Claudia E. Zapata

Resilient Modulus for Unsaturated Unbound Materials

ABSTRACT The suitability of the current resilient modulus test protocol (NCHRP 1-28A) for its application to unsaturated soils was assessed. Modifications to the stress state conditions of the protocol are necessary due to the axis-translation needed during the test when measuring matrix suction. This study presents the modulus of unbound materials resulting from tests performed under unsaturated soil conditions. Two different materials were tested. The base material was tested under drained and undrained boundary conditions, while the subgrade was tested under drained boundary condition. The results allowed for the enhancement of the Universal Model for resilient modulus prediction by incorporating suction as a stress state. This model predicts the resilient response of unbound materials as a function of external stress state and matrix suction levels and therefore, it is independent of moisture variation.

Geotechnical engineering practice for collapsible soils

Conditions in arid and semi-arid climates favor the formation of the most problematic collapsible soils. The mechanisms that account for almost all naturally occurring collapsible soil deposits are debris flows, rapid alluvial depositions, and wind-blown deposits (loess). Collapsible soils are moisture sensitive in that increase in moisture content is the primary triggering mechanism for the volume reduction of these soils. One result of urbanization in arid regions is an increase in soil moisture content. Therefore, the impact of development-induced changes in surface and groundwater regimes on the engineering performance of moisture sensitive arid soils, including collapsible soils, becomes a critical issue for continued sustainable population expansion into arid regions.In practicing collapsible soils engineering, geotechnical engineers are faced with (1) identification and characterization of collapsible soil sites, (2) estimation of the extent and degree of wetting, (3) estimation of collapse strains and collapse settlements, and (4) selection of design/mitigation alternatives. Estimation of the extent and degree of wetting is the most difficult of these tasks, followed by selection of the best mitigation alternative.

Prediction of the soil-water characteristic curve based on grain-size-distribution and index properties

The grain-size-distribution (GSD) of a soil is intimately related to its pore size distribution and hence, the GSD holds a close relation with the soil-water characteristic curve (SWCC). In addition, the plasticity index (PI) is a measure of the water holding capacity of the soil and therefore, it plays an important role in shaping the SWCC. This paper presents two sets of statistically derived equations that describe the SWCC of non-plastic and plastic soils. Data from 154 non-plastic soils and 63 plastic soils were analyzed. Soil samples were collected as part of the National Cooperative Highway Research Program (NCHRP) 9–23 project entitled Environmental Effects in Pavement Mix and Structural Design Systems . Samples were obtained from underneath paved roads of 30 sites located throughout the United States. The soil samples were subjected to laboratory testing that included index testing and SWCC testing. SWCCs were determined using a newly developed pressure plate device capable of overburden pressure application, continuous measurements of moisture content, and volume change monitoring. In addition to the collected field data, a database of published soil index properties and SWCCs was incorporated to the analysis. Each SWCC data set was fitted with Fredlund and Xing curve, which provided an S-shaped curve with four parameters, a f , b f , c f , and h rf . Using multiple regression analysis, equations were derived for these four parameters based on predictors derived from GSD and PI. The equations presented in this paper are useful in predicting the SWCC of any given soil without carrying out actual SWCC testing and they can easily be incorporated into computer codes to solve various unsaturated soil mechanics problems such as determining moisture beneath covered areas.

Enhanced Model for Resilient Response of Soils Resulting from Seasonal Changes as Implemented in "Mechanistic–Empirical Pavement Design Guide"

The present study deals with the revision of the current model in the Mechanistic–Empirical Pavement Design Guide (MEPDG) used to predict the environmental factor for unfrozen unbound materials (FU), which is used to adjust the resilient response of soils resulting from seasonal changes. A large database with data from the existing literature and studies at Arizona State University was developed to evaluate the model. The results suggest that the environmental factor is underestimated for fine-grained materials with high plasticity under dry (arid) conditions. However, insufficient data were available to enhance the FU models for wetter conditions. Three fundamental factors that may have impacts on the FU values were evaluated in this study: stress state, compaction energy (soil density), and soil type. The stress state was found to have little to no impact on the predictions of FU. But density changes and soil type were found to be important. The potential for soil index properties to be predictive variables was assessed. Models dependent on enhanced moisture content accounting for the effect of soil type are proposed for nonplastic and plastic materials. The range of predicted FU values is in close agreement with the actual measured FU values found from laboratory studies. It is recommended that the new models be adopted in the revision of the MEPDG model for the drier conditions described in this report and that research be conducted to enhance the FU approach in the current MEPDG for wetter conditions brought on for a variety of reasons (e.g., groundwater table change, increased rainfall, and frost effects).

Necessary Assessment of Use of State Pavement Management System Data in Mechanistic–Empirical Pavement Design Guide Calibration Process:

Calibration and validation of the Mechanistic–Empirical Pavement Design Guide (MEPDG) for state department of transportation (DOT) networks requires detailed information for a variety of pavement inputs. The biggest database resource that most states possess is their pavement management system (PMS). It is understandable for any DOT to maximize the use of this resource to the greatest extent possible. However, before any state uses PMS data in MEPDG calibration, it is important that a careful comparison be made of the equality of the measurements, data, and so forth in the PMS database with national Long-Term Pavement Performance (LTPP) values. The purpose of this paper is to review possible differences between several key data found in a state PMS database and the LTPP database. Arizona Department of Transportation (ADOT) PMS data are used for this purpose. The specific variables examined in this paper deal with distress type (rutting, cracking, and roughness) and nondestructive deflection testing (NDT) backcalculated moduli to predict in situ pavement layer properties. It was concluded that significant differences did exist between the ADOT PMS values and LTPP measurements. Differences were found between ADOTs NDT measurements and LTPP data in rut measurements, asphalt cracking, international roughness index, and all layer backcalculated moduli. Consequently, ADOT may incur significant expense to calibrate the MEPDG to Arizona conditions. The paper discusses several possible reasons for differences between the PMS data of a specific agency and those in the LTPP database.

Matric Suction Prediction Model in New AASHTO Mechanistic-Empirical Pavement Design Guide

Equilibrium moisture beneath highway pavements is critical to pavement design because it directly affects the strength and stiffness of pavement systems. Moisture is related to soil suction by means of the soil water characteristic curve (SWCC). Previous research has indicated a correlation of suction with Thornthwaites moisture index and soil type; however, these suction correlations exhibit large variability. Under NCHRP Project 9-23, Environmental Effects in Pavement Mix and Structural Design, sponsored by FHWA, soil samples were collected from beneath 30 pavement sections throughout the United States. The SWCCs and index properties of collected samples were measured at Arizona State University. The in situ degree of saturation was obtained and the corresponding in situ soil suction was found from measured SWCCs. On the basis of the field and laboratory data, two models were developed to predict equilibrium suction under pavements: one for granular non-plastic materials and another for fine-grained plastic materials. These models were adopted in the new AASHTO Mechanistic–Empirical Pavement Design Guide because they exhibited good results, with variability within acceptable limits. Model development is presented.

Study of Expansive Soils and Residential Foundations on Expansive Soils in Arizona

Construction on expansive soils is challenging and thus prone to some problems and litigation. The engineering community makes extensive use of local experience and empirical procedures to address these problems. Although there has been extensive study of expansive soils and foundations on expansive soils, data related to performance of residential structures are limited in general and limited in the Phoenix area, in particular. In this study, an overview of the Phoenix Valley, Arizona, geotechnical practice and foundation performance related to residential structures on expansive clays, was developed through surveys and interviews with geotechnical engineers, structural engineers, and homebuilders. Using data obtained from files of Phoenix area geotechnical firms and government agencies, the existing Natural Resource Conservation Service map showing expansive soil locations throughout the Phoenix region was updated through the use of correlation developed in this study relating expansion index to common soil index properties such as Atterberg limits and percent passing the No. 200 sieve. Files of forensic investigations linked to expansive soil regions were made available for this study by several geotechnical engineering firms, and Phoenix Valley areas where forensic investigations have been identified, were mapped for comparison to regions identified in the updated map as having expansive soils. Comparison of the forensic investigation map to the updated map of expansive clay locations revealed that most of the forensic investigations were in regions identified with clays labeled as high to moderately high expansion potential, with a few forensic investigations in regions of medium expansion potential. Finally, unsaturated flow analyses were conducted for an Arizona expansive clay profile for two very different landscaped conditions of well-irrigated turf and desert landscape. The results of the numerical analyses were consistent with the reported observations and modes of failure identified through the surveys and interviews conducted with engineering and homebuilder professionals, including the finding that site drainage was found to be extremely important to good foundation performance, regardless of the type of landscape selected.

SWRC Modelling Framework for Evaluating Volume Change Behavior of Expansive Soils

The soil-water Retention curve (SWRC) has been used as a tool by geotechnical researchers and practitioners to determine the properties such as the shear strength, coefficient of permeability, bearing capacity and the modulus of elasticity of unsaturated soils. Such studies are valuable to the practicing engineers as they alleviate the use of time consuming and elaborate testing techniques required for determining the unsaturated soil properties. In this paper, an attempt is made to study the relationship between the SWRC and the swelling pressure including the one dimensional swell behavior for two different expansive soils using statically compacted specimens. Test results were analyzed to evaluate the clay mineralogy and influence of initial compaction moisture content or matric suction on both swell strain and swell pressure properties in one-dimensional test conditions. Comparisons of swell pressures and swell strain potentials and related SWRC properties of these soils are also made.

LONG-TERM MOISTURE CONDITIONS UNDER HIGHWAY PAVEMENTS

Equilibrium moisture beneath highway pavements is critical to pavement design because moisture directly affects the strength and stiffness of pavement systems. Moisture is related to soil suction by means of the soil-water characteristic curve (SWCC). Previous research has indicated a correlation of suction with Thornthwaite Moisture Index (TMI) and soil type; however, these suction correlations exhibited large variability. Under an NCHRP project sponsored by the Federal Highway Administration, soil samples were collected from beneath thirty pavement sections throughout the United States: two from the WesTrack test facility, one from the MNRoad Project, and twenty-seven from the Long Term Performance sites. SWCCs and index properties were measured on collected samples at Arizona State University. The in-situ degree of saturation was obtained from soil index properties, dry unit weight, and moisture content, and the corresponding in-situ soil suction was obtained from SWCCs. Based on the field and laboratory data, an algorithm was developed to predict suction under the pavement using TMI, percent passing 200, and Plasticity Index.

Model for Seasonal Variation of Resilient Modulus in Silty Sand Subgrade Soil: Evaluation with Falling Weight Deflectometer

One of the main input parameters in the mechanistic design and analysis of pavement systems, the stiffness of unbound pavement material usually is moisture dependent. As a result, most unbound pavement layers exhibit seasonal variations in stiffness as the pavement moisture content changes over the year. Any realistic pavement design should take this variation into account. In unbound material with a high fines content, a change in moisture content can change the stress state because of suction effects. An enhanced predictive resilient modulus model is presented to account for seasonal variation by means of suction measurement. A silty sand subgrade soil was tested with a modified repeated load triaxial system under different moisture (suction) conditions, and a set of resilient modulus model regression parameters was determined. The capability of the model to capture seasonal variation in moisture content was evaluated further with field data. A series of falling weight deflectometer (FWD) tests with multilevel loads was conducted on an instrumented pavement structure in which the moisture content of the subgrade was changed by manipulating the pavement drainage condition. The resilient modulus values obtained from the model were compared with backcalculated stiffness data obtained from FWD tests conducted under different moisture conditions. Overall, agreement was good between the laboratory-based resilient modulus values and the backcalculated stiffness. The resilient modulus–suction model could efficiently capture the moisture content effects.