Geochemical Fingerprinting of Coltan Ores by Machine Learning on Uneven Datasets

Two modern machine learning techniques, Linear Programming Boosting (LPBoost) and Support Vector Machines (SVMs), are introduced and applied to a geochemical dataset of niobium–tantalum (“coltan”) ores from Central Africa to demonstrate how such information may be used to distinguish ore provenance, i.e., place of origin. The compositional data used include uni- and multivariate outliers and elemental distributions are not described by parametric frequency distribution functions. The “soft margin” techniques of LPBoost and SVMs can be applied to such data. Optimization of their learning parameters results in an average accuracy of up to c. 92%, if spot measurements are assessed to estimate the provenance of ore samples originating from two geographically defined source areas. A parameterized performance measure, together with common methods for its optimization, was evaluated to account for the presence of uneven datasets. Optimization of the classification function threshold improves the performance, as class importance is shifted towards one of those classes. For this dataset, the average performance of the SVMs is significantly better compared to that of LPBoost.     

On trend detection

A main obstacle to trend detection in time series occurs when they are autocorrelated. By reducing the effective sample size of a series, autocorrelation leads to decreased trend significance. Numerous recipes attempt to mitigate the effect of autocorrelation, either by adjusting for the reduced effective sample size or by removing the autocorrelated components of a series. This short note deals with the latter, also called prewhitening (PW). It is known that removal of autocorrelation also removes part of the trend, which may affect the signal‐to‐noise ratio. Two popular methods have dealt with this problem, the trend‐free prewhitening (TFPW) and the iterative prewhitening. Although it is generally accepted that both methods reduce the adverse effects of PW on the trend magnitude, corresponding effects on statistical significance have not been clearly stated for TFPW. Using a Monte Carlo approach, it is demonstrated that both methods entail quite different Type‐I error rates. The iterative prewhitening produces rates that are generally close to the nominal significance level. The TFPW, however, shows very high Type‐I error rates with increasing autocorrelation. The corresponding rate of false trend detections is unacceptable for applications, so that published trends based on TFPW need to be reassessed.     

Interpretation of SOFIE PMC measurements: Cloud identification and derivation of mass density,particle shape,and particle size

The Solar Occultation For Ice Experiment (SOFIE) was launched onboard the Aeronomy of Ice in the Mesosphere (AIM) spacecraft to measure polar mesospheric clouds (PMCs) and their environment. This work describes methods for identifying PMCs in SOFIE observations and determining mass density, particle shape, particle effective radius, and the parameters of a Gaussian size distribution. Results using SOFIE measurements from the northern summer of 2007 are compared with concurrent observations by the ALOMAR lidar in northern Norway. Ice particle properties determined from SOFIE are in good agreement with the lidar results, considering the differences in instrument characteristics.     

Implications of the principle of effective stress

Acta Geotechnica - While Terzaghi justified his principle of effective stress for water-saturated soil empirically, it can be derived by means of the neutrality of the mineral with respect to...     

Reconstruction of the Late Holocene climate and environmental history from North Bolgoda Lake,Sri Lanka,using lipid biomarkers and pollen records

The catastrophic impact and unpredictability of the Indian Ocean Monsoon (IOM) over South Asia are evident from devastating floods, mudslides and droughts in one of the most densely populated regions of the globe. However, our understanding as to how the IOM has varied in the past, as well as its impact on local environments, remains limited. This is particularly the case for Sri Lanka, where erosional landscapes have limited the availability of well-stratified, high-resolution terrestrial archives. Here, we present novel data from an undisturbed sediment core retrieved from the coastal Bolgoda Lake. This includes the presentation of a revised Late Holocene age model as well as an innovative combination of pollen, source-specific biomarkers, and compound-specific stable carbon isotopes of n-alkanes to reconstruct the shifts in precipitation, salinity and vegetation cover. Our record documents variable climate between 3000 years and the present, with arid conditions c. 2334 and 2067 cal a bp. This extreme dry period was preceded and followed by more wet conditions. The high-resolution palaeoenvironmental reconstruction fills a major gap in our knowledge on the ramifications of IOM shifts across South Asia and provides insights during a time of major redistribution of dense human settlements across Sri Lanka.     

Jesenice—A new meteorite fall from Slovenia

Abstract– On April 9, 2009, at 3:00 CEST, a very bright fireball appeared over Carinthia and the Karavanke Mountains. The meteoroid entered the atmosphere at a very steep angle and disintegrated into a large number of objects. Two main objects were seen as separate fireballs up to an altitude of approximately 5 km, and witnesses reported loud explosions. Three stones were found with a total weight of approximately 3.611 kg. The measured activity of short‐lived cosmogenic radionuclides clearly indicates that two specimens result from a very recent meteorite fall. All cosmogenic radionuclide concentrations suggest a rather small preatmospheric radius of <20 cm; a nominal cosmic‐ray exposure age based on ²¹Ne is approximately 4 Ma, but the noble gas and radionuclide results in combination indicate a complex irradiation. Jesenice is a highly recrystallized rock with only a few relic chondrules visible in hand specimen and thin section. The texture, the large grain size of plagioclase, and the homogeneous compositions of olivines and pyroxenes clearly indicate that Jesenice is a L6 chondrite. The bulk composition of Jesenice is very close to the published average element concentration for L ordinary chondrites. The chondrite is weakly shocked (S3) as indicated by the undulatory extinction in olivine and plagioclase and the presence of planar fractures in olivine. Being weakly shocked and with gas retention ages of >1.7 Ga (⁴He) and approximately 4.3 Ga (⁴⁰Ar), Jesenice seems not to have been strongly affected by the catastrophic disruption of the L‐chondrite parent body approximately 500 Ma ago.