Congress on Technical Advancement 2017: Cold Regions Engineering

Thermally induced stresses in composite steel-concrete bridges are higher than those experienced by their concrete and steel cousins, leading to significant damage in the concrete deck and corrosion of the steel reinforcement. Bridge design engineers use thermal profiles prescribed by codes such as AASHTO to predict future service stresses. A 3D finite element model is presented that investigates the temperature distribution in a case study bridge with pre-existing construction deck cracks. The non-linear transient simulation is performed using actual environmental loads for a geographic region with severe climate (North Dakota), and the resulting profile is compared to that of AASHTO. The results show the thermal gradient proposed by AASHTO to be overly conservative in cold regions. Existing models seem to ignore the nonlinearity of the thermal gradient, which can be critical for thermal stress calculations. The pre-service deck cracks appear to have a considerable effect on both the vertical and the longitudinal temperature distributions, and it is recommended that they be given careful consideration by design codes.

The United States (U.S.) electric grid is increasingly incorporating renewable energy resources to offset base and peak loads; renewables now account for 15% of total energy generation and nearly 20% of electric generating capacity. Geothermal energy resources are available across the U.S., and many are co-located within communities. Geo-exchange (using ground and water-source heat pumps) can be accomplished in any state in the country. Geothermal energy is optimal for projects in colder regions for several reasons: it can offset both electricity and space heating energy use, it is a consistent source that does not vary with time of day or season, it is not affected by weather, and it has the highest capacity factor of any renewable technology. Increased utilization of geothermal energy resources would benefit homes and businesses, especially for regions in the northern U.S. that have high heating demands.

Previous studies by others had proposed that corrugating porous surfaces would improve their acoustic properties. To investigate the validity of that lemma, a concrete test section built in 2008 specifically as a 4 inch thick pervious overlay on an 8 inch thick concrete pavement in the low volume test track of the MnROAD research facility was diamond ground in 2013. This section has been monitored seasonally for ride, friction, noise sound absorption, and nuclear density before and after grinding. In this paper, results of performance tests performed prior to the diamond grinding are compared to the results of performance test performed after the grinding. Summarily, remarkable improvements were obtained in ride quality and texture orientation. Significant changes occurred in the tire pavement noise of the inside lane that had been subjected to traffic loads and thus exhibited high pre-grind noise while the outside lane that was exposed only to environmental factors showed insignificant acoustic improvement. It was evident that whereas grinding improved ride quality in pervious concrete, it provided only marginal acoustic benefit in sound absorption and On Board Sound Intensity. The lemma of improvement in acoustic properties of porous surfaces due to corrugation was validated in the deteriorated pervious surface and only marginally in the non-deteriorated surface.

The coefficient of thermal contraction/expansion (CTE) of aggregates may influence mixture CTE and low temperature performance since aggregates account for about 85% of mixture volume. However, due to difficulties in measuring the CTE of aggregate and mixtures, there have not been many studies on the effect of aggregates on mixture low temperature performance. The existing CTE predictive equation and models are based on the ratio of the bulk volume and CTE of aggregates and binder ignoring the potential disproportionate contribution of coarse and fine aggregates due to their spatial arrangement. The purpose of this study was to investigate the thermal contraction characteristics of coarse and fine aggregates and their effects on the low temperature performance of asphalt pavements. Two limestone aggregates with a high and a low CTE were selected for this study. Dense graded asphalt mixtures were prepared with these two aggregates and cross-blends of the aggregates for which the coarse fraction of low and high CTE aggregates were blended with the fine fraction of high and low CTE aggregates. The CTE and cracking temperature of the mixtures were determined using the Ohio CTE device (OCD) and the asphalt concrete cracking device (ACCD), respectively. The coarse aggregates had a significantly greater influence on mixture CTE and low temperature performance as compared to the fine aggregates. The average measured CTE difference between low and high CTE aggregate mixtures was found to be lower than the difference obtained from CTE predictive equation. Measured mixture CTEs also exhibited Glass Transition behavior.

Highway engineering in saline soils in cold regions is challenging for corrosion and ground heaving due to frost and salt expansion at low ambient temperature. This paper presents an innovative construction technique, namely the bagged reinforced concrete (RC) shaft, for deep foundation construction in saline soils. The key of this technique is to use an anti-corrosion geotextile container to isolate the RC shaft from the saline soil. The main processes in this new method are summarized, key parameters are presented, and quality control issues are discussed. While this method is promising for protecting deep foundations in a saline environment, issues such as the impact of the bag to vertical and horizontal loading capacity and resistance to ground heaving due to frost or salt expansion need to be further investigated.

In permafrost terrains, ground ice volume can vary widely, with gravimetric moisture content not uncommonly differing by one or two orders of magnitude within meters horizontally and vertically. Two types of electrical earth resistivity techniques were performed on warm-discontinuous permafrost terrain in Interior Alaska to study the level of accuracy and efficiency the techniques might provide to enhance geotechnical investigations. For validation we drilled 34 boreholes at 3m spacing along the survey transect, sampling vertically every 30cm to depths of approximately 7m. We collected sediment type information and 589 moisture (ice) content values, which were mapped, analyzed, and qualitatively compared to the resistivity images. The results indicate that electrical resistivity surveys, coupled with directed borehole drilling, can produce high resolution cross-sections of the permafrost subsurface. We believe the integration of inter-borehole mapping via rapid earth resistivity surveys can greatly aid design decisions by identifying the ultimate foundation or embankment possibilities early in the design process, and can provide optimal structure siting. The increased ground ice detail may facilitate innovative and cost effective foundation designs resilient to a changing climate.

Many studies have been conducted on frozen soil, and numerous models and theories on frost heaving have been published. However, microscopic observation of the internal structure of frozen soil with the conventional technology is difficult, and verification thereof has not been sufficient. Therefore, in this study, we developed a small frozen experimental apparatus that can be placed in an X-ray CT scanner. Since the X-ray CT system used in this study had a micro-focus X-ray generator and a high-sensitivity X-ray detector, the internal structure of soil in the frost heave process could be acquired as a three-dimensional X-ray image. Additionally, this frozen experimental apparatus could freeze the specimens one-dimensionally. The specimens were cooled from the lower surface and moisture was supplied from the upper surface. We used three types of soil samples with varying grain size distribution. Experimental results showed that the frost heave amount, ice lens generation time, and the thickness and shape of the ice lens varied according to the types of soil samples. A flat-shaped ice lens was formed in weathered volcanic ash not containing gravel. However, in terrace deposits including gravel, the ice lens was formed around large particles and had an uneven shape compared with that of weathered volcanic ash. We further observed that thin ice lenses were generated in volcanic ash sand. It was therefore possible to microscopically observe the dynamic internal structure of freezing soil by using the X-ray CT scanner and frost heave experimental apparatus developed in this study.

In 2016, NASA-affiliated researchers completed the 20th consecutive field season at Haughton Crater in Canada’s High Arctic. The Haughton-Mars Project (HMP) has established a base camp to provide essential services during summer field seasons. The crater and surrounding areas provide a high-fidelity terrestrial analog for planetary exploration, with similarities in geology, weather conditions and complex logistics. This paper provides a systematic review of HMP infrastructure, design research and future development priorities. Status of camp buildings, mechanical systems, energy inputs, and site impacts are discussed, as are opportunities for integration of systems to conserve resources and reduce future operational costs. Techniques for rapid site assessment are compared. Recommendations include further design and program development for modernization of camp facilities, alongside restoration of relationships with service providers to re-engage local community members. This will ensure availability of a well-established, high quality proving ground for technologies and operational concepts for NASA’s future planetary surface exploration missions.

In 2015 Applied Research Associates, Inc. (ARA) developed a virtual roadway weather information system (vRWIS) that provides roadway maintenance staff with both current and forecasted roadway icing and freeze-thaw conditions. This virtual system uses a computerized climatic model which has evolved over the past 40+ years and can accurately predict the temperature in pavement systems based on atmospheric weather data inputs and pavement materials. Historically, the enhanced integrated climatic model (EICM) has been used for pavement performance research and as a key component in pavement performance modeling and pavement design. The EICM is one of the key modules in the AASHTO Pavement-ME Design software, which is the commercial version of the Mechanistic Empirical Pavement Design Guide (MEPDG). This paper details the level of accuracy at which the default EICM input values for asphalt and concrete pavements predict pavement temperatures compared to the measured values from actual roadway weather information systems (RWIS) in Illinois. First, several vRWIS stations were created to represent the predicted model. The process for creating a vRWIS station consists of collecting pavement structural and material information and then modeling the pavements at these locations with actual atmospheric weather data. Next, the difference between actual and modeled pavement temperatures was evaluated. The analysis determined that the default values in the EICM can accurately predict pavement temperature data. The one-to-one comparison shows an R2 above 0.85 for all selected sites. An additional finding concludes that the predicted model for the vRWIS site is more accurate at colder temperatures. The accuracy at colder temperatures is important for the vRWIS icing and freeze-thaw predictions, allowing winter maintenance crews to effectively determine when and where to use valuable labor and material resources.

Based on the knowledge accumulated from surveys on winter construction done by the Civil Engineering Research Institute for Cold Region, the major reasons for the deformation of embankments constructed under cold weather conditions are thought to be the following: 1) work cessation necessitated by nighttime, 2) the inclusion of frost-susceptible materials, and 3) the inclusion of snow and frozen soil blocks in the materials. In this report, the authors addressed the mixing of frozen soil blocks in the material and investigated the characteristics of frozen soil as an embankment material by constructing some embankments using frozen soil and others using unfrozen soil. The construction work was done in summer and winter, and the strength and deformation of the embankments during the snowmelt season were investigated. The authors concluded that the embankments constructed using frozen soil undergo long-term subsidence during and after melting. While the strength of the embankment crown gradually increases with time, the strength inside the embankment does not increase with time.

Confronting snowfall at airports is a long-standing challenge for the aviation industry. To address these concerns, electrically conductive portland cement concrete (ECON) is currently a focus in pavement design. This method applies a potential difference to a surface or near-surface conductive concrete layer, heating the pavement to melt the snow. Due to the complex nature of performance measurements, most studies use experimental methods to examine the ability to remove accumulated snow. However, these methods are resource intensive. This study addresses this challenge through the development of an experimentally-validated 3D finite element (FE) model of ECON for evaluating its thermal performance. Predicted surface temperatures of the FE model are consistent with the data, indicating that a FE model is a promising method for feasibility studies and control strategy development.

The study presented herein examines frost resistance of different types of high early-age strength concretes made for rapid repair including Type III, Type V, and calcium sulfoaluminate (CSA) cement concretes. To this aim, high early-age strength concretes were made with different cement contents, including 386, 445, 504, and 564 kg/m³ (650, 750, 850, and 950 lb/yd³) for Type V cement concretes; 326, 386, and 445 kg/m³ (550, 650, and 750 lb/yd³) for Type III cement concretes; and 326, and 386 kg/m³ (550 and 650 lb/yd³) for CSA cement concretes. This study also compares frost resistance of different types of high early-age strength concrete for a common cement factor of 386 kg/m³. The results of this study showed that while CSA cement concretes had the fastest strength developments (shortest opening time which is the time to reach a minimum compressive strength of 20 MPa), they had considerably weak frost resistance. Type III and Type V cement concretes exhibited significantly superior frost resistance when compared to CSA cement concrete. The frost resistance reduced as cement content was increased and improved as curing age was extended.

This paper investigates use of natural pozzolan as a supplementary cementitious material to replace a portion of Portland cement in self-consolidating concrete (SCC). It evaluates effects of natural pozzolan on compressive strength, water permeability, and frost resistance of SCC. For this purpose, self-consolidating concretes with natural pozzolan replacement ratios of 0, 15, 22.5, 30, and 37.5% by mass of Portland cement were prepared. These concretes were made with a constant water-to-cementitious material ratio of 0.4 and cementitious material content of 475 kg/m³. In addition to the fresh properties, compressive strength, water permeability, and mass loss and failure cycles were measured. The results of this investigation showed that the frost resistance of natural Pozzolan contained SCCs was superior to that of SCC made with purely Portland cement. They also had lower water penetration depth than SCC made without natural pozzolan. On the other hand, slight reductions were observed in early and mid-aged strength properties when natural pozzolan was included in SCC mixtures.

Liquefaction, due to strong ground motions caused by earthquakes, is typically disregarded in permafrost soils. However, in cases where permafrost degrades, it can be challenging to estimate liquefaction potential using conventional screening methods. Historically, there have been several tools engineers have utilized to estimate the potential for liquefaction in thawing permafrost soils, such as those used for the Trans-Alaska Pipeline System (TAPS) in the 1970s. Recently more than one hundred boreholes were advanced in an area of discontinuous permafrost near Fairbanks, Alaska. The area is partially developed, which has resulted in decades of localized permafrost degradation. While the boreholes in the undeveloped area remained in a warm permafrost condition, others within close proximity showed thawed soils to depths greater than 50 feet. This study utilizes the close proximity of the frozen and thawed soils to evaluate historical liquefaction screening methods of frozen, thaw-susceptible soils and compares those methods to the results of current methods for screening of thawed soils for liquefaction. The findings from this study provide a methodology that can be further refined and used and identifies key issues to consider when evaluating liquefaction susceptibility at sites where climate change or development adversely affects permafrost soils.

Thermosyphons are widely used passive cooling devices that rely on seasonally cold air to provide subgrade refrigeration to maintain permafrost. With the Arctic impacted by warming due to climate change, the quantity of cold air available to drive the thermosyphons is reduced and consequently, their cooling effect is reduced. A method to determine the design air freezing index for thermosyphon systems is presented and the results used in an example design for a subgrade cooling system beneath a slab-on-grade structure using flat-loop evaporator type thermosyphons. Additionally, an older design from 30 years ago is analyzed with the latest climate data to determine if it is adequate or if it requires an upgrade. Upgrades are discussed for older thermosyphon systems.

Landing wheeled aircraft on snow runways is uncommon and minimally documented. This report describes the modeling used to evaluate design and construction of the new compacted-snow Phoenix runway in Antarctica for the first wheeled C17 aircraft landing. Snow density from the target design and snow density of the as-built Phoenix runway structure were used to determine basic elastic parameters for use in layered elastic analysis formulation (LEAF). LEAF is part of a software package developed by the Federal Aviation Administration (FAA) that allows for forward calculation of runway stress, strain, displacement, and associated principal stress and strain based on design aircraft loading. Stress responses for the Phoenix runway were modeled for the C17, A319, and B757 aircraft. Two construction vehicles were also modeled for comparison. Results show the comparison of the stress profiles and effect of subgrade stiffness on stresses in the layers just above the subgrade. This is the first time the model was used for a snow runway and it provided valuable insight for design, construction and runway performance for the first landing of the C17 on compacted snow. The model is flexible enough for simulating additional aircraft that might use the runway in the future. The successful construction and use of the new compacted-snow Phoenix runway is significant and demonstrates the heaviest wheeled aircraft (C17) operating on a compacted snow runway, ever.

Structural engineers typically detail a concrete structure as they have always done in the past, often based on a limited understanding of thermal behavior, with the intention of structural and fabrication efficiency. However, with renewed focus on energy consumption, the desire for thermally efficient buildings is increasing. Poor connection details can create significant thermal bridging, which limits thermal efficiency and can even cause premature degradation through condensation. Thermal bridging occurs when heat transfer is given a path through a more conductive material like concrete or steel rather than insulation, resulting in increased heat loss. By nature, concrete sandwich wall panels tend to be highly efficient at preventing heat loss in the middle of the panel, with greatest heat loss occurring at the connections. This project partnered with tilt-up and precast contractors and engineers to perform infrared thermographic inspections of concrete sandwich wall panel structures for window, door, roof, foundation, and insulation termination details currently in use. This study investigates heat loss of concrete sandwich wall panel structures to identify where thermally efficient details are needed for use in future design of concrete sandwich wall panels to reduce energy use and heat loss, increase sustainability, and lessen the environmental impact of concrete sandwich wall panel structures.

To investigate the performance of steel with a heavy-duty coating against sea ice friction as an anticorrosive measure, a sliding wear test and an exposure test for the material in an ice-covered area were performed to reveal the characteristics of wear mechanisms and the wear rates, and to give some suggestions from a practical point of view. In the sliding wear test, ice caused severe adhesive wear of the resin coating and a wear rate one order of magnitude greater than the abrasive wear rate of carbon steel; however, ice containing particles caused even more severe abrasive wear, with a rate that was a few times greater than the adhesive wear rate. From the exposure test, it was confirmed that the separation of the coating progressed from the edge to reach greater deterioration in the fourth year. It was concluded that the application of such a heavy-duty coating, which has been used under normal sea areas, to ice-infested waters would be inappropriate, otherwise it requires special care and proper management for its application in ice-infested waters where active sea ice movement is seen.

Damage has been reported to coastal structures and training levees made of steel sheet-pile-type seawalls in the Sea of Okhotsk facing Hokkaido, Japan. This damage has been caused by various effects such as wear, abrasion, and deformation of steel sheet piles resulting from the action of sea ice. If rust layers due to corrosion of steel surfaces are removed by the action of sea ice in the wintertime, the corrosion amount will increase further during ice-free periods due to activation of these surfaces. Consequently, this process of corrosive wear leads to larger wear of steel structures in ice-infested sea than that in ice-free sea. We consider sacrificial steel walls with/without concrete lining in addition to the base steel structure as a useful countermeasure to sea ice action. A method for estimating the required thickness of a sacrificial steel wall is proposed. From a sliding wear test in a cold room, it is also shown that the effect of sand inclusion in sea ice on the wear rate of steel in the mild (steady) wear region is crucial.

ASCE-7 ground snow load requirements remain largely unspecified for large portions of the topographically complex western United States. Many western states have addressed this issue by independently developing ground snow load estimation methods most often described in state snow load reports. While these reports justify their methodologies, little has been done to rigorously evaluate the accuracy of such methods. The lack of accuracy comparisons between methods makes it difficult to reconcile the snow load requirement discrepancies that currently exist between state boundaries. This paper considers three methods for predicting ground snow loads in Utah: the current Utah ground snow load equations, Idaho’s normalized ground snow loads based on inverse distance weighting, and an adaptation of PRISM which models the log-linear relationship between elevation and ground snow load. The accuracy of these methods are compared by means of 10-fold cross validation on a recently developed Utah snow load dataset. It is concluded that PRISM achieves the best accuracy on this dataset, with a mean absolute error approximately 40% less than the other two methods. The improvements in prediction accuracy offered by PRISM demonstrate the feasibility of creating multi-state ground snow load prediction methods, ultimately eliminating the snow load requirement discrepancies between states.