ABSTRACT
Web-based citizen science often involves the classification of image features by large numbers of minimally trained volunteers, such as the identification of lunar impact craters under the Moon Zoo project. Whilst such approaches facilitate the analysis of large image data sets, the inexperience of users and ambiguity in image content can lead to contamination from false positive identifications. We give an approach, using Linear Poisson Models and image template matching, that can quantify levels of false positive contamination in citizen science Moon Zoo crater annotations. Linear Poisson Models are a form of machine learning which supports predictive error modelling and goodness-of-fits, unlike most alternative machine learning methods. The proposed supervised learning system can reduce the variability in crater counts whilst providing predictive error assessments of estimated quantities of remaining true verses false annotations. In an area of research influenced by human subjectivity, the proposed method provides a level of objectivity through the utilisation of image evidence, guided by candidate crater identifications.
ABSTRACT
ALH84001, a cataclastic cumulate orthopyroxenite meteorite from Mars, has been dated by Ar-Ar stepped heating and laser probe methods. Both methods give ages close to 3,900 Ma. The age calculated is dependent on assumptions made about 39Ar recoil effects and on whether significant quantities of 40Ar from the Martian atmosphere are trapped in the meteorite. If, as suggested by xenon and nitrogen isotope studies, Martian atmospheric argon is present, then it must reside predominantly in the K-rich phase maskelynite. Independently determined 129Xe abundances in the maskelynite can be used to place limits on the concentration of the atmospheric 40Ar. These indicate a reduction of around 80 Ma to ages calculated on the assumption that no Martian atmosphere is present. After this correction, the nominal ages obtained are: 3940 +/- 50, 3870 +/- 80, and 3970 +/- 100 Ma. by stepped heating, and 3900 +/- 90 Ma by laser probe (1 sigma statistical errors), giving a weighted mean value of 3,920 Ma. Ambiguities in the interpretation of 39Ar recoil effects and in the contribution of Martian atmospheric 40Ar lead to uncertainties in the Ar-Ar age which are difficult to quantify, but we suggest that the true value lies somewhere between 4,050 and 3,800 Ma. This age probably dates a period of annealing of the meteorite subsequent to the shock event which gave it its cataclastic texture. The experiments provide the first evidence of an event occurring on Mars coincident with the time of the late heavy bombardment of the Moon and may reflect a similar period of bombardment in the Southern Highlands of Mars. Whether the age determined bears any relationship to the time of carbonate deposition in ALH84001 is not known. Such a link depends on whether the temperature associated with the metasomatic activity was sufficient to cause argon loss from the maskelynite and/or whether the metasomatism and metamorphism were linked in time through a common heat source.
