This thesis is an investigation of glacial erratics derived from two small, adjacent, distinctive outcrops of essexlte near Lennoxtown, Central Scotland. Analysis of till samples has revealed that rock fragments and sand-size particles of essexite are absent up-ice of the essexite outcrops and scarce in a down-ice direction. Using dry stone walls as random samples of glacially transported stones, classic erratics train patterns of the distribution of erratics from both outcrops have been derived. The close spacing of Joints on one outcrop has led to the rapid disappearance of its erratics in the walls with distance from the source due to glacial crushing. Crushing is also shown by a size reduction of erratics from the other outcrop in the walls in a down-ice direction. Measurement of size and morphometric properties of essexite erratics has revealed that crushing is the dominant and abrasion a subsidiary process in glacial transport, whilst abrasion is dominant for particles that have undergone fluvioglacial and beach transport. Roundness appears to be the best measure differentiating between processes acting in the glacial, fluvioglacial and beach environments. Tests of compressive strength and experimental abrasion of essexite have been carried out to aid interpretation of the results. Tlll-particle preferred orientations, striae and the long axes of ice-moulded features snow a similar direction to that of the major axis of the train. From the evidence available it is suggested that the essexite wall stones underwent englacial transport in the lower layers of the last active ice and were deposited as an ablation mantle. A position of transport at the base of the ice is invoked for essexite particles in the till. The lateral spreading of essexite erratics down-ice of the source is attributed to divergence of basal ice flow around subglacial obstacles.

No description available.

No description available.

No description available.

The results of a comprehensive three-year study of the Marr Ice Piedmont, Anvers Island, Antarctica are presented. The piedmont stands on a low coastal platform ranging from slightly below sea level to 200 m. a.s.l. Ice thickness ranges from 60 to 80 m, at the coastal cliffs to more than 600 m. inland. Annual accumulation is high. There is a strong relationship between elevation and accumulation rates and a marked variation of accumulation rates from year to year. Surface ice velocities range from 14 m/year to 218 m/year and there is considerable ice streaming as a result of the subglacial topography. The mass balance of a representative part of the piedmont is considered to be in equilibrium or possibly, slightly positive. A study of a peripheral ramp shows annual fluctuations of balance and it is hypothesised that there may be a long-term tendency towards a positive regime. Ice core studies indicate that there is no dry snow fades but all other fades are identified. The saturation line lies at approximately 600 m. a.s.l. and the equilibrium line ranges from 60 to 120 m. a.s.l. Englacial ten-metre temperatures range from -0.8ºC near the coast to -4.9ºC inland. Deformation velocities have been calculated and basal sliding velocities inferred. It is hypothesised that basal conditions are not everywhere the same and that parts of the piedmont are frozen to bedrock. It is suggested that basal sliding and erosion are related and that the piedmont is selectively eroding its bed and accentuating the subglacial topo,j;raphy. Evidence of erosion, debris-rich ice, exists in the piedmont but is below sea level at the coastal cliff. The piedmont is not a "Strandflat Glacier" which is cutting a planed surface at a level controlled by the sea.

A model is proposed for the study of general two-dimensional 'polythermal' glacial ice flows. It consists of equations of conservation of mass, momentum and energy together with a constitutive relation between stress and strain rate and a full set of boundary conditions. A novel aspect is the consideration of the moisture content of temperate ice as an enthalpy variable: this represents a first attempt at incorporating the glacial hydrology in a dynamical model. One of the bedrock boundary conditions requires a knowledge of the so-called 'sliding law' relating the basal shear stress to the basal velocity (when the ice is temperate). A detailed model is proposed to determine this law, and upper and lower bounds for the basal velocity in terms of the stress are given by using a variational principle. The effect of cavitation on the sliding law is considered in the case of Newtonian flow, and it is shown to have a dramatic effect on the basal sliding. We then turn to our analysis of the glacier flow model. Firstly we consider kinematic waves: an analogue of Nye's (1960) equation is derived and analysed from a nonlinear viewpoint. The formation and evolution of surface shocks is studied, and explicit results given for small perturbations to the surface profile. Secondly, we show how the modal may be used to predict a finite slope at the glacier snout by use of the method of strained coordinates. Finally, steady state solutions are examined in the cases of large and small conduction. We demonstrate that there generally exists a large patch of basal ice which is almost temperate, but which slides at a velocity lower than that predicted by the sliding law: this is probably the most important result of the present work.

No description available.

No description available.

There is a need for accurate information on patterns of distribution of surface water in agricultural landscapes in western Canada. Artificial drainage on agricultural land has become widespread as farmers attempt to maximise their available land and minimise obstacles to efficient cultivation. There is a shortage of appropriate methods for describing, analysing and simulating the flow of surface water and its accumulation in shallow closed depressions in agricultural landscapes. Such capabilities are increasingly needed to assess the benefits and impacts of both on- and off-farm drainage. The goal of the present study is therefore to develop and test procedures to quantify the capacity to store runoff in depressions in agricultural landscapes and to estimate the extent to which this depressional storage is filled at any given time. Analyses of the conditions controlling runoff were conducted at a single site characterised by low relief, glaciated terrain and a non-integrated drainage system. Field measurements of soil hydrological properties and observed ponding were collected for the farm-scale site during a period of spring snow melt and runoff in 1989. Automated extraction of geomorphological features from digital elevation data was used to delineate the location , extent and maximum volume of all depressions and to establish the sequence in which they would be expected to fill, overspill and drain. A highly-distributed, physically-based hydrological model (DISTHMOD) was assembled and used to simulate runoff and ponding for the selected site for the same period. Field studies confirmed that the formation of shallow ponds in the spring was almost wholly related to rapid melting of snow and surface runoff of snowmelt. Runoff was observed to collect in all depressions of any significant size. Ponds less than 40 cm deep disappeared by mid June through a combination of evaporation and infiltration. Ponds greater than 40 cm deep generally persisted throughout the summer. Such ponds occupied lower positions in the landscape and most were in contact with a water table.

A dynamic, vertically integrated, three-dimensional, mass continuity, computer model of the Greenland ice sheet is used to predict the ice sheet's response to climatic perturbation. The ice flow is gravity driven according to glaciological physics, whereby ice flow is claculated as the sum of deformation and sliding components averaged over the ice thickness where longitudinal stresses are considered negligible. The model has inputs of the present-day ice surface and basal topography, and is forced by changes in sea level and surface mass balance, which are modelled by separately described and forced accumulation and ablation parts. The mass balance and the forcing terms are parametized against present-day values and the present-day ice sheet is initially used as the starting condition for model experiments. By repeating model experiments from the same starting point and varying only one model parameter at a time the model's sensitivity to individual parameters is assessed. Characteristic behaviour patterns, reaction times and significant parameters are identified. The model is seen to produce a realistic simulation of present-day conditions. The ice dynamics model is robust compared to the changes that can be introduced by small variations in sea-level. The relationship between forcing temperatures and ablation rates exerts most control on the model. Confidence in the model allows a series of predictive runs to be undertaken to smiulate three types of glacial fluctuation: Short term change, behaviour under maximum ice conditions and deglaciation trends. Whilst climatic forcing is important in driving the model overall, topographic effects and the influences of calving are crucial to understanding maximum state conditions and retreat behaviour. In each of the three cases, the model results corroborate well with what is known about the real world. These matches and the closeness with which the present-day conditions are simulated are mutually supportive to the conclusion that the model is effective and realistic in the way long term ice sheet change in Greenland is represented. The theoretical processes and the model results are considered to describe real processes and events. Modelling, in conjunction with field techniques, is seen as an powerful means of understanding nature.