Short-term rockburst risk prediction using ensemble learning methods

Natural Hazards - Short-term rockburst risk prediction plays a crucial role in ensuring the safety of workers. However, it is a challenging task in deep rock engineering as it depends on many...     

Effect of coastline configuration on precipitation distribution in coastal zones

The effects of coastline, together with the other geographical and topographical factors such as longitude, latitude, elevation and distance from sea, on mean annual precipitation are investigated for the coastal area of the eastern Black Sea region. In this location, mountains run parallel to the sea coast. The coastline configuration is represented by a new variable, the coastline angle. A multiple linear regression (MLR) analysis is performed on mean annual precipitation recorded in coastal and inland gauges. In the study area, three variables—longitude, latitude and coastline angle—can explain 81% of the spatial variability of precipitation for the coastal gauges and 93% of that for the inland gauges. When the entire study area considered, without grouping the gauges, 80% of the spatial variability is explained by these variables. Copyright © 2009 John Wiley & Sons, Ltd.     

A statistical assessment on international landslide literature (1945–2008)

In the present study, the analysis of the international landslide literature is aimed. The landslide is perhaps one of the most complex natural phenomena. Also, due to landslides, a considerable amount of loss of lives and economic losses is encountered throughout the world. Although a vast amount of landslide papers is published in international journals, a statistical assessment on this literature is not encountered. For this reason, this study is performed. For the assessment of the international landslide literature, the Science Citation Index Expanded (WOS; Web of Science) published by Institute of Scientific Information (now Thomson Scientific), USA for the period 1945–2008, is considered. A total of 3,468 publications are found and this data is stored into Oracle XE database and queried by using Structured Query Language and Procedural Language/Structured Query Language. In the following stages, some statistical analyses are performed and the possible trends are discussed.     

Stratigraphic organisation,spatial distribution,palaeoenvironmental reconstruction,and demise of Lower Cretaceous (Barremian-lower Aptian) carbonate platforms of the Western Pontides (Black Sea region,Turkey)

Barremian-Lower Aptian platform carbonates (“Urgonian limestones”) of the northern margin of the Istanbul zone extend from Zonguldak to the Kurucasile area along the Black Sea coast. New stratigraphic data on the “Inpiri” Formation of the Inpiri-Kurucasile area are based on the identification of calcareous algae, foraminifera, and rudists. They show that this lithostratigraphic unit is stratigraphically and lithologically equivalent to the Ökü?medere Formation from Zonguldak. Some of the biostratigraphic markers are reported for the first time in Anatolia. Foraminifera are represented by several forms with a significant biostratigraphic potential used to distinguish the Barremian from the lower Aptian. Lower Aptian beds also yield relatively advanced caprinid rudists.The Ökü?medere Formation is relatively thin, terrigeneous-rich, and rudist-free or rudist-poor in the Kurucasile sector, and thick, terrigeneous-poor, and rudist-rich from Amasra to Zonguldak, with a set of marker beds including either charophytes or Palorbitolina and capped by a coral unit underlying ammonite bearing marls. Terrigeneous-rich carbonates from the eastern sector are interpreted as marginal marine coastal, infralittoral environments and grade distally, northward, to marly basinal sediments. By contrast “Urgonian type” limestones from the Zonguldak-Amasra region possess a wide extent and no transition to coastal or basinal sediments has been observed. A transition from a typical platform westward to a mixed siliciclastic-carbonate ramp eastward was controlled by both the nature of the adjacent exposed area and tectonic factors affecting the overall continental margin that is a northward downwarping. The exposed area was flanked southward by a belt of coastal siliciclastics grading southward and eastward to deep water sediments of the Ulus basin. In mid-Bedoulian time, carbonate platform demise from the western region was drowned below deeper marly sediments whereas the eastern siliscilastic-carbonate ramp was buried below coastal clastics.     

Spatial distribution and assemblage structure of macrobenthos in a tidal creek in relation to industrial activities

The impact of petrochemical special economic zone activities on the health status of Jafari Creek was studied by assessing the changes in macroinvertebrate assemblages in nine sites during September 2006–January 2008. The relationship between spatial pattern of macroinvertebrate assemblages and ambient factors (i.e. water temperature, salinity, pH, dissolved oxygen, sediment grain size distribution, sediment organic content, heavy metals contents) was measured. Background enrichment indices, contamination factor and contamination degree, were used to assess the health status in the study area based on nickel, lead, cadmium and mercury contents of the sediments. The macrobenthic communities had a low diversity and were dominated by opportunistic taxa. The BIO-ENV analysis identified salinity, dissolved oxygen, pH and silt/clay content of sediments as the major environmental variables influencing the infaunal pattern. This suggests that management should attempt to ensure minimal disturbance to environmental variables underlying the spatial variation in macroinvertebrate assemblages. Background enrichment indices showed that the health of Jafari Creek has declined over time due to the constant discharge of heavy metals to the Creek system. These indices also identified a significant degree of pollution in the study area. The decrease in the ecological potential of Jafari Creek was best highlighted by the alteration in macrobenthic assemblages.     

Reinterpretation of an archaeological pebble culture from the Middle Mekong River valley,Cambodia

In 1963, E. Saurin and J.‐P. Carbonnel discovered the Sre Sbov site on an alluvial terrace of the Mekong River in central Cambodia. Saurin described a lithic typology dating to the Lower/Middle Pleistocene from this site. Although the original lithic assemblage has been lost, this typology has been used continuously as a reference by Southeast Asian prehistorians. In 2007, a Khmer–French team conducted excavations at Sre Sbov that yielded numerous pebbles and cobbles showing apparently convincing handmade removals, as Saurin had previously described. However, an in‐depth study of this assemblage, combined with a geological survey of the area, led to the conclusion that the stones were, in fact, of natural origin, and that for this reason their typology should be disregarded. Using satellite imagery and geological surveys, we explain how such a misinterpretation may have occurred and define a “buffer zone,” corresponding to the maximal extent of the proto‐Mekong River, where fluvially reworked pebbles and cobbles resembling artifacts may be recovered. © 2009 Wiley Periodicals, Inc.     

Stochastic Modeling Approach for the Evaluation of Backbreak due to Blasting Operations in Open Pit Mines

Backbreak is an undesirable side effect of bench blasting operations in open pit mines. A large number of parameters affect backbreak, including controllable parameters (such as blast design parameters and explosive characteristics) and uncontrollable parameters (such as rock and discontinuities properties). The complexity of the backbreak phenomenon and the uncertainty in terms of the impact of various parameters makes its prediction very difficult. The aim of this paper is to determine the suitability of the stochastic modeling approach for the prediction of backbreak and to assess the influence of controllable parameters on the phenomenon. To achieve this, a database containing actual measured backbreak occurrences and the major effective controllable parameters on backbreak (i.e., burden, spacing, stemming length, powder factor, and geometric stiffness ratio) was created from 175 blasting events in the Sungun copper mine, Iran. From this database, first, a new site-specific empirical equation for predicting backbreak was developed using multiple regression analysis. Then, the backbreak phenomenon was simulated by the Monte Carlo (MC) method. The results reveal that stochastic modeling is a good means of modeling and evaluating the effects of the variability of blasting parameters on backbreak. Thus, the developed model is suitable for practical use in the Sungun copper mine. Finally, a sensitivity analysis showed that stemming length is the most important parameter in controlling backbreak.     

Shaping the Kuiper belt size distribution by shattering large but strengthless bodies

The observed size distribution of Kuiper belt objects (KBOs)—small icy and rocky Solar System bodies orbiting beyond Neptune—is well described by a power law at large KBO sizes. However, recent work by Bernstein et al. (2004, Astron. J. 128, 1364-1390) indicates that the size distribution breaks and becomes shallower for KBOs smaller than about 70 km in size. Here we show that we expect such a break at KBO radius ∼40 km since destructive collisions are frequent for smaller KBOs. Specifically, we assume that KBOs are gravity-dominated bodies with negligible material strength. This gives a power-law slope q?3 where the number N>r of KBOs larger than a size r is given by N>r∝r1−q; the break location follows from this slope through a self-consistent calculation. The existence of this break, the break's location, and the power-law slope we expect below the break are consistent with the findings of Bernstein et al. (2004, Astron. J. 128, 1364-1390). The agreement with observations indicates that KBOs as small as ∼40 km are effectively strengthless.     

Fate and biodegradability potential of an emerging micropollutant diclofenac in subsurface environment

The present study deals with the assessment of the fate of an emerging contaminant (diclofenac), present in the subsurface environment at specific concentration levels, and its biodegradation potential under different environmental configurations. Diclofenac is a widely consumed nonsteroidal anti-inflammatory drug used as a painkiller prescribed as pills or ointments and among the most frequently detected pharmaceuticals in wastewater treatment plants’ effluents. Fermentative/methanogenic cultures were enriched from previously contaminated sediment samples. Although partial biodegradation was achieved in all enrichment cultures, complete degradation was not observed. The results reflected that less than 6% of the diclofenac sorbed on the solid phase and the rest was either biotransformed (25–40%) or remained in the liquid phase (55–70%) at the end of 45-day incubation period. Although partial cometabolic degradation of diclofenac occurred in the presence of glucose, biodegradation was not observed in the presence of readily biodegradable carbon source (i.e., acetate). Diclofenac concentration up to 1000 µg/L did not affect the methanogenic activity of the enriched culture. The results of this study will have a significant impact in the designation of the permitted concentration limits of diclofenac before the discharge through wastewater treatment plants.     

Hot spot analysis based on network spatial weights to determine spatial statistics of traffic accidents in Rize,Turkey

The increased numbers of vehicles using roads in the world today are cause of traffic-related problems, and in this respect, road traffic accidents are an important topic relating to public health. Especially on the road connecting two border provinces, traffic accidents are increasing substantially in parallel with the quantity of transport facilities. By determining areas where traffic accidents result in deaths or injuries, accident prevention strategies can be developed. This study applies the spatial statistics techniques using Geographical Information Systems (GIS) to determine the intensity of traffic accidents (hot-spot regions) over 45 km of main routes in Rize Province, Turkey. Traffic accidents recorded in data spanning 5 years are combined with a geographical dataset for evaluation using hot spot statistical analysis. Unlike other studies, this study used hot spot analysis based on network spatial weights (an innovative review in the methods of determining traffic accident hot spots: “novel application of GIScience”) to identify black spots for traffic safety. To perform the analysis using Hot Spot Analysis: Getis-Ord Gi*, a generated network dataset and the spatial weights of the road data are used to generate network spatial weights. Then, Kernel Density method is used to define traffic accident black spots. Finally, these two methods are compared each other with visually.