Markov-random-field-based super-resolution mapping for identification of urban trees in VHR images

Identification of tree crowns from remote sensing requires detailed spectral information and submeter spatial resolution imagery. Traditional pixel-based classification techniques do not fully exploit the spatial and spectral characteristics of remote sensing datasets. We propose a contextual and probabilistic method for detection of tree crowns in urban areas using a Markov random field based super resolution mapping (SRM) approach in very high resolution images. Our method defines an objective energy function in terms of the conditional probabilities of panchromatic and multispectral images and it locally optimizes the labeling of tree crown pixels. Energy and model parameter values are estimated from multiple implementations of SRM in tuning areas and the method is applied in QuickBird images to produce a 0.6 m tree crown map in a city of The Netherlands. The SRM output shows an identification rate of 66% and commission and omission errors in small trees and shrub areas. The method outperforms tree crown identification results obtained with maximum likelihood, support vector machines and SRM at nominal resolution (2.4 m) approaches.     

The remarkable Re–Os chronometer in molybdenite: how and why it works

The Re–Os (rhenium–osmium) chronometer applied to molybdenite (MoS₂) is now demonstrated to be remarkably robust, surviving intense deformation and high‐grade thermal metamorphism. Successful dating of molybdenite is dependent on proper preparation of the mineral separate and analysis of a critical quantity of molybdenite, unique to each sample, such that recognized spatial decoupling of ¹⁸⁷Re parent and ¹⁸⁷Os daughter within individual molybdenite crystals is overcome. Highly precise, accurate and reproducible age results are derived through isotope dilution and negative thermal ion mass spectrometry (ID‐NTIMS). Spatial decoupling of parent–daughter precludes use of the laser ablation ICP‐MS microanalytical technique for Re–Os dating of molybdenite. The use of a reference or control sample is necessary to establish laboratory credibility and for interlaboratory comparisons. The Rb–Sr, K–Ar and ⁴⁰Ar/³⁹Ar chronometers are susceptible to chemical and thermal disturbance, particularly in terranes that have experienced subsequent episodes of hydrothermal/magmatic activity, and therefore should not be used as a basis for establishing accuracy in Re–Os dating of molybdenite, as has been done in the past. Re–Os ages for molybdenite are almost always in agreement with observed geological relationships and, when available, with zircon and titanite U–Pb ages. For terranes experiencing multiple episodes of metamorphism and deformation, molybdenite is not complicated by overgrowths as is common for some minerals used in U–Pb dating (e.g. zircon, monazite, xenotime), nor are Re and Os mobilized beyond the margins of individual crystals during solid‐state recrystallization. Moreover, inheritance of older molybdenite cores, incorporation of common Os, and radiogenic Os loss are exceedingly rare, whereas inheritance, common Pb and Pb loss are common complications in U–Pb dating techniques. Therefore, molybdenite ages may serve as point‐in‐time markers for age comparisons.     

Calendar year age estimates of Allerød–Younger Dryas sea‐level oscillations at Os,western Norway

A detailed shoreline displacement curve documents the Younger Dryas transgression in western Norway. The relative sea‐level rise was more than 9 m in an area which subsequently experienced an emergence of almost 60 m. The sea‐level curve is based on the stratigraphy of six isolation basins with bedrock thresholds. Effort has been made to establish an accurate chronology using a calendar year time‐scale by ¹⁴C wiggle matching and the use of time synchronic markers (the Vedde Ash Bed and the post‐glacial rise in Betula (birch) pollen). The sea‐level curve demonstrates that the Younger Dryas transgression started close to the Allerød–Younger Dryas transition and that the high stand was reached only 200 yr before the Younger Dryas–Holocene boundary. The sea level remained at the high stand for about 300 yr and 100 yr into Holocene it started to fall rapidly. The peak of the Younger Dryas transgression occurred simultaneously with the maximum extent of the ice‐sheet readvance in the area. Our results support earlier geophysical modelling concluding a causal relationship between the Younger Dryas glacier advance and Younger Dryas transgression in western Norway. We argue that the sea‐level curve indicates that the Younger Dryas glacial advance started in the late Allerød or close to the Allerød–Younger Dryas transition. Copyright © 2004 John Wiley & Sons, Ltd.     

Gaussian approximations to conditional distributions for multi-Gaussian processes

Suppose a multi-Gaussian process is observed at some set of sites, and we wish to obtain the conditional block grade distribution given some observations. We show that this conditional distribution is approximately Gaussian under certain conditions. In particular, given a single observation from a continuous multi-Gaussian process, the conditional distribution under a small change of support is approximately Gaussian unless, roughly speaking, the observed process is twice differentiable and the observation site is at the center of mass of the support region. A Gaussian approximation for the conditional prediction error of the total ore in a fixed region is considered also, although an example demonstrates that a naive analysis can give incorrect limiting conditional means.     

An earthquake swarm on the Chagos—Laccadive Ridge and its tectonic implications

Summary. An unusual, isolated swarm of earthquakes occurred on the Chagos-Laccadive Ridge in the Central Indian Ocean between 1965 and 1970. Sixteen earthquakes were "located at approximately 6.0° S, 71.3°E on the steep west facing scarp of the Chagos Bank. This swarm forms the only major seismic activity in the period 1963–76 on the otherwise aseismic ridge. The mechanisms of the three largest earthquakes were studied using body- and surface-wave data. All are extremely similar shallow normal fault events on an east—west fault plane. Such faulting is difficult to reconcile with that expected from either the trend of the Chagos-Laccadive Ridge or the spreading direction at the nearby Central Indian Ridge. The swarm may have occurred at depth on a cross fracture remaining from the breakup of the Chagos Bank and the Mascarene Plateau and the formation of the present Central Indian Ridge.     

Some asymptotic properties of kriging when the covariance function is misspecified

The impact of using an incorrect covariance function on kriging predictors is investigated. Results of Stein (1988) show that the impact on the kriging predictor from not using the correct covariance function is asymptotically negligible as the number of observations increases if the covariance function used is compatible with the actual covariance function on the region of interestR. The definition and some properties of compatibility of covariance functions are given. The compatibility of generalized covariances also is defined. Compatibility supports the intuitively sensible concept that usually only the behavior near the origin of the covariance function is critical for purposes of kriging. However, the commonly used spherical covariance function is an exception: observations at a distance near the range of a spherical covariance function can have a nonnegligible effect on kriging predictors for three-dimensional processes. Finally, a comparison is made with the perturbation approach of Diamond and Armstrong (1984) and some observations of Warnes (1986) are clarified.     

Large earthquakes kill coral reefs at the north-west Gulf of Aqaba

Down‐faulting at the north‐west margins of the Gulf of Aqaba is inferred to have triggered a catastrophic sedimentary event at 2.3 ka that killed the Elat fringing coral reef. Whereas segments of the Holocene reef were perfectly fossilized and preserved beneath a veneer of siliciclastic sediments, other segments were abraded, settled by nomads, and later re‐submerged under 4 m of water. Repeated damage triggered by down‐throwing earthquakes degenerate the fringing reefs of the north‐west end of the gulf. Conversely, on the north‐eastern and southern parts of the gulf, where earthquakes uplift the margins, modern reefs are thriving, attached to uplifted fossil reef terraces. Therefore, coastal subsidence moderates the development of fringing coral reefs during the late Holocene sea‐level stand still.     

Granitic dimensional stones in Uruguay: evaluation and assessment of potential resources

In Uruguay commercial granite varieties comprise mafic rocks, granitoids, and syenitoids. There is a long tradition in Uruguay, as well as worldwide, of using dimensional stones in architecture and art, specially granitic ones. Some of the present applications of these dimensional stones are as façade cladding, countertops, and outdoor and indoor floor slabs. The color spectrum of the Uruguayan granitic dimensional stones varies from black to light gray, covering a wide variety of red and pink and minor greenish-gray. The décor of these granitic dimensional stones is mainly determined by their fabric, fundamentally the grain size and the color distribution between the different minerals that compose the rocks. In the present research the most important commercial granites were sampled to analyze their petrography and petrophysical properties. A detailed structural analysis has been performed in several deposits, as well as the application of the software 3D Block Expert for modeling the possible raw block size distribution. Other factors controlling the mining viability of the deposits were also studied (e.g., homogeneity/heterogeneity of color and décor) and the possible reserves were calculated.     

Use of GIS for a spatial and temporal analysis of Kenyan wildlife with generalised linear modelling

This paper applies generalised linear statistical techniques in a GIS to analyse wildlife data from a Kenyan wildlife reserve and its surrounding areas. Attention focuses on the spatial distribution of elephant during nine successive surveys, analysing their temporal and spatial relationship to 12 environmental covariates. A principal component analysis identifies five major determining factors, thereby reducing dimensionality in the data, while a simple spatial analysis procedure, suitable for wildlife data obtained from airborne surveys, quantfies clustering for different animal species. The number of explanatory variables appearing in abundance models is found to be subject to large variations during successive surveys with a minimum and maximum of four and eight variables, respectively. Species from highly clustered populations are found to have over 20 times more observations within short distances compared to the rest. The study concludes that a combination of generalised linear modelling and GIS gives deeper insight into the dynamics of wildlife species in and around well-defined nature reserves.     

Spatial variability in classification accuracy of agricultural crops in the Dutch national land-cover database

Variability in per cell classification accuracy is predominantly modelled with land-cover class as the explanatory variable, i.e. with users' accuracies from the error matrix. Logistic regression models were developed to include other explanatory variables: heterogeneity in the 3×3 window around a cell, the size of the patch and the complexity of the landscape in which a cell is located. It was found that per cell, the probability of correct classification was significantly (α=0.05) higher for cells with a less heterogeneous neighbourhood, for cells part of larger patches and for cells in regions with a less heterogeneous landscape. To validate the models, a leave-one-out procedure was applied in which the absolute difference between the actual and the model-estimated number of cells correctly classified was summarized over 55 regions in the Netherlands. The sum of differences reduced from 60.9 to 48.1 after adding the variables ‘patch size’ and ‘landscape dominance’ to the land-cover class model. Spatial variability thus modelled therefore led to a substantial improvement in the estimation of the per cell classification accuracy.