DIGITAL MAPPING, QUANTITATIVE ANALYSIS AND VISUALIZATION OF DRUMLIN MORPHOLOGY IN THE PETERBOROUGH DRUMLIN FIELD, ONTARIO

The digital mapping of glacial landforms from DEM and satellite data allows for the automated calculation of morphological variables that can be used to analyse the landforms, their spatial distribution and their relationships with other geological parameters. This study focuses on the digital mapping of drumlins within the Peterborough Drumlin field, located in southern Ontario, Canada. The drumlin field covers an area of over 5000km2 and contains many thousands of individual drumlins formed during late Wisconsin glaciation of the region. A digital map of part of the drumlin field (approximately 1800 km2) was extracted from 10m resolution DEM data made available from the Ontario Geospatial Data Exchange (OGDE). Drumlins are identified from the DEM data by extracting groups of closed contours and morphological variables such as length, width and circularity are calculated which can then be subjected to traditional and spatial statistical analyses in order to identify spatial trends in drumlin characteristics within the study area. Pearson correlation coefficients can be calculated to measure how the values of any two variables vary together in contrast to how they vary independently. For example, the drumlins analyzed in the study area show a significant positive relationship between drumlin circularity and drumlin area.

Digital maps of drumlin morphology can also be compared with the distribution of subsurface sediment and bedrock types, drift thickness and ice-flow characteristics using SPSS software. Spatial statistical techniques such as kernel density analyses, which identify regions in which certain drumlin forms are concentrated, can be displayed as three dimensional continuous maps. Data can be further tested for clustering using neighbourhood statistics.

This method of digital mapping of glacial landforms allows quantitative statistical analysis of drumlin morphology and its spatial variability across a broad area and may provide insight into the nature of subglacial processes operating below large ice sheets and the mechanisms responsible for the formation of drumlin fields. The methodology may also be applied to the analysis of other drumlin fields where adequate digital data are available, or to the spatial analysis of other landforms such as sand dunes or karst features.