Sed i men tary ev i dence of ex treme storm surge or tsu nami events in the south ern Bal tic Sea (Rogowo area, NW Po land)
Andrzej PIOTROWSKI1, *, Witold SZCZUCIÑSKI2, Pawe³ SYDOR1, Bartosz KOTRYS1, Monika RZODKIEWICZ3 and Jarmila KRZYMIÑSKA4
1 Pol ish Geo log i cal In sti tute – Na tional Re search In sti tute, Pom er a nian Branch, Wieniawskiego 20, 71-130 Szczecin, Po land
2 Adam Mickiewicz Uni ver sity in Poznañ, In sti tute of Ge ol ogy, Bogumi³a Krygowskiego 12, 61-680 Poznañ, Po land
3 Adam Mickiewicz Uni ver sity in Poznañ, In sti tute of Geoecology and Geoinformation, Bogumi³a Krygowskiego 10, Poznañ 61-680, Po land
4 Pol ish Geo log i cal In sti tute – Na tional Re search In sti tute, Ma rine Ge ol ogy Branch, Koœcierska 5, 80-328 Gdañsk, Po land
Piotrowski, A., Szczuciñski, W., Sydor, P., Kotrys, B., Rzodkiewicz, M., Krzymiñska, J., 2017. Sed i men tary ev i dence of ex - treme storm surge or tsu nami events in the south ern Bal tic Sea (Rogowo area, NW Po land). Geo log i cal Quar terly, 61 (4):
973–986, doi: 10.7306/gq.1385
The Bal tic Sea is not typ i cally con sid ered as an area af fected by tsu na mis. How ever, dur ing the Late Pleis to cene and Ho lo cene sev eral tsu nami events have been in ter preted from the sed i men tary re cord, mainly in Swe den and Es to nia. Fur ther more, on the south ern coast of the Bal tic Sea, there are his tor i cal ac counts of catastrophical ma rine floodings called “der Seebär” (“the Sea Bear”). Their de scrip tions re veal many fea tures typ i cal for tsu nami, but their gen e sis re mained un known and sed i men tary ev i dence for such events has not been found. Here we pro vide ev i dence of sandy event lay ers from the area of Rogowo, NW Po land – the area of his tor i cal cat a strophic storms as well as “der Seebär” events. The study area is a low-ly ing coastal plain with an av er age el e va tion of –0.5 to +0.5 m a.s.l., pro tected from the open sea by beach and coastal dune sys tems up to 5 m high. Sedimentological, micropalaeontological and geo chem i cal anal y ses along with AMS 14C dat ing were ap plied to sed i men - tary suc ces sions seen in 5 ma jor trenches and 198 sed i ment cores up to 1.5 m long. Two sandy lay ers were iden ti fied in the peat de pos its that de vel oped on the plain dur ing the last ~2000 years. They re veal a num ber of typ i cal fea tures of tsu nami de - pos its (sig nif i cant lat eral ex tent and thick ness, rip-up clasts, chem i cal and micropalaeontological ev i dence of ma rine or i gin), how ever, 14C dat ing along with the his tor i cal ac counts re vealed that the ma jor layer, ex tend ing at least 1.2 km from the mod ern coasts, was prob a bly de pos ited by ar gu ably the larg est storm surge dur ing the last 2000 years, which took place in 1497 AD.
These storm de pos its were likely formed dur ing in un da tion of the low-ly ing coastal plain af ter ma jor breach ing of coastal dunes re sult ing in tsu nami – like flow pat tern and thus sim i lar sedimentological ef fects. A dis con tin u ous sand layer of youn ger age (18th cen tury) and shar ing sim i lar prop er ties to the pre vi ous one may be re lated to “der Seebär” event or an other storm surge.
The study re vealed that the south ern Bal tic Sea coast may be af fected by much greater coastal flood ing than known from more re cent ac counts and ob ser va tions. Thus, the pre sented geo log i cal re cord should be taken as an ex am ple of a worst-case sce - nario in coastal zone risk as sess ment from nat u ral haz ards. These events left sed i men tary de pos its that re sem ble tsu nami de - pos its. It is likely that, in sim i lar set tings where storm surges cause uni di rec tional in un da tion of a coastal plain, it may not be pos si ble to es tab lish whether the re sult ing de pos its were laid down from storms or tsu na mis.
Key words: tsu nami de pos its, storm surge de pos its, grain size anal y sis, geo chem is try, ra dio car bon dat ing, Bal tic Sea.
INTRODUCTION
Cat a strophic ma rine floodings (storms, hur ri canes, tsu na - mis) may cause dra matic di sas ters as ex em pli fied by a num ber of re cent cat a strophic events (e.g., the 2004 In dian Ocean tsu - nami; Hur ri cane Catherina in 2005; the 2011 Tohoku-oki tsu - nami). Many of these ap peared to be un ex pected since no his -
tor i cal re cords ex isted of events of sim i lar size, or ex ist ing re - cords were of lim ited ac cu racy and were not taken into ac count in coastal haz ard as sess ments. Study of palaeotsunami and palaeostorm de pos its in coastal sed i men tary ar chives may pro - vide cru cial in for ma tion on the oc cur rence and min i mum mag ni - tude of such events in the past (e.g., Jankaew et al., 2008; Goto et al., 2010, 2014).
How ever, at the same time it was found that both tsu nami and storm de pos its are source-de pend ent and ex hibit large vari abil ity and their dis tinc tion may be dif fi cult. They can be iden ti fied us ing mul ti ple proxy cri te ria in ter preted in the lo cal con text of sed i men tary fa cies. Some di ag nos tic cri te ria have been de vel oped by study ing re cent and past cat a strophic events (e.g., Mor ton et al., 2007; Shiki et al., 2008; Bour geois,
* Corresponding author, e-mail: andrzej.piotrowski@pgi.gov.pl Received: December 19, 2017; accepted August 30, 2017; first published online: September 26, 2017
2009; Chagué-Goff, 2010; Pe ters and Jaffe, 2010; Chagué - -Goff et al., 2011, 2017; Goff et al., 2012; Szczuciñski et al., 2012b). How ever, there is no easy way to dis crim i nate tsu nami from storm de pos its (Andrade, 1992; Nanayama et al., 2000;
Goff et al., 2004; Luque et al., 2004; Tuttle et al., 2004;
Dominay-Howes et al., 2006; Kortekass and Dawson, 2007;
Mor ton et al., 2007; Switzer, 2008; Switzer and Jones, 2008;
Lario et al., 2010).
Al though the low-seis mic ity Bal tic Sea set ting is not con sid - ered to be typ i cal for tsu nami gen er a tion, re ports from the north ern shores (Mörner, 1995, 1996, 1999, 2003, 2008, 2013;
Nikonov, 2004; Mörner and Dawson, 2011) pro vide in for ma tion on a num ber of the past tsu nami-like events. Re cently, Rotnicki et al. (2016) dis cussed the pos si ble tsu nami or i gin of an event layer found in the mid dle part of south ern Bal tic Sea coast line, which was dated at ~6 ka, while re cent re ports from Fin land pro vide in for ma tion on “meteotsunami” phe nom ena (Pellikka et al., 2014). Atyp i cal ma rine in un da tions of un spec i fied or i gin have been also known from the west ern part of the Bal tic Sea and the North Sea (e.g., Credner, 1889). The lat ter is re lated to his tor i cal ac counts of “der Seebär” (“the Sea Bear”) (Piotrowski, 2007a, b). More over, storms in the Bal tic Sea have been dis - cussed by many au thors (Majewski, 1986, 1989, 1998a, b;
Stigge, 1994; Dziadziuszko and Jednora³, 1996; Hupfer et al., 2003; Suursaar et al., 2003, 2006; Averkiev and Klevannyy, 2007, 2010; Gurwell, 2008; Jensen and Müller-Navarra, 2008;
Rosenhagen and Bork, 2009; Wiœniewski and Wolski, 2009;
Rich ter et al., 2012; Wolski et al., 2014). In most of the ac counts storm de pos its were de scribed as washover fans with a lat eral range lim ited to a few tens of metres (e.g., Rudowski and Wróblewski, 2000). How ever, some his tor i cal ac counts de scri - bed much more ex ten sive in un da tions.
The pres ent study fo cuses on geo log i cal ev i dences of tsu - nami-like events and storms on a low-ly ing (at sea level) coastal plain pro tected by a low coastal dune belt in the Rogowo area, south ern Bal tic Sea coast (Fig. 1). The ma jor goals are to ver ify his tor i cal ac counts of ex traor di nary cat a strophic floodings in the past to pro vide better coastal haz ard as sess ment, as well as to con trib ute to dis cus sion of the sedimentological char ac ter is tics of tsu nami and storm de pos its.
STUDY AREA
The Bal tic Sea is a shal low (55 m on av er age) semi-en - closed mar ginal sea. It is com posed of sev eral sills and bas ins.
Con se quently, wa ter ex change be tween var i ous parts, and be - tween the Bal tic and the North Sea, is lim ited. This makes the Bal tic Sea the larg est brack ish-wa ter ba sin in the world. It is con sid ered a non-tidal sea, since the am pli tude of tides rarely ex ceeds a few centi metres (Medvedev et al., 2013). Cur rently the Bal tic Sea drain age area is pop u lated by ~85 mil lion peo ple from 14 coun tries and the ex ploi ta tion of Bal tic Sea re sources causes many con flicts be tween socio-eco nomic in ter ests and the unique eco log i cal en vi ron ment.
The Pol ish coast is lo cated in the south ern Bal tic and has a to tal length of ~500 km. It is built of Qua ter nary gla cial and glaciofluvial de pos its, and mod ern coastal de pos its. Ap prox i - mately 80% of the coast line is com posed of sandy beaches and dunes, while the rest is a cliff coast. Most of the coast line is not af fected by en gi neer ing struc tures. Many low-ly ing parts of the shore line are ex posed to flood ing caused by storm surges, which re sult from a com bi na tion of high wind ve loc ity, deep low-pres sure sys tems and thus a higher water level in the sea.
In the study area (Fig. 1) the mor phol ogy is dom i nated by a large flat val ley (the Old Rega River Val ley) formed by gla cial meltwaters func tion ing at the end of the Late Vistulian (Pleis to - cene). The val ley is filled with Ho lo cene de pos its (mostly peat and gyttja). Be low the layer of peat are grey me dium-grained sands and dark grey sands with or ganic mat ter of flu vial or i gin (Dobracka, 1990, 1992; Cedro, 2012). The Old Rega River Val - ley floor lies at a height of 0.2–0.5 m a.s.l. The val ley is sep a - rated from the sea by a belt of coastal dunes (with heights of up to 5 m in the east ern part and up to 17 m in the west ern part) and also by in land dunes in the west ern part at heights of up to 31 m. The dune belt pro tects the low-ly ing plain from ma rine flood ing. The rate of coastal re treat in this re gion is 0.5–0.8 m/y.
The width of the beach in the study area is 40–50 m (Dobracki and Zachowicz, 1997).
At the mouth of mod ern Rega River ma rine in flows are ob - served dur ing storm surges. The cur rent Rega mouth was ar ti fi - cially formed in 1457 AD (Riemann, 1873; Stoewer, 1897).
Along with this in vest ment an em bank ment road was made con nect ing the new port and fish ing vil lage of Treptowschen Tip with Trzebiatów, a city lo cated 10 km south of the sea.
His tor i cal ac counts re port at least two cat a strophic ma rine in un da tion events in this area. The first event oc curred along most of the south ern Bal tic Sea coast and took place on 17th of Sep tem ber 1497. It caused breach ing of the Resko Przy - morskie Lake bar rier (Fig. 1), as well as coastal flood ing re - ported in ar eas a few tens of km eastwards to reach up to 4 km in land. Kaliningrad (Königsberg) and Gdañsk (Danzig) were flooded, more over Dar³owo (Rügenwalde), Szczecin (Stettin), Ko³obrzeg (Kolberg), Stralsund, Wismar and other coastal towns at the Bal tic Sea were at least partly dam aged (Riemann, 1873; Stoewer, 1897).
The sec ond event oc curred on the 3th April 1757.
Brüggemann (1779) re ported it as fol lows: “That hap pened on the 3rd of April, 1757 about noon, in calm and bright weather, the Bal tic coast near Trzebiatów on Rega River sud denly was roll ing so much that a ship, moored in the har bor, was snatched away by high waves and shifted far on to the land. Af ter that (roll ing) was re peated three times, the sea be came calm again.
Lo cal peo ple call this phe nom e non the Seebär”. Trzebiatów is a town lo cated 10 km away from the Bal tic coast. The de scrip tion of this event from the mid-18th cen tury is very sim i lar to mod ern re ports of tsu na mis of earth quake or land slide or i gin.
MATERIALS AND METHODS
FIELDWORK
Field work fol lowed de tailed anal y sis of ex ist ing data, in clud - ing his tor i cal maps and notes. The key study sites iden ti fied (Fig. 1) were vis ited sev eral times. The pres ent study in cluded doc u men ta tion based on 4 ma jor trenches and 198 bore holes up to 1.5 m deep (Fig. 1). Sam ples for fur ther anal y ses were col lected from trenches and bore holes, as well as from the mod ern beach and dunes.
LABORATORY ANALYSIS
Grain size anal y sis. Al to gether 115 sam ples taken from the mod ern beach and dunes, as well as from event de pos its col lected in transects A, B and C (Fig. 1); they were dried and sieved on a Fritsch Vi bra tory Sieve Shaker Analysette 3 PRO.
dekram setis gn iroc dna sehcnert fo snoi ta col htiw aera yduts eht fo yh pa rgo poT .1 .giF ;er ugifeht no deman era )ezis niarg dna yr ts imeh coeg ,smo t aid ,gn itad( si s ylana yro ta r obal ra l uci trap rof del pmas erew hcihw setis ylno ;’C–C ,’B–B ,’A–A snoi tces-ssorc etoN aera aeS ci tlaB eht nihtiw dna noi ger eht ni etis yduts eht fo noi ta col swohs pam te sni eht ;atad RADIL morfsi pam ecruos eht
The mesh di am e ter in ter val of the sieves was 0.25 phi. Sam - ples from trench no. 53 – ded i cated to geo chem i cal anal y ses – were ana lysed us ing an op ti cal diffractometry method on a la - ser dif frac tion-based Mastersizer 2000 Par ti cle An a lyzer. Sta - tis ti cal pa ram e ters of grain size dis tri bu tions (Folk and Ward, 1957) were cal cu lated for each sam ple us ing the GRADISTAT soft ware (Blott and Pye, 2001).
Geo chem is try. The 17 sam ples se lected of beach and dune sands, sandy event de pos its, peat and soils were ana - lysed for loss of ig ni tion (LOI), ma jor el e ments (SiO2, TiO2, Al2O3, Fe2O3, MnO, MgO, CaO, Na2O, K2O, P2O5, SO3, Cl and F) and trace el e ments (Ba, Cd, Co, Cr, Cu, Mo, Ni, Pb, Sn, Sr, V, Zn). The mea sure ments were con ducted at the Cen tral Chem i cal Lab o ra tory of the Pol ish Geo log i cal In sti tute – Na - tional Re search In sti tute in War saw (Po land). Ma jor el e ments were mea sured by XRF and trace el e ments by ICP-OES af ter aqua regia di ges tion.
Micropalaeontological anal y sis. Nine teen sam ples for di - a tom, mol lusc and ostracod anal y ses were col lected from the low er most parts of the sandy event lay ers in se lected sed i ment cores and trenches. The di a tom sam ple prep a ra tion and anal y - sis fol lowed the stan dard method out lined by Battarbee (1986).
In short, sed i ment sam ples were treated with 10% hy dro chlo ric acid (HCl), 30% hy dro gen per ox ide (H2O2) and heated to 170°C in or der to re move car bon ates and or ganic mat ter. The mi cro - scopic slides mounted with Naphrax were ana lysed with a Zeiss Axio Scope A1 mi cro scope at a mag ni fi ca tion of 1000x. Up to 300 di a tom valves were iden ti fied and counted per sam ple. The di a tom spe cies were clas si fied to eco log i cal groups us ing OMNIDIA soft ware (Ver sion 4.2) (Lecointe et al., 1993, 1999), and then the re sult ing groups were dis tin guished ac cord ing to Kolbe clas si fi ca tion (see Denys, 1991; van Dam et al., 1994).
For ostracod and mol lusc anal y ses, the sed i ment was treated with 30% H2O2 to di gest all the or ganic mat ter, and then washed with wa ter on a 0.1 mm sieve. A Nikon bin oc u lar mi cro - scope was used to iden tify the spe cies pres ent in each of the sam ples.
Ra dio car bon dat ing. The chro nol ogy of the sed i ment sam ples stud ied and the ap prox i mate age of the event lay ers
were based on high-pre ci sion AMS 14C dat ing per formed on bulk or ganic mat ter from soils and peat. The 26 sam ples were se lected and af ter care ful re moval of fresh rootlets they were mea sured in the Poznañ Ra dio car bon Lab o ra tory (Po land) us - ing the 1.5 SDH – Pelletron Model “Com pact Car bon AMS”.
The dates were con verted into cal i brated ages us ing the cal i - bra tion pro gram CALIB Rev 7.1 (Stuvier et al., 2016) and the IntCal13 cal i bra tion dataset (Reimer et al., 2013). Sev eral sam - ples con tained mod ern, post-bomb car bon in di cat ing a post-1960 age. To con strain the age of the sam ples, they were cal i brated us ing the CALIBomb Ra dio car bon Cal i bra tion pro - gram and the north ern hemi sphere 14C datasets (Hua et al., 2013). The mod ern sam ples are pre sented in per cent mod ern car bon (pMC). The cal i brated re sults are re ported with a two stan dard er ror age range.
RESULTS
GENERAL SEDIMENT DESCRIPTION
The study fo cused on an area ~600 m to >1.4 km from the coast, as the area closer to the coast is cov ered by dunes and the area fur ther in land is strongly af fected by ame lio ra tion works and ag ri cul ture. The coastal plain stud ied is cov ered by peaty soils, with a rel a tively high sand con tent, which is in ter ca lated with one con tin u ous sandy event layer, called here layer A (Figs. 2–4). More over at least one dis con tin u ous sandy layer was doc u mented at in di vid ual sites above layer A, e.g. in trench no. 55 (Fig. 3). The max i mum lat eral ex tent of the most ex ten - sive layer, A, is >600 m, reach ing a dis tance of 1.2 km from the mod ern coast (Figs. 2 and 4). The thick ness of this event layer var ies be tween 0.5 and 50 cm (Fig. 4). The larg est vari a tions in the event layer thick ness were ob served at a dis tance of 850 to 1200 m from the sea shore (Fig. 4). The bound ary of the sandy event layer A with the un der ly ing peat is sharp and ero sional.
The struc ture is mas sive; no in ter nal mac ro scopic strat i fi ca tion was noted within the layer, and it con tains root traces. The layer
Fig. 2. An ap prox i mately 380 m long cross-sec tion A–A’ (see Fig. 1 for lo ca tion), lo cated next to the dated trenches and con sid ered to be rep re sen ta tive of the study area
The up per part pres ents a sim pli fied geo log i cal cross-sec tion; the num bers above re fer to cor ing site num ber; note that the ver ti cal scale is ex ag ger ated; the lower panel shows grain size dis tri bu tions of bulk sam ples taken from the ma jor sandy event layer (A)
in cludes frag ments of peat in the form of rip-up clasts (Fig. 3).
The fol low ing grain size, geo chem i cal and di a tom anal y ses fo - cused on this layer.
Event layer B was found in in di vid ual trench and sev eral sed i ment cores (e.g., Figs. 2 and 3). It shared sim i lar mac ro - scopic char ac ter is tics as layer A, dif fer ing in the lim ited spa tial dis tri bu tion and smaller thick ness. It was found to be <10 cm thick. It can not be ex cluded that lay ers A and B are amal gam - ated in some places.
Event layer C was found only in trench 28 (Fig. 3) and pi lot trenches lo cated up to a few metres away. It is also com posed of a dis con tin u ous layer of well-sorted sand up to 3 cm thick with rip-up clasts of the un der ly ing clayey silt de pos its. The lat ter are char ac ter ized by the com mon pres ence of iron ox ides.
GRAIN SIZE ANALYSIS
Event layer A is com posed mostly of fine-grained sand (90% of the sam ples ana lysed; Figs. 4–6). Me dium-grained sand oc curred in 9% of the sam ples ana lysed, and coarse sand
in 1% of the sam ples. The stan dard de vi a tion in most sam ples (78%) shows them to be well-sorted de pos its, al though a slight bimodality is com mon in most of the sam ples with ma jor modes at 2.0 and 2.5 phi (Figs. 2 and 5). Very well-sorted de pos its com prise 11% of sam ples and mod er ately sorted de pos its com prise 10% of sam ples. 79% of sam ples have val ues typ i cal for very neg a tive-skewed dis tri bu tions and in 11% they are neg - a tive-skewed. In 7% of sam ples the dis tri bu tions are sym met ri - cal and in 2% they are pos i tively skewed. Kurtosis in 82% of sam ples shows mesokurtic dis tri bu tion curves. In 17% of sam - ples kurtosis was platykurtic and in 1% of sam ples it was leptokurtic.
The grain size char ac ter is tics of event layer A does not re - veal clear trends in land ward di rec tion (Fig. 4), or be tween par - tic u lar cross-sec tions (Figs. 4–6). Ver ti cal vari a tions within the event layer are small, with a slight fin ing-up wards trend (Fig. 7).
Grain size anal y sis re vealed that the event layer de pos its are very sim i lar to dune and beach de pos its in terms of grain size dis tri bu tion and sta tis tics (Figs. 4–6), in clud ing the same pat tern of biomodal dis tri bu tion (Fig. 5). The event layer de pos - its re vealed grain size dis tri bu tions and sta tis tics in be tween the Fig. 3. Pho to graphs of the event lay ers stud ied in trenches no. 28, 53, 54, 55 and their geo chron ol ogy
A – ex am ples of strati graphic suc ces sions in trenches with iden ti fied sandy event lay ers A (trench 53, 54 and lower sandy layer in trench 55), B (up per sandy layer in trench 55) and C (trench 28); the sam pling lo ca tions and 2 sigma-cal i brated 14C ages are in cluded; B – over view of all the cal i brated 14C dat ing re sults for sites 53, 54 and 55 (trench and cor ing sites next to them) from sam ples brack et ing the ma jor event layer (A) and youn ger layer (B), see Ap pen dix 1* for de tail re sults; see Fig ure 1 for trench lo ca tions
* Supplementary data associated with this article can be found, in the online version, at doi: 10.7306/gq.1385
typ i cal val ues for mod ern beach and dune de pos its in the study area. The mod ern beach de pos its are gen er ally slightly coarser and more poorly sorted than the event de pos its. The dune de - pos its are very uni form. They are char ac ter ized by slightly finer mean grain size and better sort ing than the event and beach sed i ments.
GEOCHEMISTRY
The geo chem i cal com po si tion of beach, dune, soil and event de pos its is sum ma rized in Ta ble 1 and in Fig ure 7. The con cen tra tion ranges of the ma jor and trace el e ments ana lysed in the event de pos its and in the beach sed i ments over lap. This is par tic u - larly the case for po ten tial sea wa ter in di ca tors such as so dium and chlo rine. How ever, the lat ter are only very slightly above the lim its de tected in dune sand and soils. It must be stressed that in the case of dune sands the geo chem i cal com po si tion is also very sim i - lar, how ever, the soils and peats sandwiching the event layer are quite dif fer ent. The event layer de pos - its are richer in SiO2 and Cl than the peat and soil sam ples, while they are poorer in most of the re main - ing ma jor and trace el e ments ana lysed (Ta ble 1).
The ver ti cal vari abil ity of geo chem i cal com po si tion in event de posit A is small (Fig. 7).
MICROPALAEONTOLOGICAL ANALYSIS
All the sam ples taken from sandy layer A re veal very sim i lar di a tom com po si tion (Fig. 8), and con - tained up to 56 taxa per sam ple. Most of the spe cies rep re sent a fresh wa ter hab i tat. How ever, sev eral fresh wa ter di a tom spe cies such as for ex am ple Pseudo staurosira brevistriata, Staurosira constru - ens, Diploneis ovalis, Fragilaria sopotensis or Calo - neis bacillum have been of ten en coun tered in the Bal tic Sea (Hällfors, 2004). The most abun dant (non Bal tic Sea) fresh wa ter taxon is Pseudo stauro sira brevistriata with a per cent age abun dance of be - tween 8.8 and 31.1%. The main con tri bu tors to the brack ish taxa com mu nity are Fragilaria martyi (9.4–56.4%), Navicula cincta (10.1%), N. cryptocephala (13.9%) and F. sub salina (8.4%). The char ac ter is tic ma rine spe cies Diploneis stro emii was found in high abun dance (17.3%) only in one sam ple.
In or der to show the ra tio be tween the Bal tic Sea di a tom com mu ni ties and other fresh wa ter and ter res trial taxa, all brack - Fig. 4. Thick ness, mean grain size and stan dard de vi a tion of grain size
dis tri bu tion of the ma jor sandy event layer (A), as doc u mented in cross-sec tions A–A’, B–B’ and C–C’
For lo ca tions see Fig ure 1
Fig. 5. Grain size dis tri bu tion curves av er aged for all the sam ples an a lysed taken from the event de posit layer A sam pled in cross-sec tions A–A’, B–B’, C–C’,
as well as mod ern dune and beach de pos its
The grain size dis tri bu tions are pre sented as cu mu la tive curves (A) and fre quency curves (B)
ish, ma rine and Bal tic Sea fresh wa ter spe cies have been merged into one group (Fig. 8). The av er age per cent age abun - dance of the Bal tic Sea di a toms was 70%; the high est was 87%
in core no. 255, while the low est (55%) was in core no. 26 (16–22 cm and 45–46 cm) (Fig. 8).
Nei ther ostracods nor molluscs were found in the sam ples ana lysed.
RADIOCARBON DATING
All to gether 28 bulk sed i ment 14C dates were ob tained (Ap - pen dix 1). The dat ing strat egy fo cused on iden ti fi ca tion of the ages of par tic u lar sandy event lay ers, thus most of the sam ples for the dat ing were col lected just be low and above the event lay - Fig. 6. Re la tion ships of grain size sta tis tics for sam ples taken from mod ern beach
and dune, as well as from the sandy event layer sam pled along transects A–A’, B–B’, and C–C’
Fig. 7. Ver ti cal changes in mean grain size, or ganic mat ter con tent rep re sented by loss on ig ni tion in 550°C (LOI), and ox ides of se lected ma jor el e ments: Fe, Na and Cl in trench no. 53
Sam ples in depth in ter val 16 to 33 cm rep re sent sandy event de pos its A, while the top most and low er most sam ples are taken from peaty soils above and be low the event layer
ers. The low er most part of the coastal peat se quence was dated in or der to pro vide likely time span of its de vel op ment.
The old est ages from the low er most part of the peat se quence are from 370 to 91 cal i brated years BC. Thus, the suc ces sion stud ied rep re sents ap prox i mately the last two thou sand years.
The age of layer A is es tab lished on the ba sis of 8 sam ples taken just be low the layer and 6 sam ples col lected above it (Ap - pen dix 1 and Fig. 3). All the sam ples pre dat ing the event pro - vide very sim i lar cal i brated age ranges cov er ing the 14th and 15th cen tury AD. The old est cal i brated age range is 1301–1368 AD, while the youn gest is 1414–1499 AD. The peat sam ples col lected above the event layer rep re sent two groups of ages. The first group rep re sents sam ples from sites where event layer A is a sin gle layer. In these sam ples the ages are mod ern (19th and 20th cen tury), likely due to con tri bu tion from mod ern plants. In sites where the youn ger event layer B is pres - ent (youn ger than 17th cen tury), the cal i brated age ranges of the peat above layer A are in a sim i lar range to the sam ples just un der neath the layer, namely in the range of 1392–1449 AD.
The old est sam ple taken some dis tance above layer A is peat un der ly ing layer B, dated to 1620–1675 AD. The most likely his - tor i cal event which may cor re spond to the for ma tion of the layer A is the cat a strophic storm of the 17th of Sep tem ber 1497. The slightly older ages of some sam ples col lected just above the sandy layer may be re lated to con tri bu tion of older or ganic mat - ter, likely from ero sion of pre-event peat, re de pos ited af ter the em place ment of the sandy layer.
The age of layer B is based on two sets of sam ples taken from un der neath and above the event layer (Ap pen dix 1 and Fig. 3). The cal i brated age ranges of the peat be low the layer are 1719–1819 AD and 1621–1675 AD. The dif fer ence may be due to ero sion of some pre-event peat, which is likely due to the pres ence of rip-up clasts in the event de pos its. How ever, con - tam i na tion with youn ger or ganic mat ter due to mod ern plants root pen e tra tion can not be com pletely ruled out. Both sam ples taken above the layer pro vide mod ern ages. Tak ing into ac - count the youn gest ages of the pre-event peat, the age of the event likely took place dur ing the 18th cen tury or shortly af ter - wards. The his tor i cal re cord in cludes at least two events which could be re lated to that layer, the “Seebär” from 1757 AD al - ready noted or a ma jor re gional storm in 1779 AD.
T a b l e 1 Geo chem is try of sam ples from mod ern beach, dune, mod ern and old soils and peat de pos its as well as of the in ves ti gated sandy event layer A (pro vided are ranges of val ues in sam ples
from par tic u lar groups)
Beach Dune Soil/peat Event layer
No sam ples 2 1 4 10
SiO2% 95.3–95.8 87.2 51.8–86.2 93.7–95.3
TiO2% 0.20 0.27 0.17–0.23 0.08–0.21
Al2O3% 1.7–2.1 2.26 2.6–3.5 2.1–2.5
Fe2O3% 0.36–0.39 0.49 0.7–1.9 0.2–0.4
MnO% 0.02 0.02 0.02–0.07 0.01–0.02
MgO% 0.07–0.08 0.05 0.11–0.37 0.03–0.05
CaO% 0.29–0.3 0.27 0.4–1.3 0.2–0.27
Na2O% 0.43–0.5 0.43 0.44–0.53 0.46–0.56
K2O% 0.52–0.68 0.72 0.83–0.96 0.76–0.94
P2O5% 0.03 0.04 0.08–0.31 0.02–0.03
(SO3)% <0.01 <0.01 <0.01–0.05 <0.01
(Cl)% 0.02–0.03 0.02 0.01–0.02 0.02–0.03
(F)% 0.06–0.07 0.02 <0.01–0.05 <0.01–0.1
LOI% 0.35–0.51 8.34 7.92–39.32 0.47–1.65
As mg/kg <3 <3 <3–6 <3
Ba mg/kg 16–18 12 20–50 10–23
Cd mg/kg <0.5 <0.5 <0.5 <0.5
Co mg/kg <1 <1 <1–1 <1
Cr mg/kg 1 1 7–18 <1–2
Cu mg/kg <1–1 6 3–8 <1–2
Mo mg/kg <0.5 <0.5 <0.5–0.9 <0.5
Ni mg/kg <1 2 2–7 <1
Pb mg/kg <2 8 11–25 <2–3
Sn mg/kg <2 <2 <2 <2
Sr mg/kg 7 5 10–37 4–6
V mg/kg 2 2 11–32 <1–3
Zn mg/kg 3–4 13 9–46 2–5
Fig. 8. Pro por tions of di a toms com mon in the mod ern Bal tic Sea and other fresh wa ter di a toms pres ent in sam ples col lected from sandy event layer A
For sam pling sites see Fig ure 1
layer. They pro vide over lap ping age ranges: 1166–1266 AD and 1204–1285 AD, re spec tively. Thus the event likely took place at the be gin ning of the 13th cen tury. No his tor i cal re cord of this event has been found.
RESULTS AND DISCUSSION
ORIGIN OF THE EVENT LAYERS
Geo log i cal iden ti fi ca tion of past ma rine cat a strophic in un - da tions (tsu na mis and storms) is im por tant for coastal zone risk as sess ment stud ies, es pe cially in ar eas where the his tor i cal re - cord is lim ited or ab sent (e.g., Jankaew et al., 2008; Goto et al., 2014). The main prob lem when us ing the geo log i cal ev i dence is to cor rectly iden tify the sed i men tary re cord, in par tic u lar in dis - tin guish ing be tween tsu nami and storm de pos its. Both of these are high-en ergy events that may leave coarser event de pos its con tain ing ma rine in di ca tors in coastal strati graphic se quences (e.g., Kortekaas and Dawson, 2007; Mor ton et al., 2007; Goff et al., 2012; Chagué-Goff et al., 2017; Soria et al., 2017). More - over, one al ways needs to take into ac count po ten tial post - -depositional changes of the de pos its (e.g., Szczuciñski, 2012).
The pres ent study re vealed the pres ence of three sand lay - ers (Fig. 3) de pos ited and pre served on the coastal plain near Rogowo and Mrze¿yno. All of these share sev eral char ac ter is - tics of event de pos its. They are com posed of coarser, sandy sed i ments con tain ing rip-up clasts of un der ly ing soil or peat.
The 14C dates of or ganic mat ter from be low and above them rep re sent a short time span and thus sug gest that the sand lay - ers rep re sent short-last ing events. Since the study area is in the coastal zone, close to a river val ley, sev eral pos si ble or i gins of the event lay ers are pos si ble in clud ing storms, tsu nami, river floods and wind trans port.
Sand layer A is con tin u ous, it ex tends for over 600 m in land, and is very ho mo ge neous in terms of grain size dis tri bu tion along shore as well as in a land ward di rec tion (Figs. 2–4). The thick ness of the layer var ies be tween a few and 20 cm, but in some cores it is up to 50 cm thick. Its grain size dis tri bu tions and sta tis tics sug gest that the sed i ment may be of mixed dune and beach or i gin (Figs. 5 and 6). The struc ture is mas sive and ver ti - cal grain size vari a tions (Fig. 7) re veal a slight fin ing-up wards trend. The basal con tact is sharp and ero sional, as rip-up clasts of un der ly ing peat are pres ent in the layer. The layer con tains slightly in creased con cen tra tions of Na and Cl (Fig. 7) and sig - nif i cant con tent of di a toms typ i cal of the Bal tic Sea (Fig. 8).
Ostracods and molluscs were not pres ent in the sam ples ana - lysed. The ages of the un der ly ing peat pro vide cal i brated age ranges en com pass ing the 14 and 15th cen tu ries AD. While some above-layer ages are sim i lar, where a sand layer B is also pres ent, mod ern ages were found in cases of where no over ly - ing event layer was pres ent.
The geomorphological po si tion, ex tent, grain size char ac - ter is tics and pres ence of ma rine in di ca tors in layer A seem to fa vour a storm or tsu nami rather than a flood (e.g., Skolasiñska et al., 2015; Matsumoto et al., 2016) or wind-blown or i gin of the event layer. How ever, one can not ex clude some post -depo - sitional mod i fi ca tion of the event layer due to wind trans port af - ter de po si tion, as was ob served for in stance af ter the 2011 Tohoku-oki tsu nami on Sendai Plain, where sandy tsu nami de - pos its were partly re worked by winds (Rich mond et al., 2012).
This may be for in stance a rea son for the ob served in creased
re cent North At lan tic hur ri cane de pos its, which were partly re - de pos ited land ward due to on shore di rected winds dur ing the fi - nal phase of hur ri cane.
A ma rine or i gin for the event layer is not eas ily dem on - strated in the case of the Bal tic Sea, as it is a brack ish sea with very low sa lin ity, 5 to 18 PSU in coastal wa ters of the south ern Bal tic Sea (Krzymiñski et al., 2004). So, it is not sur pris ing that the geo chem i cal sig na ture of sa lin ity is only slight (Fig. 7). How - ever, as sa lin ity in di ca tors are usu ally poorly pre served in sandy de pos its (e.g., Chagué-Goff et al., 2017), the doc u mented in - crease in Na and Cl may be mean ing ful. Ostracods and molluscs were not pres ent in the sam ples ana lysed. How ever, their pres er va tion de pends on post-burial con di tions. Peaty low - land de pos its are char ac ter ized by a large amount of humic acid re sult ing in low pH, pos si bly lead ing to dis so lu tion of car bon ate shells. The com plete dis so lu tion of car bon ate fos sils and microfossils is a com mon phe nom e non in acidic coastal set - tings (e.g., Szczuciñski et al., 2016). In such a case si li ceous rem nants of di a toms are use ful in di ca tor. The Bal tic Sea due to its low sa lin ity is a hab i tat for ma rine, brack ish spe cies but also some spe cies con sid ered to be fresh wa ter. All these groups are also found in the event layer A (Fig. 8). More over, a ma rine or i - gin of the de pos its may be in ter preted from the grain size of the de pos its, which seem to be of mixed source in clud ing coastal dune and beach prov e nance.
It proved very dif fi cult to dis tin guish be tween a tsu nami de - posit and storm de pos its, es pe cially in sed i ment cores, but also in some trench sec tions. Com par a tive stud ies (e.g., Nanayama et al., 2000; Goff et al., 2004; Tuttle et al., 2004; Kortekaas and Dawson, 2007; Mor ton et al., 2007) showed that al though both types of de pos its share many char ac ter is tics, sev eral fea tures may be con sid ered more typ i cal of storm or tsu nami de pos its.
Storm de pos its are of ten lam i nated, have shorter land ward ex - tent and reach thick nesses of sev eral tens of centi metre or more. Tsu nami de pos its are of ten more ex ten sive and thin ner (from a few mm to <0.5 m) than storm de pos its. Tsu nami de - pos its are also of ten mas sive or fine up wards, con tain intra - clasts from un der ly ing ma te rial (rip up clasts), are poorer sorted, and are capped with mud laminae. How ever, in doc u - men ta tion of re cent storms and tsu na mis, fea tures con sid ered to be typ i cal of storms are found in tsu nami de pos its (e.g., Szczuciñski et al., 2012a, b) and those con sid ered to be char - ac ter is tic of a tsu nami may be pres ent in storm de pos its (e.g., Soria et al., 2017).
Tak ing into ac count the cri te ria noted above the sand layer A seems to con tain more fea tures typ i cal of tsu nami de pos its: a sharp lower con tact, wide ex tent, mas sive struc ture with slight grain size up ward fin ing and rip-up clasts. How ever, the dat ing of the event sug gest that it is likely re lated to 17th Sep tem ber 1497 event. That event is re ported as one of the stron gest storms in his tor i cal re cords, caus ing ex ten sive flood ing of the coastal zone along a large part of the south ern Bal tic Sea coast (Piotrowski, 2007b). Due to its ex tent and size with runup height in Dar³owo re ported to pos si bly reach up to 20 m a.s.l., as well as one of the big gest his tor i cal earth quakes in Swe den be ing re ported in the same year, Piotrowski (2007b) con sid ered a pos si ble con tri bu tion to that event also from tsu nami. How ever, the more re cent his tor i cal earth quake cat a logue (Kulhanek and Persson, 2011) spec i fied the date of the 4.8 mag ni tude earth - quake to be 20th Jan u ary 1497, so ex clud ing its con tri bu tion to that storm event. The lo cal his tor i cal de scrip tions from the Rogowo re gion men tion the com plete ero sion of a bar rier spit next to the nearby Resko Przymorskie Lake (Fig. 1) by a storm
last ing more than 2 days (Ta ble 2). Thus it was likely the storm surge event that most plau si bly formed the ex ten sive sand sheet on the coastal plain.
The pos si ble mech a nism of for ma tion of tsu nami-like de - pos its by storm surge is shown in Fig ure 9. We con sider the lo - cal to pog ra phy to be of im por tance. The dune at Rogowo is rel - a tively low in con trast to high bar ri ers else where along the coast. More over, low-ly ing ter rain land ward of the dunes caused spe cific con di tions for for ma tion of the event layer. Dur - ing the ex treme, long-last ing storm the wa ter level may be set-up by a cou ple of metres along the coast. This leads to a sit - u a tion where sea level is sev eral metres higher than the coastal low land be hind the dunes. The over top ping of dunes, or dune breach ing could cause a mas sive salt wa ter in un da tion of the coastal plain, likely sup ported by strong on shore wind. Thus, the low-ly ing coastal plain could have been in un dated by uni di - rec tional flow – sim i lar to that ob served in case of tsu na mis.
Low el e va tions could pro mote un con fined flow and con struc tion of washover ter races or sheetwash de pos its (Mor ton and Sallenger, 2003). Back wash prob a bly did not oc cur un til the
storm ceased, but this is also the case in tsu na mis flood ing coastal plains, where wa ter is ponded for a long pe riod of time (e.g., Chagué-Goff et al., 2012; Tappin et al., 2012).
Sand layer B shares many sedimentological sim i lar i ties with sand layer A, in clud ing an ero sional con tact, a thick ness of >10 cm and the pres ence of rip-up clasts. How ever, its ex tent is much more con fined, al though it is not pos si ble to ex clude some amal gam ation of event lay ers. The dat ing sug gests that the or i gin of the layer could be re lated to strong storms which were com mon along the south ern Bal tic Sea coast in the 18th cen tury, re sult ing in a sim i lar sce nario to the one de scribed above for the 1497 event. How ever, an al ter na tive gen e sis of these de pos its is the “Seebär” from 1757 AD. A com par i son of his tor i cal ac counts and sedimentological ex pres sions of this event is shown in Ta ble 2. The ev i dence pro vided can not al low cer tain de ter mi na tion of the or i gin of the de pos its stud ied. Tak - ing into ac count that a tsu nami (“Seebär”) could take place dur - ing lower sea level (calm con di tions) it might be more likely that a storm surge with set-up of sea wa ter level by a cou ple of metres is re spon si ble for the for ma tion of the layer.
Layer C was found only in a small area. This may be due to its lim ited pres er va tion or its lim ited orig i nal ex tent. The lo ca tion of the layer close to the Rega River may sug gest that it is a re cord of a for - mer flood event. Tak ing into ac count the tim ing of the event at the be gin ning of the 13th cen tury, one must keep in mind that the Rega River was en ter ing the Bal tic not at its pres ent po si tion, as it then flowed through the Resko Przymorskie Lake (Fig. 1), so ma rine in un da tion along the river val ley is less likely.
The be gin ning of the 13th cen tury was also a time of in creased pre cip i ta tion in Po land (Ga³ka et al., 2013; Starkel et al., 2013).
IMPLICATIONS FOR COASTAL HAZARD ASSESSMENT
This case study shows that some ex treme salt - wa ter in un da tions re ported in his tor i cal ac counts may also be partly re con structed us ing the geo log i - cal re cord. The re cord has some lim i ta tions due to pres er va tion is sues, changes of coast line po si tion with time and hu man en gi neer ing ac tions. More - over, the ob ser va tions sug gest that it is likely im pos - T a b l e 2 Com par i son of his tor i cal in for ma tion and geo log i cal re cord of the tsu nami/storm events
Event His tor i cal de scrip tions Geo log i cal re cord
17th of Sep tem ber 1497
–storm
–many ships dam aged,
–noted along >500 km of south ern Bal tic Sea coast,
–lasted more than 2 days,
–lo cal runup height maybe up to 20 m a.s.l. in Dar³owo,
–salt wa ter in un da tion
–mas sive sand layer from dune and beach ero sion,
–the layer is up to ~50 cm thick and ex tends up to 1.2 km in land
–soil rip-up clasts,
–con tains ma rine di a toms,
–en riched in Na and Cl,
–in ter preted as a storm surge flood ing of a low land plain (~–0.5 to +0.5 m a.s.l.) with ma rine wa ters af ter coastal dunes ero sion
3th of April 1757
–calm and bright sky,
–pe cu liar roar or thun der,
–in other de scrip tions the sound ef fect, is like am mu ni tion stor age ex plo sion,
–re treat of the wa ter be fore surge,
–three in un da tions (waves)
–non-con tin u ous, mas sive sand layer,
–soil rip-up clasts,
–in ter preted as a flood ing of low land, plain (~–0.5 to +0.5 m a.s.l.) with ma rine wa ters dur ing tsu nami-like event (meteotsunami, land slide-gen er ated tsu nami) or storm (not re lated to the noted his tor i cal ac count)
Fig. 9. A sim ple con cep tual model show ing the pos si ble sit u a tion dur ing
“nor mal” con di tions, storm con di tions and a pos si ble sce nario dur ing the 1497 AD ex treme storm surge event lead ing to for ma tion of ex ten sive sandy de posit layer A, doc u mented in the pres ent con tri bu tion
geo log i cal re cords to con strain a worst case sce nario. This is par tic u larly needed along the Bal tic Sea coasts where coastal haz ards due to cat a strophic events are lit tle rec og nized while at the same time large in fra struc ture in vest ments are be ing con - sid ered, e.g. nu clear power plants. The prob a ble geo log i cal re - cord of the 1497 storm event re ported here, which was pos si bly the great est such event of the last 2000 years, pro vides a kind of min i mum es ti mate of what may hap pen dur ing such an event, when the coastal plain may be flooded more than 1 km in land.
CONCLUSIONS
Three sandy event lay ers have been iden ti fied in the coastal peat and clayey de pos its de vel oped dur ing the last ~2000 years. The most ex ten sive layer re veals a num ber of typ i cal fea tures of tsu nami de pos its (sig nif i cant lat eral ex tent, thick - ness, rip-up clasts, chem i cal and micropalaeontological ev i - dence of ma rine or i gin). Pre cise 14C dat ing along with his tor i cal ac counts showed that the ma jor layer, ex tend ing at least 1.2 km from the mod ern coast, was likely de pos ited by the big gest storm surge dur ing the last 2000 years, which took place in 1497. These storm de pos its were likely formed dur ing in un da -
sedimentological ef fects. Two dis con tin u ous sand lay ers were likely de pos ited by river flood (at the be gin ning of the 13th cen - tury) and by storm surge or a “Seebär” tsu nami-like event of youn ger age (18th cen tury). The study re vealed that the south - ern Bal tic Sea coast may be af fected by much larger coastal floods than are known from more re cent re cords and ob ser va - tions, some thing that should be taken into ac count in coastal haz ard as sess ment.
The event de pos its doc u mented, which re sem ble tsu nami de pos its, are likely of storm surge or i gin. In spe cific set tings (coastal plain, low coastal dunes) a storm surge may cause uni - di rec tional in un da tion of a coastal plain and leave de pos its alike tsu nami de pos its. Thus in such cases storm or tsu nami or i gin may not be distinguishable.
Ac knowl edge ments. The re search was funded by the Na - tional Sci ence Cen tre in Po land grant No.
2011/01/B/ST10/07220. We would like to thank J. Seidler for help in di a tom anal y sis, M. Mitrêga for as sis tance in grain size anal y sis, and T. Goslar and his team for AMS 14C dat ing. The con struc tive re views by R. Jagodziñski, A. Kotilainen and N.-A.
Mörner, as well as ed i to rial cor rec tions by E. D¹browska - -Jêdrusik are kindly ac knowl edged.
REFERENCES
Andrade, C., 1992. Tsu nami gen er ated forms in the Algarve bar rier is lands. Sci ence of Tsu nami Haz ards, 10: 21–33.
Averkiev, A.S., Klevannyy, K.A., 2007. De ter min ing cy clone tra jec - to ries and ve loc i ties lead ing to ex treme sea level rises in the Gulf of Fin land. Rus sian Me te o rol ogy Hy drol ogy, 32: 514–519.
Averkiev, A.S., Klevannyy, K.A., 2010. A case study of the im pact of cy clonic tra jec to ries on sea-level ex tremes in the Gulf of Fin - land. Con ti nen tal Shelf Re search, 30: 707–714.
Battarbee, R.D., 1986. Di a tom anal y sis. In: Hand book of Ho lo cene Paleoecology and Paleohydrology (ed. B.E. Berglund):
527–570. John Wiley and Sons.
Blott, S.J., Pye, K., 2001. GRADISTAT: a grain size dis tri bu tion and sta tis tics pack age for the anal y sis of un con sol i dated sed i ments.
Earth Sur face Pro cesses Land forms, 26: 1237–1248.
Bour geois, J., 2009. Geo logic ef fects and re cords of tsu na mis. In:
The Sea. Tsu na mis, 15 (eds. A.R. Rob in son and E.N. Ber nard):
55–91. Har vard Uni ver sity Press, Cam bridge, USA.
Brüggemann, L.W., 1779. Ausfuerliche Beschreibung des Gegen - wärtiges Zustandes des Koeniglieschen Preußischen Herzog - tums Vor- und Hinterpommern. Stettin.
Cedro, B. ed., 2012. PóŸnoglacjalne i holoceñskie przemiany œrodowiska przyrodniczego zarejestrowane w osadach pro filu T28 z okolic Mrze¿yna na podstawie badañ wielodyscy - plinarnych (in Pol ish). Przedsiêbiorstwo Produkcyjno-Handlowe ZAPOL Dmochowski, Sobczyk Sp.j.
Chagué-Goff, C., 2010. Chem i cal sig na tures of paleotsunamis: a for got ten proxy? Ma rine Ge ol ogy, 271: 67–71.
Chagué-Goff, C., Schnei der, J.-L., Goff, J.R., Dominey-Howes, D., Strotz, L., 2011. Ex pand ing the proxy toolkit to help iden tify past events – les sons from the 2004 In dian Ocean Tsu nami and the 2009 South Pa cific Tsu nami. Earth-Sci ence Re views, 107:
107–122.
Chagué-Goff, C., Niedzielski, P., Wong, H.K.Y., Szczuciñski, W., Sugawara, D., Goff, J., 2012. En vi ron men tal im pact as sess -
ment of the 2011 Tohoku-oki tsu nami on the Sendai Plain. Sed i - men tary Ge ol ogy, 282: 175–187.
Chagué-Goff, C., Szczuciñski, W., Shinozaki, T., 2017. Ap pli ca - tions of geo chem is try in tsu nami re search: a re view. Earth-Sci - ence Re views, 165: 203–244.
Credner, R., 1889. Über den „Seebär” der westlichen Ostsee vom 16/17 Mai 1888. Jahrbuch der Geographischen Gesellschaft Greifswald, R.3.
Denys, L., 1991. A check list of the di a toms in the Ho lo cene de pos its of the west ern Bel gian coastal plain with a sur vey of their ap par - ent eco log i cal re quire ments. Belgische Geologische Dienst, Pro fes sional Pa per, 1991/2: 246.
Dobracka, E., 1990. Szczegó³owa mapa geologiczna Polski 1:50000, arkusz Trzebiatów (78) (in Pol ish). Pañstwowy Instytut Geologiczny, Warszawa.
Dobracka, E., 1992. Objaœnienie do Szczegó³owej mapy geologicznej Polski w skali 1:50 000, arkusz Trzebiatów (78) (in Pol ish). Pañstwowy Instytut Geologiczny, Warszawa.
Dobracki, R., Zachowicz, J., 1997. Mapa geodynamiczna polskiej strefy brzegowej Ba³tyku po³udniowego w skali 1:10 000 arkusze Mrze¿yno i DŸwirzyno (in Pol ish). Pañstwowy Instytut Geologiczny, Szczecin.
Dominay-Howes, D.T.M., Humphreys, G.S., Hesse, P.P., 2006.
Tsu nami and palaeotsunami depositional sig na tures and their po ten tial value in un der stand ing the late-Ho lo cene tsu nami re - cord. The Ho lo cene, 16: 1095–1107.
Dziadziuszko, Z., Jednora³, T., 1996. Flood ing threat caused by storm surges at the coast of Bal tic Sea and Vistula La goon (in Pol ish with Eng lish sum mary). Wiadomoœci Instytutu Meteoro - logii i Gospodarki Wodnej, 19: 123–133.
Folk, R.L., Ward, W.C, 1957. Brazos River bar, a study of sig nif i - cance of grain size pa ram e ters. Jour nal of Sed i men tary Pe trol - ogy, 27: 3–26.
Ga³ka, M., Miotk-Szpiganowicz, G., Goslar, T., Jêœko, M., van der Knaap, W.O., Lamentowicz, M., 2013. Palaeohydrology, fires
and veg e ta tion suc ces sion in the south ern Bal tic dur ing the last 7500 years re con structed from a raised bog based on multi-proxy data. Palaeo ge ogra phy, Palaeoclimatology, Palaeo - ec ol ogy, 370: 209–221.
Goff, J., McFadgen, B.G., Chagué-Goff, C., 2004. Sed i men tary dif fer ences be tween the 2002 Easter storm and the 15th-cen - tury Okoropunga tsu nami, south east ern North Is land, New Zea - land. Ma rine Ge ol ogy, 204: 235–250.
Goff, J., Chagué-Goff, C., Nichol, S., Jaffe, B., Dominey-Howes, D., 2012. Prog ress in palaeotsunami re search. Sed i men tary Ge ol ogy, 243–244: 70–88.
Goto, K., Kawana, T., Imamura, F., 2010. His tor i cal and geo log i cal ev i dence of boul ders de pos ited by tsu na mis, south ern Ryukyu Is lands, Ja pan. Earth-Sci ence Re views, 102: 77–99.
Goto, K., Fujino, S., Sugawara, D., Nishimura, Y., 2014. The cur - rent sit u a tion of tsu nami ge ol ogy un der new pol i cies for di sas ter coun ter mea sures in Ja pan. Ep i sodes, 37: 258–264.
Gurwell, B., 2008. Coastal pro tec tion along the Bal tic Sea coast – Mecklenburg-Vorpommern. Die Küste, 74: 179–188.
Hällfors, G., 2004. Check list of Bal tic Sea phytoplankton spe cies (in clud ing some heterotrophic protistan groups). Bal tic Sea En - vi ron ment Pro ceed ings, 95.
Hua, Q., Barbetti, M., Rakowski, A.Z., 2013. At mo spheric ra dio - car bon for the pe riod 1950–2010. Ra dio car bon, 55: 2059–2072.
Hupfer, P., Harff, J., Sterr, H., Stigge, H.J., 2003. Wasserst¬nde an der Ostsee Küste. Die Küste, 66: 4–331.
Jankaew, K., Atwater, B., Sawai, Y., Choowong, M., Charoentitirat, T., Mar tin, M., Prendergast, A., 2008. Me di - eval fore warn ing of the 2004 In dian Ocean tsu nami in Thai land.
Na ture, 455: 1228–1231.
Jensen, J., Müller-Navarra, S.H., 2008. Storm surges on the Ger - man Coast. Die Küste, 74: 92–124.
Kortekaas, S., 2002. Tsu na mis, storms, and earth quakes: dis tin - guish ing coastal flood ing events. Ph.D. the sis, Coven try Uni ver - sity, UK.
Kortekaas, S., Dawson, A.G., 2007. Dis tin guish ing tsu nami and storm de pos its: an ex am ple from Martinhal, SW Por tu gal. Sed i - men tary Ge ol ogy, 200: 208–221.
Krzymiñski, W., Kruk-Dowgia³³o, L., Zawadzka-Kahlau, E., Dubrawski, R., Kamiñska, M., Lysiak-Pastuszak, E., 2004.
Typology of Pol ish ma rine wa ters. Coast line Re ports, 4: 39–48.
Kulhanek, O., Persson, L., 2011. Macroseismic ob ser va tions of Swed ish earth quakes from 1375 to 2000. SGU (Geo log i cal Sur - vey of Swe den) Re search Pa per, C 837.
Lario, J., Luque, L., Zazo, C., Goy, J.L., Spencer, Ch., Cabero, A., Bardaji, T., Borja, F., Dabrio, C.J., Civis, J., Gon za - lez-Delgado, J.A., Borja, C., Alonso-Azcarate, J., 2010. Tsu - nami vs. storm surge de pos its: a re view of the sedimentological and geomorphological re cords of ex treme wave events (EWE) dur ing the Ho lo cene in the Gulf of Cadiz, Spain. Zeitschrift fur Geomorphologie, 54, Suppl., 3: 301–316.
Lecointe, C., Coste, M., Prygiel, J., 1993. “OMNIDIA”: soft ware for tax on omy, cal cu la tion of di a tom in di ces and in ven to ries man - age ment. Hydrobiologia, 269/270: 509–513.
Lecointe, C., Coste, M., Prygiel, J., Ector, L., 1999. Le logiciel OMNIDIA ver sion 3, une puissante base de données pour les inventaires de diatomées et pour le calcul des diatomiques européens. Cryptogamie Algologie, 20: 132–134.
Luque, L., Zazo, C., Lario, J., Goy, J.L., Civis, J., Lopez-Gon za - les, F.M., Silva, P.G., Dabrio, C.J., 2004. El efecto del tsu nami de 1755 en el litoral de Conil de la Frontera (Cadiz). Zona Arqueologica, 4. Miscelanea en Homenaje a Emiliano Aguirre.
V-I, Geologia: 72–82.
Majewski, A., 1986. Ex treme wa ter level fluc tu a tions along the Bal - tic coast (in Pol ish with Eng lish sum mary). In¿ynieria Morska, 2:
46–50.
Majewski, A., 1989. Un usu ally short-lived sea wa ter level os cil la - tions on the south ern and east ern coasts of the Bal tic Sea (in Pol ish with Eng lish sum mary). Przegl¹d Geofizyczny, 34:
191–199.
Majewski, A., 1998a. The high est storm surges along the south ern coast of the Bal tic Sea (in Pol ish with Eng lish sum mary).
Wiadomoœci Instytutu Meteorologii i Gospodarki Wodnej, 21:
81–98.
Majewski, A., 1998b. Di sas trous storms and floods on south ern coasts of the Bal tic Sea (in Pol ish with Eng lish summary).
In¿ynieria Morska i Geotechnika, 2: 67–69.
Matsumoto, D., Sawai, Y., Yamada, M., Namegaya, Y., Shinozaki, T., Takeda, D., Fujino, S., Tanigawa, K., Nakamura, A., Pilar - czyk, J.E., 2016. Ero sion and sed i men ta tion dur ing the Sep tem - ber 2015 flood ing of the Kinu River, cen tral Ja pan. Sci en tific Re - ports, 6, Ar ti cle num ber: 34168. doi:10.1038/srep34168 Medvedev, I.P., Rabinovich, A.B., Kulikov, E.A., 2013. Tidal os cil -
la tions in the Bal tic Sea. Oceanology, 53: 526–538.
Mörner, N.A., 1995. The Bal tic Ice Lake – Yoldia Sea tran si tion.
Qua ter nary In ter na tional, 27: 95–98.
Mörner, N.A., 1996. Liq ue fac tion and varve dis tur bance as ev i dence of paleoseismic events and tsu na mis, the au tumn 10 430 BP event in Swe den. Qua ter nary Sci ence Re views, 15: 939–948.
Mörner, N.A., 1999. Paleo-tsu na mis in Swe den. Phys ics and Chem is try of the Earth, 24: 443–448.
Mörner, N.A., 2003. Paleoseismity of Swe den – a Novel Par a digm.
Stock holm Uni ver sity, Stock holm.
Mörner, N.A., 2008. Tsu nami events within the Bal tic. Pol ish Geo - log i cal Spe cial Pa pers, 23: 71–76.
Mörner, N.A., 2013. Drain age varves, seismites and tsunamites in the Swed ish varve chro nol ogy. GFF, 135: 308–315.
Mörner, N.A., Dawson, S., 2011. Traces of tsu nami events in off- and on-shore en vi ron ments. Case stud ies in the Mal dives, Scot - land and Swe den. In: The Tsu nami Threat – Re search and Tech - nol ogy (ed. N.A. Mörner): 371–388. InTech.
Mor ton, R.A., Sallenger, A.H. Jr., 2003. Mor pho log i cal im pacts of ex treme storms on sandy beaches and bar ri ers. Jour nal of Coastal Re search, 19: 560–573.
Mor ton, R.A., Gelfenbaum, G., Jaffe, B.E., 2007. Phys i cal cri te ria for dis tin guish ing sandy tsu nami and storm de pos its us ing mod - ern ex am ples. Sed i men tary Ge ol ogy, 200: 184–207.
Nanayama, F., Shigeno, K., Satake, K., Shimokawa, K., Koitabashi, S., Miyasaka, S., Ishii, M., 2000. Sed i men tary dif - fer ences be tween the 1993 Hokkaido-nansei-oki tsu nami and the 959 Miyakojima ty phoon at Taisei, south west ern Hokkaido, north ern Ja pan. Sed i men tary Ge ol ogy, 135: 255–264.
Nikonov, A.A., 2004. Ev i dence of paleotsunami in the early Ho lo - cene Lake Kunda (south ern coast of the Gulf of Fin land).
Doklady Earth Sci ences, 396: 477–480.
Pellikka, H., Rauhala, J., Kahma, K.K., Stipa, T., Boman, H., Kangas, A., 2014. Re cent ob ser va tions of meteotsunamis on the Finn ish coast. Nat u ral Haz ards, 74: 197–215.
Pe ters, R., Jaffe, B., 2010. Iden ti fi ca tion of Tsu nami De pos its in the Geo logic Re cord: De vel op ing Cri te ria Us ing Re cent Tsu nami De pos its. US Geo log i cal Sur vey Open-File Re port: 2010–1239.
Piotrowski, A., 2007a. Tsu nami events in Ko³obrzeg in the light of his tor i cal ev i dences. Biuletyn Pañstwowego Instytutu Geolo - gicznego, 424: 73.
Piotrowski, A., 2007b. Tsu nami AD 1497. Biuletyn Pañstwowego Instytutu Geologicznego, 424: 77.
Reimer, P.J., Bard, E., Bayliss, A., War ren Beck, J., Blackwell, P.G., Bronk Ramsey, C., Buck, C.E., Cheng, H., Ed wards, R.L., Friedriech, M., Grootes, P.M., Giuilderson, T.P., Hafilidason, H., Hajdas, I., Hatte, C., Heaton, T.J., Hoffman, D.L., Hogg, A.G., Hughen, A.G., Kai ser, A.K., Kromer, B., Man ning, S.W., Niu, M., Reimer, R.W., Rich ards, D.A., Scott, E.M., Sothon, J.R., Staff, R.A., Turney, C.S.M., van der Plicht, J., 2013. INTCAL13 and Marine13 ra dio car bon age cal i bra tion curves 0-50 000 years cal BP. Ra dio car bon, 55: 1869–1887.
Rich mond, B., Szczuciñski, W., Chagué-Goff, C., Goto, K., Sugawara, D., Witter, R., Tappin, D.R., Jaffe, B., Fujino, S., Nishimura, Y., Goff, J., 2012. Ero sion, de po si tion and land - scape change on the Sendai coastal plain, Ja pan, re sult ing from the March 11, 2011 Tohoku-oki tsu nami. Sed i men tary Ge ol ogy, 282: 27–39.
Rich ter, A., Groh, A., Dietrich, R., 2012. Geo detic ob ser va tion of sea-level change and crustal de for ma tion in the Bal tic Sea re - gion. Phys ics and Chem is try of the Earth Pt., A/B/C: 43–54.
Rosenhagen, G., Bork, I., 2009. Rekonstruktion der Sturmflut - wetterlage vom 13. No vem ber 1872. Die Küste, 75: 51–70.
Rotnicki, K., Rotnicka, J., Goslar, T., Wawrzynniak-Wydrowska, B., 2016. The first geo log i cal re cord of a palaeotsunami on the south ern coast of the Bal tic Sea, Po land. Geo log i cal Quar terly, 60 (2): 417–440.
Rudowski, S., Wróblewski, R., 2000. Sed i men tary struc ture of the storm surge washover fan on the Lake Bukowo spit. Oceano - logical Stud ies, 29: 101–107.
Shiki, T., Tachibana, T., Fujiwara, O., Goto, K., Nanayama, F., Yamazaki, T., 2008. Char ac ter is tic fea tures of tsunamiites. In:
Tsunamiites – Fea tures and Im pli ca tions (eds. T. Shiki, Y. Tsuji, T. Yamasaki and K. Minoura): 319–340, Elsevier.
Skolasiñska, K., Szczuciñski, W., Mitrêga, M., Jagodziñski, R., Lorenc, S., 2015. Sed i men tary re cord of 2010 and 2011 Warta River sea sonal floods in the re gion of Poznañ, Po land. Geo log i - cal Quar terly, 59 (1): 47–60.
Soria, J.L.A., Switzer, A.D., Pilarczyk, J.E., Siringan, F.P., Khan, N.S., Fritz, H.M., 2017. Ty phoon Haiyan overwash sed i ments from Leyte Gulf coast lines show lo cal spa tial vari a tions with hy - brid storm and tsu nami sig na tures. Sed i men tary Ge ol ogy, 358:
121–138.
Starkel, L., Michczyñska, D., Kr¹piec, M., Margielewski, W., Nalepka, D., Pazdur, A., 2013. Prog ress in the Ho lo cene chrono - climatostratigraphy of Pol ish ter ri tory. Geochronome - tria, 40: 1–21.
Stigge, H.J., 1994. Die Wasserstande an der Küste Mecklenburg - -Vorpommern. Die Küste, 56: 1–24.
Stoewer, R., 1897. Geschichte der Stadt Kolberg. C. F. Post’schen Buchhandlung und Buchdruckerei.
Stuiver, M., Reimer, P.J., Reimer, R.W., 2016. CALIB 7.1 (WWW pro gram) at http://calib.org, ac cessed 2016-12-2
Suursaar, U., Kullas, T., Otsmann, M., Kouts, T., 2003. Ex treme sea level events in the coastal wa ters of west ern Es to nia. Jour - nal of Sea Re search, 49: 295–303.
Suursaar, U., Kullas, T., Otsmann, M., Saaremae, I., Kuik, J., Merilain, M., 2006. Cy clone Gudrun in Jan u ary 2005 and mod - el ing its hy dro dy namic con se quences in the Es to nian coastal wa ters. Bo real En vi ron ment Re search, 11: 143–159.
Switzer, A.D., 2008. 20 years of paleotsunami stud ies on coastal sandsheets – a re view. 2nd In ter na tional Tsu nami Field Sym po - sium. IGCP Pro ject 495. GI2S Coast Re search Pub li ca tion, 6:
163–165.
coast: sea-level change, tsu nami or ex cep tion ally large storm?
The Ho lo cene, 18: 787–803.
Szczuciñski, W., 2012. The post-depositional changes of the on - shore 2004 tsu nami de pos its on the Andaman Sea coast of Thai land. Nat u ral Haz ards, 60: 115–133.
Szczuciñski, W., Kokociñski, M., Rzeszewski, M., Chagué-Goff, C., Cachao, M., Goto, K., Sugawara, D., 2012a. Sed i ment sources and sed i men ta tion pro cesses of 2011 Tohoku-oki tsu - nami de pos its on the Sendai Plain, Ja pan – in sights from di a - toms, nannoliths and grain size dis tri bu tion. Sed i men tary Ge ol - ogy, 282: 40–56.
Szczuciñski, W., Rachlewicz, G., Chaimanee, N., Saisuttichai, D., Tepsuwan, T., Lorenc, S., 2012b. 26 De cem ber 2004 tsu - nami de pos its left in ar eas of var i ous tsu nami runup in coastal zone of Thai land. Earth, Plan ets and Space, 64: 843–858.
Szczuciñski, W., Paw³owska, J., Lejzerowicz, F., Nishimura, Y., Kokociñski, M., Majewski, W., Nakamura, Y., Pawlowski, J., 2016. An cient sed i men tary DNA re veals past tsu nami de pos its.
Ma rine Ge ol ogy, 381: 29–33.
Tappin, D.R., Ev ans, H., Jor dan, C.J., Rich mond, B., 2012.
Coastal changes from the im pact of the 2011 Táhoku-oki tsu - nami: in ter pre ta tions of time se ries sat el lite im ages, he li cop - ter-borne video foot age and field ob ser va tions. Sed i men tary Ge ol ogy, 282: 151–174.
Tuttle, M.P., Ruffman, A., An der son, T., Jeter, H., 2004. Dis tin - guish ing tsu nami from storm de pos its in east ern North Amer ica:
the 1929 Grand Banks tsu nami ver sus the 1991 Hal low een storm. Seis mo log i cal Re search Let ters, 75: 117–131.
Van Dam, H., Mertens, A., Sineldam, J., 1994. A coded check list and eco log i cal in di ca tor val ues of fresh wa ter di a tom from the Neth er lands. Neth er lands Jour nal of Aquatic Ecol ogy, 28:
117–133.
Wil liams, H.F.L., 2015. Con trast ing styles of Hur ri cane Irene washover sed i men ta tion on three east coast bar rier is lands:
Cape Look out, North Carolina; Assateague Is land, Vir ginia; and Fire Is land, New York. Geo mor phol ogy, 231: 182–192.
Wiœniewski, B., Wolski, T., 2009. Katalogi wezbrañ i obni¿eñ sztormowych poziomów morza oraz ekstremalne poziomy wód na polskim wybrze¿u (in Pol ish). Wydawnictwo Naukowe Akademii Morskiej, Szczecin.
Wolski, T., Wiœniewski, B., Giza, A., Kowalewska-Kalkowska, H., Boman, H., Grabbi-Kaiv, S., Hammarklint, T., Holfort, J., Lydeikaite, Z., 2014. Ex treme sea lev els at se lected sta tions on the Bal tic Sea coast. Oceanologia, 56: 259–290.
