SP 44 Chapter 15: Statistical, Mathematical & Geostatistical Methods for Dealing with Glacial Dispersal Application of GIS Technology to Till Data fr
by J.R. Harris.
Requires 2 downloads. Click on 'Members Only', then 'My Information' to see available downloads after payment is processed.
This chapter is from GAC's Special Paper 44: "GIS For the Earth Sciences" editted by Dr. Jeff Harris. To purchase the entire book in hardcopy or disc formats, please see the Special Papers section of our bookstore.
ABSTRACT: Different methods for identifying directional effects, due primarily to glacial dispersal, in geochemical data sampled from till, are reviewed. These methods include mathematical, geostatistical and proximity (search) algorithms available in many Geographic Information Systems (GIS), or associated statistical and geostatistical software packages. New search methods for identifying directional effects in till data, identifying potential dispersal trains and for generating a map of potential up-ice source zones of geochemical anomalies are also presented.
These techniques are initially evaluated using simulated data characterized by varying background noise levels in which simulated ribbon and fan-shaped dispersal trains are embedded. The techniques are then applied to Cu and Zn data,sampled from till over the Swayze greenstone belt in Ontario, and Pb and Zn data sampled from till over southeastern Cape Breton Island, Nova Scotia.
The primary cause of directional biases in the till data is down-ice glacial dispersal. The effects of the orientation (strike) of stratigraphy also exerts an influence on the data, especially for Zn which commonly substitutes as a trace element in rock-forming mafic minerals.
The search methods were the most successful at identifying documented Pb and Cu dispersal trains in the Cape Breton till dataset. However, the geologist must still carefully screen the selected points (representing candidate locations for the heads of dispersal trains) and areas (up-ice source zones of till anomalies) identified by the search algorithms.
With respect to the Swayze till data, all the methods that were reviewed identified a southward directional trend in the Cu and Zn data that parallels the major ice-flow direction, as recorded by the orientation of glacial striae. Minor variations in spatial trends are also noted between the different methods that can be attributed to differences in the way the algorithms operate statistically and spatially. The Cu and Zn anomalies derived from the down-ice search algorithm (DISA) are generally better predictors of the known base-metal prospects than are the in-situ anomalies derived from the interpolated data.
Geological Association of Canada
<< Go Back