Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran

The identification of landslide-prone areas is an essential step in landslide hazard assessment and mitigation of landslide-related losses.In this study,we applied two novel deep learning algorithms,the recurrent neural network(RNN)and convolutional neural network(CNN),for national-scale landslide susceptibility mapping of Iran.We prepared a dataset comprising 4069 historical landslide locations and 11 conditioning factors(altitude,slope degree,profile curvature,distance to river,aspect,plan curvature,distance to road,distance to fault,rainfall,geology and land-sue)to construct a geospatial database and divided the data into the training and the testing dataset.We then developed RNN and CNN algorithms to generate landslide susceptibility maps of Iran using the training dataset.We calculated the receiver operating characteristic(ROC)curve and used the area under the curve(AUC)for the quantitative evaluation of the landslide susceptibility maps using the testing dataset.Better performance in both the training and testing phases was provided by the RNN algorithm(AUC=0.88)than by the CNN algorithm(AUC=0.85).Finally,we calculated areas of susceptibility for each province and found that 6%and 14%of the land area of Iran is very highly and highly susceptible to future landslide events,respectively,with the highest susceptibility in Chaharmahal and Bakhtiari Province(33.8%).About 31%of cities of Iran are located in areas with high and very high landslide susceptibility.The results of the present study will be useful for the development of landslide hazard mitigation strategies.     

Optimum structural design of full-scale steel buildings using drift-tribe-charged system search

Earthquake Engineering and Engineering Vibration - In this paper, the potential of utilizing improved metaheuristic approaches in optimal design of building structures is concerned. In this regard,...     

Early Mesozoic sedimentary‒tectonic evolution of the Central-East Iranian Microcontinent: Evidence from a provenance study of the Nakhlak Group

In Central Iran, the mixed siliciclastic?carbonate Nakhlak Group of Triassic age is commonly seen to have a Cimmerian affinity, although it shows considerable resemblances with the Triassic Aghdarband Group in far northeastern Iran, east of Kopeh-Dagh area, with Eurasian affinity. The Nakhlak Group is composed of the Alam (Late Olenekian?Anisian), Baqoroq (Late Anisian??Early Ladinian), and Ashin (Ladinian??Early Carnian) formations consisting mainly of volcanoclastic sandstone and shale and fossiliferous limestone. The Baqoroq Formation contains also metamorphic detritus. Sandstone petrofacies reflect the detrital evolution from active volcanism to growing orogen and again active volcanism. Textural and modal analyses of volcanic lithic fragments from the Alam Formation reflect the eruption style and magma composition of a felsic to intermediate syn-sedimentary arc activity. The detrital modes of the Baqoroq Formation sediments suggest a recycled orogenic source followed by arc activity in a remnant fore-arc basin. The sandstone samples from the Ashin Formation demonstrate a continuity of felsic to intermediate arc activity. Major and trace element concentrations of the Nakhlak Group clastic samples support sediment supply from first-cycle material and felsic magmatic arc input. The enrichment in LREE, the negative Eu anomalies, and the flat HREE patterns indicate origination from the old upper continental crust and young arc material. The chemical index of alteration (CIA ～51–70 for sandstone and 64–76 for shale samples) indicates medium degrees of chemical weathering at the source. Petrographical and geochemical evidence together with facies analysis constructed the following depositional conditions for the Nakhlak Group sediments: In the Olenekian, a fore-arc shallow to deep marine depositional basin developed that later was filled by recycled and arc-related detritus and changed into a continental basin in the Anisian. Ladinian extension let to a deepening of the basin. With respect to the similarities between the Nakhlak and Aghdarband (NE Iran) groups and unusual present-day position of the Nakhlak Group with no stratigraphic connection to the surrounding area, the development of first a fore-arc basin and later change into a back-arc depositional basin in close relation with the Aghdarband basin at the southern Eurasian active margin in the Triassic are here proposed. Understanding the basin development recorded in the Nakhlak Group provides constraints on the closure history of Palaeotethys and of the tectonic evolution of early Mesozoic basins at the southern Eurasian margin before the Cimmerian Orogeny.     

Numerical Analysis of Pile–Soil–Pile Interaction in Pile Groups with Batter Piles

Battered piles are usually used to counteract lateral forces in a pile group. As there is little spacing between piles, they are affected by one another, and there is interaction between them. In this study, pile–soil–pile interaction in a group of battered piles was numerically simulated using finite element analysis. Double and frictional pile groups under static lateral and axial loadings were analyzed separately. The effects of batter angle, slenderness ratio, spacing between piles, pile–soil stiffness ratio, and soil plasticity on interaction factors were computed and presented in curves.     

Regional improvement of global geopotential models using GPS/Leveling data

Studia Geophysica et Geodaetica - Global geopotential models are widely used in the remove-compute-restore technique for local gravity field modeling. In this paper, a method for regional...     

A Fuzzy Rule-Based Approach to Address Uncertainty in Risk Assessment and Prediction of Blast-Induced Flyrock in a Quarry

Strict control of the environmental impacts of blasting operations needs to be completely in line with the regulatory limits. In such operations, flyrock control is of high importance especially due to safety issues and the damages it may cause to infrastructures, properties as well as the people who live within and around the blasting site. Such control causes flyrock to be limited, hence significantly reducing the risk of damage. This paper serves two main objectives: risk assessment and prediction of flyrock. For these objectives, a fuzzy rock engineering system (FRES) framework was developed in this study. The proposed FRES was able to efficiently evaluate the parameters that affect flyrock, which facilitate decisions to be made under uncertainties. In this study, the risk level of flyrock was determined using 11 independent parameters, and the proposed FRES was capable of calculating the interactions among these parameters. According to the results, the overall risk of flyrock in the studied case (Ulu Tiram quarry, located in Malaysia) was medium to high. Hence, the use of controlled blasting method can be recommended in the site. In the next step, three optimization algorithms, namely genetic algorithm (GA), imperialist competitive algorithm (ICA) and particle swarm optimization (PSO), were used to predict flyrock, and it was found that the GA-based model was more accurate than the ICA- and PSO-based models. Accordingly, it is concluded that FRES is a very useful for both risk assessment and prediction of flyrock.

     

Mediterranean Sea level trends from long-period tide gauge time series

The main purpose of this study is to highlight, on the basis of statistical tests, the significant long-term changes of the Mediterranean Sea level, through the analysis of historical tide gauge records. In this framework, 14 tide gauge monthly series selected from the Permanent Service of the Mean Sea Level （PSMSL） database were used. The search for the presence or not of trends within these series, that have a temporal coverage from 59 to 142 years, was carried out using the Mann-Kendall test and the Sen''s slope estimator. The obtained results show that the Split Rt Marjana series are the only ones which does not exhibit a significant trend. The other 13 series show significant increasing trends. This result seems sufficient to suppose the presence, in the past century, of a new climatic phase on the scale of the Mediterranean basin, where the rising sea level is one of the consequences.     

Geochemical and mineralogical pattern recognition and modeling with a Bayesian approach to hydrothermal gold deposits

The Bayesian approach is an effective method of identifying the probability of mineralogical and geochemical type (MGT) mineralization of trace elements in galena, pyrite and other distributions in ore mineralization. Monomineralic samples have been identified using a computer-based Bayesian method and exploration geochemical techniques of Au deposits for MGT. In order to employ the method, a data bank was used consisting of the results of analysis of more than 12,000 monomineralic samples collected from the main hydrothermal Au deposits in Tajikistan (a territory of CIS). The Bayesian approach applied to geochemical data, such as posterior probabilities and discriminant analysis, provide numerical and graphical means through which the relationships between the trace elements and samples can be studied. The method used here, along with GIS, to find MGT can be used as geochemical indicators of regions with Au mineralization. The results of analyzing 100 monomineralic samples of pyrite from the Au–Ag Shkolnoe deposit (Tajikistan) show a multi-MGT anomaly superposition which is a combination of three MGT: (1) Au–Ag type (85% and more), (2) Au–sulfide-polymetallic type (46%), and (3) Au–sulfide type (40%). Mineralogical and geochemical maps (MGM) can be drawn based on results of MGT anomalies in a GIS environment. These maps can replace traditional metallogenic maps. The advantage of MGM substitutions is that a qualitative tool is replaced by a quantitative one. This helps one to make optimal managerial and more economical decisions.     

A destructuration theory and its application to SANICLAY model

Many natural clays have an undisturbed shear strength in excess of the remoulded strength. Destructuration modeling provides a means to account for such sensitivity in a constitutive model. This paper extends the SANICLAY model to include destructuration. Two distinct types of destructuration are considered: isotropic and frictional. The former is a concept already presented in relation to other models and in essence constitutes a mechanism of isotropic softening of the yield surface with destructuration. The latter refers to the reduction of the critical stress ratio reflecting the effect of destructuration on the friction angle, and is believed to be a novel proposition. Both the types depend on a measure of destructuration rate expressed in terms of combined plastic volumetric and deviatoric strain rates. The SANICLAY model itself is generalized from its previous form by additional dependence of the yield surface on the third isotropic stress invariant. Such a generalization allows to obtain as particular cases simplified model versions of lower complexity including one with a single surface and associative flow rule, by simply setting accordingly parameters of the generalized version. A detailed calibration procedure of the relatively few model constants is presented, and the performance of three versions of the model, in descending order of complexity, is validated by comparison of simulations to various data for oedometric consolidation followed by triaxial undrained compression and extension tests on two structured clays. Copyright © 2009 John Wiley & Sons, Ltd.     

Stochastic methods for safety assessment of the flood defense system in the Scheldt Estuary of the Netherlands

In this study, we address the effective method to apply a novel reliability method integrated with finite element models to the safety assessment of pilot site Scheldt in the Netherlands. This site was considered as one of the three main pilot sites in Europe to assess the application of newly suggested techniques in order to reduce and manage the flood risk in the Floodsite project. , 2004–2009). The novel method of dynamic bounds (DB) is applied to this site after a successful experience in (Rajabalinejad in Reliability methods for finite element models, 1 edn. IOS Press, Amsterdam, 2009). In this study, the bi-functional response of the finite element model is considered, and the dimensional uncertainty is defined presenting the expected uncertainty for a certain dimension in the DB method. The uncertainty is used as a judgment tool to choose the dimension for the DB method for the desired accuracy. The results obtained by applying this technique are presented in this paper.