Attribute uncertainty modelling in lunar spatial data

Abstract

A novel methodology to evaluate uncertainties in lunar elemental abundances is presented. Contrary to most terrestrial applications, lunar geographic information system (GIS) data cannot be verified by in situ measurements because of the limited number of ground control points and their reduced spatial extent. This work reports on investigations to evaluate the uncertainty in lunar abundance measurements without the use of ground checks but by statistical evaluation and comparison of datasets. We find that by using correlations between the major elements, charts of estimated oxygen abundances can be derived that can then be used to evaluate uncertainties in the soil composition. Our results show that the majority of the derived values for oxygen fit well to the measured ones. We also found that the patterns of cells derived from our model and the orbital measurements are concentrated mainly at locations in the near–side mare.

Acknowledgements

We thank the Hong Kong Polytechnic University for support of this study. The datasets that were used were produced by the Lunar Prospector mission and are disclosed by the National Aeronautics and Space Administration (NASA) and Washington University in St Louis on the website of the PDS Geoscience Node (http://pds-geosciences.wustl.edu/missions/lunarp/reduced_special.html). Making this information freely available to the scientific community is an example that we hope will be followed by others in the coming ‘Lunar Decade’ and we would like to express our gratitude to the team that developed these products. P. Weiss also thanks E. Seguin from COMEX S.A. for the help and hints throughout the development of the lunar GIS.