Search

A systematic research program was undertaken with the objective of developing quantitative geotechnical parameters to support soil-pipe interaction assessment for buried pipelines in muskeg. For this purpose, a field geotechnical investigation program comprising cone penetration testing (SCPT) with shear wave velocity (Vs) measurements, electronic field vane shear testing (eVST), full-flow ball penetration testing (BPT), and pressuremeter testing (PMT), along with fixed-piston tube soil sampling was undertaken in a muskeg soil terrain. The data from field testing were initially interpreted to obtain typical stiffness and strength parameters for the subject soils. These parameters were then used to numerically simulate pressuremeter tests and the results were compared with those obtained from field pressuremeter testing; the intent was to calibrate a suitable constitutive model to represent the muskeg soil mass. These ascalibrated constitutive model was then applied on numerical models developed to simulate buried pipelines in muskeg soil subject to relative lateral ground movements. The work is aimed at developing a framework to generate soil restraint versus relative ground displacement relations (“soil springs”) to assess soil-pipe interaction of pipelines buried in muskeg soils. Initial results from the research are presented herein, with a comparison made between soil springs developed from numerical analyses and those generated from current practice guidelines.

A highway project of up to 100 km/h is currently being constructed between Colombo and Katunayake International Airport across a Sri Lankan muskeg area. At this site, peat deposit was initially 0.8∼15.3 m thick and was underlain by sand, clay, or gneiss. The ground improvement methods adopted in the project were combined drainage consolidation-preloading methods, pipe pile foundation, and geogrids. This paper provides a detailed insight into the implementation of combined drainage consolidation-preloading methods used in the project, including sand pile, gravel pile, and plastic drainage plate as the prefabricated vertical drains. Periodical field-level observations were taken during the ten years, including the construction and postconstruction periods. The results show that peat soils’ consolidation coefficient has been increased several times to tens of times due to ground improvement. After removing the temporary surcharge, the highway embankments did not heave and was followed by long-term settlements totaling 1.3∼7.4 cm over the following seven years of observations. Analysis of the settlement records shows that combined drainage consolidation-preloading methods have helped accelerate drainage consolidation and reduce postconstruction settlement.

This study interprets deformations of indurated 10s of metres thick bedrock strata by subglacial meltwater pressures at maximum levels, resulting in types of hydrodynamic structures not previously recognized. The structures provide insights into the range of extreme pressures possible with the backup of meltwaters sufficient to deform multi-metre thick indurated beds, unlike elsewhere. Subglacial meltwater flows into the subsurface below the 50–100 m thick bitumen platform aquiclude of the Cretaceous Athabasca Oil Sands deposit were driven by the hydraulic head of the 1.5-km-thick Laurentide Ice Sheet. These meltwaters over-pressured the regional Devonian aquifer waters, but the low permeability of the aquifer below the site of the Muskeg River Mine was insufficient to accommodate the voluminous influxes of subglacial meltwater. The resulting meltwater flows backed-up, resulting in elevating aquifer pressures to maximum levels along the margins of the underlying Devonian Keg River reef mound and within water-saturated Cretaceous sand beds offset to the east. The meltwater pressure build to extreme levels hydrofractured strata at sites along the margins of the Keg River mound preconditioned by dissolution-induced subsidence in underlying salt beds. Release of the confining pressure upon withdrawal of the Laurentide Ice Sheet resulted in 45-m-high open blowout structures that punctured the bitumen platform at sites above the northern margin of the Keg River mound. Other blowout chimney sites above the southern margin of the mound were plugged because of insufficient pressure build to clear vents of all ejecta. Concurrently, pressured meltwater flows along the eastern margin resulted in dykes that cross-cut and buckled Devonian limestone beds. Some dykes extended into overlying water-saturated unconsolidated Cretaceous sand beds, channelling pressured water and hydroplastic mudflows that dissipated the extreme meltwater flow pressures and prevented puncturing of the overlying bitumen platform. These unusual deformation structures in western Canada resulted from catastrophic failure of hydrofractured rock zones responding to extreme meltwater pressures, in contrast to unconsolidated sediment deformations resulting from only moderately elevated meltwater pressures commonly observed elsewhere.

A series of laboratory tests have been performed to explore the geotechnical and geophysical characteristics of fibrous muskeg (i.e., grassy bog), using undisturbed block samples obtained from Alberta, Canada. The basic index properties of muskeg show similar values to the reference data for other organic soils, and they significantly differ from those of typical inorganic soils. Due to high organic content, muskeg has high water content, high compressibility, and high initial permeability. The effective stress and porosity play a major role in controlling the stress-dependent geophysical properties of shear wave velocity, real permittivity, and electrical conductivity. The values of peak drained friction angle obtained in the direct shear tests are in the range of 26° to 51°.

The dearth of baseline, near-surface (

Polycyclic aromatic hydrocarbons (PAHs) originating from natural sources, and potentially from the Athabasca Oil Sands development, are of concern for the Athabasca River and Lake Athabasca delta ecosystems. In order to model the transport of fine sediments (and associated PAHs), it is important to describe the sediment dynamics within the river system. Flocs possess different settling characteristics compared to individual particles. A key aspect in modelling floc settling behaviour is the mathematical linkage of the floc density to floc size. In this paper, a rotating annular flume is used to determine the settling characteristics of Muskeg River (a tributary of the Athabasca River) sediments under different shear conditions. Simulations of the settling and flocculation behaviour of these sediments were used to calibrate a density vs. floc size model. A relationship of the parameters relating floc size and density with the fractal dimension F shows that as diameter increases flocs become weaker. Recommendations for the practical application of the model are further formulated in this paper. The deposition tests offer a quantitative measure of the relative amount of sediment that is likely to be transported through the river for given flow conditions.

At the Muskeg River Mine, bitumen is hosted in the clastic sediments of the lower Cretaceous McMurray Formation. Within the mine area, the McMurray Formation is divided informally into mappable units representing fluvial, continental floodplain, open estuarine, estuarine channel complex (ECC), and marine environments. Fluvial, open estuarine, and ECC deposits host more than 90% of the mineable bitumen reserves. Bitumen grade is more consistent within the fluvial and open estuarine units (12–15 mass%), whereas ECC sediments are characterized by significant lateral and vertical grade variability (0–15 mass%). In the ECC deposits, bitumen grade is controlled by significant reservoir heterogeneity. Facies assemblages including point-bar deposits (PB), abandoned channel-fills (AC), and tidal flat deposits (TF), create complex internal geometries, architectures and associated reservoir properties. Traditional facies mapping and correlation has proven to be difficult even in closely spaced wells for the ECC deposits of the McMurray Formation; thus, an alternative technique using concepts of Stratigraphic Dip Analysis (SDA) was developed to assess bitumen grade for the deposits at the Muskeg River Mine. This approach involves three main steps: (l) juxtaposing azimuth maps (rose diagrams) over horizon slice facies maps for selected stratigraphic intervals to identify major channel trends (paleo-current directions); (2) comparison of dips, with corresponding sedimentary structures allows for a better prediction and geometries of point bars and abandoned channel-fills; and (3) comparison of dip trends with dominant lithology of facies assemblages and available bitumen grades provides a base for accurate delineation of architectural elements. A detailed case study is presented and shows that this approach provides a base for accurate delineation of architectural elements and confirms that bitumen grade decreases laterally with inferred maturity of point bar successions.

Previous studies indicate that the Middle Devonian of the Rainbow sub-basin represents a closed petroleum system with the Muskeg evaporite acting as the regional seal. To investigate the spatial and compositional characteristics of hydrocarbons in this area, multivariate statistical methods were used to study the geochemical signatures of a suite of Middle Devonian oils and analyze the geographic distribution characteristics of the non-associated gas accumulations in the sub-basin. The results suggest that there are two genetic groups of oils derived from slightly different source rocks. Geographic correlation of an oil group with a particular depositional facies indicates that communication between source and reservoir is the primary control on the variation of spatial characteristics of the oils. Geographic distribution characteristics of the non-associated gas accumulations, oil pool size and reservoir character suggest that cross-formational migration of hydrocarbon may have occurred in major fracture systems. The findings of this study may provide useful clues for future oil and gas exploration in this and adjacent, relatively unexplored, regions.

The Tongass National Forest of Southeast Alaska, USA, provides a unique environment for monitoring the impact of the cave system on water quality and biological productivity. The accretionary terrane setting of the area has developed into a complex and heterogeneous geologic landscape which includes numerous blocks of limestone with intense karstification. During the Wisconsian glaciation, there were areas of compacted glacial sediments and silts deposited over the bedrock. Muskeg peatlands developed over these poorly drained areas. The dominant plants of the muskeg ecosystem are Sphagnum mosses, whose decomposition leads to highly acidic waters with pH as low as 2.4. These waters drain off the muskegs into the cave systems, eventually running to the ocean. In accordance with the Tongass Land Management Plan, one of the research priorities of the National Forest is to determine the contributions of karst groundwater systems to productivity of aquatic communities. On Northern Prince of Wales Island, the Conk Canyon Cave insurgence and the Mop Spring resurgence were continuously monitored to understand the buffering capacity of the cave system. Over the length of the system, the pH increases from an average 3.89 to 7.22. The insurgence water temperature, during the summer months, ranged from between 10oC to 17oC. After residence in the cave system, the resurgence water had been buffered to 6oC to 9oC. Over the continuum from insurgence to resurgence, the specific conductance had increased by an order of magnitude with the resurgence waters having a higher ionic strength. The cave environment acts as a buffer on the incoming acidic muskeg water to yield resurgence water chemistry of a buffered karst system. These buffered waters contribute to the productivity in aquatic environments downstream. The waters from this system drain into Whale Pass, an important location for the salmon industry. The cool, even temperatures, as well as buffered flow rates delivered by the karst systems are associated with higher productivity of juvenile coho salmon.

This chapter describes the results of the measurement and characterization of the shearing behavior of a variety of terrains, including mineral terrain, muskeg, and snow-covered terrain. The measurements are made using a vehicle-mounted bevameter with the associated computerized data acquisition and processing system. When a multiaxle wheeled vehicle (or a tracked vehicle) is in straight-line motion over an unprepared terrain, an element of the terrain is subject to the repetitive shearing of the consecutive wheels (or roadwheels). In order to predict the shear stress distribution on the vehicle–terrain interface, the response to repetitive shear loading of the terrain should be known. The response of a frictional terrain (a dry sand) to repetitive shear loading under a constant normal load indicates that when the shear loading is reduced to zero and is then reapplied, the shear stress–displacement relationship during reshearing is similar to that when the terrain is being sheared in its virgin state. The variation of the shearing force beneath a rectangular shear plate on dry sand subject to a vertical harmonic load with a frequency of 10.3 Hz indicates that during the loading portion of each cycle, the shearing force S does not reach its maximum value instantaneously.

This chapter describes the measurement and characterization of the response of a variety of terrains, including mineral terrain, muskeg, and snow-covered terrain. One of the fundamental tasks in the characterization of terrain behavior is to establish functional relationships that can realistically describe the responses of the terrain to various types of loading, such as the stress–strain relationship, pressure–sinkage relationship, and shear stress–shear displacement relationship. The responses of a variety of mineral terrains in Eastern Ontario, Canada, including sand (LETE sand), sandy loam, clayey loam, and loam to normal and repetitive loadings were measured and characterized. The results indicate that there is little difference between the values of the pressure–sinkage parameters obtained using a set of rectangular plates of large aspect ratios and those obtained with a set of circular plates having radii equal to the widths of the rectangular plates. The responses of two muskegs in the Petawawa area, Ontario, Canada, to normal load and to repetitive loading were measured using the bevameter together with the automatic data acquisition and processing system. A well-defined critical pressure at which the surface mat of the muskeg is broken can be identified.

With 150 Million Barrels of recoverable reserves, the Nexen Hay River Bluesky oil pool is among the largest oil discoveries in Western Canada in the past two decades. Initial pilot production testing lead to the conclusions that, in order to produce the field in its most economical fashion, the following would be needed: • Triple lateral horizontal producing wells • In field horizontal water injection/water disposal wells for pressure maintenance • Artificial Lift from the outset Further complicating production of the Hay Pool, which is located in a remote area of North-Eastern British Columbia, was the fact that road access would be limited to 4 months of the year because any road through the muskeg would need to be “frozen in” prior to use. It was therefore extremely important to the economic viability of the project that whatever means of artificial lift were chosen should be highly reliable and require minimal maintenance. The method chosen, apart from being reliable and safe, would also need to be capable of lifting large volumes of fluid with a significant pressure drawdown. In addition it would need to be able to operate successfully in the extreme well deviations common in the “build” sections of the development wells. This paper will review the development of the Bluesky oil pool and present the case for the choice of Jet Pumps for artificial lift. The authors will go on to detail the performance of the Jet Pump system installed in the field that has met or exceeded all performance expectations. Background The Hay development project exploits a medium to heavy oil (24 API) reservoir located about 40 miles northwest of Rainbow Lake in northeast British Columbia near the Alberta border (Figure 1). The productive formation is the Lower Cretaceous Bluesky sand encountered at a true vertical depth of about 1,060 ft. The Bluesky is an offshore to lower shoreface sandstone deposited in an overall transgressive marine environment. Porosity values range from 18-27% and permeability values range from 60–250 md. The oil zone, which is 13-15 feet thick, is underlain by an 18-foot water zone and overlain by 560 – 885 ft of Buckinghorse shale followed by glacial deposits.

Summary I performed a gravity survey on Admiralty Island in Southeastern Alaska looking for massive sulfides. The survey was conducted for Kennecott Minerals Greens Creek Mining Company to map any possible extension of the present works to the south and on three exploration targets in areas of geologic interest. Admiralty Island is characterized by very steep topography, muskeg, and poor topographic information. Despite these obstacles, good quality data were achieved by careful data acquisition and processing. The data confirmed the extension of the mined ore body at depth and provided valuable information for exploration drilling.

The Rainbow evaporite basin, centered on latitude 58°N and longitude 119°W, is situated approximately 900 km from Calgary in the remote northwest corner of Alberta, Canada. It is approximately 30 × 30 km in areal extent with a relatively uneventful surface topography which is covered by muskeg and stunted coniferous forest. The climate is continental with severe winters (temperatures dip to −50° C or lower).

Slender wheatgrass and jack pine were grown in the greenhouse in cores containing a bottom layer of extracted oil sands with four overburdens individually layered over the sand. The overburdens included a muskeg peat, two shallow mineral overburdens and a deep overburden. Mycorrhizal development, microbial respiration and biomass and the degree of decomposition of slender wheatgrass roots in litter bags were determined in each plant species-overburden combination. Both ecto- and vesicular-arbuscular (VA) mycorrhizal inoculum was present in all four overburdens. The symbionts of slender wheatgrass were the "fine endophyte" and Glomus aggregatum. VA development was very low in peat whereas plants in the shallow overburdens became heavily mycorrhizal. Infection did not spread from the overburden layer to roots in the tailing sand. Jack pine roots in the peat and two shallow overburdens were heavily infected after 4 months. The most common symbiont was an ascomycete known as the E-strain. Microbial respiration was highest in the peat and was not influenced by plant species. Microbial biomass was also highest in the peat and much lower in the mineral overburdens. Only in the peat was the amount of microbial biomass larger with slender wheatgrass than with jack pine. Slender wheatgrass roots decomposed most rapidly in the peat overburden and least rapidly in the deep overburden. Key words: Microbial activity, jack pine, slender wheatgrass, mycorrhizae, reclamation, oil sands

Pollen assemblages from surface samples of lake bottom sediment and northern muskeg indicate that the major vegetation regions (boreal forest, taiga, forest‐tundra, and tundra) that extend from central Quebec to the eastern Arctic can be identified on the basis of palynological information. Although it is possible to recognize modern palynological tundra in relation to present distribution limits of vegetation, it should not be concluded that unqualified identification can be made of “palaeotundra” that presumably existed in the periglacial zone that bordered the southern margin of the continental ice sheet in late Pleistocene time. Our present data base is not adequate with respect to arctic palynological surface sample coverage and the understanding of atmospheric pollen dispersal over the northern regions. This information is required for the study of the history of tundra environment in the Cenozoic palynostratigraphic record.

This paper discusses the application of data on stream water chemistry to regional groundwater evaluation, hydrograph separation, and evaluation of the effects of basin disturbance. We have selected for analysis the Firebag, Steepbank and Muskeg River basins, and Hartley and Thickwood Creek basins in northern Alberta, Canada. They cover slightly more than 9000 km 2 within a region containing vast quantities of surface mineable, oil sand deposits. All basins are drift covered and underlain by bedrock that can range in age from Precambrian to Cretaceous. Comparison of chemical data for stream water from all basins with that for groundwater from Hartley Creek basin confirms that groundwater is a major source of streamflow in winter. Indications are that this groundwater inflow is almost entirely from glacial drift. Hydrograph separations based on chemical parameters indicate that during the spring and summer the proportion of groundwater in the streamflow is less than winter values with the largest contribution by far coming from the drainage of muskeg. Muskeg thus exerts important control on stream discharge and chemistry. If surface mining in these basins causes the removal of muskeg and replacement by mineral soil, stream discharge from affected areas will be less during summer and more during spring runoff and stormflow periods. As well, major-ion concentrations should be generally higher in the summer and fall months. However, the impact of local disturbances in a larger basin will be diminished downstream as waters from affected areas are mixed with other waters.

The distribution of peatland as a percentage of land area and the distribution of open and treed peatlands as percentages of peatland area have been determined for the Province of New Brunswick, For this work, aerial photographs, a dot planimeter, a mirror stereoscope and a computerized mapping procedure, SYMAP, in conjunction with the University of New Brunswick IBM system/360 computer were used. The results have been displayed in the form of symbolized computer printout maps (SYMAP) showing the distribution of peatland in New Brunswick. About 10% (about 700 000 ha) of the land area of New Brunswick is covered by peatland.

Seventy surface pollen samples from coastal forest, coastal meadow, muskeg, tree line, and alpine tundra communities form a basis for interpreting fossil pollen assemblages in the Malaspina Glacier district, Alaska. Poflen and macrofossil analyses of three radiocarbon-dated fossil sections from Icy Cape indicate that vegetational changes resulting from plant succession can be distinguished from those of migrational and climatic origin. Vegetation of the early Holocene xerothermic interval (10,000–7600 yr B.P.) was dominated by Alnus communities. Wetter conditions ensued, enabling generative muskeg surfaces to develop and first Picea sitchensis, then Tsuga heterophylla to expand from areas southeastward. Climatic cooling in more recent millennia (3500 yr B.P. to the present) is indicated by the appearance and persistent growth of Tsuga mertensiana and Selaginella selaginoides along this portion of the Gulf of Alaska coastline.

The paper investigates the relationships that exist between the airform pattern, hydrology, and ontogeny of the muskegs of Canada. A new interpretive classification based on the concept of hydroform is proposed and some practical examples of its application are given. Cet article etudie les relations existant entre 1' "airform pattern", I'hydrologie, et I'ontogenese des fondrieres de mousse du Canada. Une nouvelle classification d'interpretation basee sur le concept d' "hydroform" est proposee. Quelques exemples pratiques de son application sont donnks. [Traduit par le journal]

In situ vane tests were carried out in a number of different muskeg deposits to examine the mechanism of failure and the effect of variation in vane size and speed of testing. The failure mechanism in peat was found to be similar to that in soft clays. The apparent shear strength varied with vane size but was independent of testing speed. There was an optimum vane size of 10 cm diameter.Block samples were obtained from one site for laboratory vane tests to investigate anisotropy in the strength of the peat. In situ tests for anisotropy using vanes of varying shape confirmed the laboratory results, and showed that strengths on the vertical plane could be 100 % higher than those on the horizontal plane.

The Bohemia Basin nickel-copper deposits are segregations of pentlandite, pyrrhotite and chalcopyrite in the lower portions of a norite mass, the bottom of which is basin-shaped. Eight traverses were made with an Askania vertical intensity magnetometer across the projected position of the deposit beneath a covered area. Magnetic anomalies were observed at the inferred position of the deposit along six of the traverses and indicate that the body is continuous beneath most of the covered area. Magnetic profiles are consistent with the hypothesis, proven at two places by outcrops and drill holes, that the deposit dips inward. The profiles and outcrops indicate that the deposit averages about 80 feet thick. The profiles also suggest that the deposit, under the covered area, extends down the dip for a distance greater than its thickness and that the muskeg cover is not deep. These inferences are consistent with available drill data. The inferred quantity of sulfide-bearing material in the deposits was increased by the results of magnetometer exploration from 5,000,000 tons to 8,500,000 tons. The magnetic data will facilitate planning of any future development of the deposits and would have been useful in the more effective placing of some of the holes already drilled.

Consideration is given to the prediction of undrained movements which may occur due to contained yielding beneath embankments on muskegs. The study shows that insignificant improvement in the predicted undrained movements can be expected by assuming the soil to be an elastic cross-anisotropic material unless the independent shear modulus is measured. All predictions were made using the finite element technique and assuming the soil to be an elastic isotropic yielding material. The settlement predictions are in good agreement with the recorded construction movements of the two case records studied.The first analysis clearly shows that the yielded zone is restricted to a soft clay layer immediately beneath about 20 ft (6.1 m) of peat and marl. The predictions show that berms have a beneficial effect in containing this zone and reducing differential settlement. Horizontal movements are significant in comparison with the vertical movements except near the centerline.In the second case, analysis of a shallow peat layer underlain by stiff clay shows that the predicted settlement and shape of yield zone are largely dependent on whether the lower boundary is assumed perfectly rough or shearless.

Results of tests performed using a 1.8-m (6-ft) diameter flexible based tank capable of being filled with 3.6 m (12 ft) of water are presented. Two series of tests were performed on a muskeg with an undrained shear strength of 5–7.5 kN/m2 (100–150 lb/sq.ft). The first consisted of filling the tank alone and the second with the tank positioned at the center of a 4.5-m (15-ft) diameter 0.6 m (2 ft) high flexible based reservoir. The results thus simulate the comparison of the effect of berms on the stability of embankments. Consideration is given to the short term effect of berms where they are constructed rapidly at the same time as the central fill and the long term effect where they and the central fill are constructed slowly.The resulting vertical movements were measured using a torpedo settlement gauge. The horizontal movements were measured using an inclinometer as well as a horizontal movement device. The pore pressures were measured using fine stoned piezometers.In addition suitable laboratory tests were conducted to compare predicted and measured results.

Seismic shothole drillers’ logs, record the near-surface (avg. 18.6 m deep) lithostratigraphy encountered when drilling holes to place explosive charges. These records offer a largely unrecognized wealth of geoscience information in areas for which little may be otherwise known. Stored in the Basic Files archives of petroleum exploration and seismic acquisition companies, this study first convinced companies of the potential utility of this data, then recovered the hard copy and digitally scanned records (paper, fiche, microfilm) and rendered these into a digital database and GIS. The final database of 343,989 records provides the largest source of geoscience information of its kind in northwestern Canada, and in many cases contains unique and original records on a host of subjects including surficial-, bedrock-, and hydro-geology, permafrost, and geohazards. The drillers’ log records have further been used to create geospatial models of drift, till, muskeg, massive ice and ground ice thicknesses, and continue to be applied to new avenues of research such as temporal variations of bottomfast ice extents in offshore shallow marine environments. Published in freely downloadable Geological Survey of Canada Open File reports and providing commonly used database and GIS file formats, this data rescue exercise preserves and greatly enhances what was becoming an increasingly discarded corporate data set of unrecognized potential.

General American Oils, Ltd. conducted a wet in-situ combustion pilot in the Wabiskaw Formation of the Wabasca Heavy Oil Sands Deposit from February, 1981 until February, 1983. The purpose of this test was to assist in determining if this process would be economically feasible in the Wabiskaw Formation. A single 20 acre (8.09 hectare) inverted seven spot was chosen as the pattern for the pilot. The operational concept was that the producing wells would be cyclic steamed to stimulate them at the same time that the wet in-situ combustion process was being conducted. To monitor the movement of the combustion front, a series of 3-D seismic surveys were scheduled. The initial survey was conducted prior to air injection and was used as a reference to compare to the surveys conducted after ignition of the pilot. The pilot performance using cyclic steam and wet in-situ combustion as well as the results of the 3-D seismic monitoring are discussed in this paper. The well completions and surface facilities design as well as operational concepts to allow year round operation in a muskeg environment are also discussed. Additional information was also gathered on the rock and fluid properties of the Wabiskaw. Even thoughmore » the project was terminated after two years of the planned five year program because of the poor performance exhibited using air injection, the information gathered will help in the design of future Wabiskaw pilots.« less

Marl occurs in and adjacent to at least five lakes in the Knik Arm area in south-central Alaska. The most significant deposits are near Wasilla, about 30 miles north of the city of Anchorage. Accumulation of the calcareous material has taken place in sheltered embayments and shallow marginal areas of the lakes; algae probably act as the precipitating agents. In some places the embayments have been completely filled with, marl and subsequently covered by a layer of muskeg, effectively sealing off the deposits from the lakes proper. The marl is probably chemically suitable for the manufacture of Portland cement, but the inferred reserves about one million tons (dry weight) would be insufficient for modern large-scale operations. The deposits described represent only the more obvious occurrences of marl, found during a brief reconnaissance of a small part of the Knik Arm area. If further investigation should become warranted, particular attention should be directed toward finding buried shoreline deposits that are usually higher in CaCO3 than the other lacustrine materials found in the area.

The Zama area of northwestern Alberta is in the Middle Devonian Black Creek basin. At the beginning of late Elk Point deposition, open-marine conditions prevailed and organic reefs of the Keg River Formation grew to thicknesses of 350 ft. Reef growth was terminated abruptly during a period of uplift and marine regression, and the basin was partly filled with halite of the Black Creek Member of the Muskeg Formation. With the renewal of subsidence, the Black Creek basin ceased to exist and the Zama area became part of the more extensive Elk Point evaporite basin. With two exceptions evaporitic conditions represented by the cyclical anhydrite and dolomite of the Muskeg Formation prevailed through most of the remaining time of Elk Point deposition. During early Muskeg deposition, a well-laminated, nonreef calcarenite, the Zama Member, developed as a shoal over the dormant Keg River reefs, and a second, more widespread nonreef calcarenite, the Bistcho member, was deposited in a final transgressive phase. Deposition of green, brackish-water shale of the Watt Mountain Formation was succeeded by the transgressive limestone sequence of the Slave Point Formation. Contemporaneous solution of the Black Creek salt resulted in thickening of post-Keg River units in reef-flank and offreef positions. Structural draping of the Slave Point, Watt Mountain, and Muskeg Formations across the Keg River reefs reflects the continued removal of Black Creek salt during Late Devonian time. The upper reef member of the Keg River Formation and the Zama Member of the Muskeg Formation are oil bearing. Above full-reef buildups there is no separation between the two reservoirs. In reef-flank positions the Zama Member is separated from the Keg River by a lower anhydrite member of the Muskeg Formation. The Bistcho member of the Muskeg Formation and thin calcarenite units in the Slave Point Formation are gas reservoirs.

Towards a reassessment of muskeg, an appraisal of the problems of muskeg showing the complexity of the subject and the implications for engineering is presented based on field experience and a study of the literature. A theory of the origin of muskeg is given from the engineer's point of view by considering climatic, biotic, and geologic factors. Muskeg research in other countries is reviewed with emphasis on Canadian research. The Radforth Muskeg Classification System, its application to problems and modifications are discussed. Empirical data on the engineering properties of peat are broadly described in four main categories: (1) index properties, for identification; (2) strength and deformation properties; (3) thermal properties; and (4) geophysical properties. Engineering problems, construction equipment, and costs to be considered in vehicle trafficability, road construction, corrosion and drainage operations, and frozen muskeg and permafrost construction when muskeg is involved and fully discussed. Comprehensive conclusions and recommendations on the present state of knowledge of muskeg and needs for further research are outlined.

The Muskeg Formation, as established by J. Law in 1955, is a complex of intricately interbedded and interfingering carbonate and evaporite rocks, with interspersed insoluble residual beds. The complex can be divided into several time-stratigraphic entities, which indicate a consistent upward increase in salinity. The vertical succession is consistently repeated in the horizontal dimension as normal marine sediments grade into progressively more saline sediments of an interior barred basin. In contrast to many recent observations on supra- or intertidal deposition of evaporites, those of the Muskeg Formation are believed to have been laid down from stratified brines filling the deepest part of the basin. Physicochemical considerations permit interpretation of the evaporites of the deeps as roughly synchronous lithic equivalents of carbonate banks and reefs of the shallows. The seaward position of a broad pre-evaporation shelf in northwestern Alberta was favored for prolific organic activity. Barrier, pinnacle, fringe, and patch reefs growing at shallow depth kept pace with subsidence. Simultaneously, much of the organic substance accumulated under anaerobic conditions in adjacent deeps. Source, reservoir, and trapping conditions thus were provided in a single environment.

Degrading permafrost below roadway embankments is a widespread problem in the north. Thermal modeling can help to determine thermally stable embankment configurations; however, this modeling typically does not include the effects of groundwater flow on the geosystem, which will cause permafrost degradation to occur faster than atmospheric warming alone. As part of a larger, ongoing research project, we present preliminary results of heat transfer modeling for an Alaska Highway test section near Beaver Creek, Yukon Territory, Canada. This experimental highway test section is located in an area characterized by muskeg vegetation underlain by ice-rich permafrost. While the overall project includes field work and laboratory measurements, this paper focuses on the results of exploratory modeling. A two- dimensional finite element program capable of mathematically coupling both heat and groundwater flow was used for modeling. Comparing model results produced using conductive heat flow only to model results using both heat and groundwater flow (heat advection) indicates that groundwater has a significant effect on the configuration of the thaw bulb and the temperature distribution within and below the roadway embankment. For example, using 50-year model results, including groundwater flow increases modeled thaw depths below the embankment by approximately 8 m, and can increase temperatures to an excess of +2.5°C at the bottom of the embankment. For wet terrain conditions, these preliminary modeling results indicate that it is essential to incorporate groundwater flow into thermal modeling, in order to understand better the complex interactions between roadway embankments and underlying permafrost.

Shallow geological, geotechnical, and environmental investigations in sub-arctic regions require a unique approach because of complex near-surface conditions. The surficial geology in these regions includes a thin layer of muskeg, silt, or sand. Bedrock can be sedimentary, metamorphic, or igneous, and is usually shallow. Permafrost is present as either a continuous or discontinuous layer with a thin, seasonally active layer. A combination of surface and airborne geophysical surveys, supported by drilling, sampling, and borehole geophysical logging, has proven to be an effective way of characterizing the subsurface extent of the permafrost and cover. Terrain conductivity surveys showed the seasonal development of an active (unfrozen) zone of highly variable depth. Electrical resistivity surveys produced cross-sections of the subsurface down to depths of 30 m, that showed the presence of permafrost lenses and taliks (wedges of thawed overburden). Borehole geophysical logs in shallow drill holes (

Test trenches excavated into muskeg near Fairbanks in 1969 exposed a polygonal network of active ice wedges. The wedges occur in peat that has accumulated since about 3500 yr BP and have grown episodically as the permafrost table fluctuated in response to fires, other local site conditions and perhaps regional climatic changes. Radiocarbon dates suggest one or two episodes of ice-wedge growth between about 3500 and 2000 yr BP as woody peat accumulated at the site. Subsequent wedge truncation evidently followed a fire that charred the peat. Younger peat exhibits facies changes between sedge-rich components that filled troughs over the ice wedges and woody bryophytic deposits that formed beyond the troughs. A final episode of wedge development took place within the past few hundred years. Pollen data from the site indicate that boreal forest was present throughout the past 6000 yr, but that it underwent a gradual transition from a predominantly deciduous to a spruce-dominated assemblage. This change may reflect either local site conditions or a more general climatic shift to cooler, moister summers in late Holocene time. The history of ice-wedge growth shows that wedges can form and grow to more than 1 m apparent width under mean annual temperatures that probably are close to those of the Fairbanks area today (-3.5?C) and under vegetation cover similar to that of the interior Alaskan boreal forest. The commonly held belief that ice wedges develop only below mean annual air temperatures of -6 to -80C in the zone of continuous permafrost is invalid.

Little information has been published regarding the general geology of the Mackenzie Delta region. On the western side of the Delta bedrock appears to consist largely of Mesozoic sedimentary formations. Coal-bearing rocks occur some 25 miles northwest of Aklavik and a six-foot coal seam with almost vertical dip is worked at Moose River Coal Mine. A description of these workings is given. The eastern side of the Delta shows some variety in its geology. Bedrock areas between Campbell Lake and Dolomite Lake are largely formed of dolomite and magnesian limestone. Dolomite, limestone and reddish slaty shale are seen on the north shore of Dolomite Lake and exposures of shaly material are seen on the east bank of the East Channel eight miles south of the entrance channel leading to Dolomite Lake. Between Dolomite Lake and the southern end of the Caribou Hills bedrock is concealed beneath unconsolidated deposits covered by bush and muskeg. The deposits appear to consists largely of silt, clay and gravel, and wrap round and conceal the junction of the Caribou Hills formation with the underlying formations. The undulating and rather formless character of this tract of ground and the relation of the deposits to the bedrock at Dolomite Lake suggest that they are glacial drifts. The Caribou Hills form the most striking feature on the east side of the Delta. They form a steep face or escarpment overlooking the East Channel and rise to their maximum height south of the Reindeer Station. Northwards their altitude gradually diminishes and they terminate against the tundra south-east and the swamps of the Delta of Tununuk. The rocks consist of soft poorly consolidated silty clay, and fine silt and sand with some pebbly or gravelly sand capping the hills locally. Beds of lignite and very thin layers of ironstone were also seen and, where the lignite has been burned out reddened patches and bands appear on the steep hill slopes. The Caribou Hills formation recalls Tertiary lignite-bearing formations described from other parts of Northern Canada and Alaska. The bearing of these and other observations upon the history of the Mackenzie Delta is briefly discussed.