This study investigates an innovative application of microbial-induced calcium carbonate precipitation (MICP) for enhancing erosion control on unpaved roads, leveraging the unique capabilities of native microorganisms. Aiming to advance disaster-resilient infrastructure, this research employs bio-stimulating treatments to facilitate calcium carbonate (CaCO3) precipitation by way of a novel bio-weathering pathway. Unlike the conventional urea-hydrolysis method commonly used in MICP studies, which generates toxic ammonia as a byproduct, the bio-weathering approach in this study consumes CO2, offering a potentially carbon-negative solution while effectively reducing erosion. Granular material from an unpaved road embankment underwent comprehensive geotechnical characterisation. Inderbitzen tests, which simulate rainfall and runoff, were used to assess erosion resistance. Undisturbed samples were treated either with a bio-stimulating MICP solution or with distilled water as a control. Results indicate significant reductions in soil loss and enhanced stability in bio-stimulating treated samples. While the study primarily focuses on erosion control, it also acknowledges the potential for long-term CO2 sequestration as a beneficial byproduct of the CaCO3 precipitation process. The findings suggest that MICP can be effectively integrated into geotechnical practices to improve the durability and performance of unpaved roads, contributing to sustainable and resilient infrastructure development.
INTRODUCTION
As climate change intensifies, the resilience of infrastructure, particularly unpaved roads, becomes increasingly critical. In Brazil, like in many other developing countries, roads are the backbone of transportation, accounting for approximately 65% of the supply chain and 95% of passenger travel (CNT, 2022). With 86% of Brazil’s 1·7 million km road network being unpaved, erosion poses a significant challenge (CNT, 2022; DNIT, 2022).
In regions like Roraima in the Amazon, environmental protections and vast distances often restrict access to quarries for road construction materials, which drives up costs and environmental impacts (Moizinho, 2007). Consequently, there is increased reliance on local soils; however, these soils typically lack the necessary properties for construction and require stabilisation to be used effectively (Firoozi et al., 2017).
An additional issue of the sector is that both pavement and unpaved roads are susceptible to erosion of embankment margins. Erosion, influenced by environmental conditions, road usage, and geotechnical properties, compromises the structural integrity of roads and increases maintenance needs (Ngezahayo et al., 2019). Traditional soil stabilisation methods, such as the use of cement, lime, or polymers, are effective but come with high costs, energy requirements, and significant environmental pollution, particularly CO2 emissions from cement production (Consoli et al., 2001; IEA, 2018).
As public opinion and policy increasingly favour sustainable solutions, bio-geotechnics offers promising alternatives by leveraging natural processes to address geotechnical challenges in an eco-friendly manner (Anderson et al., 2021). Within this field, microbial-induced calcium carbonate precipitation (MICP) emerges as an innovative method for enhancing soil properties.
Among the MICP processes (Table 1), urea hydrolysis and denitrification are the most widely reported in the literature. Both demonstrate significant potential for improving geotechnical soil properties, such as increasing strength, mitigating liquefaction, and controlling permeability. However, they also present important environmental limitations. Urea hydrolysis produces byproducts such as ammonium (NH4+) and ammonia (NH3), which can cause negative environmental impacts. Similarly, denitrification releases nitrogen oxides (NO and N2O), which are potent greenhouse gases. Furthermore, these methodologies often involve the introduction of exogenous microorganisms into the soil, which can disrupt the local microbiota and lead to the death of the introduced bacteria, reducing the process efficiency and increasing operational costs (Erşan et al., 2015; Osinubi et al., 2020; Young et al., 2021).
Summary of MICP methods
| Metabolic pathway | Negative by products | Frequently associated microorganisms | References |
|---|---|---|---|
| Urea hydrolysis | Ammonium ion (NH4+). | Ureolytic bacteria; urease-positive microorganisms. | (Osinubi et al., 2020; Stocks-Fischer et al. 1999) |
| Photosynthesis | — | Photosynthetic microorganisms; cyanobacteria algae. | (Dittrich et al., 2003; Erşan et al., 2015; Osinubi et al., 2020) |
| Denitrification | Nitric oxide (NO); nitrous oxide (N2O). | Nitrate-reducing bacteria. | (Erşan et al., 2015; Osinubi et al., 2020) |
| Ammonification | Ammonia (NH3). | Myxobacteria. | (Krause et al., 2018; Osinubi et al., 2020) |
| Sulfate reduction | Carbon dioxide (CO2); hydrogen sulfide ion (HS−). | Sulfate-reducing bacteria. | (Osinubi et al., 2020) |
| Bio-weathering (carbonic anhydrase) | — | Bacillus. | (Bose & Satyanarayana, 2017; Mwandira et al., 2024) |
Unlike the conventional urea-hydrolysis and denitrification pathways that produces toxic byproducts, the bio-weathering approach used in this study consumes CO2, potentially providing a carbon-negative solution. Natural weathering processes, which involve the dissolution of CO2 in water to form carbonic acid (Equation 1), play a crucial role in soil stabilisation by facilitating calcium carbonate precipitation (Renforth et al., 2011; Washbourne et al., 2012).
Under natural conditions, the rate of CO2 hydration and the subsequent formation of carbonic acid (H2CO3) are extremely slow, significantly limiting the efficiency of the process (Mirjafari et al., 2007; Supuran, 2017). Since biomineralisation directly depends on the products of Equation 1, as shown in Equations 2 and 3, it requires catalytic agents to accelerate the conversion, efficiently promoting carbonate precipitation.
where M refers to cations, such as Ca2+ and Mg2+.
In this context, carbonic anhydrase (CA) emerges as a highly effective biological catalyst. This enzyme, widely present in eukaryotic and prokaryotic organisms, including soil microorganisms, has demonstrated significant potential to overcome this limitation (Bose & Satyanarayana, 2017; Hazarika & Yadav, 2023). Recent studies, such as Mwandira et al. (2024), indicate that CA-producing microorganisms isolated and applied to soil worked successfully as an alternative to conventional MICP methods.
Stoichiometric calculations indicate that the amount of CO2 sequestered is directly proportional to the availability of reactive cations in the soil. For example, Washbourne et al. (2012) reported that a site with 1·46 mol/kg of net cations over an area of 10 hectares had the potential to remove approximately 64 804 Mg of CO2 through carbonate formation. These findings suggest substantial CO2 capture potential when sufficient cations are present. However, empirical data on the efficiency and practicality of this approach in situ are limited.
This study focuses on the use of methodologies that stimulate native soil microorganisms to accelerate the CO2 hydration process, captured from the environment, leading to the production of CaCO3. This approach avoids introducing exogenous species, minimising risks of environmental imbalance, and facilitating large-scale application (Osinubi et al., 2020). Moreover, by exploring the potential of bio-stimulation, this study aims to provide a more sustainable solution tailored to Amazonian soil conditions, contributing to erosion mitigation and promoting carbon dioxide capture in an efficient and environmentally responsible manner.
Thus, this research investigates the use of bio-stimulating MICP to reduce erosion susceptibility by leveraging the unique capabilities of native microorganisms in the bio-weathering process, which also contributes to CO2 sequestration.
MATERIALS AND METHODS
Geotechnical characterisation
The study area is located on the embankment of an unpaved access road constructed with granular material in Roraima, Brazil, within the Amazon region. Undisturbed and disturbed samples were collected from the slope of this embankment.
The soil characterisation process was carried out in accordance with the Brazilian Technical Standards (NBR). According to (ABNT, 2016a), sample preparation involved air drying, disaggregation, crushing, and homogenisation of the soil. To determine the specific gravity, the procedures established by (ABNT, 1984) were followed. Duplicate tests were carried out and the average specific gravity value obtained was 2·64.
The particle size analysis was performed according to (ABNT, 2018). Particle size distribution curves combining sieving and sedimentation of duplicated samples are presented in Fig. 1, while Table 2 summarised the key information extracted from particle size distribution curves. The diameters D10, D30, and D60 represent the particle sizes smaller than 10%, 30%, and 60% of the sample, respectively. The uniformity coefficient (Cu) calculated as Cu = D60/D10, indicates the range of particle sizes, with a higher Cu suggesting a wider range. The coefficient of gradation (Cc), calculated as Cc = D302/(D10 × D60), indicates how well-graded the soil is. Table 2 also shows the relative proportions of sand, silt, and clay that influences soil behaviour. In addition, fines content, representing the percentage of particles smaller than 0·06 mm, which affects the soil’s plasticity and compaction characteristics, is also presented.
Key parameters from particle size distribution curves, where D10, D30, and D60 represent the particle sizes smaller than 10%, 30%, and 60% of the sample, respectively
| D10: mm | D30: mm | D60: mm | Cu | Cc | % Sand | % Silt | % Clay | Fines content: % |
|---|---|---|---|---|---|---|---|---|
| 0·014 | 0·18 | 0·3 | 21·71 | 7·71 | 85·5 | 5·6 | 8·9 | 14·5 |
Cu = uniformity coefficient; Cc = coefficient of gradation
The determination of the liquid limit was performed following the guidelines of (ABNT, 2016b), while the plastic limit was determined by following the (ABNT, 2016c) standard. The liquid limit of the material was 14·10% and plastic limit could not be determined suggesting non-plastic material. According to the Unified Soil Classification System, the material collected is classified as poorly graded sand.
Bio-stimulating solution
The bio-stimulating solution used in this study was the B4 enrichment medium, known for its effectiveness in promoting bioprecipitation (Valencia-Gonzáles et al., 2014; Valencia-González et al., 2015). César et al. (2024) investigated the impact of calcium source concentration and incubation time on calcium carbonate precipitation through the bio-weathering process. They concluded that a calcium source concentration of 1·5% and an incubation time of 28 days were the most effective. Consequently, the bio-stimulating solution was prepared using 4 g/L of yeast extract, 5 g/L of dextrose, and 15 g/L of calcium acetate, with an incubation period maintained at 28 days. Undisturbed soil samples were sprayed with the bio-stimulating solution only once, as detailed in the following section.
Erosion testing
Surface erosion can be associated with the detachment of soil particles caused by excess precipitation that does not infiltrate the ground and flows over the surface as a sheet of water (Camapun de Carvalho et al., 2006).
Soil is disaggregated from the mass by the flow of the water sheet due to shear stresses generated between water and soil particles. The extent of this surface soil removal depends on factors such as rainfall potential, topography, slope gradient, vegetation, degree of soil compaction, and soil erodibility. This latter is the susceptibility of soil to resist erosive processes (Vilar & Prandi, 1993). One way to measure this erodibility is through direct surface erosion tests, such as the Inderbitzen test that uses the Inderbitzen apparatus shown in Fig. 2.
The Inderbitzen test involves simulating surface laminar flow over a soil sample and measuring the rate of soil disaggregation over time (Inderbitzen, 1961). The Inderbitzen tests conducted in this study aimed to compare the effects of using bio-stimulating solution to mitigate erosive processes against control samples treated with distilled water. The tests were performed with undisturbed samples (cylindrical samples of 150 mm in diameter and 100 mm in height) collected in the slope of the test road site. Three samples were treated with bio-stimulating solution, while three other samples were treated with distilled water as control samples for comparison. All specimens were treated with a single application of 820 ml of bio-stimulating solution or distilled water using a sprayer for uniformity, achieving 85% saturation.
After application, specimens were kept in a controlled environment at 21°C for 28 days (incubation time). Environmental factors such as temperature and soil moisture content can influence the rate of calcium carbonate precipitation in bio-mediated approaches. These factors affect microbial activity and, consequently, the efficiency of the treatment. By controlling these variables, the effects of the bio-stimulating solution on soil erosion mitigation can be isolated. This approach allows the establishment of a foundational understanding, which can then be built upon in subsequent research that considers a wider range of environmental scenarios.
In addition, it is important to note that in tropical regions, soil temperatures are relatively stable throughout the year. According to Corlett (2024), almost all soils within the geographic tropics have seasonal temperature variations of less than 6°C. This suggests that the seasonal and spatial temperature fluctuations in the study area are minimal, and their impact on microbial activity is likely to be limited.
The flow rate used in all tests was 2 L/min, which is equivalent to a precipitation of 10·79 mm/hr. Eroded soil was collected at 1 min, 5 min, 10 min, and 20 min during the test. The variations of ramp inclinations and corresponding water sheet height, flow velocity, and shear stresses can be seen in Table 3.
RESULTS
Figure 3(a) shows a marked reduction in cumulative soil loss over time for samples treated with the bio-stimulating solution compared with control samples treated with distilled water. This reduction is consistent across different ramp inclinations (10°, 20°, and 30°), highlighting the robustness of the bio-stimulation treatment in various terrain conditions. The lower cumulative soil loss indicates that the bio-stimulating treatment significantly enhances the soil’s resistance to erosion, making it a viable method for stabilising unpaved road embankments.
Inderbitzen results: (a) soil loss over time, where 10°, 20°, and 30° indicate ramp inclinations, (b) erodibility curve, and (c) erodibility criterion (after Bastos, 1999)
Inderbitzen results: (a) soil loss over time, where 10°, 20°, and 30° indicate ramp inclinations, (b) erodibility curve, and (c) erodibility criterion (after Bastos, 1999)
Figure 3(b) illustrates that the rate of soil loss in bio-stimulating samples is substantially lower than in control samples. The slope of the soil loss rate for treated samples (0·0112 g/cm2/min/Pa) is significantly less than that for control samples (0·0385 g/cm2/min/Pa), indicating that bio-stimulating treatment improves soil stability. The high R2 values for both treated (0·9991) and control samples (0·9762) suggest a strong linear relationship between shear stress and soil loss rate, reinforcing the reliability of the observed trend. This improvement in soil stability is crucial for the longevity and safety of unpaved roads, reducing the frequency and cost of maintenance.
Bastos (1999) conducted both direct and indirect evaluations to classify residual tropical soils according to their susceptibility to erosion. The direct method involved the use of Inderbitzen tests, based on which the author proposed a classification criterion. Figure 3(c) places the erodibility rates of both treated and control samples within Bastos (1999)’s criterion. The erodibility rate of the treated sample (0·0112 g/cm2/min/Pa) is significantly lower than that of the control sample (0·0385 g/cm2/min/Pa), indicating a reduction of approximately 69%. Although both materials are classified as moderately erodible, the substantial reduction in the erodibility rate due to bio-stimulating treatment confirms its effectiveness in enhancing soil resistance to erosion.
DISCUSSIONS
This experimental design ensures consistency in treatment application and incubation conditions, allowing for a clear comparison between treated and control samples. The use of undisturbed soil samples preserves the natural structure and heterogeneity of the material, which is critical for accurately assessing the effects of the bio-stimulating solution on erosion mitigation. The decision to maintain the specimens at a constant temperature of 21°C during the incubation period reflects an effort to standardise microbial activity and minimise confounding effects from temperature variability. Furthermore, the controlled saturation level and single application protocol enhance the reproducibility of the results. Future studies could build upon this foundational work by investigating the effects of repeated treatments or varying saturation levels under field conditions to better simulate long-term environmental impacts.
The results from this study can be further contextualised by comparing them with findings from the literature on conventional anti-erosion methods, which can broadly be categorised into grey and green solutions. Grey solutions, often engineering-focused, include soil stabilisation through chemical additives, geosynthetics, and structural materials, while green solutions rely on vegetation and natural processes to mitigate soil erosion.
Among grey solutions, soil–cement stabilisation has shown significant effectiveness. Pheng et al. (2019) reported that soil–cement mixes at 3% and 5% by mass improved erosion resistance by 87% and 98%, respectively, in experiments simulating rainfall intensities of 50 mm/h over 4 h. Similarly, Dafalla & Obaid (2013) demonstrated that cement–clay grout and polypropylene fibre–reinforced grout achieved 47% and 37% improvements, respectively, in erosion resistance under controlled flow conditions, while polypropylene geotextiles achieved similar results. Lime stabilisation, as studied by Costa et al. (2021), proved particularly effective in reducing soil erodibility to as low as 12% of untreated values under reduced compaction and 7% under normal compaction, with performance highly dependent on curing time and soil mixing quality. Riprap, another grey solution, demonstrated improved effectiveness when combined with geotextile filters, as highlighted by Shidlovskaya & Briaud (2023). These results collectively indicate that engineered solutions can provide substantial erosion resistance, often through enhanced soil cohesion, reduced permeability, or physical barriers against runoff.
The most popular green solutions, on the other hand, rely on the natural capabilities of vegetation to reduce erosion through mechanisms such as intercepting rainfall, increasing water infiltration, stabilising soil with roots, and reducing runoff velocity (Igwe et al., 2017). Grass-covered soils, as demonstrated by Shidlovskaya & Briaud (2023), provide highly effective erosion control, with Bermuda and St. Augustine grasses reducing erosion rates by up to 92% when they are healthy, have at least 75% coverage, and have been established for over 3 years. These findings align with broader literature highlighting the benefits of vegetation. Zhou & Shangguan (2007) studied the effects of dry grass roots and shoots on loess erosion under simulated rainfall and concluded that increasing the amount of vegetation can lead to runoff generation and erosion control. Similarly, Gyssels et al. (2005) reported that the effect of plant roots on increasing the resistance of the soil to concentrated flow erosion mainly depends on the presence and distribution of effective plant roots. Vegetation not only reduces the mechanical impact of raindrops but also stabilises soil particles. Rey (2003), who studied the influence of vegetation distribution on sediment yield in forested Marly gullies in France, stated that plant cover protects soil against erosion by reducing water runoff.
Hybrid approaches that combine grey and green solutions have also demonstrated significant promise. For example, Muthukumar et al. (2022) showed that coir geotextiles and asphalt emulsion reduced erosion by 61%–68% and 54%–62%, respectively. When used in combination, these methods achieved erosion reductions of up to 75% under simulated rainfall conditions.
In comparison, the results from this study using bio-weathering MICP showed an improvement in erosion resistance of 69%, which aligns with or surpasses many conventional methods reported in the literature. While green infrastructure competes well with this approach, offering comparable or even superior erosion resistance in some cases, its effectiveness is dependent on factors such as vegetation coverage, species type, health, and maturity. Both approaches contribute to carbon sequestration, but the mechanisms differ. Vegetation temporarily stores CO2 in biomass, which can be released back into the atmosphere through decomposition or disturbances, making it less permanent. In contrast, the bio-weathering process used in bio-stimulated MICP fixes CO2 directly into the soil matrix as stable mineral forms, providing a more permanent carbon sink. This dual functionality of erosion resistance and long-term CO2 sequestration positions bio-weathering MICP as a compelling alternative that complements and addresses some limitations of vegetation-based approaches while offering additional environmental benefits.
In this light, an additional consideration for future implementation involves the potential synergy between bio-weathering MICP and vegetation, as both offer robust, yet distinct, pathways for erosion control. Recent studies have demonstrated that combining soil additives with vegetation can optimise soil hydromechanical properties, especially under changing climatic conditions (Guo et al., 2024; Liu et al., 2024). However, the interaction between MICP and vegetation is complex. While MICP can enhance root–soil composite strength without negatively affecting vegetation growth (Hodges & Lingwall, 2020; Zheng et al., 2022), high levels of cementation may hinder root growth and vegetation coverage (Ghasemi & Montoya, 2022). The effects of MICP on plant health and germination rates vary depending on soil type and treatment solution composition (Hodges & Lingwall, 2020). Optimising MICP treatment solutions, such as adding phosphorus, can mitigate potential negative impacts on plant health (Ghasemi & Montoya, 2022). Nonetheless, these complexities should not discourage further investigations. Adapting such integrated approaches to bio-weathering MICP treatment could further enhance erosion mitigation while broadening the environmental co-benefits, indicating a promising avenue for research and practical application.
By leveraging the natural capabilities of native microorganisms, bio-stimulating MICP treatment reduces the environmental footprint of infrastructure projects. The enhanced erosion resistance and soil stability it provides can lead to longer-lasting unpaved roads, thereby reducing the need for frequent repairs and maintenance – a cost-effectiveness particularly important in regions with limited resources. Furthermore, its ability to perform under various inclinations and environmental conditions allows for wide application across different geographic areas, enhancing infrastructure resilience to climate change impacts. Although not the primary focus of this study, the bio-weathering process involved in the treatment consumes CO2, potentially contributing to carbon sequestration. This dual benefit of erosion control and environmental sustainability makes bio-stimulating MICP a promising area for further research.
CONCLUSIONS
The study demonstrated the potential of bio-stimulating using native microorganisms to enhance soil stability through MICP. The erosion testing results indicated a significant reduction in soil loss for bio-stimulating samples compared with control samples. The cumulative soil loss and the rate of erosion were markedly lower for bio-stimulating samples, demonstrating enhanced resistance to erosion. The bio-stimulating treatment reduced the erodibility rate by approximately 69% compared with untreated control samples, indicating a substantial improvement in the soil’s ability to withstand erosive forces.
In addition, the bio-weathering metabolic pathway explored in this study involves the sequestration of CO2, thus contributing to carbon dioxide capture and storage efforts. This environmental benefit adds a significant dimension to the application of MICP, aligning erosion mitigation practices with broader sustainability goals. Future studies will focus on measuring CO2 capture, contributing to the understanding of MICP as a dual-benefit solution for both erosion mitigation and carbon sequestration.
ACKNOWLEDGEMENTS
This research was supported by the 2023–2024 Fellowship of the Coalition for Disaster Resilient Infrastructure (CDRI), Application ID: 2301061780, ISPF ODA Newcastle University Institutional Fund and Brazilian National Council for Scientific and Technological Development (CNPq) – Call CNPq/MCTI/FNDCT N° 22/2024, Application ID: 445166/2024-0.
