Pockets of Ediacaran–Cambrian clastic sedimentary rocks are preserved across Fennoscandia, but the provenance, depositional setting and age of many such deposits remain uncertain. We report the first detrital zircon provenance study of the lowermost sediments deposited on sub-Cambrian bedrock in southern Sweden. We performed 285 ion microprobe U–Pb analyses on zircons from the Mickwitzia Sandstone (File Haidar Formation, Cambrian Series 2). Age peaks at c. 2100, 2000, 1800, 1550, 1500, 1225, 1150 and 950 Ma are consistent with Sveconorwegian and Fennoscandian source rocks, whereas a major peak at c. 550 Ma is attributed to the Timanian orogen to the NE. The youngest dates in the dataset, including multiple consistent dates from single grains, indicate a maximum depositional age of c. 550 Ma. This first documentation of Timanian-aged grains in southern Sweden connects reports of Timanian detritus as far afield as northern Norway, Estonia and even Poland. The Timanian detrital signature across thousands of square kilometres suggests that much of Baltica probably consisted of an extremely low-relief plain over which Timanian-sourced detritus spread extensively during the late Ediacaran and early Cambrian.

Supplementary material: Supplementary Table 1 (Isotopic ratios and dates for zircon grains), Supplementary Table 2 (Isotopic ratios and dates for standard zircon Temora 2), Supplementary Tables 3–12 (Previously published U–Pb data with our recalculations shown in orange columns), and Supplementary Figures 1 and 2 (Back-scattered electron and cathodoluminescence images of zircon grains) are available at https://doi.org/10.6084/m9.figshare.c.7616513

The sub-Cambrian bedrock of Sweden consists of gneissic granite weathered to a remarkably low-relief surface that still dominates much of the landscape of the Baltic craton to this day (Holm 1901; Flodén 1980; Lidmar-Bergström 1995, 1996; Jensen 1997). This Precambrian surface is widely considered to be the result of peneplanation that produced an extremely flat regional unconformity, generated via mechanical erosion (surface wash) and long-term chemical weathering throughout the late Mesoproterozoic and early Neoproterozoic (Goodfellow et al. 2019; Hall et al. 2019). In this view, the palaeocontinent had been extensively scoured and denuded by the start of the Phanerozoic and whatever previous topography had existed on Baltica had been all but erased (Puura et al. 1997).

Following this long-term hiatus, deposition resumed during the late Neoproterozoic and early Paleozoic, which record a series of marine transgressions (Keller and Rozanov 1980). The low relief meant that even minor changes in sea-level led to submergence and the development of extensive epeiric seas across Baltica during this time period. Consequently, the sediments deposited in this interval are recorded as thin, laterally extensive units that can be mapped across broad areas of southern Sweden and into the Baltic Sea (Tuuling et al. 1997; Nielsen and Schovsbo 2011; Slater et al. 2017; Guilbaud et al. 2018). Because these epeiric basins were particularly shallow and the coastlines were distal, the central regions were probably starved of the source sediments. This is reflected in the relatively condensed sequences of Cambrian strata across much of southern Scandinavia, which record low rates of net deposition. An exception is the Danish island of Bornholm, where c. 180 m of Cambrian sediments are preserved (Martinsson 1974).

In southern Scandinavia, the vast majority of the Paleozoic sedimentary cover has since eroded away, exposing either the crystalline basement rocks, or has been supplanted by more recent Quaternary deposits. Nevertheless, Cambrian age sediments in this region survive in small pockets, protected from erosion within fault-graben, fissure fills or as outliers capped by resistant igneous rocks (Munthe 1905; Priem et al. 1968; Friese et al. 2011). One site where both the Precambrian basement and lower Cambrian sediments are easily accessible is at Lugnås in Västergötland, Sweden (Fig. 1). Here, the lower Cambrian sequence has survived beneath a protective igneous sill, which has now eroded away at Lugnås, but is still visible on the neighbouring hills of Kinnekulle and Billingen (Priem et al. 1968).

At Lugnås, the lower Cambrian is represented by the File Haidar Formation (Bergström and Gee 1985; Hagenfeldt 1994; Jensen 1997). On the Swedish mainland, the File Haidar Formation is typically subdivided into a lower Mickwitzia Sandstone Member and an upper Lingulid Sandstone Member (Jensen and Bergström 1995; Jensen 1997; Nielsen and Schovsbo 2006). The Mickwitzia Sandstone consists of a c. 10 m thick package of thinly bedded heterolithic coarse sandstones, with fine-grained mud and kaolinite-rich clay interbeds. The high kaolinite content is likely sourced from the underlying saprolite gneiss, representing an extensive interval of deep weathering during peneplanation (Petrov 1991; Lidmar-Bergström et al. 1997). This kaolinite may relate to the exceptional fossil preservation seen in clay-rich layers of the Mickwitzia Sandstone (Slater 2024). The exposure at Lugnås provides an exceptionally well-preserved and accessible example of the contact between the Mickwitzia Sandstone Member and the underlying crystalline basement (Fig. 1).

The provenance of the sediments comprising the Mickwitzia Sandstone remains unclear. At least some of the material in the most basal portion of this unit appears to be derived from the basement itself (from the Sveconorwegian orogen, 1140–960 Ma). Rounded pebble-sized clasts of gneiss occur within a basal polymict conglomerate layer that also contains pebbles of quartz, feldspar, siderite and clay, as well as potentially wind-faceted stones (Nathorst 1885; Nathorst 1886; Hadding 1927; Lindström and Vortisch 1978; Jensen 1997; Lidmar-Bergström et al. 1997). This conglomerate has traditionally been interpreted as a transgressive lag, deposited over the denuded gneissic basement. Nevertheless, little is known of the source of these siliciclastic rocks: the low-relief basement and lack of any known nearby Neoproterozoic strata means an obvious source is lacking.

The precise age of these sediments is poorly understood and is essentially based on biostratigraphic correlations. Although the fossil contents of these strata have been useful in this regard (the sediments are clearly early Cambrian in age), these consist principally of trace fossils (e.g. Jensen 1990, 1997; Kesidis et al. 2019a, b) and organic microfossils (Hagenfeldt 1989; Eklund 1990; Slater et al. 2017) that have relatively long stratigraphic ranges. Constraining the maximum depositional age would be useful in elucidating the timing of early Cambrian transgressions and the deposition of these sediments.

Here, we approach these questions via a geochronological analysis of detrital zircons in the lowermost Mickwitzia Sandstone. The aim is to shed light on the depositional context of this shallow epeiric basin that once covered much of the palaeocontinent of Baltica.

The sample, a medium-grained sandstone with clay-rich layers (Fig. 1), was disaggregated in a jaw-crusher and ring-and-puck-style mill before its heavy mineral fraction was concentrated by magnetic separation (with a hand magnet and Frantz magnetic separator) and heavy liquid density separation (using methylene iodide with a density of c. 3.3 g cm−3). The zircon grains were picked under a binocular microscope with the aim of analysing all the sizes and morphologies of zircon grains in the sample. The grains were placed on double-sided sticky tape alongside the 91500 zircon standard of known age (isotope dilution thermal ionization mass spectrometry (ID-TIMS) 206Pb/238U age 1062.4 ± 0.8 Ma; Wiedenbeck et al. 1995), which was analysed as the calibration reference material, and Temora 2 zircon standard (ID-TIMS 206Pb/238U age 416.78 ± 0.33 Ma; Black et al. 2004), which was analysed as a secondary reference material.

The grains were cast in two epoxy mounts (laboratory numbers 1931 and 2179), which were ground and polished with a 3 µm diamond suspension to expose the grain interiors and then polished with a 1 µm diamond suspension. After the mounts had been cleaned and coated in carbon, the exposed cross-sections of the grains were imaged at the Swedish Museum of Natural History (Stockholm, Sweden) using the cathodoluminescence and back-scattered electron modes on an FEI Quanta FEG 650 scanning electron microscope equipped with a Gatan ChromaCL2 system. After imaging, the carbon coat was removed by a 15 s polish with a 1 µm diamond polishing paste and the mounts were cleaned again and coated in Au in preparation for in situ U–Pb analysis.

The zircon grains were analysed for their U–Pb isotopic composition and age by secondary ion mass spectrometry in four analytical sessions: two sessions for each mount (the grains that gave young dates in the first session on each mount were revisited and targeted with additional analyses during the later sessions). For the U–Pb analyses, an Oregon Physics Hyperion H201 RF plasma high-brightness oxygen source was used to generate O2− ions and the primary column was tuned in the critical focusing mode with a small (5 × 5 μm) raster applied to produce an analysis pit c. 15 µm across. The analyses were performed using a peak-switching routine, with a single ion counting electron multiplier as the detection device. An energy window of 45 eV was used throughout, with energy adjustments made using the 90Zr216O peak.

Precise mass calibration was maintained by using an automatic routine in CAMECA CIPS software to scan over the large peaks and extrapolate the mass to the B-field curve for peaks between these reference points – for example, the Pb isotopes were calibrated by centring the 94Zr216O peak at nominal mass 204 and the 177Hf16O2 peak at nominal mass 209. The Temora 2 standard returned weighted mean 206Pb/238U ages within the uncertainty of its established ID-TIMS age in all three sessions in which it was analysed (Temora 2 was not analysed in the first analytical session, in which just ten zircon grains were analysed): 416.8 ± 3.4 Ma (mean square of weighted deviates (MSWD) 1.2, probability of fit 0.30, n = 6) for session #2; 418.9 ± 3.6 Ma (MSWD 0.37, probability of fit 0.90, n = 7) for session #3; and 420.9 ± 4.1 Ma (MSWD 0.14, probability of fit 0.98, n = 6) for session #4. The isotopic ratios and dates for zircon grains from the sample are reported in Supplementary Table 1 and the isotopic ratios and dates for Temora 2 are reported in Supplementary Table 2.

We used the decay constant values of Steiger and Jäger (1977) and, unless stated otherwise, the uncertainties are presented at the 2σ level in the text and figures and at the 1σ level in the data tables. Concordia diagrams, histograms and kernel density estimates were made in IsoplotR (Vermeesch 2018), with the kernel density estimates in each figure having the same bandwidth (20 Ma, the approximately average 2σ uncertainty of the concordia ages calculated for each data point). Filtering of the U–Pb data involved removing data points that had common Pb (PbC) contents >2% f206 (the percentage of common 206Pb in the measured 206Pb) or were >2% discordant (concordia distance, as per Vermeesch 2021). All data are reported in Supplementary Table 1.

The U–Pb concordia ages from the Mickwitzia Sandstone range from c. 2.95 Ga to c. 480 Ma (Fig. 2). The sample is dominated by Proterozoic dates, with prominent peaks centred at c. 550 Ma (with a shoulder at c. 600 Ma), 950, 1150, 1225, 1500, 1550, 1800, 2000 and 2100 Ma (Fig. 3). Four further concordant analyses gave Archean concordia ages of 2536 ± 16, 2689 ± 10, 2716 ± 7 and 2945 ± 10 Ma. At the younger end of the spectrum, the concordia ages are spread between c. 700 and c. 530 Ma, with a single younger date of 482 ± 13 Ma. Two further analyses on the grain that gave the c. 480 Ma date gave older dates (534 ± 11 and 551 ± 12 Ma), indicating that the c. 480 Ma date may reflect partial Pb loss from a metamict domain. Thus the c. 480 Ma date is not considered likely to reflect a real geological event. Six grains that were each analysed multiple times gave weighted mean concordia ages of c. 550–540 Ma (Supplementary Table 1, Supplementary Fig. 1). For example, grain z230 gave a weighted mean concordia age of 546.7 ± 7.6 Ma (MSWD 1.1; P = 0.36) and grain z325 gave a weighted mean concordia age of 542.3 ± 9.6 Ma (MSWD 0.89; P = 0.44). Alternative methods for calculating the maximum depositional age of the sediment, such as calculating the youngest cluster at 1σ uncertainty (YC1S), the youngest cluster at 2σ uncertainty (YC2S) and the maximum likelihood age, gave ages of 539.6 ± 3.4, 544.4 ± 3.6 and 557 ± 15 Ma, respectively. Considering the uncertainties on the weighted mean ages and the various methods of calculating the maximum depositional age of the sediment, the detrital zircon dataset indicates that sediment must have been deposited after c. 550 Ma.

Absolute age constraints on timing of sedimentation

In Sweden, the precise ages of Cambrian siliciclastic strata are often poorly constrained due to a lack of absolute dating and the Mickwitzia Sandstone is no exception (Bergström 1971; Jensen 1997; Nielsen and Schovsbo 2011). Shelly macrofossils are relatively scarce in the Mickwitzia Sandstone (Bergström 1971) and, although concentrations of the eponymous brachiopod Mickwitzia, the tube-shaped Torellella and the cap-shaped Mobergella are found locally (Jensen 1997), there are vanishingly few biostratigraphically useful taxa (e.g. trilobites). Constraining the age of the File Haidar Formation (of which the Mickwitzia Sandstone is the lowermost member) has therefore been difficult and has largely relied on attempts to correlate this unit with geographically adjacent Cambrian sediments in Skåne (southernmost Sweden), southern Norway or in the Baltic States (Hagenfeldt 1989, 1994; Nielsen and Schovsbo 2011). An upper (youngest) bound has hinged on the age of the overlying Borgholm Formation, which contains Wuliuan age trilobites.

The acritarch contents appear to place the File Haidar Formation within the long-ranging Heliosphaeridium dissimilar–Skiagia ciliosa acritarch Zone (Moczydłowska and Vidal 1986). Comparisons of trace fossil (Jensen 1997) and other organic microfossil contents (Hagenfeldt 1989; Eklund 1990; Slater et al. 2017; Slater and Bohlin 2022; Slater 2024), as well as sequence stratigraphic schemes (Nielsen and Schovsbo 2011), have tentatively placed the Mickwitzia Sandstone within Cambrian Series 2, stages 3–4, likely equivalent to the Holmia kjerulfi and ‘Ornamentaspis’ linnarssoni trilobite zones established in adjacent Scandinavian strata (Bergström and Ahlberg 1981; Ahlberg and Bergström 1993). Because the underlying basement is substantially older (Proterozoic, formed during the Sveconorwegian orogeny), a lower maximum depositional age for the Mickwitzia Sandstone has been previously lacking.

The youngest zircon grains in the Mickwitzia Sandstone sample suggest that the deposition of the lowermost sediments covering the basement of southern Sweden began after c. 550 Ma. Although the actual depositional age of the Mickwitzia Sandstone is likely to be substantially younger (based on biostratigraphic data), this first radiometric constraint on the timing of deposition of the Mickwitzia Sandstone is consistent with previous biostratigraphic correlations.

The Timanides

The source material for the Mickwitzia Sandstone Member (and, by extension, other portions of the File Haidar Formation) has traditionally been considered to derive from the Precambrian gneissic basement (including Archean, Paleoproterozoic, Mesoproterozoic and Tonian sources) (Lindström and Vortisch 1978; Jensen 1997). Nevertheless, in addition to these older dates, our results highlight a distinct younger population of late Ediacaran zircon grains (Figs 2–4). This raises the obvious question of the source of this late Ediacaran peak. Major sediment sources on Baltica, such as the Ural, Grampian and Caledonide orogens, are far too young (Ordovician–Carboniferous) and the most likely source therefore appears to be the Timanide orogen.

The Timanide orogen is a Neoproterozoic (Ediacaran) orogen that extends along the northeastern margin of Baltica from the Varanger Peninsula (Finnmark) in the north, down through the northern to southern parts of the Ural Mountains (Siedlecka and Roberts 1996; Puchkov 1997; Roberts and Siedlecka 2002; Gee 2004; Gee and Pease 2004; O'Leary et al. 2004; Roberts and Olovyanishnikov 2004; Pease and Scott 2009; Gee et al. 2014; Zhang et al. 2016). This orogen must therefore have affected vast expanses of Fennoscandinavia and the northwestern portion of the East European Platform. Syn- and post-tectonic granitoids in the orogen date to c. 560 Ma (Pease et al. 2004). The Timanide orogen has previously been proposed as the source of late Ediacaran populations of zircons contributing to the late Ediacaran to early Cambrian sedimentary rocks of northern Baltica (Andresen et al. 2014; Zhang et al. 2015; Ershova et al. 2019).

Timanide-associated detrital zircon populations have been reported from late Ediacaran to early Cambrian sedimentary rocks as far afield as the Stáhpogieddi Formation in far northern Norway (Zhang et al. 2015), the Upper Dividal Group of northern Sweden and Norway (Andresen et al. 2014), the Ringstrand Formation from southern Norway (Slama and Pedersen 2015; Slama 2016), various formations in the Lake Ladoga and St Petersburg areas in Russia (Kuznetzov et al. 2011; Miller et al. 2011; Ershova et al. 2019), the Lükati and Tiskre formations in Estonia (Isozaki et al. 2014; Põldvere et al. 2014) and possibly even Poland (Valverde-Vaquero et al. 2000) (Fig. 5). However, it is important to note that zircon ages of c. 700 to c. 550 Ma (i.e. consistent with a Timanian source) in lower Cambrian rocks from further south (e.g. southeastern Poland) have been interpreted as not having been sourced from the Timanides, but from fragments of the similarly aged Cadomian orogen on the periphery of Gondwana (Żelaźniewicz et al. 2020; Callegari et al. 2025) (Fig. 5). Although Żelaźniewicz et al. (2020) considered it unlikely that the late Neoproterozoic zircon grains in basins of the Trans-European Suture Zone were mostly sourced from the Timanides, they acknowledged that Ediacaran zircons in Scandinavian rocks may indeed have such a provenance.

In contrast with the Upper Dividal Group of northern Sweden and Norway, the underlying Lower Dividal Group sediments have traditionally not shown Ediacaran U–Pb dates (e.g. Kirkland et al. 2011; Andresen et al. 2014; Zhang et al. 2015). This has been interpreted to indicate that the c. 560 Ma plutons in the Timanides had not been exposed when the Lower Dividal Group was deposited. However, the recent discovery of a single Ediacaran aged zircon in the Lower Dividal Group (McLoughlin et al. 2021) suggests that some Timanide plutons were beginning to be exposed during the deposition of this unit. Reworking and transport of Timanide detritus in the late Ediacaran are also supported by characteristic zircon populations recovered from latest Ediacaran (Kotlin age) sediments from the Lake Ladoga region (Ershova et al. 2019).

The majority of detrital zircon provenance studies focused on the Cambrian sedimentary rocks of Baltica have been from the margins of the palaeocontinent (e.g. the Scandinavian Caledonides, Finnmark), with only a few studies targeting the interior parts of Baltica (e.g. around Lake Ladoga; Ershova et al. 2019). In particular, there has been a paucity of reports on Timanian sources in southern Scandinavia (Fig. 5). Indeed, detrital zircons from the lower Cambrian Hardeberga Formation in southernmost Sweden (Skåne) show no such Timanian peak (Lorentzen et al. 2018, their fig. 3). Instead these Scanian deposits appear to be dominated by older peaks at 1010, 1140 and 1620 Ma, implying that they were principally sourced from an extensive reworking of peneplain material originating from the Sveconorwegian orogenic belt (Lorentzen et al. 2018). Our results from the Mickwitzia Sandstone imply that there was input from the Timanides to the sediments deposited in south-central Sweden (Västergötland), a distance of c. 1800 km. Although a portion of the Timanide-generated sediments would have been sequestered in the subsiding foreland basin, a substantial volume of sediment must have been dispersed southwards, most likely by storm suspension and remobilization.

Although the simplest explanation for the Ediacaran age zircon grains in the Mickwitzia Sandstone is that they are sourced from the Timanides (and not, for example, from the Cadomian orogen, which has been invoked as a source of Ediacaran age zircon identified further south; Żelaźniewicz et al. 2020; Callegari et al. 2025), many questions remain. For example, why is the Hornavan–Vattudal basin of the north-central Scandinavian Caledonides devoid of Ediacaran zircons (Greiling et al. 2024 argued that the so-called Akkajaure–Tysjord culmination acted as a barrier against zircons shed from the Timanian foreland in the north), whereas Ediacaran zircons are observed further south in Norway (Slama and Pedersen 2015) and also in Sweden (this study)?

This is of interest when reconstructing the broader setting of late Ediacaran and early Cambrian deposition on Baltica. Cambrian marine sediments in this part of Baltica were deposited within expansive epeiric seas in which most sediments may have been deposited near the storm wave base. The substantially higher sea-levels during the Ediacaran–Cambrian are likely to have impacted the physiography of marine shelves more broadly. The controls on accommodation space, sediment generation (pre-vegetated continents), routing and provenance in such epicontinental settings may differ substantially from the depositional environments found on modern narrow continental margins (Peters 2007). Indeed, Ediacaran–Cambrian deposition on Baltica provides an excellent case study to explore the alternative dynamics of sediment transport within shallow epicontinental seas.

Depositional setting

There have been many efforts to analyse detrital zircon U–Pb data for insights into the nature of the tectonic setting in which the sediment was deposited. In some schemes, the distribution of zircon crystallization ages relative to the time at which the grains were deposited is used to infer whether a sediment was likely to have been deposited in a convergent, collisional or extensional tectonic setting (Cawood et al. 2012), whereas, in other schemes, previous knowledge of the sediment's depositional age is not required (Barham et al. 2022). Generally, convergent depositional settings may be characterized by a large number of zircon grains that date close to the time of sediment deposition, whereas extensional settings, such as a passive margins, may be characterized by a greater proportion of older ages that reflect the history of the underlying basement (Cawood et al. 2012).

Barham et al. (2022) presented a bivariate discrimination plot of active convergent and divergent/passive margin tectonic settings. In this scheme, the Mickwitzia Sandstone plots in the divergent field (Fig. 6a). The fact that the Mickwitzia Sandstone plots in the ‘divergent’ field on the Barham et al. (2022) plot raises the question of whether the sediments were deposited in a rift-related basin formed in response to the opening of an ocean or in, for example, a collisional basin such as a foreland basin. Like the discrimination plot of Barham et al. (2022), that of Cawood et al. (2012) also rules out a convergent setting for the deposition of the Mickwitzia Sandstone. However, the plot of Cawood et al. (2022) further suggests that the Mickwitzia Sandstone may have been deposited in a collisional setting (Fig. 6b, c).

We prefer a conservative interpretation of the data, suggesting that the Mickwitzia Sandstone was, at least, not deposited in a convergent setting dominated by unimodal detrital zircon populations related to large amounts of zircon crystallization (and, for example, associated pluton emplacement) close in time to sediment deposition. Although the zircon age profile may suggest a collisional setting (such as a foreland basin), the detritus could arguably be sourced from a collisional setting, but then have been deposited (or recycled and redeposited) in the surrounding areas. The zircon age profiles of the other sediments around northern Baltica that record a Timanian sediment source (Figs 4, 5) generally plot in the same fields on the discrimination diagrams (Fig. 6), supporting the interpretation that, in addition to sharing a Timanian source, they were likely deposited in a similar, if not the same, basin. We note that Andresen et al. (2014) and Zhang et al. (2015) previously argued for a collisional setting for the deposition of sediments containing Timanian detritus on the basis of how the U–Pb populations plotted on the discrimination plot of Cawood et al. (2012).

The basement rocks in the immediate vicinity of the Mickwitzia Sandstone belong to the Sveconorwegian orogen, with the Trans-Scandinavian igneous belt nearby. These are prime candidates for the source of at least some of the pre-Timanian components of the Mickwitzia Sandstone, particularly the <1.8 Ga grains. This contrasts with the situation further north, where the basement beneath late Ediacaran to early Cambrian sediments consists of rocks of the Svecofennian orogen, which are older than 1.8 Ga (e.g. the Upper Dividal Group, Andresen et al. 2014). In the latter situation, Andresen et al. (2014) argued that <1.8 Ga zircon grains in the Upper Dividal Group were probably sourced, as the Ediacaran age grains, from the Timanide orogen. In the case of the Mickwitzia Sandstone, it is impossible to rule out a local source for at least some of the <1.8 Ga grains in the sample. However, the high proportion of c. 540 to c. 700 Ma zircon grains in the Mickwitzia sample also suggests that the Timanide orogen was a major source of detritus and therefore also likely to have contributed a large proportion of the older grains in the sample, which would have been eroded from the country rock of the Timanides (as has been inferred for the Upper Dividal Group; Andresen et al. 2014).

The Mickwitzia Sandstone Member of the File Haidar Formation has historically been the target of extensive study for its sedimentological and palaeobiological contents and is one of the classic sections of lower Cambrian geology in Sweden (Nathorst 1885, 1886; Hadding 1927; Størmer 1956; Bergström 1971; Bergström and Gee 1985; Eklund 1990; Seilacher 1992; Jensen 1997; Kesidis et al. 2019a). Nevertheless, the provenance of this unit (as well as other Cambrian siliciclastic units across south-central Sweden) has remained unknown. Our 285 U–Pb analyses of detrital zircon material have revealed that the provenance of the Mickwitzia Sandstone (File Haidar Formation) includes a significant Timanian source (as well as providing a first radiometric constraint on the lower Cambrian in south-central Sweden). By extension, this strongly suggests that the Timanides were a major source of sediment to a broad region of Cambrian siliciclastic strata in Sweden.

The source of detritus for many of these units had previously been assumed to have been derived from local basement material; however, such lower Cambrian siliciclastic packages are clearly a mixture of ancient basement and material from a collisional tectonic setting. Sampling of the detrital zircons in Ediacaran–Cambrian strata across Baltica is still patchy, yet this study fills an important gap in southern Scandinavia (Fig. 5), linking previously scattered reports of Timanian detritus from units as far afield as Finnmark and Lake Ladoga. Significantly, the distance of the Mickwitzia Sandstone from the northeastern margin of Baltica indicates that Timanian detrital material dispersed vast distances during the late Ediacaran and then even further into the interior of the palaeocontinent during the early Cambrian (Fig. 5). The detection of a Timanide signal in the Cambrian sediments of southern Sweden highlights the value of provenance studies for understanding how detrital material was transported and deposited in the shallow, extensive epeiric basins characteristic of early Paleozoic continents.

We thank the curators of Lugnåsberget millstone quarry (Minnesfjallet) and James Holmes for help with sample collection. We thank Kerstin Lindén for casting the zircon grains in epoxy and Heejin Jeon for assistance with the U–Pb analyses. We thank the editor and three reviewers (David Chew, Paweł Poprawa and one anonymous reviewer) for their time and effort in providing thoughtful comments and suggestions.

BJS: conceptualization (equal), data curation (equal), investigation (equal), project administration (lead), validation (equal), visualization (equal), writing – original draft (lead), writing – review and editing (lead); GGK: conceptualization (equal), data curation (equal), formal analysis (lead), investigation (equal), methodology (lead), software (lead), visualization (equal), writing – original draft (supporting), writing – review and editing (supporting); GEB: conceptualization (equal), investigation (equal), supervision (supporting), writing – original draft (supporting), writing – review and editing (supporting); MJW: formal analysis (supporting), investigation (supporting), methodology (equal), resources (supporting), software (equal), supervision (supporting), writing – original draft (supporting), writing – review and editing (supporting).

BJS acknowledges the support of Swedish Research Council (Vetenskapsrådet) grant 2020-03314, GGK acknowledges the support of Swedish Research Council grant number 2020-04862 and MJW acknowledges Swedish Research Council infrastructure grant 2021-00276. This is NordSIMS laboratory contribution number 786.

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.

All data generated or analysed during this study are included in this published article (and if present, its Supplementary information files).