Integrated environmental, phytochemical, and genetic assessment of Avicennia marina populations along the Arabian Gulf coast of eastern Saudi Arabia Authors: Aisha S Al-Shammari Published: 2026 ABSTRACT AbstractBackgroundAvicennia marina dominates mangrove communities along the hypersaline Arabian Gulf, yet links among local habitat conditions, secondary metabolite profiles, and genetic structure remain poorly characterized. This study examined four A. marina stands—Tarout (P1), Saihat (P2), Al Qatif (P3), and Safwa (P4)—near Dammam, eastern Saudi Arabia, integrating field observations, water and pore-soil physico-chemical analyses, GC–MS/MS profiling, SCoT markers, and rbcL barcoding.ResultsSites differed strongly in stand structure, water chemistry, and sediment ion loading, with Tarout, Saihat, and Safwa exhibiting marine–hypersaline waters and Al Qatif more dilute, high-alkalinity conditions but highly saline sediments. GC–MS/MS revealed 27 compounds and a shared flavonoid-rich baseline dominated by 7,3’,4’,5’-tetramethoxyflavanone, superimposed on distinct site-specific terpenoid and phenolic signatures. SCoT markers detected moderate nuclear diversity and clear genetic differentiation among populations, while rbcL sequences, though conservative, resolved low but informative chloroplast structuring.ConclusionsCollectively, the environmental, chemical, and molecular evidence indicates that A. marina in this region comprises a mosaic of locally differentiated, genetically structured stands with distinct chemotypes, providing a preliminary basis for managing these stands as locally differentiated conservation units rather than a single homogeneous resource.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12870-026-09607-2.Keywords: Avicennia marina, Mangroves, Secondary metabolites, GC–MS/MS, SCoT markers, rbcL barcoding, Genetic diversity, Salinity Background Mangrove ecosystems occupy the intertidal zones of tropical and subtropical coasts, where they perform key ecological functions including shoreline stabilization, carbon sequestration, nutrient cycling, and provision of nursery habitats for fish and invertebrates [1, 2]. Despite their importance, mangroves are increasingly threatened by coastal development, pollution, hydrological alteration, and climate-driven sea-level rise [3, 4]. Understanding how mangrove species respond to local environmental conditions—particularly soil and water chemistry—and how this is reflected in their morphology, genetic structure, and secondary metabolite profiles is therefore critical for conservation, restoration, and sustainable utilization [2, 5]. Classical taxonomy of mangroves has largely relied on morphological traits such as leaf shape and succulence, root architecture, reproductive structures, and wood anatomy; however, phenotypic plasticity under variable salinity and inundation often complicates species delimitation and population-level comparisons [6, 7]. Integrating detailed morphometric analyses with molecular markers and chemical profiling offers a more robust framework for assessing intraspecific variation, local adaptation, and cryptic diversity [8–10]. At the genetic level, nuclear markers such as microsatellites, ISSR, SCoT and SNPs, together with plastid and mitochondrial sequences, have been widely used to infer genetic diversity, connectivity, and demographic history in mangrove taxa including Avicennia, Rhizophora, Sonneratia and Bruguiera [5, 11]. These tools are indispensable for identifying evolutionarily significant units and for guiding restoration efforts in fragmented coastal landscapes. DNA barcoding complements these approaches by providing a standardized, sequence-based system for species identification and phylogenetic inference. However, commonly used plastid barcodes such as rbcL are relatively conserved and generally offer only coarse resolution at or below the species level. Chloroplast barcodes such as matK, rbcL and trnH–psbA, sometimes combined with ITS or other nuclear loci, have been successfully applied to mangrove floras to resolve taxonomic ambiguities, authenticate propagules and nursery material, and detect hybridization and introgression [12, 13]. Barcoding data, when analyzed alongside population-level molecular markers, enable comparison of plastid versus nuclear patterns and help distinguish historical from contemporary processes shaping genetic structure [14, 15]. Environmental context is a major driver of mangrove variation. Sediment (soil) and pore-water chemistry—particularly salinity, pH, redox potential, organic matter, major cations/anions, and trace metals—strongly influence mangrove growth, nutrient acquisition, and community composition [16, 17]. Detailed soil analysis (e.g. texture, conductivity, available NPK, micronutrients, and heavy metals) together with surface and pore-water analysis (salinity, pH, dissolved oxygen, nutrients, and contaminants) is essential to link edaphic and hydrological conditions with plant performance, physiological status, and contaminant uptake [10, 18]. Such datasets are particularly important in impacted estuaries, where industrial effluents and urban runoff may alter sediment geochemistry and bioaccumulation patterns in mangrove tissues [19, 20]. Mangroves are also rich in primary and secondary metabolites—such as phenolics, tannins, terpenoids, alkaloids and fatty acids—that underpin their antimicrobial, antioxidant, and anti-inflammatory activities and support their use in traditional medicine and as potential sources of novel bioactive compounds [21, 22]. Gas chromatography–mass spectrometry (GC–MS) provides a powerful platform for profiling volatile and semi-volatile constituents in leaves, bark, roots, and propagules, enabling qualitative and quantitative characterization of fatty acid methyl esters, sterols, terpenes, and small aromatic compounds [23, 24]. When GC–MS data are integrated with environmental, morphological, and genetic information, it becomes possible to examine how site conditions and genetic background influence metabolite composition, chemotypes, and potential pharmacological properties [25, 26]. In this context, an integrated approach applied to A. marina that combines (i) soil and water analyses to characterize the physicochemical environment; (ii) detailed assessment of key vegetative and reproductive traits; (iii) molecular markers and DNA barcoding to evaluate genetic diversity, population structure, and taxonomic identity; and (iv) GC–MS profiling of mangrove tissues to resolve metabolite composition can yield a holistic picture of how this species responds to local environmental gradients. By jointly examining habitat quality, morphology, genetics, and chemistry in A. marina, such a framework provides robust baseline information to support conservation prioritization, guide restoration strategies, and inform the sustainable utilization of mangrove-derived bioactive compounds. Materials and methods Study sites and field sampling Fieldwork was conducted at four mangrove sites (Fig. 1) along the Arabian Gulf coast near Dammam, Eastern Province, Saudi Arabia: Tarout (P1), Saihat (P2), Al Qatif(P3), and Safwa (P4). All stands are dominated by A. marina but differ in tree density, canopy cover, pneumatophore abundance, proximity to urban/industrial areas, and tidal exposure. Geographic coordinates for Tarout, Saihat, Al Qatif, and Safwa were recorded and are shown in the site location map. Sampling site locations of four A. marina populations (P1–P4) along the Arabian Gulf coast near Dammam, eastern Saudi Arabia: Tarout (P1), Saihat (P2), Al Qatif (P3), and Safwa (P4). Coordinates for each site are shown Surface soil (0–5 cm) was collected at each site using sterile spatulas or corers into pre-labelled sterile tubes after removal of loose litter, avoiding large roots, stones, and macrofauna. Approximately 50–100 g of soil was collected per subsample and pooled to form one composite sample per site; tools were disinfected with 70% ethanol between samples. Overlying water was collected at sites in sterile 500 mL bottles held just below the surface and facing upstream, with bottles pre-rinsed in site water where necessary and basic in situ parameters recorded during sampling. For each site, pore soil and water physical chemical parameters were measured in triplicate on the single composite sample (technical replication), and the values reported in Tables are means of these replicate measurements. Because only one composite soil and one composite water sample represent each site, these data provide descriptive, site level characterizations and are not used for formal statistical comparisons among sites. Leaves were collected from healthy, mature A. marina trees at each site at ~ 1–1.5 m height, taking three to five fully expanded leaves per tree with sterile scissors and placing them in sterile bags labelled by site and tree. Damaged or heavily fouled leaves were excluded; scissors and gloves were changed or disinfected with 70% ethanol between trees and sites. All samples were transported to the laboratory on ice and processed within 24 h or stored at − 20 °C (soil and leaves) and 4 °C (water) until analysis. Pore-water collection and chemical analysis Pore-water was sampled from the rhizosphere at Tarout, Saihat, Al Qatif, and Safwa using a hand-held plastic syringe attached to a perforated stainless-steel probe inserted to ~ 10–20 cm depth. Collected pore water was filtered in the field through 0.45 μm membrane filters into acid-washed polyethylene bottles, stored on ice, and transported to the laboratory. pH and electrical conductivity (EC) were measured immediately with calibrated portable meters [27]. Major cations (Ca²⁺, Mg²⁺, Na⁺, K⁺) and anions (Cl⁻, SO₄²⁻, HCO₃⁻, CO₃²⁻) were determined using standard methods: bicarbonate and carbonate by titration with standardized acid, chloride by argentometric titration, sulfate by turbidimetric or ion chromatographic analysis, and cations by flame photometry or inductively coupled plasma–optical emission spectrometry (ICP-OES) [27]. All glassware and containers were rinsed with deionized water before use, and blanks, standards, and replicate samples were included for quality control. Soil sampling and analysis Soil samples were collected from the 0–20 cm layer at each site using a stainless-steel auger, composited by site, and stored in clean polyethylene bags. In the laboratory, soils were air-dried, sieved (< 2 mm), and analyzed for pH and EC in soil–water suspensions using calibrated meters [28]. Soluble anions (Cl⁻, SO₄²⁻, HCO₃⁻/CO₃²⁻) and cations (Ca²⁺, Mg²⁺, Na⁺, K⁺) were determined in soil extracts following standard protocols: titration for bicarbonate/carbonate, argentometric titration for chloride, turbidimetric or ion chromatography for sulfate, and atomic absorption spectrometry or ICP-OES for cations [29, 30]. Ion concentrations were converted to milliequivalents per liter and checked by ionic charge balance (± 10%), with blanks, duplicates, spikes, and certified standards used for quality control. GC–MS/MS analysis For metabolite profiling, approximately 15–20 g of air-dried A. marina leaves from each site (pooled across individuals within that site) were extracted by maceration in absolute methanol in a tightly closed container kept in the dark for eight days. Thus, one biological composite leaf extract was prepared per population (P1–P4). Each site-specific extract was analyzed by GC–MS/MS with two replicate injections, and the relative abundances (%) reported in Table represent the mean of these replicate chromatograms. Consequently, the GC–MS data describe the chemical composition of each site level composite sample and are not subjected to formal statistical tests among sites. The methanolic extract was concentrated to dryness using a rotary evaporator to yield 7 g of crude residue, which was fractionated by column chromatography on activated silica gel (60–200 mesh). Elution was performed sequentially with hexane, hexane/ethyl acetate (90:10), hexane/ethyl acetate (80:20), and finally ethyl acetate [31]. Chemical analysis was carried out using an Agilent 7890 A gas chromatograph coupled to an Agilent 7000 triple quadrupole mass spectrometer. Separation was achieved on an HP-5ms capillary column (30 m × 0.25 mm i.d., 0.25 μm film; 5% phenyl methylpolysiloxane), with helium as carrier gas at 1 mL/min. Injector and detector temperatures were 200 °C and 250 °C, respectively, and 1 µL of each sample was injected. The mass spectrometer operated with electron impact ionization at 70 eV, interface temperature 250 °C, and scan range m/z 50–800. Compounds were identified by comparison of retention times and mass spectra with authentic standards, automated matching against NIST and WILEY libraries, and inspection of fragmentation patterns against published data. For most flavonoids and phenolic compounds, identifications should be regarded as putative (Level 2, putatively annotated compounds), as LC–MS/MS or NMR confirmation was not performed; only compounds with high library match scores (generally ≥ 80%) and consistent fragmentation patterns were retained in Table. DNA extraction and SCoT PCR amplification Genomic DNA was extracted from young, fresh leaves of A. marina collected at four sites using the DNeasy Plant Mini Kit (QIAGEN) according to the manufacturer’s instructions. At each site, leaves from 3 adult trees (n = 12 individuals in total) were sampled for SCoT analysis, with one individual representing one genotype. DNA concentration and purity were assessed with a NanoDrop 2000 spectrophotometer. Ten SCoT primers (Table 1) were used for amplification. Clear, reproducible bands were scored visually for each individual as present (1) or absent (0), and the resulting binary matrix (12 individuals × 67 loci) formed the basis for all subsequent genetic analyses. Each 25 µL PCR reaction contained 12.5 µL of 2× PCR Master Mix (Sigma), 2 µL of primer (10 pmol/µL), 3 µL of genomic DNA (10 ng/µL), and 7.5 µL of distilled water [13]. Reactions were run on a Perkin Elmer/GeneAmp® PCR System 9700 with an initial denaturation at 94 °C for 3 min; 40 cycles of 94 °C for 40 s, 50 °C for 50 s, and 72 °C for 60 s; and a final extension at 72 °C for 7 min. PCR products were separated on 1.5% agarose gels containing ethidium bromide (0.5 µg/mL) in 1× TBE buffer at 95 V, using a 100 bp DNA ladder as size marker; bands were visualized under UV and documented with a BIO RAD 2000 Gel Documentation System. Table 1SCoT primers and rbcL barcoding primers used for A. marina, with sequences and annealing temperaturesPrimerPrimer sequenceAnnealing (oC)SCoT SCoT-015’- ACGACATGGCGACCACGC − 3’61 °C SCoT-025’- ACCATGGCTACCACCGGC − 3’62 °C SCoT-035’- ACGACATGGCGACCCACA − 3’60 °C SCoT-045’- ACCATGGCTACCACCGCA − 3’60 °C SCoT-055’- CAATGGCTACCACTAGCG − 3’53 °C SCoT-065’- CAATGGCTACCACTACAG − 3’50 °C SCoT-075’- ACAATGGCTACCACTGAC − 3’52 °C SCoT-085’- CAACAATGGCTACCACGT − 3’53 °C SCoT-095’- ACAATGGCTACCACTGCC − 3’55 °C SCoT-105’- ACAATGGCTACCACCAGC − 3’55 °CBarcoding genes rbcL -F5’-ATGTCACCACAAACAGAGACTAAAGC-3’61 °C rbcL -R5’-TCGCATGTACCTGCAGTAGC-3’61 °CPrimer sequences for SCoT and rbcL were adapted from published sources [13] and the original SCoT primer description (see text for details) SCoT primers and rbcL barcoding primers used for A. marina, with sequences and annealing temperatures Primer sequences for SCoT and rbcL were adapted from published sources [13] and the original SCoT primer description (see text for details) rbcL chloroplast DNA barcoding and amplification For chloroplast DNA barcoding, the rbcL region was amplified from genomic DNA of A. marina individuals from each site. We initially amplified rbcL from 3 individuals per population; high quality sequences were obtained for at least three individuals per site. For downstream phylogenetic analyses, we selected one representative high quality rbcL sequence per population (P1–P4), corresponding to the accession numbers PZ364408, PZ350453, PZ353174, and PZ358527. Each 50 µL PCR reaction contained 1× buffer (Promega), 15 mM MgCl₂, 0.2 mM of each dNTP, 20 pmol of each rbcL primer (Table 1), 1 U GoTaq DNA polymerase (Promega), 40 ng genomic DNA, and nuclease-free water to volume [13]. Amplifications were performed on a GeneAmp® PCR System 9700 with an initial denaturation at 94 °C for 5 min; 40 cycles of 94 °C for 10 s, 61 °C for 1 min, and 72 °C for 1 min; and a final extension at 72 °C for 7 min. Products were resolved on 1.5% agarose gels stained with ethidium bromide (0.5 µg/mL) in 1× TBE buffer at 95 V, using a 100 bp ladder as marker. Bands were visualized under UV light and photographed with a BIO RAD 2000 system. Amplified fragments were purified using the EZ-10 Spin Column PCR Purification Kit according to the manufacturer’s protocol. Sequencing of rbcL barcodes Purified rbcL amplicons from 4 populations were sequenced on an ABI PRISM 3730XL Analyzer using BigDye™ Terminator Cycle Sequencing Kits. Single-pass sequencing was performed for each template using the corresponding forward primer. Fluorescently labelled extension products were cleaned by ethanol precipitation, resuspended in distilled water, and analyzed by capillary electrophoresis on the ABI 3730XL at the School of Sciences and Engineering, The American University in Cairo. Phylogenetic analysis rbcL sequences from A. marina and reference taxa were aligned using MAFFT [32]. Phylogenetic relationships were inferred under a maximum likelihood framework using raxmlHPC, and the resulting tree was visualized with the Interactive Tree of Life (iTOL) platform [33]. Genetic similarity, clustering, and diversity analysis For SCoT data from 4 populations and 12 individuals (3 per population), clear, reproducible bands were scored visually as present (1) or absent (0), with both polymorphic and monomorphic loci included in a binary data matrix. Genetic relationships among populations were quantified using Dice’s similarity coefficient. The binary SCoT matrix (presence/absence) was used to compute a Dice similarity matrix in PAST v3.26, and this matrix was converted to a distance matrix as (1 – Dice similarity) for clustering. A dendrogram was constructed using the UPGMA (Unweighted Pair Group Method with Arithmetic Mean) algorithm, with default PAST settings (UPGMA, no bootstrapping). Principal component analysis (PCA) of the SCoT data was also performed in PAST v3.26 using a covariance matrix of standardized binary variables; eigenvalues and the percentage of variance explained by the first two principal components (PC1 and PC2) are reported. An additional tree based on the Euclidean similarity index was generated using PAST3 software [34]. Genetic diversity was summarized using percentage of polymorphic bands (P%), polymorphic information content (PIC), Shannon’s information index (I), Nei’s gene diversity (H), effective number of alleles (Ne), number of observed alleles (Na), and expected heterozygosity (He). All diversity indices were calculated with GenAlEx, providing a preliminary assessment of genetic variation within and among A. marina populations [35]. Results Study area and site description The study area comprises four mangrove sites— Tarout (P1), Saihat (P2), Al Qatif(P3), and Safwa (P4)—along the Arabian Gulf coast near Dammam, eastern Saudi Arabia, characterized by arid coastal conditions, elevated salinity, and intertidal mudflat habitats dominated by A. marina. Visual assessment revealed clear spatial variation in vegetation structure, substrate, and pneumatophore density (Fig. 2). Tarout was characterized by relatively dense A. marina stands with abundant foliage and numerous pneumatophores emerging from soft, fine, light-colored sediment, with slightly turbid, calm water around the roots. Saihat showed moderately dense, healthy green mangroves with intermediate pneumatophore density and somewhat firmer sediment, accompanied by clearer water than at Tarout. In contrast, Al Qatif exhibited a more open canopy with fewer trees, lower pneumatophore density, and more exposed sediment between roots, suggesting a more sparsely vegetated and possibly more exposed intertidal zone. Safwa displayed tall, well-developed mangrove trees and extensive, evenly distributed pneumatophore fields on relatively dry, pale sediment, representing the most structurally complex and well-established stand among the four sites. These observations highlight substantial heterogeneity in mangrove development and habitat structure across the study area. A. marina habitats at four sites (P1–P4), showing differences in vegetation structure and pneumatophore density across panels (A) Tarout (P1), (B) Saihat (P2), (C) Al Qatif (P3), and (D) Safwa (P4) Water physical-chemical characteristics Because water chemistry measurements are based on a single composite sample per site with technical replication, differences among sites are described qualitatively and are not supported by formal statistical tests. Water physical-chemical properties varied among the four A. marina sites, with Tarout, Saihat, and Safwa showing typical marine–hypersaline signatures and Al Qatif clearly distinct and more dilute (Table 2). Conductivity, TDS, and salinity were high and comparable at Tarout, Saihat, and Safwa (conductivity 60.1–62.4 dS/m; TDS 37,863–39,312 mg/L; salinity 40.0–41.7 ppt), whereas Al Qatif showed markedly lower values (9.14 dS/m, 5,760 mg/L, 5.1 ppt), indicating a less saline environment. Dissolved oxygen increased from Tarout to Safwa (5.1–6.6 mg/L), while pH was slightly alkaline at all sites (7.30–8.01), with the lowest value at Al Qatif. Specific gravity was similar among sites, though slightly higher at Safwa. Table 2Physico chemical properties of surface water at four A. marina sites along the Arabian Gulf coastParameter (units)P1P2P3P4Physical characteristics Conductivity (dS/m)60.862.49.1460.10 TDS (mg/L)38,30439,3125,76037,863 Salinity (ppt)40.541.75.140.0 Dissolved Oxygen, DO (mg/L)5.15.76.16.6 pH8.017.807.307.84 Specific gravity (g/mL)0.983970.987500.986291.01626Chemical characteristics (mg/L) Fluoride F0.6820.901.501.00 Chloride Cl26,455.927,198.72,396.927,893.0 Phosphate PO49,822.713,458.81,180.212,274.8 Sulfate SO4NDND1,557.9ND Sodium Na18,857.916,878.41,735.717,879.6 Potassium K659.51639.6089.67669.50 Magnesium Mg22,146.71,843.1277.202,136.8 Calcium Ca21,572.31,241.8932.801,606.8 Total alkalinity (as CaCO3 mg/L)130126190140 Carbonate CO3400016 Bicarbonate HCO390126190134EC electrical conductivity (dS/m), TDS total dissolved solids (mg/L), DO dissolved oxygen (mg/L), TA total alkalinity (mg/L as CaCO₃). Salinity is expressed in practical salinity units (ppt). All major cations and anions are in mg/L unless otherwise indicated. ND = not detected (below analytical detection limit of 0.01 mg/L for fluoride, nitrate, nitrite, bromide, lithium, ammonium, and ammonia; 0.1 mg/L for phosphate and sulfate). Values represent the mean of three technical replicate measurements per site on a single composite water sample. Because only one composite water sample represents each site, these values provide descriptive, site level characterizations and are not used for formal statistical comparisons among sites Physico chemical properties of surface water at four A. marina sites along the Arabian Gulf coast EC electrical conductivity (dS/m), TDS total dissolved solids (mg/L), DO dissolved oxygen (mg/L), TA total alkalinity (mg/L as CaCO₃). Salinity is expressed in practical salinity units (ppt). All major cations and anions are in mg/L unless otherwise indicated. ND = not detected (below analytical detection limit of 0.01 mg/L for fluoride, nitrate, nitrite, bromide, lithium, ammonium, and ammonia; 0.1 mg/L for phosphate and sulfate). Values represent the mean of three technical replicate measurements per site on a single composite water sample. Because only one composite water sample represents each site, these values provide descriptive, site level characterizations and are not used for formal statistical comparisons among sites Among major ions, chloride, sodium, magnesium, calcium, and potassium were consistently much higher at Tarout, Saihat, and Safwa than at Al Qatif, mirroring the salinity gradient (e.g. Cl–: 26,455.9–27,893.0 mg/L vs. 2,396.9 mg/L at Al Qatif; Na+: 16,878.4–18,857.9 mg/L vs. 1,735.7 mg/L at Al Qatif). Phosphate concentrations were also substantially elevated at the more saline sites (9,822.7–13,458.8 mg/L at Tarout, Saihat, and Safwa vs. 1,180.2 mg/L at Al Qatif), whereas sulfate was only detected at Al Qatif (1,557.9 mg/L). Fluoride ranged from 0.682 to 1.50 mg/L, with the highest value at Al Qatif. Nitrate, nitrite, bromide, lithium, ammonium, and ammonia were not detected at any site. Total alkalinity was moderate (126–140 mg/L as CaCO₃) at Tarout, Saihat, and Safwa and higher at Al Qatif (190 mg/L), with carbonate present only at Tarout and Safwa and bicarbonate accounting for most of the alkalinity, especially at Al Qatif. Overall, these data indicate a strong contrast between the hypersaline marine conditions at Tarout, Saihat, and Safwa and the more diluted, higher-alkalinity waters at Al Qatif. Soil analysis Soil chemistry values represent a single composite pore soil sample per site measured in triplicate; we therefore describe site level patterns descriptively without applying inferential statistics. Pore soil chemistry varied markedly among sites (Table 3). Soils were mildly alkaline (pH 7.2–8.0). Electrical conductivity was substantially higher at Al Qatif and Safwa (84.30–86.90 dS/m) than at Tarout and Saihat (14.22–18.60 dS/m), while reported salinity (30–38 ppt) was broadly similar. Major anions and cations were enriched at Al Qatif and Safwa: chloride (739.5–814.5 mg/L) and sodium (1,000 mg/L) were an order of magnitude higher than at Tarout and Saihat, magnesium and calcium were also elevated at Al Qatif and Safwa (Mg 91.2–196.8 mg/L; Ca 27.0–74.0 mg/L), and sulfate was notably high at Al Qatif and Safwa (438.4–440.3 mg/L) compared with low values at Tarout and Saihat. Bicarbonate concentrations were relatively low and similar across sites (3.30–5.19 mg/L). Overall, Al Qatif and Safwa exhibit much higher salinity/electrolyte content and ion loading, indicating stronger marine influence or salt accumulation relative to Tarout and Saihat. Table 3Pore soil (0–20 cm) physico chemical properties at four A. marina sites along the Arabian Gulf coastParameterP1P2P3P4pH7.47.67.28.0EC (dS/m)18.6014.2284.3086.90Salinity, SP (ppt)38313030Anions (mg/L) HCO3 −4.253.305.195.19 Cl −145.16147.58739.52814.52 SO4 −55.294.51438.41440.34Cations (mg/L) Ca2 +11.338.6774.0027.00 Mg2 +30.2425.0491.17196.82 Na +16012010001000 K +3.121.6917.9535.89EC electrical conductivity (dS/m). Soil ion concentrations are reported as mg/L in saturated paste extracts unless otherwise indicated. Salinity is expressed in ppt. ND = not detected (below analytical detection limit of 0.1 mg/L for anions and cations). Values represent the mean of three technical replicate measurements per site on a single composite soil sample. Because only one composite soil sample represents each site, these values provide descriptive, site level characterizations and are not used for formal statistical comparisons among sites. Carbonate (CO₃²⁻) concentrations were 0.0 mg/L at all sites and are therefore not shown in the table Pore soil (0–20 cm) physico chemical properties at four A. marina sites along the Arabian Gulf coast EC electrical conductivity (dS/m). Soil ion concentrations are reported as mg/L in saturated paste extracts unless otherwise indicated. Salinity is expressed in ppt. ND = not detected (below analytical detection limit of 0.1 mg/L for anions and cations). Values represent the mean of three technical replicate measurements per site on a single composite soil sample. Because only one composite soil sample represents each site, these values provide descriptive, site level characterizations and are not used for formal statistical comparisons among sites. Carbonate (CO₃²⁻) concentrations were 0.0 mg/L at all sites and are therefore not shown in the table GC–MS/MS analysis GC–MS profiling of one pooled leaf extract per population (P1–P4) revealed 27 compounds with notable descriptive variation in relative abundance among the four A. marina samples (Table 4). The flavanone 7,3’,4’,5’-tetramethoxyflavanone was the dominant constituent in all samples (Tarout 32.25%; Saihat 25.58%; Al Qatif 25.14%; Safwa 15.06%). Other abundant constituents included quercetin 3,5,7,3’,4’-pentamethyl ether (11.03% in Tarout; 6.48% Saihat; 8.28% Al Qatif; 3.90% Safwa), flavoayamenin (8.77% Tarout; Saihat 4.26%; 6.65% Al Qatif; 16.26% Safwa) and hexa-hydro-farnesol (7.63% Tarout; absent in Saihat and Al Qatif; 6.34% Safwa). Table 4GC–MS/MS identified compounds and their relative quantities (%) in leaf extracts from four A. marina populations (P1–P4)No.RTCompoundsQuantityP1P2P3P4113.0266,7,3’,4’-Tetramethoxyflavone1.320.811.550.46213.5643,2’,4’,5’-Tetramethoxyflavone1.050.511.060.46314.2043,2’,4’,5’-Tetramethoxyflavone0.420.380.510.18414.3385,7-Dimethoxycoumarin0.470.450.690.27514.477Cedrol0.50.460.340.16614.621Isolongifolol0.620.490.890.38715.001β Carotene0.550.370.420.23815.377Cubebol1.042.96.240.76915.711Corymbolone0.54.563.690.271015.8782-Hexadecanol0002.041116.23Phytol2.075.273.281.841216.5645β,7βH,10α-Eudesm-11-en-1α-ol1.284.252.942.041317.148α-Tocopherol0004.921417.7Quercetin 3,5,7,3’,4’-pentamethyl ether11.036.488.283.91518.34Gardenin5.66.0206.191618.67Vitexin0.854.522.460.481719.0927,3’,4’,5’-Tetramethoxyflavanone32.2525.5825.1415.061819.694Gentisic acid1.9007.551920.13Luteolin 6-C-glucoside4.3612.46.661.742020.905Tricosanoic acid4.350002121.4942’,4’-Dimethoxy-3-hydroxy-6-methylflavone4.535.693.649.922222.7513-Hydroxy-7,8,2’-trimethoxyflavone3.767.624.744.022323.103Hexa-hydro-farnesol7.63006.342423.442Gallic acid00011.542523.913’,4’,5’,5,6,7-Hexamethoxyflavone5.143.479.120.862624.493,7,8,2’-Tetramethoxyflavone03.529.352.142724.912Flavoayamenin8.774.266.6516.26Values are mean relative abundances (% of total ion current) per population, based on two replicate injections of one pooled leaf extract per site (biological composite). Because only one composite extract was prepared per site, these data are descriptive and are not subjected to formal statistical testing among sites. Compounds that showed 0.0% relative abundance in all four populations (P1–P4) would be omitted from the table for brevity; in the present dataset no such compounds were detected GC–MS/MS identified compounds and their relative quantities (%) in leaf extracts from four A. marina populations (P1–P4) Values are mean relative abundances (% of total ion current) per population, based on two replicate injections of one pooled leaf extract per site (biological composite). Because only one composite extract was prepared per site, these data are descriptive and are not subjected to formal statistical testing among sites. Compounds that showed 0.0% relative abundance in all four populations (P1–P4) would be omitted from the table for brevity; in the present dataset no such compounds were detected Site-specific features included elevated cubebol (6.24%) and high levels of several methoxyflavones (e.g. 3,7,8,2’-tetramethoxyflone 9.35%) at Al Qatif; relatively high corymbolone (4.56%) and phytol (5.27%) at Saihat; and unique occurrence of α-tocopherol (4.92%), gallic acid (11.54%), and 2-hexadecanol (2.04%) at Safwa. Several compounds (e.g. gentisic acid, tricosanoic acid) occurred predominantly or exclusively in Tarout or Safwa. Overall, the profiles are flavonoid rich with site-specific terpenoid and phenolic variation, indicating notable chemical diversity among A. marina populations. SCoT marker analysis and genetic structure Across the 12 individuals (3 per population), the ten SCoT primers used for A. marina generated a total of 67 bands, of which 39 were polymorphic, with total bands per primer ranging from 4 to 12 and polymorphic bands from 2 to 11 (Fig. S1; Table 5). The percentage of polymorphism (%P) varied widely among primers (25–100%), and the number of monomorphic bands ranged from 0 to 6. Polymorphism information content (PIC) showed low-to-moderate informativeness (0.10–0.42). Genetic diversity parameters likewise exhibited moderate variation across primers: Nei’s gene diversity (H) ranged from 0.10 to 0.42, Shannon’s information index (I) from 0.14 to 0.61, and the effective number of alleles (Ne) from 1.15 to 1.73, while the observed number of alleles (Na) varied from 1.25 to 2.00. Expected heterozygosity (He) followed a similar pattern (0.10–0.42). Among the primers, SCoT-6 produced the highest number of total bands (12) with high polymorphism (92%), whereas SCoT-10 exhibited 100% polymorphism and the highest He (0.42), indicating strong discriminatory power. SCoT-7 showed the lowest diversity indices (25% polymorphism and He = 0.10). Overall, SCoT markers were effective in detecting moderate genetic diversity in A. marina, with SCoT-6 and SCoT-10 particularly informative for population genetic analyses. Table 5Primer wise genetic diversity parameters of A. marina for four populations based on SCoT markersPrimerTBPB%PMBPICIHNeNaHeSCoT‑1548010.360.530.381.721.800.38SCoT-2524030.180.250.181.321.400.18SCoT‑3548010.330.480.331.561.800.33SCoT‑4524030.150.230.151.241.400.15SCoT-5658310.310.470.311.51.830.31SCoT-612119210.350.530.351.581.9170.35SCoT-7822560.10.140.11.151.250.1SCoT‑8437510.280.420.281.451.750.28SCoT‑9658310.380.530.381.71.830.38SCoT‑106610000.420.610.421.732.000.42TB total bands, PB polymorphic bands, P% percentage of polymorphic bands, MB monomorphic bands, PIC polymorphic information content, H Nei’s gene diversity, I Shannon’s information index, Ne effective number of alleles, Na observed number of alleles, He expected heterozygosity. All indices were calculated from the binary SCoT presence/absence matrix (12 individuals; 3 per population) using GenAlEx. Given the small sample size per population and the limited number of dominant loci, these diversity estimates should be regarded as preliminary and exploratory Primer wise genetic diversity parameters of A. marina for four populations based on SCoT markers TB total bands, PB polymorphic bands, P% percentage of polymorphic bands, MB monomorphic bands, PIC polymorphic information content, H Nei’s gene diversity, I Shannon’s information index, Ne effective number of alleles, Na observed number of alleles, He expected heterozygosity. All indices were calculated from the binary SCoT presence/absence matrix (12 individuals; 3 per population) using GenAlEx. Given the small sample size per population and the limited number of dominant loci, these diversity estimates should be regarded as preliminary and exploratory Multivariate analyses consistently revealed clear genetic structuring among A. marina populations from Tarout, Saihat, Al Qatif, and Safwa (Fig. 3; Table 6). The UPGMA dendrogram based on SCoT data separated the populations into distinct clusters, with Tarout (P1) and Saihat (P2) forming one cluster, Safwa (P4) grouping as a closely related but separate branch, and Al Qatif (P3) appearing as the most distinct population. Across all pairwise comparisons, Dice similarity coefficients ranged from 0.61 to 0.69, indicating moderate differentiation among populations. PCA of the SCoT data yielded PC1 and PC2 explaining 37.1% and 34.9% of the total variance, respectively, clearly separating the four populations into discrete groups. The heatmap of genetic diversity parameters across primers and populations showed contrasting patterns of allele frequencies and diversity indices, with some populations consistently higher or lower for particular primers, as reflected by distinct red–blue gradients. Together, the clustering, PCA, and heatmap analyses corroborate the presence of noticeable genetic differentiation among the four sites. Genetic relationships among four A. marina populations based on SCoT marker data. A UPGMA dendrogram constructed from Dice’s similarity coefficients (Dice similarity converted to distance as 1 – Dice similarity) using PAST v3.26, showing Tarout (P1) and Saihat (P2) clustering together, Safwa (P4) as a closely related branch, and Al Qatif (P3) as the most distinct population. B Principal component analysis (PCA) plot based on the SCoT presence/absence matrix for 12 individuals (3 per population), showing separation of populations along PC1 (37.1% of variance) and PC2 (34.9% of variance). C Heatmap of SCoT band presence/absence illustrating contrasting patterns of allele occurrence across populations. Note that agarose gel images of SCoT PCR products are provided as Supplementary Fig. S1 Table 6Nei’s genetic similarity matrix among four A. marina populations (P1–P4) based on SCoT markersP1P2P3P4P110.610.670.69P20.6110.680.67P30.670.6810.69P40.690.670.691Values are Nei’s genetic similarity coefficients computed from the binary SCoT presence/absence data for 12 individuals (3 per population). Given the limited sample size and marker set, these values provide a preliminary indication of genetic relatedness among populations Nei’s genetic similarity matrix among four A. marina populations (P1–P4) based on SCoT markers Values are Nei’s genetic similarity coefficients computed from the binary SCoT presence/absence data for 12 individuals (3 per population). Given the limited sample size and marker set, these values provide a preliminary indication of genetic relatedness among populations Given the very small sample size per population (n = 3) and the limited number of dominant SCoT loci, these diversity estimates and clustering patterns should be regarded as preliminary and exploratory, rather than definitive measures of population genetic structure. rbcL barcoding and phylogenetic analysis Analysis of rbcL barcoding sequences for the four A. marina populations (P1–P4: Tarout PZ364408, Saihat PZ350453, Al Qatif PZ353174, Safwa PZ358527) revealed very limited but detectable chloroplast variation that provided a coarse indication of relationships among sites (Fig. 4). The maximum-likelihood tree showed short branch lengths overall, consistent with a conserved rbcL region, yet separated the samples into three distinct groups (G1–G3). Tarout (P1) and Safwa (P4) formed a single clade (G1) with nearly identical branch lengths, whereas Saihat (P2; G2) and Al Qatif (P3; G3) each formed separate lineages, indicating subtle haplotypic divergence superimposed on a shared chloroplast background. These results indicate that, despite the conserved nature of rbcL, there is low but detectable chloroplast divergence among the four A. marina populations. Maximum likelihood phylogenetic tree of A. marina populations from Tarout (P1, PZ364408), Saihat (P2, PZ350453), Al Qatif (P3, PZ353174), and Safwa (P4, PZ358527) based on rbcL barcoding sequences. Branch lengths are short overall, reflecting the conserved nature of rbcL, but the tree resolves three clades: Tarout (P1) and Safwa (P4) form a single clade (G1), while Saihat (P2; G2) and Al Qatif (P3; G3) each form separate lineages. Bootstrap values (where ≥ 50%) are indicated at nodes. This tree provides a coarse chloroplast context and is not intended for fine-scale population genetic inference Discussion Environmental heterogeneity among Arabian Gulf mangrove stands The four A. marina stands along the Arabian Gulf coast near Dammam— Tarout (P1), Saihat (P2), Al Qatif (P3), and Safwa (P4)—showed marked small-scale variation in structure and habitat conditions. Structural differences ranged from dense, well-developed canopies with abundant pneumatophores at Tarout and Safwa to a more open, sparsely vegetated stand at Al Qatif. Such contrasts are consistent with the strong influence of tidal regime, inundation frequency, and sediment texture on A. marina growth and rooting patterns [36, 37]. The tall trees and extensive pneumatophore fields at Safwa indicate a relatively stable, mature stand, whereas the open canopy and exposed sediment at Al Qatif suggest greater physical exposure, reduced sediment accretion, or recent disturbance [38, 39]. Water chemistry clearly separated hypersaline marine sites— Tarout, Saihat, and Safwa—from the more dilute, higher-alkalinity waters at Al Qatif. Elevated conductivity, TDS, salinity, and major ions at Tarout, Saihat, and Safwa typify the extreme marine–hypersaline conditions of the Arabian Gulf, driven by high evaporation, shallow depth, and restricted circulation [40]. By contrast, Al Qatif exhibited much lower water salinity but higher alkalinity, consistent with localized freshwater or brackish inputs from runoff or groundwater in an otherwise saline setting [41]. Pore-soil chemistry revealed another dimension of heterogeneity. Al Qatif and Safwa had the highest soil electrical conductivity and ion loadings, including chloride, sodium, magnesium, calcium, and sulfate, indicating strong evaporative concentration and salt accumulation in the root zone. This decoupling between water-column and pore-water salinity at Al Qatif mirrors patterns reported for arid-zone mangroves where episodic surface dilution does not immediately relieve subsurface salt stress [42]. Such conditions strongly influence root aeration, redox status, and sulfide dynamics, thereby affecting mangrove performance and zonation [42]. Overall, the combined water and soil data indicate that local hydrology and salt accumulation create sharp environmental gradients among Tarout, Saihat, Al Qatif, and Safwa, likely contributing to the observed structural and physiological differences. Our environmental and GC–MS datasets are based on one composite sample per site with technical replication only, and thus provide descriptive, site level characterizations rather than statistically tested differences among sites. Accordingly, the contrasts we discuss among Tarout, Saihat, Al Qatif, and Safwa should be interpreted as exploratory patterns that require confirmation with independent biological replication. Secondary metabolite profiles and potential environmental drivers GC–MS/MS profiling showed that A. marina at all four locations possesses a flavonoid-rich metabolome, dominated by methoxylated flavanones and flavanols, together with terpenoids, phenolic acids, and lipid-soluble antioxidants. The consistent prevalence of compounds such as 7,3’,4’,5’-tetramethoxyflavanone and quercetin methyl ethers agrees with earlier reports that Avicennia species are rich sources of polyphenolics with antioxidant and antimicrobial activities [37, 43, 44]. Detection of gallic and gentisic acids, as well as tocopherol and phytol derivatives, further highlights the importance of phenolic and lipid-phase antioxidant systems for coping with oxidative stress under high salinity, temperature, and irradiance typical of the Arabian Gulf [45, 46]. Quantitative differences among sites suggest that secondary metabolism is shaped by local environmental conditions. Tarout was enriched in several methoxylated flavonoids and unique acids, Saihat showed relatively high phytol and corymbolone, Al Qatif had elevated cubebol and multiple methoxyflavones, and Safwa uniquely accumulated α-tocopherol and high gallic acid. Similar site-specific shifts in flavonoid and phenolic profiles have been linked to spatial gradients in salinity, nutrient limitation, and redox regime in mangroves and other halophytes [47, 48]. Methoxylated flavonoids, in particular, enhance membrane association and antioxidant efficiency and are often upregulated under salinity and oxidative stress [43, 49]. The combination of high pore soil salinity and ion loading at Al Qatif and Safwa with their distinctive chemical signatures is consistent with the hypothesis that A. marina adjusts its secondary metabolite profile to local edaphic stress; however, we cannot exclude contributions from other unmeasured factors (e.g. microclimate, herbivory, pathogens, or ontogenetic stage). Experimental studies under controlled conditions and broader metabolomic surveys would be required to demonstrate causal links between specific environmental drivers and the observed chemotypic patterns. This metabolic plasticity is consistent with a role in osmotic adjustment, reactive oxygen species scavenging, and defense against pathogens favored in anoxic, sulfidic sediments [43, 50], although these functional interpretations would need to be confirmed by targeted physiological and bioassay experiments. Integrating these metabolomic patterns with environmental data from Tarout, Saihat, Al Qatif, and Safwa underscores the value of chemical traits as indicators of site-specific stress and potential adaptive responses in mangrove populations. Genetic diversity and population structure based on SCoT markers SCoT marker analysis revealed moderate genetic diversity and clear structuring among the four populations from Tarout, Saihat, Al Qatif, and Safwa. The proportion of polymorphic bands and diversity indices (PIC, H, I, He) are comparable to values reported for A. marina and other mangroves using similar marker systems, reflecting long life span, outcrossing, and geographically constrained propagule dispersal [9, 10, 13]. The particularly high informativeness of SCoT-6 and SCoT-10 confirms the suitability of SCoT markers for resolving intra-specific diversity in coastal trees, in agreement with studies on other woody taxa and salt-tolerant crops [9, 13]. Although SCoT markers detected moderate genetic diversity and clear clustering among the four sites, the small number of individuals genotyped per population (n = 3) and the limited set of dominant loci impose a strong limitation on the robustness of our population genetic inferences. With such sample sizes, estimates of parameters such as Nei’s gene diversity, Shannon’s index, and expected heterozygosity are subject to considerable sampling error. Accordingly, our SCoT based results should be interpreted as an exploratory indication of fine scale structuring along this coastal sector, and not as definitive evidence of pronounced population divergence. Larger sample sizes, additional populations, and higher resolution marker systems (e.g. microsatellites or genome wide SNPs) will be needed to more accurately quantify genetic diversity, connectivity, and potential barriers to gene flow in eastern Arabian Gulf A. marina. Multivariate analyses (UPGMA, PCA, heatmap) consistently separated the four locations into distinct genetic clusters with moderate to high similarity, indicating that while gene flow likely occurs along this coastal sector, local demographic processes and environmental filtering have produced detectable differentiation. Similar patterns of shallow but structured diversity have been documented for A. marina and other mangroves across the Indo-West Pacific, driven by coastline configuration, current systems, and habitat fragmentation [51, 52]. The partial correspondence between environmental contrasts—such as extreme pore soil salinity at Al Qatif and Safwa—and genetic clustering is consistent with the possibility of environment–genetic associations or incipient local adaptation, but these patterns could also arise from neutral processes (e.g. restricted dispersal, genetic drift) and demographic history. Because we did not perform common garden or reciprocal transplant experiments, and we did not use genome wide markers or formal genotype–environment association analyses, we cannot demonstrate local adaptation. Formal tests, such as common garden experiments or genome wide environmental association analyses, would be needed to disentangle selective from neutral explanations for the observed molecular variability [53, 54]. For conservation and management, this structure implies that Tarout, Saihat, Al Qatif, and Safwa each contribute non-redundant components of genetic diversity. Preserving stands under contrasting environmental conditions, including the more marginal environment at Al Qatif, is likely important for maintaining the potential for evolutionary responses under accelerating climate change, sea-level rise, and coastal development in the Arabian Gulf [52, 55, 56]. rbcL barcoding, phylogenetic signal, and data contribution The rbcL barcoding analysis, based on newly generated sequences from Tarout (PZ364408), Saihat (PZ350453), Al Qatif (PZ353174), and Safwa (PZ358527), revealed low but informative chloroplast divergence among the four Avicennia marina populations. The close clustering of Saihat and Safwa, followed by Al Qatif, and the distinct branch of Tarout reflect the conservative nature of rbcL, which typically resolves broad haplotype structure rather than fine-scale population differences [10, 13]. Nonetheless, the higher genetic distance of Tarout suggests a comparatively distinct maternal lineage or colonization history, echoing reports of geographically structured cDNA lineages in A. marina associated with oceanographic barriers and historical refugia [57, 58]. We emphasize that rbcL alone is not suitable for resolving fine scale, population level genetic diversity in A. marina, and our inferences about intraspecific structure and differentiation rely primarily on the more variable nuclear SCoT markers. Given both the conservative nature of rbcL and the small geographic extent of our sampling, the chloroplast signal we detect is necessarily weak and should be interpreted as a preliminary indication of local haplotypic patterning. These rbcL sequences constitute the first barcoding records for A. marina from Tarout, Saihat, Al Qatif, and Safwa and expand the global reference library for this species. Such additions are particularly important for under-represented regions like the Arabian Gulf, where they enhance species identification, enable more robust meta-analyses, and provide a baseline for future phylogeographic and conservation genetic work [59, 60]. When combined with nuclear markers such as SCoT and, ultimately, genome-wide SNP datasets, these barcodes will support more detailed inferences about connectivity, demographic history, and adaptive variation across environmental gradients in eastern Saudi Arabian mangroves [10, 13]. Overall, our environmental, phytochemical, and molecular datasets are correlative, and the mechanistic links we propose between site conditions, secondary metabolite profiles, and genetic structure should be regarded as working hypotheses rather than demonstrated causal relationships. Future experimental and longitudinal studies will be essential to verify the specific drivers of the observed variability. Overall, the integration of environmental, chemical, and genetic evidence indicates that A. marina along the coasts of Tarout, Saihat, Al Qatif, and Safwa forms a mosaic of environmentally differentiated and genetically structured stands with distinct secondary metabolite profiles. This multilevel heterogeneity supports managing Arabian Gulf mangroves as a network of locally differentiated units, rather than a homogeneous resource, in order to safeguard both ecological function and evolutionary resilience, while recognizing that local adaptation has not yet been formally demonstrated. From a conservation and restoration perspective, our results support treating these populations as a network of locally differentiated management units. We recommend that restoration and afforestation projects along the eastern Saudi Arabian coast prioritize the use of local propagules or nursery stock sourced from the same site or from environmentally similar, nearby stands, to preserve existing genetic and chemotype structure. Environmentally marginal but distinct sites such as Al Qatif—characterized by relatively dilute surface water but highly saline, ion-rich pore soils—should receive particular attention, as they may harbor stress-tolerant genotypes and chemotypes of high conservation value. Maintaining and, where possible, expanding such contrasting stands could enhance the adaptive capacity and resilience of Gulf mangrove forests under ongoing climatic and anthropogenic pressures. These management recommendations are necessarily preliminary and should be refined as broader-scale and higher-resolution genetic and demographic data become available. Conclusions This study demonstrates that A. marina populations along the eastern Saudi Arabian coast form a finely structured mosaic of environmental, chemical, and genetic diversity. The four sites—Tarout, Saihat, Al Qatif, and Safwa—differed markedly in stand structure, water and pore-soil chemistry, and ion loading, with Al Qatif standing out as a more dilute, high-alkalinity water environment but strongly saline, ion-rich sediment, and Safwa representing the most structurally developed stand. GC–MS/MS profiling revealed a shared, flavonoid-rich baseline dominated by 7,3’,4’,5’-tetramethoxyflavanone, overlaid by clear site-specific signatures in phenolic acids, terpenoids, and tocopherol, indicating environmentally modulated secondary metabolite patterns. Molecular analyses corroborated this heterogeneity. SCoT markers detected moderate genetic diversity and clear nuclear differentiation among populations, with several primers (notably SCoT-6 and SCoT-10) providing high discriminatory power. rbcL barcoding, despite its conserved nature, revealed low but informative chloroplast divergence and resolved distinct haplotypic groupings. Together, these results show that A. marina along the Arabian Gulf coast consists of locally differentiated, genetically structured stands with distinct chemotypes. However, because our study spans a relatively short stretch of coastline and is based on a small number of individuals and loci, the observed molecular differences represent fine scale, local structuring and should be confirmed and placed in a broader context by larger, region wide surveys using higher-resolution markers. For conservation and restoration, these populations may be treated, at least provisionally, as a network of locally differentiated management units, to maintain both current ecosystems functioning and the potential for evolutionary responses of Gulf mangrove forests under ongoing environmental change. Supplementary Information Supplementary Material 1.