The distribution, morphometry and biophysical characteristics of turf-banked lobes and terraces are described from a small subarctic alpine plateau (1300--1900 m a.s.l.) overlooking Kluane Lake in the southwest Yukon Territory, Canada. Observations on the distribution of lobes and terraces at this site indicate that the main controlling factors are the complexity of the topography and the properties of snow cover. Morphometric analysis of 85 lobes subdivided into five different slope orientations indicates a morphogenetic underpinning in which lobe form remains consistent (length-to-width ratio below unity) at different scales and between aspects, however, morpho-biophysical correlations could not provide an explanation for this. Nonparametric and multivariate parametric statistical tests indicate significant inter-aspect differences in the dimensions and biophysical characteristics of the lobes. These appear to be related to differences in site-specific factors. Measurements made along certain transects reveal a definitive downslope trend in the size of lobes. This organization is developed as a result of downslope gradients of various biophysical factors, primarily soil moisture and frost-susceptible soil (silt and clay), which are created by slopewash processes.

An important source of information about climate change comes from sedimentary deposits from the arctic region. However, there are few lacustrine fossil records from the central mid Arctic region of Canada. A 485 cm sediment core, from a lake unofficially called JR01, Boothia Peninsula, Nunavut, yielded a high-resolution diatom record. Five radiocarbon dates provided the chronology. Three distinct zones are revealed in the relative percent abundance and diatom concentration diagrams. A more diverse and productive flora in the first zone implies warmer temperatures in the middle Holocene. A complete shift in diatom composition to a predominantly Fragilaria flora and a reduction in diversity and production suggests a cooler climate in this region after approximately 4360 years B.P. Smaller-scale climatic trends, such as the Little Ice Age (LIA) (ca. 550--150 yr. B.P.) and the Medieval Warm Period (ca. 1000--700 yr. B.P.), caused shifts in the diatom flora and production. Subtle shifts in floristic diversity within the LIA may reflect climate variability at a decadal scale. This variability is also recorded in other high-resolution paleoclimate records from the Arctic. A gradual shift to a more diverse and productive flora in the last 180 years B.P. (1790 years A.D.) suggests a response to the recent warming trend. This study permits us to address ambiguities with respect to the timing, the geographic extent and the severity of the recent warming. Changes in total diatom concentrations, magnetic susceptibility, organic and carbonate content may also reflect changes in the diatom assemblages and indicate shifts in vegetation, soil erosion, or in hydrology from the surrounding catchment. The morphological variance of Fragilaria and other taxonomy issues are also discussed.

Fifty-one frost mounds were examined in Wolf Creek, Yukon Territory (60°30'N, 135°13'W) at an elevation of 1235 m a.s.l. Cryostratigraphic analyses and aerial photographic interpretation were undertaken to investigate the origin and longevity of the mounds, and to assess their utility as climatic indicators. It was determined that 37 mounds were palsas, as evidenced by their dimensions, cryostructure of segregated ice and location within a fen; one mound was a frost blister, as evidenced by its core of intrusive ice; one mound may have been a compound form, with segregation and possibly intrusive ice; and 12 mounds were termed aggradational permafrost mounds as their cores were of segregated ice but they did not fit the locational requirements to be palsas. Aerial photographs, spanning the period from 1946 to 2001, showed that palsas at the study site have been aggrading and degrading continually over the past 55 years, and are continuing to do so. (Abstract shortened by UMI.)

Ground ice formation in the Far Northwest of Canada was investigated using stable isotopes (delta18O and deltaD) and 14C dating. These have proven to be valuable tools when inferring the origin of ground ice bodies. The major findings are: (1) During the Late Pleistocene, massive tabular ground ice bodies formed extensively in the Willow River area, Richardson Mountains. These ice bodies are currently exposed in the headwalls of retrogressive thaw slumps as a result of climate warming. Two units were recognized on the basis of ice structure: a debris-rich ice overlain by a diamicton. The debris-rich ice, characterized by low delta 18O values (-30‰ to -27‰), is believed to have formed by in situ freezing of subglacial meltwater derived from the Laurentide ice sheet. The meltwater flowed through water-saturated till between relict and aggrading permafrost. A decrease in subglacial meltwater flow from the ice sheet and a decrease in enthalpy transfer from the meltwater caused the freezing of the meltwater within the proglacial water-saturated till. The ice in the diamicton is characterized by relatively high delta 18O values (-26‰ to -20‰) and formed from the freezing of isotopically enriched Holocene waters. (2) In the northern Yukon, a delta18O analysis of surface and subsurface ice was undertaken. The results indicate that the delta18O composition of the various ice types is dependent upon a number of factors, such as surface air temperature, relative humidity and freezing rates. Holocene age waters are characterized by delta18O values >-27‰, while Pleistocene age waters have delta18O values <-27‰, indicating colder air temperature at the time of precipitation. The effect of evaporation on the isotopic composition of water was best observed on open surfaces where enrichment in the order of 5--11‰ in 18 O was measured.

This paper proposes an estimating method of snow reflectance based on Landsat-5 Thematic Mapper (TM) data (June 21 1988) and a 30-meter resolution digital terrain model that correct and adjust the problems mentioned above. A radiative transfert model (LOWTRAN-7) in conjunction with radiosonde data was used to eliminate the radiometric distortion on the signal created by the presence of the atmosphere. A software package called PRODEM was use to model the terrain shape from the digital elevation data. PRODEM was also very helpful for integrating the digital elevation model to the digital analysis of the remote sensing data. The results were compare to a series of data extracted from the literature. The estimated values of reflectance and albedo are in good correspondence with the values in the literature. However, we observed various problems to estimate the reflectance and albedo of surfaces that present extreme topographic characteristics. First, we think that those problems can be related to an under evaluation of the diffuse component of the incident radiation. Second, it can also be attributed to an imperfect match between the satellite data and the incident solar radiation image. Also, the differences between the estimated and the observed snow reflectance values (literature) can be imputed coarsely to the approximation of the different model constituents and the spatial and temporal correspondence of satellite and in situ data. (Abstract shortened by UMI.)

The purpose of this thesis is to investigate the interactions between meltwaters and sedimentary material at the base of an alpine glacier which are indicated by intra- and interseasonal variations of discharge, solute and sediment outflow from the terminus. During the 1989 and 1990 ablation seasons, discharge, suspended sediment concentration and dissolved solids draining from Maxwell Glacier, St. Elias Mountains, Yukon Territory, were obtained at hourly intervals. The discharge time series of Maxwell Creek reveals that flow varies at a period of one to two weeks, at a diurnal level and over one to three hours. Over all three time scales, electrical conductivity is inversely proportional to discharge suggesting that during high during high diurnal flows, outwash water is derived primarily from surface melt while at low nocturnal discharge, water originates from subglacial and englacial stores. In spring, when supraglacial meltwater begins to access the subglacial drainage system the basal water pressure is raised. Large areas previously hydraulically isolated are integrated releasing stored basal sediment. During periods of increasing discharge in the early summer hydroglacial events resulting from gradual expansion and/or isolation of subglacial cavities, temporary blockage of subglacial conduits or from input from adjacent gravity deposits occur. As flow increases, conduit diameters increase and water gradually spreads out over the glacier bed as a diffuse network of linked cavities. Throughout the ablation season, different areas of the sole are integrated with flow. Towards the end of the ablation season, meltwater supply declines resulting in the reduction of flow in subglacial passageway and reduced sediment and solute supply. (Abstract shortened by UMI.)

The main objective of this thesis was to offer explanations on the evolution of the plateau surfaces in the northern Ruby Range, Yukon Territory. Specifically, to determine whether the geological and geomorphological evidence from the plateau surfaces could support the belief that these regions were subjected to glaciation, or does the evidence indicate other evolutionary processes. The more traditional geological techniques of geochemical and heavy mineral compositional analyses have been used previously in both geological and glacial related studies in the Canadian Shield and other regions of Canada, and have aided in the description of environments. However these types of analyses have had limited application in alpine environments with expensive plateau regions, such as those experienced in the northern Ruby Range. The data also made it possible to delineate, in a regional context, mineralized zones that have not yet been identified. This pursuit constituted an ancillary objective in this thesis. It was determined that the average proportion of heavy minerals identified was similar for both plateau and valley sediment samples. The range of the proportions was found to vary considerably in most minerals. The significance of these comparisons was not clear, but probably indicated that the sediments from both plateau and valley sites did not vary greatly in heavy mineral species and proportions, but demonstrated a variable range. As well, there was no appreciable difference in the heavy mineral assemblages between plateau and valley sediment samples, except for the relative lack of allanite from plateaus, and goethite from valleys. The data did not support a clear association between heavy mineral assemblages from plateau and valley sediments, and those derived from bedrock within or outside the study area. However, the sediment assemblages may not be exclusively related to local bedrock sources. The provenance of orthopyroxene, goethite, and allanite may be attributed, in part, to bedrock from outside the study area, volcanic ash, and meteoritic impacts, or a combination of these. The occurrence of these minerals may also be associated with mineral alterations. Both the heavy mineral and geochemical data indicated there were no discernable dispersal patterns in either an up-valley or down-valley direction. The random dispersals of minerals and elements in the valleys can be characterized as secondary, epigenetic dispersals, and can be further classified as resembling clastic, hydromorphic, or biogenic patterns. Heavy mineral and element analyses performed on surficial sediments from the plateau surfaces indicated that some of the material may not be of local origin, and discernable dispersal trends were evident in certain heavy minerals and elements which were traced to known sources of mineralizations. The most striking characteristics of these trends were their continuous form, spanning many kilometres, and their patterns which extended over extensive plateau surfaces that were dissected by numerous valleys. (Abstract shortened by UMI.)

Three satellite images originating from MSS (06-19, 1983), TM (09-19, 1984) and SPOT (04-22, 1987) were analysed in a forestry context for a region part of north-western Quebec. The result of this work was mainly based on visual analysis; the TM image best performed according to the parameters on which this research was conducted.

Limestone tablets placed in various aquatic environments yielded a mean rate of erosion of 22,8 mm/1000 years. Within different soil media, the same method gave a 12,4 mm/1000 years intensity of erosion. The hydrochemical approach gave a mean rate of 21,4 mm/1000 years, but the median rate of 17,2 mm/1000 years is believed to be closer to reality. The karsts developed on the crystalline (metamorphic) limestone of the study area are found at the surface as well as underground. Fossil caves hang on the valley slopes far from the actual streams. Many active caves show in their path fossil phreatic tubes superimposed on active vadose corridors. Rates of erosion topographic position, and the sediment filling of the fossil caves, suggest that these could be pre-Holocene in age. One model implies a sub-glacial dissolution by meltwater (phreatic tube) and a subsequent Holocene erosion (vadose corridor). The second model implies a washing-away of sediments by glacial meltwater within a pre-holocene cave, followed by Holocene erosion. (Abstract shortened by UMI.)

Parametric methods, currently used in regional flood frequency analysis, have numerous drawbacks and limitations, especially with regard to flood distribution selection and regional relationship form. Alternative approaches involving nonparametric methods are investigated in this thesis on a set of New Brunswick annual maximum floods. Nonparametric methods were employed at the three steps of regional analysis: at-site flood frequency analysis, homogeneous region delineation and regional relationship development. Nonparametric flood frequency analysis indicated that an annual maximum flood data set from New Brunswick contained some unimodal distributions along with many mixed distributions of bimodal and heavy-tailed shapes. A simulation study showed that sampling variability from a unimodal distribution could not account for the bimodality in nonparametric frequency analysis, confirming the existence of mixed distributions. L-moment analysis, a parametric method, confirmed that the entire set of floods from New Brunswick could not be appropriately described by a unimodal distribution. In this study, a new method is proposed for the purpose of homogeneous region delineation which effectively combines geographical considerations and flood data characteristics. The technique is based on the grouping of stations with similar density function shape, which reflect similar flood generating mechanisms. In New Brunswick, flood densities of three different shapes were grouped on a geographical basis to delineate homogeneous regions. Statistical tests based on L-moment analysis confirmed that the stations within a homogeneous bimodal region came from the same distribution. But L-moment analysis would propose either the Generalized Logistic or the Generalized Extreme Value as the regional distribution. Nonparametric frequency analysis revealed, however, that the floods within that region actually came from a mixed distribution. Nonparametric regression was employed for regional relationship development in New Brunswick; however, no significant improvement over the parametric approach of linear regression resulted. Using bootstrapping of pairs, a new method to compute the confidence interval at the center of a nonparametric regression was investigated. A comparison of linear and nonparametric regression confidence intervals can assist in evaluating the appropriateness of a linear model, and thus the need to employ nonparametric regression. Nonparametric regression was shown to be useful in screening irrational relationships that could be developed with the parametric approach. A new regional analysis methodology, involving nonparametric methods at the three steps of regional analysis, is proposed in this study, resulting in improved homogeneous region delineation, in more accurate at-site quantile estimates, and more realistic regional relationships.