Multiple Classifier Systems for Hyperspectral Remote Sensing Data Classification

One of the most widely used outputs of remote sensing technology is Hyperspectral image. This large amount of information can increase classification accuracy. But at the same time, conventional classification techniques are facing the problem of statistical estimation in high-dimensional space. Recently in remote sensing, support vector machines (SVMs) have shown very suitable performance in classifying high dimensionality problem. Another strategy that has recently been used in remote sensing is multiple classifier system (MCS). It can also improve classification accuracy by combining different classifier methods or by a diversity of the same classifier. This paper aims to classify a Hyperspectral data using the most common methods of multiple classifier systems i.e. adaboost and bagging and a MCS based on SVM. The data used in the paper is an AVIRIS data with 224 spectral bands. The final results show the high capability of SVMs and MCSs in classifying high dimensionality data.     

Investigating the capability of deformation rate analysis method in stress estimation: a case study of water conveyance tunnel of Gotvand Dam

Knowledge of the magnitude and orientation of the initial in situ stress of rock mass in underground spaces in mining, construction, and oil projects are so vital; hence, putting it aside could not only cost a lot rather incur some irrecoverable damage. Various methods are available to estimate in situ stress in rock mass. However, the most commonly used one, i.e., hydraulic fracturing (HF) method is considered expensive and time consuming. As a matter of fact, laboratory methods based on drilled “core” have become prevalent these days considering them simple, cheap, and quick. Taking into account one such procedure, i.e., deformation rate analysis (DRA), the current research tries to review the DRA capability in determining the magnitude of initial in situ stress is in different parts of stress–strain curve. Further, an investigation was made about the usage of DRA method for both brittle and ductile rocks. To compare the DRA and hydraulic fracturing methods in in situ stress measurement, the water conveyance tunnel of Gotvand Dam was selected as a case study. The DRA tests were conducted on core samples prepared from blocks of tuff (as brittle) and soft sandstones (as ductile) from shallow quarry. The results show that the DRA method is suitable for all types of intact rock and that this could easily estimate in situ stress values. A comparison between in situ stress values obtained by DRA and those of HF method show the feasibility of geotechnical project, simplicity, speed, and low cost.     

Fluoride pollution in soils and waters of Koohbanan region, southeastern Iran

Dental fluorosis occurs because of fluoride over-absorption during tooth calcification and maturation. We studied fluoride concentration in water and soil samples of the Koohbanan region in Kerman province of southeastern Iran and the effects of calcium chloride and gypsum treatments in decreasing the amount of fluoride in water samples of this region. The results indicate that the high amount of fluoride in the water samples of Koohbanan region is not in agreement with the recommended amount of fluoride concentration for drinking water by World Health Organization (that is 1–1/5 mg/l). Applying calcium chloride and gypsum treatments decreased the amount of fluoride in the water samples showing that utilizing calcium chloride (6 mg/l) or gypsum (12 mg/l) can lower the fluoride concentration in the water samples of Koohbanan, and thus solve the observed dental fluorosis problem.     

Heavy metal contamination in water and sediment of the Port Klang coastal area, Selangor, Malaysia

This investigation presents the temporal and spatial distribution of heavy metals (As, Cd, Cr, Cu, Ni, Pb, Hg, and Zn), in water and in sediments of Port Klang, Malaysia. Water and sediment samples were collected from 21 stations at 3-month intervals, and contamination factor $ (C_{\text{f}} ) $ and contamination degree $ (C_{\text{d}} ) $ were calculated to estimate the contamination status at the sampling stations. Cluster analysis was used to classify the stations based on the contamination sources. Results show that concentrations of As, Cd, Hg, and Pb in sediment and As, Cd, Hg, Pb, Cr, and Zn in water were significantly higher than the background values at which these metals are considered hazardous. The main sources of heavy metal contamination in Port Klang were industrial wastewater and port activities.     

Delineation of groundwater potential zones using remote sensing,GIS, and AHP technique in Tehran–Karaj plain,Iran

Evaluation of groundwater resources in dry areas without enough data is a challenging task in many parts of the world, including Tehran–Karaj plain in Iran, which includes Tehran, the capital city of Iran and Karaj, one of Iran’s biggest cities. Water demand due to increasing agricultural and industrial activities caused many problems in the field of water resources management. In this study, the potential of groundwater resources was evaluated using remote sensing, geographic information system (GIS), and analytic hierarchy process (AHP) for the first time. Digital Elevation Model from Shuttle Radar Topography Mission was used to generate a slope map and drainage density map. Three Landsat-8 satellite images were utilized to provide lineament density and land cover/land use maps. Geological and soil type maps were provided from the Geological Survey and Mineral Explorations of Iran (GSI). Tropical Rainfall Measuring Mission data were used to prepare average annual precipitation map. Discharge values from 102 pumping wells in the time period of 2002–2014 were used to evaluate the results. Seven data layers were prepared, and the geodatabase was made in GIS. The layers and their classes were assigned weights using AHP method. Finally, the layers were overlaid based on their weights, and the potential map of groundwater resources was generated. The area was classified into five zones with very high, high, moderate, low, and very low potentials. The zones covered 5.95, 32.90, 22.70, 10.20, and 28.25% of the study area, respectively. The results showed good agreement with the field data obtained from discharge wells.     

The performance of SPI and PNPI in analyzing the spatial and temporal trend of dry and wet periods over Iran

Drought monitoring is carried out using various drought indices, including SPI, to generate time series of dry and wet periods. Furthermore, the dispersion of dry and wet periods was embossed with different intensities (high, medium, and low) over the data record years. Although these results were very necessary for planning and predicting future droughts, it appeared that the application of any trend over dry and wet periods could provide more accurate and unbiased or safer predictions in terms of analysis process. Generally, most of the researchers believed that the results of a drought trend analysis have been influenced by short-term persistence or significant autocorrelation with different lags on drought event time series and the mentioned impact should be preferably removed. Accordingly, drought monitoring was accomplished using SPI and PNPI drought indices to extract time series of dry and wet periods in terms of 50-year (1965–2014) annual rainfall data of 40 synoptic stations over Iran. Having used the basic and modified Mann–Kendall nonparametric tests, it was attempted to analyze the trend of dry and wet periods extracted from mentioned indices. The results represent the relative advantage of using the modified Mann–Kendall test in drought trend analysis. Furthermore, it was shown that the trend of dry and wet periods was negative in the majority of selected stations and that this trend was significant at 95% confidence level in northwest of Iran. Also, the results indicated the similar performance of SPI and PNPI indices in trend analysis of dry and wet periods.

     

Inverse modeling of variable-density groundwater flow in a semi-arid area in Iran using a genetic algorithm

Flow and mass transport parameter estimation was done by creating an inverse model of a seawater intrusion system using a genetic algorithm (GA) method as the optimization procedure. Firstly, the SEAWAT code was used for the forward solution part and then a program was written in MATLAB for coupling the forward and inverse processes. The auto-calibration objective function was defined with the root mean square errors (RMSE) between the observed and the simulated values. A simple GA was used to minimize the RMSE criterion. The methodology was applied to a coastal aquifer with heterogeneous formations in a semi-arid area near salty Tashk Lake (electrical conductivity 61,420 µS/cm), Fars province, Iran. In the last two decades, the overexploitation of groundwater has caused a major water level drawdown and, consequently, salt-water intrusion. Firstly, flow and transport parameters (hydraulic conductivity, porosity, specific storage coefficient and longitudinal dispersivity) were estimated simultaneously in steady-state and, secondly, in the developed code, these results were used as initial values of the parameters in transient-state. Results show a good match for observed and simulated data. It can be concluded that GA is a helpful tool for automatic calibration of variable density fluid systems such as seawater intrusion cases.     

Modeling Uranium Transport in Koongarra, Australia: The Effect of a Moving Weathering Zone

Natural analogues are an important source of long-term data and may be viewed as naturally occurring experiments that often include processes, phenomena, and scenarios that are important to nuclear waste disposal safety assessment studies. The Koongarra uranium deposit in the Alligator Rivers region of Australia is one of the best-studied natural analogue sites. The deposit has been subjected to chemical weathering over several million years, during which many climatological, hydrological, and geological changes have taken place, resulting in the mobilization and spreading of uranium. Secondary uranium mineralization and dispersed uranium are present from the surface down to the base of the weathering zone, some 25 m deep. In this work, a simple uranium transport model is presented and sensitivity analyses are conducted for key model parameters. Analyses of field and laboratory data show that three layers can be distinguished in the Koongarra area: (1) a top layer that is fully weathered, (2) an intermediate layer that is partially weathered (the weathering zone), and (3) a lower layer that is unweathered. The weathering zone has been moving downward as the weathering process proceeds. Groundwater velocities are found to be largest in the weathering zone. Transport of uranium is believed to take place primarily in this zone. It appears that changes in the direction of groundwater flow have not had a significant effect on the uranium dispersion pattern. The solid-phase uranium data show that the uranium concentration does not significantly change with depth within the fully weathered zone. This implies that uranium transport has stopped in these layers. A two-dimensional vertically integrated model for transport of uranium in the weathering zone has been developed. Simulations with a velocity field constant in time and space have been carried out, taking into account the downward movement of this zone and the dissolution of uranium in the orebody. The latter has been modelled by a nonequilibrium relationship. In these simulations, pseudo-steady state uranium distributions are computed. The main conclusion drawn from this study is that the movement of the weathering zone and the nonequilibrium dissolution of uranium in the orebody play an important role in the transport of uranium. Despite the fact that the model is a gross simplification of what has actually happened in the past two million years, a reasonable fit of calculated and observed uranium distributions was obtained with acceptable values for the model parameters.