An evaluation of Guided Regularized Random Forest for classification and regression tasks in remote sensing

New Earth observation missions and technologies are delivering large amounts of data. Processing this data requires developing and evaluating novel dimensionality reduction approaches to identify the most informative features for classification and regression tasks. Here we present an exhaustive evaluation of Guided Regularized Random Forest (GRRF), a feature selection method based on Random Forest. GRRF does not require fixing a priori the number of features to be selected or setting a threshold of the feature importance. Moreover, the use of regularization ensures that features selected by GRRF are non-redundant and representative. Our experiments based on various kinds of remote sensing images, show that GRRF selected features provides similar results to those obtained when using all the available features. However, the comparison between GRRF and standard random forest features shows substantial differences: in classification, the mean overall accuracy increases by almost 6% and, in regression, the decrease in RMSE almost reaches 2%. These results demonstrate the potential of GRRF for remote sensing image classification and regression. Especially in the context of increasingly large geodatabases that challenge the application of traditional methods.     

Strain localization in sand: an overview of the experimental results obtained in Grenoble using stereophotogrammetry

Experimental results are presented from the extensive program of drained plane strain compression tests on sand carried out in Grenoble over the last two decades. Systematic analysis of photographs of the deforming specimen allowed for measuring deformations and determining strain fields throughout the test, that is: prior to, at, and after the onset of strain localization. The principles, details and accuracy of the procedure are described, as well as its suitability to properly depict the patterns of deformation. Findings concerning the occurrence and progression of strain localization are discussed. The issues of shear band orientation and thickness are addressed, as well as temporary and persistent complex localization patterns, and the volumetric behaviour inside a band after its formation. The influence of such variables as initial state of the sand (effective stress and relative density), specimen size and slenderness, as well as grain size, is discussed. Copyright © 2004 John Wiley & Sons, Ltd     

Testing the circumstellar disc hypothesis: a search for H2 outflow signatures from massive young stellar objects with linearly distributed methanol masers

The results of a survey searching for outflows using near-infrared imaging are presented. Targets were chosen from a compiled list of massive young stellar objects associated with methanol masers in linear distributions. Presently, it is a widely held belief that these methanol masers are found in (and delineate) circumstellar accretion discs around massive stars. If this scenario is correct, one way to test the disc hypothesis is to search for outflows perpendicular to the methanol maser distributions. The main objective of the survey was to obtain wide-field near-infrared images of the sites of linearly distributed methanol masers using a narrow-band 2.12-μm filter. This filter is centred on the  H₂ v = 1–0 S(1)  line; a shock diagnostic that has been shown to successfully trace CO outflows from young stellar objects. 28 sources in total were imaged of which 18 sources display H₂ emission. Of these, only two sources showed emission found to be dominantly perpendicular to the methanol maser distribution. Surprisingly, the H₂ emission in these fields is not distributed randomly, but instead the majority of sources are found to have H₂ emission dominantly parallel to their distribution of methanol masers. These results seriously question the hypothesis that methanol masers exist in circumstellar discs. The possibility that linearly distributed methanol masers are instead directly associated with outflows is discussed.     

Hydrated sulphuric acid in dense molecular clouds

We consider sulphur depletion in dense molecular clouds, and suggest hydrated sulphuric acid, H₂SO₄ ·  n H₂O, as a component of interstellar dust in icy mantles. We discuss the formation of hydrated sulphuric acid in collapsing clouds and its instability in heated regions in terms of the existing hot core models and observations. We also show that some features of the infrared spectrum of hydrated sulphuric acid have correspondence in the observed spectra of young stellar objects.     

The FAST/SKA site selection in Guizhou province

Many karst depressions with diameters of 300 m to 500 m, suitable for constructing Arecibo-style radio telescopes, were identified in the south of Guizhou Province by Remote Sensing (RS) and Geographic Information System (GIS) technologies together with field investigations. Fundamental topography and landform databases were established for 391candidate depressions, and using GIS the3-dimensional images of depressions, at a scale of 1:10000, were then simulated to fit a spherical antenna. This revised version was published online in July 2006 with corrections to the Cover Date.     

Ab initio infra-red spectra of iron-polycyclic aromatic hydrocarbon compounds: a model case

C₁₀H₆-Fe⁺ complexes were observed by Fourier transform mass spectrometry by Marty et al. (Chem. Phys. Lett. 256 (1996) 669). In this article, we present the infra-red (IR) spectra of the two isomers calculated by ab initio methods of quantum chemistry. In the past, the same approach has predicted successfully unexpected features of the IR spectra of polycyclic aromatic hydrocarbon (PAH) cations which were confirmed by measurements in matrices (J. Phys. Chem. 98 (1994) 9187). The C₁₀H₆-Fe⁺ systems are models for larger PAHs-Fe⁺ complexes, which are believed to play a key rôle in space chemistry.     

A stepwise method for determining the number of component distributions in a mixture

This paper considers the problem of estimatingm, the number of components in a finite mixture of distributions from a parametric family. A step-up procedure using the bootstrap method is proposed. Some properties of the procedure are illustrated with simulation studies. An example of the method, applied to orientation of beach clasts, is given.     

Screening and ranking framework for geologic CO<Subscript>2</Subscript> storage site selection on the basis of health,safety, and environmental risk

A screening and ranking framework (SRF) has been developed to evaluate potential geologic carbon dioxide (CO₂) storage sites on the basis of health, safety, and environmental (HSE) risk arising from CO₂ leakage. The approach is based on the assumption that CO₂ leakage risk is dependent on three basic characteristics of a geologic CO₂ storage site: (1) the potential for primary containment by the target formation; (2) the potential for secondary containment if the primary formation leaks; and (3) the potential for attenuation and dispersion of leaking CO₂ if the primary formation leaks and secondary containment fails. The framework is implemented in a spreadsheet in which users enter numerical scores representing expert opinions or published information along with estimates of uncertainty. Applications to three sites in California demonstrate the approach. Refinements and extensions are possible through the use of more detailed data or model results in place of property proxies.     

Oceanographic influences on seabirds and cetaceans of the eastern tropical Pacific: A review

This paper is part of a comprehensive review of the oceanography of the eastern tropical Pacific, the oceanic region centered on the eastern Pacific warm pool, but also including the equatorial cold tongue and equatorial current system, and summarizes what is known about oceanographic influences on seabirds and cetaceans there. The eastern tropical Pacific supports on the order of 50 species of seabirds and 30 species of cetaceans as regular residents; these include four endemic species, the world’s largest populations for several others, three endemic sub-species, and a multi-species community that is relatively unique to this ecosystem. Three of the meso-scale physical features of the region are particularly significant to seabirds and cetaceans: the Costa Rica Dome for blue whales and short-beaked common dolphins, the Equatorial Front for planktivorous seabirds, and the countercurrent thermocline ridge for flocking seabirds that associate with mixed-species schools of spotted and spinner dolphins and yellowfin tuna. A few qualitative studies of meso- to macro-scale distribution patterns have indicated that some seabirds and cetaceans have species-specific preferences for surface currents. More common are associations with distinct water masses; these relationships have been quantified for a number of species using several different analytical methods. The mechanisms underlying tropical species–habitat relationships are not well understood, in contrast to a number of higher-latitude systems. This may be due to the fact that physical variables have been used as proxies for prey abundance and distribution in species–habitat research in the eastern tropical Pacific.Though seasonal and interannual patterns tend to be complex, species–habitat relationships appear to remain relatively stable over time, and distribution patterns co-vary with patterns of preferred habitat for a number of species. The interactions between seasonal and interannual variation in oceanographic conditions with seasonal patterns in the biology of seabirds and cetaceans may account for some of the complexity in species–habitat relationship patterns.Little work has been done to investigate effects of El Niño-Southern Oscillation cycles on cetaceans, and results of the few studies focusing on oceanic seabirds are complex and not easy to interpret. Although much has been made of the detrimental effects of El Niño events on apex predators, more research is needed to understand the magnitude, and even direction, of these effects on seabirds and cetaceans in space and time.     

Genetic Diversity in Marine Molluscs: A Test of the Niche-Width Variation Hypothesis

Abstract. A test of the niche width variation hypothesis was conducted in the three Mediterranean littoral species of the family Trochidae: Monodonta turbinata, M. turbiformis and Gibbula richardi. Allozymic variation encoded by 26 gene loci was analyzed, and the levels of polymorphism (P), heterozygosity (H) and allele diversity per locus (A) were estimated and appear to support the niche-width variation hypothesis. The broad-niche species, M. turbiformis and G. richardi are more polymorphic and heterozygous than the relatively narrow-niche M. turbinata (P = 0.571; 0.731 and 0.417; H = 0.108; 0.116 and 0.078; A = 2.025; 2.489 and 1.965, respectively).