Ph.D. - Civil Engineering

Permanent URI for this collectionhttps://hdl.handle.net/10125/880

Browse

Now showing 1 - 20 of 45

Carbon diversion for energy efficiency via biosorption in the high rate biological contactor - activated sludge process
(University of Hawai'i at Manoa, 2025) Wong, Tiow Ping; Babcock Jr., Roger W.; Civil Engineering
The high rate biological contactor (HRBC) is an advanced primary treatment that can remove particulate, colloidal, and soluble fractions of organic matter via biosorption and flotation, and divert the removed organic matter to anaerobic digestion for methane production. Simultaneously, this reduces the secondary aeration energy by reducing the organic load to the downstream secondary treatment. The pilot and bench-scale results demonstrated that, in addition to the activated sludge (AS) processes, waste activated sludge (WAS) from a biofilm secondary treatment technology could also serve as a biosorbent in the HRBC process. The tests were conducted at a range of contact times (15-60 minutes) and dissolved oxygen (DO) (0.2-2.0 mg/L) using WAS from a trickling filter/solids contact (TF/SC) process. The biosorption performance was the lowest when contact times were shorter than 30 minutes and was unstable at DO concentrations below 0.5 mg/L. The overall average of 20% soluble chemical oxygen demand (sCOD) capture was similar to findings reported by others using WAS from conventional AS processes. The biomethane potential (BMP) of the HRBC float can be as high as that of primary sludge (300-400 mL CH4/g VS), which is significantly higher than that of WAS. When operating the HRBC with a long contact time (>30 min) or with a high DO (>1 mg/L), the amount of sCOD removal increases, but the BMP of the float decreases. This phenomenon could suggest that the removal mechanism under these conditions is more oxidation than biosorption. Based on bench-scale tests, it was also found that biosorption is effective only when a WAS is paired with wastewater from the same facility. According to the literature, extracellular polymeric substances (EPS) reportedly comprise approximately half of the organic matter in AS and, therefore, strongly influence the properties of AS. The second phase of the study evaluated the component fraction of EPS normalized to volatile suspended solids (VSS) in WAS from a TF/SC process and its ability to biosorb organic matter from raw wastewater with a 30-minute contact time. Biosorption is defined as a process in which organic matter (carbohydrates, proteins, humic acids, DNA, uronic acids, and lipids) in raw wastewater sorbs onto a sorbent such as WAS. Based on the results, a statistically significant correlation was found between the total concentration of EPS and the protein component of the EPS, and the biosorption removal of soluble sCOD and truly soluble COD (ffCOD). Thus, the biosorption of soluble forms of COD can accurately be predicted by quantifying just the amount of proteins in WAS-associated EPS. However, no significant correlations were found for the biosorption of colloidal COD (cCOD). The results showed that the TF/SC WAS biosorbed 45-75 mg/L of COD in 30 minutes. WAS absorbed or stored the protein fraction of the soluble microbial products (SMP) during the biosorption process. The biosorption process effluent contained higher concentrations of humic acids than the untreated wastewater, warranting further study. Longer cation exchange resin (CER) extraction times yielded greater total EPS from the WAS: 90 ± 9, 158 ± 3, and 316 ± 44 mg/g VSS for 45-minute, 4-hour, and 24-hour extraction times, respectively. Thus, the EPS extracted accounted for only 9%, 15.8%, and 31.6% of the VSS, respectively, raising questions about whether the accurate characterization of EPS can be performed using the typical extraction time of 45 minutes, given the different extraction rates for different components. It was found that the humic acids fraction had a much slower extraction time than the other fractions. The third phase of the study characterized the composition of EPS in different types of AS processes and analyzed the biosorption of soluble organic matter when WAS was mixed with raw wastewater for primary carbon diversion. The fraction of AS organics identified as EPS was 26% in a membrane bioreactor (MBR), 54% in a conventional AS (CAS), and 51% in a TF/SC process. For CAS and MBR, the EPS were found to be approximately 15% carbohydrates, 40% proteins, 40% humics, and 5% uronics in the AS. Although biosorption was not correlated with VSS of the WAS, statistically significant correlations were found with the total amount of EPS (for TF/SC and CAS) and the protein fraction (for TF/SC and MBR) in the VSS. EPS from different types of WAS biosorbed the same amount of soluble organics, removing 1.43 ± 0.15 (n = 16) mg of sCOD, and 1.20 ± 0.18 (n = 16) mg of ffCOD, per mg of CER total extracted EPS. When utilizing multiple extraction methods in series, the CER method, followed by base extraction and then sulfide extraction, resulted in a nearly 100% increase in EPS extraction yield relative to the CER method alone, providing different EPS fractionation for CAS. The final part of this dissertation involves developing a wastewater model that can simulate the biosorption performance in the HRBC process. The HRBC is a process characterized by having a short HRT, a short SRT, a low DO, and a high food-to-microorganism ratio (F/M). The EPS constituent concentrations were converted to COD using stoichiometric equations. The conversion did not alter the finding that normalized total EPS concentrations showed a positive relationship with normalized sCOD removal, with a correlation coefficient of 0.91. The fraction of AS organics identified as EPS in terms of COD was 37% in CAS, 33% in TF/SC, and 18% in the MBR process. Each mg of EPS in COD (CODEPS) removes 1.02 mg sCOD. A commercial full-plant wastewater model was modified by incorporating the EPS in the WAS, and it was used to simulate the pilot test data. The maximum model error was -12.6%, and the remaining data points were within ±10% of the measured values. The modified model predicted 16% sCOD removal, compared with 21% sCOD removal by another proprietary wastewater model using the design data for a full-scale HRBC facility.
Methodological framework to empower artificial intelligence-embedded adaptive traffic signal control optimization
(University of Hawai'i at Manoa, 2025) Wang, Yiwei; Zhang, Guohui; Civil Engineering
Traffic congestion remains a persistent and critical challenge worldwide, shaping environmental outcomes, economic productivity, and overall quality of life. To address this issue, a wide range of traffic management strategies have been developed to enhance traffic flow stability and efficiency under varying traffic conditions. In urban areas, traffic signal control at intersections serves as the principal mechanism for managing traffic movements and often represents the key bottleneck that constrains road network performance. With advancements in network-wide sensing and communication technologies, data-driven AI methods offer significant potential to further enhance mobility, smooth traffic flow, and improve safety. Motivated by these opportunities, this study proposes a comprehensive methodological framework to enable AI-embedded adaptive traffic signal control optimization for real-world implementation. First, a hybrid modeling framework is proposed for MIMO traffic system modeling, where its hybrid structure with linearity between control input and predicted output is able to simplify the controller design to ensure operational feasibility for fast feedback signal control. Building upon this framework, a hybrid spatial-temporal graph attention network (HSTGAN) is proposed, aiming to capture complex spatial-temporal dependencies among traffic dynamics at signalized intersections. The experiment results illustrate that the hybrid modeling approaches yield clear performance gains over standalone deep learning models, such as the multilayer perceptron (MLP) model and the long short-term memory (LSTM) model. Additionally, HSTGAN achieves the best testing performance, which indicates that introducing a phase-level graph-based information transition mechanism and transformer attention mechanism for capturing spatial and temporal dependencies could improve the model performance. Subsequently, the computational effectiveness of the proposed hybrid modeling approaches is evaluated for the signal control optimization problem, in which the proposed analytical method based on a hybrid model is significantly faster than traditional optimization algorithms, such as trust-region optimization, especially upon deep learning model structures. With the advantage of a hybrid modeling framework in prediction performance and computational effectiveness, an AI-enabled closed-loop feedback optimal control (AI-CFOC) algorithm is used to systematically reduce total control delay across an entire corridor. To deploy the AI-CFOC into the existing physical signalized system, a comprehensive field implementation workflow is designed for an urban arterial along Nimitz Highway and Ala Moana Boulevard in Honolulu, Hawaii, with 4 corridors totaling 30 intersections. Finally, a before-and-after test is conducted to evaluate the performance of the proposed algorithm in real-world implementation. The test results demonstrate that the AI-CFOC system achieved an overall 4.3\% improvement in vehicle delays across all corridors, with improvement as high as 7.10% observed in the second corridor. Moreover, the computational time and latency of the total control loop are well within acceptable operational bounds for the cycle-based signal control mechanism. Furthermore, uncertainties in traffic system modeling and control processes can further complicate traffic signal system controllability. To partially address this challenge, a novel system modeling approach enhanced with error probability density function (PDF) shaping techniques is proposed, where a PDF shaping-based loss function with kernel density estimation is formulated for modeling training and adaptive learning. Additionally, two alternative approaches are provided based on the concepts of Kullback-Leibler distance and entropy. The numerical experiment is conducted, and results demonstrate that the error PDF shaping kernel-enhanced traffic system model can achieve prediction results comparable to the MSE-minimization-oriented model, which prioritizes minimizing modeling uncertainties. On the other hand, the HNN-PDF-based control approach outperforms the baseline control strategies and reduces overall average delays by 11.64% on average in a digital twin simulation environment.
Subsurface characterization using deep learning theory and techniques
(University of Hawai'i at Manoa, 2025) Bao, Jichao; Lee, Jonghyun; Civil Engineering
Estimating spatially distributed properties such as permeability from limited sparse measurements is a great challenge in subsurface characterization. In this dissertation, a data assimilation framework combining generative models and ensemble methods is explored. The proposed data assimilation framework is applied for subsurface characterization in CO2 storage sites and enhanced geothermal sites.Chapter 2 introduces the data assimilation framework, which combines the Ensemble Smoother with Multiple Data Assimilation (ES-MDA) and generative models such as the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) and the Denoising Diffusion Implicit Model (DDIM). The generators of the trained generative models are used to generate complex permeability fields from latent space, and ES-MDA is then used to update the latent variables by assimilating available measurements. Additionally, a deep learning surrogate model is used for faster simulations. Chapter 3 applies the proposed data assimilation framework to CO2 storage sites characterization. Several subsurface site characterization examples including Gaussian, channelized, and fractured reservoirs are used to evaluate the accuracy and computational efficiency of the proposed method and the main features of the unknown permeability fields are characterized accurately with reliable uncertainty quantification. Furthermore, the estimation performance is compared with a conventional inversion approach, and the proposed approach outperforms the inversion method in several benchmark cases. The superior performance is investigated by visualizing the objective function in the latent space: because of nonlinear and aggressive dimension reduction via generative modeling, the objective function surface becomes extremely complex while the ensemble approximation can smooth out the multi-modal surface during the minimization. Chapter 4 applies the proposed data assimilation framework to the enhanced geothermal site characterization at the Utah Frontier Observatory for Research in Geothermal Energy (FORGE) site. Discrete Fracture Networks (DFNs) are generated from measured data to represent the enhanced geothermal reservoir. DDIM is trained using the permeability fields upscaled from DFNs. Furthermore, a surrogate model is trained for faster thermal-hydraulic simulations. Data assimilation is performed using pressure data from the 9-hour circulation test conducted in April 2024 at the Utah FORGE site, and a one-month simulation is conducted using the estimated permeability field to investigate the long-term performance of the enhanced geothermal system. The results demonstrate that the predicted injection pressure and production temperature are consistent with the measured data.
Estimation of scour depth around bridge piers using innovative machine learning approaches
(University of Hawai'i at Manoa, 2025) Eini, Nasrin; Bateni, Sayed M.; Civil Engineering
Bridge scour, the removal of sediment around bridge foundations due to flowing water, is one of the primary causes of bridge failure worldwide. Accurate estimation of scour depth is essential for designing safe and resilient bridge foundations. Traditional empirical equations often fail to capture the complex, nonlinear interactions among hydraulic, geometric, and sediment parameters that govern scour processes. This dissertation integrates advanced hybrid machine learning (ML) modeling with field-based monitoring to enhance scour depth prediction, interpretability, and practical applicability for bridge safety management.A comprehensive dataset of laboratory and field observations was compiled to evaluate equilibrium scour depth under diverse flow and pier conditions. Several optimized and interpretable ML frameworks were developed, including XGBoost, CatBoost, and metaheuristic algorithms, to model scour behavior for circular, non-circular, and debris-affected bridge piers. The models were optimized several metaheuristic optimization algorithms to improve prediction accuracy and generalization. The use of SHapley Additive exPlanations (SHAP) enabled clear interpretation of the influence of key input variables such as flow depth, velocity, pier diameter, and sediment size, providing physical insight into the scouring mechanisms. The results demonstrate that optimized hybrid ML models outperform traditional equations in prediction accuracy and successfully capture the influence of debris and flow variability. This research advances current understanding of scour dynamics and supports resilient infrastructure management under changing hydraulic and climatic conditions.
Developing an artificial intelligence-based near real-time wildfire prediction system for the State of Hawaiʻi
(University of Hawai'i at Manoa, 2025) Tran, Thi Kieu Trang; Bateni, Sayed M.; Civil Engineering
In Hawaiʻi, wildfires pose a growing threat driven by invasive grasses, land use changes, and climate variability. The lack of reliable vegetation data, combined with the need for high-resolution fire susceptibility mapping and near real-time predictive tools, presents major challenges for wildfire management in the islands. This dissertation addresses these challenges through a series of studies that integrate remote sensing, machine learning, and optimization techniques. The first study develops a gap-filling framework for normalized difference vegetation index (NDVI) time series, combining multiple imputation and machine learning methods to reconstruct missing observations. This approach improves the accuracy and continuity of vegetation monitoring. The second, third, and fourth studies develop wildfire susceptibility mapping across four Hawaiian Islands: Oʻahu, Hawaiʻi , Molokaʻi, and Kauaʻi. By applying state-of-the-art machine learning models enhanced with nature-inspired optimization algorithms, these studies produce highly accurate susceptibility maps and identify the environmental drivers associated with wildfire occurrence. The final study develops a near real-time wildfire ignition prediction system for Hawaiʻi State. Using hybrid random forest models with daily climate and land cover inputs, the models generate dynamic maps of ignition probability. The integration of explainable artificial intelligence, such as SHAP, ensures that predictions are interpretable and actionable for wildfire managers and decision-makers. Together, these five studies form a comprehensive framework for advancing wildfire science in Hawaiʻi. The contributions include methodological innovations for vegetation monitoring, improved susceptibility mapping across Hawaiʻi State, and the creation of a near real-time ignition system. The findings support proactive, data-driven strategies for wildfire prevention and risk reduction, while offering transferable approaches applicable to other fire-prone regions worldwide.
Investigating the causes and mechanisms of a concrete pavement distress in Hawaiʻi
(University of Hawai'i at Manoa, 2025) Rahman, Muhammad Abdur; Ooi, Phillip S K; Civil Engineering
The soils underneath a PCC pavement on the island of Oahu, Hawaii were sampled and tested extensively because adjacent slabs had experienced vertical and horizontal separation. Based on the site geology, historical sea level rise and ebb may have caused the alluvial soils of volcanic origin to mix with calcareous fines (calcite), the source of which are the remains of dead coral giving rise to possible cementation in the soil. Hydrochloric acid and XRD tests suggest the presence of calcite in the upper reaches of the soil deposit and halloysite throughout. Index tests criteria proposed by the U.S. Army Corp of Engineers to assess soil collapsibility showed that the soils are collapsible based on dry unit weight, porosity and plasticity index values. However, the liquid limit test criterion was not consistent with the other criteria suggesting that the U.S. Army Corp of Engineers liquid limit criteria may not be as reliable in identifying collapsible tropical soils. The soils in the median are not as collapsible compared to soils directly underneath the pavement possibly due to the fact that the soils in the median have been subjected to a decade more exposure to atmospheric conditions and possibly because the soils below the pavement are constantly moist due to the canopy effect. The matric suction values under the pavement indicate that the soils below the pavement are saturated or nearly saturated. In contrast, the matric suction of the soil in the median fluctuated depending on the precipitation and the weather. Another notable feature is that the moisture content is higher at 0.5 ft than at 2.5 ft below the pavement but the reverse is true in the median . This phenomenon can also be attributed to the canopy effect at the pavement location and conversely, there is no canopy effect in the median. Suction controlled consolidation tests ran using a constant rate of strain consolidometer indicated soil collapse during three different phases of testing: saturation, drying and cyclic loading. A batch of samples collapsed by about 4% to 6% when saturated at a seating stress of only about 5 kPa. This occurred in the topmost samples below the pavement. The samples were first saturated so that the starting point on the soil-water retention curve is known. Since these samples are from the shallows, it can be implied that the shallowest soil is highly collapsible. It is postulated that the collapse in these samples is due to the breaking of calcite cementation bonds upon dissolution. A second batch of samples did not collapse during saturation. Instead, they collapsed by about 1% during suction application. In both samples, the suction applied was 200 kPa. One possible explanation for this is that the halloysite present loses the layer of water molecules in between the kaolinite units upon drying or application of suction. As a result, the soil collapsed when the water layer is removed irreversibly. Admittedly, the collapse due to suction application is smaller than that during saturation. The third batch of samples collapsed due to load-unload-reload cycles on saturated samples. The rationale for this is that PCC pavements heat up during the afternoons causing the pavement slab to curl downwards. They cool down in the early hours of the morning causing the pavement slab to curl upwards. These temperature cycles cause stresses to escalate at the edges and center of the pavement slab during the day and night, respectively. To simulate this in a consolidometer, the soil is loaded to a target vertical stress and then unloaded at a rate of 1 cycle per day for many cycles to simulate the repeated upward and downward curling. It was found that 60 to 80 cycles were adequate to yield the asymptotic strain. The collapse settlement or permanent deformation due to repeated loading and unloading depends on factors such as the number of cycles, duration of loading, and magnitude of stress difference in a cycle. A methodology is presented to estimate the long-term collapse settlement due to this mechanism. A hyperbolic curve can be fitted through each collapse versus cycle number curve, from which the asymptotic value of collapse strain can be estimated. The asymptotic strain can then be plotted versus the stress difference to allow an engineer to easily estimate the long-term load-unload-reload collapse for any value of stress difference. Curling-induced stresses were estimated utilizing first a finite difference analysis to estimate the temperature profile and then applying the temperature profile in a finite element analysis. The diurnal curling-induced subgrade stress can be as high as 25 kPa at the pavement edges with truck loading during the hottest time of the day in relation to an unloaded pavement at the coldest time of the morning. Based on a diurnal cyclic stress difference of 25 kPa and the cyclic consolidation test results, the collapse settlement due to curling-induced subgrade stress cycling is estimated to be about 2%. In sum, the 3 different sources of collapse strains are as follows: a. Collapse Strain due to Curling-Induced Subgrade Stress Cycling ≈ 2%. b. Collapse Strain due to Saturation and Loss of Cementation ≈ 4 to 6%. c. Collapse Strain due to Drying or Loss of Water Between Kaolinite Units ≈ 1%. Summing the 3 sources of collapse strain leads to a total collapse of 7% to 9%. According to Jennings and Knight (1975), the severity of the problem for these values of strain is considered “Trouble.” These stresses induced in a collapsing soil can lend an explanation of why the pavement distress was observed in the concrete pavement. Finally, some methods of mitigating collapse of similar soils are proposed in the Implementation section of the report (Section 5.3). They include dynamic compaction, pre-wetting and pre-loading, partial excavation and replacement, and a variety of soil improvement options.
Comprehensive approach to vulnerability assessment of structures and communities under tsunami hazards
(University of Hawai'i at Manoa, 2025) Abdelhafeez, Mostafa Hassan Fathi; Moon, DoSoo; Civil Engineering
Coastal regions face significant threats from tsunamis, which cause severe damage to infrastructure and human life. Due to their infrequent occurrence, accurately assessing tsunami hazards remains challenging, complicating risk evaluation despite their potentially catastrophic impact. This dissertation aims to enhance the understanding of structural and community vulnerability to tsunamis by developing advanced assessment methodologies. The research first evaluates structural fragility under tsunami loads and then expands to community-level vulnerability, offering insights for risk mitigation and resilience enhancement. A key component of this study is the use of fragility and vulnerability curves, which quantify the probability of structural failure under tsunami forces. These curves, based on cumulative log-normal distributions, provide critical insights for risk assessment and disaster management decision-making. This dissertation develops a comprehensive analytical approach for evaluating the fragility of structures subjected to tsunami loads. Through advanced numerical simulations, it examines structural behavior under extreme tsunami conditions, providing critical insights into their response. The proposed simulation approach is applied to selected structures, with the results analyzed to enhance understanding of structural fragility under tsunami loading, supporting the development of more resilient infrastructure in tsunami-prone regions. This dissertation presents three studies focused on the vulnerability of structures subjected to tsunami waves. The first study focuses on the tsunami vulnerability analysis of a reinforced concrete (RC) building classified as Tsunami Risk Category II. It examines the impact of different load distributions on the structural response during a tsunami event. The study also defines the inundation depth and flow velocity for dynamic assessments of the structural model. To quantify the effects of tsunamis on building performance, fragility relationships are derived from nonlinear dynamic response history analysis, utilizing a novel structural reliability method. The findings highlight that the number of tsunami cycles significantly influences the vulnerability of reinforced concrete structures, with a uniform load distribution being recommended for its conservatism. The second study examines the tsunami vulnerability of structures through numerically simulated tsunami waves, analyzing their propagation and impact on coastal dynamics and structural fragility. This study investigates the effects of various factors, including bathymetric features, alongside tsunami wave parameters. The simulation tool FUNWAVE-TVD is employed to model tsunami propagation and inundation. These tsunami intensity measures are then used to develop fragility curves to evaluate the structural probability of failure under different tsunami conditions. The study shows that higher Manning coefficients, steeper slopes, and longer wave periods reduce fragility, while more tsunami cycles and higher crest amplitudes significantly increase vulnerability, providing valuable insights for coastal resilience. The third study focuses on assessing the vulnerability of structures to 2D tsunami waves, with the goal of enhancing resilience in the coastal community of Kahului, Maui. A numerical simulation approach is developed to evaluate the fragility of selected structures under various tsunami scenarios through nonlinear dynamic time history analyses. The study aims to develop fragility curves for commercial, residential, and industrial buildings within the community, specifically examining the effects of wave amplitude, wave period, and other structural parameters. The findings indicate that both offshore tsunami amplitude and wave period significantly influence structural vulnerability, while building characteristics, such as design and materials, further impact the vulnerability of individual structures. Emphasizing the importance of estimating structural vulnerability, the study seeks to improve the accuracy of risk assessments and inform decision-making in disaster management. In conclusion, this research aims to advance the resilience of structures subjected to tsunami loads and to develop methods for assessing the vulnerability and performance of these structures. It highlights the importance of accurate vulnerability evaluation under varying tsunami wave conditions, considering factors such as bathymetric changes, wave characteristics, and structural types, to enhance risk assessment and decision-making in disaster management. The findings provide critical insights into the behavior of structures under tsunami loads and contribute to improved designs for both new and existing structures, including commercial, residential, and industrial buildings in coastal areas. Ultimately, these studies offer the potential to mitigate the economic and societal impacts of tsunami-related damage and inform effective coastal planning and disaster mitigation strategies.
Development of a numerical model for tsunami-driven debris transport and hazard assessment in a coastal community
(University of Hawai'i at Manoa, 2025) Koh, Myung Jin; Park, Hyoungsu; Civil Engineering
Tsunamis pose a critical risk to coastal communities, not only through inundation but also via debris transport, which intensifies damage by impacting infrastructure, blocking evacuation routes, and creating damming effects. This dissertation introduces a Debris Impact Hazard Assessment (DIHA) Framework to systematically evaluate tsunami-driven debris hazards at a community scale. The DIHR integrates numerical modeling and experimental data to generate actionable hazard metrics, including debris dispersion ratios, maximum impact loads, and intensity mapping. These metrics provide critical insights into vulnerable zones and high-risk areas, supporting disaster resilience planning and infrastructure design. The framework is applied to Honolulu Harbor, Hawai‘i, under a hypothetical 2,500-year tsunami scenario, demonstrating its effectiveness in identifying hazard hotspots.The newly developed numerical model has two key features. First, it introduces a semi-analytical solution for debris transport (tracking) model enabling to simulations of DIHA at a community-scale domain. Second, it incorporates a novel "randomness" variable at debris collision. Especially, this variable account for inherent uncertainties in the debris motion tracking, such as angular momentums, and varied reflection angles, enhancing predictive reliability while maintaining computational efficiency. The model simplifies debris representation through disk-shaped elements, validated against scaled laboratory experiments, ensuring robust simulation of debris dynamics. By capturing the probabilistic nature of debris transport, the model aligns closely with observed phenomena and improves the assessment of complex, real-world scenarios. The integration of DIHA and the randomness-enhanced debris transport model represents a significant advancement in tsunami hazard modeling. The findings demonstrate the importance of coupling site-specific hazard metrics with adaptive modeling techniques, bridging gaps between theoretical research and practical disaster management applications. This dissertation contributes to the scientific understanding and mitigation of tsunami-driven debris hazards, enhancing coastal community resilience against future tsunami events.
Multi-hazard fire vulnerability assessment and resilience quantification
(University of Hawai'i at Manoa, 2025) Sherif, Mohamed Ayman; Moon, Dosoo; Civil Engineering
The increasing frequency and severity of multi-hazard events, particularly those involving fire and seismic loading, pose significant challenges to the resilience of structural systems. Fire following an earthquake, concurrent fire-seismic events, and fire following earthquakes with aftershocks introduce complex degradation mechanisms that compromise material properties, structural stability, and load-bearing capacity. Traditional design methodologies often treat these hazards in isolation, failing to capture their interdependent effects on structural behavior. This dissertation advances the field of multi-hazard vulnerability assessment by integrating fire-inclusive material modeling, probabilistic fragility analysis, and resilience quantification to improve performance-based design strategies. A central contribution of this research is the development of advanced nonlinear material models for steel and concrete under fire and cyclic loading conditions. These models capture critical thermal-mechanical interactions, including strain reversal effects, cyclic degradation, and temperature-dependent stress-strain behavior. The study introduces a bilinear-quadratic cyclic model for steel, incorporating plastic strain accumulation to enhance predictive accuracy under sequential and concurrent dynamic loading conditions. For concrete, a modified Mander’s model is extended to account for high-temperature degradation, including spalling, stiffness reduction, and progressive strength deterioration. These material models are validated through numerical simulations and experimental data to ensure their reliability in hazard assessments. Building upon these material formulations, a multi-hazard fragility assessment framework is developed to quantify the probability of structural failure under fire and seismic loading conditions. Probabilistic fragility curves are generated using finite element reliability using MATLAB method, incorporating uncertainty in fire intensity, seismic ground motion, and structural parameters. Case studies examine the fragility of reinforced concrete (RC) structures under various hazard sequences, including fire-following-earthquake, fire following earthquake with aftershock, and concurrent fire-seismic loading. Results highlight significant reductions in structural capacity due to fire-induced stiffness degradation, yield strength reduction, and progressive material failure mechanisms. To extend the utility of fragility analysis, this research introduces a resilience index as a computational metric for quantifying post-hazard structural performance. This index integrates fragility-based failure probabilities with structural damage states, providing an analytical tool for assessing recovery potential and structural adaptability under multi-hazard exposure. The framework demonstrates how elevated temperatures and cyclic loading interactions exacerbate structural vulnerabilities, reinforcing the need for multi-hazard-inclusive design approaches. The findings of this research contribute to the advancement of fire-aware fragility modeling and performance-based multi-hazard assessment, providing engineers, policymakers, and researchers with quantitative tools for designing resilient infrastructure in fire-prone and seismic regions. The proposed methodologies and models enhance the accuracy of multi-hazard structural performance predictions, bridging critical gaps in current hazard assessment frameworks.
Enhancing Wastewater Surveillance Through the Development of Novel Detection Assays for Variants of SARS-CoV-2 and Influenza Viruses
(University of Hawai'i at Manoa, 2024) Jeon, Min Ki; Yan, Tao; Civil Engineering
The outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), responsible for the COVID-19 pandemic, has posed a significant threat to global public health. The transmission of this virus through respiratory droplets and aerosols has led to the rapid spread of infections. Since the COVID-19 pandemic outbreak started in late 2019, numerous studies have demonstrated the presence of SARS-CoV-2 viral particles or genomic RNA in bodily wastes, including feces, urine, and respiratory fluids, originating from both symptomatic and asymptomatic patients. Due to these results, wastewater surveillance has been growing as a supportive tool to monitor and determine the presence of respiratory viruses in communities. This approach can help identify the emergence of outbreaks before clinical cases are reported, allowing public health officials to take early measures to contain the spread of the virus. However, despite its potential benefits, wastewater surveillance also has some limitations that need to be addressed for accurate and effective surveillance. The overall objective of this dissertation was to enhance wastewater surveillance for enveloped respiratory viruses such as SARS-CoV-2 and influenza viruses through advancements in data adjustments and molecular analyses. First, we conducted cross-correlation analysis on time-series data of SARS-CoV-2 RNA abundance and clinical case data from two major wastewater treatment plants (WWTPs) in Honolulu, Hawaiʻi. The detection of time lags between the data was enhanced through prewhitening and population normalization strategies. Second, molecular detection inhibitors in wastewater, especially humic acid, were used to determine the inhibition effect on hydrolysis probe-based real-time polymerase chain reaction (qPCR). This indicates the significance of comprehending the inhibitory mechanism of reporter dye of hydrolysis probes in the presence of humic acid in wastewater to ensure accurate quantification of targets. Third, we developed a new nested RT-PCR-PCR method to amplify and sequence the spike (S) protein gene region of SARS-CoV-2 (532 bp), including the receptor binding domain (RBD) for mutation identification from monthly wastewater samples collected from a small residential sewershed in Honolulu, Hawai‘i since the beginning of the COVID-19 pandemic. The study aimed to determine the viability of detecting novel variants in wastewater and to evaluate the performance of removal of PCR inhibitors with the Inhibitor Removal Technology® (IRT) and internal amplification standard (IAS). Fourth, a new amplicon sequencing method was developed to simultaneously amplify 576 bp of the polymerase basic 1 (PB1) encoding gene from various influenza virus variants present in wastewater. To address amplification bias resulting from varying primer-template affinities among variants and the presence of PCR inhibitors, IASs with multiple genes and different primer-template affinities were used to evaluate the amplification bias. Collectively, the outcomes of this dissertation will illustrate the potential for enhancing wastewater surveillance of SARS-CoV-2 and influenza viruses, leading to more accurate detection of their abundance and variants in wastewater.
EVALUATION OF FOOD WASTE AND SEWAGE SLUDGE ANAEROBIC CO-DIGESTION: KINETIC MODELING, META-ANALYSIS, AND LONG-TERM OPERATION WITH MICROAERATION
(University of Hawai'i at Manoa, 2023) Chuenchart, Wachiranon; Khanal, Samir Kumar; Civil Engineering
Machine learning modeling has recently gained attention in bioprocess engineering research for its precision prediction, optimization, and failure detection. However, due to its “black box” nature, interpretability approaches are needed to integrate to improve their understandability. In our study, the conventional approach of bioprocess assessment (i.e., kinetic modeling, a systematic review with meta-analysis, and long-term continuous operation) assisted with statistical analysis, and machine learning modeling was employed. In the batch experiments, higher food waste content resulted in higher specific methane yield (SMY), indicating higher biodegradability during co-digestion. The superimposed model with the first-order kinetic and modified Gompertz structure exhibited better accuracy among others in co-digestion. Meta-analysis reveals synergistic interactions of lignocellulosic biomass with animal manures and food waste with animal manures and lignocellulosic biomass (relative synergistic index, RSI > 1.20). Based on correlation analysis, multilinear regression, and tree-based regression, temperature was identified as a key parameter to improve methane yield in co-digestion of lignocellulosic biomass and fats, oils, and grease. However, food waste content is more important in food waste co-digestion. Long-term anaerobic co-digestion reaffirms that higher food waste content resulted in higher methane yield due to its rapid biodegradability and reveals the interactions of microaeration on hydrogenotrophic methanogenesis. The time-series model, specifically the trained nonlinear autoregressive network with exogenous inputs (NARX), also showed promising application on continuous systems with R2 = 0.8–0.9.
ADAPTIVE CONTROL STRATEGY FOR TRAFFIC PLATOON COORDINATION EMPOWERED BY CONNECTED AND AUTOMATED VEHICLES
(University of Hawai'i at Manoa, 2023) Yuan, Runze; Zhang, Guohui; Civil Engineering
Traffic congestions caused by increasing population, car ownership, and corresponding traffic demands have been a serious issue that not only hazards the safety and comfort of the commuters, but also increases the economic losses and environmental damages. Platoon control enabled by connected and autonomous vehicles (CAV) is regarded as a powerful solution to alleviate the traffic pressure and congestions caused by road bottlenecks. However, it will take a long time before the CAV can share a high market penetration rate to achieve pure CAV platooning. Therefore, both CAVs and the human driven vehicles (HDV) without Vehicle-to-Vehicle (V2V) communication mechanism will coexist and form mixed platoons in the future for a long period of time. It places urgent demands on reliable and robust strategies for this kind of mixed platoons to achieve steady and smooth platoon control. In this study, a specific mixed platoon composition with a leading CAV, a following CAV, and several sandwiched HDVs between them is adopted to research the mixed platoon characteristics. A spring-mass-damper-clutch physical system inspired linear car following model with delay is used to account for the delayed reaction and the varying driving modes of human drivers. Since the HDVs are not equipped with V2V communication devices, the last following CAV is regarded as degraded CAV (DCAV). Therefore, an Adaptive Numerical algorithm for Subspace system Identification (AN4SID) method is proposed to identify the model of direct preceding HDV in front of the last following DCAV. The model could be used to predict the future states of the preceding HDV given the estimated delay through vector fitting and the planned future trajectory of the leading CAV. Then a hybrid control strategy combining the Model Predictive Control (MPC) and the optimal feedback control methods is applied to the following DCAV. The predicted HDV states will be used to calculate the reference trajectory of the following DCAV for the MPC process. Then the MPC control signal will be used as the feedforward input together with the feedback control input to eliminate the deviation caused by prediction errors.By conducting simulations based numerical experiments on MATLAB platform and the real vehicle trajectory data from the Next Generation Simulation (NGSIM) open dataset, the proposed AN4SID method shows reliable prediction performance compared with the Recursive Least Square (RLS) method and the Iterative Extended Kalman Filter (IEKF) method. Then, further experiments judged that our proposed hybrid control strategy overperformance both the Linear Quadratic Regulator (LQR) and the regular MPC algorithm, in terms of the safety, comfort of passengers, and fuel efficiency.
A Study Of Machine Learning Applications For Solving Problems In Structural Engineering
(University of Hawaii at Manoa, 2023) Song, Xi; Cho, Chunhee; Civil Engineering
Structural engineering, a sub-discipline of civil engineering, involves the design and analysis of structures. The goal of structural engineering is to ensure that these structures are safe, stable, and capable of performing their intended functions throughout their lifespan. Meanwhile, machine learning (ML), a subset of artificial intelligence, utilizes statistical methodologies to learn and generate predictions or decisions directly from data. This dissertation explores the integration of these two fields, offering innovative approaches to solve structural engineering problems through the application of ML.The work comprises three projects that showcase the application of ML in the domain of structural engineering. The first project focuses on the predicting structural strength of steel circular hollow section (CHS) X-joints. Using support vector machines and deep neural networks, this project demonstrates how machine learning can effectively manage structural strength prediction tasks, pointing towards a promising future for ML in this field. The second project ventures into seismic risk analysis, a crucial part of structural safety evaluations. The use of advanced ML algorithms, including the discussion of hyperparameter tuning and model optimization, allowed for a more efficient and accurate prediction of seismic impact on structures such as railway bridges. The last project adopts machine learning for structural damage detection, using pre-trained convolutional neural networks tailored for image-oriented input. Key structural dynamic properties are transcribed into scalograms via wavelet transform, serving as training samples for the machine learning model. The promising outcomes from this project endorse the potential of machine learning in augmenting the efficiency and accuracy of processes for detecting and evaluating structural damage. Throughout the dissertation, a progressive learning journey unfolds, detailing how the understanding and application of ML evolved from basic techniques to more advanced methodologies. Each project enhances the subsequent one, demonstrating a continuous improvement in the application and understanding of ML. This research demonstrates that machine learning can provides new perspectives and methods for tackling topics in structural engineering, greatly enhancing the efficiency and effectiveness of problem solving in the field. The integration of ML can circumvent the complex experiments, simulations, and calculations that are typically required in structural design and analysis. The work encourages future discussion in the field, refining ML applications, exploring more innovative techniques, and ultimately continuing to push the boundaries of what can be achieved in structural engineering.
Integrated Approaches For Scalable Aquifer Numerical Modeling, Inversion, And Upscaling
(University of Hawaii at Manoa, 2023) Seo, Young-Ho; Lee, Jonghyun; Civil Engineering
In this dissertation, three interconnecting research themes in the domain of groundwater modeling and characterization are explored. The dissertation represents a significant integration of novel approaches and computational tools for groundwater modeling and characterization. It not only improves our current understanding but also presents considerable new directions for future study, making a significant contribution to groundwater modeling. Chapter 2 focuses on the development of a distinctive joint-inversion methodology, which utilizes hydrogeological, self-potential, and magnetotellurics data, to estimate hydraulic conductivity and electrical resistivity. The proposed technique doesn't necessitate any assumptions related to petrophysical relationships and demonstrates a 25\% improvement in the estimation of hydraulic conductivity in comparison to single data-type inversions, providing crucial insights into regions beyond immediate observation wells. In Chapter 3, a significant focus is placed on developing a reliable hydraulic conductivity upscaling tool for high-dimensional groundwater flow models. Recognizing the vital role of accurately representing hydraulic conductivity at an appropriate scale, the study strived to develop a computational tool that effectively balances computational efficiency while preserving key features of the detailed hydraulic conductivity field. The tool, based on Kitanidis' (1990) hydraulic conductivity upscaling approach, has the capability to calculate upscaled hydraulic conductivity values in the tensor form and account for anisotropy. Rigorous tests were carried out to assess the performance of the tool, and its resilience under various flow conditions, providing a reliable resource for high-dimensional groundwater modeling. Chapter 4 addresses the development of the PISALE software. This tool is specifically designed to manage the complexities of groundwater flow processes in Pacific islands that are marked by dynamic interactions between freshwater and seawater in highly heterogeneous volcanic rocks. The software integrates advanced mathematical techniques and parallel programming models to accelerate solutions and offer precision in reproducing freshwater-seawater interfaces in large-scale coastal aquifers.
Testing And Evaluation Of Self-consolidating Concrete (SCC) Using Volcanic Aggregates
(University of Hawaii at Manoa, 2023) Hosseinpanahi, Azadeh; Shen, Lin; Civil Engineering
Self-consolidating concrete (SCC) flows under its own weight, passes through intricate geometrical configurations, and fills the formwork without vibrating or consolidating. In comparison with conventional concrete mixes, SCC should be closely controlled in terms of its composition and rheological properties to meet the fresh property requirements simultaneously. However, large-scale applications and acceptance have not yet been well-established in Hawaii for SCC materials. The problem can be attributed to Hawaii's unique aggregate, which is locally mined and crushed with absorption capacities ranging from 1 to 5%. Because volcanic aggregate has different absorption capacities, gradations, and textures than mainland aggregate, conventional SCC mix designs will not work. As a result, this study examined how aggregate properties affect SCC rheology. It was found that higher aggregate volume, higher fine aggregate to coarse aggregate ratio, smaller aggregate size, and lower aggregate packing density may increase the yield stress of the SCC mixture. Aggregate size had an insignificant effect on plastic viscosity. SCC mixes were also studied for their mechanical and durability properties to ensure their sustainability. Furthermore, since the rheological properties of SCC have been proven in Chapter 2, machine learning models were conducted to predict plastic viscosity. During the last phase of this project, liquid carbon dioxide admixture was used to test concrete carbon neutrality in the field. According to the research, CO2 admixtures do not adversely affect the fresh, mechanical, or durability properties of concrete, and they may be useful for carbon sequestration in the construction industry.
Integrated Framework For Multi-hazard Resilience Assessment Of Infrastructures
(University of Hawaii at Manoa, 2023) Ghanem, Amr; Moon, DoSoo; Civil Engineering
Natural hazards such as earthquakes and tsunamis can have a severe impact on societies globally, causing significant economic losses and leading to casualties. In marine environments, earthquake-induced fault ruptures can trigger devastating tsunamis that amplify the scale of the disaster. Therefore, it is crucial to assess the structural fragility of infrastructure and buildings in regions that are prone to seismic and tsunami hazards. Previous research has shown that earthquakes and tsunamis have caused thousands of deaths and billions of dollars in losses. Thus, it is imperative to continue investing in research aimed at developing resilient infrastructure systems that can withstand multiple natural hazards, especially earthquakes, and tsunamis. By understanding the risks and implementing measures to enhance infrastructure resilience, communities can mitigate the impact of these hazards and reduce their devastating effects on their economies and societies.In the context of structure loss estimation and hazard risk assessment, fragility or vulnerability curves are commonly utilized. These curves represent the probability of failure for a specific hazard and are typically displayed using cumulative log-normal distribution functions that express the likelihood of achieving predetermined damage criteria. By utilizing fragility or vulnerability curves, decision-makers can better understand the potential for structural damage or failure during the hazard and develop strategies to mitigate risk and enhance infrastructure resilience. This dissertation developed a comprehensive analytical fragility framework for the assessment of structures subjected to multi-hazards. As well as to investigate the behavior of structures under these extreme loads using numerical simulation techniques and to provide valuable insights into the behavior of structures under seismic and tsunami loads. The developed simulation frameworkis applied to selected structures, analyzed, and compared the results of the research, and offers insights into the reliability and robustness of current assessment methodologies. The results of the research study are expected to contribute to the broader research goals of improving our understanding of structural fragility under extreme loads and developing more resilient structures in seismic and tsunami regions. This dissertation presents three studies that investigate the vulnerability of different types of infrastructures (buildings and bridges) when subjected to consecutive hazards (seismic and tsunami loads). The first research focuses on the seismic vulnerability analysis of skewed reinforced concrete bridges, which are commonly used in transportation networks due to their complex and non-orthogonal nature. Despite their geometric irregularities, such bridges are often necessary, and avoiding them is not a viable solution. To quantify the effects of skewness on bridge performance, fragility relationships for various bridge models are derived from nonlinear dynamic response history analysis using a new structural reliability method. The results show that the skew angle has a direct effect on the seismic vulnerability of reinforced concrete (RC) bridges, and this research can be useful for designing new skew RC bridges. The second research investigates the seismic fragility of reinforced concrete frames with mass irregularities resulting from possible changes in structure use. Such changes can significantly affect live load distributions, making the structure more vulnerable to earthquake damage. Comprehensive three-dimensional structural models are analyzed for accurate vulnerability evaluation, considering twenty-one frame models with varying vertical and in-plan irregularities under different live load distribution scenarios. The results highlight the effects of vertical and in-plan mass irregularities on the structural vulnerability and performance of reinforced concrete frames when subjected to earthquake loading. The third research addresses the vulnerability of structures under combined seismic and tsunami loads, which is a critical issue for enhancing resilience in coastal regions. Neglecting the interaction between joint intensity-measure fragilities can lead to significant inaccuracies when estimating damage or structural loss. Therefore, a numerical simulation framework is developed to assess the fragility of selected structures under various earthquake and tsunami scenarios using successive nonlinear dynamic time history analyses. The study emphasizes the importance of considering the complex interaction between various intensity measures to improve the accuracy of risk assessment and decision-making in disaster management. In conclusion, this research aims to enhance the resilience of structures under different loading conditions and to develop the knowledge and approaches to assess the vulnerability and performance of such infrastructures. The research emphasizes the need for an accurate evaluation of vulnerability under different loading conditions, including structural irregularities and complex loading scenarios, to improve the accuracy of risk assessment and decision-making in disaster management. The results of these studies provide valuable insights into the behavior of structures under different hazards loads and are expected to be useful for new designs of reinforced concrete bridges and frames subjected to seismic and tsunami loads and to assess the vulnerability of existing structures to hazard events., with the potential to reduce the economic consequences of damage to transportation systems and coastal regions.
Applications Of Innovative Building Material And Computer Vision Methods In Geotechnical Engineering
(University of Hawaii at Manoa, 2022) Han, Xiaole; Jiang, Ningjun; Civil Engineering
The Hawaiian Islands have been a world-famous traveling spot for their unique tropical island view. But the particular geological formation of the islands and their unique locations have always proposed challenges for geotechnical engineers and geologists. For example, coral sand is widely encountered in coastal areas of tropical or subtropical regions. It can be found on most beaches in Hawaiian islands. Compared with silica sand, it usually exhibits weaker mechanical performance from the perspective of engineering geology. Thus, necessary soil improvements shall be applied to the coral. Considering the fragile and unique ecosystem, sustainable material with less carbon footprint and less environmental impact would be developed and selected priorly. Moreover, the infrastructures along the island have been facing coastal erosion issues from both physical erosion (waves) and chemical erosion (sea wind and seawater). The road embankment, the house embankment, the harbor, etc often require maintenance to sustain their service time. Due to the topography and climate, the windward side of the coastal area on Oahu is suffering from marine microplastics (MP) pollution issues. Furthermore, as an island state, hurricanes and tsunamis could also threaten the safety of islanders and infrastructures. Therefore, as a geotechnical major Ph.D. student, this dissertation would devote some potential solutions to the challenges. Firstly, a novel alkali-activation-based sustainable binder was developed for coral sand stabilization. The alkali-activated slag (AAS) binder material was composed of ground granulated blast furnace slag (GGBS) and hydrated lime with the amendment of biochar, an agricultural waste-derived material. The biochar-amended AAS stabilized coral sand was subjected to a series of laboratory tests to determine its mechanical, physicochemical, durability, and microstructural characteristics as well as durability. Results show that the addition of a moderate amount of biochar in AAS could improve soil strength, elastic modulus, and water holding capacity by up to 20%, 70%, and 30%, respectively. Moreover, the addition of biochar in AAS had a marginal effect on the sulfate resistance of the stabilized sand, especially at high biochar content. However, the resistance of the AAS-stabilized sand to wet-dry cycles slightly deteriorated with the addition of biochar. Based on these observations, a conceptual model showing biochar-AAS-sand interactions was proposed, in which biochar served as an internal curing agent, micro-reinforcer, and mechanically weak point. Secondly, a state-of-the-art deep-learning algorithm, Mask R-CNN, was utilized for the clayey soil crack detection, locating and segmentation. A comprehensive dataset including 1200 annotated crack images of 256×256 resolution was prepared for the algorithm training and validation. The proposed Mask R-CNN algorithm achieved precision, recall and F1 score of 73.29%, 82.76% and 77.74%, respectively. Besides, the algorithm gained a mean locating accuracy (APbb) of 64.14% and a mean segmentation accuracy (APm) of 47.59%. The detection performance of the Mask R-CNN was also compared with the U-Net in three different scenarios. The test results have demonstrated the superiority of the Mask R-CNN over the U-Net algorithm in crack detection, locating and segmentation and the algorithm could automatically process the crack characterization. Then, this dissertation proposed a state-of-the-art deep learning-based approach, Mask R-CNN, to locate, classify, and segment large marine microplastic particles (fiber, fragment, pellet, and rod). The fully trained Mask R-CNN algorithm was compared with U-Net in characterizing microplastics against various backgrounds. The results showed that the algorithm could achieve Precision=93.30%, Recall=95.40%, F1 score=94.34%, APbb=92.7%, and APm = 82.6% in a 250 images dataset with white background. The algorithm could also achieve a processing speed of 12.5 FPS. The results obtained in this study implied that the Mask R-CNN algorithm is a promising microplastics characterization method that can be potentially used in the future for large-scale surveys. Finally, a video instance segmentation algorithm was trained to locate, identify, and segment soil cracks in a real-time video stream. The algorithm could record the cracks' locations and numbers simultaneously. Besides, the crack ratio of clay could be calculated by crack pixels divided by total clay pixels among the entire soil cracking process. Furthermore, Structure from Motion (SfM) has been applied to reconstruct the 3D soil desiccation models. The soil crack can be detected in a 3D point cloud graph and highlighted. A series of 3D parameters like depth, volume, and cross-section profile can be obtained for future analysis. The proposed video instance segmentation method has demonstrated the potential application for real-time crack alerts and monitoring of geotechnical infrastructures via surveillance cameras.
Sandy Soil Stabilization by Microbial Induced Calcite Precipitation (MICP) using Bio-stimulation Method
(University of Hawaii at Manoa, 2022) Wang, Yijie; Jiang, Ningjun; Civil Engineering
Calcareous sands are often considered as very unstable and evolving materials with large particles allowing an abundance of air space. Their physical properties are usually defined as weak or no structure, free draining with high permeability, poor water retention, and high sensitivity to compaction. In tropic areas, the physical properties of sandy soils are more complicated. A more frequent cycling of wetting and drying can affect the physical properties significantly. These adverse engineering properties make it difficult to manage in the construction field. Chemical stabilization has been applied in the field for years. Frequently used stabilizers include Portland cement, lime and industrial waste lime, asphalt, and others. However, the major problem of these tradition materials is highly dependent on mass industrial production which requires substantial energy. Some potential environmental issues can be induced due to the high alkaline level of many of these chemical stabilizers. Considering the abovementioned limitations of those traditional additives, alternative technique is required to make the sandy soil stabilization more economic and environmental-friendly. Microbiologically Induced Calcite Precipitation (MICP) as a new interdisciplinary method combines the microbiological, geochemical, and geotechnical research to fulfill the increasing demands for the sandy soil improvement. The stabilization process can be achieved by inducing calcite precipitation within the sand matrix via microbial ureolysis procedure by ureolytic bacteria. The produced calcite precipitation preferentially accumulates at particle-particle contacts, which can provide extra connection and gain more strength. Because of the widely distribution of ureolytic microbes in the natural soil, the in-situ bio-stimulation of indigenous ureolytic bacteria becomes feasible. The study in this thesis firstly investigated the effect of enrichment media on the stimulation of native ureolytic bacteria in calcareous sand under solution condition. A series of batch tests were conducted, and the generic and three different selective enrichment media were compared from the biochemical aspects. The results show that the selective enrichment media rich in urea could successfully stimulate and enrich the native ureolytic bacteria by monitoring the ureolytic activity, pH, and electric conductivity. The nutrients with higher nitrogen sources show better efficiency in improving ureolytic activity. Secondly, the biochemical and direct shear behaviors of bio-cemented sand treated by bio-stimulated MICP were investigated under sand column condition. It was found that during the enrichment phase, the indigenous ureolytic bacteria can be enriched significantly within 48 hours, though the ureolysis rate was varying with the initial urea concentration. The microbial community changed significantly after enrichment stage. The ureolytic species could be found in the sand. Furthermore, the cementation content increased with the flushing number of cementation solution. The shear strength of bio-cemented sand could be significantly enhanced after MICP treatment. The cohesion and peak friction angle increased with elevated cementation level while declined with the increasing normal stress. Then, the compressive characteristics of bio-cemented sand was studied under one-dimensional compression tests. It was found that the compressibility of bio-cemented reduced with the increasing cementation content. The samples prepared under higher initial relative density showed less compression. Based on microscopic observations, a conceptual framework based on the interparticle contact modes and their corresponding damage modes during the compression tests was proposed. After that, to find out a way to preserve the enriched urease within soils as much as possible, and further increase the bio-cementation content, the biochar-amendment was incorporated into MICP. The shear behavior of biochar-amended bio-cemented calcareous sand treated via bio-stimulation was investigated through a series of direct shear test. The results showed that the addition of biochar powders could effectively increase the cementation content as the extra nucleation sites for native ureolytic bacteria. However, too much biochar within bio-cementation may become weak points and thus diminish the contribution of interparticle bonding to the shear strength. Finally, a preliminary exploration of bio-stimulated MICP in rainfall-induced erosion prevention was conducted by a series of laboratory model tests. The artificial sandy slope was made and subjected to the rainfall. Meanwhile, the photography-based SfM (Structure from Motion) as the new technique was firstly applied to evaluate the slope deformation under rainfall-induced erosion. The results showed that bio-stimulated MICP could significantly reduce the erosion on a sandy slope, especially in case treated by YE-based enrichment medium with 170 initial urea concentration. It is a good trial to explore a feasible monitoring way for the future full-scale application in field.
Diversity Of Fecal Pathogens In Hawaii’s Coastal Waters And Detection With Improved Performance By Next Generation Sequencing
(University of Hawaii at Manoa, 2020) Saingam, Prakit; Yan, Tao; Civil Engineering
Among all pollutants in coastal waters, fecal pathogens have caused a lot of concerns due to their disease transmission risks. An investigation on the impact of pollution sources could provide guidelines for us to protect coastal water quality. However, the use of fecal indicator bacteria (FIB) in monitoring the presence of fecal pathogens can be misleading because of their prevalence in the environment. In the meantime, our understanding about the impact of storm water sources is still limited. While direct detection of pathogen has been an alternative monitoring system, the detection methods need to be improved. In order to address the above issues, I conducted four research projects, which will be shown and discussed in this dissertation. In Chapter Two, the use of FIB in monitoring the quality of tropical urban marine estuary waters was investigated. The investigation included distribution of three FIB and four fecal pathogens at Ala Wai Canal (Oahu, Hawaii). The results showed that there was no correlation detected among the three FIB. Also, most of the tested typical fecal pathogens were not found. However, a pathogen and a marine microbe Vibrio parahaemolyticus was found prevalent but not correlated with the FIB. These findings indicated inconsistency between the FIB and fecal pathogens at a tropical urban marine estuary. In Chapter Three, the immediate impact of major storms such as hurricanes on coastal waters was researched. The investigation included the prevalence and genotypes of FIB enterococci in coastal waters at Hilo bay immediately after the Hurricane Lane. Total enterococci across the investigated sites were detected with levels meeting the standard criteria and gradually decreased over time. However, the variation of enterococci concentrations and genotype distribution were more distinctive among sites. The findings uncover the immediate impact of a hurricane on temporal and spatial variation of coastal water quality. In Chapter Four, the impact of hurricane on occurrence of pathogenic enterococci in Hilo bay water was examined. In this study, 49 enterococci isolates from Hilo bay were scrutinized for their species identification, virulence genes and antibiotic resistance susceptibility. Among the isolates, prevalence of E. faecalis and E. faecium/E. durans and their resistance to eight antibiotics were observed and four virulence genes were detected. This suggests the hurricane impact on the occurrence of pathogens and identified potential pathogens in coastal waters. In Chapter Five, a novel PCR-NGS method of detecting pathogens in environmental samples was explored. This included tests on sensitivity and robustness against water samples. It will be shown that NGS can improve sensitivity of PCR and qPCR-based detection of Salmonella. Also, the PCR-NGS indicated that Salmonella can be successfully detected from coastal stream samples. These results demonstrated that PCR-NGS is a promising method of pathogen detection in environmental samples. Overall, this dissertation sheds light on how we can enhance coastal water quality monitoring. The following topics were investigated and discussed: (i) evidences of misleading use of FIB in monitoring tropical urban marine estuary waters, (ii) impact of storm water on coastal water quality and occurrence of pathogen, and (iii) a novel approach of pathogen detection essential for the development of the pathogen-based monitoring system.
Design of alkali-activated materials based on quantitative microanalysis of precursors and reaction kinetics
(University of Hawaii at Manoa, 2020) Mirmoghtadaei, Reza; Shen, Lin; Civil Engineering
Alkali-activated materials (AAMs) provide a new virtuous and green solution to the employment of waste materials, avoiding their harmful impacts on environment and ecology. Compared with conventional concrete mixes with cement, the composition and impurities of precursors, as well as the type and concentrations of alkalis should be carefully controlled in order to satisfy fresh properties, hardened properties, and durability simultaneously. Moreover, in order to optimize the mix design, it is essential to employ a reliable, accurate, and user-friendly approach to investigate precursors’ compositions as well as the reaction kinetics. Geopolymerization occurs through chemical reactions of powders containing aluminosilicate oxides with alkalis leading to the formation of Si-O-Al bonds. These three-dimensional networks have amorphous to semi-crystalline silicoaluminate structures. Material scientists classify any binder system obtained from the reaction of alkali sources with solid silicate powders as AAMs. The precursor can be calcium silicate as in the hydration of clinkers, or aluminosilicate-rich powders like blast furnace slag (BFS), fly ash, natural pozzolan, or bottom ash. Moreover, the alkali can be any soluble materials, which increase the pH in the mix and accelerate the dissolution process. To study the kinetic of alkali-activation and design of AAMs, which is robust regardless of type of precursors and alkalis, it is critical to be able to quantify different phases of precursors quickly, monitor the process of alkali-activation, and design AAMs with a reliable and easy method. In this study, a quantitative phase analysis by Raman Spectroscopy is introduced as a simple, fast, and reliable method for analyzing precursors and monitoring the kinetic of reactions. On the other hand, a new method based on the initial pH of concrete mixes was developed to design AAMs. It was found that Raman Spectroscopy is capable of successfully quantifying critical compositions and impurities of precursors. Different mixing procedures have been studied on precursors with high percentages of impurities in their compositions. Undesired elements in precursors could be sulfur, chloride, and unburnt carbon. It was found that applying alkali activators separately from silicate activators could significantly enhance the fresh and hardened properties of AAMs made from precursors with high impurities. Finally, the effects of different levels of initial pH on the mechanical properties and final products of AAMs were investigated. It was found that initial pH lower than 12 led to unstable soluble silica, a mixture of different crystals, and lower compressive strengths. It was also concluded that in the presence of optimal initial pH ranged between 12.2 and 12.4 the best selection for sodium silicate solution would be the products with a SiO2/Na2O ratio of 2.1.

Browse

Recent Submissions