The study was conducted on three artificial, homogeneous, and active slopes with different soil textures (sandy, clayey, and silty) located in Loja, Ecuador. Soil samples were collected from selected locations, ensuring representation of distinct textural classes. The specific sampling points, including coordinates and altitude, are detailed in Table 1.

To provide a visual representation of the sampling sites, Figure 1 presents a georeferenced map indicating the exact locations of soil collection.

The soil sampling procedure followed the NTE INEN-ISO 10381–2 standard for soil quality and sampling. In the central section of each slope, the vegetation cover was removed, and disturbed soil samples were collected from a depth of 30 cm using a shovel. Approximately 6 kg of soil was extracted from each site. Additionally, undisturbed soil samples were taken in accordance with ASTM D4220. All samples were placed in labeled plastic bags and transported to the Soil Chemistry Laboratory at Universidad Técnica Particular de Loja for further analysis.

The bacterium used in this study, Sporosarcina pasteurii, was isolated from samples collected at a wastewater treatment plant in Manta, Ecuador. A total of 2 kg of sludge was obtained from both the influent and effluent sections of the plant. The sludge samples were collected using sterile tools, placed in sterile plastic bags, properly sealed, and stored in an ice box at temperatures below 4 °C during transportation to the laboratories of the Universidad Técnica Particular de Loja.

The enrichment culture technique was used to isolate urease-producing bacteria, following the procedure described by Omoregie et al. (2016). The isolated strains were identified through molecular characterization.

Genomic DNA from the isolated bacterial strains was extracted using The Wizard^® Purification Kit (Promega, United States). Universal primers 27F and 1492R (Baker et al., 2003) were employed to amplify the 16S rRNA region, and the resulting amplicons were sent to Macrogen (Seoul, Republic of Korea) for Sanger sequencing. Raw sequencing data were manually curated to remove adapters and low-quality nucleotides. The processed sequences were analyzed using the EzTaxon-e server (https://www.ezbiocloud.net), and closely related sequences (96% threshold) were downloaded (Chalita et al., 2024). Sequence alignments were performed using MAFFT v.7.397 (Katoh et al., 2002), and the best-scoring maximum likelihood (ML) phylogenetic tree was constructed using IQ-tree 2.2.0-beta with the -m MFP option. Phylogenetic relationships were inferred, and topological robustness was evaluated with 10,000 ultrafast bootstrap replicates (-bb 10,000). The resulting phylogenetic trees were visualized with ITOL v.6 (https://itol.embl.de/).

Additionally, phylogenetic distances were calculated using the p-distance model. Ambiguous positions were removed using the pairwise deletion method, resulting in a final dataset comprising 1483 positions. All evolutionary analyses were performed using MEGA11.

Disturbed soil samples were air-dried at room temperature (25 °C) for 48 h, then sieved and homogenized using a No. 10 sieve (AS 200, RETSCH, Germany) to achieve a particle size smaller than 2 mm, based on the soil’s textural class, targeting the fine fraction.

Soil pH was measured following ASTM D4972 with a 1:2 mass-to-volume ratio using a Seven Easy pH meter (Mettler Toledo, Switzerland). Carbonate content was determined by titration (GLOSOLAN-SOP-05). Soil texture was analyzed via the Bouyoucos method using a vertical agitator (H-4260A, Hamilton Beach, United States) (Bouyoucos, 1936). Exchangeable bases ( Na + , K + , Ca 2 + , and Mg 2 + ) were measured using the US EPA 3052 method, employing a flame atomic absorption spectrometer (AA400, Perkin Elmer, United States). Cation exchange capacity (CEC) was measured using the ammonium acetate saturation method (Chapman, 1965).

For undisturbed samples, moisture content was determined according to AASHTO T 265. The Atterberg limits were determined in accordance with AASHTO T089-96 for the liquid limit (LL) and AASHTO T090-96 for the plastic limit (PL). The plasticity index (PI) was then calculated as the difference between the liquid limit and the plastic limit, as follows: P I = L L − P L

Particle size distribution was analyzed according to the Unified Soil Classification System (USCS) following ASTM D422. Triaxial compression tests, performed per AASHTO T234, determined soil cohesion and internal friction angle using remolded specimens (5 cm diameter, 10 cm height) in a triaxial apparatus (Wykeham Farrance, TRITECH, UK).

To evaluate the effect of microbially induced calcite precipitation (MICP), the Sporosarcina pasteurii strain isolated in this study was used. The bacteria were cultured in 500 mL Erlenmeyer flasks containing 100 mL of a liquid medium composed of Nutrient Broth (M002, HiMedia Laboratories Pvt. Ltd., India) and 2% urea, as commonly used for the cultivation of Sporosarcina pasteurii in MICP applications (Gat et al., 2014). The flasks were incubated overnight in an orbital shaker (Qmax 3000, Thermo Scientific, United States) at 30 °C and 180 rpm to obtain sufficient biomass for inoculation.

For carbonate precipitation tests, a solution containing 1 N calcium chloride (CaCl₂) as the calcium source was prepared, following the procedure described by Cheng et al. (2014). Remolded soil specimens were prepared using PVC molds (5 cm diameter, 10 cm height). These specimens were inoculated twice with the isolated Sporosarcina pasteurii strain, at 0 and 24 h, respectively.

The volumes of inoculum and cementing solution applied to each soil specimen varied according to the soil’s textural class, Atterberg limits, and initial moisture content under normal conditions (Table 2). The initial cell concentration was adjusted to 2 × 10 ⁸ cells/mL, as verified by Neubauer chamber counting (0.02 mm depth), ensuring a consistent microbial load across treatments.

To ensure comparability among soils with different textures, a single bacterial concentration was used in all experiments. This approach allows the influence of soil texture and pore structure on MICP-induced carbonate precipitation to be evaluated independently, while avoiding confounding effects associated with variable microbial inoculation levels (Cheng et al., 2013).

All experiments were conducted using independent soil specimens treated under identical conditions. The bacterial inoculum was prepared from a single microbial culture and used for all treatments; therefore, the replicates represent independent experimental (technical) replicates rather than independent biological replicates. This design was selected to reduce biological variability and to isolate the effects of soil texture and treatment conditions on MICP performance.

The specimens were left to react for 14 days, after which the following parameters were quantified: cation exchange capacity (CEC), carbonates, exchangeable bases ( Na + , K + , Ca 2 + , and Mg 2 + ), pH, electrical conductivity, texture, Atterberg limits, and triaxial compression tests.

The curing period was set to 14 days, as previous studies have demonstrated that MICP-treated soils typically reach their maximum mechanical strength and stable carbonate cementation within this timeframe. Earlier curing stages show limited strength development due to incomplete calcium carbonate precipitation, whereas strength gains beyond 14 days tend to plateau as microbial activity decreases and cementation stabilizes (Omoregie et al., 2024; Wu et al., 2024; Payan et al., 2024).

To verify calcium carbonate precipitation and assess bacterial cementation, a soil sample from the MICP-treated group was analyzed using scanning electron microscopy (SEM). The sample was mounted on an aluminum stub with carbon tape, sputter-coated with gold, and examined using a Tescan Mira 3 SEM (Tescan, Czech Republic) equipped with an energy-dispersive X-ray spectroscopy (EDS) detector.

Secondary electron images were captured at an accelerating voltage of 5 kV to visualize the morphological characteristics of the calcium carbonate deposits and bacterial-induced cementation. EDS analysis was performed with a Bruker X-Flash 6–30 detector (Bruker, Germany) operating at 25 kV to determine the elemental composition of the precipitates.

All experiments in this study were conducted in triplicate, with duplicate analytical measurements (n = 6). Data were analyzed before and after treatment, and results are expressed as the mean ± standard error of the mean (SEM). Statistical analysis was performed using a two-way ANOVA with a significance level set at p ≤ 0.05 , conducted with Prism GraphPad Version 9.5.1.

The EzTaxon-e server was used to identify the bacterial isolates by searching against quality-controlled databases of 16S rRNA sequences. The top hit strain for both A1 and E4 sequences (Accession numbers PV173707 and PV173706, respectively) was Sporosarcina pasteurii (strain NCIMB 8841) with 100% and 99.57% identity, respectively.

To construct a 16S rRNA phylogeny, 30 bacterial sequences sharing at least 96% identity with A1 and E4 isolates were downloaded. The resulting maximum likelihood (ML) tree (Figure 2) shows that A1 and E4 form a distinct clade within a cluster containing Sporosarcina pasteurii NCIMB 8841T. This phylogenetic placement, supported by the bootstrap values shown in the same figure, confirms the assignment of A1 and E4 to the genus Sporosarcina. A similar phylogenetic arrangement has been previously reported by Sun et al. (2017). This finding is further supported by genetic distances calculated (Supplementary Figure S1), where a low distance was observed between S. pasteurii and A1 and E4 isolates.

Soil pH plays a crucial role in microbially induced calcite precipitation (MICP), influencing carbonate formation and overall soil stability. As shown in Figure 3A, pH increased significantly in sandy and clayey soils after biostabilization. Specifically, sandy soils exhibited a mean pH increase of approximately 1 unit (from 5.43 ± 0,116 to 6.15 ± 0.045), while clayey soils showed a more pronounced increase of nearly 2 units (from 6.01 ± 0.027 to 7.92 ± 0.069). In contrast, silty soils exhibited a pH decrease of approximately 2 units (from 7.54 ± 0,263 to 5.12 ± 0.048), suggesting a different biochemical response to the treatment. All differences are significant with p < 0.05 .

Figure 3B presents the carbonate content ( CaCO 3 ) before and after treatment. The results indicate a significant increase in carbonate content across all soil types ( p < 0.05 ) , though the magnitude of this increase varied. Clayey soils displayed the highest accumulation of carbonates, followed by sandy and silty soils.

Scanning electron microscopy (SEM) analysis revealed distinct morphological variations in calcium carbonate precipitates formed during microbially induced carbonate precipitation (MICP). Well-defined rhombohedral calcite crystals were the predominant phase, densely packed within the soil matrix, acting as cementing agents that enhanced mechanical stability by binding soil particles together. Additionally, spherical vaterite precipitates of varying sizes were observed, often forming aggregates. This suggests active microbial precipitation with a potential transformation into calcite over time. The coexistence of these distinct morphologies indicates a heterogeneous biomineralization process, influenced by localized microenvironmental conditions that govern the formation of calcium carbonate polymorphs (Figure 4).

In this study, the cation exchange capacity (CEC) of all three soil types increased after the biostabilization treatment (Figure 5), indicating modifications in their physicochemical properties. These increases were significant with p < 0.05 .

Clayey soil exhibited the most significant increase in CEC, rising from 0.23 ± 0.022 meq/100 g before treatment to 19.54 ± 0.467 meq/100 g post-treatment. Similarly, silty soil exhibited an increase in CEC from 2.14 ± 0.053 meq/100 g to 13.54 ± 0.757 meq/100 g. In sandy soil, a more moderate increase in CEC was measured, rising from 3.44 ± 0.136 meq/100 g to 5.89 ± 0.177 meq/100 g.

Regarding exchangeable bases, calcium ( Ca 2 + ) concentrations increased significantly ( p < 0.05 ) in all soil samples, with initial values ranging from 3116.64 ± 172.31 mg/kg to 6883.89 ± 306.60 mg/kg, reaching up to 62438.26 ± 2080.19 mg/kg after treatment (Figure 6A). Potassium ( K + ) also showed a substantial increase (Figure 6C), with final concentrations exceeding 14000 mg/kg, particularly in coarser-textured soils.

Sodium ( Na + ) concentrations decreased significantly ( p < 0.05 ) in sandy and clayey soils (Figure 6B), with final values below 700 mg/kg. Conversely, a slight increase was recorded in silty soils, reaching a final value of 1537.13 ± 58.8 mg/kg. Magnesium ( Mg 2 + ) also exhibited a general decrease across most soils (Figure 6D), with reductions exceeding 50% compared to initial values. However, an increase was observed in silty soils, where concentrations reached 1820.39 ± 71.74 mg/kg. All differences are significant with p < 0.05 .

Soil texture was assessed before and after biostabilization. The results (Table 3) show notable modifications in the textural classification of all three soil types following treatment.

Initially, the soils were classified as sandy clay loam, clay loam, and silty loam. After treatment, textural shifts were observed, with the sandy clay loam transitioning to sandy loam, the clay loam to loamy sand, and the silty loam to sandy loam. These changes are associated with variations in the proportions of sand, silt, and clay.

The clay-sized fraction showed an apparent reduction to below the detection limit of the hydrometer analysis in all soil types after MICP treatment, while the sand fraction increased consistently across all samples. In sandy soils, the silt content increased from 28.8% ± 1.44%–40.75% ± 0.04, contributing to an apparent shift toward a coarser textural classification.

The apparent reduction of the clay-sized fraction after MICP treatment is attributed to particle aggregation and calcium carbonate cementation, which promote the incorporation of fine particles into larger, hydraulically coarser aggregates. These aggregation effects influence sedimentation-based particle size measurements and do not imply clay removal or mineralogical transformation.

The most pronounced transformation occurred in clayey soils, where the clay fraction dropped from 28.9% ± 1.11% to 0%, and the sand fraction increased from 41% ± 0.83%–72.48% ± 0.81. This indicates a significant alteration in particle distribution, resulting in a transition to loamy sand. Similarly, silty soils experienced a marked shift, with the clay fraction reduced from 15.4% ± 1.10% to 0%, and the sand fraction increasing from 33.4% ± 1.03%–60.00% ± 0.01. This resulted in a reclassification from silty loam to sandy loam.

The Atterberg limits, including the liquid limit (LL) and plastic limit (PL), were measured before and after the biostabilization treatment for sandy, clayey, and silty soils. The Plasticity Index (PI) was then calculated as the difference between LL and PL. The results, presented in Figure 7, show the changes in LL (A), PL (B), and PI (C) across all soil types. These results indicate that the treatment led to an increase in both LL and PL, with varying degrees of change depending on the initial soil composition, while the PI showed a marked reduction for all soil types after treatment. All differences are significant with p < 0.05 .

The LL increased across all soil types after treatment. Sandy soil exhibited a moderate increase of 11.5% (from 26.12 ± 0.59 to 29.12 ± 1.08), while clayey soil showed a more pronounced rise of 16.2% (from 26.00 ± 0.08 to 30.20 ± 0.37). In silty soil, LL increased by 5.5% (from 30.33 ± 0.1 to 32.00 ± 0.7), indicating a less substantial yet notable change. Similarly, the PL increased consistently across all soil types, with the most significant change observed in clayey soil, where it rose by 71.8% (from 15.43 to 26.50). In sandy and silty soils, PL increased by 16.6% (from 21.44 to 25.00) and 2.5% (from 27.80 to 28.50), respectively, reflecting modifications in soil plasticity due to the biostabilization treatment.

As a result of these changes, the Plasticity Index (PI), calculated as the difference between LL and PL, showed a marked reduction across all soil types. The most significant decrease was observed in clayey soil, where PI dropped by 65% (from 10.57 ± 0.42 to 3.70 ± 0.71), indicating a substantial reduction in plasticity. In silty and sandy soils, PI decreased by 57% (from 4.20 ± 1.1 to 1.80 ± 0.5) and 12% (from 4.68 ± 0.65 to 3.09 ± 0.56), respectively.

The triaxial compression test was performed to assess the stress-strain response of sandy, clayey, and silty soils before and after biostabilization. This test measured cohesion and the internal friction angle ( ϕ ) , both key parameters in determining soil strength and stability. The obtained results indicate a significant improvement in soil mechanical properties after treatment (Figure 8).

Cohesion increased across all soil types, with sandy and clayey soils increasing from 0.75 to 1.00 kg/cm² and 0.80–1.00 kg/cm², respectively. Silty soil exhibited the most pronounced change, increasing from 0.22 to 2.00 kg/cm². This corresponds to an increase in cohesive forces of 33% in sandy soil, 25% in clayey soil, and 800% in silty soil.

A similar trend was observed in the internal friction angle, which increased from 20° to 41° in sandy soil, 25°–32° in clayey soil, and 25°–34° in silty soil. These increments represent a 100% increase in sandy soil, 28% in clayey soil, and 36% in silty soil.

Large percentage increases should be interpreted in the context of the very low initial shear strength of the untreated soils; therefore, absolute post-treatment values provide a more meaningful indicator of mechanical improvement. While the laboratory results indicate a positive contribution to shear strength, field-scale slope stability depends on site-specific conditions and cannot be directly inferred from laboratory parameters alone.

The increase in pH observed in sandy and clayey soils is consistent with the expected MICP mechanism, where urea hydrolysis generates ammonia and hydroxide ions, creating alkaline conditions that favor carbonate precipitation (Safdar et al., 2021). During MICP, urea hydrolysis generates ammonia ( NH 3 ), which further hydrolyzes into ammonium and hydroxide ions (OH⁻), leading to an overall pH increase in the surrounding environment (Stocks-Fischer et al., 1999). The more pronounced pH shift in clayey soils may be attributed to their higher cation exchange capacity, which enhances ion retention and buffering effects, stabilizing alkaline conditions.

In contrast, the pH decrease observed in silty soils suggests an alternative biochemical pathway, likely driven by increased heterotrophic microbial respiration. This leads to CO 2 and organic acid accumulation, which acidifies the medium (Fang et al., 2018). This finding is consistent with studies by Ferris et al. (2004) and Omoregie et al. (2020), who reported similar acidification effects under anoxic conditions due to bacterial metabolism. The reduced pH in silty soils may indicate a lower urea hydrolysis rate, potentially necessitating bacterial augmentation to sustain MICP efficiency (Ashraf et al., 2021). Furthermore, the lower pH correlates with a reduced carbonate accumulation in silty soils, suggesting increased carbonate dissolution or shifts in microbial metabolic activity that may limit overall MICP performance.

Despite the pH decrease in silty soils, the increase in carbonate content across all soil types confirms successful biostabilization. The addition of calcium chloride ( CaCl 2 ) resulted in calcium carbonate ( CaCO 3 ) precipitation, promoting the aggregation of fine particles into coarser structures, modifying the textural class of the soils. This effect was particularly notable in clayey soils, where higher carbonate accumulation contributed to a reduction in clay particles and an increase in coarser fractions. The greater carbonate accumulation in clayey soils is likely due to their finer texture and larger surface area, which promote nucleation and retention of precipitated calcite. Conversely, the lower carbonate content in silty soils may result from enhanced carbonate dissolution under acidic conditions or reduced nucleation efficiency.

The mineralogical variations observed in the precipitates confirm that MICP promotes the formation of multiple calcium carbonate polymorphs, influenced by localized microenvironmental conditions. The predominance of rhombohedral calcite suggests that the biomineralization process favors its stabilization as the thermodynamically most stable phase. In contrast, the presence of vaterite in certain regions indicates transient mineralization stages, aligning with previous studies where vaterite often appears as an intermediate phase before transforming into calcite (Anbu et al., 2016; Tourney and Ngwenya, 2009).

The cementing role of calcite was evident in the formation of dense aggregates binding soil particles, reinforcing its contribution to soil biostabilization. Energy-dispersive X-ray spectroscopy (EDS) confirmed the successful precipitation of calcium carbonate, while also detecting silicon (Si) and aluminum (Al), likely originating from the soil matrix. This suggests that interactions between microbial carbonate precipitation and existing soil components may influence mineralization pathways, potentially affecting the stability and distribution of carbonate phases.

The cation exchange capacity (CEC) quantifies the ability of soil minerals to adsorb and retain cations, a property closely linked to specific surface area. A higher cation exchange capacity usually enhances soil stabilization by promoting particle aggregation and reducing soil plasticity (Bhurtel et al., 2024).

The increase in CEC after treatment suggests an enhancement in the soil’s ability to retain cations, likely due to the incorporation of calcium carbonate and other precipitated minerals. Clayey soils, with their inherently high specific surface area, exhibited the most significant increase in CEC, while sandy soils showed the least improvement due to their lower charge density and larger particle size. The increase in silty soils was also notable, albeit to a lesser extent, reflecting their intermediate texture and surface reactivity.

Regarding exchangeable bases, the marked increase in calcium concentration is directly linked to the addition of CaCl 2 , facilitating calcium adsorption onto soil particles. This corroborates previous studies highlighting the role of calcium in enhancing soil stability through carbonate precipitation (Fu et al., 2023).

Sodium ( Na + ) concentrations decreased in sandy and clayey soils, likely due to cation exchange reactions where calcium ( Ca 2 + ) replaced sodium ( N a + ) in soil adsorption sites, a well-documented phenomenon in soil stabilization (Ranjbar and Jalali, 2015). However, a slight increase in sodium concentration in silty soils suggests differential ion exchange dynamics influenced by initial soil composition and cation availability.

Soil texture is a key physical property that influences water retention, permeability, and mechanical stability. Textural changes occur when the relative proportions of sand, silt, and clay are altered, affecting soil behavior. Textural changes observed in all soils are attributed to particle aggregation and redistribution driven by MICP. The reduction in clay content suggests that fine particles were incorporated into larger aggregates, shifting the texture toward coarser fractions. In clayey soils, calcium carbonate accumulation likely bound clay particles, effectively reducing the fine fraction and increasing aggregate formation. This effect has been previously reported in soil stabilization studies, where an increase in calcium carbonate ( CaCO 3 ) content leads to an apparent reduction in clay and an increase in coarser particles (Virto et al., 2018).

The apparent reduction of the clay fraction observed after MICP treatment should be interpreted as an effect of particle aggregation and carbonate cementation influencing hydrometer-based measurements, rather than as a true mineralogical alteration or removal of clay minerals. Similar aggregation and contact enhancement effects have been reported in previous MICP studies. For instance, Mahawish et al. (2018) demonstrated that improved particle contacts and cementation significantly affect the mechanical behavior of MICP-treated soils, supporting the interpretation that particle aggregation, rather than mineralogical alteration, governs the observed changes.

For sandy and silty soils, the increase in the sand fraction may result from the cementation of finer particles into larger aggregates. Calcium carbonate precipitation enhances flocculation and aggregation, leading to reduced clay dispersion and an overall shift toward a coarser texture. This aligns with findings from previous studies on biostabilization, which suggest that calcite formation promotes grain binding and alters soil texture (Xie et al., 2024).

The Atterberg limits include the liquid limit (LL), which defines the transition from a semi-liquid to a plastic state, and the plastic limit (PL), which marks the shift from plastic to semi-solid. The plasticity index (PI), defined as the difference between LL and PL, reflects the soil’s plasticity and its response to moisture content variations. Evaluating these limits is essential for understanding the behavior of soils in engineering applications (Shimobe and Spagnoli, 2022).

The increase in LL and PL after biostabilization suggests structural modifications in the soil matrix, likely due to particle aggregation and calcite precipitation. The most significant shift in PL was observed in clayey soils, indicating a reduction in plasticity and a transition toward a more stable structure. The moderate increases of PL in sandy and silty soils indicate that the treatment also affected their ability to retain water at the liquid limit, potentially due to particle aggregation and calcite precipitation.

The reduction in plasticity index across all soil types reflects a transition toward more stable soil behavior under varying moisture conditions. This response is consistent with MICP-induced mechanisms in which calcium carbonate precipitation and associated cation exchange promote clay particle flocculation and pore-space modification, thereby limiting the mobility of fine particles and reducing plastic deformation. Similar reductions in soil plasticity following calcite precipitation have been reported in previous stabilization studies, where the binding and flocculation of clay minerals contribute to improved mechanical stability (Cokca, 2001).

The mechanical properties of biostabilized soils varied with texture, influencing both cohesion and internal friction angle ( ϕ ) , which exhibited different degrees of enhancement depending on soil type. Friction coefficients correlate with shear resistance, and ϕ typically ranges between 25° and 45° in most soils (Liu et al., 2011). Cohesion, influenced by molecular forces and water films binding soil particles, tends to be higher in finer soils, such as clays (Havaee et al., 2015).

The pronounced increase in cohesion observed in silty soils highlights the texture-dependent effectiveness of MICP, indicating that intermediate-textured materials provide favorable conditions for carbonate precipitation and interparticle bonding. In such soils, calcite formation can efficiently bridge particles and fill pore spaces without the diffusion limitations commonly associated with clay-rich matrices, leading to a more cohesive and mechanically stable soil framework. Similar texture-dependent responses have been reported in previous MICP studies, where silty and fine-grained soils exhibited notable gains in shear strength as a result of effective calcite bonding and pore-scale cementation (Cui et al., 2017; Park et al., 2020).

The biostabilization treatment significantly improved mechanical properties, with cohesion and internal friction angle ( ϕ ) increasing across all samples, particularly in silty soil, which exhibited the greatest enhancement. These results align with previous studies demonstrating that increased CaCO 3 content enhances soil strength and stiffness (Lin et al., 2016; Wu et al., 2021). In clayey soils, the friction angle reached 32°, and cohesion increased to 1 kg/cm², further confirming the effectiveness of the biostabilization process. This improvement underscores the critical role of CaCO 3 in enhancing soil stability, consistent with findings that even low CaCO 3 concentrations (e.g., 1%) can significantly improve soil resistance (Lin et al., 2016).

Overall, the observed improvements in cohesion and friction angle confirm the effectiveness of biostabilization in enhancing soil stability, supporting its potential application in geotechnical engineering for slope stabilization and erosion control. The mineralogical and compositional analysis further validates the underlying mechanisms driving this stability, demonstrating that microbial carbonate precipitation effectively promotes soil cementation. The combined evidence from SEM and EDS confirms the structural integrity of the precipitated calcium carbonate, reinforcing the role of Sporosarcina pasteurii in MICP-based soil stabilization.

While the laboratory-scale results demonstrate the potential of MICP for improving soil mechanical behavior, these findings should be interpreted with caution when considering field-scale slope stability. Although the observed increases in cohesion and shear resistance suggest that MICP can contribute to enhanced near-surface stability and reduced susceptibility to shallow failures, laboratory-derived parameters alone are not sufficient to directly quantify slope stability at the field scale. Field performance is governed by site-specific factors such as slope geometry, groundwater conditions, treatment depth, spatial uniformity of cementation, and the long-term durability of carbonate bonds.

In addition, the experiments were conducted under controlled conditions that may not fully capture the environmental variability of natural slopes, including fluctuations in moisture, temperature, and long-term biological activity. The use of a single bacterial concentration also does not account for potential texture-specific optimization of microbial dosage. Consequently, future research should focus on field-scale validation, long-term performance assessment, and optimization of MICP treatment parameters. Within this framework, the integrated physicochemical and geomechanical dataset presented in this study provides a soil-specific baseline to support preliminary slope-stability assessments and the design of pilot-scale applications in erosion-prone environments.