Paleogeothermal Gradients Across an Inverted Hyperextended Rift System: Example of the Mauléon Fossil Rift (Western Pyrenees)
Abstract
The fossil rift in the North Pyrenean Zone, which underwent high temperature-low pressure metamorphism and alkaline magmatism during Early Cretaceous hyperextension, was studied to explore the geothermal regime at the time of rifting. In this work, we combined Raman lab analysis and thermal numerical modelling to shed light on the distribution of geothermal gradients across the inverted hyperextended Mauléon rift basin during Albian and Cenomanian time, its period of active extension. Data were acquired from a set of 155 samples from densely spaced outcrops and boreholes, analyzed using Raman spectroscopy of carbonaceous material. The estimated paleogeothermal gradient is strongly related to the structural position along the Albian-Cenomanian rift, increasing along a proximal-distal margin transect from ~34°C/km in the European proximal margin to ~37–47°C/km in the two necking zones and 57–60°C/km in the hyperextended domain. This pattern of the paleogeothermal gradient induced a complex interaction between brittle and ductile deformation during crustal extension. A numerical model reproducing the thermal evolution of the North Pyrenees since 120 Ma suggests that mantle heat flow values may have reached 100 mW/m2 during the rifting event. This model reveals that above the thermal pulse, the temperature gradient varied within a small range of 55 to 62°C/km, as inferred from RSCM peak temperatures. We demonstrate that the style of reactivation during subsequent convergence influenced the thermal structure of the inverted rift system.
Key Points
- We mapped paleogeothermal gradients in the Mauléon hyperextended rift from Raman spectroscopy of carbonaceous material in 155 samples
- The paleogeothermal gradient increases from cratonal values of 30°C/km in the proximal margin to 60°C/km in the hyperextended domain
- Inherited Early Cretaceous rift structure controlled the post collisional thermal history of the Mauléon basin
1 Introduction
The processes responsible for hyperextension of the continental crust have been widely studied in the present-day Atlantic conjugate margins based on interpretation of borehole data and petroleum seismic reflection profiles, for example, in central Norway and East Greenland (Kodaira et al., 1998; Mjelde et al., 2001, 2008; Peron-Pinvidic et al., 2012a, 2012b, 2013; Reston, 2007; Weigel et al., 1995), in Iberia and Newfoundland (Boillot et al., 1987, 1989; Driscoll et al., 1995; Haupert et al., 2016; Manatschal et al., 2001; Pérez-Gussinyé, 2013; Péron-Pinvidic et al., 2007; Péron-Pinvidic & Manatschal, 2009; Reston, 2009), and in Angola and Brazil (Aslanian et al., 2009; Aslanian & Moulin, 2013; Contrucci et al., 2004; Karner et al., 2003; Karner & Gambôa, 2007; Moulin et al., 2005, 2010; Unternehr et al., 2010). These systems have been numerically modeled to reproduce the paleogeometry of the continental margins and the detachment faults responsible for crustal thinning (Bai et al., 2019; Brune et al., 2014, 2016; Huismans & Beaumont, 2003, 2008, 2011, 2014; Lavier & Manatschal, 2006). These studies have shown that the Atlantic passive margins are characterized by three domains that increase in continental crustal thickness toward the craton: the hyperextended rift domain (less than 10 km thick), the necking zone (10–25 km thick), and the proximal margin (~30 km thick) (e.g., Péron-Pinvidic et al., 2015). The paucity of in situ tectonic and stratigraphic data available from the Atlantic passive margins has led researchers to study fossil passive margins in the Alps to better constrain the tectonic, sedimentary, and structural evolution of these margin domains (Beltrando et al., 2014; Decarlis et al., 2015; Froitzheim & Manatschal, 1996; Lemoine et al., 1987; Manatschal et al., 2000, 2006, 2011; Manatschal & Nievergelt, 1997; Masini et al., 2011, 2013; Mohn et al., 2012). However, few studies have described the thermal structure of such hyperextended systems (Clerc, 2012; Clerc et al., 2015; Corre, 2017; Hart et al., 2017; Nirrengarten et al., 2019; Vacherat et al., 2014).
Recent works on the fossil passive margin of the North Pyrenean Zone have established that outcrops of subcontinental mantle identified there (Fabriès et al., 1991, 1998) are the result of hyperextension of the continental crust during the Albian-Cenomanian rifting stage (Clerc et al., 2014; Clerc & Lagabrielle, 2014; Corre et al., 2016; Grool et al., 2018; Jammes et al., 2009; Labaume & Teixell, 2020; Lagabrielle et al., 2010, 2020; Lagabrielle & Bodinier, 2008; Masini et al., 2014; Mouthereau et al., 2014; Saspiturry, 2019; Saspiturry, Razin, et al., 2019; Teixell et al., 2016; Tugend et al., 2015). In contrast to the fossil margins in the Alps, where the hyperextended rift and oceanic domains were lost to Cretaceous subduction (Handy et al., 2010; Rubatto et al., 1998; Stampfli et al., 1998), the prerift to postrift sediments preserved in the inverted Pyrenean rift margins retain the imprint of the synrift thermal regime. The North Pyrenean Zone is a privileged place to trace the thermal imprint of hyperextension from the proximal rift margins to the hyperextended domain. The high-temperature, low-pressure metamorphism recorded on the North Pyrenean Zone rift system is interpreted as the consequence of continental crustal thinning during Early Cretaceous time (Albarède & Michard-Vitrac, 1978a, 1978b; Choukroune & Mattauer, 1978; Clerc et al., 2015; Debroas, 1990; Golberg et al., 1986; Golberg & Leyreloup, 1990; Golberg & Maluski, 1988; Montigny et al., 1986; Ravier, 1957; Vielzeuf & Kornprobst, 1984).
The work reported in this paper, based on Raman spectroscopy of carbonaceous material (RSCM) on the prerift to postrift sediments of the Mauléon basin in the North Pyrenean Zone (Figure 1), integrated paleotemperature data from 102 borehole samples and 54 outcrop samples. All the major domains of the Mauléon rift system were sampled in order to (1) obtain RSCM peak temperatures on the precollisional sedimentary infill, (2) examine the paleogeothermal gradients related to hyperextension in the proximal margin, necking zone, and hyperextended domain of the rift, and (3) define the effect of rift inheritance on the post collisional thermal regime. The RSCM methodology was coupled with a numerical thermal model, allowing us to (1) validate the coherence of the paleogradient estimated by RSCM, (2) estimate at what date the maximum temperature was reached by a rock that underwent the series of tectonic events recorded in the Mauléon basin, (3) constrain the present-day mantle heat flow values beneath the basin, and (4) estimate the synrift mantle heat flow beneath the basin.

2 The Mauléon Basin
2.1 Tectono-Sedimentary Evolution
The Mauléon basin is characterized by prerift deposits composed of (1) an Upper Triassic shale, evaporite and ophite complex (Curnelle, 1983; Lucas, 1985) and (2) Jurassic carbonate platform limestones and marls developed in a relatively stable tectonic context (Lenoble, 1992) (Figure 2). The top of the Jurassic section is characterized by large-scale exposure of the Aquitaine-Pyrenean domain, which was responsible for the removal by erosion of the Jurassic carbonate platform top and the absence of Neocomian deposits (Combes et al., 1998) (Figure 2).

During the early Barremian, localized subsidence of the previously emerged domain favored the transgression of carbonate platform deposits that continued until the earliest Albian (e.g., Arnaud-Vanneau et al., 1979). The end of this period was marked locally by extensive halokinetic deformation, associated with the appearance of normal faults (~120 Ma; e.g., Canérot et al., 2005). The deposition of a thick, deep-marine conglomeratic sequence on the southern margin of the newly formed Mauléon basin marked the onset of Albian rifting in the basin, indicating uplift of the Axial Zone of the Pyrenees (e.g., Saspiturry, 2019; Souquet et al., 1985). In the basin axis, more distal turbidites of the Albian to lower Cenomanian Black Flysch group, time-equivalent to the Mendibelza Conglomerates on the southern margin, constitute the first stratigraphic unit of the North Pyrenean rift. Along the more gently dipping northern margin, coeval Albian shallow-marine carbonate deposits grade southward to more distal marl-dominated sedimentation (Serrano et al., 2006).
The Mauléon basin widened during the middle Cenomanian to middle Santonian postrift stage (Razin, 1989). In the basin axis, a thick carbonate turbiditic system was deposited, supplied by the European platform to the north. At the southern margin, the transgressive “Calcaires des Cañons” carbonate platform onlapped the previously emerged Paleozoic basement, currently exposed in the Axial Zone (e.g., Alhamawi, 1992) (Figure 2).
From late Santonian (~80 Ma) to middle Miocene time (~15 Ma), the Mauléon basin and the Aquitaine domain were affected by Pyrenean compression (Angrand et al., 2018; Labaume et al., 1985; Ortiz, 2019; Ortiz et al., 2019; Teixell, 1996, 1998; Vergés et al., 2002) responsible for the inversion of the Albian-Cenomanian hyperextended rift system (Garcia-Senz et al., 2019; Saspiturry, Allanic, et al., 2020; Teixell et al., 2016) and the deposition of a thick deep-water sequence derived from syntectonic siliciclastic systems to the east (Figure 2). This succession is no longer exposed in the Mauléon basin as the Pyrenean domain underwent exhumation and erosion starting in the Bartonian (~40 Ma; Bosch et al., 2016; Labaume et al., 2016; Vacherat et al., 2017). The Western Pyrenees ceased being affected by compression in the middle Miocene (15 Ma) during the postcollisional stage (e.g., Macchiavelli et al., 2017; Ortiz, 2019; Ortiz et al., 2019; Vergés et al., 2002).
2.2 Inverted Rift Domains
In the western Pyrenees, the North Pyrenean Zone corresponds to the inverted Mauléon basin (Figure 1; Ducasse & Vélasque, 1988; Souquet et al., 1977). The Mauléon basin is offset by the N20°E trending Saint-Jean-Pied-de-Port, Saison, and Barlanès transfer faults inherited from the Permian (Saspiturry, Cochelin, et al., 2019) and developed during the Albian rifting stage (Figure 1c; Canérot, 2008, 2018a, 2018b; Debroas et al., 2010). The Iberian (southern) proximal margin corresponds to the Axial Zone, consisting of a Paleozoic substratum overlain by a postrift Upper Cretaceous carbonate platform (Souquet, 1967), and the Paleozoic “Massifs Basques” (Muller & Roger, 1977) represent its western prolongation (Figure 3).

The Iberian necking zone, currently overthrusting the Axial Zone along the Lakhoura thrust (Teixell, 1993, 1998), is represented by the Mendibelza and Arbailles Units (Figures 3b and 3c). The Mendibelza Unit is composed of deep-sea Albian (synrift) and Upper Cretaceous (postrift) deposits overlying a Devonian to Permian substratum (Figure 3a; Boirie, 1981). The Arbailles Unit forms a syncline in which Paleozoic to Upper Triassic sediments are overlain by Jurassic (prerift) and Lower Cretaceous (synrift) carbonates and marls (Figure 3; Casteras et al., 1971; Lenoble, 1992).
The Mauléon basin hyperextended domain is bounded by the North Arbailles and the Saint-Palais reactivated synrift normal faults (Figure 3b; Saspiturry, Allanic, et al., 2020). This domain is composed of (1) a Jurassic to Lower Cretaceous carbonate cover displaced by gravity slides upon underlying evaporites of Triassic to Hettangian (earliest Jurassic) age (Bouquet, 1986; Corre et al., 2016; Ducasse et al., 1986; Lagabrielle et al., 2010), (2) Albian to lower Cenomanian (synrift) marls and fine-grained turbidites of the Black Flysch group (Fixari, 1984; Souquet et al., 1985), and (3) Upper Cretaceous (postrift) turbidites derived from the North Aquitaine carbonate platform on the European (northern) margin (Claude, 1990; Razin, 1989) (Figure 2).
The European margin is transported northward over the Arzacq basin on the Sainte-Suzanne thrust (Figure 3b; Angrand et al., 2018; Daignières et al., 1994; Issautier et al., 2020; Serrano et al., 2006; Teixell, 1998). Both sides involved are composed of a Jurassic to Albian carbonate platform and Upper Cretaceous turbidites.
3 The Cretaceous Pyrenean Metasomatic Event
3.1 High Temperature-Low Pressure Metamorphism
Synrift Pyrenean metamorphism is localized within the Internal Metamorphic Zone, a narrow east-west trending zone of metamorphosed and strongly deformed rocks (Casteras, 1933; Choukroune, 1974; Mattauer, 1968). The high temperature-low pressure metamorphism is characterized by a typical assemblage of muscovite, phlogopite, tremolite, plagioclase, potassium feldspar, and scapolite (Albarède & Michard-Vitrac, 1978a; Boulvais et al., 2006; Clerc et al., 2015; Golberg & Leyreloup, 1990; Golberg & Maluski, 1988; Montigny et al., 1986). The mineral assemblages indicate a maximum temperature of 550–650°C and a maximum pressure of 3–4 kbar (Bernus-Maury, 1984; Golberg & Leyreloup, 1990; Vauchez et al., 2013). Previous authors have established that this metamorphic event, which affected deposits of Triassic to Late Cretaceous age, featured a very high geothermal gradient resulting from continental crustal thinning (Dauteuil & Ricou, 1989; Golberg & Leyreloup, 1990). This event is also associated with deep-sea synrift sediments of the Black Flysch group (Debroas, 1978, 1987, 1990) and exposures of subcontinental mantle rocks (Clerc et al., 2012; Debroas et al., 2010).
Evidence for local denudation of the subcontinental mantle during the Early Cretaceous includes the presence of reworked clasts of mantle rocks in the late Albian to Cenomanian synrift sediments (Debroas et al., 2010; Duée et al., 1984; Fortané et al., 1986; Jammes et al., 2009; Lagabrielle et al., 2010; Roux, 1983) and the association of a positive gravity anomaly with the Early Cretaceous North Pyrenean rift system (Figure 1b; Boillot et al., 1973; Casas et al., 1997; Chevrot et al., 2018; Daignières et al., 1994; Gottis, 1972; Grandjean, 1992, 1994; Wehr et al., 2018), interpreted as indicating the presence of subcontinental mantle at shallow depths (around 10 km) under the Mauléon basin (Figure 3b; Wang et al., 2016). Geochronological evidence indicates that mantle exhumation took place between 107 and 85 Ma (Albarède & Michard-Vitrac, 1978a; Golberg et al., 1986; Golberg & Maluski, 1988; Montigny et al., 1986; Thiébaut et al., 1992).
Recent studies based on RSCM have produced evidence that the Internal Metamorphic Zone reached temperatures ranging from 400°C to 600°C in the central and eastern Pyrenees (Chelalou et al., 2016; Clerc, 2012; Clerc et al., 2015; Ducoux, 2017; Lagabrielle et al., 2016) to 200–630°C in the Nappes des Marbres at its western end, southwest of the Mauléon basin (Ducoux, 2017; Lamare, 1936; Martínez-Torres, 1989; Mendia & Ibarguchi, 1991). Hydrothermal fluid circulation can account for the local thermal homogeneity in the Internal Metamorphic Zone, for instance, in the Boucheville basin in the eastern part of the North Pyrenean Zone, where the temperature range was 530–580°C (Boulvais, 2016).
3.2 Alkaline Magmatism, Albitization, and Talc Mineralization
The North Pyrenean Zone underwent widespread Mesozoic magmatic activity, attested by Cretaceous alkaline magmatism (Montigny et al., 1986; Rossy et al., 1992). The Cretaceous igneous rocks consist of alkaline basalt and trachyte, as well as intrusive teschenyte and syenite (Azambre et al., 1992; Azambre & Rossy, 1976; Thiébaut et al., 1979). The Cretaceous volcanism was predominantly submarine, characterized by pillow basalt and pyroclastic rocks. The position of these rocks within Cretaceous sedimentary deposits ensures reliable biostratigraphic ages for this magmatic event. They range in age from late Albian to Santonian in the western Pyrenees (Castañares et al., 1997; Casteras et al., 1970; Lamolda et al., 1983; López-Horgue et al., 1999; Rat, 1959; Rat et al., 1983; Schoeffler et al., 1964) and from late Albian to Turonian in the central Pyrenees (Debroas, 1990; Dubois & Seguin, 1978) and eastern Pyrenees (Montigny et al., 1986).
These biostratigraphically constrained ages are consistent with the radiometric dates available for the Cretaceous magmatic rocks, which range from 115 to 85 Ma (Golberg et al., 1986; Henry et al., 1998; Montigny et al., 1986). Some amphibolite, pyroxenite, and hornblende dikes in the North Pyrenean mantle rocks might be derived from alkaline magmatism (Bodinier et al., 1988, 2004). Radiometric ages of these rocks are similar to those of the alkaline magmatism: 116 ± 5 Ma (Verschure et al., 1967) and 101 ± 2.5 Ma (Golberg et al., 1986) in Lherz, 103 ± 2 Ma in Caussou (Albarède & Michard-Vitrac, 1978a), and 103 Ma in Urdach (Azambre & Monchoux, 1998).
The high temperature-low pressure metamorphism and alkaline magmatism affecting the North Pyrenean Zone were synchronous with the albitization of the Agly, Salvezine, Saint-Barthélémy, and Arize massifs in the North Pyrenean Zone (Boulvais et al., 2007; Poujol et al., 2010) and the formation of thick talc-bearing intervals (Moine et al., 1989), characterized by fluid temperatures ranging from 250°C to 550°C, during hydrothermal alteration that took place between 117 and 92 Ma (Boulvais et al., 2006; Fallourd et al., 2014).
4 Methods
4.1 RSCM Thermometry
Raman measurements were performed at BRGM using a Renishaw inVia Reflex microspectrometer with a diode-pumped solid-state laser source excitation of 514.5 nm. The laser power reaching the sample surface, through the 100X objective (numerical aperture 0.90) of a Leica DM2500 microscope, did not exceed 0.1 mW. Before each measurement session, the microspectrometer was calibrated using the 520.5 cm−1 line of an internal silicon standard. After Rayleigh diffusion was eliminated by edge filters, the Raman signal was first dispersed using 1,800 lines/mm signal before analysis with a deep depletion CCD detector (1,024 × 256 pixels). In certain contexts, such as areas of contact metamorphism, at least 25 spectra need to be acquired (Aoya et al., 2010). Lacking evidence of contact metamorphism, we applied the Lahfid et al. (2010) analytical protocol and acquired 10–15 spectra to ensure consistent data. Renishaw Wire 4.1 software was used for instrument calibration and Raman measurements. More spectra were routinely acquired in samples that were observed to be highly heterogeneous.
The Raman spectrum of carbonaceous material is composed of first-order and second-order regions. In our study, we recorded Raman spectra only for the first-order region (700–2,000 cm–1), where we mainly observed two wide bands: the defect band (D; centered at ~1,350 cm–1) and the graphitic band (G; centered at ~1,580 cm–1) (see Henry et al., 2019, for a review). The Raman spectrum of pristine graphite displays only the G band; for less ordered structures of graphite the defect bands D1 (~1,350 cm–1) and D2 (~1,620 cm–1) also appear (e.g., Beyssac et al., 2002). Other bands are seen in Raman spectra for disordered structures of carbonaceous material that has experienced maximum temperatures of 200–350°C (e.g., Lahfid et al., 2010; Sadezky et al., 2005) such as D3 (~1,500 cm–1) and D4 (~1,250 cm–1). More strongly disordered structures of carbonaceous material display additional bands at 1,150 and 1,400 cm–1 (Ferralis et al., 2016; Rebelo et al., 2016; Schito et al., 2017).
The continuous evolution from disordered to ordered structure in carbonaceous material is mainly dependent on temperature and thus can be used as an indicator of the diagenetic or metamorphic grade of rocks. This evolutionary process can be measured using Raman microspectroscopy (e.g., Guedes et al., 2010; Jehlička et al., 2003; Marques et al., 2009; Wopenka & Pasteris, 1993; Yui et al., 1996). A geothermometer based on RSCM has been calibrated, initially for the temperature range of 330–650°C, to calculate peak temperatures of regional metamorphism (Beyssac et al., 2002) or contact metamorphism (Aoya et al., 2010) and subsequently extended to 200–330°C, temperatures typical of diagenesis and low-grade metamorphism (e.g., Lahfid et al., 2010). More recent work has extended the applicability of RSCM to temperatures lower than 200°C (Henry et al., 2018, 2019; Hinrichs et al., 2014; Lahfid et al., 2019; Lunsdorf et al., 2017; Schito & Corrado, 2018; Schmidt et al., 2017; Zhou et al., 2014), and Schito et al. (2017) has proposed a reference Raman spectrum for the range 80–200°C. These studies have highlighted several parameters that can serve as temperature indicators; however, the correlation curves between these parameters and vitrinite reflectance are significantly different. These differences may arise from the data treatment (the curve fitting procedure selected) or the analytical protocols followed.
In this study, we used a series of reference Raman spectra published by Lahfid et al. (2010) and Schito et al. (2017) for the range 100–340°C. We identified reference Raman spectra for carbonaceous material at 12 20°C steps in this temperature range (100, 120, 140, …, 300, 320, and 340°C; Figure 4). The ±20°C error bars on the figures represent the absolute uncertainty of maximum temperatures attributed to Raman spectra in the range 100–340°C. This ±20°C error is applied equally to all RSCM peak temperature data rather than to individual samples. Our determinations of RSCM peak temperature were based on comparisons of the Raman spectrum shape of each sample to the reference series in Figure 4. Thus, in this study, all maximum temperatures determined by RSCM geothermometry are qualitative estimates.

4.2 Calculation of Paleogeothermal Gradients
The geotherm, defined as the temperature profile versus depth, can be linear in the crust and the lithosphere if the medium is uniform and internal heat production is negligible. Where the crust is heterogeneous, however, contrasts in thermal conductivity may perturb the geotherm from linearity. In addition, radiogenic heat production in crustal rocks results in a curved geotherm in which the geothermal gradient decreases with depth. Advective processes of various scales can also disturb the thermal regime of the crust. Diapirism, magmatism, and hydrothermal fluid circulation can significantly affect the geotherm over periods ranging from several tens of thousands to millions of years, and tectonic events such as collision and thrusting, rifting (crustal extension), and coeval erosion or sedimentation lead to transient evolution of crustal temperatures at longer time scales (e.g., Jaupart & Mareschal, 2011; Magri et al., 2015; Nábělek & Nábělek, 2014).
In this study, we estimated paleogeothermal gradients across the Cretaceous extended precursor of the Mauléon basin, using RSCM peak temperatures measured in transects of surface rocks and in cuttings from deep petroleum wells. Because the investigated depths do not exceed 6,000 m, the relatively fixed temperature conditions at the surface force the geotherm to remain nearly linear. Using RSCM peak temperatures to estimate paleogeothermal gradients is not an easy task because rocks may theoretically record several peak temperatures over the course of geologic time. We ran a series of thermal models to test the sensitivity of the Mauléon basin to its particular tectonic history (see section 5) and found that the maximum peak temperatures recorded during Cretaceous time were the highest ones since 120 Ma. On that basis, we attempted to deconvolve the signal of the paleogeothermal gradient for each field section or well.
Of the seven boreholes used in this study, only the Ainhice-1 well was strictly vertical, thus the other six required adjustments for their deviations from vertical before evaluating their thermal gradients. The Chéraute-1 well had the largest deviation, amounting to 742 m at the bottom of the borehole. The other depth deviations were 313 m for Bellevue-1, 251 m for Les Cassières-2, 165 m for Hasparren-101, 88 m for Orthez-102, and 13 m for Uhart-Mixe-1. Linear regressions for all boreholes except the Chéraute-1 well are characterized by coefficient values of best fit (R2) of 0.95 or greater, their slopes corresponding to the apparent thermal gradient. This gradient was then corrected for each borehole on the basis of the mean stratal dip to yield the paleogeothermal gradient that existed before inversion of the Mauléon basin. When crustal deformation occurred prior to the time the RSCM peak temperature was reached, the paleogeothermal and apparent thermal gradients are the same. Considering the ±20°C error of the RSCM peak temperature (see section 4.1), the linear regression line is shifted from ±20°C but its slope and intercept are unchanged. Thus, the paleogradient estimation is unaffected by this uncertainty.
We also had to consider the case where a segment of the paleogeotherm was vertical or nearly so. A homogeneous temperature profile over a substantial vertical depth may be caused by homogenization of temperatures due to hydrothermal convection (e.g., temperature profiles in geothermal systems; Guillou-Frottier et al., 2013), localized ascent or descent of hydrothermal fluids (e.g., upflow through permeable fault zones; Roche et al., 2018), or refraction effects due to bodies with high thermal conductivity, such as salt domes or evaporite layers (e.g., Guillou-Frottier et al., 2010; Magri et al., 2008; Mello et al., 1995). In this latter case, the geotherm is not truly vertical, but the temperature gradient can be decreased by a factor of 2 to 3. When temperature profiles describe such a decreased thermal gradient (DG), the paleogeothermal gradient cannot be evaluated, as was the case for the Hasparren-101 and Chéraute-1 wells.
5 Results
Representative RSCM peak temperature spectra are shown in Figure 5, for the field samples and for each of seven boreholes, and the resulting temperatures are listed in Table 1 for the field samples and in Table 2 for the borehole samples (Saspiturry, Lahfid, et al., 2020). Uncorrected and corrected borehole depths are reported as measured depth (MD) and true vertical depth (TVD), respectively, in Table 2. Columnar sections of the boreholes and field sites, annotated with these RSCM peak temperatures, are presented along a north-south transect in Figure 6 and an east-west transect in Figure 7. Figure 8 presents the RSCM peak temperatures measured in boreholes with respect to TVD, and Figure 9 illustrates the RSCM peak temperatures measured on the field sections. The linear regression lines drawn for the seven boreholes are characterized by R2 values of 0.95 or greater, except in the case of the Chéraute-1 well.

Sample | Latitude | Longitude | Age | Lithology | Raman temperature (°C) |
---|---|---|---|---|---|
Axial Zone: Iberian Proximal Margin | |||||
LAK-1 | 43.102108 | –1.219360 | Maastrichtian | Calcschist | 160 |
LAK-2 | 43.036917 | –1.177413 | Danian | Calcschist | 160 |
LAK-3 | 43.035055 | –1.162842 | Campanian | Calcschist | 160 |
LAK-4 | 43.023748 | –1.144220 | Danian | Marl | 180 |
LAK-5 | 43.034887 | –1.135020 | Campanian | Calcschist | 180 |
LAK-6 | 42.991763 | –0.986248 | Campanian | Calcschist | 170 |
LAK-7 | 42.991763 | –0.986248 | Campanian | Calcschist | 170 |
LAK-8 | 42.998580 | –0.985852 | Campanian | Calcschist | 180 |
LAK-9 | 43.018367 | –0.937463 | Campanian | Calcschist | 230 |
LAK-10 | 42.999257 | –0.838788 | Campanian | Calcschist | 250 |
LAK-11 | 42.994837 | –0.834207 | Campanian | Calcschist | 250 |
LAK-12 | 43.005373 | –0.731272 | Campanian | Calcschist | 250 |
LAK-13 | 43.002290 | –0.725235 | Campanian | Calcschist | 260 |
LAK-14 | 43.003928 | –0.724552 | Campanian | Calcschist | 260 |
LAK-15 | 42.996247 | –0.684223 | Campanian | Calcschist | 280 |
Mendibelza Unit: Iberian Necking Zone | |||||
MEN-1 | 43.071898 | –1.019520 | Early Albian | Siltstone | 240 |
MEN-2 | 43.069752 | –1.021996 | Early Albian | Siltstone | 240 |
MEN-3 | 43.065726 | –1.023290 | Early Albian | Siltstone | 240 |
MEN-4 | 43.053575 | –1.017703 | Early Albian | Siltstone | 230 |
MEN-5 | 43.089358 | –1.111091 | Middle Albian | Siltstone | 200 |
MEN-6 | 43.080316 | –1.098856 | Middle Albian | Siltstone | 200 |
MEN-7 | 43.042209 | –1.151609 | Late Albian | Siltstone | 140 |
MEN-8 | 43.040542 | –1.172827 | Santonian | Breccias | 160 |
MEN-9 | 43.040623 | –1.172787 | Coniacian | Siltstone | 150 |
MEN-10 | 43.069492 | –1.194545 | Santonian | Breccias | 150 |
Saint-Etienne-de-Baïgorry Unit: Iberian Necking Zone | |||||
STB-1 | 43.181566 | –1.284771 | Early Jurassic | Limestone | 160 |
STB-2 | 43.170830 | –1.336082 | Late Albian | Marls | 130 |
STB-3 | 43.136767 | –1.264075 | Santonian | Calcschist | 150 |
STB-4 | 43.141252 | –1.261580 | Santonian | Calcschist | 150 |
STB-5 | 43.140792 | –1.261903 | Santonian | Calcschist | 150 |
STB-6 | 43.137648 | –1.257442 | Santonian | Calcschist | 160 |
Arbailles Unit: Iberian Necking Zone | |||||
BEL-1 | 43.151402 | –1.042786 | Early Jurassic | Marly limestone | 180 |
BEL-2 | 43.152268 | –1.045087 | Middle Jurassic | Marly limestone | 180 |
BEL-3 | 43.149475 | –1.055378 | Late Jurassic | Marly limestone | 180 |
BEL-4 | 43.147340 | –1.059262 | Barremian | Marl | 180 |
BEL-5 | 43.144710 | –1.058376 | Aptian | Marl | 220 |
BEL-6 | 43.141317 | –1.065354 | Aptian | Limestone | 200 |
BEL-7 | 43.157929 | –1.053770 | Early Jurassic | Marly limestone | 180 |
ETCH-1 | 43.138797 | –1.003380 | Middle Jurassic | Marly limestone | 160 |
ETCH-3 | 43.135267 | –0.998984 | Aptian | Limestone | 160 |
ETCH-4 | 43.135725 | –0.999518 | Aptian | Limestone | 220 |
ARBA-1 | 43.100010 | –1.051226 | Earliest Albian | Marl | 240 |
ARBA-2 | 43.121969 | –1.063731 | Earliest Albian | Marl | 230 |
Arberoue Unit: Iberian Necking Zone | |||||
ARB-1 | 43.332444 | –1.200691 | Late Jurassic | Marly limestone | 280 |
ARB-2 | 43.331583 | –1.199066 | Late Jurassic | Marly limestone | 280 |
ARB-3 | 43.331834 | –1.198787 | Barremian | Limestone | 260 |
ARB-4 | 43.329313 | –1.198841 | Barremian | Limestone | 260 |
ARB-5 | 43.327623 | –1.197450 | Aptian | Marl | 260 |
Hyperextended Domain | |||||
ORS-1 | 43.300258 | –1.075456 | Turonian | Marl | 250 |
ORS-2 | 43.301730 | –1.074582 | Turonian | Marl | 210 |
ORS-3 | 43.302269 | –1.074059 | Turonian | Marl | 210 |
IRI-1 | 43.257235 | –1.232790 | Santonian | Calcschist | 200 |
AGU-2 | 43.222597 | –1.271063 | Late Jurassic | Limestone | 160 |
HEL-1 | 43.293563 | –1.240483 | Santonian | Calcschist | 180 |
Measured depth (m) | True vertical depth (m) | Age | Lithology | Raman temperature (°C) |
---|---|---|---|---|
Hasparren-101 Well: Iberian Necking Zone | ||||
210 | 210 | Early Cenomanian | Marl | 130 |
420 | 420 | Early Cenomanian | Marl | 140 |
550 | 550 | Early Cenomanian | Marl | 160 |
700 | 700 | Early Cenomanian | Marl | 170 |
860 | 860 | Early Cenomanian | Marl | 190 |
1,060 | 1,060 | Early Cenomanian | Marl | 200 |
1,298 | 1,298 | Early Cenomanian | Marl | 220 |
1,500 | 1,500 | Early Cenomanian | Marl | 230 |
1,740 | 1,740 | Early Cenomanian | Marl | 240 |
2,040 | 2,040 | Early Cenomanian | Marl | 250 |
2,290 | 2,290 | Early Cenomanian | Marl | 260 |
2,790 | 2,790 | Early Cenomanian | Marl | 240/280 |
3,120 | 3,120 | Albian | Marl | 280 |
3,260 | 3,260 | Late Jurassic | Marl | 300 |
3,410 | 3,410 | Late Jurassic | Marl | 280 |
3,680 | 3,680 | Early Jurassic | Limestone | 270 |
4,120 | 4,117 | Early Jurassic | Clay | 280 |
4,390 | 4,387 | Late Triassic | Clay | 280 |
4,760 | 4,757 | Late Triassic | Clay | 240/280 |
5,120 | 5,117 | Late Triassic | Clay | 280 |
5,550 | 5,543 | Turonian | Calcschist | 240 |
5,870 | 5,827 | Turonian | Calcschist | 220 |
6,050 | 5,960 | Turonian | Calcschist | 220 |
6,220 | 6,055 | Permian | Clay | 240 |
6,270 | 6,105 | Permian | Clay | 240 |
6,280 | 6,115 | Paleozoic | Clay | 230 |
Ainhice-1 Well: Hyperextended Domain | ||||
450 | 450 | Middle Cenomanian | Calcschist | 180 |
1,078 | 1,078 | Late Jurassic | Marl | 220 |
1,454 | 1,454 | Early Jurassic | Marl | 240 |
1,585 | 1,585 | Early Jurassic | Limestone | 240 |
1,652 | 1,652 | Late Aptian | Limestone | 240 |
1,878 | 1,878 | Late Aptian | Limestone | 240 |
2,555 | 2,555 | Late Triassic | Clay | 280 |
3,005 | 3,005 | Late Triassic | Clay | 310 |
3,504 | 3,504 | Stephanian? | Limestone | 340 |
Chéraute-1 Well: Hyperextended Domain | ||||
512 | 512 | Cenomanian | Marl | 240 |
1,010 | 1,005 | Cenomanian | Marl | 240 |
1,495 | 1,475 | Senonian | Marl | 270 |
2,032 | 1,982 | Senonian | Marl | 260 |
2,488 | 2,428 | Senonian | Marl | 270 |
3,014 | 2,924 | Albian | Marl | 260 |
3,495 | 3,375 | Albian | Marl | 290 |
3,995 | 3,841 | Aptian | Marl | 300 |
4,500 | 4,270 | Aptian | Marl | 300 |
4,974 | 4,644 | Aptian | Marly limestone | 330 |
5,507 | 5,002 | Aptian | Marl | 330 |
5,997 | 5,341 | Late Jurassic | Marl | 330 |
6,015 | 5,355 | Late Jurassic | Marl | 330 |
Uhart-Mixe-1 Well: Hyperextended Domain | ||||
500 | 500 | Senonian | Marl | 230 |
613 | 613 | Senonian | Marl | 230 |
902 | 896 | Senonian | Marl | 240 |
952 | 946 | Senonian | Marl | 250 |
952 | 948 | Senonian | Marl | 250 |
1,264 | 1,254 | Senonian | Marl | 260 |
1,508 | 1,496 | Early Cenomanian | Marl | 270 |
1,735 | 1,722 | Early Cenomanian | Marl | 280 |
1,777 | 1,764 | Early Cenomanian | Marl | 280 |
Bellevue-1 Well: European Necking Zone | ||||
560 | 560 | Cenomanian | Marl | 120 |
1,180 | 1,180 | Albian | Marl | 140 |
1,550 | 1,550 | Aptian | Limestone | 160 |
2,040 | 2,037 | Aptian | Marl | 180 |
2,570 | 2,565 | Barremian | Siltstone | 200 |
3,000 | 2,992 | Late Jurassic | Limestone | 220 |
3,500 | 3,470 | Middle Jurassic | Marl | 230 |
4,000 | 3,934 | Middle Jurassic | Limestone | 240 |
4,180 | 4,101 | Late Jurassic | Marl | 240 |
4,690 | 4,560 | Barremian | Limestone | 280 |
4,730 | 4,596 | Barremian | Limestone | 240 |
5,210 | 5,050 | Late Triassic | Clay | 260 |
5,780 | 5,480 | Late Triassic | Clay | 300 |
6,430 | 6,238 | Late Triassic | Clay | 310 |
6,710 | 6,470 | Barremian | Limestone | 330 |
6,890 | 6,591 | Barremian | Limestone | 330 |
Les Cassieres-2 Well: European Proximal Margin | ||||
535 | 535 | Santonian | Marl | 160 |
968 | 966 | Santonian | Marl | 200 |
1,492 | 1,488 | Cenomanian | Marl | 200 |
1,994 | 1,990 | Cenomanian | Marl | 200 |
2,510 | 2,505 | Cenomanian | Marl | 220 |
3,083 | 3,077 | Aptian–Albian | Marl | 240 |
3,580 | 3,560 | Aptian–Albian | Marl | 240 |
4,005 | 3,960 | Aptian–Albian | Marl | 270 |
4,523 | 4,443 | Aptian–Albian | Marl | 270 |
5,015 | 4,865 | Aptian–Albian | Marl | 300 |
Orthez-102 Well: European Proximal Margin | ||||
510 | 500 | Senonian | Calcschist | 120 |
1,140 | 1,127 | Senonian | Calcschist | 140 |
2,000 | 1,979 | Albian | Limestone | 150 |
2,600 | 2,577 | Aptian | Limestone | 160 |
2,770 | 2,747 | Aptian | Limestone | 170 |
3,350 | 3,325 | Barremian | Marl | 200 |
4,030 | 3,995 | Late Jurassic | Marl | 220 |
4,200 | 4,156 | Early Jurassic | Marl | 240 |
4,260 | 4,214 | Early Jurassic | Limestone | 230 |
4,360 | 4,310 | Senonian | Marl | 240 |
4,390 | 4,338 | Aptian | Marl | 240 |
4,470 | 4,416 | Barremian | Marl | 120 |
4,690 | 4,632 | Senonian | Marl | 120 |
4,830 | 4,769 | Albian | Marl | 120 |
4,910 | 4,846 | Albian | Marl | 120 |
5,010 | 4,943 | Albian | Marl | 120 |
5,120 | 4,953 | Aptian | Limestone | 130 |
5,481 | 5,399 | Aptian | Limestone | 130 |




5.1 Iberian Proximal Margin
Our paleogeothermal analysis of the Iberian proximal margin is based on 15 outcrop samples of Santonian to Danian age (syncollisional) exposed beneath the Lakhoura thrust in the Axial Zone just south of the Mendibelza Unit (Figures 3 and 6). These samples yielded RSCM peak temperatures ranging from 160°C to 180°C in the west to 280°C in the east (Figure 5a and Table 1). Because most of the samples were from the same stratigraphic level, it was not possible to estimate a paleogeothermal gradient for this structural unit. We tried to estimate RSCM peak temperatures in the Cenomanian to Santonian carbonate platform deposits (Calcaires des Cañons limestones), but the samples are too recrystallized to yield reliable data.
5.2 Iberian Necking Zone
We calibrated RSCM peak temperatures in the Iberian necking zone from 33 outcrop samples from the Mendibelza (Figure 9a), Arbailles (Figure 9b), Saint-Etienne de Baïgorry, and Arberoue Units (Table 1) and cuttings from the Hasparren-101 well (Table 2). Ten samples were from Albian (synrift) to Santonian (postrift) deposits of the Mendibelza Unit, for which a Mesozoic stratigraphic section 2,400 m thick has been reconstructed on the basis of new field observations and previously published data (Figure 6; Boirie, 1981; Saspiturry, 2019; Saspiturry, Razin, et al., 2019; Souquet et al., 1985). The 800 m section at the base of the Albian synrift sequence reached temperatures of 230–240°C, values that are consistent with those obtained by Clerc et al. (2012). RSCM peak temperatures decreased upward from 230 to 140–160°C in the upper Albian siltstone and Santonian breccia (Figure 7). The temperature difference between the base and the top of the Mendibelza section defines a linear regression line with R2 = 0.95 defining a paleogeothermal gradient of 46°C/km (Figure 9a). To the west, six samples from the Saint-Etienne-de-Baïgorry Unit recorded the same range of RSCM peak temperatures (130–160°C) in samples from the top of the Mendibelza Unit section (Figures 3 and 5a).
Twelve samples from the Arbailles Unit represented Jurassic (prerift) to Albian (synrift) deposits (Figures 3 and 6). Surprisingly, the Jurassic to Barremian carbonate deposits, in the northern part, recorded a uniform RSCM peak temperature of 180°C whereas the younger Aptian and Albian carbonates and marls, in the southern part, recorded temperatures of 220–240°C (Figure 9b and Table 1). This range is close to the RSCM peak temperature range of 250°C and 256°C published by Clerc et al. (2012). The anomalous temperature distribution in the Arbailles Unit makes it difficult to estimate the paleogeothermal gradient there.
The site of the Hasparren-101 well was located between deltaic deposits to the west (Saint-Jean-de-Luz domain; Razin, 1989) and deep basin turbidites to the east (Saint-Palais domain; Souquet et al., 1985) during the Albian rifting stage. That position lies between the Iberian proximal margin and the deep turbiditic basin. This well penetrates two structural units separated by a major thrust at 5,400 m MD (Figure 7). The lower, autochthonous unit is made up of basement rocks and Permian red siltstone directly overlain by Turonian to Santonian carbonate turbidites. It yielded RSCM peak temperatures ranging from 220°C to 240 C (Figure 5b and Table 2). The tectonic contact between the two structural units is marked by a sharp paleotemperature gap of around 50°C. The upper, allochthonous unit is an Upper Triassic to Cenomanian sequence. The lower part of this unit yielded fairly uniform RSCM peak temperatures around 280°C between 2,800 and 5,400 m MD (Figure 7). This uniformity may reflect the presence of high-conductivity rocks such as evaporates, but these are not present in the depth range 2,800–3,900 m. Another explanation may lie in the possible presence of convective processes, which could occur if these formations (limestones, marls, and evaporites) are sufficiently permeable. Thermal convection could have homogenized temperatures over a thickness of more than 2,500 m. Raman spectra of the two samples from 2,790 and 4,760 m were heterogeneous, yielding RSCM peak temperatures of 240°C and 280°C, respectively (Figure 7 and Table 2). The upper part of the upper unit yielded RSCM peak temperatures varying from 280°C at the base to 130°C at the top. The resulting linear regression line defines a 62°C/km deformed thermal gradient (Figure 8a). The Hasparren-101 well is not included in the cross-section because (1) it is far from the cross-section line and (2) its upper unit is intensively tectonized and has variable stratal dips, thus estimating a paleogeothermal gradient based on a mean dip was problematic. However, this borehole has been integrated in this study as it bears upon the role of fluid circulation in controlling the local thermal record (see section 6.2.2).
The Arberoue field section is paleogeographically in the same domain as the Hasparren-101 well. The Arberoue succession, composed of Late Jurassic to Aptian carbonates, yielded RSCM peak temperatures, ranging from 260°C to 280°C, that are similar to those obtained from the Hasparren-101 well in the same stratigraphic interval (Figures 3a, 5a, and 7 and Table 1). The temperature difference between the base and the top of the Arberoue section defines a linear regression line with R2 = 0.91 defining a paleogeothermal gradient of 40°C/km (Figure 9c).
5.3 Hyperextended Domain
Peak temperatures in the hyperextended domain of the Mauléon basin were determined from field samples and boreholes. The distal parts of the hyperextended domain are represented in the Orsanco section (Figures 3a and 7 and Table 1), the outcrop samples HEL-1, IRI-1, and AGU-2, and the Ainhice-1, Chéraute-1, and Uhart-Mixe-1 wells.
The Ainhice-1 well reaches basement strata of Carboniferous (Stephanian) age at a depth of around 2,900 m MD (Figure 6). The Mesozoic cover is duplicated by a minor northward thrust at 1,600 m MD that was active during early Albian time (Figure 6; Saspiturry, Razin, et al., 2019). RSCM peak temperatures varied from 340°C at the base to 180°C at the top (Figures 5c and 8b and Table 2). These values fit a linear regression line corresponding to a 52°C/km apparent thermal gradient and, after correcting for a mean stratal dip of 25°, a paleogeothermal gradient of 57°C/km (Figure 8b). There is no significant variation in RSCM peak temperature across the décollement (Figure 8b).
The Chéraute-1 well penetrates a sequence of Jurassic to Upper Cretaceous rocks (Figure 7). The upper part of the well displays an Albian-Cenomanian cover sequence that is duplicated in reverse order above approximately 1,500 m MD of Upper Cretaceous deposits. The RSCM peak temperatures were more variable than in the other wells, ranging from 330°C to 290°C between 6,000 and 3,400 m MD (Figure 7 and Table 2). The folded Albian to Senonian sequence yielded RSCM peak temperatures of 240°C on top and 260–270°C at the base (Figures 5d and 7). These RSCM peak temperatures define a deformed thermal gradient of 21°C/km, although the linear regression line falls short of significance with R2 = 0.92 (Figure 8c). The complex fold structure makes it difficult to estimate a reliable paleogeothermal gradient for this well.
The Uhart-Mixe-1 well penetrates Cenomanian to Turonian turbidites (Figures 3 and 6). The RSCM peak temperatures vary from 280°C at the base to 230°C at the top (Figures 5e and 6 and Table 2). These values fit a linear regression line defining an apparent thermal gradient of 42°C/km and a paleogeothermal gradient of 60°C/km after taking into account a 45° mean stratal dip (Figure 8d).
The Orsanco field section exposes the same paleogeographic domain as the Uhart-Mixe-1 well. Samples from that section yielded RSCM peak temperatures of 210–250°C, which are similar to those in the corresponding part of the Uhart-Mixe-1 well (Figure 3a and Table 1). The temperature difference between the base and the top of the Orsanco section defines a linear regression line with R2 = 0.94 defining a paleogeothermal gradient of 60°C/km (Figure 9d).
Approximately 10 km west of the Orsanco section, outcrop samples HEL-1 and IRI-1, of Santonian age, yielded RSCM peak temperatures of 180°C and 200°C, respectively, and outcrop sample AGU-2 to their south, of Late Jurassic age, yielded a RSCM peak temperature of 160°C (Figure 3a and Table 1).
5.4 European Necking Zone
The paleotemperature of the European necking zone was derived from the Bellevue-1 and the Les Cassières-2 wells (Figures 3a and 6). The Bellevue-1 well penetrates a north directed major thrust around 4,700 m MD that divides the drilled interval into a basal and an upper unit.
The intensively deformed basal unit, made up of Upper Triassic evaporites and Barremian-Aptian carbonate lenses, is characterized by RSCM peak temperatures ranging from 330°C at the base (6,900 m MD) to 240°C on the thrust plane (Figures 5f and 8e). Despite the intense deformation, the RSCM values of this unit fit a linear regression line defining an apparent thermal gradient of around 47°C/km (Figure 8e). The tectonic contact between the basal and upper units corresponds to a temperature offset of 40°C.
The upper unit is repeated in reverse across an anticline affecting Jurassic to Albian carbonates and marls. This complex structure has been interpreted as the result of Tertiary reactivation of an Albian diapiric ridge (Saspiturry, Razin, et al., 2019). The upper unit yielded RSCM peak temperatures decreasing steadily from 280°C at the base to 120°C at the top (Figures 5f and 6 and Table 2). These temperatures fit a linear regression line defining an apparent thermal gradient of 38°C/km (Figure 8e).
The Bellevue fold is inherited from Albian-Cenomanian time, as it corresponds to a synrift diapir that controlled Albian sedimentation (Saspiturry, 2019). During the compression, the fold was thrusted northward along the Bellevue thrust. Thus, the fold is Albian in age and does not result from Pyrenean compression. It is thus older than the thermal event. RSCM peak temperatures in these two units define similar linear regression lines, meaning that the folded structure in the upper unit is clearly older than the thermal event that set the RSCM peak temperatures. In this case we cannot consider the structural data to restore the paleogeothermal gradient. However, the cross section (Figure 3b) shows that the northward Bellevue thrust doesn't create major Pyrenean tilting at the site of the Bellevue-1 well. Consequently, the paleogeothermal gradient is the same as the apparent thermal gradients, that is, 47°C/km in the basal unit and 38°C/km in the upper unit.
The Les Cassières-2 well penetrates a complete succession of Jurassic to Santonian sediments, unaffected by any Tertiary thrust (Figure 7). The Aptian sequence yielded a RSCM peak temperature of around 300°C whereas the Santonian turbidites had a peak temperature of 160°C (Figures 5g and 7 and Table 2). These data define an apparent thermal gradient of 30°C/km (Figure 8f). Taking into account a mean stratal dip of 35–40°, the paleogeothermal gradient of this well is estimated at 37–40°C/km.
5.5 European Proximal Margin
The paleotemperatures of the European proximal margin were defined using data from the Orthez-102 well (Figure 3c). The Orthez-102 well penetrates two structural units separated by the Sainte-Suzanne northward thrust at 4,400 m MD. The lower autochthonous unit, made up of Aptian to Upper Cretaceous limestone and marl, yielded RSCM peak temperatures of 120–130°C (Figure 6). The upper allochthonous unit is a complete sequence of Jurassic limestone to Senonian carbonate turbidites. The base and top of the upper unit yielded RSCM peak temperatures of 240°C and 120°C, respectively (Figures 5h and 6), for an apparent thermal gradient of 32°C/km. Taking into account a mean stratal dip of 20°, the paleogeothermal gradient of this unit is estimated at 34°C/km.
5.6 Thermal Evolution of the Hyperextended Domain: Numerical Simulation
We derived a simplified model to constrain the thermal evolution of the Mauléon basin area since 120 Ma that was based on geological knowledge plus reasonable thermal properties and boundary conditions. This model allowed us to (1) validate the coherence of the paleogradient estimated by the RSCM methodology, (2) estimate at what date the maximum temperature was reached by a rock that underwent the series of tectonic events recorded in the Mauléon basin, (3) constrain the present-day mantle heat flow values beneath the basin, and (4) estimate the synrift mantle heat flow beneath the basin.
The model calculates the thermal evolution of a crustal section undergoing four stages as shown in Figure 10: (1) a thermal pulse and sedimentation stage featuring increasing mantle heat flow from 120 to 80 Ma at the base of a thinning crust (initially 30 km thick) representing both the rifting stage and the postrift stage (refer to Figure 2 and section 2.1 for details of the tectonic stages), (2) a thermal cooling stage coupled to a sedimentation event from 80 to 40 Ma representing the onset of compression in the Pyrenees, (3) a thermal warming stage from 40 to 15 Ma, corresponding to the main exhumation and erosional stage of the Pyrenees, and (4) a thermal relaxation stage from 15 Ma to the present, characterizing the postcollisional stage in the Pyrenees.

5.6.1 Numerical Approach
Our model incorporates the thickness of the different crustal units, as well as the sedimentation and erosional rates, from this and previous studies (e.g., Saspiturry, Razin, et al., 2019; Vacherat et al., 2014, and references therein). The main unknown variable is the mantle heat flow and its variation through time. However, the following constraints have been considered: a maximum RSCM peak temperature of 600°C extrapolated at a depth of 10 km, and a present-day estimated temperature gradient of 25.0 ± 2.7°C/km (see Supporting Information S1).

From a thermal point of view, an erosion event corresponds to the vertical upward motion of crustal material at a velocity corresponding to the erosion rate, and a sedimentation event corresponds to downward motion of crustal material at a velocity corresponding to the sedimentaton rate (e.g., Turcotte & Schubert, 2002, section 4–20). To simulate the thermal effects of sedimentation and erosion events, we used the advective term of the heat equation and assigned the erosion/sedimentation rate to the vertical velocity component. Sedimentation is thus represented by a positive vertical velocity for the entire model, while erosion is assigned a negative velocity (the vertical axis being downward). Rates are calculated with the thicknesses and durations of erosion or sedimentation events shown in Figure 10.
5.6.2 Heat Equation, Boundary Conditions, and Thermal Properties


The chosen model box is 30 km thick, corresponding to the initial crustal thickness. The box length depends on the pulse wavelength: It is 80 km long for λ = 25 km but twice as long (160 km) for λ = 75 km. A fixed temperature condition (T = 10 °C) is imposed at the surface of the model. Lateral boundaries are insulating, and the basal boundary condition corresponds to a varying mantle heat flow. A fixed heat flow of 30 mW/m2 is imposed at time t = 0 (or t0, or 120 Ma). A thermal pulse (Equation 1) is imposed during the thinning phase (from t0 to t0 + 40 Myr, i.e., from 120 to 80 Ma). To reproduce the present-day surface temperature gradient, the mantle heat flow has to be decreased (Figure 10).


5.6.3 Thermal Pulse
The mantle heat flow condition (Equation 1) is applied at the bottom of the modeled region, regardless of variations in the continental crust thickness, to avoid a spatially variable boundary condition. Hence, the thermal pulse imposed at a depth of 30 km is transferred to the moving base of the crust, as well as the advected heat component during the extension phase.
The heat equation was solved by the finite-element method with Comsol Multiphysics software, in which temperature-dependent properties can be easily implemented. Different values for the maximum mantle heat flow (Qm0) were tested in order to reach 600°C at a depth of 10 km. For the small-scale thermal pulse (λ = 25 km), Qm0 is 100 mW/m2, and for the large-scale thermal anomaly (λ = 75 km), Qm0 is 71 mW/m2. In both cases, the maximum heat flow at the base of the crust (at time t0 + 40 Myr) ranges between 90 and 100 mW/m2, and the surface heat flow (which includes the radioactive component) is 135–140 mW/m2, a value consistent with surface heat flow values measured in continental rift zones (e.g., Jaupart & Mareschal, 2007; Lucazeau et al., 2010).
5.6.4 Thermal Evolution
Figure 11 shows the thermal regime of the crust from 120 Ma to present. Temperatures at three distinct depths (Figure 11a) and evolution of surface temperature gradient (Figure 11b) are shown. Figure 11c shows the evolution of temperature profiles together with the pressure-temperature-time path followed by a rock initially at the surface. The case shown in Figure 11 corresponds to Qm0 = 100 mW/m2, λ = 25 km and k(0) = 2.5 W/m K.

The model goes through four thermal stages: (I) The increased mantle heat flow at the base of the crust warms the entire crust while the coeval sedimentation and extension events have much smaller effects; (II) the crust cools due to the decrease of mantle heat flow and the effect of the sedimentation event; (III) temperatures increase due to the erosion event; and (IV) temperatures decrease toward equilibrium.
As shown in Figure 11c, the maximum temperature recorded by rock that initially (at t0) was at the surface (486°C at 8 km depth) is reached at the end of the heating phase (t0 + 40 Myr). The only way to reach higher temperatures would be to consider much higher erosion rates, which are not supported by field data.
The surface temperature gradient above the thermal pulse (Figure 11b) shows that the present-day value (t0 + 120 Myr) has recovered to its long-term value (25°C/km). Looking at time t0 + 40 Myr (age 80 Ma), the surface temperature gradient reaches 58°C/km, close to the paleogradient estimate for the hyperextended domain. However, RSCM peak temperatures that were used for paleogradient estimates do not correspond to surface values. A closer look at the calculated paleogradients, from the surface to the bottom of the model, reveals that above the thermal pulse, the temperature gradient varies within a small range of 55 to 62°C/km, as inferred from RSCM peak temperatures.
6 Discussion
6.1 Age of RSCM Peak Temperature
In this study, we defined the thermal evolution of the Mauléon hyperextended rift on the basis of a thermal numerical simulation and RSCM peak temperatures, derived from 155 outcrop and borehole samples, that ranged from 120°C to 340°C. The younger age limit of this thermal peak is constrained by the late Santonian onset of rift inversion and the resultant lowering of the thermal gradient (Labaume et al., 2016; Vacherat et al., 2014). The youngest rocks analyzed by RSCM are of early Santonian age and record RSCM peak temperatures around 180°C. Because this temperature postdates the early Santonian, it does not reflect the synrift geothermal gradient but must be attributed to a later gradient associated with burial under synorogenic sediments. The thermal simulation not only confirms the RSCM peak temperatures methodology but also shows that the high paleogeothermal gradient in the Mauléon basin center was acquired during the Albian-Cenomanian rifting stage, also indicated by a low-temperature thermochronology analysis (Vacherat et al., 2014). This high paleogeothermal gradient is consistent with the presence of a positive gravity anomaly (Figure 1b; Boillot et al., 1973; Casas et al., 1997; Chevrot et al., 2018; Daignières et al., 1994; Grandjean, 1992, 1994; Wehr et al., 2018), interpreted as the presence of subcontinental mantle at around 10 km depth (Figure 3b; Wang et al., 2016) that was locally exposed by denudation and reworked into the late Albian to early Cenomanian Urdach synrift deposits (Debroas et al., 2010; Fortané et al., 1986; Jammes et al., 2009; Lagabrielle et al., 2010; Roux, 1983) during the hyperextension of the continental crust (Masini et al., 2014; Teixell et al., 2016). The numerical simulation and the RSCM peak temperatures both indicate that the maximum peak temperature in the prerift to early postrift sediments was reached at the end of the postrift stage (~80 Ma). The low-temperature thermochronology analysis of the Mauléon basin by Vacherat et al. (2014) indicates that (1) the elevated Mauléon basin paleogeothermal gradient is inherited from the Albian-Cenomanian rifting stage and (2) rocks were heated to around 180°C soon after ~100 Ma and went through a nearly isothermal stage starting at ~80 Ma that lasted as long as ~30 Myr. This timing is also consistent with geochronological data from elsewhere in the North Pyrenean Zone, which put the peak at 107 and 85 Ma (Albarède & Michard-Vitrac, 1978b; Golberg et al., 1986; Golberg & Maluski, 1988; Montigny et al., 1986; Thiébaut et al., 1992). This high synrift gradient lasted at least through the postrift stage and possibly until 50 Ma, as proposed by Vacherat et al. (2014). As in the Mauléon basin, the Cameros basin in Spain has evidence of an elevated paleogeothermal synrift gradient in its deepest part (Golberg et al., 1988; Guiraud & Séguret, 1985; Rat et al., 2019). However, the high temperature-low pressure metamorphism reached its peak temperature during the postrift stage (Casas-Sainz & Gil-Imaz, 1998; Casquet et al., 1992; Golberg et al., 1988; Mata et al., 2001).
6.2 Synrift Paleogeothermal Gradient
6.2.1 Proximal Margins
The upper unit of the Orthez-102 well, in the European proximal margin, recorded a paleogeothermal gradient of 34°C/km. This gradient is consistent with the average geothermal gradient of ~30°C/km in continental domains and the average continental heat flow of 80 mW/m2 (Jaupart & Mareschal, 2007). The isotherms in this borehole document northward transport of the warmer upper unit (240°C) over the cooler autochthonous unit (120°C). This latter RSCM peak temperature affects Aptian-Albian deposits of the autochthonous unit whereas the same stratigraphic levels of the allochthonous unit recorded higher RSCM peak temperatures of around 160–200°C. Consequently, we interpret the Orthez-102 autochthonous unit as representing a more proximal part of the European margin.
As mentioned in section 5.1, the paleogeothermal gradient could not be estimated for the Iberian proximal margin. The 15 Maastrichtian outcrop samples collected on the footwall of the Lakhoura thrust recorded RSCM peak temperatures of around 180°C in the Mendibelza domain (Figure 3a and Table 1). These temperatures are comparable to those obtained in the Cenomanian to Santonian breccias of the Lakhoura thrust hanging wall (140–160°C), but they affect younger sediments in the Lakhoura thrust footwall. Taking into account (1) the ~30°C/km thermal gradient of the proximal margin, (2) the temperature reached by the Santonian breccias in the Lakhoura thrust hanging wall (140–160°C), and (3) the regional thickness of the Campanian-Maastrichtian sedimentary pile, we can estimate the peak temperature reached by the eroded Maastrichtian sequence in the Lakhoura thrust hanging wall as around 100–120°C. This is lower than the peak temperatures in the Maastrichtian sequence of the Lakhoura thrust footwall (180°C). Thus, on the Iberian margin, the peak temperature largely postdates the rifting stage. This scenario is supported by thermochronology data (Bosch et al., 2016) showing that the region of Mendibelza and Lakhoura and both hanging wall and footwall cooled from 180°C or higher (U-Th/He on zircon) during early Eocene time, recording exhumation of the Lakhoura thrust domain while the Mauléon basin core reached peak temperatures soon after the rifting stage (Vacherat et al., 2014). However, thermochronology analyses constrain exhumation in the Lakhoura region during the Eocene but the age of Lakhoura thrust initiation is not resolved by these data and could be Late Santonian in age.
6.2.2 Necking Zones
The Iberian necking zone is represented by the Mendibelza (Figure 9a), Arbailles (Figure 9b), and Arberoue (Figure 9c) field sections and the Hasparren-101 borehole. This domain had higher paleogeothermal gradients than normal stable crust (30°C/km), 46°C/km in the Mendibelza section and 40°C/km in the Arberoue section. In the Arbailles field section, the RSCM peak temperature increases toward the younger stratigraphic levels, being 180°C in the Jurassic sequence and 230–250°C in the Albian synrift sequence. This increase could be interpreted as the result of early Cenomanian southward tilting of the Iberian necking zone (Saspiturry, Razin, et al., 2019) that occurred before peak temperatures were reached. To the northwest, the Hasparren-101 borehole presents a more complex thermal record. The thrust plane at 5,380 m depth is marked by an abrupt 40°C change in the RSCM peak temperature from 240°C in the footwall to 280°C in the hanging wall (Figure 7). In our interpretation, this thrust transported a basinal section over a more proximal one.
The lower part of the allochthonous unit displayed an adiabatic temperature gradient over a 2.5 km interval with a constant 280°C temperature, whereas the upper part yielded a 62°C/km deformed thermal gradient (Figure 8a). As mentioned in section 4.2, the homogeneous temperature was probably due to upward circulation of hot fluid from 5.1 to 2.8 km (hot upwelling), but it could also correspond to downward circulation of cold fluid from 2.8 to 5.1 km (cold downwelling). Such temperature profiles, in which an elevated temperature gradient at the surface overlies a zone of constant temperature over several hundreds to thousands of meters, are often seen in geothermal systems typified by hot upwelling (e.g., Guillou-Frottier et al., 2013; Muraoka et al., 2000). However, in this borehole, we could not discriminate between hot upwelling and cold downwelling. Although the 62°C/km deformed thermal gradient was probably influenced by fluid transport, it is not necessarily higher than it would be with no fluid circulation. For example, in the case of a permeable medium where permeability is depth-dependent, the surface temperature gradient over a zone of cold downwelling may be lower than the purely conductive case (see supporting information Text S2). It is thus problematic to estimate a paleogeothermal gradient in the presence of a convective zone.
In the European necking zone, the paleogeothermal gradient was estimated using the Bellevue-1 borehole just north of the Saint-Palais thrust (Figure 3a). The Bellevue-1 well intersects a complex structure corresponding to the Tertiary reactivation of an Albian diapiric ridge (Saspiturry, Razin, et al., 2019). The apparent thermal gradient shows, however, that the RSCM peak temperatures postdate the formation of this structure. Thus, we considered the paleogeothermal gradient to be very close to the apparent gradients or 38°C/km in the allochthonous unit and 47°C/km in the autochthonous unit (Figure 8e). These values are comparable to those obtained in the Iberian necking zone.
The Les Cassières-2 well, in the European necking zone, penetrates a well-preserved sedimentary succession undisturbed by major Pyrenean thrusting (Figure 3a). We consider it as a reference section for the European necking zone that defines a paleogeothermal gradient of ~37–40°C/km.
6.2.3 Hyperextended Domain
The paleothermal structure of the hyperextended domain was calibrated using the Chéraute-1, Uhart-Mixe-1, and Ainhice-1 boreholes (Figure 3a) and the Orsanco field section (Figure 9d). The Chéraute-1 borehole is just east of the Saison structure that accommodates the major Roquiague diapir (Figure 3a; Canérot, 1988, 1989, 2008). We interpret the fold in this well as having been induced by the development of the Roquiague diapir during Senonian time (Figure 7). The well is characterized by a steady rise in RSCM peak temperatures from 240°C to 330°C across a sedimentary succession 5,350 m thick, defining a deformed thermal gradient of 21°C/km (Figure 8c and Table 2). Heat refraction by thermally conductive Upper Triassic evaporites could be responsible for the homogenization of temperatures across the sedimentary basin infill near this structure. Consequently, the measured paleogeothermal gradient is incompatible with the gradients in the Uhart-Mixe-1 and Ainhice-1 wells. These two boreholes lie between the Saint-Jean-Pied-de-Port and Saison transfer zones (Figure 1; Canérot, 2008). The Ainhice-1 well, located at the transition with the Iberian necking zone, yielded an unusually high paleogeothermal gradient of 57°C/km (Figure 8b). Because the paleogeothermal profile was not disturbed where a packet of strata ~400 m thick was duplicated above the décollement at 1,600 m, the peak temperatures there clearly postdate the displacement on the décollement. The Uhart-Mixe-1 well, in the heart of the hyperextended domain, is free of major Pyrenean deformation. It too yielded an unusually high paleogeothermal gradient of 60°C/km (Figure 8d), as did the Orsanco field section (Figure 9d).
6.3 The Mauléon Basin Internal Metamorphic Zone
Considering the 60°C/km paleogeothermal gradient obtained using the RSCM peak temperatures in the Uhart-Mixe-1 well (Figure 8d), which is fully consistent with the gradient proposed on thermochronological grounds by Vacherat et al. (2014), we can extrapolate the temperature from the base of the well into the hyperextended domain in this sector of the Mauléon basin, between the Saint-Jean-Pied-de-Port and Saison transfer faults (Figure 1c) and estimate that the prerift cover reached a temperature of 500–600°C (Figure 12). This estimate is supported by the thermal simulation that indicated a maximum temperature of 600°C at the base of the Mauléon basin prerift cover. This temperature range is similar to those documented in the Internal Metamorphic Zone in the central and eastern Pyrenees (e.g., Azambre et al., 1992; Bernus-Maury, 1984; Chelalou et al., 2016; Clerc, 2012; Clerc et al., 2015; Golberg & Leyreloup, 1990; Vauchez et al., 2013) and the Nappes des Marbres in the Basque-Cantabrian basin (Ducoux, 2017; Lamare, 1936; Martínez-Torres, 1989; Mendia & Ibarguchi, 1991). The Internal Metamorphic Zone has been interpreted as the inverted base of the North Pyrenean hyperextended rift domain (Clerc, 2012; Clerc et al., 2015; Clerc & Lagabrielle, 2014; Ducoux, 2017; Lagabrielle et al., 2016). Consequently, we can infer that marble corresponding to the Internal Metamorphic Zone marbles exists in this sector of the Mauléon basin. However, farther eastward, the base of the basin does not exceed 350°C, as indicated by RSCM peak temperatures in the Chaînons Béarnais prerift cover (Corre, 2017) and Paleozoic basement (Asti et al., 2019). Contrary to the great east-west extent of the Internal Metamorphic Zone in the central and eastern Pyrenees, its extension in the Mauléon basin appears to be restricted to a very small area. Thus, we can deduce that the Internal Metamorphic Zone hyperextended basin was much wider than the hyperextended Mauléon system, raising a question about the way Cretaceous extension is distributed along the Pyrenean rift system.

6.4 Distribution of the Synrift Paleogeothermal Gradient and Implications
The paleogeothermal gradients we obtained vary in the different structural units of the inverted hyperextended Mauléon basin. This variation is clearly evident along a proximal-distal margin transect: the Cretaceous paleogeothermal gradient increases basinward from ~34°C/km near the European proximal margin to ~37–47°C/km in the two necking zones to 57–60°C/km in the hyperextended domain (Figure 12b). The distribution of synrift paleogeothermal gradients is similar in the Camèros basin in Spain, where the rifting stage developed under a high thermal gradient (~70°C/km) in the basin core (Del Río et al., 2009; Mata et al., 2001). As in the Mauléon basin, the paleothermal gradient decreases to 41.5°C/km near the edge of the Camèros basin, along with the intensity of the high temperature–low pressure metamorphism (Omodeo-Salé et al., 2017). In the Mauléon basin, Lescoutre et al. (2019) and Lescoutre et al. (2019) have proposed that the paleogeothermal gradient had an asymmetric distribution in response to Early Cretaceous simple shear thinning. However, this study indicates that the Cretaceous paleogeothermal gradient was symmetric. This gradient is responsible for differing deformation styles across the basin that resulted in the formation of a pseudo-symmetric smooth-slope type extensional basin (Lagabrielle et al., 2020) in which the proximal margins underwent brittle deformation while the distal hyperextended domain underwent dominantly ductile thinning. This bimodal deformation style was linked to the basinward increase in the thermal gradient and burial of the central basin under thick sedimentary cover. The movement of the thick prerift sedimentary cover and synrift sediments into the hyperextended domain, aided by salt tectonics, accentuated the existing thermal gradient and led to ductile thinning of the upper continental crust under high-temperature/low-pressure metamorphic conditions at temperatures higher than 450°C (Lagabrielle et al., 2020).
6.5 Role of Rift Inheritance on Postcollisional Thermal Imprint
The Pyrenean compression on the Mauléon basin was not intense (Figures 1c and 12), and high-temperature marbles were not removed from the base of the hyperextended domain on the Saint-Palais thrust on its northern edge. Indeed, using numerical modeling, Jourdon et al. (2019) have shown that folding was limited for two reasons: (1) The deformation in the sedimentary cover was decoupled from the basement by Triassic evaporites, and (2) the position of the Mauléon basin in the Pyrenean retro-wedge led to its tectonic inversion being less pronounced, resulting in good preservation of precollision markers. Moreover, rift inversion started during late Santonian to Campanian time in the eastern Pyrenees (Choukroune & Etude Continentale et Océanique par Réflexion et réfraction Sismiques [ECORS] Team, 1989; Ford et al., 2016; Macchiavelli et al., 2017; Muñoz, 1992; Ternois et al., 2019) and in late Eocene time in the western Pyrenees (Labaume et al., 2016; Teixell, 1993). This asynchronous compression should have resulted in greater shortening in the central and eastern Pyrenees (Beaumont et al., 2000; Mouthereau et al., 2014; Muñoz, 1992; Vergés et al., 1995) than in the western Pyrenees (Teixell, 1996, 1998). Consequently, in the central and eastern Pyrenees, the Internal Metamorphic Zone was uplifted and exposed during Pyrenean compression whereas it is still buried in the Mauléon basin.
The previously acquired postrift isotherms were folded and tilted inside the Mauléon basin pop-up structure, bounded on the north by the Bellevue and Sainte-Suzanne thrusts and on the south by the Lakhoura thrust (Figure 12a). On the European margin, our evidence demonstrates that the postrift isotherms were not erased by the Pyrenean thrusting but instead were passively transported onto the proximal margin. On the Iberian margin, the Lakhoura thrust is marked by an increase of peak temperatures on its footwall, meaning that these peak temperatures were acquired after the thrusting.
The different thermal responses on the opposite sides of the pop-up structure can be linked to their tectonic reactivation style. Indeed, exhumation is limited in the northern retro-wedge, which was affected by thin-skinned deformation on an Upper Triassic salt décollement, and significantly greater in the southern prowedge, which underwent thick-skinned deformation (Figure 12a; Jourdon et al., 2019, 2020; Saspiturry, Allanic, et al., 2020). The post thrusting thermal structure along the Lakhoura thrust system was preserved due to sequential activation of thrusting southward and below the Lakhoura thrust system. This explanation is consistent with thermochronology data (Bosch et al., 2016) showing that the basement was buried at 5–6 km depth at 30 Ma. Subsequently, deformation propagated southward and the Axial Zone was exhumed more than the North Pyrenean Zone.
On the European margin and in the hyperextended domain of the Mauléon basin, a major decrease in the thermal gradient to the 25.0 ± 2.7°C/km gradients in modern boreholes (see supporting information S1) followed the onset of compression. In the Iberian margin, the northward motion of the Iberian slab under the Mauléon hyperextended domain greatly decreased the heat flow from the asthenosphere. The European margin, to the contrary, never underwent thick-skin tectonics, explaining the current Moho depth of ~27 km there (Figure 12b). Instead, shortening of the European margin was accommodated by the Bellevue and Sainte-Suzanne thin-skin thrusts. The unusually low thermal gradient there is directly controlled by the thickness of the European continental crust and the presence of nonradiogenic subcontinental mantle at shallow depths.
6.6 Implications for Petroleum Systems
The peak temperature of rocks is a key parameter in the thermal evolution of basins. Many geothermometers, such as illite crystallinity, fluid inclusion microthermometry, and vitrinite reflectance, are not easily applicable in continental margin contexts such as the inverted Pyrenean system. In this study, we propose an integrated thermal approach, combining new analytical (RSCM thermometry) and numerical methods, to reconstruct the thermal history of hyperextended rift systems. The Mauléon inverted hyperextended basin represents an analog of current tectonostratigraphic settings targeted by petroleum companies; thus, its thermal evolution is of broad interest. Indeed, laboratory and numerical experiments directed at thermal interactions between the conducting continental crust and the convective mantle (e.g., Grigné et al., 2007; Guillou & Jaupart, 1995) have long indicated the presence of high heat flows and thermal gradients at continental margins. The insulating tendency of continents favors low mantle heat flow (10–20 mW/m2) in intracontinental settings and much higher mantle heat flows at continental margins (more than 80 mW/m2; e.g., Nirrengarten et al., 2019). Our study shows that continental margins have high mantle heat flows of ~100 mW/m2 at the base of the hyperextended domain, as indicated by our basin thermal modeling (section 5.6). This high mantle heat flow is consistent with the 55–60°C/km thermal gradient, acquired at the end of the rifting stage, estimated by RSCM thermometry and thermal modeling. These values are consistent with other estimated paleotemperature gradients in continental rifts, which range from 50 to 100°C/km in Antarctica (Berg et al., 1989), 80–100°C/km in Iceland (Bertani, 2017), and 70°C/km in Spain (Del Río et al., 2009; Mata et al., 2001). Our estimated paleogeothermal gradients for the Mauléon hyperextended domain are comparable to other recent estimates: An estimate of 60–75°C/km by Corre (2017) was based on RSCM data, and estimates of 80°C/km by Vacherat et al. (2014) and Hart et al. (2017) were based on detrital zircon fission-track data and (U-Th-Sm)/He thermochronology data, respectively. In the Mauléon basin, the prerift cover was heated to temperatures as high as 500°C in the hyperextended domain and 150–200°C in the proximal margin. Thus, potential source rocks are overmature in the hyperextended domain and within the oil to gas window near the basin margins at the end of the rifting stage.
7 Conclusion
In this study, we estimated the paleothermal structure of the Mauléon hyperextended rift, using thermal numerical simulation and RSCM peak temperatures obtained from 155 outcrop and borehole samples. The Mauléon basin recorded an unusually high heat flow following Albian-Cenomanian hyperextension. Although the basin's elevated paleogeothermal gradient is inherited from the Albian-Cenomanian rifting stage, the numerical simulation and the RSCM peak temperatures indicate that the prerift to early postrift sediments reached their maximum temperature at the end of the postrift stage (~80 Ma). Thus, peak temperature postdates hyperextension, and the prerift to postrift sedimentary infill was thus affected by a paleogeothermal gradient whose isotherms crosscut all older tectonic structures.
The paleogeothermal gradients estimated in this study probably reached their maximum during Paleocene to early Eocene time. They rise in magnitude from the proximal domains to the center of the Albian-Cenomanian rift, starting from ~34°C/km on the European proximal margin (typical of stable cratons) to ~37–47°C/km in the two necking zones and to 57–60°C/km in the hyperextended domain. This finding is strengthened by the numerical simulation that reveals temperature gradients above the thermal pulse varying within a small range (55–62°C/km). The area of highest gradient corresponds to the position of a positive gravity anomaly, interpreted as the presence at shallow depth (~10 km) of subcontinental mantle. Extrapolating the temperature at the base of the hyperextended domain, we infer that it reached 500–600°C in the segment of the basin between the Saint-Jean-Pied-de-Port and Saison transfer faults. These temperatures are similar to those measured in marbles of the Internal Metamorphic Zone in the central and eastern Pyrenees. However, the east-west divisions of the Mauléon basin high temperature are very narrow compared to those of the Internal Metamorphic Zone of the central and eastern Pyrenees. This finding suggests that the Albian-Cenomanian Metamorphic Internal Zone hyperextended basin was much wider than in the western Pyrenees.
The pop-up structures on the northern and southern flanks of the Mauléon basin, formed during later compression across the rift, present differing postcollisional thermal responses. The European margin was affected by northward thrusting that transported the paleoisotherms across the proximal margin without heating the thrust footwalls, a sign that the shortening rate was slow or that erosion was intense during the compression. As a result, the precollisional maximum paleogeothermal gradients of both the hyperextended domain and the European margin were preserved. On the Iberian margin, the geothermal gradient increased as the footwall of the Lakhoura thrust underwent heating after the collision, which reset its precollisional paleogeothermal record. This gradient increase can be linked to thickening of the Iberian continental crust following the formation of the antiformal crustal stack in the Axial Zone. The northward motion of the Iberian slab beneath the Mauléon hyperextended domain is responsible for a steep decrease in the asthenospheric heat flow and explains the low postcollisional geothermal gradient (25.0 ± 2.7°C/km) in the Mauléon basin core and European margin. Unlike the Iberian margin, the European margin never underwent thick-skinned tectonics and retained its current Moho depth of 25–27 km. Its low geothermal gradient is the consequence of a thin European continental crust and the presence at shallow depth of nonradiogenic subcontinental mantle.
Acknowledgments
This work is part of the Orogen geological research project cofunded by Total S.A., BRGM, and Institut national de sciences de l'Univers (INSU). We thank Orogen project managers Emmanuel Masini (Total), Olivier Vidal (CNRS), and Isabelle Thinon (BRGM). Review comments by Patrice Baby, Juan Contreras and three anonymous reviewers, and Djordje Grujic and Laurent Jolivet of the editorial board significantly improved the initial manuscript.
Open Research
Data Availability Statement
Data sets for this research are available in a Mendeley Data repository (https://data.mendeley.com/datasets/47kgv7r9wm).