Integrated stratigraphy and clay mineralogy of the Owadów–Brzezinki section (Lower–Upper Tithonian transition, central Poland): implications for correlations between the Boreal and the Tethyan domains and palaeoclimate

The Owadów–Brzezinki palaeontological site located near Tomaszów Mazowiecki in the NW margin of the Holy Cross Mts (Fig. 1) is one of the most important recent palaeontological discoveries from Poland (Kin et al. 2013; Błażejowski et al. 2016). Unusually well preserved fossils of marine and terrestrial organisms of Late Jurassic (Tithonian) age, many of them new to science, provide a good opportunity for studying the taphonomy of the ecosystem, evolution and migration of taxa, and palaeoenvironmental changes (cf. Kin and Błażejowski 2014; Błażejowski et al. 2016, 2019; Wierzbowski et al. 2016). The section (c. 30 m thick) has been studied for biostratigraphy, sedimentology and faunistic variations as well as stable carbon and oxygen isotopes and basic geochemical proxies, which has allowed reconstruction of its depositional settings (Kin et al. 2013; Matyja and Wierzbowski 2016; Pszczółkowski 2016; Wierzbowski et al. 2016, 2019).

The Owadów–Brzezinki section provides important clues for stratigraphical correlation between the NW Europe, Russian and Tethyan domains in the Tithonian, linking calpionellid occurrences (a typical Tethyan stratigraphic proxy) with the well established, British and Russian zonal schemes based on ammonites (Matyja and Wierzbowski 2016). The stratigraphical subdivision of the Upper Tithonian in the Tethyan domain is mostly based on integration of calpionellid and calcareous nannofossil stratigraphy in relation to ammonite stratigraphy and magnetostratigraphy (e.g. Pruner et al. 2010; Grabowski et al. 2019; Casellato and Erba 2021; Szives and Fözy 2022). In contrast, outside Tethys, calibration of biostratigraphical and magnetostratigraphical zonal schemes has been only partially established (Houša et al. 2007; Bragin et al. 2013). Calpionellids and Tethyan-type calcareous nannofossils do not occur in the Boreal realm (Hesselbo et al. 2021). Magnetostratigraphic data are often scarce and of moderate quality (e.g. Ogg et al. 1994; Schnyder et al. 2012) and not available from well dated, ammonite-bearing marine sequences (e.g. Hesselbo et al. 2009). In addition, they are calibrated using floating timescales (Huang 2018). However, the Boreal sections provide consistent palaeoclimatic trends with humid/arid intervals (Abbink et al. 2001; Schnyder et al. 2006, 2012; Hesselbo et al. 2009; Schneider et al. 2018), which may also be identified in the Tethyan realm (Schnyder et al. 2005; Grabowski et al. 2017).

The aim of the present paper is to refine the biostratigraphy of the Owadów–Brzezinki section, based on new discoveries of chitinoidellids, and to provide new data on its magnetostratigraphy and inter-regional correlations, especially with the Tethyan domain. The reliability of the palaeomagnetic signal is evaluated on the basis of geochemical proxies that show the effect of diagenesis on the magnetic component of the rocks. In addition, clay minerals and chemical proxies of weathering are studied to reconstruct palaeohumidity variations and their timings against palaeoclimatic changes documented from NW Europe.

The uppermost part of the Brzostówka Marl Member (Mb) of the Pałuki Formation (Fm) and the overlying limestones of the Kcynia Fm, including the Sławno Limestone Mb, ‘Corbulomima limestones’ and a fragment of ‘serpulid’ beds, are exposed in the Owadów–Brzezinki section (Kutek 1994; Matyja and Wierzbowski 2014, 2016; Fig. 1). The section shows a gradual marine regression revealed by a transition from offshore to coastal and lagoonal settings but its uppermost part was deposited during a short-term marine transgression and the re-appearance of coastal environments (Błażejowski et al. 2016; Wierzbowski et al. 2016).

The uppermost part of the Brzostówka Marl Mb of the Pałuki Fm from the Owadów–Brzezinki quarry (c. 4 m thick) consists of black, blue-greyish and yellow-bluish marls with the intercalation of thin oyster-bearing and marly limestone beds (Błażejowski et al. 2016; Wierzbowski et al. 2016; Fig. 1a, c). The marls yielded abundant marine microfossils, bivalves and ammonites (Figs 1–5). The overlying limestones of the Kcynia Fm have been subdivided into four lithological units (cf. Salamon et al. 2006; Kin et al. 2013; Błażejowski et al. 2016). Thick-bedded chalky and micritic limestones of the unit I and unit II, which contain marine fauna, were included into the Sławno Limestone Mb by Matyja and Wierzbowski (2016). Overlying well-bedded micritic limestones of unit III, which contain common Corbulomima bivalves and were formed in a lagoonal environment, are, informally, called the ‘Corbulomima limestones’ (Fig. 1) according to the earlier classification (Kutek 1994). A narrow interval of unit IV from the uppermost part of the Owadów–Brzezinki section consists of oyster–bryozoan–serpulid organodetrital limestones assigned to the ‘serpulid’ beds, which were formed in coastal settings (Matyja and Wierzbowski 2016; Wierzbowski et al. 2016). Units III and IV of the Owadów–Brzezinki section have characteristics of the marine-brackish Purbeckian facies widely distributed in NW Europe during the latest Jurassic (Marek et al. 1989; Hallam et al. 1991; Schnyder et al. 2012). The marine ‘serpulid’ beds of central Poland pass gradually upwards into brackish and evaporate deposits, known from boreholes, which is a result of long-term marine regression and progressive isolation of the restricted, epicontinental, mid-Polish Basin (Dembowska 1973; Grabowski et al. 2021).

According to the stratigraphical studies of Kutek (1994) and Matyja and Wierzbowski (2016) based on ammonite fauna (Fig. 1), the lower part of the Owadów–Brzezinki deposits is dated to the regularis horizon (the uppermost part of the Brzostówka Marl Mb of the Pałuki Fm) and zarajskensis horizon (unit I of the Sławno Limestone Mb of the lowermost part of the Kcynia Fm) of the Zarajskensis Subzone of the Scythicus (Panderi) Zone of the Middle Volgian, as well as to the Fittoni Zone from the ‘Bolonian’ zonation of England (Matyja and Wierzbowski 2016). The upper part of the Owadów–Brzezinki section (units III and IV belonging to the ‘Corbulomima limestones’ and ‘serpulid’ beds, respectively) has, in turn, been assigned to both the Gerassimovi Subzone of the Virgatus Zone of the Middle Volgian and the Albani Zone of the ‘Portlandian’ (Matyja and Wierzbowski 2016).

Chitinoidellids were previously found in the middle part of the Sławno Limestone Mb of the Kcynia Fm in the middle part of the Owadów–Brzezinki section (Matyja et al. 2016; Pszczółkowski 2016). They occur at the top of the chalky limestones of the unit I of the Sławno Limestone Mb, within an 0.3 m thick interval (Fig. 1a). Therefore, the Chitinoidella Zone was very thin compared to the whole quarry section. Examination of the uppermost part of the Brzostówka Marl Mb has, however, revealed the presence of chitinoidellid internal moulds (Figs 1a, 2–4 and 5j) as well as ostracods and foraminifers (Fig. 5).

Organisms typical of different marine bioprovinces indicate that the Owadów–Brzezinki site represents a palaeobiogeographical link due to its location within an open marine passage between the Tethyan and northern Boreal/Subboreal realms (Błażejowski et al. 2016; Matyja and Wierzbowski 2016). The appearance of chitinoidellids points to the episodic connection of the study area with the Mediterranean Province, which was partially blocked by the Štramberk-type coral reef barrier (Matyja 2009). Fossil insects and turtles found in Owadów–Brzezinki have Submediterranean–Mediterranean affinities (Szczygielski et al. 2017).

The positioning of the Lower and Upper Tithonian boundary at the base of the Microcanthum ammonite Zone and near the base of the M20n magnetochron was applied according to uppermost Jurassic biostratigraphical correlations of Wierzbowski et al. (2017) and Hesselbo et al. (2021).

Samples for biostratigraphy (BB 1-5 and B in Fig. 1) were mechanically disintegrated using a liquid nitrogen method (Remin et al. 2012), cleaned in an ultrasonic bath and subsequently sieved to obtain the 63–600 $μ$m fraction. Foraminifera and ostracoda specimens were picked from the fraction and analysed for the taxonomy and abundance of taxa using optical and scanning electron microscopes.

The analysis and microphotographs of the chitinoidellids from thin sections were made with a NIKON Polarizing Microscope (ECLIPSE LV100POL) at the Institute of Geological Sciences (Research Centre in Warsaw) of the Polish Academy of Sciences.

Twenty seven oriented hand samples (Fig. 6) were taken for palaeomagnetic investigations, with spacing of c. 0.5 to 1 m. Core specimens of 2.5 cm diameter and 2.1 cm length were drilled from each hand sample, three to seven specimens from each. Palaeomagnetic and rock-magnetic investigations were carried out at the palaeomagnetic laboratory of the Polish Geological Institute – National Research Institute in Warsaw. Natural remanent magnetization (NRM) and other laboratory induced magnetization parameters were measured using a JR6A spinner magnetometer (sensitivity 5 × 10⁻⁶ A m^–1). Thermal demagnetization was performed in a MMTD1 non-magnetic oven and reversing tumbling alternating field (AF) demagnetization using a Molspin device (peak demagnetization field 100 mT). Magnetic susceptibility (MS) was measured using a KLY2 kappabridge and mass-normalized.

Rock magnetic investigations included acquisition of anhysteretic remanent magnetization (ARM) and isothermal remanent magnetization (IRM). They were conducted on the 27 specimens collected for magnetostratigraphy (OB sample collection), as well as on a set of 71 rock chips cut from geochemical samples studied by Wierzbowski et al. (2016, SHW collection). Static ARM was induced in a Molspin device with peak alternating field of 100 mT, a steady field bias of 0.1 mT and the decay time of the field of 43 s. The IRM was measured along the specimen Z axis in a field of 1 T, and then antiparallel in a field of 100 mT (both using MMPM10 pulse magnetizer). The S-ratio was calculated as IRM_-100mT/IRM_1T, indicative of the proportions of low- and high-coercivity minerals (e.g. Opdyke and Channell 1996). Positive values of the S-ratio are characteristic of low-coercivity minerals, while more negative S-ratio values indicate an admixture with a high-coercivity fraction. For representative samples with different S-ratios, stepwise acquisition of IRM (maximum field of 1.4 T), followed by thermal demagnetization of a three-axis IRM acquired in 1.4, 0.4 and 0.1 T pulse fields (Lowrie 1990), was performed in order to identify magnetic minerals through their unblocking temperatures.

Additional interpretation of the geochemical data of Wierzbowski et al. (2016) was carried out. We focus on the composition of the lithogenic fraction, considering Ti/K and K/Al ratios (proxies of chemical weathering), Zr/Rb and Ti/Al ratios (grain-size proxies) and Ca/Ti ratios (a measure of terrigenous input) of the rocks (see Dypvik and Harris 2001; Calvert and Pedersen 2007; Arnaud et al. 2012; Bassetti et al. 2016; Thöle et al. 2020).

The U/Th, Fe/Al and Mo/Al ratios are also discussed as palaeoredox indicators (Algeo and Liu 2020). The significance level of correlation between geochemical and magnetic proxies was checked using tables in Rollinson (1993).

The <2 µm clay fraction of 26 samples of carbonates from the Owadów–Brzezinki section (OB collection) were analysed using X-ray diffraction. After grinding, samples were decarbonated with a 0.2N HCl solution. The clay fraction was extracted with a syringe after deflocculation and decantation of the suspension for 95 minutes following Stokes’ law; this fraction was then centrifuged. The clay residue was smeared on oriented glass slides and analysed using a Bruker D8 diffractometer with CuK_$α$ radiations, a LynxEye detector and a Ni filter, at a voltage of 40 kV and an intensity of 25 mA (Biogéosciences laboratory, Université Bourgogne/Franche-Comté in Dijon, France). Goniometer scanning ranged from 2.5° to 28° for each analysis. Three runs were performed for each sample to discriminate the clay phases: (1) air-drying; (2) ethylene-glycol solvation at room temperature for 24 hours; and (3) heating at 490°C for 2 hours, as recommended by Moore and Reynolds (1997). Clay minerals were identified using their main diffraction (d_00l) peaks compared in the three diffractograms. Relative proportions of the clay minerals are estimated using peak intensity ratios (error margin of the method is c. ±5%).

Chitinoidellids occur in the blue-grey silty marls above the oyster layer (Fig. 1a) in the uppermost part of the Brzostówka Marl Mb of the Pałuki Fm (cf. Błażejowski et al. 2016). The chitinoidellid taxa identified in thin sections from sample B (Fig. 1a) are as follows:

Bonetilla elongata (Pop, 1997) n. comb. Benzaggagh, 2021 – Figure 2d, e 

Bonetilla cf. sphaerica Benzaggagh, 2021 – Figure 3b 

Bonetilla sp. ex gr. B. boneti-carthagensis – Figure 2a 

Bonetilla sp. ex gr. B. svinitensis-elongata – Figure 3f 

Borziella slovenica (Borza, 1969) – Figure 3a 

Borziella cf. slovenica (Borza, 1969) – Figure 2c 

Borziella cf. tithonica (Borza, 1969) n. comb. Benzaggagh, 2021 – Figure 3e 

Carpathella cf. longirumanica Benzaggagh, 2021 – Figure 3g 

Daciella banatica Pop, 1998 amend. Benzaggagh, 2021 – Figure 3d 

Daciella cf. banatica Pop, 1998 amend. Benzaggagh, 2021 – Figure 2i 

Daciella cf. danubica (Pop, 1998) amend. Benzaggagh, 2021 – Figure 2f–h 

Dobenilla dobeni (Borza, 1969) var. dobeni Benzaggagh, 2021 – Figure 3i 

Dobenilla colomi (Borza, 1966) var. longicolomi Benzaggagh, 2021 – Figure 3h 

Furrazolaia cristobalensis (Furrazola-Bermúdez, 1965) n. comb. Benzaggagh, 2021 – Figure 2b 

Scarce chitinoidellids identified in thin sections made from the samples BB2-BB5 (Fig. 1a) are as follows:

?Bonetilla sp.

Borziella cf. slovenica (Borza, 1969)

Daciella danubica Pop, 1998 amend. Benzaggagh, 2021 – Figure 3c 

Daciella cf. danubica Pop, 1998 amend. Benzaggagh, 2021 

Daciella sp. ex gr. D. banatica-danubica

Dobeninae gen. et sp. indet.

Chitinoidellids are sometimes poorly preserved. Their internal moulds are frequently observed in the studied samples. Other microfossils have been identified as calcareous dinoflagellate cysts and didemnid spicules:

Pirumella multistrata (Pflaumann and Krasheninnikov, 1978) Lentin and Williams, 1993 (Cadosina semiradiata semiradiata Wanner, 1940) – Figure 4a 

Pirumella multistrata (Pflaumann and Krasheninnikov, 1978) Lentin and Williams, 1993 (Cadosina cf. semiradiata semiradiata Wanner, 1940)

Pirumella multistrata (Pflaumann and Krasheninnikov, 1978) Lentin and Williams, 1993 (Cadosina semiradiata fusca Wanner, 1940) – Figure 4b 

Pirumella cf. multistrata (Pflaumann and Krasheninnikov, 1978) Lentin and Williams, 1993 (Cadosina cf. semiradiata fusca Wanner, 1940)

Pirumella piriformis (Keupp 1977) Lentin and Williams, 1993 (Carpistomiosphaera cf. tithonica Nowak 1968) – Figure 4g 

Pirumella cf. piriformis (Keupp 1977) Lentin and Williams, 1993 (Carpistomiosphaera cf. tithonica Nowak 1968) – Figure 4h 

Pirumella thayeri (Bolli, 1974) Lentin and Williams, 1993 (Colomisphaera lapidosa [Vogler, 1941]) – Figure 4c 

Pirumella thayeri (Bolli) Lentin and Williams (or Colomisphaera tenuis [Nagy]) – Figure 4d, e 

Pirumella sp. (Cadosina sp.)

Orthopithonella gustafsonii (Bolli, 1974) – Figure 4f 

Didemnoides moreti (Durand-Delga, 1957) – Figure 4i 

Didemnum sp.

Ostracod fauna from the uppermost part of the Pałuki Fm (samples BB4-BB5, Fig. 1) is represented by a relatively low-diversity assemblage restricted to Galliaecytheridea compressa (Partial Range Zone) ranging from the mid-Tithonian (the upper part of the Fittoni ammonite Zone) to the lower part of the Upper Tithonian (the Okusensis ammonite Zone of Britain; see Wilkinson 2008; Wilkinson and Whatley 2009). It is composed of three species belonging to Cytheroidea: Macrodentina sp. aff. M. (Polydentina) rudis Malz (Fig. 5a–c), Galliacytheridea sp. (Fig. 5d) and Galliacytheridea compressa Christiensen and Kilenyi (Fig. 5e). Galliaecytheridea compressa occurs in various parts of Europe in the mid- to Upper Tithonian (between uppermost Fittoni Zone and Albani to Glaucolithus ammonite zones) (Wilkinson and Whatley 2009). Macrodentina (Polydentina) rudis occurs within the same stratigraphical interval (Schudack 2004; Wilkinson and Whatley 2009). All species are fully marine taxa (Wilkinson and Whatley 2009).

The Pałuki Fm yielded a relatively low-diversity foraminifera fauna similar to that from the Kcynia Fm, with most abundant species of Lenticulina sp. (Fig. 5f), Paleomiliolina sp. (Fig. 5i), and Cornuspira sp. (Fig. 5g, h), also representatives of genus Marginulopsis, Tristix and others (see Wierzbowski et al. 2016).

The rocks have NRM intensities below 10⁻⁴ A/m (Fig. 6d) and a general upward-decreasing trend of MS values is observed (Fig. 6a). Positive values of MS are only observed in the lowermost part of the section comprising the Pałuki Fm and lower beds of Unit I of the Kcynia Fm – with MS up to 12 × 10⁻⁹ m³ kg^–1. Most younger samples revealed MS values close to 0 or slightly negative. Correlation between MS, ARM and IRM_1T is positive; however MS correlates better with ARM (Pearson r coefficient = 0.72) than with IRM_1T (r = 0.47) (Supplementary Fig. 1). The degree of correlation between magnetic parameters is different in two intervals of the section. The correlation is highly positive in the lower part of the section (0–11 m) and only moderate for the remaining upper part of the section 11–29 m (Fig. 7a–c).

The largely positive values of the S-ratio indicate a predominance of low-coercivity magnetic minerals (i.e. magnetite). High-coercivity minerals shown by the lower S-ratios occur in the uppermost part of the section (Fig. 6b). The presence of probable magnetite (saturation by c. 300 mT, maximum unblocking temperatures below 600°C) is confirmed by stepwise acquisition of the IRM and thermal demagnetization of the three-axis IRM, for samples with S-ratio between 0.6 and 0.9 (Supplementary Fig. 2A, B). The sample OB24 (S-ratio = 0.49) shows the presence of hematite because of a large increase of IRM between 400 and 1400 mT and maximum unblocking temperatures above 600°C (Supplementary Fig. 2C). An increased magnetization above 350°C was observed in this sample.

The NRM of six pilot specimens were demagnetized both thermally and with AF. Since the results obtained using both methods were similar, the rest of the collection was demagnetized using AF. The quality of demagnetization behaviour was different between the lower and upper part of the section. In the lower 11 m (up to sample OB9) one component of normal polarity was revealed at 35–40 mT or 350°C (Fig. 8a, b; Table 1). In the rest of the section, especially in unit III, the magnetization was mostly unstable (Fig. 6c). Exceptions were samples OB10 and OB13 from the uppermost part of unit I and the lowermost part of unit III of the Kcynia Fm, which revealed a reversed polarity component, until 70 mT or 400°C (Fig. 8c, d) and specimens from sample OB 24, which were again resistant to AF demagnetization (almost no NRM intensity decrease until 60 mT).

The quality and reliability of polarity interpretations were assessed based on the following criteria. High-quality specimens (like those presented in Fig. 8) reveal consistent demagnetization behaviour (in at least two twin specimens) and a characteristic direction, which could be calculated using a fitted line. Poorer-quality samples reveal more chaotic demagnetization directions; however their polarity is still possible to estimate as directions cluster or wander along great circle paths close to the ‘expected’ position on the stereonet (Table 1). The polarity is rated as unknown when contradictory trends appear in two specimens from the same sample, or the direction is clearly anomalous, deviating from the ‘expected’ position of the palaeomagnetic direction (e.g. shallow inclination).

The normal and reversed directions measured from high-quality samples group together with northerly or southerly declinations near the expected position of the palaeopole for the Tithonian of the European Platform (Torsvik et al. 2012; see Fig. 6g and Table 1) and Purbeckian deposits in England (Ogg et al. 1994).

An upward-decreasing lithogenic input and increasing carbonate content is observed throughout the Scythicus (Panderi) and Virgatus Zones based on concentrations of Al, Ti, Zr and Ca (Fig. 9a, b, see also fig. 13 in Wierzbowski et al. 2016). The Ca/Ti ratio is low in the Pałuki Fm and units I and II of the Kcynia Fm (between 0 and 2 × 10³) and elevated in units III and IV (up to 14 × 10³, Fig. 9b). The lower 13.5 m part of the section, including the Pałuki Fm and unit I of the Kcynia Fm is relatively enriched in Ti and Zr and depleted in K and Rb. This is shown by elevated Ti/K, Ti/Al and Zr/Rb ratios and decreased K/Al ratios for this interval (Fig. 9c–f). Conversely, units II–IV of the Kcynia Fm have mostly low Ti/K, Ti/Al and Zr/Rb ratios and an elevated K/Al ratio. In the uppermost part of the unit III, again elevated values of Ti/K, Ti/Al and Zr/Rb ratios are observed. A sharp decrease of the Ti/K ratio, from 0.25 to 0.15 occurs at the boundary between the units I and II (Fig. 9c). The boundary is also marked by an increase of the K/Al ratio (from 0.25 to 0.35, Fig. 9d) and a sharp decrease of the Ti/Al ratio (from 0.065 to 0.045) (Fig. 9e). The Zr/Rb ratio reveals a more stepwise decrease between the uppermost part of the unit I and the middle part of the unit III, from 0.9 to 0.6 (Fig. 9f).

Despite the positive linear correlation between Al and Fe (r = 0.73. Wierzbowski et al. 2016) the Fe/Al ratio fluctuates between 0.5 and 8, largely following the U/Th and Mo/Al trends (Fig. 10a).

The clay mineral assemblages are dominantly composed of illite (20 to 63%, average 31.5%), random illite–smectite mixed layers (I–S) (10 to 65%, average 28.4%) and kaolinite (12 to 60%, average 40.3%) (Fig. 10c). Clay minerals are systematically associated with small amounts of quartz. Iron oxy-hydroxides including goethite and lepidocrocite commonly occur in the clay fraction.

The section can be divided into two parts. In the lower part, including the Pałuki Fm and unit I of the Kcynia Fm, the clay assemblages are rich in kaolinite (>40%) while in the upper part of the section, including units II and III of the Kcynia Fm, kaolinite does not exceed 35% (Fig. 10c). The upsection decrease of kaolinite is mainly balanced by an increase in the proportion of I–S. In addition, in the lower part of the section, I–S is relatively rich in illite (R1 type), whereas in the upper part of the section I–S is richer in smectite sheets (R0 type). Supplementary Figure 3 shows the typical X-ray diagrams of the two parts of the section.

The former Chitinoidella Zone (Borza 1984; Olóriz et al. 1995; Pop 1997; Reháková 2002) was substituted for the Chitinoidellidae Zone because the genus Chitinoidella Doben, 1963 was replaced by Bonetilla n. gen. (Benzaggagh 2021). The chitinoidellid taxa found in the uppermost part of the Pałuki Fm (Fig. 1a, sample B) belong to the subfamily Bonetinae and occur in the Upper Tithonian Boneti Subzone of the Chitinoidellidae Zone (Benzaggagh 2021). This is supported by the presence of: Bonetilla elongata (Pop, 1997) n. comb. Benzaggagh, 2021, Bonetilla cf. sphaerica Benzaggagh, 2021, Bonetilla sp. ex gr. B. boneti-carthagensis, Bonetilla sp. ex gr. B. svinitensis-elongata and Furrazolaia cristobalensis (Furrazola-Bermúdez, 1965).

The chitinoidellid-bearing horizon identified in the uppermost part of the Pałuki Fm along with the horizon occurring at the top of the unit I of the Sławno Limestone Mb of the Kcynia Fm (cf. Matyja et al. 2016; Pszczółkowski 2016) allows us to designate the 9.7 m thick Bonetilla boneti Subzone in the Owadów–Brzezinki section (Fig. 1a). Therefore, the interpreted Boneti Subzone comprises the uppermost part of Brzostówka Marl Mb of the Pałuki Fm and likely all of unit I of the Sławno Limestone Mb of the Kcynia Fm, although specimens of Chitinoidellidae have been found so far in its uppermost part only. The chalky limestones of unit I contain also echinoderm debris with Saccocoma sp. skeletal fragments (Pszczółkowski 2016).

Other specimens of Chitinoidellidae found in the samples BB-2 to BB-5 (Fig. 1a) from the upper part of the Pałuki Fm (located below the sample B; Fig. 1a) belong to the subfamily Dobeninae Benzaggagh, (2021). Only one specimen identified as ?Bonetilla sp. in the sample BB-3 might represent the subfamily Bonetinae Benzaggagh (2021). Therefore, it is probable, although it could not be firmly established, that the Dobeni/Boneti subzonal boundary occurs between samples BB-2 and BB-3 (about 0.5 m below the top of the Pałuki Fm, Fig. 1a).

The stratigraphical range of calcareous dinoflagellate cysts found in the samples (Fig. 4a–h) is consistent with the range of the Boneti Subzone, as Cadosina semiradiata semiradiata and Cd. semiradiata fusca occur from the base of the ‘middle’ Tithonian up to the Albian (Reháková 2000). Both taxa have been included in the synonymy of Pirumella multistrata (Pflaumann and Krasheninnikov 1978) Lentin and Williams, 1993 (Ivanova and Keupp 1999). The taxon Colomisphaera lapidosa was recorded from the Oxfordian up to the Valanginian (Borza 1969) and was included in the synonymy of Pirumella thayeri (Bolli 1974) Lentin and Williams 1993 (Ivanova and Keupp 1999). Also the taxon Colomisphaera tenuis (Nagy), recorded starting from the Lower Tithonian up to the Berriasian, was transferred to the synonymy of Pirumella thayeri (Ivanova and Keupp, 1999). In the biostratigraphical subdivision of the Tithonian Stage, based on calcareous dinocysts, the Colomisphaera tenuis Zone was correlated with the Chitinoidella Zone (Lakova et al. 1999) or with its upper part (Reháková 2000), or even directly with the Boneti Subzone (Jach et al. 2014).

The microfossils found in the deposits of the Brzostówka Marl Mb of the Pałuki Fm and the Sławno Limestone Mb of the Kcynia Fm confirm the conclusion of Wierzbowski et al. (2016, p. 81) concerning the normal marine conditions during the sedimentation of this part of the Owadów–Brzezinki section. Considering the ammonite zonation established for the Owadów–Brzezinki quarry (Matyja and Wierzbowski 2016), the Boneti Subzone correlates with the uppermost part of the regularis horizon and the zarajskensis horizon (distinguished in the uppermost part of the Pałuki Fm and Unit I of the Sławno Limestone Mb, respectively) of the Zarajskensis Subzone of the Scythicus (Panderi) Zone. This apparently confirms the finding of Nowak (1971) who documented the occurrence of genus Zaraiskites together with Chitinoidella boneti. The Boneti Subzone also correlates with the upper part of the Fittoni Zone (Matyja and Wierzbowski 2016).

The ostracod findings (section: Ostracoda and Foraminifera) are in a good agreement with the ammonites, placing the boundary interval of the Pałuki and Kcynia Fms between the upper part of the Fittoni and Okusensis ammonite zones, exactly between the uppermost Lower Tithonian and the lower part of the Upper Tithonian (Hesselbo et al. 2021).

The correlation between magnetic and terrigenous proxies is usually applied to test whether magnetic minerals are of detrital origin (e.g. Riquier et al. 2010; Da Silva et al. 2012; Grabowski et al. 2019). MS, ARM and IRM_1T correlate strongly with terrigenous proxies (e.g. Ti content, see Fig. 7d–f; Al content, see Supplementary Fig. 4) in the lower part of the section, 0–11 m (r₁ = 0.97, 0.85 and 0.84, respectively, Fig. 7d–f), so the magnetic carriers of this interval are mostly related to clayey input. Conversely, magnetic proxies from the upper part of the section (11–29 m) reveal moderate to weak correlation with Ti (r₂ = 0.65 for MS, 0.42 for ARM and only 0.15 for IRM_1T, Fig. 7d–f). The contrasting correlations of ARM v. Ti and IRM v. Ti are not related to different Ti content of ferromagnetic minerals in both parts of the section, since correlations of ARM and IRM_1T with Al reveal the same features (Supplementary Fig. 4). It should be rather concluded that the upper part of the section is enriched in ferromagnetic minerals of non-detrital (authigenic, biogenic or weathering) origin.

High values of S-ratio indicate that magnetite is a dominant magnetic mineral, although its variations indicate an admixture of a high-coercivity phase. The values of S-ratio below 0.5 might be related to the presence of hematite (Supplementary Fig. 2C). Fluctuations of S-ratio between 0.6 and 0.9 might indicate grain-size differentiation of magnetite (e.g. Channell et al. 2013), since the presence of hematite is not proved (Supplementary Fig. 2A, B). Samples with low S-ratio occur in the uppermost part of the section (above 23.5 m) and apparently correlate with increased values of redox proxies: U/Th and Fe/Al ratios (Fig. 10a). This suggests a secondary (diagenetic) origin of hematite, likely related to weathering (e.g. Elmore et al. 2012) and decomposition of pyrite (e.g. Heller and Channell 1979; Johnson et al. 1984). An increase of IRM intensity above 350°C during heating (Supplementary Fig. 2C) might also be a result of the presence of remnant iron sulfides or siderite and their transformation into iron oxides (Henry 2007).

Palaeomagnetic and magnetostratigraphic interpretations pose some problems. Reversed polarity components of three samples (OB10, OB13 and OB24, Fig. 6c) reveal higher coercivity (Fig. 8c, NRM stable until 70 mT) than the normal polarity component from the lowermost part of the section (Fig. 8a, NRM stable untill 40 mT), which indicates the NRM is carried by different magnetic minerals. This might be related to a weathering event prior to the Brunhes epoch when the succession might have been exhumed (cf. Heller 1978; Johnson et al. 1984). Van Velzen and Zijderveld (1995) documented that weathering seriously affects coercivity of primary magnetite grains, which is higher in significantly weathered samples and associated with secondary magnetization components of stronger stability than depositional magnetization. Relatively less stable normal magnetization from the lower part of the section is interpreted as primary, depositional (detrital) remnant magnetization since magnetic proxies correlate there well with clay input (Fig. 7d–f, Supplementary Fig. 4). Therefore, its magnetostratigraphic record (up to the sample OB9) is interpreted as magnetosubzone M20n.2n, which is consistent with the typical occurrence of B. boneti (Figs 6, 11). The lagoonal environment of unit III was most probably unfavourable for the preservation of primary magnetization and magnetic minerals, similarly to the lowermost Purbeckian deposits studied by Ogg et al. (1994).

The presence of I–S R0 points to low thermal alteration of deposits during their burial, since these minerals transform into I–S R1 and then into illite as soon as the burial temperature reaches 60–70°C (Środoń et al. 2009). This is consistent with the low thermal maturity of the Upper Jurassic rocks of the central part of the Polish Basin deduced from low T_max values of kerogen, which are placed above the oil window, and from the moderate burial depth of the Upper Jurassic strata (Dellisanti et al. 2010; Resak et al. 2010; Łuszczak et al. 2020).

The formation of authigenic minerals is often related to the enhanced rock porosity. This is observed in the case of clay/sandstone alternations, where authigenic kaolinite appears in sandstone beds, or in marl/limestone alternations, where diagenetic chlorite occurrences are noted in limestone beds. Therefore, diagenetic impact can be revealed by a possible link between clay mineralogy and lithology (Chamley 1989). In the studied section, we do not observe any consistent relationship between the clay mineralogy and major lithological changes (Fig. 10b, c). Therefore the clay-mineralogical variations including the decreasing rate of kaolinite, balanced by increasing rate of smectite, do not appear to be related to diagenetic alteration.

Consequently, the clay minerals identified are inferred to be mainly detrital and their fluctuations can be interpreted in terms of environmental change.

The upward transition from a clay assemblage largely dominated by kaolinite to another one characterized by the high abundance of I–S R0 suggests a change from humid to drier climatic conditions. Indeed, kaolinite forms in hot and regularly humid climates, while I–S forms rather in hot climates with contrasting seasonal humidity (Chamley 1989; Ruffell et al. 2002a). This mineralogical change coincides with a significant decrease in Ti/Al and Ti/K ratios (see below) and is consistent with the onset of a drier climate at the transition between the Scythicus (Panderi) Zone and the Virgatus Zone. The change in clay mineral assemblage coincides with a deterioration of oxygen availability manifested by increased U/Th and Fe/Al ratios (Fig. 10a) and slight increase in carbonate content (Wierzbowski et al. 2016 and Fig. 9b).

The humid to dry climate transition is also observed in southern England and northern France, throughout the Rotunda/Fittoni and Albani ammonite biochrons where the proportions of kaolinite decrease sharply in favour of I–S R0 (Hesselbo et al. 2009; Turner and Huggett 2019). Coeval aridification is also recorded in the Volga Basin by the occurrence of palygorskite (Ruffell et al. 2002b). According to Matyja and Wierzbowski (2016) the transition between the Scythicus (Panderi) Zone and the Virgatus Zone is correlated with the transition between the Fittoni and Albani Zones from the NW of the sub-Boreal province. Therefore, the onset of the latest Jurassic aridification phase seems to be almost synchronous in mid-latitudes of the northern hemisphere (see below).

The geochemical record of the Owadów–Brzezinki section can be stratigraphically correlated with the record of the Kcynia IG 2 borehole section from the north-central part of the Polish Basin (Grabowski et al. 2021). The lower part of the Kcynia IG 2 borehole section is assigned to the Pałuki Fm. The lowest part of the Kcynia Fm from the Kcynia IG 2 borehole section covers the uppermost part of the Lower Volgian (Puschi Zone) and the lower part of the Middle Volgian (Scythicus/Panderi) Zone and probably a part of the Virgatus Zone, see Fig. 12). Although the boundary between the Lower and Middle Volgian in the Kcynia IG 2 section is firmly established, the presence of the Virgatus Zone of the Middle Volgian is uncertain in this core (Grabowski et al. 2021). A general decrease of lithogenic input is observed throughout the uppermost Lower and the Middle Volgian in both Kcynia IG 2 and Owadów–Brzezinki sections (Fig. 12). The provenance area of fine-grained clay and silty material transported to the northern part of the Polish basin is disputable; however, it may be derived from the Fennoscandian and Ringkøbing–Fynn Highs (Bembenek et al. 2021). It was likely different from the source area of Upper Jurassic–Lower Cretaceous clastic material accumulated in Southern England, which is linked to the Anglo–Brabant Massif (Sladen and Batten 1984).

A decrease in the Ti/K ratio occurs in the uppermost part of the Scythicus (Panderi) Zone of the lowermost part of the Middle Volgian in the Owadów–Brzezinki section (Fig. 9c), while in the Kcynia IG 2 section this change falls in the lower half of the Scythicus (Panderi) Zone, c. 15 m below the base of Corbulomima limestones (cf. Fig. 12 and Grabowski et al. 2021). The position of the Ti/K shift in both sections might, however, be coeval assuming that unit II in the Owadów–Brzezinki section still belongs to the Scythicus (Panderi) Zone and that a stratigraphical gap occurs between units I and II. A sudden change of ratios Ti/K, Ti/Al, K/Al and Zr/Rb and single peaks of Al, Ti and Zr content at boundary between units I and II of the Owadów–Brzezinki section (Fig. 9 and Wierzbowski et al. 2016) might point to the presence of a condensed interval at that location (cf. Fig. 9 and fig. 13 in Wierzbowski et al. 2016).

The decrease of Ti/K in the Kcynia IG 2 section was interpreted by Grabowski et al. (2021) as a result of climate aridification. It is comparable to a similar phenomenon reported from the uppermost part of the Kimmeridge Clay Fm, starting from the boundary of the Rotunda and Fittoni ammonite zones (Hesselbo et al. 2009). A fall of Ti/K ratio from the Owadów–Brzezinki section, accompanied by an increase of K/Al ratio (Fig. 9d), indicates an increasing amount of K retention in clays and feldspars due to decreasing intensity of chemical weathering (Calvert and Pedersen 2007).

Ti/Al and Zr/Rb ratios are usually interpreted as grain-size proxies. Ti resides in both clay mineral fraction (together with Al) and heavy minerals. Variations of Ti/Al ratio in clay fraction are typically interpreted as resulting from grain-size variations, since larger grains are associated with more intensive transport, the presence of heavy minerals and increasing quartz content, which results in an increase of both Ti/Al and Si/Al ratios (Calvert and Pedersen 2007). Rb is usually associated clay minerals and micas (Dypvik and Harris 2001; Bassetti et al. 2016) while Zr is enriched in heavy minerals commonly occurring in fine sand and silt fraction (Calvert and Pedersen 2007). Therefore the Zr/Rb ratio can also be used as grain-size indicator (Thöle et al. 2020). Its decrease might result from diminished transport efficiency either due to limited riverine flow (dependent on rain intensity) in the provenance areas or an effect of marine transgression correlating with maximum flooding surface (Thöle et al. 2020).

The interpretation of minimum values of Ti/Al and Zr/Rb ratios in the Owadów–Brzezinki section as a result of transgression is, however, inconsistent with the shallowing of the depositional environment, as decreasing trends of both grain-size proxies correlate with the onset of lagoonal environment (Kin et al. 2013; Wierzbowski et al. 2016 and Fig. 9e–g). These circumstances suggest that the climatic shift (major mineralogical change in Fig. 10) recorded in the Owadów–Brzezinki section and Kcynia IG 2 borehole section is likely associated with climate aridification (Grabowski et al. 2021). It corresponds to a gradual passage of the depositional settings of the Kcynia Limestone Fm in the Owadów–Brzezinki section from the offshore environment to coastal and lagoonal ones and a significant restriction of the mid-Polish basin (Dembowska 1973; Niemczycka 1997; Wierzbowski et al. 2016). This interpretation is supported by the decrease of kaolinite content (between units II and III), which slightly predates the change of Ti/K and Ti/Al ratios. A similarly positioned aridification trend from southern England started at the transition between the Rotunda and Fittoni ammonite biochrons, and possibly continued throughout the Fittoni–Albani biochron transition (Hesselbo et al. 2009). It may be linked to the major ‘Portlandian’ regression, which is well documented in the Albani Zone of the Boulonnais area of northern France (Herbin et al. 1995). In this area, the decreasing kaolinite/illite ratio, which occurs in the middle part of the ‘Assise de Croï’ Formation (Albani Zone according to Townson and Wimbledon, 1979), is probably coeval with the decrease of kaolinite/illite ratio recorded in the Polish Basin (Fig. 13). The Portlandian regression culminated with the widespread occurrence of ‘Purbeckian’ facies in NW and central European basins (e.g. Marek et al. 1989; Hallam et al. 1991; Schnyder et al. 2012). A slight increase of Ti/K, in the uppermost part of unit III (and into unit IV; Fig. 9) might be linked to another brief humid episode. Indeed this geochemical shift correlates with a short-term return of marine conditions (Wierzbowski et al. 2016) and an increase of grain-size proxies (Ti/Al and Zr/Rb). However, this interpretation is not supported by any changes in clay mineralogy, therefore the nature of this geochemical event remains uncertain. The correlation of the latest Jurassic/earliest Cretaceous humid and arid episodes from the Polish Basin with, respectively, transgressions and regressions has been already noted by Grabowski et al. (2021, see also Fig. 9g). This suggests a glacial or tectonoeustatic cause of oscillations in the Tithonian climate humidity and their link to sea-level variations (Kutek 1994; Price 1999; Brysch, 2018; Scotese et al. 2021).

A comparison of dating of the zonal boundaries in precisely calibrated Tethyan sections (Fig. 11) suggests, assuming some uncertainty, that the base of the Boneti Subzone is located either within the M20n.2n magnetosubzone (Puerto Escaño and Velykyi Kamianets) or slightly lower in the M20r magnetozone (Brodno and Lókút sections). As the Dobeni/Boneti subzonal boundary has not been found in the M21n magnetozone (Casellato and Erba 2021) the correlation of the normal polarity interval from the lower half of the Owadów–Brzezinki section with M20n.2n magnetosubzone is justified. The documented Boneti Subzone indicates the earliest Late Tithonian age of the lower part of the Owadów–Brzezinki section, starting from the uppermost beds of the Pałuki Fm up to the topmost part of unit I of the Sławno Limestone Mb (Fig. 1). Therefore, the upper part of the regularis horizon and the whole zarajskensis horizon of the Zarajskensis Subzone of the Scythicus (Panderi) Zone of the Middle Volgian (Matyja and Wierzbowski 2016) should be assigned to the Upper Tithonian probably as partial stratigraphical equivalents of the lower part of the Microcanthum ammonite Zone (Enay and Geyssant 1975; Zeiss 2003; Pruner et al. 2010; Főzy et al. 2011; Price et al. 2016; Szives and Fözy 2022). The Gerassimovi Subzone from the upper part of the Owadów–Brzezinki section (Matyja and Wierzbowski 2016) is also Late Tithonian in age. It should be placed, most probably, still within the Chitinoidella Zone and M20n.2n Subzone, as suggested by correlation with the Nordvik section (Fig. 14). The Nordvik section is the only Boreal section whose magnetostratigraphy can be correlated with Tethyan sections (Houša et al. 2007; Bragin et al. 2013). The Chitinoidella/Praetintinnopsella zonal boundary corresponds to the top of the Boneti Subzone and occupies a similar position within the M20n.2n magnetosubzone, just below the M20n.1r magnetosubzone, in all Tethyan sections studied previously (Fig. 11 and Casellato and Erba 2021). The M20n.1r magnetosubzone in the Nordvik section correlates with the Middle/Upper Volgian boundary which falls stratigraphically much higher than the top of the Owadów–Brzezinki section (Fig. 14).

The Late Tithonian age of the regularis and zarajskensis horizons of the Zarajskensis Subbiochron of the Scythicus (Panderi) Biochron was suggested earlier (Kutek 1994; Zeiss 2003), but their correlation with the Crassicollaria Biochron was questioned (Rogov 2004; Matyja and Wierzbowski 2016).

The Owadów–Brzezinki section can also be correlated with the Purbeckian succession in the NW Aquitaine Basin, France (Schnyder et al. 2012) and the ‘Bolonian/Portlandian' transition in southern England (Ogg et al. 1994). The Fittoni/Albani boundary interval in the Phare de Chassiron section falls in the normal polarity interval interpreted as M21n (Schnyder et al. 2012). However, as the interpreted Fittoni/Albani transition at Owadów–Brzezinki is correlated with M20n2n magnetosubzone, it is possible that the top of the Phare de Chassiron section also falls in this subzone. The Fittoni/Albani transition at the Phare de Chassiron records similar palaeoenvironmental changes as the Owadów–Brzezinki section, with regression and climate aridification, including the transition from fully marine conditions in the Pectinatus Zone, through the serpulids patch-reefs of the Fittoni Zone to the overlying evaporitic interval, observed at the Fittoni/Albani Zone boundary (Schnyder et al. 2012). Our data indicate that the Fittoni/Albani boundary (the boundary of informal ‘Bolonian’ and ‘Portlandian’ stages, or the top of ‘Kimmeridgian sensu anglico’) should be situated within the lower part of M20n.2n magnetosubzone or close to its base, similarly to the correlation presented by Hesselbo et al. (2021). Ogg et al. (1994) suggested that the base of the ‘Portlandian stage’ may be correlated with magnetozone M21r, which would imply correlation of the Fittoni/Albani zonal boundary with the Tethyan Richteri Zone (lower part of the present Fallauxi Zone). However, the Fittoni/Albani boundary has been subsequently moved higher, into the upper part of the Fallauxi Zone (Ogg 2004) or even to the Ponti/Peroni Zone (Ogg and Hinnov 2012), and nowadays it almost coincides with the base of the Microcanthum Zone (Hesselbo et al. 2021), which is in a very good agreement with our data.

1. Chitinoidellids from the uppermost part of the Pałuki Fm in the Owadów–Brzezinki section belong to the topmost part of the Dobeni and lower part of the Boneti Subzones of the Chitinoidellidae Zone.

2. The interpreted Boneti Subzone is about 9.7 m thick in the Owadów–Brzezinki section. The Boneti Subzone documents the earliest Late Tithonian age of the uppermost part (0.5 m thick) of the Brzostówka Marl Mb of the Pałuki Fm up to the topmost part of unit I of the Sławno Limestone Mb of the Kcynia Fm. Therefore, the upper part of the regularis and zarajskensis horizons of the Zarajskensis Subzone of the Scythicus (Panderi) Zone of the Middle Volgian should be assigned to the lower part of the Upper Tithonian, being probably coeval with the lowermost part of the Microcanthum Tethyan ammonite Zone. This part of the Owadów–Brzezinki section corresponds to the upper part of the Fittoni Zone.

3. Although the Owadów–Brzezinki section does not possess complete magnetostratigraphy, its lower, c. 11 m thick part (including the uppermost part of the Pałuki Fm and the lowermost part of the Kcynia Fm) has revealed consistent and stable normal polarity magnetization, which is interpreted as a record of the magnetosubzone M20n2n. This is consistent with the usually reported position of the Dobeni and Boneti Subzones in Tethyan sections. An interval with primary magnetization reveals a strong positive correlations between both ARM and IRM_1T and Ti and Al contents, showing a detrital origin of the magnetic minerals. The upper part of the section is more strongly affected by secondary magnetic phases and does not carry primary magnetization.

4. Clay mineral assemblages of the Owadów–Brzezinki section predominantly consist of illite and kaolinite, with subsidiary quantities of random, I–S minerals. The clay mineral assemblages are considered mainly detrital and their fluctuations can be interpreted in terms of environmental changes. A major transition from kaolinite- to I–S R0-dominated clay mineral assemblages is observed at the boundary between the Scythicus (Panderi) and Virgatus zones of the Middle Volgian. This phenomenon is related to aridification of the latest Jurassic climate, which is also recorded in NW–NE Europe, as inferred from geochemical proxies.

5. Palaeohumidty (Ti/K, K/Al) and grain-size proxies (Ti/Al and Zr/Rb) account for decreasing chemical weathering and terrestrial runoff close to the Scythicus (Panderi) and Virgatus Biochron transition of the Middle Volgian. More humid conditions reappear later in the earliest Virgatus Biochron. Apparently humid intervals correlate with fully marine conditions, while arid conditions correspond to lagoonal depositional settings interpreted as regressive events.

6. The Owadów–Brzezinki section can be correlated, using magnetostratigraphy, with Tethyan sections and some Boreal sections. The Scythicus (Panderi) and Virgatus ammonite chron aridification and regressive events are also observed in the Kcynia IG 2 borehole core (Polish Basin), as inferred on the basis of Ti/K and Zr/Rb ratios. Similar trends can be traced in the uppermost part of Kimmeridge Clay Fm (southern England) and the Purbeck Beds of the NW Aquitaine Basin (France), throughout the boundary interval of the Fittoni and Albani ammonite zones.

We warmly acknowledge Jerzy Kwiatek (Managing Director) and Sławomir Melka (Quarry and Production Manager of the Nordkalk Company) for the invaluable support and enabling research in the Owadów–Brzezinki quarry. The comments of two journal referees, M. Benzaggagh and M. Hounslow, significantly helped to improve the manuscript.

BB: conceptualization (lead), data curation (equal), formal analysis (equal), funding acquisition (lead), investigation (equal), methodology (equal), project administration (lead), supervision (lead), visualization (equal), writing – original draft (equal); AP: conceptualization (equal), data curation (equal), formal analysis (equal), writing – original draft (supporting); JG: conceptualization (equal), data curation (equal), formal analysis (equal), funding acquisition (supporting), writing – original draft (equal); HW: conceptualization (equal), data curation (equal), formal analysis (equal), methodology (equal), project administration (equal), writing – original draft (equal); J-FD: data curation (equal), formal analysis (equal), methodology (equal), writing – original draft (equal); EO: formal analysis (equal); AT: formal analysis (supporting); JN: data curation (equal), formal analysis (equal), methodology (supporting)

This work was supported by the Polish National Science Centre (grant no. 2020/39/B/ST10/01489) and statutory funds of the Polish Geological Institute – NRI (grant nos 61.2201.1601.00.0 and 75.2808.2101.00.0).

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No additional external funding received for this study.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.