Journal of Coastal Research 537-554 West Palm Beach, Florida Summer 2002
Recent Evolution of Surge-related Events in the Northern Adriatic Areal
Paolo Antonio Pirazzoli] and Alberto Tomasinj i"CNRS-Laboratoire de tCNR Geographic Physique Istituto per 10 Studio della 1 Place Aristide Briand Dinamica delle Grandi 92195-Meudon, France Masse [email protected] Venezia and Universita di Venezia Venezia, Italy [email protected]
ABSTRACT_••••••••••••••••••••••••••••_ PIRAZZOLI, P.A., and TOMASIN, A., 2002. Recent evolution of surge-related events in the Northern Adriatic area. ,tflllllll:. Journal o(Coastal Research, 18(3), 5~n-554. West Palm Beach (Florida), ISSN 0749-0208. ~ The recent increase in the frequency of coastal flooding in Venice mainly depends on loss ofland elevation (subsidence euus~ :....I and eustasy) and man-induced hydrodynamic changes in the lagoon area, but can also be strengthened in the near ---=-=-&--*" future by changes in climate. _+ In this paper, after a short review of recent changes in the relative mean and maximal levels of the sea and their causes, the main meteorological factors (atmospheric pressure and winds of sirocco and bora), which produce sea surges in the Gulf of Venice, are identified statistically. The recent evolution of these meteorological factors in the Adriatic area shows some favorable trends (the atmospheric pressure is increasing, thus provisionally masking eu static sea-level rise, and bora is sharply lessening). However, the effects of sirocco, which is increasing in frequency in the mid-Adriatic, seem to be prevailing. On the whole, the frequency of sea surges 2:5 em to 2:30 em, which are the most frequent, is increasing in the North Adriatic and this implies more "moderately high tide levels", which are however liable to flood the lowest parts of the city of Venice. Such increase in frequency seems related to recent climatic changes (possibly related to global warming) and may therefore be expected to worsen in the near future.
ADDITIONAL INDEX WORDS: Lagoon hydrodynamics, flooding, surges, global warming.
INTRODUCTION CHANGES IN LAND ELEVATION AND IN LAGOON HYDRODYNAMICS Flooding by sea surges (locally called "acqua alta", literally "high water") is a frequent phenomenon along the low coastal In the city of Venice the ground level is today very close to areas of the northern Adriatic. In lagoon areas, coastal flood sea level. Compared to the astronomical tidal range, that may ing is known to have occurred occasionally at least since the reach about one meter at spring tide, and to storm surges, descriptions by Cassiodorus (6th c. ADl, but in the city of which exceed a height of one meter once every five years on Venice the phenomenon has much worsened during the 20th average, the lowest point of the city (in St. Mark's Square, century. The worsening can be ascribed to the superposition just outside the basilica) is now only 40 cm above the present of several causes: lowering of the elevation of the land in re MSL (which is today about 23 cm above the local Datum). lation to the sea (BONDESAN et al., 1995); changes in the prop This means that all surges in the Adriatic, even moderate, agation ofthe sea (tide and surges) inside lagoon basins (PTR may produce some flooding. At street level, about :~O(Yr· of the AZZOLl, 1987, 1991); and possible changes in the climatic fac city is lower than the elevation of 120 ern (above Datum), tors which tend to produce surge events. In this paper, after which is reached by sea level (tide plus surge) 1.3 times per a briefsummary on some factors which have affected the land year on average. This makes that each centimeter of eleva elevation and the lagoon hydrodynamics, a statistical study tion is precious. All the events with tide level 2:110 cm are is presented of hourly surges (differences between the ob officially considered "acqua alta". served tide and the astronomical tide at the same moment) Regular tide-gauge records are available in Venice since and of the recent evolution of the meteorological factors 1872 (DoRIGo, 1961; PIRAZZOLI, 1982) and in Trieste (at the which contribute to surge development and to "acqua alta" northern end of the Adriatic, Figure 1) since 1890 (MOSETTJ, occurrences. 1961). In Figure 2 the MSL evolution in the two stations is compared. A 19-year running mean has been used to filter the effects of the astronomical cycles of lunar declination 01072 received 1 July 2001; accepted in revision .5 February 2002. I Invited paper presented orally at the AGU 2001 Spring Meeting, (18.6 years), Saros (18 years and 11 days) and Meton (about Boston, Ma., USA. 19 years). It can be observed that the 19-yr curves of Venice Pirazzoli and Tn mu siu
o 75 150 km I I I
Figun ' I . l.oca t iun map of' t ill' an':1 st udi erl . For mu rr- dt't ail :-: 0 11 ti ll' Ve-n ice Lagoon, :-'('P Fi gu n' ·t,
a nd 'I'riestu a re a lmost parall el unti l th e firs t hal f of t he a lso been observed t hut during th e lust three decad es MSL I92()'; and th en aga in from t he second half ofthe 19(iO,;. Th e rem a in ed on avera ge ulmost. stuhle in both Triest e a nd Ven difference in level th a t gra dua lly incr eases between Tri este ice, Thi s is in contrast not on ly with th e gradual sen-level a nd Veni ce, reaching about 10 em in th e la te 1960 s, is as rise of th e preceding decad es. but a lso with th e pres en t ly ex cribod mo,;t ly to la nd subs ide nce in th e Ven ice a rea , ca used pect ed globa l sea- Ieve ] rise. by r-xccs si v« pumping of g round wa te r, es pecia lly in th e in Th e recent a ppa re nt seu-Iovcl stubility at T ri este a nd du st ria l zuno of IV1a rghera ( C,\HB( )( ;" IN l'! (/1.. 1976 1, Signifi Ven ice may be s hort-livcd how ev er. T SI ~II' I. I S a nd B..\ KEH ca nt la nd si nkim; ph enomen a. due to methane ext ract ion . oc 1:2 00()I ha ve s uggested th at a recent sea-Icve l decr ea se in cur in ot he r coustul areas of' t hc nort.h crn Adriatic (Bi\lll'! (/1.. Mediterranean sta t inn s may be th e result of in cr ea ses in 2()OOI. rc- sultiru; in extensive lowla nds below lVISL ( B ONIlES,\N t e m pe ra t ure a nd sa lin ity of Moditorru ncnn Deep Water. l'! «t.. 199il I. How e ver, s ince th e no rt he rn Adriatic Sea is not deep I less From th e tidl'-gauge record s of th e period 1872- 2000. it t ha n fiO m : other ca uses for t ill' recent interruption in th e a ppea r,; th at tlu. tota l 10';'; in eleva tion of Venic e has been sea-Ievel ri ,;t, s hou ld be a d voca te d. It wi ll be shown below ab ou t :2 il em from 1HH1 to 1991 (wit h a mean running over (Fig ure 1:2 ) th at th ere hus been in th e Adriatic a rea an l~) yea r,;. th e first a nd th e last 9 yea rs of t he series a rc lost inc rea se in th e a ve rage su rfa ce uu nosphuric pressure of in a vera ging opernt ions r, of wh ich about 10 ern ca n be as a bo ut 2 h Pa during th e second part of the 20th century cri hcd to man -induced la nd subsidence a nd a bout I f) em (a t a nd that most of this i ncrcuse occu rred since t he 19605. Veni ce an d Tri esu» to eus tasy a nd regional tectonics, It has The hyd rosta ti c effect of s uch a n in crea se in pressure wa s
.Iou rn ul of Coast ul Ik sl'ar l'h. Vol. IH. "'0,:l. ~() () ~ Surg'p Evolution in t.hc- Adriatic
MSL at Venice (1872-2000) compared to MSL at Trieste (1890-2000) 30
, E 25 " " ~ E ,- " ..::J 20 ltl 'tl Ql ,-10 em o 'c 15 Ql "" > ,,' ... , r- 24,7 em 0) 10 • ',i " ....CIO ,IY ' Ql , ' .c:.. 5 '~', Ql o> .c , "'~':Y~'I ..ltl o ~.. .. , s: ;. Cl .. . "Qj :I: -5
-10 "
Time (years) Annual MSL at Venice -19-yr running mean of MSL at Venice -19-yr running mean of MSL at Trieste
Figur« 2, MSL variations at Venice (thick lin« I,ll' the IH-yr running mean i and at Trn-stc-. The Trieste curve has be('n re-porte-d twice: close to till' oldor part of the Venice curve. at t.h« sanl(' scale. and displaced by about 10 em. in order to 1)(' closer to Ow most rece-nt part of tho Vr-nicc- curve.
to depress sea level about 2 cm , thus explaining the recent They are here updated and corrected fill' 3-cm subsidence apparent sea-level stability. since 1969 at DSL that was revealed by a comparison of' In Figure 3 the maximal annual sea-level heights re MSLs at the two stations (CANESTHI';LLI and PII{J\ZZOLI. 19991 corded in Venice from 1872 to 2000 are considered. Though and plotted in Figure 5, together with their 7-yr and 19-yr single values leave an impression of' dispersion, the 19-yr running means. The two series of extreme events are com running mean enables to distinguish significant evolution piled separately and are therefore statistically independent. trends. The overall increase of the maximal annual tide though in many cases they correspond to the same events. It levels from the end of the 19th century to the middle of can be seen that the differences decrease ii.c. surges PPIll' the 1970s has amounted to about :39 em, i.c. 14 em more trate more easily inside the lagoon) in the 19:30s (after the than the corresponding MSL rise (Figure 2 I. This differ creation of a new industrial harbor at Marghera) and in tho ence can be ascribed to the increased easiness with which 1950s (new deepening of navigation channels aftor t hc last the highest Adriatic storm surges could enter and propa world war), whereas they increase ii.c. the surge cn'st is par gate into the lagoon, following a deepening of the lagoon tially cut in the inlet area by the interruption of'muintennuro inlets and of the main navigation channels (for a summary dredging: around the second world-wartime period and sinc« of some of these works. see PIHAZZOLI, 1987 I. Since the mid the middle of the 1970s, after the removal of th,' oil traffic dle of the 1970s the situation has slightly improved, from the Lido inlet. On the whole, it can be said, in spite of' though the increase of extreme tide Il'vels (about :n crn i some missing data, that still in the 1960s, till' maximal tide remains much larger than tho MSL riso. levels were on average G ern higher in Venice than at D8L. Correlation between periods of deepening and of major while they are now some 3 ern higher at DSL. This changl' is works and the worsening of Hooding has been confirmed by probably due to the hydrodynamic resistance offered by sonu several authors. For example, this point can be verified in new sand banks, which have formed in some arpas of the Lido tuitively from the evolution of the differences in level between inlet, thus contributing by about 8 ern to lower surgp leve-ls the maximal annual tide levels recorded at Diga Sud Lido in Venice in relation to DSL. IDSLI. at the end of a jetty, just outside the lagoon, and the Recently, a mathematical model of the lagoon (U;VH:II·;SSI·:H. ones in Venice at Punta della Salute (PDS) (Figure 4\. These 19991 has demonstrated that the morphology of the inlets is had already been analysed previously (PIHAZZOLl, 1970, 1991). essential to control the propagation of storm surges inside t.he
.lournnl 01' (;oastal ]{esearch. Vol. Is. No, :l. 2002 540 Pirazzoli and Toma sin
Maximal annual tide heights at Venice (1872-2000)
200 ~------_. _----~
180
~ 160 E ~ E ::J ni "C ell (.J 140 c ell > t- en .....eo ell s: 120 -ell > 0 .c C1l .. .r:: 39cm -OJ ' Qj :I: 100 '·1 ..
80
60 - TTIlTl T1TTlI I TTTTTll rrr i rt t rnrrrrnn Tn TnrrTTTn n TTnfTTTTI T1TTTn mmTTnTn rTlH,1TTTTI I 1IT I nnTIITrn rt I rl IT ITTT N~N~N~Nt-N~N~N t-Nt-N~N~Nt-Nt-Nt ~~ rorom m O O ...... NN~~~ ~~~wm~t-roromm rororororo rom mmmmmmm mmmmmmmmmmmm Time (years)
[ ~X i m a l annual height - 7-yr run-;;ing mean --19-yr running meiflJ
Figure 3. Maximal annua l hei ghts record ed in Venice from 1872 to 2000 and running mean s over 7 and 19 years.
lagoon. It is therefore possible to estimate, from a compa rison SURGES betwee n Figu res 2 a nd a, that the morph ological cha nges which have affected the lagoon of Veni ce after 1872 have In spite of the recent loss in a ltimetric elevation, a spring cause d an increase in flooding levels, which is at pres ent still high tide is still usually un abl e to cause exte nsive flooding in about 8 em, but after a maximum of 14 cm in th e 1970s. Such Venice. A sea surge of meteorological origin will also add its increase is of th e same order of tha t ca use d by subsi de nce effects to the astronomical tid e. However , importan t surges du e to groundwater exploitation (about 10 em ) or to eustasy may occur also at low tid e and in this case their effects are and regio nal tectonics (about 15 em). much reduced. Nevertheless, at any tid e level, surges resul t
.Iournal of Coasta l Research, Vol. 18, No.3, 2002 Su rge Evolut ion in th e Adriatic 541
[WJ W ~ 4. CZJ -
'\ Malamocco
Adriatic S ea
o 2
Figure 4. The lagoon of Ven ice. ( 1) dike d oalli (fishe ries); (2) reclam ation areas dik ed for agricu ltur al use; (3) btirene (tidal flats) diked for ind ustrial use (part was later reopened to tid e); (4) urban ized are as, du mpin g grounds (since 1900); (5) barene open to tide; (6 ) je tties at the inlets; (7): artificia l cha nnel; (8) tid e-gauge sta tion; '1': oil terminal.
from a limited set of meteorologica l factors which are more Pu nta dell a Sa lute betw een March 1940 a nd December frequ ent tha n flooding events. Th e study of their evolution 2000, observed tide was iden ti cal to th e a stro nomica l tide may th erefore provide useful indications not only on pr esent (wit h c'; 1 cm accuracy) only for 3.4 % of' obse rvations and day climate cha nges bu t a lso on foreseeabl e evolution of "ac mi ssing da ta (in the se ns e t hat they a re not av ai la ble to qua alta " occur rences. It should also be noted th at if in th e the a ut hors) a re of th e same or de r. In 22 5,173 ca se s th e open sea surge amount a nd frequen cy depend mostl y on me tid e has been hi gh er t ha n the astronomic a l tide an d in teorological and hydro logical conditions , ins ide a lagoon they 269 ,324 cases lower, bu t in ab out 70% of' the ca ses t he de depend also on hydrodyn ami c resistance, es pecially in the in via ti on has been less th an :!:: 5 cm. The di stribut ion (at 5 lets, which , as discussed above, may slow th e tide propaga cm interval s) of a ll positive a nd negative surges is sum tion and contribute to lower th eir height. marized in Fig ure 6. For devia t ions sma lle r t ha n 10 cm, Tid e-gauge data used here are hourly record s digit ized a t nega tive surges a re more fr equ ent t han positive ones, Punta della Sa lute (1940-2000), Tri este ( 1939- 2000) and wh ereas for gr ea ter (a bsolu te) values, positi ve surges Diga Sud Lido (1968- 2000). In ord er to mak e possible a com clea rly pred ominate a nd may reach heigh ts more than parison of surges from different yea rs, inter-annual cha ngcs twice t hose of nega ti ve surges. For exa mp le, th ere are 9 in sea level (which may vary from one sta tion to a nother) occurrences of negative surges :5 - 65 ern , but 73 2 occur have been taken into account by subt racting th e local yearly rences with positive surges 2: + 65 ern , 8 of whi ch are MSL from th e data of eac h station. 2: + 130 cm . Meteorolo gical data includ e t hree-hou rly wind a nd air pressure observations rec orded at Trieste (1951- 1996), METEORO LOGICAL CAUSES OF SU RGES Ronchi (1967-1996), Tessera (Venice a irfield) (19 61-1996), Rim ini (195 1- 1996), Falconara (Ancona a irfield) (1960 It is well known that t he ma in meteorological factors in 199 6), Tcrmoli (195 1- 1996), Bari (1966- 1986) and Leu ca volved in coast al surge s are ai r pressure and wind . In mar (1951- 1996). Records of maximal dai ly wind gu sts in th e ginal basins, significant contributions can also be du e to rain same stati ons, as reportcd by WMO messages, hav e als o fall and river down-flow. Air pressure cha nges cause a been used . "static" response of th e sea, which reacts as an inverted ba Out of over ha lf a million hourly tide data av a ilabl e a t rom eter, with about 1 cm rise corresponding to a pressure
J ourn al of Coas tal Resear ch, Vol. 18, No.3, 2002 :i· l:'! Pil':lzzoli a nd Tumus iu
Differences in level between the highest annual tide levels recorded at Diga Sud Lido (DSL) and the ones at Venice (Punta della Salute), from 1924 to 2000 20 ; , " ~ " 15 " ! • , '
E 10 ~ ~ "iii • > .!!! 5 c: : til •... CIl ' .. • u 0 c: ...CIl .. CIl -5 •+ ; , == " ••'. i:5 " " -10 , • • .•"
-15 ' . • I 1 I 1 " 1 '1 ~ , r II :I 111 1 1 1 : ' ! I Ii :! I I 1 1 I I I 1 '11 1 11 r 1 . "t co N I'i gll n ' ,>. Em llliion ,;inl''' 19:'!·1 or di Hi'n'nl''' '; in " ,tn'lll" ,;"a lev el s hot woen t ho >'l'a I Il S !, 1a nd th « la goon I I' Il S I. d l'lT l' a ~ e of 1 h ra. S uch a cor re lation, genera lly well estab hPa (mea n world va lue at sea levcl ). th e sea surfa ce rises Iis hcd on t he open sea, becomes however approxima te near tsurge l a nd when it is > Hl1:3 hPa , it tends to fall (nega tive t lu- coast. owing to hydrodyn ami c effects depending on the surgeL cnas tu ] topogra phy, which may st reng t hen or attenua te the Wind is a lso a n importa nt param eter th a t ca use s pili ng "st atic" response . To simplify, when a ir pressu re is < 101:3 up of w at.er at t he coa st. lts effe ct s , whi ch added to t hose of th e pressure, dep end on t h ree param eters : t he direct ion from wh ich t he wind is blowin g (wh ich may pu sh the wa ter towards the coa st): its ve locit y (which a m plifies t he ef Semi-logarithmic distribution of surges greater fect s ): a nd it s la st.ing, which cont.rnls how long th e ph enom o r eq ual (smaller or equal) than a given height at Ven ice eno n will occu r a nd may ca use busi n-scule osci llations of g 1000000 sea le vel. ~ Th e im pact of a surge on t he coast will depend on th e tid e s o 100000 level at th e mom ent of th e su rge peak a nd impacts will be at -e '" a max imum for surges occu rring at spring high tide , whereas 10000 significa nt surges may escape notice when th ey occur at low er t ide levels. PI I(:l1 and V ASSII': ( 1979 ) ha ve shown th at surges 1000 and ti des are stut.is t.icu lly inde pendent. It is possib le to cal culate th eir pr obnhi lit.ics a nd tre nds se parate ly usin g hourly 100 ti de record s. Figure 7 show s t hat at th e time of th e high est surges at 10 PDS, atmosph eric pr essure is usu ully at a min imum over th e .0" E :> northern Adriatic. z ~~~~~~~~~o~~~ ~ ~ ~~ ~ ~ ~~ ~~~ ~~~~ Wind distributi on in th e Ven ice a rea. as deduced from the ~ ~ ~ ~ ~~~ ~ ' ~NM v ~ID~ oom~ ~~ ~ ~ ~~ ~ Tessera t hree-ho urly data, shows a hiuher frequency from the Surg e hei ght at Punta de lla Salute (em ) first (bora ), t he n from th e seco nd tsirocco: qu adran t (F igure Fig-un ' n. S('llli-logClr il h mir di st ribut ion of' all hou rly su rg l':-; l lH'gat i \,(' . S), It ca ll be observed that duri ng till' 24 hours preceding a ('q u;I! til / ( ' 1'11, ,111 (/ pn- iti v« : a t Von ic«, :-; in<:l' l ~ H () . surge of a t least 80 cm at rns ( 2 1 9 c a s e ~ obse rved. to which .luu ru a ] " I' ('oa ,;la l 1(,',;" a 1'('h . V,,1. IS. """ . :1. :,!()():,! Sur ge Evolu tion in the Adriatic Sea-level atmospheric pressure on the Adriatic A) for 40 hourly cases with surge >=100 em in Venice 1020 ------. 1015 ---- a.-CO .> /~ 1010 .- s: // ~ ~R~~_ = .-- M a x i ~ u mI -...Ql 1005 0 - :::s ~ D -< ~~~/. - D-- 3rd quartile CIl ~------______./ CIl 1000 ---+- Mean Ql -- ~ ~ ... - D .i-> Q. ~ -: . -- ... 995 ------M- 1st quartile ~ ---~---- - . - Minimum 990 ------w . 985 --- Trieste Tessera Rimini Falconara Termoli Sari Leuca B) for 219 hourly cases with surge >=80 em in Venice - 1020 ~ IJ 1015 - ~ - - Maximum n;a. 1010 - :S 1005 ~- - 11- 95th percentile ~ --- ~;~~'=- - Ia----- 3rd quartile :::s 1000 ~ 995 : ~--- -Mean sa-a 7/-""-- I - a - 1st quartile ~ 990 ______.I -====---== _____ j ... ------1-- - ~ - - 0 - 5th percentil e 985 o ~ ------+- Minimu m 980 ~ ------+- 975 - I I Trieste Tessera Rimini Falconara Termoli Bari Leuca C) for 2589 hourly cases with surge >50 em in Venice 1030 ~ 1025 • - - ~- .- .- 1020 CO - .- Maximum a. 1015 III-__ _-a------______11- s: 1010 a II - D - 95th percentil e Ql ---... ~= ~ ... 1005 .. .. A A .. ~• - A- 3rd quartile "'------:::s 1000 --- . ,- CIl - -Mean CIl 995 .------Ql -- • . ... .-.--- - a- 1st quart ile Q. 990 - . • o · ... 985 -+--- -+ ~ - 0- 5th percentil e 980 --- Minimu m 975 - -+------+------I 970 -- Trieste Tess era Rim ini Falconara Termoli Sari Leuca Figure 7. Atmo spheric pressure in various Adr iat ic stations at th e time of surges at PDS greater or equal to 100 em (AI, 80 em IE ), and 50 em !C I. J ourn a l of Coastal Research, Vol. 18, No. a. 2002 544 Pira zzoli and Tomasin Wind directions at Tessera for surges >=0.8 m at Venice Normal distribution (calms =35%) For surges >=0.8 m (calms = 4%) ~ 25 -,------, ~~ ~~ ~ 15 t "' j i:~ ~t~rMl"~1 ,~~ -;a+++H-1 j ' ~ l,.~~~~"~I~~r" rl r l III r i H-rl"" 10 50 90 130 170 210 250 290 330 .10 50 90 130 170 210 250 290 330 Degrees North Degrees North 3 hours before the surge (calms =2%) ~ ~efo ,e ~ 25 ...,------. 25 the surge (calms = 0%) I :2 20 III rqd ·•.J ... .•.. .. ~ 15 . . ~ ~ trml~ t_Jr.~J~ ~~ 1 I 1I I II" ~'~ ~,,""r l + " IIIIIIIIIIIII I 10 50 90 130 170 210 250 290 330 10 50 90 130 170 210 250 290 330 Degree s North Degrees North ~ 25 ~mh before the surge (calms = 2%) 24 h before the surge (calms =16%) s 20 1- ~ 15 t . ~l~ tJ m-~. t +~ ~~ Jml l 'IM ffl+++~~++ 1 1 i 111 II1 10 50 90 130 170 210 250 290 330 10 50 90 130 170 210 250 290 330 Degrees North Degrees North Figur e 8. Wind directions at Tessera : normal distribution, and at th e moment and in the hour s precedin g surge peak s 2 80 ern in Veni ce. correspond 97 occurrenc es with observed tid e ::::: 110 em ), wind produce differences in level of severa l decimeters, with falls from 2()0_70 ° angle (bora) is clearly predominant, wher eas in its northern part and rises in its southern part, near wind from SE (110°- 150°) only sta rts to be pr esent a few Chioggia (PIRAZZOLI, 1981; DAZZI et al., 1987; FERLAand Rus. ho urs before th e surge peak . In oth er Adri atic areas more to CONI, 1994; ZECCH ETTO et aI., 1997). th e S, sirocco is every whe re predominant (Figure 9) at th e In Figure 10, bora evolution in th e nor th ern Adri atic dur tim e of su rges in Veni ce. More loca lly, if observations during ing th e 48 hou rs pr eceding surges ::::: 80 CIll in Venice is de th e hours immediately preced ing the surges arc considered, scribed. At Tri este changes are limited and bora freque ncy pr evailing directions are from 110°- 220° at Rimini, 120°-180° (about 30% of th e cases) is less th an normal (45% of th e at Falconara , 130- 200° at 'I'ermoli, 120°-200° at Bari and cases). The situation is different at Ronchi and Tessera, 1:100- 180° a t Leu ca. For lesser surges th e wind distribution where bora frequ ency before surge s is gr eater than norm al rem ains essentially th e same, with however a slight leveling frequ en cy. Du ring the immediate hours before surges , how of th e peaks in the gra phics. ever, it tends to reduce rapidly and to be replaced by sirocco. Sirocco, and to a less degr ee bora , are determining factors Twelv e hours before a surge peak ::::: 80 ern at PDS, bora is for surge developm ent in th e Gulf of Venice. Bora, genera lly still blowin g a t Tessera in 75% of the cases. strong a nd per sistent, tends to disp lace northern Adriatic wa Genera lly less strong than bora , sirocco blows a long the tor towards th e lagoon of Venice with a fetch of about 110 lon gitudinal a xis of the Ad riatic, with a fetch which can be km. According to a numerica l model proposed by STRA VI SI grea te r t ha n 800 km . Sirocco te nds th erefore to be chan (19 77 ), a sta tionary wind of 10 mls from NE will cause a sea ne led between the Apennines and t he Dinaric Alps and to level fall of about 7 em at Tri este and a ri se of 5 cm out of crea te surges in all the Adriatic, with maxima l va lu es in th e Chioggia inlet. According to another mod el, proposed by t he Gulf of Venice, from wh ere no way out is possible. Ac 2 ORLIC et al. (1994 l, when bora has a stress of 0.5 N/m , dif cordi ng to th e model by ORLI C et al . (1994), sirocco with a fer enc es in level in the northern Adriatic increa se, with a rise st ress of 0.5 N/m 2 causes a su rge of about 30 em alo ng the of 12 centi meters outside th e Venice La goon . However, in th e Ven etian coa st. ab sen ce of channeling, su rges due to bora tend to move south Sirocco is pr edominant in all th e stations of the middle a nd wards along the Ita lian coast. Inside the lagoon of Ven ice, southe rn Adri atic du rin g the 48 hours pr eceding surges ::::: 80 bora , which blows followin g th e main ax is of th e ba sin , may em in Venice (Figure 11). At Tessera, however, it is only dur- J ourn al of Coas ta l Research, Vol. 18, No. 3, 2002 Surge Evolution in the Adri atic 515 Wind directions on the Adriatic coast for sea surges >=0.8 m at Venice Rimini (calms = 13%) ~ 27 -,------.------, o~2 1~~S"0%) ~ o '18 ...... ~ 18 ell ell III III ··· ······ ······· ···········1 ····· ······ ····· ····· . n 9 ... .. -g 9 j 0 11:- 1I- I ~ I-++H+ ' "# 0 +H-H-t-tit--JJ ~t-" t-UjIjli"'\- ~~-+-- 10 40 70 100 130 160 190 220 250 280 310 340 10 40 70 100 130 160 190 220 250 280 310 340 Degrees North Degree s North Falconara (calms = 5%) Termoli (calms = 0%) .2~ 27 t ': == , I, I, I.fuJJ 'I I II~~~=~ ", 1 1 J IHi I:~ t ;~~~= ~mt' ~"H=+HJ 10 40 70 100 130 160 190 220250 280 310 340 10 40 70 100 130 160 190 220 250 280 310 340 Degrees North Degree s North Bari (calms = 2%) ~ 27 Leuca (calms ) " 0%) I o f :~ tl l l: I II I I ". , . JIDi , I ~ 1 ~ 18 - , . , JiliJ~, . I . , . i ~ , . IIIIIII,,I,1 : ,,., I:,., II I,II IIII: I, I 10 40 70 100 130 160 190 220 250 280 310 340 10 40 70 100 130 160 190 220 250 280 310 340 Degrees North Degrees North Figure 9. Wind dire ction s at Triest e, Rimini, Falconara, Terrnoli, Bari and Leuca at the moment of surge pea ks ;:' SO cm in Veni ce. Bora evolution in the northern Adriatic during 48 hours preceding surges >=80 cm at Venice 90 80 +------ 70 III t: 60 · 0 ..l'I:l e 50 +------/ ------ell III • .Q 0 40 1 '0 ;F. --l!. g;-~ -- :~ j II L-__~---~---- , -- ~--- ,-- - -'--- -r -- No rma l -48 h -24 h -12 h -6 h -3 h At surge distribution peak >=80 cm 1__ Tri este 50°-90· -iii- Ronchi 20·-70 · --lr-Tessera 20·- 70· ' Figure 10. Evolution of bora wind in th e northern Adri atic before surge peak s ;:' SO cm in Venice. .Iourn al of Coastal Research, Vol. IS,No. :1 , 2002 546 Pir azzoli and Tomasin Sirocco evolution in the Adriatic area during 48 hours preceding surges >=80 cm at Venice 100 .,------ 90 - 80 - III 70 r:: 0 .~ 60 e ., 50 .cIII 0 40 0 -~ 30 20 .--- 1- 1: J No rma l -48 h -2 4 h -12 h -6 h -3 h At surge distr ibution pea k >=80 em ___ Leuca 130·-180· -llI-Bari 120°-200° - --A-T~m oli1300 -200 0 - 1 --*- Falconara 120°-180° __ Rimini 110°-220° Tessera 110°-150° Figur e 11. Evolution of sirocco wind at Adriatic stations before surge peaks :0> 80 em in Venice. ing the imm edi ate hours before the surge peak that sirocco half of th e 1950s, betw een 1964 an d 1967 a nd around 1979 replaces bora. Simulta neous blowin g of bora on th e northern and 1986; a nd next trough seems imm inent. Adri atic a nd sirocco on th e mid dle a nd southe rn Adriatic is Si nce the e nd of the 1960s the a verage surface pressure not rare (locally called scontraura, i.e. collision sit uation) a nd on t he northern Adriatic has increased about 2hPa, t he also the two wind-driven wave sys tems ma y superimpose "sta tic" effect of wh ich ha s been to lower th e MSL by about their set up a long the Venetian coast, incre asing the flooding 2 em, thus masking a n eus ta tic ri se of t he sa me amount. risk . Th e a ppa rent recent stability of se a level (Figure 1) de pends the re fore on the varia bility of atmospheric cir cu la RECENT EVOL UTION O F SEA-LEVEL AIR tion, whi ch is essentia lly unstable, a nd may be not lasting. PRESS URE AN DOF BORA AN D SIROCCO WIND S During t he second half of t he 20th century, bora a nd ot h er eas terlies hav e shown a strong decrease in bot h fre When all th e ava ilable digitized air pressu re record s are qu enc y a nd velocity a nd a similar trend may exist since conside red, a clear increasing trend appea rs during th e last th e beginning of in strumen tal observati ons (PIRAZZO Ll a nd decad es over all the Italian Adriatic coast (f igure 12), for Tm IAsIN, 1999 ), At Trieste, for exa mple , the freq uen cy of high values (e.g. the 95th percentile of all the values of each winds from 500 _900 has sha r ply decreased sinee 1~) 51 , in yea r), as well as fill' mean va lue s, or for low values (5th per number of three-hourly records, as well (to take into ac cent .i le )". On ly at Tcrm oli, for th e 5t h percen tile, th e increa s count mi ssin g da ta ) as in percen t of all th e records avail ing tre nd was not significa nt . The 5-year running mean s in a ble. In particular typical bora records from 700 ha ve a l dicate th at t he risin g trend include: oscill ations, with peak s most disappeared (from about 700 three-hourly observa at the end of the 1950s, in the first half of the 1970s a nd tions per yea r a t th e beginnin g of the 1950s, to less th an around 1991 ; troughs of oscilla tions are observed in th e first 100 observations in recen t yea rs )(Figure 1:3l. Such a de crease is not acco mpanied by a strengt he ning of winds I In the following, tr en ds are determ ined mainly as regr ession from oth er di rect.ions , but by more frequen t calm situa lines, the slopes of which are compa red to th e slope sta ndard devi tio ns (from 26% of tot al obser vations in t he 1950s, to 44'(, ation (c:: o ), in order to es timate the ir statis tical significance. As is durin g the last ten yea rs, and even to over 60% of ca lms a lre-ady known , 0 gives the likelihood of the rea l slope with 68% during the ea rly 1980s ). In addition, a ll wind velociti es- probability lor 1.96" gives 95'/' ; or 2.58" gives 99%1. Becau se a good confide nce is requ ired for a posi tive/negative trend, as soon as the slope turn s out to be positive/negative, a nd > " (or 1.96", or 2.58,, ), one can trust that a clea r tendency exists with 68'/((or 95%', or 99 ~k ) " In wind record s a t certain sta tions, sudden dro ps in the num ber probabi lity. According to criteria of tre nd detection used in Ire e of cal ms occur which probab ly depend on changes in instrumen ta 120011 est ima tions, it will be as su med here that with a confiden ce tion.Such disconti nu ities disappear however if lowest wind veloci > 9 ~) 7, a trend will be virtua lly certain, with 90- 997, very likely, 66 ties are not cons ide red. POI' th is reason, when a na lysing win d evo 90'/' likely. 33- 66'/' possi ble, 10- 337, un likely, 0.0 1 - 10 7~ very un lution in Lime, frequ encies and av erage velocit ies will often corre likely and ~ O . O l ' l( virtua lly impossible. spond in this pap er to wind velocities > 4 knots. J ourna l of Coastal Research, Vol. 18, No.3, 2002 Surge Evolution in th e Adriati c 547 Sea-level air pressure on the Adriatic area A: 95th three-hourly percentile (1951-1996) 1951 1956 1961 100 6 1 ~1 1976 198 1 1986 1991 1996 - .-Trieste: 5-yr running mean - .-Rimini : 5-yr runningmean - . - Termoli : 5-yr running mean - -- Leuca: 5-yr running mean ---Trieste, regr.= +0.075+/-0.026 •• • R,mini, regr.= +0.096+/-0 .027 - Termoli, regr. = +0.044+/-0.030 -- Leuca, regr.= +0.070+/-0.024 8: three-hou rly mean (1951-1996) - .-Trieste: 5-yr running mean - .-Rimini: 5-yr running mean - .-Termoli: 5-yr running mean - - - Leuca: 5-yr running mean ---Trieste, regr.= +0.038+/-0.013 - Rimini, regr.: +0.055+/-0 .011 - Termoli, regr.: +0.014+/-0.013 - • Leuca, regr.= +0.033+/-0.010_ c. 5th three-hourly percentile (1951 -1996) r-tr- 195 1 19 56 1961 1966 1971 1976 198 1 1986 199 1 1996 Years - .-Trieste: 5-yr running mean - . - Rimini: 5-yr running mean - .-Termoli: 5-yr running mean - - - Leuca: 5-yr running mean ---Trieste, re9r.= +0.032+/-0.018 •• - Rimini, regr.> +0.053+/-0.017 -- Termoli, regr.= +0.0 11+/-0.017 - • Leuca, regr.= +0.030+/-0.014 Figure 12. Sea-level air pressure variations at Tri este, Rirn ini, Terrnoli a nd Leu ca from 1951 to 1996: ru nn ing mean s over 5 years and linear rcgrossions for the 95t h percentile, th e mean , and the 5th percen tile of all th ree-hourly records. from 50 0 _90° sho w a t Trieste, with > 0.99 probability, a declining trends for the ea sterlics can be observed a lso a t clea r deceleration (Fig ure 14)'. Simila r, a lt hough weaker, Ron chi a nd Tessera ( P IRAZZOLl an d T OMASI N, 1999 ). Sirocco evolu tio n is not the sa me a ll over th e Ad ri atic, " All wind s peeds reported in t his pap er cor res pond to record s from coas tal stations. Offshore wind s may however be much st ronger: 80% st ronger in th e m idd le a nd nor th ern Ad riatic IAcn:HHo NI a nd about 40 0/, · stronger in the southe rn Adri atic ( NANIA, 1969 ), and even M ANCA. 1973). .Ioum al of Coastal Research, Vol. 18, No. :1, 2002 548 Pirazzoli and Tornasin Evolution of bora >4 knots at Trieste 800 700 III <: 0 += 600 .,e'" III .c 500 0,., 1: ~ 0 s: 400 Q, ~ :; .... 0 G; .c E ~ z 100 M r-, 0> M r-, 0> M ...... 0> M ...... 0> M :;:; ;0 CD CD CD CD t: ...... r-; r-, ...... ;;; eo co co eo C;; 0> 0> 0> 0>""' 0> 0> 0> 0> 0>""' 0> 0> 0> 0> 0>""' 0> 0> 0> 0> 0> ""'en en en 0> en ""'0> ""' ~ ~ ~ ~ ~ ~ ~ ~ ""' ""' ""' ~ Years 1 __700 __60° ---6--80° Polyn omial trend (70°) Po lynomial trend (60°) -- - Polynomial trend (80°) I Figure 13. Evolut ion of th e bora compone nts from 60°, 70° an d 800N at Tri este, for wind velocities > 4 knots (from PIAAZZO LI and TOMASIK. 1999), adapted. Evolution of wind velocities from 50°_90° N at Trieste 70 60 50 '2 ~ 40 ~ '13 0 Qj 30 > Years __Max. gust --95th perc. ma x. gust ---6-- Max 3-h _ 95th perc. 3-h ----lIf-- Mean 3-h - - Regr. max . blast ---Regr. 95% Max blast - . - . Reg r. 3-HMax ---Regr. 3-H 95% L-- R.!gr. 3-H Mean >=4kn Figure 14. Evolution of bora wind velocities from 50°_90. angles at Triest e: maxima l daily gu st , 95th perce ntile of maximal daily gusts, maximal three hourly measurement. 95th percentil e of th ree -hourly measurem ents and mean of three-hourly mea surements with velocity > 4 knots. -Iournal of Coastal Research, Vol. 18, No.3, 2002 Surge Evolution in the Adria tic 519 Tessera: wind >=4 knots from 110°.150° 350 -,------,..30 V) c: g 300 _. - /~.\ "§ 200 = 20 I o s: _------.- E ecD 150 .. -. -- .. .. 15 0 :5 a Q) .0 E 50 :::J Z 1965 1970 1975 1980 1985 1990 1995 Years --- Number of observations (N) - Regr. (N) = +2.50 +/- 0.97 - . - % of observations -_ . Regr. % =+0.173 +/- 0.075 Figure 15. Frequency 0961 to 1996) of wind :0: 4 knots from 110°_ 150° at Tesser a, Falconara: wind>=4 knots from 120-180° 350 ,------~ 20 18 . . ' ~ ' ~," . p ~1ft~ _ - 16 i300 1\ • .\ - 14 V) 0 \i1\j ~ / =t: ~ :;250 J 12 r····· /·w,%rfJ !ft-;y 1 ~ --- Number (N) of observations -8- Regr . (N) =+2.28+/-0.81 --- % of observations - Regr . % =+0.083 +/- 0.029 Figure 16. Frequency (1960-1 996) of wind ~ 4 knots from 1200 - 18(t at Falcon ara. J ournal of Coas ta l Resea rch, VoL 18, No. 3, 2002 fi50 Pi razzoli a nd Tomasin ~~~~~-~- -~ Termoli: wind >=4 knots from 130°-200° 450 ..,-~~~~~~~~~~~ --- - ~- ·22 '"c .Q iii 400 ~ 20 tI.l .a c '"o '"0 ~350 18 .~ OJ o c: s: tI.l til .a'" ~ 300 16 0 :5 ~ ~ )1 ~1 '0 /1:-:;r oJ! ;j!. 14 i 250 i -t-fI-\t ~-1'- -+~t-t--f- t _'1 _ 1_ ,_",-j 12 200 -+-- L I - ---r 1 ~ 1~ 1~ 1 ~ 1m 1 ~ 1~ 1~ 1 ~ Years - .- Number (N) of observations - - Regr. (N) =+1.98+/-0.70 - - % of observations _.- Regr. (%) = +0,062+/-0,029 Figu re 17. Freque ncy (1951- Hl9GJ of wind ~ 4 kn ots from 130°- 200° at Terrnoli. Leuca: wind >=4 knots from 130°-180° ~ 8 0 0 1 ~~ -- - -~--~~- 40 1700 ~ ...... 35 ~ - 30 §'" ! 600 -1\\ . ~ c: 25 ~ .o o ~: :~ ~j~-~. ,~ ,=~; ; 20 '0 !::: h i ;j!. Q; .tfl ~ /~L~j ~\ J . ~ 300 - • ,", . ~ ~.r- ' 15 z 200 - ----+-- I-- I-~ t--f--t-- ,--t~- , 10 1950 1960 1970 1980 1990 2000 Years - -~--~------Number (N) of observat ions - .- Regr. (N) = -4.18 +/- 0.93 - .- % of observations ~ - Regr. (%) = -0.24 +/- 0.04 1 Figure 18. Frequency (1951-1996 101'wind ~ 4 kn ots from 130°-180° at Leuca . •Journ al of Coastal Research. Vol. 18, No. 3. 2002 Surge Evolution in the Adr iatic 551 Wind name Station (see Fig. I) Frequency trend Velocity trend General trend and wind direction (xo=angle) Trieste (50°-90°) '- <, Bora Ronchi (20°-70°) -+ '- <, <, \ Tessera (20°-70°) Tessera (110°-150°) / -+ Rimini (110°-120°) -+ Sirocco Falconara (120°-180°) / -+ -: Terrnoli (130°-200°) -+ Bari (120-200°) -+ Leuca (130°-180°) -, -, LE GEND:~ falling ; -+ steady ; /increasing ; ..." insignificant trend, Figure 19. Summa ry of recent trends for winds related to surges 2 80 cm at Punt a della Sa lute . At Tessera (Figure 15), as well as in th e centra l Adri atic on th e other ha nd, the freque ncy of su rge -rela ted south a t Falc ona ra (Figure 16) a nd as far as a t Term oli (Figure ea sterlie s decreases sharply, a t Bari , as well as (with 17), sout h-eas terly winds that ca use surges in Venice > 0.99 pr obab ility) at Leu ca (Figure 18 ). Th e va rious ob show, with > 0.97 pr obabi lit y, a n in crease in frequenc y se rved t re nds a re summa rize d in Fi gure 19. t rend , in the total number as well as in the percen t of ob se rv ations", whereas t heir velocit ies do not show sta tis ti SU RGE EVOLUTION INT H E ADRI ATI C A1'1DIN cally significa nt t rend of cha nge . In t he sout he rn Adria ti c, T HE LAGO ON OF VEN ICE Since 1968, simultaneo us digitized record s of hou rly tide 4 Due to lack of hom ogeneity between wind and tide data, since a re a va ila ble from various ti da l stati ons: DSL (45°25 ' N certain wind data are mi ssi ng, graphs related to wind tren ds give 12°26 ' E ) immedi ately outside the lagoon of Venice, PD S perc en t values of obser vation as well as number of observations. Tid al data series, wh ich are almost compl et e, do not create th e sa me (45 °26 ' N- 12°20 ' E), 111 t he city of Ven ice, a nd Trieste difficulty. (45°39 ' N- 13°46' EJ, at the north-eastern bord er of t he Surge frequency at Diga Sud Lido (1968-2000) 3500 3000 /\ III t-; ...... -. E' 2500 ,,! :J III >- / ;: 2000 :J o •• • • / " . . Regr.=+9.27+/-3.42 .c: \ / ' / , .' \ .' . ~ 1500 j \'.../'«... ..-. < :'>. ; ~ . , ·/ I Regr.=+6.60+/-2.96 .c I · ' / \ ," l § 1000 +;/ "./\" ' ;\";.1,-" /\. /+~ . ' u"''';\ /.,* -+- Surges >=20 em - - Surges >=25 em - 0- Surges >=30 em - - .- Figur e 20. Frequency of surges 2 ;; to 2:30 Clll at Diga Sud Lido from 1968 to 2000. J ournal of Coastal Research, Vol. 18, No, 3. 2002 552 Pirazzoli and Tornasin Surge frequency at Punta della Salute (1968-2000) 3500 ,------l 3000 ... '" ~ 1000 li":o..-,-..,..:,.,=""...",,----:-cPd~==~~O:;::7T_t='7c''\::_:t~~71 z ~. '.-- o +- ,---r---,,-'--.-.-,--"r-T',-l-,-,------,----, ---'--- ~ I -r- ...... --r - '---'- T - f - 1968 1972 1976 1980 1984 1988 1992 1996 2000 Years ------.- Surges >=5 em - 0 - Surges >=10 em ---+- Surges >=20 em ~- Surg es >=25 em Figu re 21. Fr equency or sur ges 2:5 to 2:30 cm at Pun ta della Sa lute from 1968 to 2000. Adria tic Sea. In Veni ce, the ye a rs since 1968 are esp eciall y ha ve been a na lyzed a nd divid ed into cla sses at 5-cm inter in teresting to deduce na tural surge varia tions becau se va ls. Greate r su rges (2:35 cm) did not show statistica lly th ey are later th an th e la st man -induced main morpholog significa nt evolution trends, indicating that the clim ate ical cha nges in th e lagoon (dee pening of t he Ma larn occo variabi lity observed has not ye t reached a noticeable im inl et a nd dredging of the "Oil Cha nnel" across the la goon pa ct on maj or flooding eve nts. If how ever lower surges ( 2:5 in t he ea rly 1960s ). In a first a pproxima tion, t he t rends ern to 2:30 ern, whi ch a re most fr equen t) a re con sidered , obser ved since 1968 can therefor e be te ntative ly ascr ibed t rends a re more ala r ming. At DSL , the frequenc y of hourly to clim a te variab ility a lso in th e lagoon . surges 2:5 ern tends to in cr ease over 12 hours per yea r a nd At eac h station, a ll the surges for the period con sidered th at of surges 2::30 ern still 2 hours per year (Figure 20 ). Regr.= +6 03+/-2.44 Regr.=+6.63+/-2.79 Regr.=+5.54+/-3.12 Regr.=+ 4.54+/-2.73 Regr.=+ 3.77+/-2.33 500 Regr.= +2.32+/-1.85 •. - . .... '\ ;; '. - . / "'e - . I I ~~ o I j I T' I 1---.--1-Y r -r-r-r-r-r- 1968 1972 1976 1980 1984 1988 1992 1996 2000 Years • - Surges >=5 em - 0 - Surges >=10 em - .- Surges >= 15 em I - +--- Surges >=20 em - _.- Surges >=25 em - 0 - Surges >=3 0 em Figure 22. Fr equency or surges 2:5 to 2:30 em at Tr iest e from 1968 to 2000 . Journ al or Coas tal Resear ch. Vol. 18, No.3, 2002 Surge Evolution in the Adriatic This means that they are respectively over 400 hours and Venice) for reviewing an earlier version of this paper and very 60 hours more frequent today than in 1968, with a signif helpful suggestions; and to Ms. Jane Frankenfield (CNR icance level of increase greater than 99lk, for surges 2::5 and ISDGM, Venice) for help in editing. 2:: 10 ern and greater than 95(/(, for surges 2:: 15 cm. In the Gulf of Venice, the sirocco increase (favorable to surges) LITERATURE CITED seems therefore predominant in relation to the contrary effects produced by the bora abatement and the increase ACCERBONI, E. and MANCA, B., 1973. Storm surges forecasting in the in atmospheric pressure. Adriatic Sea by means of a two-dimensional hydrodynamical nu merical model. Bollettino di Geofisica Teorica ed Applicata, 15(57) Inside the lagoon the increasing surge trends are slightly 25-33. smaller in relation to DSL, probably due to the hydrody BAtT, D.; GAMBOLATI, G., and TEATINI, P., 2000. Residual land subsi namic resistance provided by the Lido inlet, which, as dis dence near abandoned gas fields raises concern over north Adriatic cussed above (Figure 5), has been increasing since the end coastland. Eos, 81(22) 245-249. of the 1970s. At Punta della Salute, surges 2::5 cm tend to BONDESAN, M.; CASTIGLIONI, G.B.; ELMI, C.; GABBIANELLI, G.; MAROC CO, R.; PIRAZZOLI, P.A., and TOMAsIN, A., 1995. Coastal areas at increase 9 hours per year, whereas the variability of surges risk from storm surges and sea-level rise in northeastern Italy. 2::30 em is less statistically significant than at the pier pro Journal 01' Coastal Research, 11(4) 1354-1379. jecting into the sea (Figure 21). Due to its geographical CANESTRELLI, P. and PIRAZZOLI, P.A., 1999. L'evoluzione delle dif situation, the Trieste station is relatively sheltered from ferenze tra i massimi livelli annuali a Diga Sud Lido ed a Punta direct set up from sirocco, though exposed to general surg della Salute. Comune di Venezia, Gruppo di lavoro di cui alla delibera 8 rnarzo 1999 del Comitate Lntermini.steriale ex. Art. 4 es produced by sirocco in the northern Adriatic. As con legge 798/84 (part of a Report circulated for the meeting of 20 cerns the bora decrease, it implies a decrease of negative July 1999),5 p. surges (i.e. an increase of the positive ones). The result is CARBOGNIN, L.; GATTO, P.; MOZZI, G.; GAMBOLATI, G., and RICCEIU, that at Trieste the frequency of surges 2::5 cm tends to in G., 1976. New trend in the subsidence of Venice. Publ. I.A.H.S. crease by 6 hours per year and the surges 2::30 em by over 121, 65-81. DAZZI, R.; ROSSI, G.; RUSCONI, A., and TOMASIN, A., 1987. Meteorologia 2 hours per year (Figure 22 L e laguna: l'ammonimento delle pili recenti bufere per I'efficacia degli interventi di difesa. Rapporti e Studt, vol. 10, Istituto Veneto CONCLUSIONS di Scienze, Lettere ed Arti , Venezia, pp. 103-127. DORIGO, L., 1961. Le osservazioni mareografiche in laguna di Vene A decrease in the northern Adriatic Sea area of bora fre zia. Rapporti preliminari, vol. 1, Istituto Veneto di Scienze, Let quency and velocity means less-frequent invasions of arctic tere ed Arti, Venezia, pp. 11-38. cold air over central/eastern Europe, and less-frequent strong FERLA, M. and RUSCONI, A., 1994. L'evento di "acqua alta" 921208 in laguna di Venezia-Indagine sui sovralzi differenziati. XXIV Con atmospheric pressure gradients between the highs usually oegno Idraulica e Costrueioni Idrouliche. Napoli, pp. 1-16. accompanying such polar drifts, and air pressure lows devel HOUGHTON, J.T.; MEIRA FILHO, L.G.; CALLANDEl{, B.A.; HARRIS, N.; oping in the central Mediterranean area (PIRAZZOLI and To KATTENBERG, A., and MASKELL, K., 1996. Climate Change 1995- MASIN, 1999), This reduction may be correlated with the grad The Science 01' Climate Change. Cambridge University Press. ual rise in the global surface air temperature, which has been IPCC, 2001. Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Inter (e.g. recently documented not only in the Adriatic area STRAV governmental Panel on Climate Change (HoucHToN, ~J.T. et al., lSI, 1987), but also on a global scale (HOUGHTON et al., 1996 ; EditorsL Cambridge University Press. IPCC, 2001) during the same period. MOSETTI, F., 1961. Sulla tendenza secolare del livello medio marino Knowing that in the northern hemisphere winds usually a Trieste. Atti Istituto Veneto di Scierize, Lettere ed Arti, 119, 425 turn anti-clockwise around depression centers, the sirocco de 434. NANIA, A., 1969. 11 vento e l'attrito esterno nelle burrasche adriati crease in the southern Adriatic, accompanied by an increase che. Analisi statistics. Rivista di Meteorologia Aeronauiica. 29, 4. in the central and northern Adriatic and by the bora decline ORLI(:, M.; KUZMI(;, M., and PASARI(:, Z., 1994. Response of the Adri can be partly explained by a slight northward displacement atic Sea to bora and sirocco forcing. Continental Shell Research.. of storm trajectories in the Adriatic area. This suggests that 14,91-116. the recent increase in Adriatic moderate surges has a climatic PIRAZZOLI, P.A., 1970. Valori estrerni di maroa nella laguna di Ve nezia. Tecnica Italiana, 35(9), 4:3:3-466. origin, possibly related to the recent global warming (IPCC, PIRAZZOLI, P.A., 1981. Bora e acqua alta. Acqua & Aria, 10, 1115- 2001), and that, if this conclusion is correct, it may be ex 1118. pected to worsen and become less moderate in the near fu PIRAZZOLI, P.A., 1982. Maree estreme a Venezia (periodo 1872-19R11. ture, also disregarding the possible effects of a near-future Acqua & Aria, 10, 1023-1039. PIRAZZOLI, P.A., 1987. Recent sea-level changes and related engi sea-level rise of climatic origin, which would be especially of neering problems in the lagoon of Venice (Italy). Progress in concern in the Venice area. Oceanography, 18, 323-346. PIHAZZOLI, P.A., 1991. Possible defenses against a sea-level rise in ACKNOWLEDGEMENTS the Venice area, Italy. Journal 01' Coastal Research, 7(1) 231-248. PIRAZZOLI, P.A. and TOMASIN, A., 1999. Recent abatement of easterly The authors are grateful to the Servizio Meteorologico winds in the northern Adriatic. International -lournal o] Climatol dell'Aeronautica Militare of Italy, for providing weather ar ogy, 19, 1205-1219. chives; to the Ufficio Idrografico e Mareografico di Venezia, PUGH, D.T. and VASSIE, J.M., 1979. Extreme sea levels from tide and surge probability. Proc. 16th Coastal Engineering and CO!1 the Centro Maree of the Comune di Venezia and the CNR [erence (Hamburg: American Society Civil Engineering); 1,911 ITT Institute of Trieste for tidal data; to A. Marani (Univ 930. ersita Ca' Foscari, Venice) and S. Zecchetto (CNR-ISDGM, TSIMPLIS, M.N. and BAKER, T.F., 2000. Sea level drop in the Medi- Journal of Coastal Research, Vol. 18, No.3, 2002 Pirazzoli a nd Tumus in It'I'l':II11':1I1 Sl'a : a ll ind icnt ur 01' dt ,t,p wu u- r s u lin it.v a nd tom poru l h ll ;II ' : ~SI ' :I L ( ;.. I ~)~H Va lut a ziom: dq(li l'f],'ui dl' gli inu-rventi morbidi l' 1 111"< ' c ha llgl' ~ . ( ; / 'ol )II ." ~ i/'(/ 1 U,'s/'url'lI l.cttcrs , 271 12 1 17:l1 -1 7:14 . dilfu si sull« riduzion« .Iourunl or (' o:" la l 1("';l':\I'(·I1 . V,,1. IH. No. :\. 21111 2