February 2015 LIP of the Month

The Franklin Large Igneous Province and Initiation of the Sturtian Snowball Earth Glaciation

Francis A. Macdonald1, Athena E. Eyster1 and Grant M. Cox2

1 - Harvard University, Cambridge, Massachusetts, U.S.A.

2 - McGill University, Montréal, Québec, Canada.


The Franklin Large Igneous Province (LIP) was emplaced between 730 and 710 Ma and covers an area of over 2.25 Mkm2 (Ernst et al., 2008) with lavas, sills, and dikes extending over much of northern Laurentia from Alaska through northern Canada to Greenland (Fig. 1) and potentially to Siberia (Ernst et al., 2013). The age of the Franklin LIP overlaps with the onset of the Sturtian Snowball Earth glaciation, which began between 717 and 716 Ma (Macdonald et al., 2010) and marked the first glaciation in over 1 billion years (Pierrehumbert et al., 2014; Rooney et al., in press). The Franklin LIP is the largest Neoproterozoic LIP preserved in the geological record and one of the largest in Earth History. Additionally, it was emplaced at equatorial latitudes with associated sills that invaded epicontinental evaporate basins, potentially maximizing environmental effects. Below we review the petrology, geochemistry, geochronology, and paleomagnetism of the Franklin LIP, discuss its relationship to the Sturtian Snowball Earth glaciation in space and time, and propose tests for models relating the two.

Figure 1: a) Features and locations in the Arctic important for the Franklin LIP, b) Extent of the Franklin LIP, following: Denyszyn et al., 2009a, b, c; Christie & Fahrig, 1983; Fahrig et al., 1971; Fahrig & Schwarz, 1973; Park, 1981; Robertson & Baragar, 1972; Palmer & Hayatsu, 1975; Palmer et al., 1983; Macdonald et al., 2009, Macdonald et al., 2010, Macdonald et al., 2011, Cox et al., 2015b.


Igneous rocks associated with the Franklin LIP are dominated by mafic dikes that intrude the Canadian shield. The dikes presumably fed large basaltic provinces that have since been eroded. Preserved extrusive volcanics include the Natkusiak Formation basalt of Minto Inlier, Victoria Island, the Mount Harper Volcanics of the Coal Creek inlier in the Ogilvie Mountains, Yukon, the Pleasant Creek Volcanics of the Tatonduk inlier, which straddles the Yukon-Alaska border, and the Kikiktak Volcanics of Arctic Alaska (Fig. 1). These extrusive rocks are composed predominantly of basalt to basaltic andesites (hypersthene to quartz-normative tholeiites), typical of contine­­ntal flood basalts (CFBs); many of which are too evolved to be in equilibrium with peridotitic mantle (i.e. in equilibrium with olivines of less than Fo89 composition). In fact, with average MgO compositions of ~6% and SiO2 of ~52% the average composition is close to being a basaltic andesite, consequently most of these CFBs have undergone significant gabbroic fractionation (i.e. olivine + plagioclase + clinopyroxene), making it difficult to reconstruct a composition that would have been in equilibrium with mantle peridotite and ascertain the relative contribution of the sub-continental lithospheric mantle (SCLM), asthenospheric mantle, or plumes in generating these significant magmatic events (Cox et al., 2015a).

The Natkusiak Magmatic Assemblage

The Natkusiak magmatic assemblage consists of dolerite sills that intruded into carbonate and evaporites of the Shaler Supergroup (Young et al., 1981; Rainbird et al. 1994; Fig. 2A), northwesterly striking dikes, and a succession of plateau basalts (Fig. 2B, 2C; Christie, 1964; Baragar, 1976; Dostal et al., 1986). Sedimentary strata of the underlying Shaler Supergroup were updomed, beveled and faulted (Fig. 2C, 2D) before and during the emplacement of the Natkusiak magmatic assemblage (Bedard et al., 2012; Rainbird, 1993). Sills range in thickness from 5 to 100 m; some are differentiated with olivine accumulation at their base and pegmatites in their upper levels (Dostal et al., 1986). Volcanic rocks of the Natkusiak magmatic assemblage are metamorphosed to subgreenschist facies with plagioclase and pyroxenes fresh to moderately altered, and olivine replaced by smectite or chlorite.

Exposure of the Natkusiak Formation basalts has been separated into a northern lobe (Jefferson et al., 1985) and a southern lobe (Williamson et al., 2013). The Natkusiak Formation basalts are thickest in the northern lobe, with a maximum preserved thickness of ~1100 m (Jefferson et al., 1985), and consist of massive flows that are a few meters to up to 70 m thick with amygdaloidal flow tops. In the southern lobe, the Natkusiak Formation is up to 200 m thick and has been separated into four units: 1) a basal unit 1 to 10 m thick flows with rare pillows; 2) a massive volcaniclastic conglomerate; 3) a bedded volcaniclastic sandstone; and 4) three 10 to 30 m thick, laterally continuous sheet flow basalts (Williamson et al., 2013). These four units in the southern lobe have been correlated with the Jefferson et al. (1985) ‘Basal Member’, ‘Pyroclastic Member’, and ‘Lower Massive Member’ in the northern lobe (Williamson et al., 2013). Thus, it appears that units above the ‘Lower Massive Member’ in the northern lobe and not preserved in the southern lobe.

Figure 2: Examples of the Natkusiak Formation and diabase sills from the Minto Inlier of Victoria Island: A) stepped, ~30 m-thick, diabase sill and overlying tabular sill intruding the Kilian Formation, uppermost Shaler Supergroup; B) Thick, picritic, flow near base of the Natkusiak Formation (geologist for scale); C) Exhumed contact between fluvial quartzarenite of the Kuujjua Formation and basal flows of the Natkusiak Formation, southwestern Minto Inlier (section in foreground ~150 m thick); D) Aerial view of basaltic plateau overlying block-faulted carbonate rocks of the Kilian Formation, northeastern Minto Inlier (width of view ~2 km). Photos courtesy of Rob Rainbird.

Mount Harper and Pleasant Creek Volcanics

The Mount Harper Volcanics of the Coal Creek inlier of the central Ogilvie Mountains is a mildly calc-alkaline magmatic suite that has been interpreted as the product of a partial melting at the margin of a dispersed mantle plume (Cox et al., 2015a).  The bimodal Mount Harper Volcanics of the Mount Harper Group are divided into six members (Mustard and Roots, 1997; Strauss et al., 2014). Members A-C are basaltic, ryholitic member D, followed by andesitic members E and F. Member A is comprised of pillowed and minor massive flow units 30-70 m thick and dominated by subaqueous tuffs. Member B is composed of two facies, a chloritic facies with large pillows overlain by a hematic facies. The chloritic facies is marked by grey basaltic flows, including massive pillowed flows that are more than 3 m thick. The hematitic facies involves purple-red, brown, highly vesicular flows that are 1-5 m thick and have pahoehoe flow structures and other features indicative of sub-aerial eruption. Member C comprised of pyroclastic to epiclastic breccia. Member D consists of light pink-grey-green, columnar-jointed rhyolite domes. Members E and F are composed primarily of massive flows, tuffs and breccia that are andesitic (Mustard and Roots, 1997). Flows from member F have been observed to intrude and interfinger with the Rapitan Group (Mustard and Roots, 1997; Macdonald et al., 2010). There are also dikes in the region associated with the Mount Harper Volcanics. 

A potentially equivalent basaltic pile is exposed ~50 km west of Mount Harper in the Tatonduk inlier referred to as the Pleasant Creek Volcanics (Macdonald et al., 2011). The Pleasant Creek Volcanics (formerly Upper Tindir unit 1; Young, 1992) are a series of pillowed and massive basalt lava flows along with substantial volcaniclastic breccia. Maximum thickness of the volcanic pile is ~330 m.  Stratigraphically, the volcanics sit conformably on top of shallow water carbonates and have a sharp but conformable upper contact with either carbonate or ironstone correlated with the Rapitan Group.  Numerous sub-vertical dykes that are mapped extending to the north for a further 15 km (Tindir Dyke Swarm), but the relationship between the dikes and the Pleasant Creek Volcanics has not been established.

Figure 3: Examples of the Mt. Harper Volcanics from the Coal Creek inlier. A) Ropey pahoehoe flow top of Member B at Mt. Harper, hammer for scale; B) Columnar joints of Member B, geologist in green for scale; C) Pillow basalt of Member A, geologist for scale. D) Mafic tuff of Member C, hammer for scale. Photo A was taken by Francis Macdonald, photos B & C are courtesy of Justin Strauss, and photo D was taken by Athena Eyster.

Kikiktak Volcanics

The Kikiktat Volcanics (previously termed the Mount Copleston volcanics by Moore, 1987) of the North Slope of Arctic Alaska represent a 10-1000 m thick suite of exceptionally well-preserved lavas that lie directly below 717-662 Ma Sturtian glacial deposits (Macdonald et al., 2009; Macdonald et al., 2010; Strauss et al., 2013; Cox et al., 2015b). Occurring as both sub-aerial flows with preserved pahoehoe flow tops as well as sub-aqueous pillow basalts, these lavas where first described by Moore (1987) as continental tholeiites. However, detailed petrographic and geochemical analyses reveal that these volcanics were picritic melts, including high MgO picrites that have only undergone olivine fractionation, whereas other samples are more evolved picro-basalts that lie along a gabbroic fractionation trend presenting a coherent liquid line of descent from a high MgO (~12%) parental melt (Cox et al., 2015b). The maximum metamorphic grade reached by the Kikiktat Volcanics is greenschist facies, with chlorite alteration a ubiquitous feature of all samples. Serpentinization is locally present in high MgO basalt samples, and any original groundmass is now composed of a fine-grained sericitic mineral assemblage. There is evidence for secondary silicification in some samples, and average loss on ignition values for the sample suite is ~3%.

Figure 4: Field photos of the Kikiktak Volcanics. A) Pahoehoe textures preserved on flow tops at ~ 168 m in the Nanook Creek section; hammer for scale. B) Pillow basalts from the Kikiktak Volcanics along Nanook Creek in the Shublik Mountains; hammer for scale.  Contact between the Kikiktak Volcanics, Hula Hula diamictite, and Katakturuk Dolomite at Kikiktak Mountain, helicopter for scale. D) Contact of the upper Kikiktat Volcanics and overlying Hula Hula diamictite, near the Hula Hula River; hammer for scale. Green volcaniclastic layer was dated with CA-ID-TIMS on zircon at 719.5 ± 0.3 Ma (Cox et al., 2015b). Photos A and B are courtesy of Justin Strauss. Photos C and D were taken by Francis Macdonald.


Extensive geochemical studies have been carried out on the Natkusiak lavas on Victoria Island (Dostal et al., 1986; Dupuy et al., 1995; Shellnutt et al., 2004), and on the Mount Harper Volcanics of the Yukon (Cox et al., 2015a) and the Kikiktak Volcanics of Arctic Alaska (Cox et al., 2015b). These studies have demonstrated a fertile lherzolite source for the Natkusiak lavas, whereas the Kikiktat’s volcanics compositional distinctness of low Ti, low Fe, low P, high Mg (and Mg#) and depleted high field strength elements (characteristics of a depleted source) and relatively high Al and Na (with respect to the Natkusiak lavas). Such a combination is difficult to reconcile with either assimilation or fractional crystallization, implying that these volcanics possibly represent small percentage melting of a harzburgitic source. This interpretation is supported by high 147Sm/144Nd ratios, progressive melting of mantle peridotite from lherzolite to harzburgite will be accompanied by increasing 147Sm/144Nd ratios of the residual source, basalts derived from such a depleted source will consequently inherit elevated 147Sm/144Nd ratios.  The most primitive 147Sm/144Nd ratios of the Kikiktat Volcanics are higher than most MORB and OIB examples and significantly higher than the majority of CFBs (Goldstein et al., 1984), this is in  despite the fact that assimilation drives melt compositions to lower 147Sm/144Nd ratios (Goldstein et al., 1984).

Considering that the parental melts for neither the Natkusiak basalts or the Kikiktat Volcanics contained clinopyroxene as a liquidus phase at any pressure (Cox et al., 2015b), and that the former required a lherzolite source (Dostal et al., 1986), the Natkusiak basalts must represent melting past the clinopyroxene-out boundary. In contrast, the Kikiktat Volcanics must represent a smaller percentage melt of a harzburgitic source. Consequently, the significant compositional contrasts between the two can be reconciled with their broadly comparable phase equilibrium, which predict liquids that do not coexist with clinopyroxene.

One of the distinguishing features of the Kikiktat Volcanics is their low TiO2 content, which has been interpreted to reflect unusually low Ti abundances in the source composition (Cox et al., 2015b). While these low Ti Kikiktat Volcanics are presumably associated with the breakup of Rodinia, continental flood basalts associated with the breakup of Pangea also have spatially distinct low-Ti magmatic provinces. For example, the Gondwanan Ferrar low-Ti province stretches some 3000–4000 km across Australia and Antarctica (Hergt et al., 1991). Furthermore, the Parana continental flood basalts of Brazil are an example of a smaller, but still significant Gondwanan low-Ti province (Gibson et al., 1995). Gibson et al. (1995) argued that the combination of low Ti, Fe, P and an evolved εNd signature require a mantle source for the Ferrar that was depleted in major and trace elements, with the sub-continental lithospheric mantle (SCLM) the most likely candidate. In contrast, the combination of coexisting low Ti and high Ti basalts for the Parana CFB indicated plume-SCLM interaction (Gibson et al., 1995). Following similar arguments, the Kikiktat Volcanics can be plausibly linked to melting of harzburgitic mantle.

Following similar arguments outlined above for the Kikiktat volcanics, the MHVC has likewise be plausibly linked to melting of harzburgitic mantle (Cox et al., 2015a). Together the Kikiktat volcanics and MHVC is reminiscent of the Gondwanan low-Ti provinces (Fig. 5) in that two presumably widely separated basaltic provinces share the common traits of low Ti, low Fe, depleted HFSE, high Al, high MgO (and Mg#), consistent with small degrees of partial melting of a harzburgite mantle source and are geochemically distinct (Fig. 5) from the Natkusiak basalts (near the plume center).

Cox (2015a) proposed two scenarios that can integrate these volcanic provinces into the larger Franklin LIP: 1) that melting occurred within pre-existing harzburgitic mantle underlying Laurentia (i.e. depleted SCLM), possibly heated by an underlying plume, or 2) that this may represent second stage melting of the original lherzolite source envisaged for the Franklin plume. This second scenario is appealing insofar as large-scale melting of the Franklin plume head would have left behind a melt-depleted, low-density source, which would then migrate laterally away from the plume centre.

Figure 5: (A-D) Mg8.0 analysis for the MHVC (red circles) (Cox et al., 2015a), Kikiktat Volcanics (KV) (green circles) (Cox et al., 2015b) and the Natkusiak volcanics (NV) (blue triangles) (Dostal et al., 1986).


U-Pb zircon and baddeleyite ages from the Franklin LIP rocks span from ca. 730-710 Ma (Fig. 6), however, this large spread may be due in part to imprecise multigrain techniques. The most precise, single grain chemical abrasion-ion dilution-thermal ion mass spectrometry (CA-ID-TIMS) on zircon and baddeleyite associated with the Franklin LIP yield ages between 720 and 716 Ma (Macdonald et al., 2010). Most recently, in Arctic Alaska, zircons from volcaniclastic strata at the top of the Kikiktak Volcanics were dated with CA-ID-TIMS at 719.5 ± 0.3 Ma (Cox et al., 2015b). It has been proposed that the ca. 725 Ma Dovyren Intrusive Complex in the Lake Baikal region of Russia represents a component of the Franklin LIP in SW Siberia (Ernst & Soderland, 2012; Ernst et al., 2013). Although this correlation remains speculative, if true it would provide further evidence that Franklin magmatism initiated well before the onset of the Sturtian glaciation and covered an area beyond that defined by the radiating dyke swarm in north-central to northeastern Laurentia.

Figure 6:  Ages of the Franklin LIP. U-Pb zircon and baddeleyite ages on magmatic rocks in northern Laurentia and environs inferred to be related to the Franklin large igneous province. Sources of radiometric data: a (Denyszyn et al., 2009b), b (Denyszyn et al., 2009a), c (Pehrsson and Buchan, 1999), d (Macdonald et al., 2010a), e (Heaman et al., 1992), f (Cox et al., 2015b) g (Ernst et al., 2013), h (Ariskin et al., 2013).


Paleomagnetic data from the Franklin LIP provide the most robust Neoproterozoic pole for Laurentia and an important tie point for Neoproterozoic paleogeography. These data are complicated by overprints, secular variation of the magnetic field and Cenozoic Arctic rotations of microcontinental blocks in the Arctic. The contamination from two-polarity Cenozoic overprints was acknowledged to cause a large scatter in inclination for many paleomagnetism studies of the Franklin LIP (Pehrsson and Buchan, 1999). In some cases, it was found to completely dominate the paleomagnetic data (Christie & Fahrig, 1983, Pehrsson and Buchan, 1999). In addition to Cenozoic overprints, the Franklin direction is difficult to distinguish from the very similar younger Cambrian directions (Torsvik et al., 2012). Interpretations of the paleomagnetic data from the Arctic islands and Greenland are complicated by Cenozoic rotations, such as during the opening of the Labrador Sea and possible motions along the Nares Strait. In pre-Cenozoic reconstructions of Arctic North America, the closing of the Labrador Sea brings Greenland closer to Canada (Roest & Srivastava, 1989). Moreover, the Kikiktak Volcanics of Arctic Alaska are also likely displaced from their original position relative to Laurentia. Artic Alaska is thought to be a peri-Laurentian terrane that was displaced along the Arctic margin during the Paleozoic and Mesozoic (Strauss et al., 2013), however, paleomagnetic studies of the movements of Arctic Alaska are complicated by Cretaceous overprints (Halgedahl and Jarrard, 1987).

Paleomagnetism Sample Locations

Many studies have collected and analyzed Franklin aged paleomagnetic samples from northwest Canada. These studies span six key regions, the Canadian Mainland, Victoria Island, Baffin Island, Devon Island, Ellesmere Island and northern Greenland (Fig. 1). Here we have reevaluated the paleomagnetic data from these regions, excluding sites where the circle of 95% confidence about the mean is greater than 25 degrees, unless the data was already more stringently filtered by the authors. We then computed paleomagnetic poles for the sites an­­d a site-mean paleomagnetic pole for the region (Table 1 & Figure 7a).  These are shown along with two recent grand-mean poles. Note, we follow previous authors by defining westerly-directed magnetizations as normal polarity, and easterly-directed magnetizations as reversed polarity (Denyszyn et al., 2009).

On the Canadian Mainland, the dated coronation sills (Robertson & Baragar, 1972) and Lasard River intrusions of the Brock inlier have been sampled for paleomagnetism (Park, 1981). These sites appear to span enough time to average out secular variation, as both normal and reverse directions were found over a wide area. The Coronation sills are dominated by reverse polarity directions, while the likely younger Lasard River Dikes and sills are mixed (Robertson & Baragar, 1972, Park, 1981, Ernst et al., 2001, Heaman et al 1992).

On Victoria Island, the Natkusiak Formation basalts and associated intrusive rocks were the focus of two studies. The initial study of the Natkusiak basalts revealed only reversed magnetization directions (Palmer & Hayatsu, 1975). A later study of about 740 m of the Natkusiak flows, revealed a double reversal, with the lower reverse direction badly contaminated by a downward steep overprint (Palmer et al., 1983). As the dikes in the area both fed and cut sills and flows, dikes were interpreted to be active throughout the duration of the Franklin LIP (Palmer et al., 1983). Diabase sills from the Minto Inlier yielded multigrain U-Pb TIMS (zircon and baddeleyite) ages from 718 ± 2 Ma to 723 + 12/- 2 Ma (Heaman et al., 1992), and were almost all reversed in magnetization, with only one displaying a normal direction (Palmer & Hayatsu, 1975, Palmer et al., 1983).

Baffin Island dikes have been intensively sampled all across the island and have been split into several groups: the Franklin dikes, the Borden dikes and the Strathcona sound dikes. The northwest trending Franklin dikes dominate Baffin Island and multiple studies (Christie & Fahrig, 1983, Fahrig et al., 1971, Fahrig & Schwarz, 1973) show that the dikes preserve both normal and reverse polarities. Initially, the northwest trending Borden dikes (on the Borden Peninsula) were thought to be part of a much older event, due to an older ca 950 Ma age, cross-cutting relationships and a steep paleomagnetic direction (Christie & Fahrig, 1983). Revaluation of these dikes by Pehrsson and Buchan (1999) updated the age to Franklin times (ca 720 Ma), and suggested that the steep magnetizations had resulted from the superposition of a Cretaceous–Tertiary aged chemical remanent magnetization on normal and reversed primary Franklin components. The Strathcona Sound dike swarm is likely slightly younger than the Franklin swarm, as it is tangential to and cross-cuts the Franklin swarm (Christie & Fahrig, 1983). Despite its younger age, the Strathcona Sound dikes have a similar paleomagnetic direction to the rest of the Franklin dikes (Christie & Fahrig, 1983).

The mean paleomagnetic directions from Mainland Canada, Victoria Island and Baffin Island have some overlap at the 95% confidence level (Figure 7a), but this is not the case for paleomagnetic poles from Ellesmere Island, Devon Island and Greenland. Interpretation of the Devon Island dikes and Ellsemere Island dikes is complicated by the presence of a possible Ellesmere microplate that includes Devon and Ellsemere Islands and may have moved relative to the rest of North America during early Cenozoic motions (Denyszyn et al., 2009).

The Devon Island dikes display directions generally shallowly down to the west, with one dike shallow up to the east (reversed), with no overprints observed (Denyszyn et al., 2009). One dike has an anomalous direction, based on its location it is unlikely caused by a local rotation. On Ellesmere Island, three dikes were sampled for paleomagnetic study; two of them had anomalous directions, one of which yielded a Franklin LIP age of 721+/-2Ma (Denyszyn et al., 2009). As the anomalous directions were from dikes of Franklin age and were unlikely due to local rotations, they were interpreted as recording secular variation (Denyszyn et al., 2009).  Across the Nares Strait, Franklin aged dikes and sills were sampled from the northern Thule area of Greenland, where the resulting paleomagnetic directions were both normal and reversed (Denyszyn et al., 2009).

A fundamental assumption for paleomagnetic studies is that the Earth’s magnetic field averaged over enough time can be approximated by a geocentric axial dipole. It is thought that paleomagnetic sampling should take care to cover at least 10kyr in order to average out secular variation (Tauxe, 2010). As LIPs are by definition short duration, it is important to ensure that secular variations of the magnetic field are adequately preserved. For the Franklin LIP, as ages from sampled units span 721-712 Ma, it is likely that secular variation is adequately sampled (Denyszyn et al., 2009). Despite this, as mentioned by Denyszyn et al. (2009), it is also possible that the directions from Ellesmere and Devon islands are biased by the few anomalous dikes, and that the associated mean paleomagnetic pole is not adequately averaging out this secular variation.

Rotations in the Arctic

There is longstanding discussion regarding the amount of tectonic movement along the Nares Strait and the original location of Greenland. Although it is acknowledged that seafloor spreading in the Labrador Sea resulted in tectonic motions and rotations, the magnitude of possible strike slip motions is debated, and the tectonic rotations of the other Arctic islands is still undecided (Roest & Srivastava 1989, Torsvik et al., 2001). Across the Nares Strait, the similarities in the Paleozoic geology of Ellesmere Island and northwest Greenland suggest that no more than 70 km of offset occurred (Denyszyn et al., 2009). However, the main dike concentrations on both sides of the Nares Strait (on Devon Island and near Thule on Greenland) are offset by 200km, suggesting large sinistral motion (Denyszyn et al., 2009).

Denyszyn et al. (2009) suggested that the differences in the Ellesmere Island, Devon Island and Greenland paleomagnetic data from the regions to the west was best resolved by early Cenozoic block rotations around the Nares Strait region (Figure 7b). Their data supports a 20 degree rotation between Mainland Canada and an Ellesmere microcontinent (that includes at least Ellesmere and Devon islands), along with a 14 degree rotation of Greenland (Denyszyn et al., 2009). The magnitude and direction of the Greenland rotation agree with that needed to close the Labrador Sea (Roest and Srivastava 1989). However, the differences between the Greenland rotation and that needed to explain the data from Ellesmere and Devon islands suggest an extinct plate boundary between Ellesmere and Greenland that accommodated addition movements. In addition, if one assumes that Greenland, Ellesemere Island and Devon Island were part of the same tectonic block and applies the same 14 degree restoration to all three of them, there is considerable overlap of Devon Island over Baffin island.  This suggests that one must treat Devon and Ellesmere Islands separately from Greenland (Denyszyn et al., 2009).   As Lancaster Sound (between Devon and Baffin islands) is filled with thick Cenozoic sediments in a westward narrowing basin, it is possible that this area also accommodated movement of the microcontinent (Denyszyn et al., 2009). The microcontinent rotation is also consistent with a ~25 degree difference between Permian poles from Ellesmere Island and North America (Torsvik et al., 2001). This rotation is also in agreement with the restoration by Rowley and Lottes (1988) where Ellesmere is displaced from North America, helping to explain extension in the Arctic Oceans.

In addition to tectonic motions of Greenland, Ellesmere and Devon Islands, Baffin Island may have moved relative to North America along faults in the Hudson Strait during the opening of the Labrador Sea (Pinet et al., 2013). As this motion is not thought to be very large, it is unlikely to change the Baffin pole position too drastically.

The unadjusted paleomagnetic data from Ellesmere Island, Devon Island and Greenland all act to pull the paleomagnetic pole of the Franklin LIP to the west. As mentioned earlier, these sites may have been biased by an anomalous field or tectonic motions and rotations, the extent to which is still undecided. Thus, when piecing together Neoproterozoic paleogeography, a more cautious calculation of the Franklin Grand Mean Pole would use only sites from regions unaffected by these issues.  Regardless of which pole is chosen, Laurentia is restored to low paleolatitudes. Using the adjusted/rotated Grand Mean pole of Denyszyn et al., (2009), or the Grand Mean pole of Buchan et al., (2000), yield almost indistinguishable results. With the restoration, ca. 717 Ma Laurentia is centered near the equator and the main basalts and dikes of the Franklin LIP are within ±9 degrees latitude (Figure 7c).

Figure 7: a) Mean paleomagnetic poles and Grand Mean poles of the Franklin LIP . See Table 1 for details. b) Greenland and the Ellesmere Microcontinent and associated poles are rotated relative to those of the Canadian Mainland, Victoria Island and Baffin Island following Denyszyn et al., 2009.  c) Restored ca 717 Ma location of Laurentia, Greenland and Ellesmere Microcontinent following Denyszyn et al., 2009. Figure was prepared using Gplates.


The Franklin LIP appears to overlap in age with the onset of the onset of the Sturtian glaciation, which began between 717.4 ± 0.1 and 716.5 ± 0.2 Ma (CA-ID-TIMS on zircon; Macdonald et al., 2010). Additional recent ages on the Franklin LIP and the onset of the Sturtian glaciation confirm this age model. In China, an ash bed within stratified diamictite of the lowermost Chang-an Formation was dated with secondary ion mass spectrometry (SIMS) on zircon at 715.9 ± 2.8 Ma (Lan et al., 2014). In Arctic Alaska, a 719.5 ± 0.3 Ma CA-ID-TIMS age on zircon from volcaniclastic strata at the top of the Kikiktak Volcanics and directly below the Hula Hula diamictite is consistent with both a Sturtian-age of the Hula Hula diamictite and a tight temporal relationship between the Franklin LIP and the onset of the Sturtian glaciation (Cox et al., 2015b). Is this a coincidence or is there a mechanistic relationship between the largest Neoproterozoic LIP and the initiation of the Sturtian glaciation?

The Franklin LIP and the Sturtian glaciation may have been causally related through the rifting of Rodinia at low latitude, the concomitant emplacement of LIPs, and an increase in basalt weathering, all of which could have resulted in an increase of CO2 consumption through silicate weathering reactions (Donnadieu et al., 2004; Godderis et al., 2003; Macdonald et al., 2010b). Additionally, the Franklin LIP was emplaced at ±9 degrees latitude on an east-facing margin (Denyszyn et al., 2009), ideally positioned for high rainfall and high temperatures in the tropics. This is consistent with recent work that suggests that the weathering of widespread magmatic provinces associated with the poly-phase breakup of Rodinia could have increased global weatherability as seen in Sr, Os, and Nd isotope systematics (Cox et al., 2015a; Halverson et al., 2010; Rooney et al., 2014), leading to effective CO2 consumption and the initiation of global glaciation.

Although the paleogeographic setting, and the widespread weathering of basalt are consistent with increased silicate weathering causing a cooler climate, was it the proximal trigger for the onset of the Sturtian glaciation? Weathering of basalt may have lead to additional changes in the carbon cycle through the enhanced delivery of P and Fe to the ocean (Cox et al, 2015a). The in flux of limiting nutrients to the ocean may have stimulated primary productivity and resulted in enhanced carbon burial and sequestration in newly formed rift basins. Enhanced organic carbon export could also lower atmospheric CO2 concentration through changes in carbonate alkalinity, potentially triggering an ice age (Tziperman et al., 2011).

Along with being emplaced at an equatorial latitude, the Franklin LIP also intruded strata deposited in an epicontinental basin within Rodinia, which include thick sulfate evaporite deposits of the Shaler Supergroup in northwestern Canada (Young, 1981). Evans (2006) estimated a volume of 120,000 km3 of evaporites in these basins. The invasion of dikes and sills in these basins associated with the Franklin LIP may have liberated additional sulfur and led to the release of high concentrations of sulfate aerosols that increase the optical depth of the atmosphere (c.f. Self et al., 2005; Thordarson et al., 2009). The low latitude of the Franklin LIP would have extended climatic effects to both hemispheres and maximized albedo effects. If the Earth was already in a relatively cool climate state prior to the Franklin LIP, due to the low latitude paleogeography, it also would have been more sensitive to sudden changes in solar insolation (Bendsten, 2002).

Distinguishing between these hypotheses will require additional geochronology and modeling. The most precise, single grain chemical abrasion-ion dilution-thermal ion mass spectrometry (CA-ID-TIMS) ages on zircon and baddeleyite associated with the Franklin LIP have yielded ages between 720 and 716 Ma (Macdonald et al., 2010; Cox et al., 2015b). Although these ages are consistent with a relationship between the Franklin LIP and the onset of the Sturtian Snowball Earth glaciation between 717.4 ± 0.1 and 716.5 ± 0.2 Ma (Macdonald et al., 2010), these data do not distinguish whether the bulk of eruptions occurred well before onset, perhaps more consistent with weathering hypotheses, or more precisely coincident with onset. Additional modeling of weathering of the Franklin LIP in coupled topographic and ocean geochemical models could help determine the limits of increasing CO2 consumption through weathering reactions. Moreover, further modeling of the atmospheric effects of LIPs that invade evaporitic basins at low latitude in a low oxygen world may also constrain if it is feasible to release and maintain sulfur aerosols at a high enough concentration to initiate climate catastrophe. However, we must remember that a successful model for the initiation of the Sturtian Snowball Earth glaciation will also have to address the fact that a second Neoproterozoic Snowball Earth, referred to as the Marinoan glaciation, occurred between ca. 645 and 635 Ma. Together new data and modeling experiments will lead to exciting insights into the relationships between Neoproterozoic geography, LIPs, climate, and life during what is perhaps the greatest transition in surface environments in Earth History.

Table 1. Mean Paleomagnetic Poles and ages by Region



Age (Ma)

Plat (oN)

Plong (oE)




A95 (o)


Canadian Mainland Mean

Coronation Sills

Brock Inlier (Lasard River) Dikes+sills

N+R (dominated)








Park, 1981

Fahrig et al., 1971

Robertson & Baragar, 1972

Heaman et al., 1992

Ernst et al., 2004

Victoria Mean

Victoria Sills/Diorites and Natkusiak Basalts

N + R

(at least 2 reversals)


723 + 12/- 2






Palmer & Hayatsu, 1975

Palmer et al., 1983

Heaman et al., 1992

Baffin mean

Strathcona + Franklin dikes

Span western (Borden), central and eastern parts of island








Christie & Fahrig, 1983

Fahrig et al., 1971

Fahrig & Schwarz, 1973

Pehrsson & Buchan, 1999

“Ellesmere microplate”

Ellesemere +Devon islands










Denyszyn et al., 2009

Thule (Greenland)

Dikes and 2 sills

N (dominated) +R








Denyszyn et al., 2009

Buchan et al., 2000

Mean pole









Denyszyn et al., 2009

Mean Pole

Site filtered, variably rotated

Data from Canada, Baffin Is, Victoria Is, Ellesmere Is., Devon Is., and Greenland









Ariskin, A. A., Kostitsyn, Y. A., Konnikov, E. G., Danyushevsky, L. V., Meffre, S., Nikolaev, G. S., et al. 2013. Geochronology of the Dovyren intrusive complex, northwestern Baikal area, Russia, in the Neoproterozoic. Geochemistry International, 51(11), 859-875.

Baragar, W. R. A., 1976, The Natkusiak basalts, Victoria Island, District of Franklin, Current Research, Part A, Geological Survey of Canada Paper 76-1A, 347-352.

Bedard, J. H., Naslund, H. R., Nabelek, P., Winpenny, A., Hryciuk, M., Macdonald, W., Hayes, B., Steigerwaldt, K., Hadlari, T., Rainbird, R. H., Dewing, K., and Girard, E., 2012, Falut-mediated melt ascent in a Neoproterozoic continental flood basalt province, the Franklin sills, Victoria Island, Canada. Geological Society of America Bulletin, 124(5/6), 723-736.

Buchan, K. L., Mertanen, S., Park, R. G., Pesonen, L. J., Elming, S. Å., Abrahamsen, N., & Bylund, G. 2000. Comparing the drift of Laurentia and Baltica in the Proterozoic: the importance of key palaeomagnetic poles. Tectonophysics, 319(3), 167-198.

Christie, K. W., & Fahrig, W. F. 1983. Paleomagnetism of the Borden dykes of Baffin Island and its bearing on the Grenville Loop. Canadian Journal of Earth Sciences, 20(2), 275-289.

Christie, R. L., 1964, Diabase-gabbro sills and related rocks of Banks and Victoria Islands, Arctic Archipelago. Geological Survey of Canada Bulletin, 105, 13.

Cox, G. M., 2015a, Linking tectonics, LIP Emplacement and Environmental Change in the Cryogenian, [P.h.D. Thesis]: McGill University, 301 p.

Cox, G. M., Strauss, J. V., Halverson, G. P., Schmitz, M. D., McClelland, W. C., Stevenson, R. S., and Macdonald, F. A., 2015b, Kikiktat Volcanics of Arctic Alaska  – Melting of harzburgitic  mantle associated with the Franklin Large Igneous Province: Lithosphere, v. 7. no. X; p. XXX–XXX.

Denyszyn, S. W., Halls, H. C., Davis, D. W., & Evans, D. A. 2009a. Paleomagnetism and U–Pb geochronology of Franklin dykes in High Arctic Canada and Greenland: a revised age and paleomagnetic pole constraining block rotations in the Nares Strait region. Canadian Journal of Earth Sciences, 46(9), 689-705.

Denyszyn, S. W., Davis, D. W., & Halls, H. C. 2009b. Paleomagnetism and U-Pb geochronology of the Clarence Head dykes, Arctic Canada: Orthogonal emplacement of mafic dykes in a large igneous province. Canadian Journal of Earth Sciences, 46(3), 155-167.

Denyszyn, S. W., Halls, H. C., & Davis, D. W. 2009c. A paleomagnetic, geochemical and U-Pb geochronological comparison of the Thule (Greenland) and Devon Island (Canada) Dyke Swarms and its relevance to the Nares Strait problem. Polarforschung, 74(1-3), 63-75.

Donnadieu, Y., Godderis, Y., Ramstein, G., Nedelec, A., and Meert, J., 2004, A 'snowball Earth' climate triggered by continental break-up through changes in runoff. Nature, 428, 303-306.

Dostal, J., Baragar, W. R. A., and Dupuy, C., 1986, Petrogenesis of the Natkusiak continental basalts, Victoria Island, Northwest Territories, Canada. Canadian Journal of Earth Sciences, 23(5), 622-632.

Dupuy, C., Michard, A., Dostal, J., Dautel, D., and Baragar, W. R. A., 1995, Isotope and trace-element geochemistry of Proterozoic Natkusiak flood basalts from the northwestern Canadian Shield. Chemical Geology, 120 (1–2), 15-25.

Ernst, R. E., Bleeker, W., Söderlund, U., and Kerr, A. C., 2013, Large Igneous Provinces and supercontinents: Toward completing the plate tectonic revolution. Lithos, 174, 1-14.

Ernst, R. E., and Soderlund, U., 2012, A proposed 725 Ma Dovyren-Kingash LIP of southern Siberia and possible reconstruction link with the 725-715 Ma Franklin LIP of northern Larentia, Geological Association of Canada -  Mineralogical Association of Canada Joint Annual Meeting, Geoscience at the Edge: St John's, Newfoundland and Labrador, Canada.

Ernst, R. E., Wingate, M. T. D., Buchan, K. L., and Li , Z. X., 2008, Global record of 1600-700 Ma Large Igneous Provinces (LIPs): Implications for the reconstruction of the proposed Nuna (Columbia) and Rodinia supercontinents. Precambrian Research, 160, 159-178.

Evans, D. A., 2006, Proterozoic low orbital obliquity and axial-dipolar geomagnetic field from evaporite palaeolatitudes. Nature, 444(7115), 51-55.

Fahrig, W. F., Irving, E., & Jackson, G. D. 1971. Paleomagnetism of the Franklin diabases. Canadian Journal of Earth Sciences, 8(4), 455-467.

Fahrig, W. F., & Schwarz, E. J. 1973. Additional paleomagnetic data on the Baffin diabase dikes and a revised Franklin pole. Canadian Journal of Earth Sciences, 10(4), 576-581.

Gibson, S. A., Thompson, R. N., Dickin, A. P., and Leonardos, O. H., 1995, High-Ti and low-Ti mafic potassic magmas: Key to plume-lithosphere interactions and continental flood-basalt genesis. Earth and Planetary Science Letters, 136(3–4), 149-165.

Godderis, Y., Donnadieu, Y., Nedelec, A., Dupre, B., Dessert, C., Grard, A., Ramstein, G., and Francois, L. M., 2003, The Sturtian 'snowball' glaciation: fire and ice. Earth and Planetary Science Letters, 6648, 1-12.

Goldstein, S. L., O'Nions, R. K., and Hamilton, P. J., 1984, A Sm-Nd isotopic study of atmospheric dusts and particulates from major river systems. Earth and Planetary Science Letters, 70(2), 221-236.

Halgedahl, L. L., and Jarrard, R. D., 1987, Paleomagnetism of the Kuparuk River Formation from oriented drill core: Evidence for rotation of the Arctic Alaska plate, in Tailleur, I., and Weimer, P., eds., Alaskan North Slope Geology, Book 50: Santa Barbara, California, Pacific Section SEPM, p. 581-617.

Halverson, G. P., Wade, B. P., Hurtgen, M. T., and Barovich, K. M., 2010, Neoproterozoic chemostratigraphy. Precambrian Research, 182(4), 337-350.

Heaman, L. M., LeCheminant, A. N., & Rainbird, R. H. 1992. Nature and timing of Franklin igneous events, Canada: implications for a Late Proterozoic mantle plume and the break-up of Laurentia. Earth and Planetary Science Letters, 109(1), 117-131.

Hergt, J. M., Peate, D. W., and Hawkesworth, C. J., 1991, The petrogenesis of Mesozoic Gondwana low-Ti flood basalts. Earth and Planetary Science Letters, 105(1–3), 134-148.

Jefferson, C. W., Nelson, W. E., Kirkham, R. V., Reedman, J. H., and Scoates, R. F. J., 1985, Geology and copper occurrences of the Natkusiak basalts, Victoria Island, District of Franklin, Current Reseach, Part A, Volume Paper 85-1A, Geological Survey of Canada, 203-214.

Lan, Z., Li, X., Zhu, M., Chen, Z.-Q., Zhang, Q., Li, Q., Lu, D., Liu, Y., and Tang, G., 2014, A rapid and synchronous initiation of the wide spread Cryogenian glaciations. Precambrian Research, 255(0), 401-411.

Macdonald, F. A., Schmitz, M. D., Crowley, J. L., Roots, C. F., Jones, D. S., Maloof, A. C., Strauss, J. V., Cohen, P. A., Johnston, D. T., and Schrag, D. P., 2010, Calibrating the Cryogenian: Science, 327, 1241-1243.

Macdonald, F. A., McClelland, W. C., Schrag, D. P., & Macdonald, W. P. 2009. Neoproterozoic glaciation on a carbonate platform margin in Arctic Alaska and the origin of the North Slope subterrane. Geological Society of America Bulletin, 121(3-4), 448-473.

Macdonald, F. A., Smith, E. F., Strauss, J. V., Cox, G. M., Halverson, G. P., and Roots, C. F., 2011, Neoproterozoic and early Paleozoic correlations in the western Ogilvie Moutnains, Yukon, in MacFarlane, K. E., Weston, L. H., and Relf, C., eds., Yukon Exploration and Geology 2010: Whitehorse, Yukon Geological Survey, 161-182.

Moore, T. E., 1987, Geochemistry and tectonic setting of some volcanic rocks of the Franklinian assemblage, central and eastern Brooks Range, in Tailleur, I., and Weimer, P., eds. Alaskan North Slope Geology, Volume 2, SEPM, Pacific Section, California and Alaska Geological Society, Alaska, 691-710.

Mustard, P. S., and Roots, C. F., 1997, Rift-related volcanism, sedimentation, and tectonic setting of the Mount Harper Group, Ogilvie Mountains, Yukon Territory. Geological Survey of Canada Bulletin, 492, 0-92.

Palmer, H. C., Baragar, W. R. A., Fortier, M., & Foster, J. H. 1983. Paleomagnetism of Late Proterozoic rocks, Victoria Island, Northwest Territories, Canada. Canadian Journal of Earth Sciences, 20(9), 1456-1469.

Palmer, H. C., & Hayatsu, A. 1975. Paleomagnetism and K-Ar dating of some Franklin lavas and diabases, Victoria Island. Canadian Journal of Earth Sciences, 12(8), 1439-1447.

Park, J. K. 1981. Paleomagnetism of basic intrusions from the Brock inlier, Northwest Territories, Canada. Canadian Journal of Earth Sciences, 18(10), 1637-1641.

Pehrsson, S. J., & Buchan, K. L. 1999. Borden dykes of Baffin Island, Northwest Territories: a Franklin U-Pb baddeleyite age and a paleomagnetic reinterpretation. Canadian Journal of Earth Sciences, 36(1), 65-73

Pierrehumbert, R. T., 2004, Climate dynamics of a hard snowball Earth. Journal of Geophysical Research, 110, no. DO1111, 10.1029/2004JD005162.

Pinet, N., Lavoie, D., & Keating, P. 2013. Did the Hudson Strait in Arctic Canada record the opening of the Labrador Sea?. Marine and Petroleum Geology, 48, 354-365.

Rainbird, R. H., 1993, The sedimentary record of mantle plume uplift preceding eruption of the Neoproterozoic Natkusiak flood basalt. Journal of Geology, 101, p. 305-318.

Robertson, W. A., & Baragar, W. R. A. 1972. The petrology and paleomagnetism of the Coronation sills. Canadian Journal of Earth Sciences, 9(2), 123-140.

Roest, W. R., & Srivastava, S. P. 1989. Sea-floor spreading in the Labrador Sea: a new reconstruction. Geology, 17(11), 1000-1003.

Rooney, A. D., Macdonald, F. A., Strauss, J. V., Dudas, F. O., Hallman, C., and Selby, D., 2014, Re-Os geochronology and coupled Os-Sr isotope constraints on the Sturtian snowball Earth: Proceedings of the National Academy of Science, 1073, 1-6.

Rooney, A. D., Strauss, J. V., Brandon, A. D., and Macdonald, F. A., in press, A Cryogenian Chronology: Two long-lasting, synchronous Neoproterozoic Snowball Earth glaciations. Geology.

Rowley, D. B., & Lottes, A. L. 1988. Plate-kinematic reconstructions of the North Atlantic and Arctic: Late Jurassic to present. Tectonophysics, 155(1), 73-120.

Self, S., Thordarson, T., and Widdowson, M., 2005, Gas fluxes from flood basalt eruptions: Elements, 1(5), 283-287.

Shellnutt, J. G., Dostal, J., and Keppie, J. D., 2004, Petrogenesis of the 723 Ma Coronation sills, Amundsen basin, Arctic Canada: implications for the break-up of Rodinia. Precambrian Research, 129(3–4), 309-324.

Strauss, J. V., Macdonald, F. A., Taylor, J. F., Repetski, J. E., and McClelland, W. C., 2013, Laurentian origin for the North Slope of Alaska: Implications for the tectonic evolution of the Arctic. Lithosphere, 5(5), 477-482.

Strauss, J.V., Roots, C.F., Macdonald, F.A., Halverson, G.P., Eyster, A., and Colpron, M., 2014b, Geological map of the Coal Creek inlier, Ogilvie Mountains (NTS 116B/10-15 and 116C/9, 16): Yukon Geological Survey, Open File 2014-15, 1:100,000.

Tauxe, L. 2010. Essentials of paleomagnetism. Univ. of California Press.

Thordarson, T., Rampino, M., Keszthelyi, L. P., and Self, S., 2009, Effects of megascale eruptions on Earth and Mars. Geological Society of America Special Papers, 453, 37-53.

Torsvik, T. H., Van der Voo, R., Meert, J. G., Mosar, J., & Walderhaug, H. J. 2001. Reconstructions of the continents around the North Atlantic at about the 60th parallel. Earth and Planetary Science Letters, 187(1), 55-69.

Torsvik, T. H., Van der Voo, R., Preeden, U., Mac Niocaill, C., Steinberger, B., Doubrovine, P. V., & Cocks, L. R. M. 2012. Phanerozoic polar wander, palaeogeography and dynamics. Earth-Science Reviews, 114(3), 325-368.

Williamson, N; Bédard, J; Ootes, L; Rainbird, R; Cousens, B; Zagorevski, A. 2013. Volcanostratigraphy and significance of the southern lobe Natkusiak Formation flood basalts, Victoria Island, Northwest Territories; Geological Survey of Canada, Current Research 2013-16(Online); 15 pages, doi:10.4095/292706

Young, G. M., 1981, The Amundsen Embayment, Northwest Territories: Relevance to the Upper Proterozoic evolution of North America, in Campbell, F. H. A., ed., Proterozoic Basins of Canada, Geological Survey of Canada Paper 81-10, 203-218.

Young, G. M., 1982, The late Proterozoic Tindir Group, east-central Alaska; Evolution of a continental margin. Geological Society of America Bulletin, 93, 759-783.