  Detachment
faults (barbs indicate downthrown side).

GRAVITY INSTABILITIES IN CANYON, POTENTIAL WAVE AND TSUNAMI RECONSTRUCTION (BAY OF NAPLES)

GEOLOGICA CARPATHICA

GEOLOGICA CARPATHICA, 2011, 62, 1, 55—63

has  recently  been  carried  out.  The  relative  bathymetric  data
were  acquired  during  the  years  1997  and  2002,  using  Multi-
beam systems with an average vertical resolution of  0.25 %
water  depth  and  a  position  accuracy  of

10 m. The survey

data were successively merged with a Digital Terrain Model
(DTM) created from topographic maps of the Bay of Naples
onshore coastal area and islands, to produce a Digital Eleva-
tion Model (DEM) based on a homogeneous grid with cell-
spacing of 20 m (D’Argenio et al. 2004: fig. 2).

The shaded relief map has provided new, detailed infor-

mation on the morphology of the Campania coasts. This seg-
ment  of  the  Italian  continental  margin  displays  evidence  of
the  latest  Neogene-Quaternary  interplay  between  tectonics
and  volcanism,  as  well  as  of  the  depositional  processes
largely  developing  as  a  consequence  of  frequent  and  large
volcano-sedimentary  supply.  The  major  morphological  fea-
tures  revealed  by  the  3D  digital  maps  are:  i)  the  system  of
marine canyons (Dohrn and Magnaghi) that cut the continen-
tal slope at a depth between 250 m and 1100 m; ii) the conti-
nental slope system of the Ischia volcanic structure (Chiocci &
de Alteriis 2006; Aiello et al. 2009a); iii) the onshore and off-
shore volcanoes of the Campi Flegrei; iv) the rugged seafloor
area of the outer shelf of the city of Naples (Banco della Mon-
tagna – Sacchi et al. 2000; D’Argenio et al. 2004); v) the de-
bris flow/avalanche deposits on the inner continental shelf off
Mt  Vesuvius  (Milia  et  al.  1998,  2008),  laterally  grading  into
the “Torre del Greco” volcanic structure (Aiello et al. 2010).

The DEM of the Bay of Naples, representing the base of

the geological and morpho-bathymetric interpretation, has a
20 20 m cell and has been based on the integration of dif-
ferent grids. This cell has resulted adequate in detecting the
most prominent topographic and bathymetric features of the
coastal zone larger than a few tens of meters.

Despite the great number of geological and volcanological

studies  on  the  Gulf  of  Naples,  among  which  Latmiral  et  al.
(1971),  Finetti  &  Morelli  (1973),  Cinque  et  al.  (1997),
Pescatore et al. (1984), Fusi et al. (1991), Milia (1996), Milia
et al. (1998, 2008), Aiello et al. (2001, 2004, 2005, 2009a,b,
2010),  Bruno  et  al.  (2003),  the  morpho-bathymetry  and  the
submarine  gravity  instabilities  of  the  Dohrn  and  Magnaghi
Canyons  have  not  been  investigated  in  detail,  nor  have  the
areas of incipient submarine sliding in the surrounding of the
canyons even been accurately defined.

The Dohrn Canyon formation was triggered off by the tec-

tonic  uplift  of  both  the  outer  shelf  and  the  fluvial  valley
mouths,  during  periods  of  eustatic  fall  of  sea  level  (Milia
2000). During the Late Quaternary the Bay of Naples continen-
tal slope was characterized by slumping and canyon formation
and  therefore  it  may  be  considered  to  represent  an  erosional
slope-system (Ross et al. 1994; Galloway 1998). The geologi-
cal conditions for the formation of the Dohrn Canyon appear
to be incompatible with models based on oversteepening, high
sediment  supply,  sea-level  rise  and  retrogressive  slumping.
The Dohrn Canyon developed along a central slope showing
low gradients and the landward rotation of the platform block
resulted in a substantial decrease of the slope. Moreover, the
canyon  wall  cuts  the  main  body  of  the  slumps,  whereas  the
most  prominent  scars  show  no  correlation  with  the  canyon
walls.  So  the  Dohrn  Canyon  formation  was  controlled  by  a

sea-level eustatic fall, inducing a seaward migration of the riv-
er systems of the Late Pleistocene and the formation of the
valleys, coupled with a fault block rotation, responsible for the
outer shelf uplift (Milia 2000).

Based on the interpretation of Multibeam bathymetry we

have outlined a schematic geomorphological map of the
Dohrn Canyon and of the surrounding continental slope,
which shows, in particular, the slide scars and the submarine
gravity instability areas (Fig. 2).

The main morpho-structures of the canyon system consist

of volcanic structural highs, relic morphologies of the Middle-
Late Pleistocene continental shelf, turbiditic slope fans and
Mesozoic carbonate structural highs. Some morphological lin-
eaments, such as the canyon’s walls, the shelf break, the slope
of the “Ammontatura” paleo-canyon, the slide scars and the
canyon’s axis have also been represented (Fig. 2).

The Dohrn Canyon’s width ranges from a few hundred

meters to more than 1 km, its depth from 250 m at the shelf
edge to some 1300 m at the merging with the bathyal plain;
the dip of its walls attains some 35° in the steepest sectors.
This  canyon  starts  with  two  major  curved  branches.  The
western branch merges into the shelf through a 1.5 km wide
and 20—40 m deep channel (“Ammontatura” channel), char-
acterized  by  a  flat  bottom  and  asymmetrical  levees,  located
along  the  —200 m  isobath  and  by  a  sinuous  shape  in  plan
view.  The  Dohrn  eastern  branch  shows  a  meandering  trend
and starts from the shelf break of the Sorrento Peninsula, lo-
cated  along  the  —120 m  isobath.  The  Dohrn  western  branch
is broader than the eastern one and more deeply incised; the
two branches form a typical Y-structure.

The abrupt termination of the “Ammontatura” shallow

channel against the Nisida volcanic bank, whose growth was
older or contemporaneous with the eruption of the volcanic
deposits of the NYT (Neapolitan Yellow Tuff – Scarpati et
al. 1993) suggests that the Dohrn canyon system is older than
the volcanic deposits of the NYT, which forms the main basis
of the city of Naples. This implies that most part of the canyon
system activity was probably older than 11—12 ka B.P., age of
the NYT deposits in the city of Naples (Di Girolamo et al.
1984; Rosi & Sbrana 1987).

The Magnaghi Canyon shows a triple incised head and is

pervasively affected by small-scale slope instabilities. It runs
parallel  to  the  southern  flanks  of  the  Procida  and  Ischia  Is-
lands and then is buried below the Ischia Debris Avalanche
(Chiocci & de Alteriis 2006; Aiello et al. 2009a). Three main
tributary channels join basinwards into the main axis.  Ero-
sion and transport of volcanoclastics in the western sector of
the  gulf,  near  the  Ischia  and  Procida  Islands,  developed
along the Magnaghi Canyon axis and it appears unrelated to
the present  fluvial drainage system on land.

As already noted, the Bay of Naples canyons start from the

shelf break near the Phlegrean Fields volcanic district, located
along  the  —140 m  isobath.  Their  trends  are  controlled  by  the
main  morpho-structures  of  the  bay:  the  “Banco  di  Fuori”
structural  high,  bounding  southwards  the  whole  canyon  sys-
tem and the Capri Island structural high, bounding eastwards
the Dohrn Canyon, near its confluence with the bathyal plain.
Both  the  Magnaghi  Canyon  and  the  Dohrn  western  branch
show morphological evidence of retreat of the canyon’s head.

DI FIORE, AIELLO and D’ARGENIO

GEOLOGICA CARPATHICA

GEOLOGICA CARPATHICA, 2011, 62, 1, 55—63

Based on the interpretation of the Multibeam bathymetry the

main features that we interpret as slide scars are located (Fig. 2):

a) on the head of the Dohrn western branch, showing a

double retrogressive head, controlled by extensive subma-
rine erosion;

b) on the western slope of the Dohrn western branch (Fig. 3),

at its boundary with the eastern flank of the Banco di Fuori
morpho-structural high (here a set of coalescent, large slide-
scars, not related to the canyon’s thalweg, may be observed);

c) on the continental slope, north of the Capri structural

high, where large scars are suggested by the trending of the
isobaths next to the Dohrn Canyon thalweg.

Here the canyon system is characterized by several terrac-

es and three rectilinear gullies converge into the canyon from
the northern slope of the Capri Island. An acoustic basement,
probably composed of Mesozoic-Cenozoic carbonate rocks
almost reach the sea bottom, under a thin drape of Holocene
sediments.

Fig. 2. Morpho-bathymetry of the Bay of Naples canyons and gravity instability map. Note that chief submarine instability areas are located
at the Dohrn Canyon’s head, on the slope surrounding the western Dohrn branch, on the slope southwards of the Magnaghi Canyon and on
the north-eastern slope of the Banco di Fuori structural high. Figures on the side refer to kilometric coordinates.

GRAVITY INSTABILITIES IN CANYON, POTENTIAL WAVE AND TSUNAMI RECONSTRUCTION (BAY OF NAPLES)

GEOLOGICA CARPATHICA

GEOLOGICA CARPATHICA, 2011, 62, 1, 55—63

The geomorphological interpretation and the analysis of the

canyon morphometric parameters suggest that the submarine
instabilities of the canyon system are located (Fig. 2):

a) around the Dohrn western branch, from the canyon’s

head to the middle of the branch, north of the “Banco di Fuo-
ri” morpho-structural high;

b) on the continental slope, southwards of the Magnaghi

Canyon, where two large areas of instability, not connected
with the canyon’s thalweg, may be observed;

c) on the north-western slope of the “Banco di Fuori” mor-

pho-structural high, where the concave trending of the iso-
baths suggests the occurrence of an incipient and/or fossil
slide scar.

No slumped masses are preserved in the main thalweg,

suggesting  a  probable  activity  of  low-density  turbidity  cur-
rents  during  the  Late  Quaternary  evolution  of  the  canyon.
Large  isolated  remnants,  showing  average  dimensions  of
300 400 m across are present on the canyon floor which, in
its  upper  part,  is  scoured  by  a  minor  meandering  channel.
These  rounded  morphologies  are  interpreted  as  relic  struc-
tures,  probably  due  to  a  selective  erosion  acting  along  the
canyon’s valley. Moreover, terrace rims, located respectively
at  —340 m  and  —300 m  of  water  depth,  suggest  at  least  two
phases in the activity and retreat of the canyon head.

A relative abundance of the V-shaped erosional profiles

has  been  observed  in  both  the  branches  of  the  Dohrn  Can-
yon.  The  flat-bottomed  valley  depositional  morphologies,
suggesting  recent  phases  of  canyon  filling,  occur  mainly  in
the  wider,  southernmost  part  of  the  Dohrn  western  branch

and at the confluence of the Dohrn northern and southern
branches.

The drainage system of the canyons is composed of a

dense  network  of  tributary  channels,  controlling  the  over-
flow of sediments in the surrounding areas of the continental
slope. Two main tributary channels originate from the shelf
break  off  the  Phlegrean  submarine  volcanic  banks,  located
along the 140 m isobath and run along the continental slope
between  the  Dohrn  and  the  Magnaghi  Canyons,  giving  rise
to channel-lobe systems (Fig. 2). A complex system of tribu-
tary channels, located at water depths ranging from —200 m
and  —500 meters,  fed  the  Dohrn  southern  branch,  starting
from  the  shelf  break  off  the  Sorrento  Peninsula  and  it  ap-
pears partly fault-controlled. This interpretation is suggested
by  the  rectilinear  shape  in  plan  view  of  the  southernmost
four  channels.  A  system  of  lobes,  genetically  linked  with
some  of  these  channels,  has  been  recognized  on  a  morpho-
logical  terrace  located  at  water  depths  of  340 m.  Moreover
fossil tributary channels hung over the main branches testify
stages  of  rapid  re-incision,  switching  off  the  feeding  from
lateral sources and forming suspended valleys.

In conclusion, several submarine slides and scars are evi-

dent  on  the  canyon’s  walls,  especially  along  the  western
flank  of  the  Dohrn  northern  branch  and  on  the  continental
slope  as  well.  The  north-western  branch  of  the  Dohrn  Can-
yon is affected by instability, with an incipient slump caus-
ing  a  broad  depression,  200—300 m  across  and  away  from
the  canyon’s  edge,  and  a  semi-circular  scar  on  the  canyon
walls, showing lateral coalescence and defining a large area

Fig. 3. Detailed bathymetric map of the canyons showing the location of modelled submarine landslide (arrow).

DI FIORE, AIELLO and D’ARGENIO

GEOLOGICA CARPATHICA

GEOLOGICA CARPATHICA, 2011, 62, 1, 55—63

where (see also nomenclature below):
µ is the frequency-dispersion parameter (h

), i.e. when

the ratio of water depth to wavelength is small. In general the
guideline for dispersive properties is that the depth-integrat-
ed equations are valid for wavelengths greater than two wa-
ter depths;

)

( v

is the horizontal velocity vector;

is the horizontal gradient vector;

p is the depth-dependent pressure;

is the nonlinearity parameter (a/h

), i.e. when a<<h

, the

nonlinearity is not important (the initial wave amplitude is rela-
tivity small compared to the wavelength and the water depth).

Fig. 5 shows the parameters used in the algorithm imple-

mentation.

In many cases, the seafloor displacement is small with respect

to the local water depth. Since the free-surface displacement is
directly proportional to the seafloor displacement, namely

)=D( h).

The boundary conditions applied are:
On the free surface z=

x y, t), the usual kinematic and

dynamic boundary conditions, considering the depth integra-
tion interval between z=[—h,

], we set:

on (4.4)

p=0,
where p is water pressure.
Along the seafloor where z=—h, the kinematic boundary

condition requires

on z=—h (4.5)

Integrating from z=—h, to z=

, the depth-integrated

continuity equation, we obtained:

of instability, located westward of the north-western slope of
the “Banco di Fuori” morpho-structural high (Fig. 2).

A bathymetric profile has been constructed in correspon-

dence to the scars involving the Dohrn western branch in or-
der to give quantitative constraints to the numerical
modelling. A detachment area, about 415 m across occurs at
water depths ranging between —250 m and —370 m. Also de-
bris accumulation develops in water depths ranging from
—380 m and —450 m, while the junction with the foot of the
thalweg occurs, at water depths of about —430 m (Fig. 4).

Methods of analysis and modelling

In this paragraph we discuss the problems of the possible

tsunami generation and characteristics, as we may infer from
the above analytical description of the Bay of Naples canyon
system.  We  apply  here  a  mathematical  model  already  pro-
posed  by  Lynett  et  al.  (2002)  in  which  the  generation  and
propagation  of  tsunamis  is  reconstructed  from  ancient  sub-
marine landslides. In such a model we assume that there is a
weak  frequency  dispersion,  which  means  that  the  ratio  of
water  depth  to  wavelength  is  small  or  <<1.  In  general,  for
dispersive properties the depth-integrated equations are valid
only for wavelengths greater than two times the water depth,
whereas the depth-averaged model is valid for lengths great-
er  than  five  times  the  water  depth  (e.g.  Nwogu  1993).  In
Lynett  et  al.  (2002)  the  full  nonlinear  effect  is  included,
namely the ratio of wave amplitude and water depth is order
one, and therefore, this model is more general as compared to
other  models  (Liu  &  Earickson  1983)  where  the  Boussinesq
approximation was used.

The motion can be described by Euler’s equations:

Fig. 4. (a) Model scheme referred to the parameters
utilized  in  this  paper  and  evolution  of  submarine
slope  surface  shape  during  landslide  movement.
The detachment area relative to the slide and the de-
bris accumulation area, covering a total distance of
620 m,  have  also  been  represented.  (b)  Detailed
bathymetric  map  of  the  area  and  slice  location
(A—B) of the ancient marine landslide.

(4.3)

;

(4.2)

;

(4.1)

;

⎟⎟

⎠

⎞

⎜⎜

⎝

⎛

;

⎥⎦

⎤

⎢⎣

⎡

∫

(4.1)

(4.2)

(4.3)

(4.6)

GRAVITY INSTABILITIES IN CANYON, POTENTIAL WAVE AND TSUNAMI RECONSTRUCTION (BAY OF NAPLES)

GEOLOGICA CARPATHICA

GEOLOGICA CARPATHICA, 2011, 62, 1, 55—63

where H =

h .

After integral resolution of the two-dimensional governing

equation, we applied the finite difference algorithm to solve
the general model equation (Lynett et al. 2002). In their pa-
per Lynett et al. (2002) developed an algorithm for the gen-
eral two-horizontal-dimension problem analogously to Wei
& Kirby (1995) and to Wei et al. (1995). The difference be-
tween the two models is that in Wei et al. (1995) terms due
to a time-dependent water depth and some nonlinear-disper-
sive terms have been added, while in this study we use an
impulsive bottom movement in a constant water depth.

The previous governing equation was approximated onto a

two-dimensional horizontal plane. Some assumptions are
listed below to resolve the approximate two-dimensional
governing equation (Lynett et al. 2002):

1. only one-horizontal-dimension is examined;
2. nonlinearity is assumed so is small;
3. frequency dispersion is weak so µ

is small;

4. all spatial derivatives are differenced to fourth-order in

numerical analysis.

Fig. 5. Schematic representation of the geometrical parameters used in
wave-run up computation. h

– characteristic water depth and vertical

length scale; l

– characteristic length of the submarine slide region

and horizontal length scale; l

/(gh

)

1/2

– time scale; a – char-

acteristic wave amplitude and scale of the wave motion.

Fig. 6. Water wave amplitude vs. time-series located in x/h = 0 (horizontal coordinates of the midpoint of the seafloor movement). The rapid
decadence of the amplitude after 8 s is evident. Note that the amplitude wave run-up expressed in terms of depth of sea floor percentage
ranges from 0 to 2.5 %. In absolute terms the wave height amplitude corresponds to 5—6 m.

Finally, the numerical analysis was applied resolving two-

dimensional equation so obtaining the relative run-up.

The bottom movement, consisting of a length l

= 620 m

(Fig. 5), pushed the water masses vertically upwards. The
change in depth for this experiment is about 0.1 ( h/h

), and

therefore nonlinear effects should play a small role near the
source region, and 20 times the water depth from the edge of
the source region. The landslide is located near the Dohrn
Canyon head (Fig. 3) and we assume that the landslide may
move rapidly.

To analyse this model we have introduced the characteris-

tic water depth h

, as the vertical length-scale, the character-

istic length of the submarine slide region l

as the horizontal

length-scale, l

/ (gh

)

1/2

as the time-scale, and the character-

istic wave amplitude a as the scale of wave motion.

Fig. 6 shows the results of the analysis in terms of ampli-

tude vs. time-series at x /h = 0. It is possible to notice that the
amplitude wave run-up expressed in terms of depth of seaf-
loor percentage, ranges from 0 to 2.5 %. In absolute terms
the wave height amplitude corresponds to 5—6 m.

Discussion and conclusion

Submarine landslides play a fundamental role in originat-

ing  tsunamis,  especially  near  the  coast.  Our  experiment  in
the  Bay  of  Naples  could  be  used  as  example  in  the  areas
where  there  is  a  high  probability  for  these  events  to  occur
(Tinti et al. 2003). The idea is to produce an offshore hazard
map  delimiting  both  the  potential  tsunamis  areas  (located
near the instability zone) and the influence zone of the water
wave.

The morphometric analysis of the Dohrn Canyon slope

provides insight into tsunami hazard, including the locations
of mass movements, the size of mass failures, their relative
importance  for  the  structure  of  a  given  margin,  and  the  po-
tential  for  landslide-generated  tsunami  hazard.  We  do  not
have enough evidence to consider all these detachment areas
located  on  the  Dohrn  Canyon  edges  as  generated  by  land-
slides with impulsive character, or if there has been an evo-
lution over time controlled by small detachments, not signif-
icant  if  compared  to  the  potential  needed  to  generate
anomalous waves. The nature of the materials that character-

DI FIORE, AIELLO and D’ARGENIO

GEOLOGICA CARPATHICA

GEOLOGICA CARPATHICA, 2011, 62, 1, 55—63

ize  the  area  is  both  volcanic  (Campanian  Ignimbrite)  and
sedimentary. Therefore we deal with a non-coherent material
that,  together  with  the  presence  of  the  active  volcanoes
(Somma-Vesuvius  and  Phlegrean  Fields),  need  to  be  analy-
sed with regard to its ability to trigger landslides.

The tsunami predictions according to Booth et al. (1993)

have  to  be  related  not  so  much  to  the  slope  gradient  alone,
but  instead  to  sedimentation,  erosion,  local  geology,  fluid
overpressure  and  regional  seismicity  as  complex  causative
triggers. This whole complex is critical in determining land-
slide  location  and  ultimately  size  of  the  derived  tsunami,
considering that  the continental shelf of the Gulf of Naples
is  dipping  in  average  6°  and  has  a  width  of  20—22 km,  ex-
tending  for  560 km

, while the continental slope dips up to

45°. Our study suggests that the tsunami hazard is higher on
the canyon margin: in fact, Fig. 2 shows several slide scars
and large submarine instability areas near to canyon borders.

In conclusion we observe that the morphological evolution

of the analysed area depends on complex systems where vol-
canism, sedimentation and tectonics affect the coast and the
marine bottoms, that show the potential to develop slope in-
stability, even though high urbanization has impacted signif-
icantly on the sedimentation, reducing the accumulation rate
of materials potentially prone to slide.

Due to the lack of more detailed morpho-bathymetric stud-

ies as well as of innovative methods for hazard evaluation in
coastal areas, the balance of these opposite trends is difficult
to assess. However we stress the importance of the analysis
of submarine gravity instability, mainly for its implications
in terms of geo-hazard related to landslides as the cause of
tsunamis. Much remains to be done to reach an accurate un-
derstanding of these problems.

Nomenclature

wave amplitude
displacement function
gravity
characteristic water depth or baseline water depth, function

of space

water depth profile, function of space and time
total water depth (

characteristic horizontal length-scale of the submarine slide
depth-dependent pressure
time
depth-dependent components of velocity in (x, y, z)
horizontal velocity vector, (u, v)
Cartesian coordinates
characteristic, or maximum, charge in water depth due to

seafloor motion

nonlinearity parameter (a/ho)
frequency-dispersion parameter (ho/lo)
free-surface displacement

margin during the Plio-Pleistocene. In: Geological Develop-
ment of the Sicilian-Tunisian platform. Unesco Reports in Ma-
rine Sciences, Urbino, Italy 58, 25—30.

Aiello G., Marsella E. & Sacchi M. 2000: Quaternary structural

evolution of Terracina and Gaeta basins (Eastern Tyrrhenian
margin, Italy). Rend. Fis. Acc. Lincei 9, 11, 41—58.

Aiello G., Budillon F., Cristofalo G., D’Argenio B., de Alteriis G.,

De Lauro M., Ferraro L., Marsella E., Pelosi N., Sacchi M. &
Tonielli R. 2001: Marine geology and morpho-bathymetry in
the Bay of Naples. In: Faranda F.M., Guglielmo L. & Spezie
G. (Eds.): Structures and processes of the Mediterranean Eco-
systems. Springer-Verlag, 1—8.

Aiello G., Angelino A., Marsella E., Ruggieri S. & Siniscalchi A.

2004: Carta magnetica di alta risoluzione del Golfo di Napoli.
Boll. Soc. Geol. Ital. 123, 333—342.

Aiello G., Angelino A., D’Argenio B., Marsella E., Pelosi N., Rug-

gieri  S.  &  Siniscalchi  A.  2005:  Buried  volcanic  structures  in
the  Gulf  of  Naples  (Southern  Tyrrhenian  sea,  Italy)  resulting
from  high  resolution  magnetic  survey  and  seismic  profiling.
Ann. Geophysics 48, 1—15.

Aiello G., Marsella E. & Passaro S. 2009a: Submarine instability pro-

cesses on the continental slopes off the Campania region (South-
ern Tyrrhenian sea, Italy): the case history of Ischia island
(Naples Bay). Boll. Geof. Teor. Appl. 50, 2, 193—207.

Aiello G., Marsella E., Di Fiore V. & D’Isanto C. 2009b: Stratigraphic

and structural styles of half-graben offshore basins in Southern
Italy: multichannel seismic and Multibeam morpho-bathymetric
evidences on the Salerno Valley (Southern Campania continen-
tal margin, Italy). Quaderni di Geofisica 77, 1—34.

Aiello G., Marsella E. & Ruggieri S. 2010: Three-dimensional mag-

neto-seismic reconstruction of the “Torre del Greco” sub-
merged volcanic structure (Naples Bay, Southern Tyrrhenian
sea, Italy): Implications for Vesuvius’s marine geophysics and
volcanology. Near Surface Geophysics 8, 1, 17—31.

Argnani A. & Trincardi F. 1990: Paola slope basins: evidence of re-

gional contraction on the Eastern Tyrrhenian margin. Mem.
Soc. Geol. Ital. 44, 93—105.

Bartole R. 1984: Tectonic structure of the Latian-Campanian shelf.

Boll. Oceanol. Teor. Appl. 2, 3, 197—230.

Bartole R., Savelli C., Tramontana M. & Wezel F.C. 1983: Struc-

tural and sedimentary features in the Tyrrhenian margin off
Campania, southern Italy. Mar. Geol. 55, 163—180.

Booth J.S., O’Leary D.W. & Popenoe P.D. 1993: U.S. Atlantic conti-

nental slope landslides; their distribution, general attributes, and
implications. In: Schwab W., Lee H. & Twichell D. (Eds.): Se-
lected studies in the U.S. Exclusive Economic Zone. U.S.G.S.
Bull. 2002, 14—22.

Brancaccio L., Cinque A., Romano P., Rosskopf C., Russo F., San-

tangelo N. & Santo A. 1991: Geomorphology and neotectonic
evolution of a sector of the Tyrrhenian flank of the Southern
Apennines. Z. Geomorphol. 82, 47—58.

Bruno P.P.G., Rapolla A. & Di Fiore V. 2003: Structural setting of

the Bay of Naples (Italy). Tectonophysics 372, 193—213.

Chiocci F.L. & de Alteriis G. 2006: The Ischia debris avalanche:

first clear submarine evidence in the Mediterranean of a volca-
nic island prehistoric collapse. Terra Nova 18, 202—209.

Cinque A., Aucelli P.P.C., Brancaccio L., Mele R., Milia A., Ro-

bustelli G., Romano P., Russo F., Russo M., Santangelo N. &
Sgambati D. 1997: Volcanism, tectonics and recent geomor-
phological change in the Bay of Napoli. Suppl. Geogr. Fis. Di-
nam. Quat. 3, 123—141.

D’Argenio B., Pescatore T.S. & Scandone P. 1973: Schema geologi-

co dell’Appennino meridionale (Campania e Lucania). In: Mod-
erne vedute sulla geologia dell’Appennino. Convegno (Roma,
16—18 Febbraio 1972). Accademia Nazionale dei Lincei, Prob-
lemi Attuali di Scienza e Cultura, Quaderni, 183, 49—72.

wave amplitude
displacement function
gravity
characteristic water depth or baseline water depth,
function of space
water depth profile, function of space and time
total water depth (

characteristic horizontal length-scale of the subma-
rine slide
depth-dependent pressure
time
depth-dependent components of velocity in (x, y, z)
horizontal velocity vector, (u, v)
Cartesian coordinates
characteristic, or maximum, charge in water depth
due to seafloor motion
nonlinearity parameter (a/ho)
frequency-dispersion parameter (ho/lo)
free surface displacement

u, v, w

x, y, z

References

Agate M., Catalano R., Infuso S., Lucido M., Mirabile L. & Sulli A.

1993: Structural evolution of the Northern Sicily continental

GRAVITY INSTABILITIES IN CANYON, POTENTIAL WAVE AND TSUNAMI RECONSTRUCTION (BAY OF NAPLES)

GEOLOGICA CARPATHICA

GEOLOGICA CARPATHICA, 2011, 62, 1, 55—63

D’Argenio B., Aiello G., de Alteriis G., Milia A., Sacchi M., Tonielli

R., Budillon F., Chiocci F.L., Conforti A., De Lauro M., D’Isanto
C., Esposito E., Ferraro L., Insinga D., Iorio M., Marsella E., Mo-
lisso F., Morra V., Passaro S., Pelosi N., Porfido S., Raspini A.,
Ruggieri S., Terranova C., Vilardo G. & Violante C. 2004: Digital
elevation model of the Naples Bay and adjacent areas, Eastern
Tyrrhenian sea. Atlante di Cartografia Geologica scala 1 : 50,000
(progetto CARG). Servizio Geologico d’Italia (APAT). 32° Inter-
national Congress “Firenze 2004”, Editore De Agostini, Italy.

Di Girolamo P., Ghiara M.R., Lirer L., Munno R., Rolandi G. &

Stanzione D. 1984: Vulcanologia e petrologia dei Campi
Flegrei. Boll. Soc. Geol. Ital. 103, 349—413.

Di Fiore V., Aiello G. & Angelino A. 2008: Wave and run-up (tsu-

nami) reconstruction from ancient submarine-landslide in the
Gulf of Naples, Italy. Extended Abstract GNGTS 2008, Trieste,
232—236.

Ferranti L., Oldow J.S. & Sacchi M. 1996: Pre-Quaternary orogen-

parallel extension in the Southern Apennine belt, Italy. Tec-
tonophysics 260, 325—347.

Finetti I. & Morelli C. 1973: Esplorazione sismica per riflessione nei

Golfi di Napoli e Pozzuoli. Boll. Geof. Teor. Appl. 16, 175—222.

Fusi N., Mirabile L., Camerlenghi A. & Ranieri G. 1991: Marine

geophysical survey of the Gulf of Naples (Italy): relationships
between submarine volcanic activity and sedimentation. Mem.
Soc. Geol. Ital. 47, 95—114.

Galloway W.E. 1998: Siliciclastic slope and base-of-slope deposi-

tional systems: component facies: stratigraphic architecture and
classification. AAPG Bull. 82, 569—595.

Latmiral G., Segre A.G., Bernabini M. & Mirabile L. 1971: Pros-

pezioni sismiche per riflessione nei Golfi di Napoli e Pozzuoli.
Boll. Soc. Geol. Ital. 90, 163—172.

Liu P.J. & Earickson J. 1983: A numerical model for tsunami gen-

eration and propagation. In: Lida J. & Iwasaki T. (Eds.): Tsu-
namis: their science and engineering Harpenden. Terra
Science, 227—240.

Lynett P.J. & Liu P. 2002: A numerical study of submarine-landslide-

generated waves and run-up. Proc. R. Soc. London A(2002) 458,
2885—2910.

Lynett P.J., Tso-Rea W. & Liu P. 2002: Modeling wave run-up with

depth-integrated equations. Coastal Engineering 46, 89—107.

Mariani M. & Prato R. 1988: I bacini neogenici costieri del margine

tirrenico: approccio sismico-stratigrafico. Mem. Soc. Geol. Ital.
41, 519—531.

Milia A. 1996: Evoluzione tettono-stratigrafica di un bacino peri-

tirrenico: il Golfo di Napoli. PhD Thesis, University of Naples
“Federico II”.

Milia A. 2000: The Dohrn canyon: a response to the eustatic fall

and tectonic uplift of the outer shelf along the eastern Tyrrhe-
nian sea margin, Italy. Geomarine Lett. 20, 101—108.

Milia A., Mirabile L., Torrente M.M. & Dvorak J.J. 1998: Volcan-

ism offshore of Vesuvius volcano (Italy): Implications for haz-
ard evaluation. Bull. Volcanol. 59, 404—413.

Milia A. & Torrente M.M. 1999: Tectonics and stratigraphic archi-

tecture of a peri-Tyrrhenian half-graben (Bay of Naples, Italy).
Tectonophysics 315, 297—314.

Milia A., Molisso F., Raspini A., Sacchi M. & Torrente M.M. 2008:

Syneruptive features and sedimentary processes associated
with pyroclastic currents entering the sea: the AD 79 eruption
of Vesuvius, Bay of Naples, Italy. J. Geol. Soc. 165, 839—848.

Nwogu O. 1993: Alternative form of Boussinesq equations for near-

shore wave propagation. J. Wtrwy Port Coastal Ocean Engi-
neering 119, 618—638.

Ortolani F. & Torre M. 1981: Guida all’escursione nell’area interessata

dal terremoto del 23-11-80. Rend. Soc. Geol. Ital. 4, 173—214.

Ortolani F. & Aprile F. 1978: Nuovi dati sulla struttura profonda

della Piana Campana a sud-est del fiume Volturno. Boll. Soc.
Geol. Ital. 97, 591—608.

Patacca E. & Scandone P. 1989: Post-Tortonian mountain building

in the Apennines. The role of passive sinking of a relic lithos-
pheric slab. In: Boriani P. et al. (Eds.): The lithosphere in Italy.
Advances in Earth Science Research, Atti dei Convegni Lincei
80, 157—176.

Pescatore T., Diplomatico G., Senatore M.R., Tramutoli M. & Mira-

bile L. 1984: Contributi allo studio del Golfo di Pozzuoli: aspetti
stratigrafici e strutturali. Mem. Soc. Geol. Ital. 27, 133—149.

Principe C., Rosi M., Santacroce R. & Sbrana A. 1987: Explanatory

notes in the geological map. In: Santacroce R. (Ed.): Somma-
Vesuvius. Quaderni De La Ricerca Scientifica 114, 8, 1—47.

Rosi M. & Sbrana A. 1987: Phlegrean Fields. Quaderni De La Ricer-

ca Scientifica, Roma 114, 1—175.

Ross W.C., Halliwell B.A., May J.A., Watts D.E. & Syvitski J.P.M.

1994: Slope readjustment: a new model for the development of
submarine fans and aprons. Geology 22, 511—514.

Royden L., Patacca E. & Scandone P. 1987: Segmentation and con-

figuration of subducted lithosphere in Italy: an important con-
trol on thrust-belt and foredeep-basin evolution. Geology 15,
714—717.

Sacchi M., Infuso S. & Marsella E. 1994: Late Pliocene—Early

Pleistocene compressional tectonics in offshore Campania
(Eastern Tyrrhenian sea). Boll. Geof. Teor. Appl. 36, 141—144.

Sacchi M., D’Argenio B., Morra V., Petrazzuoli S., Aiello G., Budil-

lon F., Sarnacchiaro G. & Tonielli R. 2000: Pyroclastic lumps;
rapidly deforming sedimentary structures beneath the sea floor
off the Naples Bay, Italy. EUG General Assembly, Abstr., Nice.

Scarpati C., Cole P.D. & Perrotta A. 1993: The Neapolitain yellow

tuff – A large volume multiphase eruption from Campi Flegrei,
southern Italy. Bull. Volcanol. 55, 343—356.

Tinti S., Pagnoni G. & Piatanesi A. 2003: Simulation of tsunamis

induced by volcanic activity in the Gulf of Naples (Italy). Nat-
ural Hazards and Earth System Sciences 3, 311—320.

Wei G. & Kirby J.T. 1995: A time-dependent numerical code for

extended Boussinesq equations. J. Wtrwy Port Coastal Ocean
Engineering 120, 251—261.

Wei G., Kirby J.T., Grilli S.T. & Subramanya R. 1995: A fully non-

linear Boussinesq model for surface waves. Part 1. Highly non-
linear unsteady waves. J. Fluid Mech. 294, 71—92.

Zaniboni F., Tinti S., Pagnoni G., Gallazzi S., Della Seta M., Fredi

P., Marotta E., Orsi G., De Vita S. & Sansivero F. 2007: Land-
slides-generated tsunamis in the Ischia island (Italy). Geophys.
Res. Abstr., Vol. 9, 06246, 2007.