Geo. Quart. 49 (2).Vp

Geo log i cal Quar terly, 2005, 49 (2): 223–234

The Wenlock -Ludlow car bon isotope trend in the Vidukle core, Lithu a ni a, and its re lations with oce anic events

Tánu MARTMA, Antanas BRAZAUSKAS, Dimitri KALJO, Donatas KAMINSKAS and Petras MUSTEIKIS

Martma T., Brazauskas A., Kaljo D., Kaminskas D. and Musteikis P. (2005) — The Wenlock- Ludlow carbon isotope trend in the Vidukle core, Lith u ania, and its rela tions with oce anic events. Geol. Quart., 49 (2): 223–234. Warszawa.

A Wenlock to Ludlow terrigenous-car bon ate succes sion in the Vidukle core in Centra l Lithu a nia repre sents a deep shelf envi ron m ent with a genera l upwar ds-shallowing trend, inter rupted by brief deep ening epi sodes. The carbon isotope trend, based on 115 whole-rock analy ses, shows three main excur sions: (1) a major ex cursion (d13C values reach 3.2‰) in the lower Wenlock, (2) low shifts (1.3‰ and 1.6‰) at two levels in the upper m ost Wenlock Siesartis For ma tion cor re spond ing to the Monograptus ludensis Biozone, (3) the most prom i nent ex cur sion (d13C values reach 8.2‰) oc curs in the upper Ludlow Mituva Fm. The upper Ludlow excur sion is dated by the last oc cur rences of Polygnathoides siluricus be low the main shift and the ap pearance of Ozarkodina wimani and O. crispa above the excur - sion. The excur sion stratigraphically coin cides with the Lau oceanic Event and is corre late d with the mid-Ludfordian Neocucullograptus kozlowskii-Bohemograptus bohemicus tenuis Biozone. Changes in the car bon isotope trend are in genera l har mony with some aspects of the rock (CaO, terrigenous com ponent) and fossil content of the sec tion. The data presente d are consis tent with an arid clim ate model for the Ludfordian isotope event.

Tánu Martma and Dimitri Kaljo, Insti tute of Geol ogy, Tallinn Univer sity of Technol ogy, 7 Esto nia Blvd., 10143 Tallinn, Es tonia, e-mail: [email protected], [email protected]; Antanas Brazauskas, Donatas Kaminskas and Petras Musteikis, De part ment of Ge ol ogy and Min er al ogy, Vilnius Uni ver sity, 21/27 Ciurlionio Str., LT-2009 Vilnius, Lithu a nia, e-mails: antanas.brazauskas@gf,vu.lt, [email protected], [email protected] (receiv ed: Decem ber 11, 2004; accepte d: March 24, 2005).

Key words: Lith u a nia, Si lu rian, car bon iso topes, oce anic events, sed i men tary fa cies.

INTRODUCTION Silu ria n facie s and faunas, with bi otic and stable isotope events linked to atm ospheri c, oce anic and tectoni c changes (Landing and Johnson, 1998, 2003) are well-studied and can The role of carbon diox ide as a cli mate driver has be come a provide a good base for an inte grat ed analy sis of the enti re en- sub ject of great in ter est in mod ern en vi ron men tal pol icy, but vi ron men tal com plex. A gen eral pat tern of the Si lu rian car bon also in palaeoclimatology (Kump, 2000; Veizer et al., 2000). In isotope trend seems to be suffi cie ntly well-founded, only in this context data on the long-term carbon cy cling and its impact part of Llandovery and Pridoli the avail able doc u men ta tion on the vari ous aspect s of global change are highly neces sary . (Azmy et al., 1998; Kaljo et al., 1998; Kaljo and Martma, 2000; The gen eral mecha nism of car bon cycling is well es tab lished Saltzman, 2001) is rel atively scarce. (Holser et al., 1995), but inter preta ti on of actual data and dif fer- Three posi ti ve carbon isotope excur sions have been rec og - ent mod els var ies greatly, in clud ing im pli ca tions of pri mary nized in the Wenlock and Ludlow se quences of the world. The pro duc tiv ity, car bon burial, bi o log i cal ox i da tion and oce anic first ex cur sion, with the d13C val ues reaching 4–5‰, occurs in pro cesses. A short-term event like the re lease of methane from the lower m ost Wenlock and is the most well known Silu ria n gas hy drates or influx of vol canic CO2 (Weissert, 2000) may shift (for a sum mary see Munnecke et al., 2003). The second change the above sce nario, even if these ef fects are hard to excur sion, observe d in the top of the Wenlock, is double measure. Another com pli cat ing ques tion is to what degree a peaked and depend ing on sedi m entar y fa cies, is showing dif- carbon isotope anom aly is induced or de formed by local and fer ent d13C val ues. In deep shelf rocks val ues re main be low 2‰ global factors. Appli cati on of carbon isotopes for strati graphi - (Kaljo et al., 1997, 1998; Porêbska et al., 2004), but in slightly cal cor re la tions is caus ing less am bi gu ity, but biostratigraphic shal lower ones in Britain and on Gotland they reach 3–4‰ control is highly desir able . (Cor field et al., 1992; Azmy et al., 1998; Samtleben et al., 2000). 224 Tánu Martma, Antanas Brazauskas, Dimitri Kaljo, Donatas Kaminskas and Petras Musteikis

A record high value of 4.6‰ is known from the Ohesaare core, (Vilnius) for geochem istry and lithology, P. M. for brachiopod Es to nia (Kaljo et al., 1997). The youngest among the three ex - community evolution. cursions is the most prom inent one, even in terms of the enti re Phanerozoic. It is rec ognize d by d13C val ues of over 10‰ in the up per Lud low (Ludfordian) of Baltica, Laurentia, Perunica and PREVIOUS STUDIES ON THE LATE LUDLOW Austra lia, giving evi dence of the truly global di mension of this d13C EXCURSION sta ble iso tope event. The same evalu ation is also ac cept able for the first two excur sions . These ex cur sions are stratigraphically related to the follow - Jux and Steuber (1992) first noted the Ludfordian pos i tive ing con sec u tive oce anic events: the early Wen lock Ireviken sta ble iso tope shift in the Burgsvik, Hamra and Sundre beds of d13 Event, the late Wen lock Mulde Event and the late Ludlow Lau Gotland (Fig. 1), with a Ccarb peak of 5.7‰ (mean of 6 bulk d13 Event, proposed by Jeppsson (1993, 1998) and by Jeppsson rock sam ples) in the Hamra Beds and Corg peak in the and Aldridge (2000) based on a vari ety of envi ron m ental sig- Burgsvik Beds. The authors did not have sam ples from the un- nals and conodont biodiversity changes. Each event involves derly ing Eke Beds. Based on analy ses of brachi opod shells, in ter est ing prob lems (Kaljo et al., 1997, 1998, 2003; Wenzel and Joachimski (1996) showed a more com plete curve d13 Munnecke et al., 2003; Porêbska et al., 2004) deserv ing spe cial of the same Ccarb excur sion be ginning in the Lower Eke study, but here we concen tra te on the Lau Event, which coin - Beds (mean value 6.9‰), reaching peak values (8.1‰) in the cides with an extra ordi nari ly strong pos iti ve carbon isotope ex - Upper Eke Beds and conti nu ing through the Burgsvik (6.8‰) cur sion (in Aus tra lia the d13C val ues reach 12‰, Andrew et al., and Hamra (3.7‰) beds. The authors of both papers high - 1994). Marked biodiversity changes in several groups of biota lighted a cor re lation be tween the global sea level curve and d13 (cor als, graptolites, ver te brates, cono donts, acritarchs; Kaljo et changes in the dis tri buti on of carbon isotopes : the high C al., 1995) have been regis ter ed in approx i m ately the same in- values, gener all y confine d to sea level low-stand epi sodes, ter val of the Ludfordian, making the event highly intri guing. were re garded as in di ca tions of en hanced pri mary pro duc tiv ity We also discuss the earli er, Wenlock excur sions in order to and possi bly increa sed burial of Corg in Si lu rian seas. better un der stand the re la tion ships be tween the en vi ron men tal In study ing the Ludfordian ex cur sion on Gotland (with a dy nam ics and ac com pa ny ing changes in car bon cycli ng, as gap corre s ponding to the Burgsvik Beds), Samtleben et al. well as its global and lo cal aspects. We as sume that the rea sons (1996) and Bickert et al. (1997) obtai ned the same range of val - for both the isoto pic and bioti c changes can be estab li shed ues as above. New data from the Burgsvik Sandstone (val ues through a detai led study of the rela ti onships betwe en these phe - up to 7.7‰) by Samtleben et al. (2000) produced a conti nu ous nom ena and their bio- and lithofacies background. This paper profil e of the excur sion. Their inter preta ti on of the C isotope was writ ten to con trib ute to this aim. T. M. was re spon si ble for trend was based on the inter preta ti on that cli mati c changes con - isotope studies , A. B. for conodont palaeon tol ogy, D. K. trolled the fa cies dis tri bu tion (Primo and Se cun do ep i sodes of (Tallinn) for stra tig ra phy and fa cies in ter pre ta tion, D. K. Jeppsson, 1990, corre spond ing to the Hum id and Arid states of

Fig. 1. Strati graphi cal clas si fi ca tion and ter mi nol ogy em ployed in the pa per (sources: stan dard units — Kaljo et al., 1995; Scan di na via — Bassett et al., 1989; Bal tic area — Paškevièius et al., 1994; Kaljo et al., 1997) B — Bjärsjölag¯rd Member The Wenlock-Ludlow carbon isotope trend in the Vidukle core, Lithuania, and its relations with oceanic events 225 the latter au thors) with less empha sis on sea level changes. way 77 sec tion cut in the Arbuckle Mountai ns of Oklahoma the High d13C val ues were linked with arid cli mate con di tions with excur sion peak (3.6‰) is locat ed just above the P. siluricus reef growth, evap o ra tion, on shore-di rected anti-estuarine cir - cono dont Biozone in the low erm ost Henryhouse Fm. In the cu lation of sur face ocean wa ters and anoxic deep ocean Pete Hanson Creek II secti on (Robert Mountai ns, centra l Ne - (Bickert et al., 1997). vada) the exact dating of the peak (3.8‰) in the Robert Moun- Azmy et al. (1998, ta ble DR1) also re ported several d13C tains Fm. is more problem atic. Accord ing to figure 3 of val ues from Gotland, in clud ing iden ti fi ca tions from the Eke Saltzman (2001) the shift occurs at the top of the range of the (4.9–5.4‰) and Hamra (4.7–7.4‰) beds, corre lat ed respec - cono dont Kockelella variabilis and within that of Saetograptus tively with the Bohemograptus bohemicus tenuis and chi maera. If the lat ter spe cies has been iden tified cor rectly, this Neocucullograptus kozlowskii graptolite biozones. The authors indi cat es a level below the bottom of the Ludfordian Stage, i.e. note cor re la tion of the shift with a sea level lowstand, but the clearly an older event than the d13C shift in the Highway 77 sec - role of enhance d prim ary produc ti on on the burial of Corg was tion. Judging from the Baltoscandian ex pe ri ence, this seems consi dered uncle ar. unli kely and the dating should be checked be fore any conclu - In the East Bal tic the Ludfordian d13C excur sion has been sions are drawn. In dis cussing the excur sion, Saltzman (2001) es tab lished (Kaljo et al., 1997, 1998) in three core sec tions noted a connec ti on with the Pentamerid or End-siluricus from Lat via (Fig. 1). In the Priekule core the peak val ues reach Bioevent and simi lar ities with the early Wen lock Ireviken 5.9‰ in the Nova Beds of the Dubysa For ma tion (fur ther ab- Event accom panied by a low ering of sea level. brevi ate d Fm.), in the Pavilosta core (4.2‰) in the lower part, Re cently, Lehnert et al. (2003) reporte d the first record of and in the Ventspils core (5.0‰) in the top, of the Mituva Fm. the Ludfordian excur sion in connec ti on with the Lau Event In the Ohesaare core the peak level is rep re sented by a gap. The from the peri-Gondwana area (Perunica). The excur sion (max. shift was cor re lated with the early and mid-Ludfordian sea d13C whole-rock values of 4.6‰) is well dated by conodonts level lowstand and the fol low ing sea level rise, with two and brachio pods in the Muslovka sec tion near Prague: the ex - bioevents (seri ous exti ncti ons in several groups of biota, Kaljo cursion started 3 m above the last oc currence of Polygna- et al., 1995) and with the Lau Event of Jeppsson (1993). thoides siluricus and ended 1 m below the first oc currence of Data from the Bol’shezemel’skaya Tundra secti on in the Ozarkodina snajdri. The peak level oc curs just above a sharp Timan–Pechora region (Russia ) on the NE shelves of Baltica sea level fall and above sev eral fau nal ex tinc tion events re - show much lower d13C val ues in the Ludfordian than usual at corded in the secti on. In terms of graptolite biozonation, the be- this level in the Bal tic gulf area. Modzalevskaya and Wenzel ginning of the excur sion was placed into the upper part of the (1999) note that the peak val ues in the Tselebej Fm. (cor re lated Neocucullograptus kozlowskii Biozone, but the sharp shift with with the lower part of the middle Ludfordian Collarothyris the highest values conti nues into the next Monograptus canaliculata Brachio pod Zone) reach only 1.5‰, but the total latilobus Biozone (fig. 2 in Lehnert et al., 2003). This date is rise from the bot tom of the curve in the under ly ing Sizim Fm. slightly younger than that discusse d below. (= Didymothyris didyma Zone) is over 4‰. Such low val ues might have been caused by diagenetic factors, but the origi nal carbon isotope trend is still pre served. Accord ing to these au- MATERIAL AND METHODS thors the oxy gen isotope content was even less (!) influ ence d by diagenesis and therefore the low d13C val ues are most likely re lated to spe cific re gional ocean o graphic and cli ma tic con di- The isotope analy ses on the Vidukle core were perform ed tions dif fer ing from those in the Bal tic area. at the Palaeoclimatological Labo ra tory of the Insti tute of Geol - In Scania the Ludfordian carbon isotope shift reaches ex - ogy at the Tallinn Uni versit y of Technol ogy. A total of 115 tremely high d13C val ues (11.2‰, Wigforss-Lange, 1999), in sam ples were analy sed from a 324 m long secti on and the iso- the oncoid limeston es of the Bjärsjölag¯rd Mem ber (Klinta tope data obtai ned are shown in Ta ble 1. The mean sampling Fm., Öved — Rams¯sa Group). Wigforss-Lange stressed that inter val was 2.8 m, though around 1 m close to event levels . such high d13C val ues cannot be explai ned only by enhance d Whole-rock samples were crushed, the mate rial was pow dered prim ary produc ti on and/or by burial of organic carbon. These and treated with 100% phos phoric acid at 100°C for 15 min and seem to be the main reasons for values up to 5‰, but values ana lysed with a Finnigan MAT Delta-E mass spec trom e ter. above 6‰ contai n a frac tion gener ate d by the photo syn the siz - The re sults are given in the usual d-nota ti on, as per mil devi a - ing ac tiv ity of cyanobacteria and al gae re lated to the abun dance tion from the VPDB standard. Reproducibility of repli cate of oncoids in the Bjärsjölag¯rd Mem ber. anal yses was gen er ally better than 0.1‰. Very high d13C levels were doc um ented also by Andrew et al. The main advan tage of the whole-rock method is that sam - (1994) in the upper Ludlow Jack Limestone in Austra lia. The ac- pling could be perform ed at regu lar inter vals not depend ing on tual values are not given in the text, but peak values measured the oc cur rence of bioclasts. Pre vi ous stud ies (Kaljo et al., 1997; from figure 3 of Andrew et al. (1994) were 12–13‰. The shift is Heath et al., 1998; Brenchley et al., 2003) show lit tle diagenetic lo cated in the MBC sec tion (Broken Hill, Queensland) just above al ter ation of Bal tic early Palaeozoic rocks, indi cat ing that good the Pentamerid (or End-siluricus) Event and 40 m above the only result s of carbon isotope analy sis of bulk rock sam ples may be known oc cur rence of Polygnathoides siluricus. Andrew et al. (op. expect ed from the Vidukle core. Accord ing to the conodont col - cit.) empha sized causal ties be tween bi otic and iso tope events. our al ter ation index (CAI 1-1.5), the Wenlock and Ludlow rocks In North America the d13C val ues of the Ludfordian ex cur - in the core were heated to less than 100°C, suggest ing that tem - sion are slightly lower than 4‰ (Saltzman, 2001). In the High- pera ture has had only a lim ited influ ence on diagenesis. 226 Tánu Martma, Antanas Brazauskas, Dimitri Kaljo, Donatas Kaminskas and Petras Musteikis

Ta ble 1 1 2 3 4 1261.0 Dubysa Fm. –0.17 –5.13 Carbon and oxy gen isotope data from the Vidukle core 1262.8 Dubysa Fm. –0.12 –4.71 1267.9 Dubysa Fm. –0.32 –5.56 Depth [m] Stratig ra phy d13C [‰] d18O [‰] 1272.9 Dubysa Fm. –0.12 –4.93 1 2 3 4 1278.3 Dubysa Fm. –0.05 –5.32 1080.0 Minija Fm., Pridoli 0.06 –5.74 1280.5 Dubysa Fm. 0.38 –5.37 1086.5 Minija Fm., Pridoli 0.10 –5.93 1281.4 Dubysa Fm. 1.42 –4.91 1092.7 Minija Fm., Pridoli –0.02 –4.76 1281.6 Dubysa Fm. 1.47 –5.23 1098.4 Ventspils Fm. 1.04 –6.72 1283.0 Siesartis Fm. 0.57 –5.16 1102.0 Ventspils Fm. 1.42 –5.66 1284.4 Siesartis Fm. 1.56 –4.94 1106.4 Ventspils Fm. 1.56 –4.90 1286.1 Siesartis Fm. 1.31 –4.65 1110.7 Ventspils Fm. –0.37 –5.89 1287.3 Siesartis Fm. 0.76 –3.27 1115.0 Ventspils Fm. –0.85 –6.24 1289.8 Siesartis Fm. 0.53 –3.38 1115.5 Mituva Fm. 0.04 –5.83 1291.0 Siesartis Fm. 0.77 –3.92 1116.2 Mituva Fm. 1.24 –4.88 1292.2 Siesartis Fm. 0.14 –4.08 1120.0 Mituva Fm. 2.21 –2.93 1293.6 Siesartis Fm. 0.75 –3.86 1125.0 Mituva Fm. 3.38 –5.57 1295.0 Siesartis Fm. 1.16 –4.08 1130.5 Mituva Fm. 4.43 –5.92 1296.5 Siesartis Fm. 1.25 –4.16 1133.0 Mituva Fm. 4.43 –6.11 1297.9 Siesartis Fm. 1.01 –5.56 1137.2 Mituva Fm. 6.41 –4.77 1298.8 Siesartis Fm. 1.23 –5.30 1138.5 Mituva Fm. 5.31 –5.05 1299.5 Siesartis Fm. 1.31 –4.34 1140.5 Mituva Fm. 7.11 –5.21 1301.9 Siesartis Fm. 0.75 –3.46 1142.5 Mituva Fm. 7.37 –5.27 1304.1 Siesartis Fm. 0.61 –3.54 1145.2 Mituva Fm. 6.36 –4.85 1305.0 Siesartis Fm. 0.69 –3.91 1146.5 Mituva Fm. 7.60 –4.66 1306.8 Siesartis Fm. 0.15 –3.44 1147.5 Mituva Fm. 8.17 –4.68 1307.6 Siesartis Fm. 0.16 –3.39 1148.9 Mituva Fm. 4.63 –4.88 1307.9 Siesartis Fm. –0.17 –3.29 1150.0 Mituva Fm. 4.56 –4.66 1309.0 Riga Fm. 0.48 –4.29 1152.5 Mituva Fm. 4.42 –4.35 1310.0 Riga Fm. –0.09 –3.11 1155.5 Mituva Fm. 3.40 –4.52 1311.2 Riga Fm. –0.01 –3.69 1157.5 Mituva Fm. 2.92 –4.46 1311.8 Riga Fm. –0.02 –3.69 1159.0 Mituva Fm. 1.87 –4.63 1313.3 Riga Fm. 0.10 –5.60 1164.0 Mituva Fm. 0.80 –5.00 1315.3 Riga Fm. –0.03 –4.19 1166.6 Mituva Fm. 0.77 –4.87 1317.3 Riga Fm. –0.40 –5.30 1169.8 Dubysa Fm. –0.53 –4.66 1319.3 Riga Fm. –0.25 –5.63 1174.8 Dubysa Fm. –1.39 –4.46 1321.3 Riga Fm. –0.34 –5.29 1178.5 Dubysa Fm. –1.54 –4.55 1323.3 Riga Fm. –0.60 –5.60 1181.6 Dubysa Fm. –0.88 –4.81 1325.3 Riga Fm. –0.36 –5.33 1184.5 Dubysa Fm. 0.40 –5.29 1327.3 Riga Fm. –0.50 –5.81 1187.0 Dubysa Fm. 0.32 –5.29 1329.3 Riga Fm. –0.23 –5.74 1192.5 Dubysa Fm. 0.71 –5.09 1332.6 Riga Fm. 0.52 –4.27 1197.2 Dubysa Fm. 0.74 –5.38 1335.3 Riga Fm. –1.05 –4.91 1200.1 Dubysa Fm. 0.21 –5.58 1339.3 Riga Fm. –1.14 –5.97 1202.5 Dubysa Fm. 0.01 –6.10 1343.3 Riga Fm. –0.54 –5.22 1207.1 Dubysa Fm. 0.17 –4.97 1347.5 Riga Fm. –0.62 –5.08 1211.7 Dubysa Fm. 0.48 –4.95 1351.5 Riga Fm. –0.76 –5.23 1216.3 Dubysa Fm. 0.11 –5.12 1355.5 Riga Fm. –0.77 –4.76 1221.7 Dubysa Fm. –0.03 –4.80 1359.3 Riga Fm. –0.72 –5.80 1227.8 Dubysa Fm. –0.32 –4.70 1363.3 Riga Fm. –0.32 –5.20 1232.8 Dubysa Fm. 0.01 –5.61 1367.3 Riga Fm. –0.20 –5.42 1235.4 Dubysa Fm. –0.19 –4.63 1371.3 Riga Fm. –0.20 –5.19 1238.0 Dubysa Fm. 0.01 –4.79 1375.3 Riga Fm. 1.59 –4.72 1240.2 Dubysa Fm. –0.55 –4.16 1379.3 Riga Fm. 2.13 –4.72 1243.3 Dubysa Fm. –1.20 –3.67 1383.3 Riga Fm. 2.67 –4.55 1248.0 Dubysa Fm. –0.26 –4.87 1387.5 Riga Fm. 3.16 –5.04 1252.9 Dubysa Fm. –0.21 –5.26 1391.5 Riga Fm. 2.62 –5.15 1255.0 Dubysa Fm. –0.70 –4.22 1395.5 Riga Fm. 2.05 –5.63 1258.2 Dubysa Fm. 0.49 –4.21 1399.5 Riga Fm. 0.80 –5.78 1259.8 Dubysa Fm. –1.14 –4.92 1401.5 Riga Fm. 0.67 –5.77 1260.0 Dubysa Fm. –1.79 –4.78 1403.5 Rasyte Fm., Lland. 0.94 –5.66 The Wenlock-Ludlow carbon isotope trend in the Vidukle core, Lithuania, and its relations with oceanic events 227

The re li abil ity of isotope signals in the Ordo vi cia n and Si lu- Ta ble 2 rian rocks of Esto nia and elsewhere in Baltoscandia has been Geochem ical data from the Ludlow rocks of the Vidukle core discusse d in several papers (Samtleben et al., 1996; Brenchley et al., 2003). The main conclu sion is that isotope values measured from the early Palaeozoic rock samples re flect the pri - Depth CaO MgO CaO/ Cal cite Do lo mite Terrigenous mary iso to pic com po si tion of sea wa ter. An other ques tion is [m] [%] [%] MgO [%] [%] com po nent how reli able whole-rock isotope data are, of the Balti c rocks in [%] par tic u lar. Com par i son of the Bal tic Or do vi cian and Si lu rian 1097.0 53.3 1.0 53 92.9 4.3 2.9 whole-rock data with brachio pod shell isotope data shows 1098.1 54.8 0.8 69 95.9 3.6 0.5 1102.6 50.0 2.4 21 83.7 10.3 6.0 slight dif fer ences in d13C val ues, but a great simi lar ity of curves 1108.5 49.4 2.0 25 83.6 8.3 8.1 produced by both approache s (Samtleben et al., 1996; Marshal l 1112.9 21.0 6.9 3 23.4 25.9 50.7 et al., 1997; Heath et al., 1998; Brenchley et al., 2003). The ox- 1116.3 46.1 4.1 11 72.6 17.9 9.5 ygen isotope rati os are more sensi ti ve to diagenesis (Mar shall, 1119.5 20.8 13.3 2 6.2 56.9 36.9 1992) and therefore data from whole-rock analy sis may not be 1120.9 52.0 1.2 43 90.1 5.0 4.9 trust wor thy. An other dif fi culty arises from the fact that Bal tic 1125.1 54.1 0.5 108 95.5 2.0 2.5 carbon ate rocks are mostly highly vari able mixture s of calcit e 1133.4 53.8 0.9 59 93.9 3.9 2.1 and dolo m ite, which have clearly differ ent oxy gen isotope frac- 1138.4 28.0 7.7 4 32.9 31.4 35.7 tionati on factors. Consid er ing these aspect s, we do not discuss 1142.9 50.8 1.9 26 86.3 8.2 5.5 the ox y gen an a lyt i cal data pre sented in Ta ble 1. 1146.8 51.0 1.8 28 86.9 7.7 5.3 Data on the content of CaO and MgO in the Ludlow rocks 1149.7 24.2 10.8 2 18.5 45.6 35.9 of the Vidukle core, pre sented in Ta ble 2, were suppli ed by the 1151.7 47.8 2.7 18 79.2 11.5 9.4 chem i cal lab o ra tory of the Geo log i cal Sur vey of Lith u a nia ac- 1155.4 45.1 3.6 13 72.5 14.8 12.7 cording to a VSEGEI standard method of s.c. “wet chemi cal 1157.4 28.0 6.9 4 35.1 27.3 37.6 anal y sis”. Terrigenous com po nent, cal cite and do lo mite were 1165.7 44.7 1.1 40 78.2 3.1 18.7 cal culat ed from the data us ing the form ulas and method ol ogy 1166.7 47.9 1.3 36 83.1 4.7 12.2 de scribed by Kaljo et al. (1997, p. 213). 1171.4 16.9 8.3 2 12.8 31.8 55.4 1174.9 39.8 4.0 10 62.6 15.7 21.8 1186.2 8.9 6.0 1 5.5 19.3 75.2 GEOLOGICAL SETTING 1191.6 49.4 1.7 29 84.5 6.9 8.6 1195.1 17.8 5.3 3 22.2 17.6 60.2 1199.5 42.0 2.0 21 71.4 6.6 21.9 EAST BALTIC STRATIGRAPHY AND CORRELATION 1203.0 4.6 4.8 1 1.3 12.6 86.1 WITH OTHER AREAS 1208.2 7.9 4.9 2 6.8 13.6 79.7 1215.7 12.6 6.0 2 11.6 19.9 68.4 The Bal tic Si lu rian stra tig ra phy, and that of Lith u a nia in 1218.0 48.5 1.8 27 82.6 7.2 10.2 par tic u lar, is sum ma rized in sev eral pa pers (Bassett et al., 1989; 1224.2 14.8 6.2 2 14.9 21.4 63.7 Paškevièius et al., 1994; Nestor, 1997; Paškevièius, 1997). For 1235.5 39.4 3.3 12 63.4 12.6 23.9 that reason, we re strict this intro duc ti on to a brief account of the 1241.0 16.4 7.2 2 14.9 26.4 58.7 most im por tant back ground in for ma tion. The strati graphi cal 1245.2 12.3 8.7 1 4.0 33.1 62.9 framework for our study is given in Fig ure 1, where the cor re - 1251.4 5.9 5.1 1 2.7 14.5 82.8 sponding local ter minol ogy and units (regional stages and for- 1251.8 35.3 2.3 15 59.2 7.0 33.8 mati ons) are corre lat ed with the Wenlock and Ludlow part of 1257.4 42.9 2.9 15 70.5 11.5 18.1 the stan dard clas si fi ca tion of the Si lu rian Sys tem. Most of the 1257.8 9.5 5.5 2 7.8 17.0 75.2 discus sion be low concerns East Balti c se quences, but Scandi - 1269.1 8.4 5.2 2 6.6 15.3 78.1 navian and Polish mate ri als are also involved. There fore the 1274.3 8.7 4.2 2 9.9 10.3 79.8 cor re lation chart comprises partly also the latter ar eas, while 1275.3 38.1 2.2 17 64.1 7.1 28.8 other data are linked to ours through standard stratigraphi cal ter mi nol ogy. The cor re la tion of the units in Fig ure 1 is based bers and beds, which make up the East Balti c regional stages in mainly on the publi cati ons menti oned above and addi ti onall y differ ent parts of the basin, are often diachronous and therefore on Jeppsson et al. (1994), Kaljo et al. (1997) and Viira (1999). their boundari es need not coin cide with those of the Only a few re fine ments were made in this study. The rea sons chronostratigraphical units. Sev eral cases of the diachrony for these changes are dealt with in detai l in the discus sion sec- play ing an essen ti al role in our topic are shown in Fig ure 1. tion below. The East Bal tic stratigrap hi cal no men clature makes use of three cate go rie s of units: bio-, litho- and chronostratigraphical GENERAL FACIES ZONATION IN THE BALTIC AREA units, but in the regional part of Fig ure 1 only lithostrati - WITH SOME DETAILS FOR LITHUANIA graphical form ati ons and a few “beds” are shown. To save space, the regional stages were left out and the nec es sary The Vidukle core is lo cated in the south ern area of the Bal tic graptolite and conodont data are include d in the descri pti on of Gulf (Fig. 2), ly ing as a part of the Baltoscandian palaeobasin on the Vidukle sec tion. The lithostratigraphic for ma tions, mem - the west ern pericratonic margins of the Baltica palaeocontinent. 228 Tánu Martma, Antanas Brazauskas, Dimitri Kaljo, Donatas Kaminskas and Petras Musteikis

con sid er able re stric tion of the basin and movem ent of the shal low shelf and ramp fa cies belts, togethe r with brachi o - pod com mu ni ties (Fig. 3), to the south-west (Bassett et al., 1989; Nestor and Einasto, 1997; Paškevièius, 1997). Fa cies changes are ap par - ent in the rela ti onships be- tween com mu ni ties, ben thic as sem blages and fa cies belts (Musteikis and Paškevièius, 1999). The shal low est-wa ter brachi o pod com mu nity listed in Fig ure 3 is that of Sphae- rirhynchia wilsoni, be long ing to both BA 2 and 3 occur ring usu ally in the shallow est part of the open shelf fa cies belt (3 in Fig. 2). All com mu ni ties listed in BA 3 and the Day- ia-Isorthis comm unit y from BA 3–4 occur in the deeper Fig. 2. Loca tion of the study area and core section s, and general fa cies zonation of the Bal tic Gulf during part of the open shelf belt. The early Lud low nilssoni time (mod ified from Bassett et al., 1989) rem aini ng comm uni t ies be- Facies belts: 1 — tidal flat/la goon, 2 — shoal, 3 — open shelf, 4 — transi tion from open to deeper shelf, 5 — shelf longing to BA 3–4, 4 and 4–5 de pres sion are dis trib uted in the tran si tion fa cies belt (4 in Fig. 2) along By the be ginning of the Wenlock epoch the conti nent, drift ing the deepen ing gra dient towards the shelf de pression (5 in Fig. 2) from middle lati tudes of the Southern Hemisphere , reached the occu pied by a pelagi c com munit y with or without rare benthi c equa to rial belt (Cocks and Torsvik, 2002). This means that the fos sils of a more or less level sea floor. lo cal clim ate be came trop ically warm, and to wards the De vo nian The general trend of sea level is barely express ed through the some signs of increa sing aridit y and evapo ra ti on have been ob- Wen lock in the Vidukle area in the deep fa cies belt. Even the served (e.g. gypsum inclu sions in lower Ludlow rocks dis trib - Homerian sea level falls well docu mented on Gotland and uted in Centra l and East Lithu a nia, Paškevièius, 1997; see facies Saaremaa (Nestor, 1997; Calner and Jeppsson, 2003) cannot be belt 1 in Fig. 2 here). The gen eral fa cies pat tern for an early Lud - traced there. In the Ludlow the sedi m entar y en viron m ent around low time-slice (Fig. 2, com posed accord ing to the facie s model the drilling site is progres sive ly more changeable (Figs. 3 and 4) by Nestor and Einasto, 1997) more or less appli es also to the and shallowe r with a more vari able bottom topog ra phy. Wen lock and espe cia lly to later times through the Ludlow. Here, The Vidukle core (Fig. 4) comprises a se ries of some dif ference s should be menti oned (seen in Fig. 3), dem on- terrigenous-car bon ate rocks, showing upsection con tin u ous strating basin de velop m ent in terms of the evolu ti on and dis tri - shallowing of the sea with a few deepen ing epi sodes in the bu tion of brachi o pod com mu ni ties. Ludlow. The Riga Fm. is repre sent ed by rather deep-water dark The ear liest Wen lock was a time (af ter a short shallowing grey to black argillites and clayey marlstones with abundant event) when sea level was in gen eral high and was char acter - graptolites. In the middle part rare Jonesea grayi, and in the up- ized by a wide distri buti on of graptolite-bear ing and other deep per part sev eral oc cur rences of Plagiorhyncha depressa, have shelf depos it s, partly even in the centra l epicratonic ar eas of been re ported. Higher lies the Siesartis Fm. consis ting of grap - Baltica. But this highstand inter val was changing in time and to lit ic ar gil la ceous marlstones with thin limeston e interbeds. At display ed some regional vari abil ity. For ex am ple, a mid-shelf the very top, rare P. depressa have been found. The Ludlow be - limestone -marl alter ation with reef mounds of the Upper Visby gins with the Dubysa Fm. subdi vided into two parts (Fig. 4). Beds on Gotland shows an upwards increa se of a carbon ate The Šešupe Beds are more ar gil laceous, con tain ing marls with con tent, but in the East Baltic the max i mum dis tri bu tion of interbeds of nodu lar limestone s rich in graptolites and shelly grapto li tic rocks was confine d to the early riccartonensis time fauna (mainly the Jonesea grayi comm u nity) and are more cal - (Bassett et al., 1989). Slow, but steady re gression of the car eous, es pe cially at the top; the Nova Beds con sist of al ter nat - Palaeobaltic Sea, even if inter rupted by brief transgressive ep i- ing nod u lar lime stones and marly in ter ca la tions. Pe lagic and sodes, began in late riccartonensis time, marked by magnif i - shelly fau nas alter nate, the lat ter pre vail ing. From the bottom cent reef/bioherm struc tures of the Högklint (Gotland) and up wards the fol low ing suc ces sion of brachi o pod com mu ni ties Ninase (= upper Riga in Fig. 1; Saaremaa Is land, Es to nia) for- is observe d: Lissatrypa obovata (BA 4) at 1215 m, Dayia mati ons. By the end of the Ludlow the regres sion had caused (BA 4) at 1204–1212 m, Dayia-Isorthis (BA 3–4) several times The Wenlock-Ludlow carbon isotope trend in the Vidukle core, Lithuania, and its relations with oceanic events 229

Fig. 3. Dis tribu tion of ben thic as sem blages (BA) in Lith u a nia through the late Wen lock to the late Lud low

In the belts marked by BA indi ces brachio pod taxa repre sent ing cer tain com mu nities are shown; the eastern m ost belts without com munity names are rep - re sented by dolomites, where no brachio pods have been identi fied; the figure is based on a serie s of com munity distri bution maps com piled by P. Musteikis betwe en 1281 and 1287 m and Protochonetes piltenesis from with thick marl interbeds. The upper Ventspils Fm. and the the Atrypoidea com munit y (BA 2) at the very top. Minija Fm. mark the begin ning of a specif ic infil ling stage of The upper Ludlow Ludfordian Stage com prises two for ma- sed i men ta tion of the Palaeobaltic basin, highli ghted by a marked tions. The lower, Mituva Fm. is repre sent ed by nodu lar and lam i- increa se in the influx of terrigenous mate rial into the sea (Nestor nated limeston es with marl in ter ca la tions; in the up per part, skel - and Einasto, 1997). e tal lime stones with oncolite in ter ca la tions pre vail. A suc ces sion of brachi o pod com mu ni ties, clearly ex press ing the shallowing tendency of the basin, is recog nize d: Alaskospira ludloviensis CARBON ISOTOPE TREND AND DATING (BA 1–2) at the lower boundary , Dayia-Isorthis (BA 3–4) at OF THE EXCURSIONS 1158–1162 m, al ter nat ing Isorthis ovalis, Atrypa reticularis, Protochonetes piltenensis and Microsphaeridiorhynchus nucula (BA 2–3) be tween 1140 and 1157 m. Higher, espe cia lly at The carbon isotope trend shows the follow ing main excur - 1120–1135 m, mem bers of the Atrypoidea com munit y (BA 2) sions: dom i nate. The Herrmannina com munit y oc curs at the upper 1. A major carbon isotope excur sion (d13C val ues reach bound ary, to gether with Atrypoidea conti nu ing also into the 3.2‰ at 1387.5 m) occurs in the lower Wenlock; higher, the over ly ing Ventspils Fm. The Ventspils Fm. con sists mainly of values re main be low or close to 0‰. clayey lami nated limeston es with silty marls. Its upper part 2. In the upper m ost Wenlock Siesartis Fm. slightly greater marks a short-lived drowning event (oc currence of BA 3 brachi- values were recorded at two levels : 1.3‰ at 1299.5 m and 1.6‰ o pods Isorthis ovalis and Atrypa reticularis). The lower Pridoli at 1284.4 m. Graptolites found (Paškevièius et al., 1994) in this Minija Fm. (above 1093 m) is repre sent ed by nodu lar limestone s bed date these weak shifts to the Monograptus ludensis Biozone. 230 Tánu Martma, Antanas Brazauskas, Dimitri Kaljo, Donatas Kaminskas and Petras Musteikis

Fig. 4. The Vidukle core sec tion: stra tig ra phy, li thol ogy, sea level changes in terpreted as ben thic assem blages (BA) succes sion and the d13C trend dated by graptolite (Paškevièius et al., 1994) and cono dont biostratigraphy

Lo cation of oce anic events af ter Jeppsson (1998); arrows — the first or last oc currenc es, in ter vals — the full ranges, open cir cles — sin gle oc cur rences of taxa

3. The lower Ludlow Dubysa Form ati on shows also low from the upper part of the Lau Event on Gotland (Jeppsson, values (0 to 0.7‰) with a nega ti ve shift reaching –1.5‰ at the 1998). The first appear ance s of Ozarkodina eosteinhornensis top. The shift is dated by the last occur rence of Bohemograptus scanica and O. ambiqua at 1130.5 m and O. wimani at 1126.7 m bohemicus tenuis and by the ap pear ance of Polygnathoides date the upper falling limb of the excur sion. The true O. crispa siluricus (Fig. 4). appears above the upper nega ti ve shift at 1107.5 m in the low er- 4. In the upper Ludlow Mituva Fm. the carbon isotope trend most Ventspils Fm. The conodont occur rence s al low us to date changes com pletely, val ues of the posi ti ve d13C ex cur sion reach the excur sion as of middle Ludfordian age. 8.2‰ (at 1147 m) and com mencing from 1178 m, the enti re shift 5. In the upper m ost Ludlow the carbon isotope values re - reaches 9.7‰. The excur sion oc cupies the enti re form ati on turned to about 1‰. (Fig. 4). A nega ti ve shift (–0.9‰ at 1115 m), sim ilar to that at the start of the excur sion, occurs at the boundary with the overly ing Ventspils Fm. The be ginning of the excur sion is dated by the last DISCUSSION oc cur rences of Polygnathoides siluricus (depth 1167 m) just be - low the main shift and togethe r with the first appear ance s of Coryssognathus dubius. The latter forms a narrow but disti nct STRATIGRAPHIC DATING OF THE MAIN SHIFTS biozone at the very begin ning of the posi ti ve shift. The highest peak of the excur sion coin cide s exact ly with the Panderodus The early Wenlock excur sion is well dated by graptolite oc- equicostatus Biozone (1152.5–1133.0 m), which is also known cur rences (Kaljo et al., 1997), and so the corre la ti on of the peak The Wenlock-Ludlow carbon isotope trend in the Vidukle core, Lithuania, and its relations with oceanic events 231 level with the Monograptus riccartonensis Biozone is com - si tion ex pressed by Urbanek and Teller (1997) based on Pol ish monly ac cepted (Azmy et al., 1998; Saltzman, 2001). The ex- core sec tions with a rich graptolite as semblage seems more cursion, however, is broad and in some secti ons (Kaljo et al., firmly grounded. Their data show that Pseudomonoclimacis 2003; Munnecke et al., 2003) its slowly rising limb begins in latilobus and Monograptus (Slovinograptus) balticus oc cur to- the topm ost Llandovery; in the Vidukle core (Fig. 4) it con tin - gether above the Neocucullograptus kozlowskii Biozone, and ues above the riccartonensis Biozone (sensu stricto). The fall- the boundary betwe en these zones marks an im portant change ing limb reaches the M. flexilis Biozone. The data from the in graptolite faunas. Vidukle core show clearly that the Ireviken oce anic Event as An addi ti onal aspect linked to this dating problem is the de fined by Jeppsson (1998) occurs be low the peak of the early cor re la tion of the Eke, Burgsvik and Hamra beds of Gotland Wen lock d13C ex cur sion. with the Mituva Fm. of Lithu a nia and with some part of the The same rela ti onships are obvi ous also in the case of the Klinta Fm. of Scania and, on the whole, with the kozlowskii- Mulde Event (Jeppsson, 1998; Calner and Jeppsson, 2003) and bohemicus tenuis Zone. Based on the loca ti on of the carbon the mid-Homerian d13C excur sion. New data from the Vidukle isotope excur sion, this corre la ti on seems secure, while only the core allow us to correct an earli er dat ing of that excur sion sug - Sundre Beds and perhaps the upper part of the Hamra Beds gested by Kaljo et al. (1997). The graptolites (Fig. 4) reported could be placed into the higher latilobus/balticus Zone. by Paškevièius et al. (1994) prove the occur rence of the double shift in the Vidukle core to lie within the Monograptus ludensis RELATIONSHIPS OF THE d13C EXCURSIONS Biozone. This dat ing is in accord with the data by Saltzman WITH SEDIMENTARY CHANGES (2001) from North America . The sit ua ti on in the Vidukle core seems very clear and the dating is well-con strained, but we are Changes in the sed i men tary en vi ron ment in West Lith u a nia not fully convince d that all these weak shifts observe d in the are in gen eral terms reflected in Fig ure 4 by the lithological log Ohesaare, Priekule and Vidukle cores are situ ated at the same of the Vidukle core and succes sion of the benthi c as sem blages level of the ludensis Biozone. A re cent pa per by Porêbska et al. in the secti on and in the area (Fig. 3). Some ad di tional char ac - (2004) reporte d about two very weak shifts in the dubius and ter isti cs could be added in part of the Ludlow rocks based on praedeubeli zones and this suggest s such a possi bil ity also ex - geo chem i cal data pre sented in Ta ble 2 and in form of trend ists in the East Bal tic area. lines for cal cite, do lo mite and terrigenous com po nent plot ted d13 13 The re la tion ships be tween the mid-Ludfordian C ex cur - against the d C ex cur sion in Fig ure 5. sion and the Lau Event seem not to be consta nt. Above, we The Wenlock part of the se quence repre sent ed by predom i- showed their nearly full coin ci dence (Fig. 4), but there exist s nantly highly argil laceous rocks (mudstones, marlstones) with some dis crepancy in the dat ing of the excur sion. Graptolites are pe lagic fos sil as sem blages (graptolites, a few ben thic brachio - very rare in the Ludfordian of the Bal tic–Scan di na vian Re gion, pods in more car bo na ceous in ter ca la tions), ex hib its a rather sta - except for in Poland, which belongs to the East Euro pean Plat - ble en vi ron men tal sit u a tion. The Vidukle core rep re sents de po si- form (Urbanek and Teller, 1997). On the basis of a few tion on the deep shelf where striking fa cies changes (e.g. appear - graptolite oc cur rences and/or on cono dont dis tri bu tions most ance of skel etal limestone s with organic buildups) observe d in authors have corre lat ed this excur sion and event with the the shal lower shelf ar eas (Gotland and Saaremaa islands ) did not Neocucullograptus kozlowskii-Bohemograptus bohemicus play any es sen tial role. Still the fa cies pat tern of the ba sin or a tenuis (Fig. 1) graptolite Biozone or a part of it (Kaljo et al., 1997, 1998; Azmy et al., 1998; Jeppsson and Aldridge, 2000). Several other authors have placed the excur sion at the more or less same level — into the post-Leintwardinian Ludfordian or up per Lud low (Wenzel and Joachimski, 1996; Bickert et al., 1997; Wigforss-Lange, 1999). Only Lehnert et al. (2003) sug- gested a slightly higher posi ti on, with the peak val ues in the lower part of the Monograptus latilobus Biozone. In the Mušlovka sec tion an erosiona l surface occurs just below the main d13C shift in top of a dark micritic limestone bed. The bed and the sur face were in ter preted as an in di cation of a marked sea level fall and the latilobus/kozlowskii zonal boundary was placed at that level. The ero sional event makes the ac tual start - ing point of the excur sion questi onable , but more im portant ly the lower boundary of the latilobus Biozone is by no means proven becaus e of the lack of graptolite oc cur rences. We think that also in Mušlovka the mid-Ludfordian shift marks the kozlowskii Biozone. The distri buti on of conodonts also allows a differ ent corre - lati on and therefore the dat ing problem deserve s further atten - Fig. 5. Changes in the con tents of do lo mite (DOL), cal cite (CAL), terrigenous com po nent (TER) cor related with the d13C trend in the tion. Paškevièius (1997) corre lat ed the Mituva Fm. and the lo- Lud low rocks of the Vidukle core cal Monograptus balticus Biozone with the kozlowskii- Contents of do lom ite and calcite is given as 5-point moving aver age based bohemicus tenuis Zone (Fig. 1). This is a pos si bil ity, but the po - on Ta ble 2 232 Tánu Martma, Antanas Brazauskas, Dimitri Kaljo, Donatas Kaminskas and Petras Musteikis large part of it should be kept in mind, when discuss ing the envi - chem is try: the mean CaO val ues, cor re spond ing to the d13C ron men tal back ground of the car bon iso tope ex cur sions. values 2–4, 4–6 and over 6‰, are 44.7 (49 when exclud ing one The early Wen lock (more precisely early Sheinwoodian) argil laceous interlayer), 41.9 (50.8) and 43.2 (51)%. This se- and the late Wen lock (= late Homerian) d13C ex cur sions have quence of num bers shows only a very general corre la ti on, as been regard ed (Bickert et al., 1997; Munnecke et al., 2003) as menti oned in the item 2 and as it might be expect ed. the A-peri ods with evapo ra ti on and low rate of conti nental In sum mary, it should be pointed out that the above dis- weath er ing at low lat i tudes in con di tions of arid cli mate. On cussed rela tion ships as well as a pattern demon strated in Fig ure shal low shelves reefs and carbon ate plat forms were expand ing, 5 show a step-by-step decrea se of influx of the terrigenous ma - while on the mid-shelf grey shales and limestone s and deeper terial into the basin and in crease of calcium carbon ate sed im en - black shales oc curred. During the A-peri ods the oceanic cir cu- tati on. The change of dom inat ing proces s occurred slightly be - la tion fol lowed an anti-estuarine (i.e. shore-wards di rected at a fore the comm ence ment of the mid-Ludfordian carbon isotope sur face) pattern and clear wa ters showed high d13C val ues, excur sion. This change could be inter preted as a replac em ent of which were de creasing with wa ter depth. Inter pret ing the hum id type of sedi m entar y en viron m ent with a more arid one. Vidukle data set in the context of an A-period as descri bed As an addi ti onal proxy we refer to the oc currence of oncolites above accord ing to the model by Bickert et al. (1997) we can in the upper part of the Mituva Fm. in the Vidukle core. note sev eral dif fi cul ties, wide dis tri bu tion of ar gil la ceous and Oncolites are com mon in this part of the secti on also on lim ited oc cur rences of car bo na ceous rocks in par tic u lar. The Gotland (Eke and Hamra beds, Samtleben et al., 1996) and in lat ter proxies , espe cia lly in a basin-wi de background, are more Scania (Bjärsjölag¯rd Mem ber, Wigforss-Lange, 1999). in dic a tive for a H-pe riod of hum id clim ate, but should have low On the other hand, these data and Fig ure 5 re veal that cor re - d13C val ues. How ever, con sid er ing also Gotland and Saare maa lati on betwe en dif ferent proxies is only very general , not data and high d13C val ues in the Vidukle core we be lieve that strictly one to one, which means that the data dis cussed here al - these two carbon isotope excur sions can be fit ted into the A-pe- low to pinpoint only some fac tors of the carbon cy cling while riod patter n. sev eral oth ers re main ob scure. The lat ter may in clude, for ex - In the Ludlow rocks of the Vidukle sec tion the content of am ple, the prim ary produc ti on and role of organic carbon the CaO, MgO, calcite, dolo mite and terrigenous com po nent burial, which seem to be im portant becaus e of the ex tinc tion of was analy sed. The data show the fol lowing distri buti on pattern: biota close to carbon isotope events (Ireviken, Lau events, 1. The terrigenous com po nent (TER) dom inate s over the Jeppsson, 1998). However, it is quite clear that the dis ap pear- car bon ates in the mudstones and marlstones (sev eral cal- ance of ten or more taxa at a da tum plane did not af fect se ri - cite-rich interbeds exclude d, Ta ble 2) of the Dubysa Fm., but in ously bioproduction in general (based more on the planktic mi - limeston es of the Mituva Fm. its content rem ains below 20% crobial produc ti on) and, therefore, it is diffi cult to identi fy their and above 1150 m even be low 10%. Excep ti ons are dolomitic influ ence . But the bi otic changes indi cat e that somethi ng is interbeds where TER content reaches 36–38%. The CaO and happen ing in the envi ron m ent and we do think that this is the calcite con tents are variable (es pe cially in the Dubysa Fm.), but cli mate in gen eral and its spe cific impli ca tions, which af fect the increa se up to the top of the Mituva Fm. (cor re spond ingly over biota in very differ ent ways. In this context the Vidukle data 50 and 90%), with a few less cal car eous lev els (cal cite and com pari sons with other ones show that the main driver of 18–33%). The MgO content is low and varies below 10%; in the carbon isotope trend is global factors, local ones play only a the Mituva Fm. it is be low 5% (dolomitic interbeds exclude d). mod i fy ing role. Fig ure 5 demon strates that in the lat ter in ter val the do lo mite is Consid er ing the high values reached and the very steep mainly tied to the terrigenous com po nent. rising limb of the excur sion, the envi ron m ental change that 2. The d13C pos itive shift co in cides with the beds (dolomitic caused the mid-Ludfordian isoto pic event must have been interbeds exclude d) where the CaO content is consis tently over rather rapid. The excur sion com menced from –1.5‰ at 40% and calcit e content 72–96%; both nega ti ve shifts be fore 1178.5 m and the peak (8.2‰) was reached at 1147.5 m; this and af ter the mid-Ludfordian excur sion oc cur in rocks with a yields 9.7‰ for the to tal shift. In terms of the IUGS 2004 rel a tively high clay con tent. time-scale for Ludlow the thickness of the rock se quence 31 3. In detai l some corre la ti ons betwe en the CaO content and m may corre spond to 700 000 years. Based on these figures the CaO/MgO rati o on the one hand, and the d13C val ues on the we cal cu lated that the rise of the d13C val ues through 31 m other, show a fairly regu lar patter n. For exam ple, in the was 0.3‰ per metre or 1‰ per 70 000 years. The topm ost part clay-rich Šešupe Beds (Fig. 4 and Ta ble 2), where the CaO of the peak was much steeper (1.44‰ per metre) and the last content is as a rule below 10% and the CaO/MgO rati o is 1–2, 4‰ were added only during 55 000 years. The fall ing limb of the d13C val ues re main be low 0‰ (–0.1 to –0.7‰). In a few the excur sion is of the same length that means the enti re ex- lime stone interbeds where the CaO/MgO rati o is about 15 the cursion corre sponds to around 1.4 Ma. This time span is a d13C val ues increa se up to 0.5–1‰. In the Nova Beds the mean con sid er able in ter val even in geo log i cal terms for en vi ron - CaO content is twice as high as in the beds below and the d13C mental changes, the same should be stressed also about the values also increa se above 0‰ up to 1‰. ini tial part of the shift and only the highest peak was formed 4. Inter vals with high d13C val ues in the Mituva Fm. show rather rapidl y, although it would be hard to classify the event the follow ing pat tern of corre la ti on betwe en Ca and C geo - as a cata strop hic one. The Wenlock-Ludlow carbon isotope trend in the Vidukle core, Lithuania, and its relations with oceanic events 233

CONCLUSIONS 3. The mid-Ludfordian pos i tive ex cur sion be gan and ended with neg a tive shifts, which seem to re flect the in her ent pat tern of the Late Si luria n carbon isotope trend. These nega ti ve shifts Based on data from the Vidukle core, the follow ing conclu - cor re late to some extent with sedimentological changes sug- sions can be made: gest ing oce anic and cli ma tic pro cesses char ac ter is tic of more 1. The Wenlock carbon isotope excur sions agree well with hu mid ep i sodes; the pos i tive ex cur sion it self may be re ferred to ear lier data from Baltica and else where, and the d13C val ues an arid epi sode and the enti re inter val studied is a part of a lon- clearly re flect the depth of the sed im en tary ba sin. In both cases ger aridisation period. the oce anic events directly pre cede the car bon iso tope events. The Ireviken Event pre cedes the major early Wen lock d13C ex - Ac knowl edge ments. The authors thank A. Munnecke and cursion and the Mulde Event the late Homerian less prom inent J. Zalasiewicz for useful construc tive com ments, T. Kiipli is but double shift. thanked for as sis tance at geo chem i cal cal cu la tions. G. 2. The peak val ues of the mid-Ludfordian d13C ex cur sion oc- Baranov, A. Noor and H. Kukk for techni cal and linguis ti c cur within the lim its of the Lau Event as define d by conodonts, help. The study was partly supporte d by the Es tonian Science but the ris ing and fall ing limbs are at least in part lo cated be fore Founda ti on (grants no. 5042 and 6127). The pa per is a contri - and af ter the Event. This in di cates that the global pro cesses caus - buti on to the IGCP Project 503. ing the carbon isotope excur sion and the Lau oce anic Event were active to a large ex tent dur ing the same time span.

REFERENCES

ANDREW A. S., HAMILTON P. J., MAWSON R., TALENT J. A. and JEPPSSON L. (1993) — Silu ria n events, the theory , and the conodonts. WHITFORD D. J. (1994) — Isoto pic cor rela tion tools in the mid-Pa- Proc. Es to nian Acad. Sc. Geol., 42: 23–27 leo zoic and their re lation to ex tinc tion events. Aus tralian Petrol. JEPPSSON L. (1998) — Silu ria n oceanic events: sum mary of genera l Explor. Ass. J., 34: 268–277. char ac ter is tics. In: Si lu rian Cy cles — Link ages of Dy namic Stra tig ra - AZMY K., VEIZER J., BASSETT M. G. and COPPER P. (1998) — Ox y- phy with At mo spheric, Oce anic and Tectonic Changes (eds. E. Land - gen and car bon iso to pic com po si tion of Si lu rian brachi o pods: im pli ca - ing and M. E. John son). N. Y. St. Mus. Bull., 491: 239–257. tions for co eval sea wa ter and glaciations. Geol. Soc., Am. Bull., 110: JEPPSSON L. and ALDRIDGE R. J. (2000) — Ludlow (late Silu ria n) oce - 1499–1512. anic ep isodes and events. J. Geol. Soc. London, 157: 1137–1148. BASSETT M. G., KALJO D. and TELLER L. (1989) — The Bal tic re gion. JEPPSSON L., VIIRA V. and MÄNNIK P. (1994) — Silu ria n cono- In: A Global Standard for the Si luria n Sys tem (eds. C. H. Holland and dont-based cor re la tions be tween Gotland (Sweden) and Saaremaa (Es - M. G. Bassett). Nat. Mus. Wales. Geol. Ser., 9: 158–170. to nia). Geol. Mag., 131: 201–218.

BICKERT T., PÄTZOLD J., SAMTLEBEN C. and MUNNECKE A. JUX U. and STEUBER T. (1992) — Ccarb und Corg — Isotopenverhältnisse (1997) — Paleoenvironmental changes in the Silu ria n indi cate d by in der silurischen Schichtenfolge Gotlands als Hinweise auf stable isotopes in brachio pod shells from Gotland, Sweden. Geochim. Meeresspiegelschwankungen und Krustenbewegungen. N. Jb. Geol. Cosmochim. Acta, 61 (13): 2717–2730. Paläont., Mh., 7: 385–413. BRENCHLEY P. J., CARDEN G. A., HINTS L., KALJO D., MARSHALL KALJO D., BOUCOT A. J., CORFIELD R. M., LE HERISSE A., KOREN J. D., MARTMA T., MEIDLA T. and NOLVAK J. (2003) — High reso - T. N., KRIZ J., MÄNNIK P., MÄRSS T., NESTOR V., SHAVER R. H., lu tion iso tope stra tig ra phy of Late Or do vi cian se quences: con straints on SIVETER D. J. and VIIRA V. (1995) — Silu ria n Bio-Events. In: the tim ing of bio-events and en vi ron men tal changes as so ci ated with Global Events and Event Stratig ra phy in the Phanerozoic (ed. O. H. mass extinc tion and glaci ation. Geol. Soc., Am. Bull., 115: 89–104. Walliser): 173–224. Springer-Verlag. CALNER M. and JEPPSSON L. (2003) — Carbon ate platform evolu tion KALJO D., KIIPLI T. and MARTMA T. (1997) — Car bon isotope event and cono dont stra tig ra phy dur ing the mid dle Si lu rian Mulde Event, markers through the Wen lock-Pridoli se quence in Ohesaare (Es to nia) Gotland, Sweden. Geol. Mag., 140: 173–203. and Priekule (Lat via). Palaeogeogr., Palaeoclimatol., Palaeoecol., COCKS L. R. M. and TORSVIK T. H. (2002) — Earth geog ra phy from 500 132: 211–224. to 400 million years ago: a faunal and palaeomagnetic review . J. Geol. KALJO D., KIIPLI T. and MARTMA T. (1998) — Corre la tion of carbon Soc. London, 159: 631–644. iso tope events and en vi ron men tal cyclicity in the East Baltic Silu rian. CORFIELD R. M., SIVETER D. J., CARTLIDGE J. E., McKERROW W. In: Si lu rian Cycles — Link ages of Dynam ic Stratig raphy with At mo - S. (1992) — Carbon isotope excur sion near the Wenlock- Ludlow (Si- spheric, Oceanic and Tectonic Changes (eds. E. Landing and M. E. luria n) boundary in the An glo-Welsh area. Ge ology , 20: 371–374. John son). N. Y. St. Mus. Bull., 491: 297–312. HEATH R. J., BRENCHLEY P. J. and MARSHALL J. D. (1998) — Early KALJO D. and MARTMA T. (2000) — Car bon isoto pic com posi tion of Si lu rian car bon and ox y gen sta ble-iso tope stra tig ra phy of Es to nia: Im - Llandovery rocks (East Bal tic Si lu rian) with en vi ron men tal in ter pre ta - pli cations for cli matic change. In: Si lu rian Cy cles — Link ages of Dy- tion. Proc. Es to nian Acad. Sc. Geol., 49: 267–283. namic Stra tig ra phy with At mo spheric, Oce anic and Tec tonic Changes KALJO D., MARTMA T., MÄNNIK P. and VIIRA V. (2003) — Im plica - (eds. E. Land ing and M. E. John son). N. Y. St. Mus. Bull., 491: tions of Gondwana glaciations in the Bal tic late Ordo vi cian and Si lu - 313–327. rian and a carbon isoto pic test of envi ron m ental cyclicity. Bull. Soc. HOLSER W. T., MAGARITZ M. and RIPPERDAN R. L. (1995) — Global Geol. France, 174 (1): 59–66. isoto pic events. In: Global Events and Event Stratig ra phy in the KUMP L. R. (2000) — What drives cli mate? Nature, 408: 651–652. Phanerozoic (ed. O. H. Walliser): 63–88. Springer-Verlag. Berlin-Hei - LANDING E. and JOHNSON M. E. eds. (1998) — Si lu rian cycles — link - del berg-New York. ages of dynam ic stratig raphy with at mo spheric, oce anic and tectonic JEPPSSON L. (1990) — An oceanic model for lithological and faunal changes. N. Y. St. Mus. Bull., 491. changes tested on the Silu ria n record. J. Geol. Soc. London, 147: LANDING E. and JOHNSON M. E. eds. (2003) — Si lu rian lands and seas. 663–674. Paleogeography outside of Laurentia. N. Y. St. Mus. Bull., 493. 234 Tánu Martma, Antanas Brazauskas, Dimitri Kaljo, Donatas Kaminskas and Petras Musteikis

LEHNERT O., FRYDA J., BUGGISCH W. and MANDA Š. (2003) — A first PAŠKEVIÈIUS J., LAPINSKAS P., BRAZAUSKAS A., MUSTEIKIS P. report of the Ludlow Lau Event from the Prague Ba sin (Barrandian, and JACYNA J. (1994) — Stratigraphi c re vision of the regional stages Czech Repub lic). In: Proc. 7th IGC-FMSSS (eds. G. Ortega and G. F. of the Upper Silu ria n part in the Bal tic Ba sin. Geologija, 17: 64–87. Acenolazas). INSUGEO, Ser. Correlat. Geol., 18: 139–144. PORÊBSKA E., KOZ£OWSKA-DAWIDZIUK A. and MASIAK M. (2004) MARSHALL J. D. (1992) — Cli matic and ocean o graphic iso to pic sig nals — The lundgreni event in the Si luria n of the East Euro pean Platform , from the carbon ate rock re cord and their preser va tion. Geol. Mag., Po land. Palaeogeogr., Palaeoclimat., Palaeoecol., 213 (3–4): 271–294. 129: 143–160. SALTZMAN M. R. (2001) — Si lu rian d13C stra tig raphy: a view from MARSHALL J. D., BRENCHLEY P. J., MASON P., WOLFF A., ASTINI North Amer ica. Ge ol ogy, 29: 671–674. R. A., HINTS L. and MEIDLA T. (1997) — Global iso to pic events as - SAMTLEBEN C., MUNNECKE A. and BICKERT T. (2000) — Devel op - soci ated with mass extinc tion and glaci ation in the Late Ordo vi cian. ment of facies and C/O-isotopes in transects through the Ludlow of Palaeogeogr., Palaeoclimat., Palaeoecol., 132: 195–210. Gotland: evi dence for global and local influ ences on a shal low-marine MODZALEVSKAYA T. L. and WENZEL B. (1999) — Biostratigraphy en vi ron ment. Fa cies, 43: 1–38. and geo chem is try of Up per Si lu rian brachi o pods from the SAMTLEBEN C., MUNNECKE A., BICKERT T. and PÄTZOLD J. (1996) Timan-Pechora region (Rus sia). Acta Geol. Pol., 49 (2): 145–157. — The Si lu rian of Gotland (Swe den): fa cies in ter pre ta tion based on sta - MUNNECKE A., SAMTLEBEN C., BICKERT T. (2003) — The Ireviken ble iso topes in brachio pod shells. Geol. Rundsch., 85: 278–292. Event in the lower Silu rian of Gotland, Sweden — re lation to simi lar URBANEK A. and TELLER L. (1997) — Graptolites and stra tig ra phy of Palaeozoic and Pro terozoic events. Palaeogeogr., Palaeoclimat., the Wenlock and Ludlow Se ries in the East Euro pean Platform . Palaeoecol., 195: 99–124. Palaeont. Pol., 56: 23–57. MUSTEIKIS P. and PAŠKEVIÈIUS J. (1999) — Brachio pod com mu nities VEIZER J., GODDERIS Y. and FRANCOIS L. M. (2000) — Evi dence for of the Lith u a nian Si lu rian. In: Paleocommunities: a Case Study from de coup ling of at mo spheric CO2 and global clim ate dur ing the the Si luria n and Lower Devo nian (eds. A. J. Boucot and J. D. Law son). Phanerozoic eon. Na ture, 408: 698–701. World Reg. Geol. Ser., 11: 305–326. Cam bridge Univ. Press. Cam - VIIRA V. (1999) — Late Si luria n conodont biostratigraphy in the northern bridge. East Bal tic. Bollettino Soc. Paleont. Italiana, 37: 299–310. NESTOR H. (1997) — Silu rian. In: Ge ol ogy and Min eral Re sources of Es - WEISSERT H. (2000) — Deci pher ing methane’ s finger print . Na ture, 406: tonia (eds. A. Raukas and A. Teedumäe): 89–106. Esto nian Acad. 356–357. Publ. Tallinn. WENZEL B. and JOACHIMSKI M. M. (1996) — Car bon and oxy gen iso- NESTOR H. and EINASTO R. (1997) — Ordo vi cian and Silu ria n car bon - to pic com po si tion of Si lu rian brachi o pods (Gotland/Swe den): ate sed i men ta tion ba sin. In: Ge ol ogy and Min eral Re sources of Es to- palaeoceanographic im pli ca tions. Palaeogeogr., Palaeoclimat., nia (eds. A. Raukas and A. Teedumäe): 192–204. Esto nian Acad. Publ. Palaeoecol., 122: 143–166. Tallinn. WIGFORSS-LANGE J. (1999) — Carbon isotope 13C enric hm ent in Up - PAŠKEVIÈIUS J. (1997) — The geol ogy of the Baltic repub lics . Vilnius per Si lu rian (Whitcliffian) marine calcar e ous rocks in Scania, Sweden. Univ. Geol. Surv. Lith u a nia. Vilnius. GFF, 121: 273–279.