The Late Cretaceous batholithic massifs of Sierra La Laguna and Sierra La Trinidad, southern Baja California, Mexico: constraints on extensional structures from geology, geochronology, and thermobarometry

ABSTRACT The Los Cabos Block, at the southern end of Baja California, exposes part of the Late Cretaceous Peninsular Ranges Batholith over 2700 km2. In the Los Cabos block, the Los Cabos Batholitic Complex is exposed in two massifs, Sierra La Laguna to the west and Sierra La Trinidad to the east. The two sierras are separated by the ~75 km long, N-S striking, San José del Cabo extensional fault system, and the homonymous basin. Here, we present field observations, new U–Pb and Ar–Ar ages, and thermobarometric determinations to constrain the magmatic history of this composite batholith and the structural relations between Sierra La Laguna and Sierra La Trinidad. Field observations show that the batholith at Sierra La Laguna comprises several granodiorite bodies, felsic sills, and dikes with complex magma mingling and mixing features typical of an upper-middle crustal system within a mush zone of intermediate to silicic composition. U–Pb zircon crystallization ages for La Laguna massif define the main period of plutonism between ~89.9 and 73.9 Ma, with progressively younger ages from west to east. Zircon chemistry and published whole-rock isotopic data point to mantle contributions to arc magmatism with significant crustal thickening with time. Hornblende thermobarometry suggests a wide range interval of crystallization (depth: 12–18 km; temperature: 670–740°C). Sierra La Trinidad has a more felsic and narrow compositional range, with crystallization ages like those from the eastern, younger part of Sierra La Laguna and lower crystallization depth, indicating that it represents the shallower portion of the Los Cabos Batholitic Complex magmatic system. The integration of our study with previous research indicates that Sierra La Trinidad is the hanging wall block of the San José del Cabo fault system, which implies a low angle geometry (<30°) and older age of activation than previously considered (>18 Ma).


Introduction
The construction of granitic batholiths preserves a record of magmatic processes that operate at shallow to mid-crustal depths (Bateman 1992;Ducea 2001;DeCelles et al. 2009;Paterson et al. 2011).Several studies suggest that plutonic igneous bodies are the product of accretion of discrete intrusions or magma batches at different crustal levels and time scales (e.g.Glazner et al. 2004;Bachman et al. 2007;Menand 2008;Annen et al. 2015;Samperton et al. 2015;Schaltegger et al. 2019).Syn-plutonic intrusions of more mafic magmas are common features in calc-alkaline batholiths, where such intrusions and mafic enclaves provide evidence of magma mixing and mingling processes (Barbarin and Didier 1992;Barbarin 2005;Rodríguez and Castro 2019).
During batholith construction, structural inheritance is important for future localization of deformation along major lithological boundaries (Butler et al. 2006) and the influence of inheritance on influences the architecture and tectonic evolution of rifted margins (Manatschal et al. 2015).The occurrence of inherited structures from previous thrust belts or remnants of subduction zones results in mechanically weak zones corresponding to ancient orogenic events that can be reactivated during rifting processes (Tommasi and Vauchez 2001;Manatschal et al. 2015).
The opening of the Gulf of California separated the Baja California peninsula from mainland Mexico and fragmented a major Cretaceous-Paleocene batholith belt, whose origin is related to the subduction of the Farallon Plate beneath the North American Plate (Gastil 1975;Silver and Chappell 1988;Ortega-Gutiérrez et al. 2014;Valencia-Moreno et al. 2021).The batholiths are presently exposed parallel to the coast in the northwestern mainland of Mexico and the eastern part of the Baja California peninsula (Figure 1).The western part of the batholithic belt, known as the Peninsular Ranges Batholith (PRB) (Silver and Chappell 1988), is well exposed in Mexico along the eastern part of northern Baja California, whereas in southern Baja California, it is mostly buried beneath Eocene to Pliocene volcanic and sedimentary deposits, with discontinuous exposures in rifted continental blocks, both emerged and submerged, within the Gulf of California (Duque-Trujillo et al. 2015).The PRB is well exposed again at the southern end of the peninsula in the Los Cabos block (Schaaf et al. 2000), where it has been included into the Los Cabos Batholithic Complex (LCBC) (Figure 1a).The exhumation of the LCBC has been related to the opening of the Gulf of California and rifting since the Middle Miocene, from the age-equivalent Puerto Vallarta batholith of mainland Mexico (Fletcher and Munguía 2000;Fletcher et al. 2007;Bot et al. 2016;Schaaf et al. 2020).
The LCBC, exposed in a 150 km x 60 km area between La Paz and Cabo San Lucas (Figure 1b), is mainly formed by a composite batholith, separated into two massifs by the N-S striking San José del Cabo Fault (SJCF) and the homonymous basin.The western massif, Sierra La Laguna, is the largest and highest, with a maximum elevation of ~2,000 m.a.s.l.The eastern massif, named Sierra La Trinidad, has a maximum elevation of ~850 m.a.s.l.Despite its exceptional exposure and its crucial role in the evolution of the Gulf of California rift, few studies have addressed the magmatic and cooling history of the Los Cabos block (Pérez-Venzor 2013; Díaz-López 2019, unpublished PhD and MSc theses, respectively).Significantly, only a few intrusion ages have been published (Díaz-López 2019;Camarena-Vázquez et al. 2022), possibly not representative of the whole age range of the batholith assembly, and the depth of crystallization of the batholith at Sierra La Laguna is unknown.On the other hand, the structural relation between Sierra La Laguna and Sierra La Trinidad and the geometry and age of the SJCF are not well known.Based on apatite and zircon fission-track ages for samples from three sites along the SJCF, Fletcher et al. (2000) proposed that the SJCF has accommodated ~5.2-6.5 km of exhumation since ~12-10 Ma.In turn, Bot et al. (2016) proposed that the northern part of the SJCF is a low-angle detachment fault active since ~18 Ma.
In this paper, we report field relations, new U-Pb and Ar-Ar ages, and thermobarometric determinations for the eastern part of Sierra La Laguna intrusive bodies.We also provide U-Pb detrital ages for the sedimentary formations of the San José del Cabo basin and U-Pb ages for igneous and volcanic rocks of Sierra La Trinidad.
Our data contribute to characterize the magmatic history of the LCBC and confirm the structural relation between Sierra La Laguna and Sierra La Trinidad.

Geological setting
The LCBC is emplaced within a pre-Cretaceous metamorphic assemblage of the Guerrero Superterrane (Aranda-Gómez and Pérez-Venzor 1988, 1989).Two granitic belts have been distinguished that constitute the Los Cabos block, the western one comprises dominantly mafic and foliated plutons in the Sierras El Novillo and La Gata, as well as in the vicinity of La Paz, in San José and Cerralvo Islands, down to the city of Todos Santos (Figure 1).The eastern belt comprises younger, almost undeformed, and more felsic rocks (Aranda-Gómez et al. 1989;Schaaf et al. 2000).Schaaf et al. (2000) provided the first geochronological and paleomagnetic study of the LCBC, reporting Rb-Sr isochron ages between 118 and 98 Ma for the deformed granitoids, ~116 Ma for a suite of mafic to ultramafic intrusive rocks near El Novillo (western belt), and Rb-Sr ages of 93-90 Ma for undeformed granitoids that make up most of the eastern belt and the LCBC.Kimbrough et al. (2014) reported two gabbro belts in the Los Cabos block with ages from 130 to 100 Ma in the west and 110-98 Ma in the east.Based on lithologic, geochronologic, and isotopic similarities, several authors correlated the LCBC with the Puerto Vallarta Batholith (PVB) located within the Mexican mainland (Figure 1a) and considered that the LCBC was juxtaposed to the northwestern part of the PVB before the opening of the Gulf of California (Schaaf et al. 2000(Schaaf et al. , 2020;;Fletcher et al. 2003Fletcher et al. , 2007;;Ferrari et al. 2013;Gutiérrez-Aguilar et al. 2021).In the following, we briefly review the main geologic units and geochronologic data of the eastern Los Cabos block (Figure 2).
The meta-igneous unit corresponds to the orthogneiss of Boca La Sierra (Figure 2) and the amphibolitic gneiss of La Palma (not differentiated in Figure 2 because of its size), associated with a migmatitic belt interpreted as a ductile shear zone (Pérez-Venzor 2013

Igneous rocks
The LCBC is an intrusive igneous suite with multiple dikes, sills, enclave swarms, and mingling structures, typical of a long-lasting and pulsating magmatic system.K-Ar and Ar-Ar cooling ages from plutonic rocks around Los Cabos block were provided by Gastil et al. (1976), Frizzell et al. (1984), Ortega-Rivera (2003) and Grove (pers.comm.)(Supp.Table S1).In the more detailed study of Sierra La Laguna by Pérez-Venzor (2013), several granitic to granodioritic plutons were identified (La Palma, Matancita, Virgencita, and Buenos Aires) plus the Los Llanitos intrusive unit, in which he grouped the more mafic rocks appearing as apophysis, enclaves, and disaggregated dikes, sometimes described as 'magmatic breccias', as well as porphyritic dikes of dacitic composition (Mata Gorda unit) and younger mafic dikes.For different lithologies, Pérez-Venzor (2013) reported Ar-Ar cooling ages in the range ~80-59 Ma and Rb-Sr isochron ages of ~109 and ~80 Ma (Supp.2019), who obtained U-Pb crystallization ages in the range 82-74 Ma and estimated a depth of intrusion in the range of 7-11 km from thermobarometric data.
The geologic information provided by Díaz-López (2019) and our own observations indicate that metamorphic rocks and mafic intrusions are absent at Sierra La Trinidad, which in turn is partly covered by felsic volcanic rocks.
transgression within the SJCB.This formation is well exposed along the northern prolongation of the SJCB (named Los Barriles fault) and consists of conglomerate and sandstone dominated by granite fragments (Schwennicke et al. 2017).El Refugio Formation is composed of limestone and shale with Pliocene marine fossils (Martínez-Gutiérrez and Sethi 1997).The youngest unit within the basin is the El Chorro Formation, which overlies all younger units in angular unconformity, is composed of sandstone and shale deposited in an alluvial system and has an estimated deposition age of Late Pliocene to Pleistocene (Arreguín-Rodríguez and Schwennicke 2013).The deposit of the sedimentary succession of the San Jose del Cabo Basin occurred in an endorheic basin and was controlled by the SJCF activity (Martínez-Gutiérrez and Sethi 1997;Bravo-Pérez, 2002;Fletcher et al. 2003Fletcher et al. , 2007;;Schwennicke et al. 2017)

Sampling strategy and preparation
Rock samples were selected from different lithologies of the Los Cabos Batholithic Complex, particularly in the eastern part of Sierra La Laguna in proximity to the San José del Cabo Fault, based on field observations and previous descriptions by Pérez-Venzor (2013).Thin polished sections were prepared at Centro de Geociencias, UNAM.A total of 20 samples were selected for U-Pb and Ar-Ar geochronology and some of them for thermobarometric studies (Tables 1  and 2).A detailed petrographic study was carried out, and granitoids were classified according to Le Maitre et al. (2005).Mineral abbreviations in tables and figures follow the nomenclature of Whitney and Evans (2010).For U-Pb dating, zircon crystals were separated using conventional crushing and separation techniques and were then handpicked under a binocular microscope.Zircons were mounted in epoxy resin and polished to expose the grain cores for photomicrography and cathodoluminescence (CL) prior to U-Pb dating.Cathodoluminescence images of representative zircon grains are shown in Supp File 1.Samples for Ar-Ar dating were prepared at Departamento de Geología, Centro de Investigación Científica y de Educación Superior de Ensenada (CICESE), Mexico, using standard methods of crushing, sieving, magnetic separation, and hand-picking.

LA-ICP-MS U-Pb and trace-element analyses
U-Pb zircon dating and trace element measurements in zircons were performed by Laser Ablation coupled to Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) at Laboratorio de Estudios Isotópicos, Centro de Geociencias, UNAM-Campus Juriquilla, Mexico.Analytical conditions are described in Solari et al. (2018) and in Supp.File 2. Measured elements include Ti, P, S, Y, Nb, La, Ce, Pr, Nd, Sm, Eu, Cd, Tb, Dy, Ho, Er, Yb, Lu, Hf, Pb, Th, and U. A laser spot diameter of 23 μm, a low energy density, and a pulse frequency of 5 Hz ensured conditions that minimize isotopic fractionation.No common lead correction was applied as the isobaric interferences on the small 204 Pb signal are not resolvable with the instrument used.For analysis, we selected only crystals without obvious damage or inclusions.Age calculations were performed using IsoplotR software (Vermeesch 2018).For magmatic samples, concordia crystallization ages were calculated from the group of U-Pb data that agree within 2 standard error uncertainties, and data were further filtered to include only those zircons with less than 30% of discordance.This procedure was applied to extract the zircon population representing the main crystallization phase in each sample, looking for a more straightforward comparison between samples with variable record of protracted crystallization.The distribution of detrital zircon ages was obtained from Kernel density estimates.A summary of obtained U-Pb ages is reported in Table 1.U-Pb data and trace element abundances are reported in Supp.Table S2.

Ar-Ar age determinations
Fifteen mineral ages were obtained for hornblende, biotite, plagioclase, and K feldspar from seven samples of the LCBC from step-heating experiments of multigrain samples at Laboratorio Interinstitucional de Geocronología de Argón (LIGAr), Centro de Geociencias, UNAM-Campus Juriquilla, Mexico.Samples and neutron fluence monitors FCT-2C (28.198 ± 0.044 Ma;Kuiper et al. 2008) and HD-B1 (24.18 ± 0.09; Schwarz and Trieloff 2007) were irradiated in position 8C at the McMaster University Nuclear Reactor (Hamilton, Ontario), using Cd shielding.A Coherent Innova 200-20 Ar-ion laser or a Pond Engineering TC-9 low blank furnace were used for gas extraction, and argon isotopes were measured with an Isotopx NGX multicollector noble gas mass spectrometer.For data reduction, the software NGX-Red 1.0® and AgeCalc 1.0® developed at CICESE were used.Detailed information on Ar-Ar procedures, analytical data, and complete results can be found in Supp.Text methods and Supp.Table S3.Obtained ages are reported in Table 1.

Amphibole mineral chemistry
Amphibole analyses for thermobarometric calculation were obtained at the Instituto de Geofísica, UNAM, Unidad Michoacán, Mexico, with a JEOL JXA-8230 Electron Probe Microanalyzer equipped with five wavelength dispersive spectrometers, using a 5 μm beam at an acceleration voltage of 15 keV, and 10 nA beam current.Detection limits depend on the concentration of the element in the sample, and there is no definition per se for the equipment.This microprobe has a response up to 10 ppm.It has a hardware error of ±1.5% W. Natural and synthetic minerals were used as calibration standards: spinel, Kakanui hornblende, VG-568 glass, and fluorite.Obtained mineral compositions and detection limits are reported in the Supp.Table S4.

Field description and petrography of Los Cabos Batholithic complex at Sierra La Laguna
The LCBC exposed at Sierra La Laguna comprises an intrusive suite ranging in composition from tonalite to alkali feldspar-granite that intruded a metamorphic suite of orthogneiss, paragneiss, calcsilicate, and schist with migmatite development.Granodiorite is the dominant rock in volume (>70%); rocks with granite, diorite, quartz diorite, and tonalite compositions are also present.Dikes and sills occur with thicknesses of tens to hundreds of metres.Abundant micro-dioritic enclaves are associated with mixing and mingling zones.In the field, we did not identify discrete plutons with sharp contacts.Instead, the LCBC can be described as a regionally extensive equigranular granite-granodiorite batholith with gradational and subtle mineralogical variations, locally intruded by dike and sill complexes.Mineral alteration is common near the SJCF footwall damage zone.

Regional granite-granodiorite batholith
The volumetrically dominant igneous lithologies in the batholith are biotite-hornblende granodiorite and biotite monzogranite, with fine-to coarse-grained equigranular textures.At the outcrop scale, these rocks host mafic to intermediate enclaves (Figure 3a,c) of variable size and morphology, as well as relatively small intrusive bodies of similar composition.Schlieren, magmatic layering, magmatic foliation, and more mafic ellipsoids with margin-parallel foliation are common features.As described by Pérez-Venzor (2013), a general N-S trend is observed in the magmatic foliation (Figure 3a).Hypabyssal dikes of granitic composition cross-cutting the rock display sharp contacts (Figure 3f).In the thin section, the observed mineral assemblage in granodiorite-granite consists of subhedral plagioclase (~40%) with sparse polysynthetic twinning and oscillatory zonation, quartz (~25%) 3-5 mm in size, K-feldspar (~15%) with Carlsbad twinning 1-2 mm in size as well as zircon, apatite, magnetite, and titanite as accessory phases.White mica, chlorite, and epidote are found locally as secondary phases.More mafic compositions are locally observed as hornblende tonalite, typically with heterogranular texture varying from fine-to medium-grained, which display complex associations with microgranular enclaves of variable size of granodiorite to diorite composition (Figure 3d,g).These outcrops are interpreted as hybridization zones where magmas of different composition, irregularly distributed mafic minerals, and mm-to cm-sized clusters of ferromagnesian minerals produce rocks of variable colour index.Microgranular enclaves are often found throughout the regional granodioritegranite outcrops (Figures. 3a,c).A modal QAPF diagram of the samples described is included to illustrate the lithologies described in this work (Figure 4e).

Enclave swarms and mingling structures
Microcrystalline dioritic enclaves are observed at several places within the granodiorite-granite unit of Sierra La Laguna, with a dominant N-S to NE-SW trend, sub-parallel to the dominant magmatic foliation (Figure 3a).At the Buenos Aires creek, near the contact with the host metamorphic units, up to 10 metre thick microdioritic dikes, which display planar to undulating contacts with the host rock, were disaggregated to form mafic enclave swarms (Figure 3b,c).These dioritic to micro-dioritic enclave swarms occur as tabular bodies dipping to the W and are also present as small, scattered rafts and cognate blocks 5 cm to 1 m in diameter (Figure 3b,c).Enclaves present mainly aphanitic to porphyritic texture, dioritic composition and are immersed in a hornblendebiotite microgranite rock.Mafic enclaves occur as amoeboid, ovoid, and ellipsoid forms with biotite rims, implying that both enclaves and host rock were semi-plastic during the intrusion (Figure 3c).Under the microscope, these enclaves show inequigranular texture, and contain euhedral to subhedral plagioclase with oscillatory zoning, biotite, orthopyroxene, hornblende, and quartz, as well as zircon, apatite, magnetite, and epidote as typical accessory phases (Figure 4).

Leucogranite dikes
Pegmatitic leucogranite dikes with peraluminous character were observed at Buenos Aires creek, west of Los Barriles (Figure 5).These dikes have variable widths from 10 cm to 5 m, present no preferential trending and crosscut the whole granodiorite-tonalite unit.Different facies were recognized: (a) fine-grained leucogranite (1 mm to over 1 cm) with almandine-rich garnet bands (Figure 3h), (b) coarse-grained leucogranite dominated by quartz + K feldspar + biotite + muscovite + garnet, (c) medium coarse-grained leucogranite with graphic texture, garnet + hornblende + biotite, and (d) skeletal hornblende + biotite arranged as an arborescent texture near the dike contact.Similar leucogranite dikes have been observed at several places in Sierra La Laguna (see descriptions in Pérez-Venzor 2013).

Porphyric dacitic dikes
A system of metre-thick dikes with sharp contacts and dacitic composition was observed in the San Dionisio and Buenos Aires creeks.The dikes exhibit porphyritic texture with plagioclase, biotite, hornblende, and quartz phenocrysts in a microcrystalline matrix (Figure 4c).Chlorite and epidote occur as products of hydrothermal alteration.Plagioclase is the dominant phase with euhedral to subhedral morphologies and compositional zoning.Biotite is subhedral and often replaced by chlorite.Hornblende, apatite, and zircon occur as accessory phases (Figure 4c).Porphyritic dikes are ubiquitous within the Los Cabos block and have been described for Sierra La Trinidad, Sierra La Gata     1 and Supp.Tables S2 and S3.

Basaltic dikes
Dikes of basaltic composition crosscut all the units described above.They are generally aphanitic and display an N-S or E-W trending, near-vertical dip, and 1 to 7 m width.

U-Pb and Ar-Ar geochronology
To better constrain the timing of emplacement and the cooling pattern of the LCBC, we collected 13 samples of intrusive rocks at Sierra La Laguna, along transects perpendicular to SJCF trace (Figure 5), one intrusive and three volcanic samples at Sierra La Trinidad, and three sedimentary samples from the San José del Cabo basin.
For these samples, we obtained Concordia U-Pb zircon ages for 16 magmatic samples, 17 Ar-Ar ages for amphibole, biotite, feldspar, and plagioclase for intrusive samples from Sierra La Laguna, and U-Pb age distributions of detrital zircons for the sedimentary samples.Obtained U-Pb crystallization and Ar-Ar cooling ages are summarized in Table 1.U-Pb Concordia diagrams (Tera-Wasserburg) for igneous samples and probability density diagrams for sedimentary rocks are presented in Figure 6a,b and 6c, respectively.Ar-Ar age spectra are plotted in Figure 7a,b.The complete data set is reported in Supp.Tables S2 and S3 Previous datings of the LCBC at Sierra La Laguna were obtained mainly by the Rb-Sr, Sm-Nd, and Ar-Ar methods on micas (biotite and muscovite) and hornblende, yielding a wide range of cooling ages between 104 and 58 Ma (Pérez-Venzor, 2013) (Figure 2; Supp.Table S1).Zircon U-Pb ages of magmatic rocks, which can be regarded as the age of onset of magma crystallization or the age of intrusion, are scarce for the LCBC.Eight U-Pb ages (82-74 Ma) have been previously reported for Sierra La Trinidad (Duque-Trujillo et al. 2015;Díaz-López, 2019), and only one U-Pb age (85.7 Ma) is known for Sierra La Laguna (Camarena-Vázquez et al. 2022) (Figure 2, Supp.Table S1).
Our U-Pb zircon dating provides a more extensive dataset of U-Pb zircon crystallization ages for the LCBC (19 U-Pb ages: 13 for intrusive rocks, 3 for volcanic rocks, 3 sedimentary).According to our results, the construction of the batholith occurred in a relatively long period, between ~89.9 and ~73.7 Ma, with the westernmost sample (LC-47) yielding an older age of 100.6 Ma.For the Sierra La Trinidad massif, a granodiorite sample (LC 18) yielded a zircon crystallization age of 79.5 Ma, which is within the range of U-Pb zircon ages of ~82 to ~74 Ma reported by Díaz-López (2019) and Duque-Trujillo et al. (2015) for this area.This age range is similar to that of the easternmost part of Sierra La Laguna.
We have also dated three volcanic rocks exposed in the eastern part of Sierra La Trinidad that were inferred to be Middle Miocene in age in Fletcher et al. (2003) and in the 1:250,000 scale geologic map of the Mexican Geological Survey (Maraver-Romero et al., 2002).Samples LC 15 and LC 42 belong to a sequence of ignimbrites cropping out NW of Cabo Pulmo, and sample LC 17 is a dacitic lava covering the batholithic rocks to the SE (Figure 5).The ignimbrites yielded a weighted mean U-Pb zircon age of 80.9 and 80.2 Ma, indistinguishable within error, whereas the dacitic lava returned a younger age of 72.7 Ma (Figure 6b).
Detrital zircons from the three lowermost sedimentary formations that fill the San Jose del Cabo basin were dated to help constrain the age of the initial activity of the San José del Cabo fault system; nevertheless, all the dated zircons yielded Late Cretaceous ages (Table 1; Figure 6c).From the older to the younger unit, ages of the fluvial sandstone of the La Calera Formation (LC 21) range between 194 and 50 Ma (peak at ~83 Ma), similar to detrital zircons ages presented by Fletcher et al. (2007), the marine shale and sandstone of the La Trinidad Formation (LC 19) between 93 and 64 Ma (peak at ~80 Ma), and those of the reworked volcaniclastics of the El Refugio Formation (LC 20) between 90 and 70 Ma (main peak at ~76 Ma).
Ar-Ar age data for hornblende, biotite, K-feldspar, and plagioclase were obtained from seven samples (Figure 7), six of them dated by U-Pb (LC 03, LC 04, LC 06, LC 13, LC 14, LC 32).These ages are used to constrain the cooling history of each sample through different closure temperatures (see below).Several samples display perturbed age spectra, being most obvious for minerals with the lowest closure temperatures (plagioclase and K-feldspar).Some age spectra display younger ages for the low-temperature steps, which become older for higher temperature steps, and for most samples, the high-temperature steps define a plateau or a 'mini-plateau' (30-50 % 39 Ar).Such age gradients can originate from a reheating event or from slow cooling through the

Figure 7a.
Step heating 40 Ar-39 Ar age spectra for different minerals from seven intrusive rock samples of Los Cabos block.Errors of the preferred ages for each mineral is given at 95% confidence level, and single step age errors are 2σ.Spectra ages were calculated using the gas fractions identified with the horizontal arrow, for which the % of released 39 Ar and the number of fractions (n) is given.tp and Wm indicate plateau and weighted mean ages, respectively; MSWD is the mean square of weighted deviates.Methods and complete results are given in Supp.Table S2.mineral is given at 95% confidence level, and single step age errors are 2σ.Spectra ages were calculated using the gas fractions identified with the horizontal arrow, for which the % of released 39 Ar and the number of fractions (n) is given.tp and Wm indicate plateau and weighted mean ages, respectively; MSWD is the mean square of weighted deviates.Methods and complete results are given in Supp.Table S2.(continued).

Zircon chemistry and thermometry
Trace element data measured in zircon were filtered out to exclude analysis probably affected by inclusions within zircon or by hydrothermalism (La > 0.7 Sm N /La N >22; Ce/Ce*>10; Hoskin et al. 2003;Hoskin 2005;Zou et al. 2019).Rare earth element (REE) patterns of analysed zircons are characteristic of igneous zircon of magmatic arcs, with strong increases towards heavy REE (Figure 8a).The Ti-in zircon thermometer of Ferry and Watson (2007) was used to estimate the crystallization temperature of the intrusive bodies.For the calculations, zircon analyses with Ti > 50 ppm, probably affected by inclusions within zircon, were filtered out (Chapman et al. 2016).In the remaining samples, the concentration of Ti varies between 0.1 and 34 ppm.Also, melt TiO 2 and SiO 2 activities of 0.5 and 1, respectively, were assumed to compensate for the absence of rutile.Ti-in-zircon thermometry indicates a crystallization interval between ~600°C and 800°C with low variation in average temperatures between 707°C and 774°C Table 2 and Figure 8b).

Amphibole thermobarometry
Thermobarometric calculations were performed on samples representative of different intrusive facies to constrain magma emplacement depth and to obtain additional information on crystallization temperature.Amphiboles in all samples are classified as magnesiohornblende to pargasite (Figure 8c).The pressure of magma crystallization was iteratively calculated with the Al-in-amphibole barometer of Mutch et al. (2016).For temperature estimations, we used the empirical calibration based on the composition of amphibole proposed by Ridolfi and Renzulli (2012).Results obtained with both procedures are shown in Figure 8(b,f) and S Table 2; detailed results can be found in the Supp.Table S4.According to Mutch et al. (2016) calibration, we obtained an average pressure ranging between 277 and 476 MPa for Sierra La Laguna, corresponding to a range from ~10 to 18 km depth (estimated with an  upper-crustal density of 2.7 g/cm 3 ).For one sample from Sierra La Trinidad (sample LC 18), we obtained an average pressure of 289 MPa, corresponding to a depth of ~11 km.Temperatures calculated with the Ridolfi and Renzulli (2012) empirical thermometer ranged from 673°C to 740°C, with the sample from Sierra La Trinidad falling in between (Table 2 and Figure 8b).

Late Cretaceous assembly of the Los Cabos Batholithic complex
Our field observations coupled with geochronologic and geothermobarometric data indicate that the Los Cabos Batholithic Complex is part of a transcrustal magmatic system formed by several magmatic pulses during the Late Cretaceous.Magmatic structures and the occurrence of mingling zones point to a mushy mid-crustal reservoir where different magmas would have coexisted at depths between 10 and 20 km.Mingling structures also indicate a plastic state with contrasting physical conditions that allowed mixing and mingling of mafic and felsic magmas (Barbarin 2005).Compositional zoning and growth textures in plagioclase, as well as quartz and feldspar intergrowths (Figure 4), indicate thermal and convective changes in the magmatic reservoirs and disequilibriumequilibrium conditions in the magmatic system, which can be associated with reheating of the system by diking and intrusion of magma batches (Annen et al. 2015;Cashman et al. 2017).These instabilities suggest an intermediate to felsic mush zone multiply reactivated by mafic magma injections (Figure 9).The above data point to a period of voluminous Late Cretaceous magmatism that may have contributed to crustal growth.Although the LCBC shows clear evidence of assimilation of crustal material along its eastern border, previous isotopic studies (Schaaf et al. 2000; Pérez-Venzor 2013; Rochín-García 2015; Díaz-López 2019) indicate variable mantle contributions to the magmas.If so, arc building was accompanied by a significant transfer of igneous material from the mantle to the crust.A precise estimation of the magma flux during the assembly of the batholith requires an estimation of the location of the batholith roots.However, based on the area of exposure of intrusive rocks in Sierra La Laguna (~4,000 km 2 ) and a time span for the emplacement of ~27 Ma, the long-term magma flux for the LCBC can be estimated in 150 km 2 /m.y., a value within the range of similar plutonic Late Cretaceous arc complexes of the US Cordillera (Paterson et al. 2011).However, it could have been an order of magnitude larger during the main pulses of activity.

Geochronological and geobarometric constraints
Our new and earlier published crystallization ages support the hypothesis that the LCBC formed in several magmatic pulses throughout the Late Cretaceous.Concordia U-Pb crystallization ages of 11 samples from the Sierra La Laguna range between 89.9 and 73.7 Ma, with only one sample, collected near the western limit of the Sierra La Laguna yielding an older age of 100.6 Ma.The spatial distribution of ages shows an eastward younging of magmatism (Figure 5), with ages of samples from the easternmost part of Sierra La Laguna comprised between ~79.2 and ~73.2 Ma (LC 13, LC 56, LC 50, LC 14; Table 1), almost identical to the age range of ~82-74 Ma of Sierra La Trinidad (our sample LC 18 and samples in Díaz-López 2019).In an early stage, granitoid magmas began to be emplaced into the pre-Cretaceous rocks (Torres-Vargas et al. 1999), leading to partial melting of the host rocks, as can be inferred from the transitional contact between these lithologies.This was followed by the emplacement of the main granite-granodiorite-tonalite bodies, with leucogranite and dacitic dikes crosscutting the main batholith in the last stage (~79.1-73.7 Ma).
Some of the dated minerals yielded spurious ages and were not further considered in the estimation of cooling rates.The hornblende of sample LC-13 yielded an apparent age (88.4 Ma) that is significantly older than the U-Pb zircon age (79.2 Ma).Possible explanations for the older hornblende age are the presence of partially reset hornblende inherited from the older wall-rock gneiss that survived as cores in magmatic hornblende or were mixed with new crystallized hornblende in the analysed mineral concentrate.An alternative is the presence of cryptic excess argon in hornblende, which is not evident in biotite with lower closure temperatures.For the sample LC-23, lacking U-Pb zircon age, the large age difference between hornblende (89.7 Ma) and plagioclase (72.6 Ma) can indicate similar processes.
We relate the discrepancy of sample LC 03, and the perturbed Ar-Ar age spectra of plagioclase from samples LC-13 and LC-14 to partial resetting of K-feldspar and plagioclase during early faulting that produced spurious ages for this sample collected close to the SJCF trace (Figure 7 and figure in Supp.Table S3).Alternatively, they may result from secondary growth of adularia in feldspars associated with shear zone fluid circulation, as suggested by Bot et al. (2016), who obtained similar ages near the SJCF.The depth of emplacement of the sample LC 18 for the plutonic rocks at Sierra La Trinidad is ~11 km, in agreement with the range from 8 to 11 km obtained by Díaz-López (2019).This indicates that the LCBC at Sierra La Trinidad was emplaced at a shallower depth than the plutonic complex at Sierra La Laguna, estimated in this work at ~10-18 km (Figure 9).
In Figure 10b, we plot the mean Ti-in-zircon crystallization temperature together with commonly used Ar-Ar closure temperatures for hornblende, biotite, plagioclase, and feldspar (e.g.Reiners et al. 2005) to estimate the cooling rate for samples for which both U-Pb and Ar-Ar ages were obtained.Four samples yielded ages consistent with cooling at rates between 42 and 117°C/ m.y.(LC 03,LC 04,LC 13,LC 32,and LC 14).With one exception (LC 03) younger samples have slower cooling rates than older samples, which may indicate a progressive heating of the crust through time.Older intrusions would have been emplaced into colder country rocks than younger intrusions leading to faster cooling rates.
The depth of emplacement of the sample LC 18 for the plutonic rocks at Sierra La Trinidad is ~11 km, in agreement with the range from 8 to 11 km obtained by Díaz-López (2019).This indicates that the LCBC at Sierra La Trinidad was emplaced at a shallower depth than the plutonic complex at Sierra La Laguna, estimated in this work at ~10-18 km (Figure 9).
Although only few data are available, a trend of increasing temperature at higher pressure is shown in Figure 8e, which would support a stable geothermal gradient.We dismiss the possible influence of low water fugacity on this trend, given that these P-T data were estimated from hornblende, a mineral that is stabilized at high magma water contents (~5-6 %; Rodríguez et al. 2007).Samples from the Peninsular Ranges Batholith from the submerged blocks in the southern Gulf of California (Duque-Trujillo et al. 2015) show comparable slow cooling rates of 52-130°C/m.y.during the Late Cretaceous.
The relative abundance of Eu in zircon is considered a calibrated crustal thickness proxy.During magmatic differentiation, the Eu content in crystallizing zircons is sensitive to pressure and is recorded with a positive correlation between the size of the Eu anomaly [Eu/ Eu*=Eu/(Sm × Gd)1/2] in zircon and the crystallization pressure (Tang et al. 2020(Tang et al. , 2021)).This method was used by Tang et al. (2020Tang et al. ( , 2021) ) to estimate the changes in the thickness of a magmatically active crust formed at convergent plate margins.In Figure 10c, we plot EuN/EuN* vs. zircon crystallization age, and we interpret the observed trend of increasing EuN/EuN* between ca. 90 and 70 Ma as the result of gradual thickening of the crust during batholith assemblage.
In summary, our data reveal significant differences between Sierra La Laguna and Sierra La Trinidad that have important implications for the reconstruction of the geometry of the SJCF.At Sierra La Laguna, the wider compositional range from tonalite to granite, the abundance of dikes and mafic enclaves, and the higher crystallization pressure are consistent with a deeper part of the magmatic system (middle crust) (Figure 10 a).By contrast, Sierra La Trinidad has a narrower compositional range (granodiorite to granite), lower crystallization pressure, and Late Cretaceous volcanic sequences are found close to the intrusive rocks; these three aspects indicate that Sierra La Trinidad represents a shallower part (upper crust) of a closest and similar magmatic system.

Correlation of the LCBC with the Jalisco Block
Although reconstructions of the position of the Los Cabos block prior to the opening of the Gulf of California may vary in latitude, most reconstructions place it to the NW of the Jalisco block (Figure 1; Stock and Lee 1994;Fletcher et al. 2007;Ferrari et al. 2013;Díaz-López 2019;Schaaf et al. 2020;Gutiérrez-Aguilar et al. 2021), which hosts the Puerto Vallarta batholith (Figure 1a).Over two decades ago, Schaaf et al. (2000) proposed a correlation between the LCBC and the Puerto Vallarta Batholith based on geochemical and geochronological similarities, although, at that time, only cooling ages were available for the two igneous complexes.Also, detrital zircon ages of Fletcher et al. (2007) suggest strong similarities in the crystallization ages from the Los Cabos and Jalisco areas.Our new crystallization ages allow a more robust comparison.Crystallization U-Pb zircon ages for the Puerto Vallarta batholith range from ~92 to 65 Ma (Fletcher et al. 2007;Valencia et al. 2013;Schaaf et al. 2020), which indicates that magmatism occurred in a similar period to that of the eastern LCBC dated in this work (~90 to 74 Ma).In both cases, the emplacement of the intrusive rocks displays an eastward progression with time as proposed by Fletcher et al. (2007).Younger ages for the western end of the Puerto Vallarta batholith with respect to the easternmost LCBC can be explained by subduction erosion of the western Jalisco Block, resulting in plutonic rocks now being directly exposed at the trench (Calmus et al. 1999) 1a).Our new ages of ~80 Ma for two ash-flow tuffs and a dacitic dike of ~73 Ma exposed in the easternmost part of the LCBC in the Cabo Pulmo region further correlate with the igneous units of the Jalisco Block.In fact, ash-flow tuffs are widespread in the eastern part of the Jalisco block and in the Islas Marías with ages of ~83 to 74 Ma (Wallace and Carmichael 1989;Righter et al. 1995;Rosas-Elguera et al. 1997;Pompa-Mera et al. 2013).In summary, our data confirm that the Puerto Vallarta batholith can be considered the SE continuation of the LCBC and the southernmost segment of the Peninsular Ranges Batholith.

Localization and geometry of the Los Cabos fault system
Although some secondary faults are located to the west of Sierra La Laguna, the SJCF is the dominant normal fault of southern Baja California and one of largest known normal faults in the Gulf Extensional Province (Fletcher and Munguía 2000).The SJCF and other faults to the west have a general N-S orientation, oblique with respect to the Main Gulf Escarpment to the north (Figure 1a), which is formed mostly by NNW -SSE trending fault systems parallel to the main orientation of the Gulf.This oblique orientation could be the result of the transtensional nature of shearing, which is associated with an EW-trending least compressive principal stress.However, the onset of a transtensional regime can be only dated back to the Late Miocene (Fletcher et al. 2007).Another complementary possibility is that the SJCF followed a pre-existing zone of weakness.In its central part, the orientation of the SJCF is roughly parallel to the intrusive contact of Sierra La Laguna Batholith with the host gneisses.In these areas, igneous structures, such as magmatic foliation and orientation of microdiorite dikes with evidence of mingling, are subparallel to the SJCF.The mechanical contrast between granitic and metamorphic rocks is an essential zone of weakness for the location of a structural discontinuity.Likewise, the difference in mechanical strength between the granitic rocks and the mafic dykes functions as a plane of weakness that allows the nucleation of fractures and, later, the development of the SJCF.The determining factors, such as mechanical contrast between the granitoids, metamorphic rocks, magmatic foliation, and presence of mingling dikes, together with the extensional state, result in the location of fractures that evolve into minor faults that grow from these lithological discontinuities (as also suggested by Pérez-Venzor 2013).However, in its northern and southern parts, the SJCF cut across the batholith without an obvious relation with the igneous structures.In summary, we consider that at least in part the SJCF owes its orientation to the existence of lithological discontinuities, although this is not always the rule throughout its trace.
Our study also contributes to confirm the geometry of the SJCF and the timing of its early activity.Fletcher et al. (2000) proposed that the SJCF has accommodated ~5.2-6.5 km of exhumation since ~12-10 Ma based on a moderately dip angle.Bot et al. (2016) proposed that the SJCF is a low-angle detachment fault already active at ~18 Ma with at least ~13 km of exhumation.A detailed study of the fault geometry and thermochronology will be presented in a forthcoming paper, but our field observations of the fault trace are consistent with the hypothesis of Fletcher et al. (2003) and Bot et al. (2016) that the SJCF has a low angle geometry.In fact, our geologic, geochronologic, and thermobarometric data confirm that Sierra La Trinidad is the hanging wall of the SJCF, and the San José del Cabo basin is a half-graben (Figure 11) as previously proposed by Fletcher et al. (2000Fletcher et al. ( , 2003Fletcher et al. ( , 2007)).Crystallization ages reported in this work, together with those of Díaz-López (2019), indicate that the ages of Sierra La Trinidad intrusive rocks (~82-74 Ma) are similar to those of the eastern part of Sierra La Laguna.Sierra La Trinidad has a narrow compositional range (i.e.mostly silicic) without mafic intrusions, which is typical of the upper part of a magmatic system.The pressure estimations obtained by Díaz-López (2019) and in this work also indicate that Sierra La Trinidad has a narrower range of crystallization depth that corresponds to the upper range of Sierra La Laguna.All the above suggest that the intrusive rocks of Sierra La Trinidad represent a shallower part of a magmatic system compared to Sierra La Laguna.The occurrence of Late Cretaceous volcanic rocks on the eastern margin of Sierra La Trinidad also indicates that this area represents the upper part of the LCBC magmatic system and the hanging wall block of the SJCF.In addition, the lower three formations filling the San José del Cabo basin (La Calera, La Trinidad, and El Refugio) are exposed only in its eastern part with tilting decreasing upsection, clearly supporting a half-graben geometry surrounded by the LCBC (Fletcher et al. 2003;Schwennicke et al. 2017) (Figure 11).Finally, the peak ages of detrital zircons from these formations are also progressively younger upsection.Given the eastward progression of Sierra La Laguna Batholith ages, this implies that the SJCF was a lowangle fault that, as the Sierra La Trinidad hanging wall moved east, exposed progressively younger rocks of Sierra La Laguna.
The horizontal distance between the present trace of the SJCF and Sierra La Trinidad is ~20 km.Assuming an average dip of 30° for the SJCF, the minimum displacement along the fault is 23 km, but it could be several kilometres more if the hanging wall had to move above the highest elevation of Sierra La Laguna.If the average slip is in the order of 1 mm/yr, the onset of the activity along the SJCF should be Early Miocene or Late Oligocene.

Conclusions
The LCBC represents a magmatic system with dikes, sills, and mingling structures covering an area of over 2700 km 2 corresponding to the southernmost part of the Peninsular Ranges Batholith in southern Baja California.The integration of U-Pb zircon ages obtained in this study with regional data from the literature shows that the locus of magma emplacement was moving eastward with time, during the construction of the batholithic complex.The Eu N / Eu N * value of dated zircons indicates a considerable crustal thickening during batholith assemblage.Geologic, geochronologic, and geobarometric data indicate that the western LCBC exposed at Sierra La Laguna represents a deeper part of the magmatic system, located in the middle crust.Sierra La Trinidad to the east represents the shallow to superficial part of the system.This is consistent with a low angle geometry for the SJCF, in which Sierra La Laguna and Sierra La Trinidad represent the footwall and hangingwall of the San José del Cabo Fault, respectively.

Figure 5 .
Figure 5. New U-Pb and Ar-Ar ages reported in this work and their distribution on the geologic map of the eastern Los Cabos block.SJCF: San José del Cabo Fault trace; BAC: Buenos Aires Creek; SDC: San Dionisio Creek.U-Pb Concordia diagrams in Figure 6.Ar-Ar age spectra in Figure 7. Geological cross-section A-A' in Figure 11.Age details in Table1and Supp.TablesS2 and S3.
.a and b: U-Pb Tera-Wasserburg Concordia and weighted mean diagrams for zircon analysis of selected igneous samples from Sierra La Laguna (a) and Sierra La Trinidad (b) in the eastern Los Cabos block.Sample age uncertainties are reported as 95% confidence intervals.Gray ellipses and bars represent spot ages excluded from the age calculation (see methods section).(continued).c)Kernel density estimates for detrital zircon U-Pb ages of samples from sedimentary formations of the San José del Cabo basin.Histograms are also shown for comparison.The bars at the base of the plots represent the age of each analysed zircon grain.Details of the U-Pb zircon experiments and cathodoluminescence images of selected zircon samples are given in the supplementary files.

c
Figure 6c.c) Kernel density estimates for detrital zircon U-Pb ages of samples from sedimentary formations of the San José del Cabo basin.Histograms are also shown for comparison.The bars at the base of the plots represent the age of each analysed zircon grain.Details of the U-Pb zircon experiments and cathodoluminescence images of selected zircon samples are given in the supplementary files.

Figure 8 .
Figure 8. a) Diagram of chondrite-normalized zircon REE contents; b) Ti-In zircon temperature based on the thermochronometer of Ferry and Watson (2007) vs. zircon crystallization age (Ma).c) Amphibole classification according to Hawthorne et al. (2012).d) Pressures-depth (km) obtained with the Al-in hornblende geobarometer of Mutch et al. (2016) vs. crystallization age.e) Temperature obtained with the Al thermometer vs. pressure obtained with Al in hornblende barometer.f) Temperatures obtained with the empirical amphibole geothermometer of Ridolfi and Renzulli (2012) vs. U-Pb crystallization age.

Figure 9 .
Figure 9. Conceptual tectonomagmatic model for the Late Cretaceous assembly of the Los Cabos Batholith Complex.
The work was supported by the PAPPIT-UNAM grant[IN108819].

Figure 11 .
Figure 11.Geological cross-section (A-A') along San José del Cabo basin, based on Fletcher et al. (2003); and our interpretation.The estimated depths and U-Pb zircon crystallization ages for Sierra La Laguna and Sierra La Trinidad massifs are also indicated.SJCF: San José del Cabo Fault.Redrawn fromBot et al. (2016).

Table S1
Bot et al. (2016)et al. (2022)2)obtained a U-Pb zircon crystallization ages of 85.4 Ma for a tonalite at Arroyo el Mezquite, in the eastern part of Sierra La Laguna.Bot et al. (2016)described the Los Llanitos unit at Boca La Sierra (Figure2) as formed by diorite, granodiorite-granite, and granitic dikes.These authors obtained K-Ar ages Ma) for the western and the eastern border of Sierra La Laguna, respectively.The LCBC at Sierra La Trinidad was studied by Díaz-López (

del Cabo Basin -U-Pb ages of detrital zircons Sample Rock type Formation Latitude Longitude Altitude Population size Max Min Peak age
Summary of the geochronological results.

age (Ma) 238 U / 206 Pb LC-32 (Alkali-Feldspar granite)
80 Figure 6a.a and b: U-Pb Tera-Wasserburg Concordia and weighted mean diagrams for zircon analysis of selected igneous samples from Sierra La Laguna (a) and Sierra La Trinidad (b) in the eastern Los Cabos block.Sample age uncertainties are reported as 95% confidence intervals.Gray ellipses and bars represent spot ages excluded from the age calculation (see methods section).

Table 2 .
Geothermobarometric data for selected samples from Los Cabos Batholithic Complex.
Duque-Trujillo et al. 2015;opera et al. 2019)erved forearc section.On the other hand, the eastern end of the LCBC has been truncated by the Gulf of California rift, but younger ages, similar to those of the eastern Puerto Vallarta Batholith, are reported for the southern part of the Sinaloa batholith (San Ignacio Batholith ~64-67 Ma;Henry et al. 2003;Montoya-Lopera et al. 2019)and its continuation in northern Nayarit (Mineral de Cucharas pluton ~65 Ma,Duque-Trujillo et al. 2015; Figure