Miocene breakup of Svalbard from Greenland formed a deep oceanic gateway that enabled circulation between the Arctic and Atlantic Oceans, significantly changing the global climate. However, the timing of events remains unclear. An excellent opportunity to constrain this timing is found onshore western Svalbard, where the Sarsbukta fault forms the eastern margin of the Eocene–Oligocene Forlandsundet basin. Here, we present new results from U-Pb dating of calcite precipitated in fault-related veins to constrain the timing of Sarsbukta fault deformation and the evolution of the basin. Our oldest calcite age is Permo-Triassic, suggesting long-lived deformation along the fault. A cluster of ages between 41 and 33 Ma overlaps with fossil-based depositional ages from parts of the Forlandsundet basin. These data indicate that onshore transtension partly pre-dated the well-established Chron 13 (magnetic polarity time scale; 35.5–33.7 Ma) reorganization of spreading ridges in the North Atlantic. Our youngest age of 13 Ma indicates that faulting persisted long after the preserved basin fill was deposited. If seafloor spreading marked the end of extension of continental crust, Molloy Ridge spreading during Chron 5 (19.6–9.8 Ma) may have initiated after 13 Ma.Final Miocene breakup of the Laurasian supercontinent culminated in the opening of the Fram Strait gateway between Svalbard and Greenland (Fig. 1A), allowing deep-water circulation between the Atlantic and Arctic Oceans and resulting in dramatic global climate change (Jakobsson et al., 2007; Engen et al., 2008; Jokat et al., 2016). Crustal thinning and final breakup in the Fram Strait have been difficult to date precisely, in contrast to the earlier and less-oblique seafloor spreading in the NE Atlantic to the south, which is well constrained at 54 Ma (Doré et al., 1999; Faleide et al., 2008). The continental bridge between Greenland and Fennoscandia underwent a phase of thinning during the Late Cretaceous that led to seafloor spreading in the Labrador Sea–Baffin Bay and the Norwegian-Greenland Sea in the early Eocene (Fig. 2). At the Eocene–Oligocene transition, spreading ridges reorganized in the North Atlantic as Greenland began moving toward the northeast (Talwani and Eldholm, 1977; Faleide et al., 1991). Deformation localized along the western Svalbard margin (Fig. 1), eventually resulting in breakup and seafloor spreading east of Greenland (Døssing et al., 2013; Hosseinpour et al., 2013). This succession of events is only indirectly dated in the Fram Strait by correlation of seafloor magnetic data along strike from the North Atlantic and the Arctic Ocean (Gaina et al., 2009). Additionally, the time from crustal thinning to breakup west of Svalbard is poorly known. Here, we present U-Pb ages of calcite from veins related to transtensional faulting to directly date the highly oblique rifting that culminated in breakup between Greenland and Svalbard.The Forlandsundet basin lies onshore the western Svalbard margin. Previous studies in the area have not dated the basin-bounding fault itself, and a gap in the onshore rock record makes it methodologically difficult to do so. We utilize U-Pb dating of calcite (re)crystallization (Roberts and Holdsworth, 2022) on fault surfaces as a novel method for dating the low-temperature deformation along the Sarsbukta fault, which bounds the Forlandsundet basin to the east. In doing so, we aim to date the processes that led to the opening of the Fram Strait. We investigate (1) when the Sarsbukta fault was active, (2) what role it played during tectonic events along the western Svalbard margin, and (3) the timing of crustal thinning versus final breakup between Svalbard and Greenland.Along the western Svalbard margin, the Sarsbukta fault (Fig. 1) delimits the Forlandsundet basin to the east (Gabrielsen et al., 1992; Kleinspehn and Teyssier, 2016; Schaaf et al., 2021). The basin is the largest of many basin features along western Svalbard (Gabrielsen et al., 1992; Maher et al., 1997), making the Sarsbukta fault an ideal candidate for constraining the timing of rifting. The fault strikes NNW-SSE and displays normal separation estimated to >1 km (Gabrielsen et al., 1992; Senger et al., 2019) between basement rocks and basin strata of mid-Eocene to lower to middle Oligocene age (Fig. 1; Livšic, 1974; Feyling-Hanssen and Ulleberg, 1984; Eidvin et al., 2014; Schaaf et al., 2021). The fault is interpreted to show transtensional kinematics based on lineations on calcite-mineralized fractures (Kleinspehn and Teyssier, 2016; Schaaf et al., 2021).The Sarsbukta fault is located immediately inboard of the De Geer Zone (Fig. 2), a transform margin that bounds the western Barents Sea. The De Geer Zone accommodated crustal-scale dextral faulting before the northward propagation of the Knipovich Ridge at ca. 20 Ma (Fig. 1; Faleide et al., 2008; Lundin and Doré, 2019) and before spreading along the Molloy Ridge segment initiated during magnetic polarity Chron 5 (19.6–8.9 Ma; Engen et al., 2008).Several tectonic events along the western Svalbard margin preceded Miocene seafloor spreading. As Greenland moved northward (Fig. 2), decoupled transpression formed the West Spitsbergen Fold-and-Thrust Belt during the Eurekan Orogeny on the main island of the Svalbard archipelago (Talwani and Eldholm, 1977; Maher and Craddock, 1988; Braathen et al., 1999). An alternative model of orthogonal compression followed by strike slip was advocated by Piepjohn et al. (2016). A fault gouge from the dextral- and reverse-slip Engelskbukta fault zone provided a K-Ar age of 53.5 ± 1.0 Ma (Figs. 1 and 2) and likely records Eurekan deformation (Schaaf et al., 2021). The age of the Central Basin in the foreland of the West Spitsbergen Fold-and-Thrust Belt (Helland-Hansen and Grundvåg, 2021) remains a topic of debate. Paleontological ages from its youngest strata range between latest Paleocene and as young as Oligocene (Livšic, 1974; Manum and Throndsen, 1986; Clifton, 2012). The most robust age of the youngest preserved strata appears to be early Eocene (Manum and Throndsen, 1986; Dypvik et al., 2011), although 1.0–1.5 km of removed section has been calculated from vitrinite reflection data, indicating that overlying strata once existed in the basin (Throndsen, 1982). The Svartfjella, Eidembukta, and Daudmannsodden lineament (Fig. 1) has been interpreted to have accommodated the strike-slip component in a decoupled transpressional setting during formation of the West Spitsbergen Fold-and-Thrust Belt (Maher and Craddock, 1988; Maher et al., 1997). Directly west of this lineament lies the Sarsbukta fault and the Forlandsundet basin. As spreading ridges in the North Atlantic became reorganized during Chron 13, extension transferred from Forlandsundet to structures to the west, ultimately separating Svalbard from Greenland (Fig. 2; Talwani and Eldholm, 1977) during Chron 5.We collected 20 oriented calcite samples from veins in a 300-m-long, 10-m-high outcrop in the damage zone of the footwall of the Sarsbukta fault (Fig. 1). Six of the samples included mineral fibers on slickenlines. The sampled veins were either parallel to the scarp or from a set of conjugate fractures (Fig. 1D). We used U-Pb isotope measurements to date calcite from the veins following the method of Hagen-Peter et al. (2021). The Supplemental Material1 provides a detailed description of the samples, the U-Pb calcite dating methods used in this study, and the interpretations of the U-Pb data. Between 35 and 76 spot analyses were obtained from each sample; results were plotted in Tera-Wasserburg concordia diagrams (Fig. 3) using IsoplotR (Vermeesch, 2018). The lower-intercept ages are interpreted to date (re)crystallization of calcite, reflecting vein opening, fluid activity, and movement of an unconstrained magnitude on the vein-hosting fault.Thirteen of 20 analyzed samples yield ages from 237 to 13 Ma (Table 1; Figs. 2 and 3). Our data set shows the following: (1) One sample yields a very old age of 237 Ma, and two other samples are 81 and 59 Ma; all these ages have relatively large 2σ uncertainties between 10 and 31 m.y. (2) A cluster of eight samples ranges in age from 41 to 33 Ma. (3) The youngest sample is 13 Ma; two other samples are of similar young age at 19 and 16 Ma but have large uncertainties between 8 and 14 m.y.All samples are relatively low in U and, consequently, radiogenic Pb. Hence, some regressions yield poorly defined lower-intercept ages. Seven of the 20 samples contained no radiogenic Pb and did not yield geochronological information. Calcite was sampled from veins representing three main orientations (Fig. 1D); two of the oldest veins strike parallel to the fault, while the rest of the sampled veins strike normal to oblique to the fault in a set of conjugate fractures. We did not observe cross-cutting relationships between the different veins or other relative age indicators, nor a relationship between the age of each sample and the location along the outcrop. Three dated samples contained mineral fibers in slickenlines (Fig. 1D) showing a local NW-SE transport direction; two of these samples belong to the 41–33 Ma age cluster, and the other is younger (13 Ma).The oldest calcite ages in our data set suggest the fault may have a Permo-Triassic or older history, perhaps reflecting early phases of rifting in the greater Barents region (e.g., Smyrak-Sikora et al., 2019). The 59 ± 8 Ma age from our data set overlaps with the 53.5 Ma K-Ar age of a clay gouge associated with Eurekan transpressional faulting in the Engelskbukta fault zone (Fig. 1C) reported by Schaaf et al. (2021). Given that uncertainties are substantial for all the pre-Eocene calcite ages, these inferences are speculative. However, the ages do indicate that the Sarsbukta fault and adjacent basin were localized along a pre-existing zone of weakness. The Svartfjella, Eidembukta, and Daudmannsodden lineament (Maher et al., 1997) is a possible candidate for a lineament that may be associated with deformation along the Sarsbukta fault as well as the Engelskbukta fault zone (Fig. 1C), indicating that it may be one of many long-lived lineaments (e.g., Smyrak-Sikora et al., 2019) in the Svalbard region.The main cluster of calcite ages spans from 41 to 33 Ma. This period overlaps with that of sedimentation in the Forlandsundet basin as dated by fossil evidence to late Eocene to early to middle Oligocene (Fig. 2; Livšic, 1974; Feyling-Hanssen and Ulleberg, 1984; Eidvin et al., 2014; Schaaf et al., 2021). In theory, compression at high angles to the fault zone during fold-and-thrust belt formation could create tension fractures normal to the fault zone. However, the dated calcite-bearing veins are associated with a conjugate set of fractures, some of which include mineral-fiber lineations indicating oblique normal slip (Fig. 1D; Supplemental Material). The transtension reported by Schaaf et al. (2021) along the Sarsbukta fault and in the adjacent basin strata is based on fractures that are similar in orientation to many of our dated samples, indicating that our samples likely date the transtension.Based on the kinematics of the Sarsbukta fault (Schaaf et al., 2021), we suggest that our 41–33 Ma calcite ages record a regional transtensional strain field. This constrains the timing of an earlier compressional strain component between Svalbard and Greenland to before 41 Ma. This result may also imply that folding and thrusting along the West Spitsbergen Fold-and-Thrust Belt and related deposition in the Central Basin to the east ceased by 41 Ma or shortly thereafter.Oceanic spreading ridges in the North Atlantic reorganized at Chron 13 (35.5–33.7 Ma; Fig. 2), and seafloor spreading in the Labrador Sea ceased (Oakey and Chalmers, 2012; Hosseinpour et al., 2013). Our 41–33 Ma age cluster mainly pre-dates the end of Chron 13, indicating that deformation along the Sarsbukta fault was coeval with oceanic spreading in the Labrador Sea. This indicates that transtensional deformation initiated in Forlandsundet while seafloor spreading was still ongoing west of Greenland. We speculate that as spreading ceased in the Labrador Sea and ridges reorganized in the North Atlantic during Chron 13, oblique crustal thinning was transferred from Forlandsundet to structures to the west around this time (Fig. 1).Engen et al. (2008) suggested that spreading initiated along the Molloy Ridge during Chron 5 (19.6–9.8 Ma). Our youngest calcite crystallization age shows likely faulting in Forlandsundet at 13 ± 1 Ma, indicating some continental deformation at this time. Given that extension tends to localize along the oceanic spreading ridge (e.g., Buck, 1991) once onshore rifting ceases, it could be argued that spreading along the Molloy Ridge initiated after our youngest age of 13 Ma and before the end of Chron 5 at 9.8 Ma. However, the potential for post-rift reactivation of continental faults (e.g., Redfield et al., 2005) makes this interpretation uncertain. Thus, it is fully possible that initial spreading in the Molloy Deep was associated with fault reactivation onshore Svalbard.From our findings, we can conclude that: (1) 41–33 Ma transtension coincided with mid-Eocene to early Oligocene basin deposition in Forlandsundet, before cessation of spreading in the Labrador Sea and ridge reorganization in the North Atlantic during Chron 13; (2) movement may have occurred along the long-lived Sarsbukta fault since the Permian to Triassic or even earlier, indicating that the present-day position of the Sarsbukta fault and thus the Forlandsundet basin was controlled by long-lived preexisting zones of weakness; and (3) the Sarsbukta fault was active as recently as 13 Ma, long after the deposition of the preserved parts of the Forlandsundet basin. Furthermore, we propose that (1) transtension in Forlandsundet at 41 Ma may constrain the timing of an earlier compressional component in the regional strain field between Svalbard and Greenland, possibly also indicating an end to fold-and-thrust belt formation and related deposition in the Central Basin; (2) at 33 Ma, a decrease in fault activity in Forlandsundet may mark a transfer of oblique crustal thinning to structures to the west; and (3) 13 Ma faulting along the Sarsbukta fault may represent an upper age limit for oceanic spreading initiation along the Molloy Ridge during Chron 5. These findings place novel constraints on the tectonic prelude to the opening of the Arctic-Atlantic seaway and thus to the onset of modern oceanic circulation patterns.We received financial support from the Research Council of Norway (RCN) through funding to The Norwegian Research School on Dynamics and Evolution of Earth and Planets, project number 249040/F60; RCN through the Arctic Field Grant from the Svalbard Science Forum; Kong Haakon den 7des utdannelsesfond for norsk ungdom managed by the University in Tromsø, The Arctic University of Norway; Petroleum Research School of Norway’s Short Term Exchange Program grant; and Equinor’s publication support for 2021 and 2022. The Logistics Department at the University Center in Svalbard provided logistical support. Analytical work was done under the support of MiMaC (Norwegian Laboratory for Mineral and Materials Characterisation Norwegian Geological Survey node), supported by RCN project number 269842/F50. Kim Senger, reviewers Alvar Braathen, Grace Shepard, Harmon D. Maher, Jr., David Chew, and one anonymous reviewer are thanked for helpful comments.

中文翻译：

---

U-Pb方解石年龄测定了北极-北大西洋门户的斜裂谷


中新世斯瓦尔巴群岛与格陵兰岛的分裂形成了一个深海门户，使北冰洋和大西洋之间的环流得以实现，极大地改变了全球气候。然而，事件发生的时间仍不清楚。限制这一时间的绝佳机会是在斯瓦尔巴群岛西部的陆上发现的，那里的萨尔斯布克塔断层形成了始新世-渐新世 Forlandsundet 盆地的东缘。在这里，我们提出了断层相关脉中沉淀的方解石 U-Pb 测年的新结果，以约束 Sarsbukta 断层变形的时间和盆地的演化。我们最古老的方解石年龄是二叠纪-三叠纪，这表明断层上存在长期的变形。 41 至 33 Ma 之间的一组年龄与 Forlandsundet 盆地部分地区基于化石的沉积年龄重叠。这些数据表明，陆上地转张力部分早于北大西洋扩张脊的已确定的Chron 13（磁极时间尺度；35.5-33.7 Ma）重组。我们最小的年龄为 13 Ma，这表明断层作用在保存的盆地填充物沉积后很长时间内仍然存在。如果海底扩张标志着大陆地壳扩张的结束，那么第 5 纪（19.6-9.8 Ma）期间的莫洛伊海脊扩张可能在 13 Ma 后开始。劳亚超大陆最后的中新世分裂在斯瓦尔巴群岛和斯瓦尔巴群岛之间的弗拉姆海峡门户开放时达到顶峰。格陵兰岛（图 1A），使得大西洋和北冰洋之间出现深水环流，导致全球气候发生剧烈变化（Jakobsson 等，2007；Engen 等，2008；Jokat 等，2016）。 弗拉姆海峡地壳变薄和最终破裂的精确年代很难确定，与此相反，大西洋东北部向南的较早且倾斜度较小的海底扩张在 54 Ma 处得到了很好的限制（Doré 等人，1999 年；法莱德等人，2008）。格陵兰岛和芬诺斯坎迪亚之间的大陆桥在白垩纪晚期经历了一个变薄阶段，导致始新世早期拉布拉多海-巴芬湾和挪威-格陵兰海的海底扩张（图2）。在始新世-渐新世过渡时期，随着格陵兰岛开始向东北移动，北大西洋的扩张脊重新组织（Talwani 和 Eldholm，1977；Faleide 等，1991）。变形集中在斯瓦尔巴群岛西部边缘（图 1），最终导致格陵兰岛以东破裂和海底扩张（Døssing 等人，2013 年；Hosseinpour 等人，2013 年）。这一系列事件只能通过北大西洋和北冰洋走向的海底磁数据的关联来间接确定弗拉姆海峡发生的时间（Gaina 等，2009）。此外，斯瓦尔巴群岛以西从地壳变薄到分裂的时间鲜为人知。在这里，我们提出了来自与张拉断层有关的矿脉的方解石的 U-Pb 年龄，以直接测定最终导致格陵兰岛和斯瓦尔巴群岛之间分裂的高倾斜裂谷。Forlandsundet 盆地位于斯瓦尔巴群岛西部边缘的陆上。该地区之前的研究尚未确定盆地边界断层本身的年代，而陆上岩石记录中的空白使得在方法上难以做到这一点。 我们利用断层表面方解石（重）结晶的 U-Pb 测年（Roberts 和 Holdsworth，2022）作为一种新方法来测定 Sarsbukta 断层沿线的低温变形，该断层将 Forlandsundet 盆地东部界定。在此过程中，我们的目标是确定导致弗拉姆海峡开放的过程的日期。我们研究了（1）萨斯布克塔断层何时活跃，（2）它在斯瓦尔巴群岛西部边缘的构造事件中发挥了什么作用，以及（3）斯瓦尔巴群岛和格陵兰岛之间地壳减薄与最终分裂的时间。 ，Sarsbukta 断层（图 1）将 Forlandsundet 盆地划定为东部（Gabrielsen 等，1992；Kleinspehn 和 Teyssier，2016；Schaaf 等，2021）。该盆地是斯瓦尔巴群岛西部众多盆地特征中最大的一个（Gabrielsen 等，1992；Maher 等，1997），使得 Sarsbukta 断层成为限制裂谷时间的理想候选者。该断层走向 NNW-SSE，显示出中始新世至中渐新世的基岩和盆地地层之间的正常分离距离估计>1 km（Gabrielsen 等，1992；Senger 等，2019）（图 1） ；Livšic，1974；Feyling-Hanssen 和 Ulleberg，1984；Schaaf 等，2021；该断层被解释为显示基于方解石矿化裂缝上的线纹的转张运动学（Kleinspehn 和 Teyssier，2016；Schaaf 等人，2021）。Sarsbukta 断层位于 De Geer 地带的内侧（图 2），是一个断层。改造巴伦支海西部的边缘。大约在克尼波维奇海岭向北扩展之前，德吉尔带容纳了地壳规模的右旋断层。 20 Ma（图 1；Faleide 等人。，2008； Lundin 和 Doré，2019）以及在磁极性 Chron 5（19.6-8.9 Ma；Engen 等人，2008）期间沿着莫洛伊海岭段扩张之前。沿斯瓦尔巴群岛西部边缘发生的几次构造事件发生在中新世海底扩张之前。随着格陵兰岛向北移动（图 2），在尤里卡造山运动期间，解耦压扭作用在斯瓦尔巴群岛主岛上形成了西斯匹次卑尔根褶皱冲断带（Talwani 和 Eldholm，1977；Maher 和 Craddock，1988；Braathen 等） .，1999）。 Piepjohn 等人提出了另一种正交压缩后走滑的模型。 （2016）。来自右旋和逆滑 Engelskbukta 断层带的断层泥提供了 53.5 ± 1.0 Ma 的 K-Ar 年龄（图 1 和 2），并且可能记录了 Eurekan 变形（Schaaf 等人，2021）。西斯匹次卑尔根褶皱冲断带前陆中央盆地的年龄（Helland-Hansen 和 Grundvåg，2021）仍然是一个争论话题。最年轻地层的古生物学年龄介于古新世晚期和渐新世之间（Livšic，1974；Manum 和 Throndsen，1986；Clifton，2012）。最年轻的保存地层的最可靠年龄似乎是早始新世（Manum 和 Throndsen，1986；Dypvik 等人，2011），尽管根据镜质体反射数据计算出 1.0-1.5 公里的移除部分，表明上覆地层曾经存在于盆地中（Throndsen，1982）。 Svartfjella、Eidembukta 和 Daudmannsodden 线状结构（图 1）被解释为在西斯匹次卑尔根褶皱逆冲带形成过程中在解耦的挤压环境中容纳了走滑成分（Maher 和 Craddock，1988；Maher 等）等，1997）。这条线的正西面是 Sarsbukta 断层和 Forlandsundet 盆地。 随着北大西洋的扩张脊在第 13 纪元期间重组，延伸从 Forlandsundet 转移到向西的结构，最终在第 5 纪元期间将斯瓦尔巴群岛与格陵兰岛分开（图 2；Talwani 和 Eldholm，1977）。我们从Sarsbukta 断层下盘破坏带内长 300 米、高 10 米的露头中的矿脉（图 1）。其中六个样品在光油线上含有矿物纤维。采样的静脉要么与陡坡平行，要么来自一组共轭裂缝（图 1D）。我们按照 Hagen-Peter 等人的方法，使用 U-Pb 同位素测量来确定矿脉中方解石的年代。 （2021）。补充材料1 提供了样品的详细描述、本研究中使用的 U-Pb 方解石测年方法以及 U-Pb 数据的解释。每个样品获得 35 至 76 个点分析；使用 IsoplotR (Vermeesch, 2018) 将结果绘制在 Tera-Wasserburg 协和图 (图 3) 中。较低截距年龄被解释为方解石（重）结晶的年代，反映了脉的张开、流体活动以及脉断层上不受约束的运动。20 个分析样品中的 13 个得出的年龄为 237 至 13 Ma（表图 2 和 3)。我们的数据集显示以下内容：（1）一个样本的年龄非常大，为 237 Ma，另外两个样本的年龄为 81 和 59 Ma；所有这些年龄在 10 到 31 Ma 之间都有相对较大的 2σ 不确定性。 (2) 一组八个样本的年龄范围为 41 到 33 Ma。 (3) 最年轻的样本为13 Ma；另外两个样本的年龄相似，分别为 19 和 16 Ma，但在 8 到 14 Ma 之间具有很大的不确定性。 myAll 样本的 U 含量相对较低，因此放射性 Pb 含量也相对较低。 因此，一些回归会产生定义不明确的低截距年龄。 20 个样品中有 7 个不含有放射性铅，也没有产生地质年代学信息。方解石是从代表三个主要方向的矿脉中取样的（图 1D）；两条最古老的矿脉与断层平行，而其余采样矿脉则在一组共轭裂缝中与断层垂直或倾斜。我们没有观察到不同矿脉或其他相对年龄指标之间的交叉关系，也没有观察到每个样本的年龄与沿露头位置之间的关系。三个已标明日期的样品在光丝中含有矿物纤维（图 1D），显示局部 NW-SE 传输方向；其中两个样本属于 41-33 Ma 年龄簇，另一个更年轻（13 Ma）。我们数据集中最古老的方解石年龄表明该断层可能具有二叠纪-三叠纪或更古老的历史，可能反映了早期阶段大巴伦支海地区的裂谷（例如，Smyrak-Sikora 等人，2019）。我们数据集中的 59 ± 8 Ma 年龄与 Schaaf 等人报道的 Engelskbukta 断层带中 Eurekan 压断层相关的粘土泥的 53.5 Ma K-Ar 年龄重叠（图 1C）。 （2021）。鉴于所有始新世前方解石年龄的不确定性都很大，这些推论都是推测性的。然而，年龄确实表明萨尔斯布克塔断层和邻近盆地位于沿着预先存在的薄弱带。 Svartfjella、Eidembukta 和 Daudmannsodden 线状线（Maher 等，1997）是可能与沿 Sarsbukta 断层以及 Engelskbukta 断层带变形相关的线状线的候选者（图 1C），表明它可能是许多长寿的轮廓之一（例如，Smyrak-Sikora 等人。，2019）在斯瓦尔巴群岛地区。方解石的主要簇年龄跨度为 41 至 33 Ma。这一时期与 Forlandsundet 盆地的沉积时期重叠，化石证据显示始新世晚期到渐新世早至中期（图 2；Livšic，1974 年；Feyling-Hanssen 和 Ulleberg，1984 年；Eidvin 等人，2014 年；Schaaf 等人）等，2021）。理论上，在褶皱逆冲带形成过程中，与断层带成大角度的压缩可能会产生垂直于断层带的张紧裂缝。然而，年代久远的含方解石矿脉与一组共轭断裂有关，其中一些断裂包括矿物纤维线状结构，表明斜向正常滑移（图 1D；补充材料）。 Schaaf 等人报道的转张力。 (2021) 沿着 Sarsbukta 断层和邻近盆地地层的裂缝基于与我们许多测年样本方向相似的裂缝，这表明我们的样本很可能确定了转张的年代。基于 Sarsbukta 断层的运动学（Schaaf 等人） ., 2021)，我们建议我们的 41-33 Ma 方解石年龄记录了区域转张应变场。这将斯瓦尔巴群岛和格陵兰岛之间较早的压缩应变分量的时间限制在 41 Ma 之前。这一结果也可能意味着，沿着西斯匹次卑尔根褶皱冲断带的褶皱和冲断作用以及东部中央盆地的相关沉积在 41 Ma 或此后不久就停止了。北大西洋的海洋扩张脊在时间 13（35.5）时重新组织。 –33.7 Ma；图 2），拉布拉多海的海底扩张停止（Oakey 和 Chalmers，2012；Hosseinpour 等，2013）。我们的 41-33 Ma 年龄组主要早于第 13 纪元末期，表明沿 Sarsbukta 断层的变形与拉布拉多海的海洋扩张同时发生。 这表明，当格陵兰岛以西海底扩张仍在持续时，福兰桑德特开始发生拉张变形。我们推测，随着拉布拉多海的扩散停止以及北大西洋海脊在第 13 纪期间重组，地壳倾斜减薄在此时从 Forlandsundet 转移到了向西的构造（图 1）。 (2008) 表明，在 Chron 5（19.6-9.8 Ma）期间，沿着莫洛伊海脊开始传播。我们最年轻的方解石结晶年龄显示，Forlandsundet 可能发生断层，时间为 13 ± 1 Ma，表明此时存在一些大陆变形。鉴于一旦陆上裂谷停止，扩张往往会沿着海洋扩张脊进行局部化（例如，Buck，1991），因此可以说，沿着莫洛伊脊的扩张开始于我们最年轻的年龄 13 Ma 之后和 Chron 5 结束于 9.8 Ma 之前。嘛。然而，大陆断层裂谷后重新活动的可能性（例如，Redfield 等，2005）使这种解释变得不确定。因此，莫洛伊深渊的初始扩张完全有可能与斯瓦尔巴群岛陆上断层的重新激活有关。根据我们的研究结果，我们可以得出结论：（1）41-33Ma转变与Forlandsundet的始新世中期到渐新世早期盆地沉积同时发生，然后在13年期间拉布拉多海停止扩张和北大西洋海脊重组； (2) 自二叠纪至三叠纪甚至更早时期，可能沿着长寿的 Sarsbukta 断层发生了运动，这表明 Sarsbukta 断层以及 Forlandsundet 盆地的当前位置是受长期存在的薄弱带控制的； (3) Sarsbukta 断层在 13Ma 时就很活跃，距 Forlandsundet 盆地的保存部分沉积很久之后。此外，我们提出（1）Forlandsundet 41 Ma 的转变可能限制了斯瓦尔巴群岛和格陵兰岛之间区域应变场中早期压缩分量的时间，也可能表明褶皱逆冲带形成和相关沉积的结束。中央盆地； (2) 33 Ma时，Forlandsundet断层活动的减少可能标志着地壳倾斜减薄结构向西转移； (3) 沿 Sarsbukta 断层的 13 Ma 断层可能代表了第 5 纪沿 Molloy 海脊的海洋扩张起始的年龄上限。这些发现对北极-大西洋海道开放的构造前奏提出了新的限制，从而对现代海洋环流模式的开始。我们通过向挪威地球和行星动力学与演化研究学院提供资金，获得了挪威研究委员会 (RCN) 的财政支持，项目编号 249040/F60； RCN 通过斯瓦尔巴科学论坛的北极领域资助； Kong Haakon den 7des utdannelsesfond fornorsk ungdom 由挪威北极大学特罗姆瑟大学管理；挪威石油研究学院短期交流计划资助； Equinor 对 2021 年和 2022 年的出版支持。斯瓦尔巴群岛大学中心的后勤部门提供了后勤支持。分析工作是在 MiMaC（挪威矿物和材料表征实验室挪威地质调查节点）的支持下完成的，并得到 RCN 项目编号 269842/F50 的支持。感谢 Kim Senger、审稿人 Alvar Braathen、Grace Shepard、Harmon D. Maher, Jr.、David Chew 和一位匿名审稿人提供的有用评论。