Landslide haza d, moni o ing and conse a ion s a egy o he sa egua d
o Va dzia Byzan ine monas e y complex, Geo gia
C. Ma go ini, N. An idze, J. Co ominas, G. B. C os a, P. F a ini, G. Gigli, D. Gio dan,
I. Iwasaky, G. Lollino, A. Manconi, P. Ma inos, C. Sca ia, A. Sonnessa, D.
Spizzichino, N. Vacheish ili
Abs ac
This pape epo s p elimina y esul s o a easibili y p ojec de eloped in coope a ion
wi h Na ional Agency o Cul u al He i age P ese a ion o Geo gia, and aimed a
en isaging he s abili y condi ions o he Va dzia monas e y slope ( upes ian ci y ca e
in he sou h-wes e n Geo gia). The aim is he implemen a ion o a low-impac
moni o ing sys em oge he wi h long- e m mi iga ion/conse a ion policies. A ield
analysis was conduc ed o econs uc geome y o he ocky cli , cha ac e is ics o
discon inui ies, main ailu e modes, and olume o po en ial uns able blocks and
geomechanical pa ame e s. Ins abili y p ocesses a e he combina ion o causa i e
ac o s such as he ollowing: li hology, equency and o ien a ion o discon inui ies,
slope o ien a ion, physical and mechanical cha ac e is ics o slope- o ming ma e ials,
and mo phological and hyd ological bounda y condi ions. The combined adop ion o
di e en su ey echniques (e.g., 3D lase scanne , g ound-based ada in e e ome y)
could be he bes solu ion in he in e disciplina y ield o cul u al he i age p ese a ion
policies. The collec ed da a will be he basis o u u e ac i i ies o be comple ed in
collabo a ion wi h local au ho i ies o a comple e haza d and isk cha ac e iza ion o
he monas e y si e and he de elopmen o an ea ly wa ning sys em o allow sa e
exploi a ion o ou is ic ac i i ies and o his o ical si e p ese a ion.
Keywo ds Plana slide and wedge ailu e . Rock all . Kinema ic analysis . Te es ial
lase scanne . Mi iga ion s a egy . Va dzia (Geo gia)
In oduc ion
The ock-cu ci y o Va dzia is a ca e monas e y si e in sou hwes e n Geo gia,
exca a ed om he slopes o he E ushe i moun ain on he le bank o he M kh a i
i e . The main pe iod o cons uc ion was he second hal o he wel h cen u y. The
ca es s e ch along he cli o some 800 m and up o 50 m wi hin he ocky wall (Fig.
1). The monas e y consis s o mo e han 600 hidden ooms sp ead o e 13 loo s, which
made possible o p o ec he monas e y om he Mongol domina ion. The main lowe
si e was ca ed om he cli cen al laye o olcanic and py oclas ic ocks (Gillespie
and S yles 1999) a an ele a ion o 1300 m abo e sea le el. The ca e ci y included a
chu ch, a oyal hall, and a complex i iga ion sys em. The ea hquake ha s uck
Sams khe in 1283 AD des oyed abou wo hi ds o he ci y ca e and he i iga ion
sys em, exposing he majo i y o he ooms o iew ou side. The si e was la gely
abandoned a e he O oman akeo e in he six een h cen u y. Now pa o a s a e
he i age ese e, he si e has been submi ed o u u e insc ip ion on he UNESCO
Wo ld He i age Lis . The si e is by he ime a ec ed by equen slope ins abili y
p ocesses along he en i e olcanic u açade o he slope. Due o his phenomena, he
Na ional Agency o Cul u al He i age P ese a ion o Geo gia (NACHPG) has
p omo ed, wi h he suppo o ISPRA, a landslide haza d assessmen o he en i e a ea
h ough ock mechanics cha ac e iza ion, geo echnical enginee ing su ey, geo-
s uc u al and kinema ic analysis, slope s abili y model, 3D lase scanne acquisi ions
and elabo a ions, and a eal- ime moni o ing sys em (GB_Rada in e e ome y) o he
iden i ica ion o de o ma ion pa h o he mos haza dous a eas.
The p esen wo k, summa izes bo h he main esul s o he s udies ca ied ou om
Decembe 2011 o Decembe 2012, and he main ou pu coming om he In e na ional
wo kshop on “Landslide haza d assessmen and moni o ing sus ainable echniques o
he sa egua d o Va dzia Monas e y si e” (Geo gia 16– 19, Oc obe 2012) wi h i s
echnical isi and expe mee ings. The wo kshop has been p omo ed wi hin he ame
o he ICL ne wo k on Landslide and Cul u al and Na u al He i age. On his occasion,
in e na ional expe s in he ield o geological and geo echnical p oblems linked wi h
he p ese a ion o his o ical and a cheological he i age we e in i ed o p opose and
de ine sho and long- e m s a egies o he a ea o Va dzia. They could deal di ec ly in
he a ea wi h he p oblems, sha e iews and knowledge wi h local echnicians (capaci y
building ac ions), and ace hypo hesis o in e en ion and implemen a ion o s udies o
he u u e managemen policies o he a ea. The syn hesis o his expe ience and i s
schema and me hodological app oach is desc ibed la e in his pape .
Collec ed da a
Lase scanning su ey
The ock-cu monas e y ci y o Va dzia is loca ed in he sou hwes e n Geo gia, nea
Akhal sikhe close o he Tu kish bo de , acing sou hwa ds on o he M kh a i i e
alley. As a p elimina y da a base, exis ing opog aphical and geological maps ha e
been collec ed and elabo a ed h ough a GIS pla o m in o de o implemen a geo-
da abase o s o e all he in o ma ion use ul o u u e analyses.
In o de o ca y ou a si e-scale speci ic analysis and o suppo he moni o ing sys em
ac i i ies, a de ailed geode ic 3D lase scanning su ey has been pe o med and
implemen ed du ing he May 2012 ield mission. The geode ic su eying was aimed a
he h ee-dimensional (3D) econs uc ion o he ca e monas e y and collec ion o da a
o u u e po en ial slope s abili y analysis o he whole ca e complex. All 3D da a we e
collec ed in a local e e ence ame by means o a e es ial high- esolu ion lase
scanne (TLS), and hen geo e e enced by using high-p ecision dual- equency geode ic
GNSS ecei e s (Ma go ini e al. 2009).
Fig. 1 The ock-cu ci y o Va dzia (Geo gia)
Su eying ac i i ies a he Va dzia si e we e pe o med using a opog aphical Riegl
Z210i lase scanne bo h o measu ing he 3D coo dina es o he models and acqui ing
he RGB in o ma ion linked o he poin clouds. Ten TLS scans we e acqui ed, in o de
o exac ly econs uc he opog aphy o he si e (Fig. 2), om wo di e en scan
posi ions by applying he mul i-scanning echnique (Baldo e al. 2009). A e age spa ial
esolu ion anges om 10 × 10 cm up o some me e s, depending on he dis ance
be ween he lase and he objec . Pu pose-buil a ge s we e used o link he scans
oge he and o build up a high- esolu ion 3D model o he ca e monas e y. The inal
model consis s o a poin cloud made o abou 8,000,000 poin s om which a DTM
wi h a 1 × 1 m cell size has been de i ed.
Fig. 2 Tex u ed 3D model o he ca e monas e y o Va dzia om TLS echniques
Geomo phology and geology
Va dzia is an excellen example o he cul u al landscape (Ma go ini and Spizzichino.
2009) in which human ac i i ies (e.g., exca a ion, cons uc ion, and implemen a ion o
he ca e monas e y, pain ing), na u al, geological, and geomo phological p ocesses, a e
s ic ly connec ed and in e dependen . O en, li hology (e.g., so ock easy o exca a e),
geomo phological p ocesses (e.g. landsliding, e osion and wea he ing), beau i ul and
imp essi e landscapes, and inaccessibili y (se lemen easie o de end and p o ec ) ha e
been he main elemen s in he his o y o humani y conce ning he choice o he
ealiza ion o owns, monumen s, eligious s uc u es, and de ense wo ks (Ma go ini
and Spizzichino 2014). The geomo phological aspec o he Va dzia monas e y is he
esul o a ious long- and sho - e m ac o s (na u al and human) ha a ec ed his pa
o Geo gia. Ac i e ec onics ela ed o he Caucasian Sys em join ly wi h he olcanic
ac i i y had a big in luence on he p esen day mo phology o he su ounding e i o y.
Va dzia slopes a e he inal esul o local seismici y, di e en olcaniclas ic and
py oclas ic alls, e osion, and deposi ion cycles o he M kh a i i e . The en i e ock
wall has a leng h o abou 800 m o a heigh o 130 m wi h a gene al EW o ien a ion.
The slopes, as a gene al ule, p esen a upes ian aspec , mainly s a i ied and
al e na i ely massi e. Ne e heless, discon inui ies o a ious ypes a e p esen ,
po en ially ela ed o he ollowing: cooling phase a e olcanic ac i i y ( e ical),
ec onic ac i i y ( aul s, mino join s, mainly sub- e ical), and geomo phological
ac i i y (s ess elease caused by alley incision). Sub- e ical and high-angle dipping
join s in e sec ing ho izon al bedding (laye s o di e en py oclas ics alls) a e qui e
equen in he a ea and ha e been obse ed du ing ield in es iga ion (E sho e al.
1999). This si ua ion causes po en ial alls, sliding, and oppling o blocks, whose ype,
dimension, and kinema ics depend on local o ien a ion, mechanical p ope ies o
py oclas ic laye s, spacing, and pe sis ence o join s.
The Va dzia monas e y is en i ely cu and ca ed in olcanic and py oclas ic ocks
(Gillespie and S yles 1999) o ma ions (uppe Moi sen–lowe Moi sen); see Fig. 3
(Gudjabidze and Gamk elidze 2003).
Fig. 3 Geological map o Va dzia a 1:15,000 scale (kindly p o ided by NACHPG)
A ep esen a i e li hologic c oss sec ion o he slope has been accomplished du ing he
las ield mission (Fig. 4). The en i e geologic sequence is he inal esul o se e al
olcaniclas ic and py oclas ic alls, cha ac e ized by di e en explosion dynamics and
chemical la a composi ions. Conce ning he s uc u al se ing, a he slope scale, i is
possible o de ine wo main discon inui y sys ems, he i s (sub e ical) due o cooling
o he olcaniclas i sequence and a second one (locally pa allel o he slope ace) mainly
due o s ess elease caused by he alley e osion. The spacing dec eases a dep h
mo ing wi hin he ock mass as isible h ough some mino unnels. P obably, bo h
laye s numbe 1 and numbe 4 (Fig. 4) can be u he subdi ided in o sub-laye s. The
join se s obse ed and p esen in laye 2 also a ec laye 1. The bounda y be ween he
di e en olcaniclas ic le els is no always clea ly ma ked bu hey a e all sub-pa allel
acco ding o he s yle o deposi ion. Fu he analyses a e needed ( hin sec ion and X- ay
di ac ome ic analysis), in o de o classi y he di e en ypes o ock om a
pe og aphical poin o iew.
Fig. 4 Li os a ig aphic c oss sec ion o he Va dzia ocky cli
Geo-mechanical pa ame e s
Geomechanical cha ac e is ics o he olcaniclas ic and py oclas ic ocks ou c opping in
he s udy a ea ha e been de e mined by means o geo echnical ield su eys and
labo a o y es s, on ock blocks. The ollowing ac i i ies ha e been ca ied ou du ing
he 2011 and 2012 ield su ey:
– Geomechanical ock mass classi ica ion, in o de o de i e he main geomechanical
cha ac e is ics and indexes (e.g. RMR, GSI, Q sys em)
– Na u al uni weigh
– Til es
– Schmid -hamme es on join su aces and in ac ock block o in si u analysis o
UCS (uncon ined comp essi e s eng h)
– Poin load es o p o ide UCS da a om sampled blocks (ISRM 1978, 1981)
– S eng h and de o ma ion pa ame e s om scien i ic and echnical li e a u e (Hoek
2007; Ba on 1973; ISRM 1981) as well as om local echnical epo s
The p elimina y si e in es iga ions p o ided he esul s summa izedin Table 1.
Li hology
Na u al uni
weigh (kN/m3)
UCS om Schmid -
hamme
es (MPa)
Basic join ic ion
angle (ϕ°) ( om il
es )
UCS om poin
load es (MPa
Laye 1, black olcanic
b eccia
18.9
19
14.8
Laye 2, ligh massi e
u
17.5
17
22º-32º
9.6
Laye 3, whi ish u
wi h da k pumices
15.1
15
22º-32º
10.8
Table 1 Main geomechanical pa ame e s o he Va dzia ocks
Li hologies a e qui e weak (Lembo Fazio e al. 1984), as usual in a eas whe e human
being ealized impo an unde g ound se lemen s. Uniaxial comp essi e s eng h and
bulk densi y o he di e en laye s ange be ween 10 and15 MPa and 15 and 19 kN/m3,
espec i ely. These alues a e compa able o hose desc ibed o simila li hologies
(Moon 1993; Tuğ ul 2004; Binal 2009; Del Po o and Hü limann 2009; Rod íguez-
Losada e al. 2009; Pola e al. 2012), and hey will be de ined in mo e de ail by means
o u u e es ing campaigns.
Geo-s uc u al analysis
The mechanical beha io o ocks is gene ally con olled by he condi ions unde which
de o ma ion akes place and by he inhe en ock p ope ies (Tu ne and Schus e 1996).
S ess condi ions o Va dzia monas e y slope p omo ed di e en ypes o discon inui ies
in he olcaniclas ic igneous ocks (Mi chell and Wes away 1999). This is due o he
s ess-s ain beha io (e.g., la e al unloading due o alley e osion coupled wi h ec onic
s ess and cooling o he py oclas ics alls) and o he ypical pe og aphical and
physical p ope ies o he so ocks (e.g., chemical and mine alogical composi ion,
g ain size, shape, hickness, homogenei y, po osi y, pe meabili y, ype o cemen ). The
p esence o b i le olcanic and py oclas ic ocks in he a ea, as a gene al ule, p omo es
blocky ock masses due o high equency o discon inui ies o a ious o igins
(Gamk elidze 1986). A geomechanical cha ac e iza ion o discon inui ies has been done
along he en i e s udy a ea in o de o desc ibe he s uc u al se ing o he a ea and
ecognize e en ual mas e join s. Rock wea he ing, slope deg ada ion, and ins abili y in
he a ea a e he esul o di e en s uc u al combina ions o he main join se s
(Spizzichino e al. 2009). Ou analyses, al hough only a p elimina y and no exhaus i e
discon inui y su ey was pe o med o e he whole a ea, allowed o cha ac e ize some
ele an slope sec o in e ms o discon inui y o ien a ion. All collec ed da a ha e been
analyzed (DIPS® Rocscience so wa e). The ock mass s eng h has been obse ed o
depend on he densi y, na u e, and ex en o ac u e and ock s eng h (i.e., lowe wi hin
he black olcanic b eccia, highe wi hin whi ish u and lapilli u ). The s e eoplo s o
Fig. 5 show ha he majo join sys ems a e s eeply dipping (70°) owa ds SE (dip
di ec ion 140°–165°). A p elimina y kinema ic analysis ocused on he iden i ica ion o
possible ins abili y modes has been pe o med o sliding (plana , wedge) and oppling
(di ec and lexu al) phenomena. The main esul s a e also shown in he s e eoplo s o
Fig. 5.
Fig. 5 S e eoplo s o join dip measu emen s om he Va dzia monas e y a ea
(le ) and global kinema ic analyses ( igh ): a di ec oppling (0.42 % o o al
planes), b lexu al oppling (no c i ical planes), c plana sliding (32.88%o o al
planes), and d wedge ailu e (44.59 % o o al planes)
Slope s abili y and e olu ion
The si e o Va dzia is a ec ed by many examples o slope ins abili ies such as ock all,
opple, and ock slide, wi h olumes anging om a ew cubic me e s o he housands
(WP/WLI 1993). Many phenomena also occu ed in ecen yea s, such as in May 1968
when a huge p o ec i e wall was cons uc ed jus abo e he main chu ch, o suppo he
ock cli ins abili y, ill he e y ecen ock slide o sp ing 2011.
Following he gene al scheme o he c oss sec ion de ined in Fig. 4 and wi h he aim o
p o ide a p elimina y assessmen acco ding o slope ailu e ype and p ocesses a ec ing
he en i e ocky cli ( om op o bo om), mos ailu es can be classi ied in o a ious
ca ego ies (C uden 1991; C uden and Va nes 1996), depending on he combina ion o
he main join se s and he geomechanical and geo-s uc u al condi ions o he di e en
laye s. The only excep ion is he la ge o o- ansla ional ock slide, po en ially a ec ing
bo h laye s 1, 2, and 3. Wi h e e ence o he di e en le els o Fig. 4, he main ock
slope ailu es can be classi ied as ollows and displayed in Fig. 6.
Fig. 6. Slope ailu e ype and p ocesses a ec ing he ocky cli o Va dzia. a
La ge o o- ansla ional ock slide in ol ing bo h laye s 1, 2, and 3. b Rock all
sca p in he uppe pa o laye 1. c Wedge ailu e a he con ac be ween laye 1
(on op) and laye 2. d High- and low-angle po en ial plana ock slide. e High- and
low-angle po en ial plana ock slide
In le el 1:
a) La ge o o- ansla ional ock slide a ec ing laye s 1, 2, and 3 (Fig. 6a)
b) Rock all, opple, o wedge ailu e o en due o he lack o suppo o he unde lying
laye equen ly a ec ed by plana sliding (Fig. 6b)
c) Rock all o clas s, de ached om he ma ix o olcaniclas ic b eccia
In le el 2:
d) Wedge ailu e and ock alling (Fig. 6c)
e) High- and low-angle plana ock sliding along he main discon inui ies (Fig. 6d)
In le el 3:
) Plana sliding along he main s uc u al se ings (Fig. 6e)
g) Rock all and opples
Fu he mo e, widesp ead supe icial e osion mo phologies ea u es ha e been
ecognized and mapped along he slope. They a e mainly due o wa e uno and mino
s eam ne wo k able o c ea e slope sedimen s and small an del a a he oe o he slope.
Depending om he slope angle, al e na ing e osion, anspo a ion, and deposi ion o
such deb is ma e ial is occu ing.
Moni o ing sys em: p elimina y esul s and u u e implemen a ion
G ound-based ada in e e ome y
Conside ing he mo phological se ings (slope ex en ca. 105 m2) and slope ins abili y
p ocesses (di e en ypologies in size, magni ude, and p obabili y o occu ence), a new
ad anced simple and lexible moni o ing sys em has been implemen ed in o de o
ob ain, namely, measu emen s, p ocessing, and emo e con ol in eal ime, and o
ans o m he moni o ing sys em in o a wa ning sys em. The sys em adop ed o he
moni o ing o he en i e cli is based on a g ound-based in e e ome ic ada (IBIS
sys em om IDS Fig. 7). This equipmen allows he moni o ing o displacemen in he
line o sigh wi h a esolu ion o millime e .
Fig. 10 Rock b idges in join s a ec ing massi e lapilli u (laye 2 o Fig. 4)
E alua ing he s eng h o he ock b idges is undamen al o he assessmen o he
haza d po en ial and associa ed isk. The ock mass is assumed o ail in shea and
ension while p o ides a s eng h componen p opo ional o he ac ion o he ailu e
plane ha is s ill in ac (Jennings 1970). The impo ance o ock b idges on he s abili y
o ock slopes has been add essed among o he s by Eins ein e al. (1983) and Nichol e
al. (2002).
Failu e in non- ully pe sis en (in e mi en ) join s may ake place along a single su ace,
o a s ep-pa h su ace, o h ough in ac ock. The s eng h (τ) along he plane o he
lis ic join s may be exp essed by he ollowing equa ion (Jennings 1970):
=k ˙ (cj + σn anΦj) + (1 – k)˙(ci + σn anΦi)
whe e cj is he cohesion in e cep o he join segmen , Φj is he ic ion angle o he
join segmen , ci is he cohesion in e cep o he ock b idge, Φi is he ic ion angle o
he in ac ock, σn is he no mal s ess uni o mly dis ibu ed be ween he join segmen
and he ock b idge, and k = lj/(lj + li) is he a io o join segmen o e he o al leng h.
Ano he possibili y is o eso o a ac u e mechanics app oach. In his case join s a e
conside ed as ac u es, which can p opaga e in o he ock b idges, bo h in mode I
( ensile s esses) and in mode II (shea s esses). The p opaga ion o he join s can be
assessed using a p opaga ion c i e ion on he basis o he ock oughness alue.
Nume ical codes ha e been de eloped (Sca ia 1995) o e alua e he possibili y o a
ock olume o ail on he basis o he simula ion o join p opaga ion in o he ock cli .
The s abili y analysis has accoun o he seismic e ec as well. As i has been al eady
men ioned, in 1283 AD, an ea hquake des oyed app oxima ely wo hi ds o he ci y
and collapsed he i iga ion sys em (Gap indash ili 1975). His o ically up o 6.5–7
ea hquake magni udes ha e been obse ed wi hin he dis ance o 10 km om he
Va dzia. A pseudo-s a ic limi equilib ium analysis is ecommended o each po en ial
block slide iden i y along he cli , whe eby he des abilizing e ec o an ea hquake is
ep esen ed by a ho izon al o ce. The alue o his o ce has o be calcula ed
conside ing he expec ed peak g ound accele a ion (PGA). To accoun o he
opog aphic ampli ica ion, he PGA will be inc eased by linea in e pola ion be ween
he base and he op o he cli , ollowing he me hodology de eloped by Ma ouli e
al. (2009). The geome y o each po en ially uns able ock block has o be de e mined
based on ield su eys o he ock mass. Scan-lines combined wi h high- esolu ion
pho os a e going o be used o quan i y he ex en o ock b idges. Join pe sis ence
is he mos di icul pa ame e o ob ain because di ec mapping o discon inui ies
in e nal o a ock mass is no possible unless he ailu e has occu ed and plane is
al eady exposed. The la e case will be used o back-analysis. Indi idual ock blocks
migh be bounded o mo e han one kinema ically uns able join . In such a case, he
s abili y analysis has o be pe o med o bo h geome ies. Measu es o he in si u cli
displacemen s p o ided by he GBSAR will p o ide a comple e iew o he on-going
ins abili y p ocess in he ock ace and will con i m he bounda ies and size o he
uns able ock block. A he same ime, i may become a p elimina y ea ly wa ning
sys em. A he same ime, he e a e su icien e idences o suppo he possibili y o
deep ailu es as occu ed and wi nessed du ing he 1283 AD ea hquake. The a ailable
de ailed opog aphic su eys o he ocky cli and o he ooms and ca i ies exca a ed
wi hin he slope could allow he econs uc ion o a ully 3D nume ical model. This
model could be used as a ool in o de o: (i) iden i y mos s essed poin s (e.g., pilla s,
aul s, walls) o be ein o ced o p o ec ed, (ii) e ec o possible ein o cemen s o be
designed and implemen ed, (iii) ex en o possible local and gene alized ailu es unde
bo h s a ic and dynamic condi ions. The e o e, i is sugges ed o pe o m a ully 3D
FEM/FDM modeling o he en i e slope.
TLS poin s cloud could be used o iden i y single uns able blocks and o cha ac e ize
hem acco ding o hei geome y and posi ion along he cli , as well as o de ec he
mos c i ical ock mass sec o s by means o de ailed ue 3D kinema ic analyses, by
assigning di e en weigh s o he discon inui ies (ei he semiau oma ically ex ac ed
om TLS da a o measu ed in si u) based on hei ue posi ion on he in es iga ed ock
mass (Abellán e al. 2010; Gigli and Casagli 2011). This in o ma ion is manda o y o a
eliable ock all modeling which could be pe o med h ough a 3D nume ical code. In
ac , local cli o ien a ion, p esence o small benches (cli pa hs be ween ooms), and
he complex geome y o he alus and lowe ocky slopes could o ce blocks o ollow
anomalous ajec o ies (C os a and Aglia di 2003, 2004; Aglia di and C os a 2003). The
Hy-S one ock all code (C os a and Aglia di 2004) could allow o ca y ou a de ailed
and obus modeling o ock all phenomena and o accomplish a eliable haza d and isk
modeling (F a ini e al. 2008; Aglia di e al. 2009) s a ing om all he a ailable
in o ma ion and es ing o di e en scena ios (block size, de achmen posi ion, block
physical mechanical cha ac e is ics). In ac , he exposu e o isi o s along he
app oaching oad and pa hs as well as along he cli is high o e e y la ge a eas an
emedial mi iga ion wo ks a e equi ed o gua an ee a su icien sa e y ma gin. In ac ,
he p oposed Hy-S one code is able o include he e ec s o passi e coun e measu es
(e.g. ca ch ne s, embankmen s, di ches) o o na u al obs acles (e.g. ege a ion) as well
as o seconda y ajec o ies de i ing om block agmen a ion (i.e. explosion a impac
poin s).
Risk mi iga ion policies and sus ainabili y
Up o now, no majo mi iga ion s a egies we e implemen ed in Va dzia. Few bol s we e
locally ins alled, bu ess and wall s uc u es we e execu ed in collapsed a eas o p e en
u he deg ada ion, he mid-le el pedes ian pa h was pa ially econs uc ed, and a
manual c ack gauge moni o ing was ope a ing in he pe iod 1968–1979. Cli ins abili y
mi iga ion measu es usually equi e he use o ei he s uc u al o non-s uc u al
measu es, o a combina ion o bo h. S uc u al measu es imply in ensi e ea hwo k and
he cons uc ion o conc e e s uc u es (VanDine 1996) and may be isually in usi e.
Design o s abiliza ion and p o ec ion wo ks o he Va dzia cli ha e he challenge o
econciling sa e y and conse a ion o he si e wi h sus ainabili y and isual in eg a ion.
Su ace ein o cemen echniques can be used o p e en small-size ock wedges and
blocks o p oduce ock alls. They aim a p o iding su ace ein o cemen and es ic
loosening o he ock mass. High-s eng h masses ha a e low isually in usi e can be
conside ed usually combined wi h nails o ock bol s o ix hem o he ock su ace
(Co ominas 2013). S abiliza ion elemen s ha e o be ca e ully selec ed. Volcanic ocks
in which he ca es ha e been exca a ed a e highly wea he able ma e ials ha
expe ience as su ace de e io a ion, a eling, and spalling (Fig. 11). Because o his,
some measu es may become ine icien . The chemical na u e o he lapilli and pumice
has o be de e mined in o de use compa ible p oduc s. Resul s om he ma hema ical
models (slope s abili y and ock all scena ios), combined wi h seismic ac i i y analysis
and his o ical ock all da abase analysis, will allow o each a comple e desc ip ion o
u u e possible ailu e scena ios and o p opose suscep ibili y and haza d zoning, up o a
isk assessmen o exposed people and s uc u es. This analysis will also suppo he
choice and design o isk mi iga ion coun e measu es o be ins alled o a sa e
exploi a ion o he his o ical si e. The main ocus o he adop ed solu ions will be he
low en i onmen al impac , o enhance he impo ance o sus ainable solu ions,
especially when dealing wi h cul u al he i ages.
Fig. 11 Spalling o he ock su ace in he oo o he oom, con olled by ock
wea he ing and s ess deg ada ion
Conclusions
Acco ding o he p esen s udy and ield missions, p edominan and mo e equen
geomo phological p ocesses in Va dzia a e, om he op o he cli and wi h e e ence
o Fig. 4, as ollows:
1. E osion and wa e un-o on he ca chmen abo e si e. These phenomena may cause
small ock-s one all, du ing hea y ain all. Limi ed slope sedimen s and small an del a
ha e been ecognized a he oe o he slope.
2. In he uppe black olcanoclas ic b eccia (le el 1 in Fig. 4), he e a e mainly ock all
o clas s, de aching om he ma ix. Tou is s a e hemos exposed o hese phenomena.
Occasionally, opple o wedge ailu e can be ecognized, o en due o he lack o
suppo o he unde lying laye , equen ly a ec ed by plana sliding.
3. In he ligh clea massi e lapilli- u (le el 2 in Fig. 4), he main ock slope ailu e a e
mainly due o wedge ailu e and ock alling and high- and low-angle plana ock
sliding along he main discon inui ies.
4. In he clea lapilli- u wi h da k pumicies, sco ias, and bombs (le el 3 in Fig. 4), he
main obse ed and po en ial ock slope ailu es a e plana sliding along he main
s uc u al se ings and opples.
5. La ge o o- ansla ional ock slide a e occasional a ec ing laye s 1, 2, and 3, and may
be du ing s ong ea hquakes.
I ems 1 and 2 may mainly in ol e ou is s in hei e olu ion whe e i ems 3, 4, and 5 a e
mainly ele an o he conse a ion o cul u al he i ages.
As a consequence o he abo e, also conside ing ha he po en ial ins abili y p ocesses
and mechanisms obse ed o he en i e ock cli can be e e ed o di e en ailu e
modes (o hei combina ion) and di e along he whole cli in e ms o ac i i y, hei
s a e ( empo al e olu ion), dis ibu ion (spa ial e olu ion), s yle (combina ion and
epe i ion o di e en ailu e mechanism), and magni ude (landslide in ensi y and
po en ial olumes), u u e in es iga ions will equi e he ollowing:
1. A be e unde s anding o he ole and combina ion o di e en p edisposing ac o s
such as he ollowing: li hology, p esence, equency and o ien a ion o discon inui ies
s. slope o ien a ion physical and mechanical cha ac e is ics o ma e ials, mois u e, and
mo phological and hyd ological bounda y condi ions.
2. A be e unde s anding o igge ing ac o s, such as, among he many po en ial,
seismici y, he mal e ec s, ain in il a ion, and un-o , o he a ie y o iden i ied
phenomena 3. Mo e accu a e and speci ic analysis o some o he de ec ed po en ial
ins abili y condi ions (slope scale isk analysis) in o de o de ine ailu e s a e in e ms
o in ensi y, po en ial e olu ion, and impac
4. The coupling o di e en su ey echniques (e.g., 3D lase scanne , enginee ing
geological and geomechanical ield su eys, g ound-based ada in e e ome y), o be
adop ed in he in e disciplina y ield o cul u al he i age p o ec ion and conse a ion
policies
5. The main aining and analysis o eal ime moni o ing sys em, sui able o he speci ic
si e condi ions being able o accoun
o all he abo emen ioned si e peculia i ies 6. The implemen a ion o a local seismic
ne wo k and he in es iga ion o local seismic esponse and ampli ica ion
Finally, a mi iga ion p ojec add essed o e alua e and mi iga e isk associa ed o slope
ailu e will be implemen ed, o bo h sho and long- e m ac ions. This will emphasize
he low en i onmen al impac o adop ed solu ions, as undamen al managemen ool o
p epa e a mas e plan o a sa e and sus ainable u u e ou is exploi a ion o he
his o ical si e.
Acknowledgmen s
This pape has been de eloped wi hin he ac i i ies o he In e na ional Conso ium o
Landslides, ne wo k on “Landslides and Cul u al & Na u al He i age (LACUNHEN)”.
The au ho s a e e y g a e ul o all he Na ional Agency o Cul u alHe i age
P ese a ion o Geo gia s a membe s o hei suppo and con inuous help.
Re e ences
Abellán A, Cal e J, Vilaplana JM, Blancha d J (2010) De ec ion and spa ial p edic ion
o ock alls by means o e es ial lase scanne moni o ing. Geomo phology
119:162–
171
Aglia di F, C os a G (2003) High esolu ion h ee-dimensional nume ical modelling o
ock alls. In J Rock Mech Min Scie 40(4):455–471
Aglia di F, C os a GB, F a ini P (2009) In eg a ing ock all isk assessmen and
coun e measu e design by 3D modelling echniques. Na Haza ds Ea h Sys Sci
9:1059–1073
Baldo M, Bicocchi C, Chiocchini U, Gio dan D, Lollino G (2009) LIDAR moni o ing
o mass was ing p ocesses: he Radico ani landslide, P o ince o Siena, Cen al
I aly. Gemo phology 105:193–201
Ba on NR (1973) Re iew o a new shea -s eng h c i e ion o ock join s. Eng Geol
7:287–332
Binal A (2009) P edic ion o mechanical p ope ies o non-welded and mode a ely
welded ignimb i e using physical p ope ies, ul asonic pulse eloci y, and poin load
index es s. Q J Eng Geol Hyd ogeol 42:107–122
C uden DM (1991) A simple de ini ion o a landslide. Bull In Assoc Eng Geol 43:27–
29
C uden DM, Va nes DJ (1996) Landslide ypes and p ocesses. In: A.K. Tu ne , R.L.
Schus e (eds.) Landslides in es iga ion and mi iga ion (Special epo 247, pp. 36-
75). T anspo a ion Resea ch Boa d, Washing on, D.C
Co ominas J (2013) A oidance and p o ec ion measu es. In: Sh ode JF (ed) T ea ise on
Geomo phology, ol 7. Academic P ess, San Diego, pp 259–272
C os a GB, Aglia di F (2003) A new me hodology o physically-based ock all haza d
assessmen . Na Haza ds Ea h Sys Sci 3:407–422
C os a GB, Aglia di F (2004) Pa ame ic e alua ion o 3D dispe sion o ock all
ajec o ies. Na Haza ds Ea h Sys Sci 4:583–598
Del Po o R, Hü limann M (2009) The dec ease in he shea s eng h o olcanic
ma e ials wi h a gillic hyd o he mal al e a ion, insigh s om he summi egion o
Teide s a o olcano, Tene i e. Eng Geol 104:135–143
Eins ein HH, Veneziano D, Baeche G, O’Reilly K (1983) The e ec o discon inui y
pe sis ence on ock slope s abili y. In J Rock Mech Min Sci Geomech Abs
20:227–236
Elash ili M, Ja akhish ili Z (2004) h p://www. dsn.o g/mee ings/2004/
Geo gia_FDSN_2004.pd
E sho AV, B une MF, Ko o ae MV, Nikishin AM, Bolo o SN (1999) La e
Cenozoic bu ial his o y and dynamics o he No he n Caucasusmolasse basin:
implica ion o o eland basin modeling. Tec onophysics 313:219–241
F a ini P, C os a GB, Ca a a A, Aglia di F (2008) Assessmen o ock all
suscep ibili y by in eg a ing s a is ical and physically-based app oaches.
Geomo phology 94:419–437. doi:10.1016/j.geomo ph.2006.10.037
Gamk elidze IP (1986) Geodynamic e olu ion o he Caucasian and adjacen a eas in
Alpine ime. In: Zoneshain LP (Ed.), Tec onics o he Eu asian Fold Bel s.
Tec onophysics 127, 261-277
Gap indash ili G (1975) (in English, Russian, Geo gian) Ancien monumen s o
Geo gia: Va dzia. Au o a A Publishe s, Lening ad, pp 7–25. ISBN 978-1-135-
68320-7
Gigli G, Casagli N (2011) Semi-au oma ic ex ac ion o ock mass s uc u al da a om
high esolu ion LIDAR poin clouds. In J Rock Mech Min Sci 48:187–198
Gillespie MR, S yles MT (1999) BGS ROCK Classi ica ion Scheme Volume 1
Classi ica ion o Igneous Rocks. B i ish Geological Su ey Resea ch Repo s
(2nd edi ion) RR 99-06
Gudjabidze GE, Gamk elidze IP (2003) Geological Map o Geo gia, 1:500.000.
Geo gian S a e Depa men o Geology and Na ional Oil Company “Saqna obi”
Hoek E (2007) P ac ical Rock Enginee ing (2007 edi ion) h p://www. ocscience.com
ISRM - Commission on S anda diza ion o Labo a o y and Field Tes s (1978) –
Sugges ed me hods o he quan i a i e desc ip ion o discon inui ies in ock
masses. In . Jou n. Rock Mech. Min. Sci. & Geomech. Abs ac s 15, n.6, 319-368
ISRM- Commission on Classi ica ion o Rocks and Rock Masses (1981) - Basic
geo echnical desc ip ion o ock masses. In . Jou n. Rock Mech. Min. Sci. &
Geomech. Abs ac s 18, 85-110
Jennings JE (1970) A ma hema ical heo y o he calcula ion o he s abili y o open
cas mines. P oc. Symposium on he Theo e ical backg ound o he Planning o
Open Pi Mines. Johannesbu g. pp. 87-102
Lembo Fazio A, Man edini G, Ribacchi R, Scio i M (1984) Slope ailu es and cli
ins abili y in he O ie o hill. 4 h In . Symp. on Landslides, To on o, 2, 115-120
Ma go ini C, Ma sella M, O lando L, Pandol i O, Soddu P, Sonnessa A, Spizzichino D,
Delmonaco G, Ca da eli G, De Donno G (2009) Low impac in es iga ion
echniques o pa ame e iza ion and conse a ion o Cul u al He i age: he case
s udy o Moai s a ues a Eas e Island (Chile). Geoi alia 2009 - ses o o um i aliano
di scienze della e a, Rimini 09-11 se emb e
Ma go ini C, Spizzichino D (2009) The managemen o cul u al and en i onmen al
he i age si es: a pi o o conse a ion and enhancemen . P oceedings o he i s
In e na ional Symposium on Danxia Land o m. The 2nd collec ion. Danxiashan,
Guadong (China), 26-28 May, 2009
Ma go ini C, Spizzichino D (2014) How geology shapes human se lemen s. In
Banda in F. & an Oe s R. (eds.), Reconnec ing he ci y. The His o ic U ban
Landscape App oach and he Fu u e o U ban He i age. Chiches e : Wiley
Blackwell
Ma ouli O, Co ominas J, Wa man J (2009) Me hodology o e alua e ock slope
s abili y unde seismic condi ions a Solá de San a Coloma. Ando a Na Haza ds
Ea h Sys Sci 9:1763–1773
Mi chell J, Wes away R (1999) Ch onology o Neogene and Qua e na y upli and
magma ism in he Caucasus: con-s ain s om K-A da ing o olcanism in
A menia. Tec onophysics 304:157–186
Moon VG (1993) Mic os uc u al con ols on he geomechanical beha io o ignimb i e.
Eng Geol 35:19–31
Nichol SL, Hung O, E ans SG (2002) La ge-scale b i le and duc ile oppling o ock
slopes. Can Geo ech J 39:773–788
Tuğ ul A (2004) The e ec o wea he ing on po e geome y and comp essi e s eng h
o selec ed ock ypes om Tu key. Eng Geol 75:215–227
Pola A, C os a G, Fusi N, Ba be ini V, No ini G (2012) In luence o al e a ion on
physical p ope ies o olcanic ocks. Tec onophysics 566–567:67–86
Rod íguez-Losada JA, He nández-Gu ié ez LE, Olalla C, Pe ucho A, Se ano A, E -
Da wich A (2009) Geomechanical pa ame e s o in ac ocks and ock masses om
he Cana y Islands: implica ions on hei lank s abili y. J Volcanol Geo he m Res
182:67–75
SHRAP (2009) APRIL P og ess Repo – 2009, Seismic Haza d and Risk Assessmen
o Sou he n Caucasus-Eas e n Tu key Ene gy Co ido s (SHRAP) h p://
www.koe i.boun.edu. /dep emmuh/eski/na o/p ojec /pd /p og ess1_983038.pd
SHRAP (2010) MAY P og ess Repo – 2010, Seismic Haza d and Risk Assessmen o
Sou he n Caucasus-Eas e n Tu key Ene gy Co ido s (SHRAP) h p://
www.koe i.boun.edu. /dep emmuh/eski/na o/p ojec /pd /p og ess3_983038.pd
Sca ia C (1995) A me hod o he s udy o c ack p opaga ion in ock s uc u es,
Géo ecnique, 45, No 3, 447-463, The ins i u ion o ci il enginee s, London
Spizzichino D, Delmonaco G, Ma go ini C (2009) Geo-s uc u al modelling o
po en ial la ge ock slide in Machu Picchu- Eu opean Geosciences Union 2009
Gene al Assembly, Vienna Aus ia, 19-24 Ap il
Tu ne AK, Schus e RL (1996) Landslides in es iga ion and mi iga ion (Special epo
247) T anspo a ion Re-sea ch Boa d: Washing on D.C., pp. 673
WP/WLI (1993) A sugges ed me hod o desc ibing he ac i i y o a landslide. Bull In
Assoc Eng Geol 47:53–57
VanDine DF (1996) Deb is low con ol s uc u es o o es enginee ing: B i ish
Columbia Minis y o Fo es s Resea ch P og am, Vic o ia, B.C., Wo king Pape
08/1996. www. o .go .bc.ca/h d/pubs/Docs/Wp/Wp22.h m. Accessed 12 No 2009
