Crystallization Conditions of Basaltic Lavas Based on Clinopyroxene and Olivine Phenocryst Petrology: A Case Study from the Neogene Lavarab Alkaline Basaltic Lavas (LABL), Eastern Iran

Abstract

This paper focuses on delineating and charactering of the magma crystallization conditions of the post-collision Lavarab Alkaline Basaltic Lavas in East Iran. The lavas consist mainly of alkali basalt and basanite, with subordinate trachybasalt. Olivine mostly shows forsterite, chrysolite and hyalo-siderite compositions. Clinopyroxenes are diopside and augite, belonging to peralkaline to subalkaline magmatic series within post-collisional tectonic settings. Estimates of temperature and pressure obtained from single clinopyroxene thermobarometers suggest that crystallization temperatures vary between approximately 1110 and 1260 °C, with pressures ranging from about 0.05 to 1.35 GPa, which correspond to depths of roughly 2 to 51 km at high oxygen fugacity in both the lower and upper continental crust. Olivine-liquid thermometry yields temperatures of ~1385 to ~1393 °C for basanites and ~1275 to ~1339 °C for alkali basalts, assuming a constant pressure of 1.4 GPa. The chemical compositions of phenocrysts in the studied basaltic lavas provide evidence of magma recharge, occurring through multiple pulses of new magma injected into the existing reservoir prior to eruptions. Petrographic evidence, including absorption features, rounded crystal morphologies, patchy zones in olivine, and sieve textures in clinopyroxene, support this interpretation. Additionally, microprobe analyses reveal oscillatory variations in crystal composition from core to rim, confirming the hypothesis of dynamic magma replenishment.

1. Introduction

Understanding the processes that govern magma generation, storage, and evolution in post-collisional tectonic settings remains poorly constrained in igneous petrology [1,2,3]. Relationships among mantle melting, trans-crustal magma plumbing systems, and geochemical diversity of alkaline basalts continue to be debated [4]. This gap is especially relevant in collisional orogens like the Alpine–Himalayan belt, where lithospheric thinning, crustal assimilation, and fault-guided magma ascent may collectively shape magmatic products. Addressing this topic not only enriches our understanding of mantle dynamics but also provides a robust framework for evaluating the continental lithosphere’s response to orogeny and subsequent extension.

The Neogene Lavarab Alkaline Basaltic Lavas (LABL), erupted within the Sistan Suture Zone following the closure of the Neo-Tethyan Ocean, record critical information about mantle dynamics, crustal interactions, and post-collisional volcanism following the closure of the Neo-Tethyan Ocean. The LABL lavas are located in the southern part of Sefidabeh Basin in the middle part of the Sistan Suture Zone, east Iran, and have formerly been interpreted as Oligocene-Miocene basalts [5,6,7]. Previous studies of this zone were focused on the characteristics, origin and tectonic setting of igneous rocks [8,9]. However, the identification of magmatic processes within the reservoir has not been addressed through the investigation of phenocrysts that crystallized in the magma chamber prior to eruption. Phenocryst mineral chemistry is a powerful tool to address this gap.

The chemistry of rock-forming minerals not only reflects the chemistry of the crystallizing melt but also can be used to investigate magma crystallization conditions. Minerals resistant against alteration and weathering, such as clinopyroxene, play a more significant role compared to other minerals. Olivine and pyroxene, normally crystallized during the early stages of magmatic systems, directly reflects the composition of the parental magma, the physicochemical conditions prevailing during crystallization in the magma chamber(s) and the tectonic environment (e.g., [10,11]). The composition of clinopyroxene is used to estimate the crystallization condition (temperature, pressure, and oxygen fugacity) during the crystallization of a magma through the various thermobarometers [11,12,13,14,15,16,17,18,19,20,21,22].

On the other hand, phenocrysts in basalts indicate that their parent melts underwent a period of crystallization in one or more hypabyssal magma chambers during ascent from the mantle to the surface [23]. Therefore, by studying the minerals that formed in the magma chamber before eruption, it is possible to better understand these magma chamber processes.

The aim purpose of this paper is to investigate the mineral chemistry and geochemical signatures of LABL. This offers insights into not only the magmatic processes active in the Sistan Suture Zone but also the broader questions concerning mantle heterogeneity and post-collisional alkaline magmatism in the Alpine-Himalayan orogenic belt. Thus, this work advances the broader understanding of how continental lithosphere responds to collisional orogeny and subsequent extension by linking mineral-scale data to large-scale tectonic processes.

2. Geological Setting and Study Area

Iran is situated in the central part of the Alpine-Himalaya Orogen, which formed as a result of the convergence between the Eurasian and Gondwanan supercontinents. This extensive orogenic belt extends from western Europe and continues through Turkey, Iran, Afghanistan, to Tibet, and Indonesia [24,25].

The Sistan Suture Zone in eastern Iran provides a notable example of the tectonic evolution of the Alpine-Himalaya Orogen, where subduction did not result in the formation of a widespread magmatic arc. Instead, the rapid destruction of a narrow strip of oceanic lithosphere led to development of the Ratuk and Neh ophiolite complexes, which formed as accretionary prisms, and the Sefidabeh basin, a forearc basin. These geological features reflect the dynamic interplay between subduction, accretion, and basin formation within the evolving Alpine-Himalayan orogenic system.

Three phases of deformation affected the marine sediments of the Neh Complex and Sefidabeh basin [9]. The first two occurred between the Early Eocene and the Early Miocene, associated with regional metamorphism (within the greenschist facies and the biotite schist zone). The third phase of deformation formed large strike-slip faults and folding in the Pliocene-Quaternary and accompanied recent volcanism [26].

The age of the deep to shallow marine deposits in the Sefidabeh basin is Maastrichtian-Paleogene and comprises lavas, pyroclastics, and volcanoclastic derivatives with detrital and carbonate rocks [8], deposited in a forearc basin. Its eastern border with the Afghan block is formed by the Harirud fault and its lower contact with the basement is the Talkhab thrust [27]. The effusives in the Sefidabeh basin are aligned in three tracts in a nearly north-south trend (ca. 20 km long and max. 2 km broad). After the eruption of the LABL the depositional environment gradually deepened, and beds of marl were deposited alternating with clastic rocks [6]. Lava morphology and the interbedded detrital rocks indicate that the lavas erupted in a continental alluvial-fluvial or lacustrine environment.

The tectonic setting of the LABL is post-collisional. Their parent melts stem from a low degree partial melting of upwelling asthenosphere which was triggered by delamination of a thickened lithosphere root [28]. Asthenospheric upwelling and lithospheric thinning resulted in a tensional setting with reactivation of the Zahedan and Harirud faults that may have acted as conduits for ascending magmas [28].

3. Field Relations and Petrography

The LABL rocks form elongated outcrops parallel to and near the Zahedan dextral fault in the west (Figure 1). In the studied area, an asymmetric syncline with a north-south axis contains lavas exposed on both its western and eastern limbs, where they are concordant with interbedded Oligocene-Miocene sediments (Figure 2a). Two eruption stages can be recognized on the eastern limb. Following the first stage, a 20-m-thick sedimentary sequence was deposited which was followed by the second stage. The first stage produced thicker and more extensive volcanic rocks, which are also exposed on the western limb. Field observations reveal columnar structures and lava flows (Figure 2b,c). Eruption likely occurred in shallow water, at depths of up to 30 m, as indicated by the presence of Ostracoda (Cytherella sp. and Cytheridea sp.) in marl deposits stratigraphically underlying the lavas near Harmak (see Figure 1 and Figure 2d). The marls also contain benthic foraminifera such as Pararotalia sp., Elphidium sp., Osangularia sp., and Asterigerina sp., suggest that these marls belong to the Early Miocene.

Lavas forming an annular outcrop in the southeastern LABL probably erupted at the intersection of a NW-SE dextral fault with a NE-SW sinistral fault in a local extensional environment. These melanocratic to mesocratic lavas are porphyritic (relatively micro/phenocryst-rich) and vesicular (Figure 2e,f), and are classified into three main groups based on petrographic characteristics:

3.1. Trachybasalt

Trachybasalts are porphyritic or vesicular with a microlithic to aphanitic groundmass, containing micro- and phenocrysts (<3 mm in length, ca. 20 vol-%) of tabular plagioclase and ca. 10 vol-% euhedral to subhedral clinopyroxene (Figure 3a). Plagioclase commonly displays polysynthetic twinning, zonation and sieve textures. Few opaque minerals are observed which may initially have been amphibole. Vesicles are mostly filled by calcite. Accessory minerals are apatite and opaque minerals.

3.2. Alkali Basalt

The alkali basalts are porphyritic or vesicular with a microlithic or glassy groundmass, where vesicles are mostly filled with zeolite. These lavas contain 35%–45% micro- and phenocrysts (<1 cm in length) comprising ca. 20–30 vol-% clinopyroxene, 5–10 vol-% olivine, ca. 5 vol-% analcime, and minor Fe-Ti oxides (Figure 3b,c). Euhedral to subhedral pyroxene, subhedral to euhedral olivine, and euhedral analcime are present as both micro-phenocrysts and phenocrysts. Clinopyroxene crystals are typically euhedral and present in two generations: one exhibiting sieve texture and the second one lacking this. Clear oscillatory zoning and hourglass twinning can be observed in thin section (Figure 3b). Analcime occurs as rounded microcrysts and euhedral micro-phenocrysts scattered in the glassy groundmass, while olivine displays embayed crystal morphologies and skeletal growth (Figure 3c). Apatite inclusions are abundant in pyroxene and also form fine needles in the groundmass. Opaque minerals are present in fine-grained, cubic, and occasionally dendritic habit, occurring both within the groundmass and as inclusions in phenocrysts, particularly olivine.

3.3. Basanite

Basanites are porphyritic or vesicular with a fine-grain to microlithic, and occasionally glassy groundmass. The lavas contain 35%–40% phenocrysts, comprising 20–30 vol-% clinopyroxene, 10–15 vol-% olivine, ca. 5 vol-% analcime, with minor amounts of Fe-Ti oxides. In some samples, olivine crystals were altered to chlorite and iddingsite (Figure 3d). Clinopyroxene micro-phenocrysts and phenocrysts are mostly euhedral. An ultramafic sample (L.57) contains 15% phenocrysts (<1.5 mm in length), consisting of ca. 14 vol-% olivine, and minor clinopyroxene.

The groundmasses of both alkali basalts and basanites contain micro-phenocrysts (<10 to 100 µm in size) of the same phases and in the same relative abundances as found in the phenocryst population. Plagioclase microliths are commonly euhedral to subhedral, whereas the other phases show anhedral morphologies. Additionally, sanidine occurs within the fine-grained (~0.25 mm) groundmass.

4. Analytical Technique

4.1. Electron Probe Micro-Analyzer (EPMA)

One trachybasalt (L.10), four alkali basalt (L.38, L.51, L.61 and L.81), and two basanite (L.57 and L.65) samples were prepared for electron microprobe analyses. Backscattered electron (BSE) imaging of clinopyroxene and olivine was used to identify zonation patterns of the different crystal populations, and to identify mineral domains suitable for chemical microanalysis. Quantitative spot analyses were performed using a CAMECA SX 100 electron microprobe at the Micro-Area Analysis Laboratory, Polish Geological Institute (Warsaw, Poland), under operating conditions of 15 kV acceleration voltage, 20 nA current, a spot size of 5 μm and 10 s counting time at peak position.

4.2. Test for Equilibrium in Mineral-Liquid Geothermobarometry

The software MagMin_PT [30] was used to calculate K_D(Fe-Mg)^ol–liq using Equations (1) and (2) (originally suggested for olivine by Roeder and Emslie [31]) to test for equilibrium between olivine-liquid and between clinopyroxene-liquid.

MgO^min + FeO^liq = MgO^liq + FeO^min

(1)

K_D(Fe-Mg)^min-liq = XFe^minXMg^liq/XMg^minXFe^liq

(2)

4.3. Programs to Calculate Temperature and Pressure

MagMin_PT was also used to determine temperature based on glass or whole-rock composition, as well as for the olivine thermometry. For clinopyroxene the thermobarometric calculations following Soesoo [19] and were performed using an MS Excel spreadsheet, while those based on Putirka et al. [32,33] and Putirka [21] were carried out using WinPyrox [34] and MagMin_PT [30].

5. Results

5.1. Whole-Rock Composition

The LABL whole-rock geochemical characteristics indicate a dominant alkaline sodic nature ([Na₂O + K₂O] = 4.13–8.11 wt.% and K₂O/Na₂O = 0.13–1.62) (see Supplementary data). The LABL show low to moderate SiO₂ contents (43.76–53.91 wt.%), high alkali contents (4–8 wt.%), and high CaO concentrations (7.39–10.83 wt.%). MgO contents range from 3.01 to 13.38 wt.%, with Mg# (Mg# = 100 Mg²⁺/[Mg²⁺ + Fe²⁺]) ranging from 53.60 to 83.24 (mean = 74.51). Al₂O₃ contents vary from 12.97 to 17.88 wt.%. TiO₂ contents range from 0.47 1.79 wt.% (mean = 0.8 wt.%), classifying these rocks as low-Ti type [35,36]. These geochemical attributes suggest a close affinity with intraplate alkaline basalts and continental rift alkaline basalts [37]. The compatible trace element Ni shows a clear correlation with MgO contents, as well as the trend between Cr and MgO. The concentrations of Cr and Ni range from 59.6 to 690.3 ppm and 31.4 to 537.3 ppm, respectively.

The lava compositions are predominantly alkali basalt, basanite, and minor trachy-andesite after Winchester and Floyd [38] (Figure 4). For a more detailed account see [28].

5.2. Mineral Chemistry

In seven LABL samples, 950 points on 272 olivine and clinopyroxene crystals were analyzed ranging in size from mega-phenocrysts (0.5–1 cm), phenocrysts (0.5–5 mm), micro-phenocrysts (0.1–0.5 mm) to microcrysts (<0.1 mm; see Supplementary data).

5.2.1. Olivine

Olivine occurs in both basanites and alkali basalts, with forsterite contents ranging from Fo₅₅ to Fo₉₄, classified as forsterite, chrysolite, and hyalosiderite (Figure 5). Its chemical composition includes CaO (0.02–0.5 wt.%), NiO (0.03–0.53 wt.%), and MnO (0.05–0.84 wt.%). Based on petrographic observations, backscattered electron (BSE) imaging, and electron probe microanalysis (EPMA) data, olivine crystals are identified as both magmatic phenocrysts and mantle-derived xenocrysts.

Olivine xenocrysts commonly display broad, chemically homogeneous cores (Fo_89–94), surrounded by thin rims characterized by sharp compositional discontinuities and lower forsterite contents (Fo_71–89). Their chemical composition includes CaO (0.02–0.10 wt.%), NiO (0.30–0.54 wt.%), and MnO (0.05–0.20 wt.%). These crystals frequently exhibit resorptive textures, suggesting partial dissolution. The abrupt decrease in forsterite content at the crystal rims is indicative of interaction with the host melt, either during magma ascent or residence within the magma chamber [39].

Magmatic olivine phenocrysts have forsterite contents ranging from Fo₇₀ to Fo₈₉, with CaO (0.10–0.50 wt.%), NiO (0.30–0.40 wt.%), and MnO (0.10–0.80 wt.%). These crystals are euhedral to subhedral in shape, exhibit generally normal zoning, and only rarely display resorption textures (e.g., embayed margins) indicative of interaction with the melt. They commonly occur in association with clinopyroxene (diopside–augite) and analcime.

In basanite samples, olivine ranges from Fo₇₀ Fa₂₉ Tp_0.8 to Fo₉₄ Fa₆ Tp_0.05 (based on 24 crystals, n = 96), and in alkali basalt from Fo₇₂ Fa₂₇ Tp_0.08 to Fo₉₂ Fa₇ Tp₁ (based on 55 crystals, n = 195). A representative olivine xenocryst from a basanite sample (L.57) shows normal zoning, characterized by high Mg contents (Fo_86–91), transitioning to outer rim zones with lower Mg concentrations (Fo_71–76) (Figure 6a).

A representative olivine phenocryst from the alkali basalt (sample L.38) displays same texture and compositional characteristics, with core regions containing ~Fo₈₆ and rims remaining invariant at Fo_85–86.5. Rim-ward trends in minor element represents depletion in NiO, enrichment in MnO, and weak enrichment in CaO (Figure 6b).

Micro-phenocrysts of olivine (100–500 μm) in both rock groups are anhedral, unzoned and exhibit limited variation in Mg contents (ca. Fo_70.4–87), with the exception of one alkali basalt sample (L.61). In this sample, olivine forsterite ranges from Fo₅₅ to Fo_88.5. The presence of micro-phenocrysts with low forsterite contents (Fo_55–67), alongside phenocryst cores reaching up to Fo₈₅, suggests significant magmatic differentiation or possible crustal contamination affecting the magma composition [37].

5.2.2. Clinopyroxene

Clinopyroxenes in the LABL are Ca-Mg-Fe rich [40] and classified as diopside and augite [41] in all rock types (Figure 7a,b). BSE images and compositional traverses of clinopyroxene crystals of each rock type are presented in Figure 8, highlighting the patchy-zoning and oscillatory zonation in the samples.

In order to investigate magma evolution during ascent and cooling, one sample from each rock was selected: L.81 (SiO₂ = 49.3 wt.%) from alkali basalts, L.57 (SiO₂ = 42.3 wt.%) from basanite, and L.10 (SiO₂ = 52.7 wt.%) from trachybasalt. According to the crystal size, the clinopyroxene mineral chemistry data from these samples are categorized into three groups, being micro-phenocrysts (30–500 µm), phenocrysts (0.5–5 mm), and mega-phenocrysts (5–10 mm).

• Clinopyroxene micro-phenocrysts

This group is observed in trachybasalt (L.10) and basanite (L.57), whereas they are absent in alkali basalt (L.81). In basanite, these crystals are Ca-Mg-Fe-rich, containing Ca- and Mg-pyroxene with diopside compositions (Wo₄₇En₃₃Fs₈ to Wo₅₂En₄₅Fs₁₅; data from 9 crystals, n = 22). According to Morimoto et al. [41], these pyroxene micro-phenocrysts are classified as augite and diopside.

In trachybasalt, pyroxene micro-phenocrysts are characterized by Ca- and Mg-rich compositions, with a higher abundance of augite and lower abundance of diopside components (Wo₄₀En₄₀Fs₁₁ to Wo₄₆En₄₇Fs₁₅).

• Clinopyroxene phenocrysts

Clinopyroxene phenocrysts are found in alkali basalt (L.81) and trachybasalt (L.10) whereas they are absent in basanite (L.57). According to the binary plot (Q-J), all clinopyroxene phenocrysts are in the range of Ca-Mg-Fe rich clinopyroxene. The alkali basalt clinopyroxene is diopside and augite (Wo₄₂En₄₂Fs₅ to Wo₄₆En₅₁Fs₁₂) as well as in the trachybasalt (Wo₄₀En₄₂Fs₈ to Wo₄₇En₄₇Fs₁₅) but with a higher abundance of augite (data from 5 crystals, n = 51 for alkali basalt, and 7 crystals, n = 57 for trachybasalt). Patchy-zoning or oscillatory-zoning are observed in the clinopyroxene.

• Clinopyroxene mega-phenocrysts

This group of crystals is found in alkali basalt (L.81), but is absent in basanite and trachybasalt. All data fall within the Ca-Mg-Fe-rich range, predominantly exhibiting diopside compositions (Wo₄₄En₄₂Fs₈ to Wo₄₆En₄₇Fs₁₂) (data from 4 crystals, n = 38).

The compositional differences of the investigated clinopyroxene crystals are shown in Figure 9.

Examination of clinopyroxene structural behavior using the Na + Al^IV versus Al^VI + 2Ti + Cr diagram reveals that a significant portion of the basanite and alkali basalt samples plot above the Fe³⁺ = 0 line (Figure 10). This distribution indicates the relative amount of ferric iron present [14], reflecting more oxidizing conditions and high oxygen fugacity during pyroxene crystallization for LABL. In contrast, most clinopyroxenes from the trachybasalt samples plot near the Fe³⁺ = 0 line, indicating more reducing conditions and a negligible contribution of Fe³⁺ to their structural formula compared to those in basanite and alkali basalt samples.

These differences in oxidation state not only confirm the variability of physicochemical conditions controlling mineral crystallization but also strengthen the reliability of the applied geothermobarometric models for the LABL volcanic rocks.

6. Discussion

6.1. Crystallization Conditions

Due to the impossibility of directly measuring intensive parameters (e.g., temperature, pressure, and oxygen fugacity) during magma generation, movement and accumulation in magma chambers, as well as its crystallization and emplacement within the crust, geoscientists commonly use indirect methods to quantify these. One such method involves utilizing the equilibrium composition between coexisting minerals and the melt from which they were crystallized. Determining these conditions is based on the thermodynamic principles of a chemical mineral-melt equilibrium, achieved by characterizing the chemical properties of experimental products analogous to the compositions of igneous minerals and their coexisting melts [22,42].

Olivine compositions are widely used to constrain magmatic conditions such as temperature, oxygen fugacity, and H₂O_melt. However, post-crystallization elemental diffusion may change the initial compositions, leading to large uncertainty in thermodynamic estimates [43]. Clinopyroxene is considered to be more resistant to post-growth alteration and is therefore an important mineral for thermo-barometric estimates, determination of oxygen fugacities, petrological interpretations, including the type of magma series, and the tectonic environment of magma formation [44]. Moreover, clinopyroxene is present in a wide range of rock types (from komatiite and picrite to basalt, andesite, dacite, and even some rhyolites) and occurs in diverse tectonic settings, including hot spots, intracontinental rifts, mid-ocean ridges, oceanic islands, arc islands, and continental margin magmatic arcs [21].

6.1.1. Whole Rock/Glass Geothermometry

Putirka [21] showed that, in spite of the simplicity and narrow calibration range, the Helz and Thornber [45] and Montierth et al. [46] thermometers work remarkably well for liquid (or glass) solidus thermometers. The formulations depend only upon the MgO content in the liquid (MgO^liq): T (°C) = 20.1MgO^liq + 1014 °C [45] and T (°C) = 23.0MgO^liq + 1012 °C [46].

According to Helz and Thornber [45], the whole-rock thermometry for the LABL basanitic lavas is estimated to be 1237–1243 °C (avg = 1239 °C) and for the alkaline basaltic lavas to be 1151–1209 °C (avg = 1180 °C) independent of pressure. Moreover, the Putirka [21] Equations (13)–(16) can be applied to glass (liquid) thermometry and are valid for olivine-bearing volcanic rock over an extensive compositional and P-T range (P = 0.0001–14.4 GPa; T = 729–2000 °C; SiO₂ = 31.5–73.64 wt.%; Na₂O + K₂O = 0–14.3 wt.%; H₂O_melt = 0–18.6 wt.%). The Equation (16), which is similar to the Yang et al. [47] Equation (4), applies specifically to liquids that are in equilibrium with olivine+plagioclase+clinopyroxene. It performs less well when additional phases, such as spinel or other oxides, are present.

Using the Putirka [21] Equation (13): T (°C) = 26.3MgO^liq + 994.4 °C), the whole-rock geothermometry estimates indicate that the basanites crystallized within a temperature range of 1286–1294 °C (avg = 1289 °C), whereas alkali basalts record temperatures between 1174–1250 °C with SEE of ±71 °C. Additionally, the glass thermometry for a group of alkali basalts with a hyalo groundmass (L.81, L.38) reveals an average temperature of 1056 °C using the Helz and Thornber [45] method and 1049 ± 71 °C using Putirka [21] Equation (13) (independent of pressure). The groundmass glass serves as an indicator of the composition of the residual melt from which crystals formed during the eruption; thus, its lower estimated solidus temperatures, in contrast to those derived from whole-rock analyses, provide insights into the cooling history of the residual melt before the eruption occurred.

6.1.2. Olivine Geothermometry

When using thermometers or barometers based on mineral-mineral or mineral-melt equilibria, it is crucial that equilibrium between the phases is achieved and maintained. Otherwise, any calculated pressure-temperature condition will be meaningless [21]. In the case of the olivine–liquid thermometer, the composition of either whole-rock or glass must be considered. If the liquid is in equilibrium with olivine, the Fe-Mg exchange coefficient between olivine and liquid is expected to be in the range of K_D (Fe-Mg)^ol-liq = 0.30 ± 0.03 [31,48], thereby confirming the applicability of the thermometer to the volcanic system.

The Fe-Mg distribution coefficients (K_D) range from 0.11 to 0.69 in basanites and from 0.27 to 0.33 in alkali basalts. Based on the Rhodes diagram (Figure 11; [49]), olivine is not generally in equilibrium with the coexisting liquid in the LABL rocks, except in some samples. Petrographic evidence, including absorption features, rounded crystal morphologies, and patchy zoned olivine crystals (in back scattered image), confirm liquid disequilibrium between olivine and the surrounding melt.

The thermometers of Putirka et al.’s [50] Equations (2) and (4), later improved in Putirka [21] Equations (21) and (22), provide the most reliable temperature estimates when water is present, with standard error of estimate (SEE) of ±53 °C and ±29 °C respectively.

Using both above equations and averaging the results at constant pressure (e.g., 1.4 GPa as the highest estimated pressure value for clinopyroxene of the LABL), the crystallization temperature of the Ol with K_D(Fe-Mg)^ol-liq ca. 0.30 ± 0.03 is calculated to be 1385-1393 °C for basanites (n = 7), and 1275–1339 °C for alkali basalts (n = 65).

6.1.3. Clinopyroxene Geothermobarometry

Determining the temperature and pressure conditions that control magma crystallization before eruption [21] is a crucial topic in igneous petrology, volcanology, and geochemistry. Due to its sensitivitfy to the temperature and pressure prevailing in magmatic systems, numerous thermobarometers have been developed and based on clinopyroxene-only [20,51,52] as well as clinopyroxene-liquid equilibria [21,32,53,54].

While both approaches provide critical insights into crystallization conditions, Wang et al. [42] emphasize that clinopyroxene-only thermobarometry is one of the most practical tools to reconstruct crystallization pressures and temperatures of clinopyroxene. Because it does not require any information about coexisting silicate melt or other co-crystallized mineral phases, it has been widely used to elucidate the physiochemical conditions of crystallizing magmas. In this section, we examine the application of clinopyroxene-based geothermobarometers to the LABL, assessing their effectiveness in constraining magmatic evolution.

In the Soesoo [19] method, which incorporates all the oxides present in the mineral, temperature estimation is performed based on the calculation of X_PT and Y_PT indices. Accordingly, trachybasalt clinopyroxene crystallized at T = ca. 1150–1200 °C and P = ca. 0.3–0.6 GPa, alkali basalt clinopyroxene at T = ca.1165 to 1270 °C and P = ca. 0.4–1.4 GPa, and basanite Clinopyroxene at T = ~1170–1230 °C and P =~ 0.5 to 1.25 GPa (Figure 12).

Putirka’s [21] Equation (32d), based on the Nimis and Taylor [20] clinopyroxene-only thermometer model and Putirka’s [21] Equation (32b), delivers the following values for the intermediate part of LABL: trachybasalt, T = 1156 − 1223 ± 58 °C, with an average of 1190 ± 58 °C and P = 0.13 − 0.9 ± 0.26 GPa with an average of 0.51 ± 0.26 GPa. Similarly, for alkali basalts, the estimated crystallization T = 1135 − 1287 ± 58 °C (average = 1209 ± 58 °C) and P = 0.1 − 1.4 ± 0.26 GPa (average = 0.65 ± 0.26 GPa). In contrast, basanites show a wide crystallization temperature interval (T= 934 − 1278 ± 58 °C) with an average of 1200 ± 58 °C, accompanied by a P = 0.15 − 1.34 ± 0.26 GPa, averaging 0.71 ± 0.26 GPa. The results derived from various geothermobarometric calculations are presented in

Table 1. P-T results of different Clinopyroxene-only geothermobarometry models for the investigated LABL rocks.

Clinopyroxene-Only Thermobarometry
	Soesoo (1997) [19]		Putirka (2008) [21] RiMG
			Equation (32a) (SEE = ±0.26 GPa)	Equation (32b) (SEE = ±0.26 GPa)	Equation (32d) (SEE = ±58 °C)
	T (°C)	P (GPa)	P (GPa)	P (GPa)	T(°C)
Trachybasalt n = 93	1150–1200	0.3–0.6	0.12–0.9 (avg = 0.55)	0.13–0.9 (avg = 0.51)	1156–1223 (avg = 1190)
Alkali basalts n = 418	1165–1270	0.4–1.45	0.03–1.99 (avg = 0.74)	0.1–1.4 (avg = 0.65)	1135–1287 (avg = 1209)
Basanites n = 169	1170–1230	0.5–1.25	0.1–1.57 (avg = 0.8)	0.15–1.34 (avg = 0.71)	934–1278 (avg = 1200)

.

X_PT = 0.446SiO₂ + 0.187TiO₂ − 0.404Al₂O₃ + 0.346FeO(total) − 0.052MnO + 0.309MgO + 0.431CaO − 0.446Na₂O

Y_PT = −0.369SiO₂ + 0.535TiO₂ − 0.317Al₂O₃ + 0.323FeO(total) + 0.235MnO − 0.516MgO − 0.167CaO − 0.153Na₂O

Over the last thirty years, various geothermobarometric methods have been developed to estimate the crystallization conditions of magma chambers, among which the use of clinopyroxene-liquid equilibrium has been widely applied [21,53,54,55,56,57,58,59]. Although this thermobarometer provides valuable insights, it remains challenging due to the open nature of magmatic systems and several magmatic processes, including contamination, mixing and mingling of distinct magma batches.

Pressure is one of the key variables that controls magmatic phase equilibria. However, estimating magma chamber pressures from erupted products is challenging [53]. The successful application of this method requires not only the accurate knowledge of the mineral composition but also of the whole-rock or glass composition. Consequently, the method becomes unreliable if clinopyroxene and the bulk rock or glass composition are not in equilibrium. These issues can be assessed via the Fe-Mg exchange coefficients of clinopyroxene-liquid K_D(Fe-Mg)^cpx-liq. When K_D(Fe-Mg)^cpx-liq is close to or equal to 0.28 ± 0.08 [49], the values can be interpreted as indicating clinopyroxene-liquid equilibrium. However, based on the Rhodes diagram (Figure 13), the majority of the clinopyroxenes in each sample are not in equilibrium with the coexisting melt. This implies that the magmatic system was open, likely influenced by multiple melt injections or rapid magma ascent. Petrographic evidence such as sieve texture, absorption features, and oscillatory zonation within clinopyroxene crystals confirm an open magmatic system for the LABL.

The Fe-Mg exchange coefficients of clinopyroxene-only composition (K_D(Fe-Mg)^cpx) range from 0.24 to 0.31 for basanites. However, the corresponding K_D(Fe-Mg)^cpx-liq range from 0.3 to 1.0, indicating there is no reliable data to justify the usage of this method for the LABL basanites. In contrast, in alkali basalts, K_D(Fe-Mg)^cpx is from 0.28 to 0.29 and K_D(Fe-Mg)^cpx-liq is 0.24 to 0.3 which are thus in acceptable ranges to apply mineral-liquid thermobarometry. In trachybasalt, K_D(Fe-Mg)^cpx ranges from 0.25 to 0.27 and K_D(Fe-Mg)^cpx-liq is 0.1 to 0.26.

Here, the Putirka et al. [32] models are used to estimate the temperature and pressure conditions prevailing in the magma chamber during the crystallization. The crystallization temperature of clinopyroxene in alkali basalt ranges between 1071 and 1215 ± 0.61 °C at a pressure of 0.77 − 1.0 ± 0.48 GPa (see

Table 2. P-T results of different clinopyroxene-liquid equilibrium geothermobarometry models for LABL.

Clinopyroxene-Whole-Rock Equilibrium Thermobarometry
	Putirka et al. (2003) [32]		Putirka (2008) [21] RiMG MODELS
	T (°C)	P (GPa)	P (GPa)	T (°C)
	(SEE * = ±61 °C)	(SEE = ±0.48 GPa)	Equation (31) (SEE = ±0.3 GPa)	Equation (33) (SEE = ±46 °C)
Trachybasalt n = 6	1092–1112 (avg = 1104)	0.00–0.2 (avg = 0.1)	0.5–0.7 (avg = 0.6)	1103–1141 (avg = 1125)
Alkali basalt n = 11	1156–1198 (avg = 1182)	0.3–0.8 (avg = 0.6)	0.2–0.8 (avg = 0.56)	1115–1185 (avg = 1155)
Basanite	N/A	N/A	N/A	N/A

* SEE considering for anhydrous system. N/A: not available.

). Moreover, the Putirka [21] Equation (31) represents a barometer based on the partitioning of Al between clinopyroxene and liquid, providing further insights into the pressure conditions of crystallization. Accordingly, the estimated crystallization pressure of clinopyroxene within alkali basalt ranges from 0.63 to 1.06 ± 0.3 GPa. Additionally, utilizing the model presented in Equation (33) by Putirka [21], the crystallization temperature of clinopyroxene is projected to fall between 1132 and 1206 ± 0.46 °C, as detailed in Table 2.

Figure 14 illustrates the P-T binary diagrams that model temperature and pressure for all rock types of the LABL, employing the clinopyroxene-only method using Mag-Min_PT.

Utilizing clinopyroxene-only thermobarometry the crystallization temperature of clinopyroxenes in the basanite ranges from T = 1120–1240 °C at P = 0.2–1.2 GPa, corresponding to depths of roughly 8–46 km. In the case of alkali basalt, the crystallization conditions are T = 1110–1260 °C and P = 0.1–1.35 GPa (corresponding to depths of about 9–51 km), while for the trachybasalt, the temperatures T = 1110 to 1190 °C under low to moderate pressures of 0.05 to 0.75 GPa (corresponding to depths of 2 to 29 km). Furthermore, thermobarometric analysis utilizing the clinopyroxene-melt method indicates that the crystallization temperature of magma in the alkali basalt ranges from T = 1071–1215 °C, at P = 0.55 and 1.05 GPa (depths of ~21–40 km).

Also, the temperature and pressure variations from the core to the rim of selected clinopyroxene phenocrysts in all three rock types (Figure 15) indicate new magma pulses into the magma reservoir.

The overall geothermobarometric results derived from both clinopyroxene-only and clinopyroxene–melt equilibria suggest variable crystallization conditions across the LABL rock types, with basanites and alkali basalts forming under higher temperatures and pressures compared to trachybasalts. However, the structural and chemical features of the clinopyroxenes—such as sieve textures, oscillatory zoning, and absorption features—along with the deviation of many K_D(Fe-Mg)^cpx-liq values from the equilibrium range, indicate an open magmatic system.

The presence of ferric iron (Fe³⁺) in many basanitic and alkaline clinopyroxenes, as evidenced in the Na + Al^IV vs. Al^VI + 2Ti + Cr diagram, further supports crystallization under relatively oxidizing conditions, which can be linked to magma recharge [60] or crustal contamination [61]. These conditions can modify the redox state of the melt and influence Fe-Mg partitioning, leading to potential biases in thermobarometric estimates if Fe³⁺ is not considered. In contrast, trachybasaltic samples show clinopyroxene signatures consistent with more reducing environments, supporting more stable and equilibrium-driven crystallization pathways in those magmas.

Taken together, the combination of mineral chemistry, structural indicators, and Fe³⁺ distribution confirms that the LABL magmatic system experienced multiple pulses, dynamic crystallization conditions, and evolving oxidation states.

6.2. Thermobarometric Constraints and Limitations

The thermobarometric estimates presented in this study provide critical insights into the crystallization conditions of the LABL magmas. They analyzed clinopyroxenes record a ~0.1–1.4 GPa pressure range (Figure 14), suggesting crystallization from Moho depths (~30 km) to shallow crustal levels (~5 km). Moreover, core-to-rim P-T zoning in olivine and clinopyroxene (Figure 15) is consistent with recharge-driven thermal perturbations. This disequilibrium likely reflects open-system processes such as magma mixing, rapid ascent, or recharge events supported also by resorbed olivine cores and clinopyroxene sieve textures. However, there are also significant constraints and limitations inherent in the methods by (1) magma composition and mineral- melt equilibrium, (2) model-specific uncertainties and (3) redox conditions. Here, we evaluate these limitations and their implications for interpreting crystallization histories.

6.2.1. Magma Composition and Mineral-Melt Equilibrium Test

Magma Composition: Whole-rock compositions, often used as melt compositions, may not represent true equilibrium liquids due to cumulate entrainment or contamination of magma. Groundmass glass thermometry (e.g., 1056 °C vs. whole-rock 1289 °C in alkali basalts) highlights this difference, with glasses preserving late-stage, near-solidus conditions.

Olivine- melt Equilibrium: The large scatter in K_D(Fe-Mg)^ol-liq (0.12–1.43) in LABL samples (Figure 10) indicates mineral-melt disequilibrium, likely driven by magma mixing, rapid ascent, or syn-eruptive degassing. Notably, only a limited number of basanites and alkali basalts yield K_D values within the equilibrium range (0.27–0.33), permitting reliable mineral-liquid thermobarometry.

Clinopyroxene- melt Equilibrium: While Putirka et al. [21] formulations (e.g., Equation (31)) are robust for equilibrium pairs, their application to LABL is restricted to limited number of trachybasalt and alkali basalts due to disequilibrium in other rocks. Even here, the ±0.3 GPa pressure uncertainty underscores the challenge of resolving mid-crustal (~0.5 GPa) vs. deep-crustal (~1.0 GPa) reservoirs.

6.2.2. Model-Specific Uncertainties

Olivine geothermometry: The inherent statistical uncertainty of olivine geothermometry arises from the specific empirical or thermodynamic calibration of the thermometer equation itself. For the model of Putirka et al. [31] Equation (2), the SEE is ±53 °C. The model represented by Putirka et al. [31] Equation (4), later refined as Equation (22) in Putirka [21], has an SEE of ±29 °C and is noted for its improved reliability in water-bearing systems. Crucially, even when mineral-melt equilibrium is perfectly maintained, these specific models carry an inherent uncertainty of at least ±29 °C or ±53 °C, respectively, due to limitations in their calibration.

Clinopyroxene-only thermobarometry: Methods like Putirka’s [21] Equation (32d) are sensitive to Al^VI/Al^IV partitioning, which varies with oxygen fugacity (fO₂) and H₂O content. The ±0.26 GPa pressure uncertainty (Putirka [21]), due to fO₂ variations that influence Al partitioning in clinopyroxene [21], corresponds to depth errors of ~10 km, complicating precise magma storage estimates. The Soesoo [19] and Putirka [21] models yield overlapping but non-identical P-T ranges (Table 1), reflecting differences in calibration datasets (e.g., hydrous vs. anhydrous systems).

6.2.3. Redox Conditions and Their Thermobarometric Implications

Iron is the main element that is readily oxidized or reduced in response to the range of redox states experienced by crustal rocks [62,63,64,65,66]. The amount of ferric (Fe³⁺) iron is commonly a good indicator of the redox budget of a rock [67], and the ratio of ferric to ferrous (Fe²⁺) iron provides a first-order indication of its oxidation state [68]. Under such conditions, assuming total Fe as Fe²⁺ without correcting for Fe³⁺ content may lead to inaccurate Fe/Mg ratios, resulting in erroneous values for K_D (Fe-Mg exchange coefficient), as well as over- or under-estimated crystallization temperatures and pressures. Therefore, incorporating Fe³⁺ estimates is essential to improve the accuracy of thermobarometric calculations.

6.3. Magma Type and Tectonic Setting Based on Clinopyroxene Composition

Numerous weathered or even slightly metamorphosed basalts exhibit fresh clinopyroxene crystals within the altered groundmass. The microprobe analysis of these relict grains can be used to identify the magma type of the host lava [11].

LeBas [69] indicated that the concentrations of Al and Ti in pyroxenes is influenced by the magma’s alkalinity. Utilizing Al₂O₃ and SiO₂ contents in the chemical composition of pyroxenes allows for the differentiation of alkaline, peralkaline, and subalkaline types of magma (Figure 16a). The diagram illustrates that the clinopyroxenes from the LABL are categorized as peralkaline (specifically for basanite and certain alkali basalts) and alkaline-subalkaline (including alkali basalt and trachybasalt) as depicted in Figure 16a. In a Ti versus Na+Ca diagram [10], basanite and alkali basalt are situated within the alkali basalt field, whereas trachybasalt is positioned along the boundary between sub-alkali basalt and alkali basalt fields (Figure 16b).

Nisbet and Pearce [11] have indicated that pyroxenes found in within-plate alkali basalts have high Na and Ti and low Si contents, whereas pyroxenes of within-plate tholeiites and volcanic arc basalts contain more Ti, Fe and Mn. Tholeiitic basalts erupted within plates in oceanic islands or continental rifts (WPT samples), and alkali basalts erupted within plates (WPA samples) [11]. The pyroxene composition based on discriminant functions (F1-F2) well distinguishes between alkaline basalts in intra-plate settings (WPA), within-plate tholeiitic magmas (WPT), and volcanic arc basalts (VAB) from one another. However, a significant overlap was observed between within-plate tholeiites and oceanic floor basalts (WPT-OFB), as well as between volcanic arc basalts and oceanic floor basalts (VAB-OFB). In this diagram, the trachybasalt and alkali basalt fall within the volcanic arc basalts (VAB) field, whereas basanite plot in within-plate alkali basalt. Also, diagrams of pyroxene composition based on Na₂O-MnO-TiO₂ and TiO₂-SiO₂ provide the basis for visual discrimination [11]. In the Na₂O-MnO-TiO₂ diagram, basanite and alkali basalt are located within the WPA field and trachybasalt is plotted in the VAB+WPA field. In the TiO₂-SiO₂ discrimination diagram, trachybasalt, alkali basalt and basanite are located within the WPA and to some extent in the VAB (Figure 16c–e). According to Leterrier et al. [10], the clinopyroxenes found in the LABL trachybasalt and alkali basalt exhibit similarities to those from orogenic tholeiitic basalt when taking into account the Ti and Ca cations concentrations. Furthermore, the Ti and Ca contents in clinopyroxene from basanite are comparable to those in clinopyroxene derived from alkali basalt (Figure 16f). The local geology in eastern Iran indicates a subduction- and collision-dominated tectonic regime. Intrusive rocks include granodioritic and biotite granite plutons, monzodiorite to dacite porphyritic dykes, which are calc-alkaline, metaluminous, and I-type, with geochemical signatures of volcanic arc and post-collisional magmas [26]. Houshmand-Manavi et al. [28] demonstrated that the LABL is related to a post-collisional setting in the Oligocene-Miocene, through comprehensive field observations and whole-rock geochemical analysis. It is postulated that an extensional environment related to the activity of two dextral basement faults that trend north-south, situated to the east and west of the region provided a pathway for magma ascent. The geology is characterized by the presence of deep reaching basement faults that facilitated the ascent and eruption of magma, which can be associated with the Zahedan and Harriroud faults, as evidenced by the alignment of the lava, which run parallel to these faults and being situated in the intervening space between them. These lavas occasionally exhibit features indicative of a pseudo-active margin (VAB), attributed to mantle metasomatism caused by fluids released during the subduction of the Sistan oceanic lithosphere (Late Cretaceous-Middle Eocene, [9,26]). Moreover, the tectonic discrimination diagrams based on clinopyroxene and whole-rock [28] chemistry more likely reflect some inherited characteristics from a heterogeneous source, which prohibits to determine the true LABL tectonic setting. Therefore, the LABL is classified as post-collisional. However, it should be noted that recent studies have demonstrated the limitations of conventional clinopyroxene-based geochemical discrimination diagrams. In this regard, Li et al. [70] applied a machine learning approach using a global clinopyroxene dataset and achieved an accuracy of over 90% in distinguishing tectonic settings. Although the present study relies on traditional approaches, applying such ML-based methods in future work could provide more precise and reliable interpretations of the tectonic setting of the LABL basalts.

F1 = −0.012×SiO₂− 0.0807×TiO₂ + 0.0026×Al₂O₃ − 0.0012×FeO(tot) − 0.0026×MnO + 0.0087×MgO − 0.0128×CaO − 0.0419×Na₂O

F2 = −0.0469×SiO₂ − 0.0818×TiO₂ − 0.0212×Al₂O₃ − 0.0041×FeO(tot) − 0.1435×MnO − 0.0029×MgO + 0.0085×CaO + 0.016×Na₂O

6.4. Magma Crystallization and Evolution Modeling

A volcanism model of LABL is presented in the Sistan Suture Zone tectonic evolution sketch in Figure 17a. It shows the delamination of thickened oceanic lithosphere root which caused to generate LABL parent magma from low degrees of partial melting of the upwelled asthenosphere beneath the studied area [28]. Basaltic magma, due to its low viscosity and high temperature, exhibits significant mobility and begins to ascend immediately after generation in the mantle.

The calculated P-T ranges (0.1–1.4 GPa; ca 940–1290 °C) and petrographic evidence (e.g., sieve-textured clinopyroxene, resorbed olivine) collectively suggest the basaltic magma ascended after generation in the mantle and accumulated at depths exceeding ca. 45 km (Moho depth in East Iran [71,72]) in the lower crust, and subsequently initiated the crystallization. We propose a multi-tiered magma plumbing system beneath the LABL through a three-stage model:

• Deep Storage (1.0–1.4 GPa; ~38–51 km):

High-P clinopyroxene mega-phenocrysts (e.g., L.81) and Fo > 90 olivine cores crystallized in the lower crust/uppermost mantle, facilitated by localized heating during lithospheric delamination [39], supported by evidences such as coexisting high-Mg# clinopyroxene (Wo₄₆En₄₇) and olivine (Fo₉₂) with Cr-spinel inclusions.

• Mid-Crustal Storage (0.3–0.8 GPa; ~11–30 km):

When fault reactivation (Zahedan/Harirud) opened pathways for magma ascent, the basaltic magma reached depths of 30–10 km (middle crust) leading to the crystallization of plagioclase and clinopyroxene. This is supported by the estimated pressures of clinopyroxene mega-phenocrysts in LABL basalts and basanites that most pressure frequency distribution represents P = 0.4–0.7 GPa, indicating periods of crystallization occurred in the middle crust magma chamber. The estimated pressure values, (P = 0.4–0.6 GPa) at depths of 15–20 km, are associated with trachybasalts (Figure 17b,c).

The mineral chemistry data also show that this mid-crustal storage refilled by deep-derived magma raised from the deep storage (1.0–1.4 GPa; ~38–51 km). The mixing between deep-derived and evolved melts, recorded by (a) reverse-zoned clinopyroxene (high-Mg rims; Figure 8a). (b) olivine with Mn-enriched rims (Figure 6a), indicating Fe-Mg exchange with fractionated melt.

• Shallow Conduits (<0.3 GPa; <11 km):

Eventually, the remaining magma, along with the phenocrysts, rises through fractures and faults to the surface. Trachybasalt crystallization (0.1–0.75 GPa) reflects rapid ascent through fractures created by a tensile mechanism resulting from the action of dextral Zahdan and Harriroud faults in the area (Figure 17b,c). This is supported by (a) Glassy groundmass (quenched melt), and (2) Lack of zoning in microlites that represent a short residence time.

In summary, this architecture explains the LABL’s chemical diversity: basanites represent deep, low-degree melts, while trachybasalts reflect shallow crustal assimilation.

In the following, sample L.81 from alkali basalts with glassy groundmass was selected to investigate the differences between crystallization temperature and pressure of clinopyroxene mega-phenocrysts and phenocrysts. As shown in Figure 18, the maximum temperature estimated for mega-phenocrysts is ~1200 °C at the maximum pressure of ~1 GPa, while the maximum temperature for phenocrysts is ~1190 °C at the maximum pressure of ~0.7 GPa, based on the clinopyroxene-only thermobarometer. In other words, the crystallization of clinopyroxene mega-phenocrysts and phenocrysts started at depths of ~38 km and ~26 km, in the lower and middle crust, respectively. There is no notable reduction in temperature during the pressure release of ca. 0.3 GPa, which indicates a continuous and rapid ascent from depths of approximately 11 km to the surface and quenching after eruption, leading to the formation of a glassy groundmass in this sample.

7. Conclusions

The mineral chemistry of the LABL indicates a peralkaline to subalkaline magma nature, suggesting an origin in an intra-plate tectonic setting and volcanic arc environment. This contradicts the local geological situation. It is therefore important to note that these implied geotectonic settings probably originate in inherited traits from a heterogeneous mantle source. Therefore, based on the real tectonic regime during the Oligocene-Miocene in Iran and field observations, a more realistic post-collision tectonic environment is suggested for the LABL. Additionally, we propose that LABL magmas were stored in a trans crustal (through-crustal) plumbing system (2–51 km depth), with eruptions triggered by fault-mediated recharge from deep reservoirs. Petrographic and thermobarometric data reveal cyclic recharge-fractionation-eruption episodes, characteristic of the LABL post-collisional alkaline volcanism.

The application of multiple thermobarometric approaches (whole-rock, olivine, and clinopyroxene chemistry) consistently indicates that the studied basaltic lavas were generated and stored under high-temperature conditions. Basanites record the highest crystallization temperatures (up to ~1390 °C) at upper mantle depths (ca. 46 km at 1.2 GPa) to lower crustal levels (ca. 8 km at 0.2 GPa), whereas alkali basalts show slightly lower temperatures (~1170–1340 °C) over broader pressure and depth ranges (0.1–1.35 GPa; 4–51 km). Trachybasalts, in contrast, crystallized at comparatively lower temperatures (~1110–1190 °C) and at shallower levels (2–29 km), reflecting their confinement to middle–upper crustal reservoirs.

These results collectively demonstrate a polybaric magma plumbing system, in which basanites and alkali basalts were stored from the lower crust to near-surface levels, while trachybasalts were mainly restricted to mid- to upper-crustal chambers. Moreover, the chemical compositions of phenocrysts in the studied basaltic lavas provide evidence of magma recharge through multiple pulses of new magma injected into the existing reservoirs prior to eruptions. The eruption history of the LABL reflects the interplay between deep melt generation associated with lithospheric thinning and fault-controlled ascent, with the observed chemical diversity resulting from polybaric fractionation and crustal assimilation.