Spatial prediction of landslide susceptibility in western Serbia using hybrid support vector regression (SVR) with GWO,BAT and COA algorithms

In this study, we developed multiple hybrid machine-learning models to address parameter optimization limitations and enhance the spatial prediction of landslide susceptibility models. We created a geographic information system database, and our analysis results were used to prepare a landslide inventory map containing 359 landslide events identified from Google Earth, aerial photographs, and other validated sources. A support vector regression (SVR) machine-learning model was used to divide the landslide inventory into training (70%) and testing (30%) datasets. The landslide susceptibility map was produced using 14 causative factors. We applied the established gray wolf optimization (GWO) algorithm, bat algorithm (BA), and cuckoo optimization algorithm (COA) to fine-tune the parameters of the SVR model to improve its predictive accuracy. The resultant hybrid models, SVR-GWO, SVR-BA, and SVR-COA, were validated in terms of the area under curve (AUC) and root mean square error (RMSE). The AUC values for the SVR-GWO (0.733), SVR-BA (0.724), and SVR-COA (0.738) models indicate their good prediction rates for landslide susceptibility modeling. SVR-COA had the greatest accuracy, with an RMSE of 0.21687, and SVR-BA had the least accuracy, with an RMSE of 0.23046. The three optimized hybrid models outperformed the SVR model (AUC = 0.704, RMSE = 0.26689), confirming the ability of metaheuristic algorithms to improve model performance.     

Global calibration of a novel,branched GDGT-based soil pH proxy

Recently, 6-methyl branched glycerol dialkyl glycerol tetraethers (brGDGTs) were separated from 5-methyl brGDGTs, which are used in brGDGT-based proxies. Here we analyzed brGDGTs in 27 soil samples along the 400 mm isoline of mean annual precipitation in China by using tandem 2D liquid chromatography. The fractional abundance of 6-methyl brGDGTs showed a positive correlation with soil pH, while that of 5-methyl brGDGTs decreased with increasing soil pH. The abundance ratio of 6-/5-methyl brGDGTs, namely the isomerization of branched tetraethers (IBT), was calculated. The correlation of IBT with pH (pH = 6.33 − 1.28 × IBT; R² 0.89; root mean squared error, RMSE, 0.24) was much stronger than that of the traditionally used cyclization index of branched tetraethers (CBT) with pH (R² 0.52; RMSE 0.49) and comparable with that of CBT′ with pH (R² 0.88; RMSE 0.25). Compiling all available data from 319 soil samples resulted in a global calibration: pH = 6.53 − 1.55 × IBT (R² 0.72; RMSE 0.65), which has a better correlation than the CBT_5ME-pH proxy (R² 0.63; RMSE 0.78), but a weaker correlation than the CBT′-pH proxy (R² 0.85; RMSE 0.52). Our result suggests that the IBT is a promising indicator for soil pH, particularly in cases when some compounds in the CBT′ index cannot be determined.     

Optimization approach to retrieve soil surface parameters from single-acquisition single-configuration SAR data

This study suggests a novel approach to the retrieval of soil surface parameters using a single-acquisition single-configuration synthetic-aperture radar (SAR) system. Soil surface parameters such as soil moisture and surface roughness are key elements for many environmental studies, including Earth surface water cycles, energy exchange, agriculture, and geology. Remote sensing techniques, especially SAR data, are commonly used to retrieve such soil surface parameters over large areas. Several backscattering models have been proposed for soil surface parameters retrieval from SAR data. However, commonly, these backscattering models require multi configuration SAR data, including multi-polarization, multi-frequency, and multi-incidence angle. Here we propose a methodology that employs single-acquisition single-configuration SAR data for the retrieval of soil surface parameters. The originality is to use single-acquisition single-configuration SAR data to retrieve the soil surface parameters using an optimization approach by the genetic algorithm (GA); we have used the modified Dubois model (MDM) in HH polarization as the backscattering model. Three HH polarization and C band data sets from Quebec (Radarsat-1), Ontario (SIR-C), and Oklahoma (AIRSAR) were analyzed. The retrieved values of soil moisture and soil surface roughness were then compared to ground truth measurements with corresponding parameters. We employed diverse criteria, including the mean absolute error (MAE), the root mean square error (RMSE), the coefficient of performance (CP), and the correlation coefficient to investigate the performance of the proposed methodology. This analysis suggests the capability of the GA for the retrieval of soil surface parameters. Based on our findings, this method presents a viable alternative approach to the retrieval of soil surface parameters when only single-acquisition single-configuration SAR data is available.     

Integrated method of RS and GPR for monitoring the changes in the soil moisture and groundwater environment due to underground coal mining

Mining affects the environment in different ways depending on the physical context in which the mining occurs. In mining areas with an arid environment, mining affects plants’ growth by changing the amount of available water. This paper discusses the effects of mining on two important determinants of plant growth—soil moisture and groundwater table (GWT)—which were investigated using an integrated approach involving a field sampling investigation with remote sensing (RS) and ground-penetrating radar (GPR). To calculate and map the distribution of soil moisture for a target area, we initially analyzed four models for regression analysis between soil moisture and apparent thermal inertia and finally selected a linear model for modeling the soil moisture at a depth 10 cm; the relative error of the modeled soil moisture was about 6.3% and correlation coefficient 0.7794. A comparison of mined and unmined areas based on the results of limited field sampling tests or RS monitoring of Landsat 5-thermatic mapping (TM) data indicated that soil moisture did not undergo remarkable changes following mining. This result indicates that mining does not have an effect on soil moisture in the Shendong coal mining area. The coverage of vegetation in 2005 was compared with that in 1995 by means of the normalized difference vegetation index (NDVI) deduced from TM data, and the results showed that the coverage of vegetation in Shendong coal mining area has improved greatly since 1995 because of policy input RMB￥0.4 per ton coal production by Shendong Coal Mining Company. The factor most affected by coal mining was GWT, which dropped from a depth of 35.41 m before mining to a depth of 43.38 m after mining at the Bulianta Coal Mine based on water well measurements. Ground-penetrating radar at frequencies of 25 and 50 MHz revealed that the deepest GWT was at about 43.4 m. There was a weak water linkage between the unsaturated zone and groundwater, and the decline of water table primarily resulted from the well pumping for mining safety rather than the movement of cracking strata. This result is in agreement with the measurements of the water wells. The roots of nine typical plants in the study area were investigated. Populus was found to have the deepest root system with a depth of about 26 m. Based on an assessment of plant growth demands and the effect of mining on environmental factors, we concluded that mining will have less of an effect on plant growth at those sites where the primary GWT depth before mining was deep enough to be unavailable to plants. If the primary GWT was available for plant growth before mining, especially to those plants with deeper roots, mining will have a significant effect on the growth of plants and the mechanism of this effect will include the loss of water to roots and damage to the root system.     

波致海床瞬态液化对再悬浮贡献率的深度学习预测方法

波致瞬态液化渗流导致海床内细粒沉积物向海水中运移,这一过程对海底沉积物再悬浮的贡献率不容忽视,但是贡献率的准确估计和预测比较困难。本研究将黄河水下三角洲的观测数据(包括水深、有效波高、有效波周期、实验舱内悬沙浓度、实验舱外悬沙浓度)作为模型输入数据集,基于长短时记忆循环神经网络建立了瞬态液化对再悬浮贡献率的深度学习预测模型。为了客观评价模型的性能,以平均绝对百分比误差、均方根误差和平均平方误差-标准偏差为评判标准,将该深度学习模型与其他预测模型(支持向量回归模型、人工神经网络)的预测结果进行了比较。结果表明,基于长短时记忆循环神经网络的深度学习模型对3.5d以内的瞬态泵送再悬浮贡献率预测误差最小,其平均绝对百分比误差、均方根误差和平均平方误差-标准偏差分别为5.87%、1.6730、0.1574。因此,该模型可以有效地减少机器学习方法在连续预测中产生的误差叠加问题。     

Spatial variability of soil water erosion: Comparing empirical and intelligent techniques

Soil water erosion (SWE) is an important global hazard that affects food availability through soil degradation, a reduction in crop yield, and agricultural land abandonment. A map of soil erosion susceptibility is a first and vital step in land management and soil conservation. Several machine learning (ML) algorithms optimized using the Grey Wolf Optimizer (GWO) metaheuristic algorithm can be used to accurately map SWE susceptibility. These optimized algorithms include Convolutional Neural Networks (CNN and CNN-GWO), Support Vector Machine (SVM and SVM-GWO), and Group Method of Data Handling (GMDH and GMDH-GWO). Results obtained using these algorithms can be compared with the well-known Revised Universal Soil Loss Equation (RUSLE) empirical model and Extreme Gradient Boosting (XGBoost) ML tree-based models. We apply these methods together with the frequency ratio (FR) model and the Information Gain Ratio (IGR) to determine the relationship between historical SWE data and controlling geo-environmental factors at 116 sites in the Noor-Rood watershed in northern Iran. Fourteen SWE geo-environmental factors are classified in topographical, hydro-climatic, land cover, and geological groups. We next divided the SWE sites into two datasets, one for model training (70% of the samples = 81 locations) and the other for model validation (30% of the samples = 35 locations). Finally the model-generated maps were evaluated using the Area under the Receiver Operating Characteristic (AU-ROC) curve. Our results show that elevation and rainfall erosivity have the greatest influence on SWE, while soil texture and hydrology are less important. The CNN-GWO model (AU-ROC = 0.85) outperformed other models, specifically, and in order, SVR-GWO = GMDH-GWO (AUC = 0.82), CNN = GMDH (AUC = 0.81), SVR = XGBoost (AUC = 0.80), and RULSE. Based on the RUSLE model, soil loss in the Noor-Rood watershed ranges from 0 to 2644 t ha^–1yr^?1.     

基于LHS方法的地下水随机模拟及开采量风险性 评价

地下水流数值模拟过程中,水文地质参数的不确定性对模拟结果影响很大。以内蒙古鄂尔多斯市某水源地为例,利用拉丁超立方抽样(LHS)方法获得了含水层渗透参数的随机组合,进行地下水流随机模拟。通过对观测资料与计算水位的绝对值平均(MAE)误差和误差均方根(RMSE)的统计分析,获得了模型较为稳定的随机模拟次数是243次。利用该随机模型对水源地设计开采量进行水位预测,并给出允许降深的风险性分布图。结果表明,预测水位和标准差分布符合实际情况,水位降深大于35 m的风险性最大达到75%。     

Prediction of blasting-induced fragmentation in Meydook copper mine using empirical,statistical, and mutual information models

One of the most important aims of blasting in open pit mines is to reach desirable size of fragmentation. Prediction of fragmentation has great importance in an attempt to prevent economic drawbacks. In this study, blasting data from Meydook mine were used to study the effect of different parameters on fragmentation; 30 blast cycles performed in Meydook mine were selected to predict fragmentation where six more blast cycles are used to validate the results of developed models. In this research, mutual information (MI) method was employed to predict fragmentation. Ten parameters were considered as primary ones in the model. For the sake of comparison, Kuz-Ram empirical model and statistical modeling were also used. Coefficient of determination (R ²), root mean square error (RMSE), and mean absolute error (MAE) were then used to compare the models. Results show that MI model with values of R ², RMSE, and MAE equals 0.81, 10.71, and 9.02, respectively, is found to have more accuracy with better performance comparing to Kuz-Ram and statistical models.     

基于WorldView-02高分影像的BP和RBF神经网络遥感水深反演

遥感水深反演是水深测量的一种重要技术和手段.以美济礁水深反演为例,选择WorldView-02高分影像为数据源,在辐射定标和大气校正的基础上,构建BP（Back Propagation）和RBF（Radial Basis Function）人工神经网络水深反演模型,以遥感影像8个波段为输入层,通过tansig、logsig、高斯函数和purelin函数变换实现从输入层到隐含层、隐含层到输出层的转换,以便反演水深.最后对反演水深与实测水深采用回归分析,求解决定系数（coefficient of determination,R2）、平均决定误差（Mean Absolute Error,MAE）、均方根误差（Root Mean Square Error,RMSE）等进行比较,评价2种模型的精度.结果表明,RBF神经网络模型结构更简单,对样本要求更低,反演精度达到0.995,更适合遥感水深反演.      

Accuracy analysis of the 2014–2015 Global Shuttle Radar Topography Mission (SRTM) 1 arc-sec C-Band height model using International Global Navigation Satellite System Service (IGS) Network

Global Shuttle Radar Topography Mission (SRTM) data products have been widely used in Earth Sciences without an estimation of their accuracy and reliability even though large outliers exist in them. The global 1 arc-sec, 30 m resolution, SRTM C-Band (C-30) data collected in February 2000 has been recently released (2014–2015) outside North America. We present the first global assessment of the vertical accuracy of C-30 data using Ground Control Points (GCPs) from the International GNSS Service (IGS) Network of high-precision static fiducial stations that define the International Terrestrial Reference Frame (ITRF). Large outliers (height error ranging from –1285 to 2306 m) were present in the C-30 dataset and 14% of the data were removed to reduce the root mean square error (RMSE) of the dataset from ～187 to 10.3 m which is close to the SRTM goal of an absolute vertical accuracy of RMSE ～10 m. Globally, for outlier-filtered data from 287 GCPs, the error or difference between IGS and SRTM heights exhibited a non-normal distribution with a mean and standard error of 6.5 ± 0.5 m. Continent-wise, only Australia, North and South America complied with the SRTM goal. At stations where all the X- and C-Band SRTM data were present, the RMSE of the outlier-filtered C-30 data was 11.7 m. However, the RMSE of outlier-included dataset where C- and X-Band data were present was ～233 m. The results suggest that the SRTM data must only be used after regional accuracy analysis and removal of outliers. If used raw, they may produce results that are statistically insignificant with RMSE in 100s of meters.     

海河流域不同下垫面土壤水分动态模拟研究

针对海河流域不同的下垫面类型,选取密云(果园林地)、大兴(城郊农田)、馆陶(平原农田)3个观测站,建立垂直方向上以含水率θ为因变量、含根系吸水项的非饱和土壤水分运动数值计算模型。该模型以一维Richards方程为基础(以下简称RE模型),采用实测的降水和蒸散数据作为模型的上边界条件,运用全隐式有限差分法,分别对不同生长期内的土壤水分进行数值模拟,得到时间序列的土壤水分廓线,并分别采用成熟软件HYDRUS-1D的模拟结果和各观测站实测土壤水分对RE模型进行交叉验证和直接验证。结果表明RE模型能够很好地模拟海河流域不同下垫面土壤水分动态变化过程,3个站模拟结果与实测土壤水分数据的均方根误差(RMSE)分别为0.03127,0.0359和0.0409 cm3/cm3。与HYDRUS-1D软件模拟结果(其与观测值的RMSE分别为0.03759,0.0647和0.0467 cm3/cm3)相比,RE模型模拟的土壤水分具有更高的精度,也显示出RE模型的可靠性。探讨3个站土壤水分的时空变异规律及其影响因子并以大兴站为例,通过优化RE模型参数,探讨犁底层对土壤水分模拟结果的影响,进一步改善RE模型的模拟精度。     

Error in digital network and basin area delineation using d8 method: A case study in a sub-basin of the Ganga

Digital elevation model (DEM) is one of the input data derived from different satellite sensors for hydrologic and hydraulic modelings. Two prime questions could be answered before using these DEMs. First, the acceptability of datasets for our use and second appropriate resolution of the dataset. Three widely used DEMs SRTM 30m, ASTER 30m and SRTM 90m are analyzed to evaluate their suitability to delineate river network and basin boundary area. The hydrology tool of spatial analyst extension inbuilt in ArcGIS 10.2 (which uses the D8 method for calculation of flow direction) has been used for the delineation of both river networks and basin boundary. The assessment of river network alignment and boundary delineation is carried out in the seven sub-catchments of Gandak river basin having different morphological characteristics. The automatically delineated boundary area for all the three DEMs reflects a significant difference when compared with the digitized basin area from the Ganga flood control commission (GFCC) map. The maximum boundary area delineation error is 39137.20 km₂ forASTER 30m, and minimum delineation error of 13239.28 km₂ for SRTM 90m. In the stream network, delineation accuracy is good for SRTM 90m while, except Gandak trunk, ASTER 30m DEM shows better delineation accuracy indicated by mean absolute error (MAE) and standard deviation (SD).     

Mapping of aboveground biomass in Gabon

The aim of this paper is to map the aboveground biomass (AGB) in Gabon. First, a random forest (RF) model that relates reference AGB values to remote sensing (RS)-derived variables (mainly radar and optical images) was built, and the significant predictive variables were determined. Second, the built RF model was applied to the significant RS-derived variables to predict AGB across Gabon. The results showed that the overall RMSE (Root Mean Square Error) on the RS-derived AGB map with a spatial resolution of 50 m was 63.3 t/ha (R² = 0.53).To improve the accuracy of the RS-derived AGB map, the integration of LiDAR data provided by the Geoscience Laser Altimeter System (GLAS) onboard the Ice Cloud and Land Elevation Satellite (ICESat) was investigated. First, an RF model that relates reference AGB values to GLAS-derived metrics and a DEM (Digital Elevation Model) was built. Second, the calibrated RF model was applied to obtain a spatially distributed estimation of AGB (GLAS footprints geolocation) covering forested areas in Gabon, with a density of 0.13 footprints/km². Third, the semivariogram of residuals (RS-derived AGB map – GLAS-derived AGB “surrogate AGB”) was computed. Later, a regression kriging interpolation was performed by taking into account the spatial structure of residuals to provide a continuous residual map. Finally, the RS-derived AGB map and the residual map were summed, and a final AGB map was obtained. The results showed that the integration of GLAS surrogate AGB data slightly improves the accuracy of the RS-derived AGB map only for AGB values lower than 100 t/ha (bias and RMSE reduced by 13.9 and 10 t/ha, respectively).     

Soil salinity and sodicity in a shrimp farming coastal area of the Mekong Delta,Vietnam

Soil salinity and sodicity are environmental problems in the shrimp farming areas of the Cai Nuoc district, Ca Mau province, Vietnam. In 2000, farmers in the district switched en masse from rice cropping to shrimp culture. Due to recent failure in shrimp farming, many farmers wish to revert to a rotational system with rice in the wet season and shrimps in the dry season. So far, all their attempts to grow rice have failed. To assess soil salinity and sodicity, 25 boreholes in shrimp ponds were analysed in four consecutive seasons from 2002 to 2004. The results showed that soil salinity was quite serious (mean EC_e 29.25 dS m⁻¹), particularly in the dry season (mean EC_e 33.44 dS m⁻¹). In the wet season, significant amounts of salts still remained in the soil (mean EC_e 24.65 dS m⁻¹) and the highest soil salinity levels were found near the sea. Soil sodicity is also a problem in the district (exchangeable sodium percentage range 9.63–72.07%). Sodicity is mainly a phenomenon of topsoils and of soils near the sea. Both soil salinity and sodicity are regulated by seasonal rainfall patterns. They could together result in disastrous soil degradation in the Cai Nuoc district.     

Large-scale spatial interpolation of soil pH across the Loess Plateau, China

Soil pH plays an important role in biogeochemical processes in soils. The spatial distribution of soil pH provides basic and useful information relevant to soil management and agricultural production. To obtain an accurate distribution map of soil pH on the Loess Plateau of China, 382 sampling sites were investigated throughout the region and four interpolation methods, i.e., inverse distance weighting (IDW), splines, ordinary kriging, and cokriging, were applied to produce a continuous soil pH surface. In the study region, soil pH values ranged from 6.06 to 10.76, with a mean of 8.49 and a median of 8.48. Land use type had a significant effect (p < 0.01) on soil pH; grassland soils had higher pHs than cropland and forestland soils. From a regional perspective, soil pH showed weak variation and strong spatial dependence, indicated by the low values of the coefficient of variation (0.05) and the nugget-to-sill ratios (<0.25). Indices of cross-validation, i.e., average error, mean absolute error, root mean square error, and model efficiency coefficient were used to compare the performance of the four different interpolation methods. Kriging methods interpolated more accurately than IDW and splines. Cokriging performed better than ordinary kriging and the accuracy was improved using soil organic carbon as an auxiliary variable. Regional distribution maps of soil pH were produced. The southeastern part of the region had relatively low soil pH values, probably due to higher precipitation, leaching, and higher soil organic matter contents. Areas of high soil pH were located in the north of the central part of the region, possibly associated with the salinization of sandy soils under inappropriate irrigation practices in an arid climate. Map accuracy could be further improved using new methods and incorporating other auxiliary variables, such as precipitation, elevation, terrain attributes, and vegetation types.     

Compilation simulation of surface water and groundwater resources using the SWAT-MODFLOW model for a karstic basin in Iran

Modeling of karstic basins can provide a better understanding of the interactions between surface water and groundwater, a more accurate estimation of infiltrated water amount, and a more reliable water balance calculation. In this study, the hydrological simulation of a karstic basin in a semiarid region in Iran was performed in three different stages. In the first stage, the original SWAT model was used to simulate surface-water flow. Then, the SWAT-MODFLOW conjunctive model was implemented according to the groundwater characteristics of the study area. Finally, due to the karstic characteristics of the region and using the CrackFlow (CF) package, the SWAT-MODFLOW-CF conjunctive model was developed to improve the simulation results. The coefficient of determination (R²) and the Nash-Sutcliffe efficiency coefficient (NSE) as error evaluation criteria were calculated for the models, and their average values were 0.63 and 0.57 for SWAT, 0.68 and 0.61 for SWAT-MODFLOW, 0.73 and 0.7 for SWAT-MODFLOW-CF, respectively. Moreover, the mean absolute error (MAE) and root mean squared error (RMSE) of the calibration for groundwater simulation using the SWAT-MODFLOW model were 1.23 and 1.77 m, respectively. These values were 1.01 and 1.33 m after the calibration of the SWAT-MODFLOW-CF model. After modifying the CF code and keeping the seams and cracks open in both dry and wet conditions, the amount of infiltrated water increased and the aquifer water level rose. Therefore, the SWAT-MODFLOW-CF conjunctive model can be proposed for use in karstic areas containing a considerable amount of both surface water and groundwater resources.

     

Reverse fault slip through soft rock and sand strata by centrifuge modeling tests

The devastating damage after the 1999 Chi-Chi and 1999 Izmit earthquakes has greatly motivated soil–reverse fault interaction studies. However, most centrifuge modeling studies have employed a single homogeneous soil layer during testing, which does not represent in situ conditions. Indeed, while geological conditions vary spatially, engineering soils are often underlain by soft rocks. Therefore, four centrifuge models were developed to evaluate the effect of soft rock layers on the ground surface and subsurface deformation. Sand–cement mixtures of varying thicknesses with a uniaxial compressive strength of 0.975 MPa, simulating extremely soft rock, were overlain by pluviated sandy soil. The model thickness was 100 mm, corresponding to 8 m in the prototype scale when spun at 80 g. Every model was subjected to a vertical offset of 50 mm/4 m (0.5 H; H: total sedimentary deposit thickness) along a reverse fault with a 60° dip. The results indicate that the presence of a soft rock stratum results in the creation of a horst profile at the ground surface. Additionally, the thinner the soil layer on top of the soft rock stratum is, the longer and higher the horst created at the ground surface. Consequently, the fault deformation zone lengthens proportionally with the increasing thickness ratio of the soft rock. Furthermore, the presence of soft rock as an intermediary stratum between bedrock and soil causes the deformation zone boundary on the hanging wall side to move in the direction of fault movement.

     

Impact of mining activities on soils and sediments at the historical mining area in Podljubelj,NW Slovenia

The abandoned Hg mine in Podljubelj was in operation with interceptions from 1557 to 1902. The entire operating period yielded about 110 000 tons of ore, from which 360 tons of Hg was produced. The objective of the research project was to establish the contents and spatial distribution of Hg in soils and stream sediments in the vicinity of the mine. On an area of 88 ha the soil was sampled in a 100 × 100 m grid. Two soil horizons (0–5 cm and 20–30 cm) were sampled in order to distinguish between geogenic and anthropogenic Hg sources. It was established that on an area of about 9 ha Hg content in soil exceeds The New Dutchlist action value for Hg (10 mg/kg). Total Hg concentrations in soil samples vary between 0.17 and 719 mg/kg, with a mean of 3.0 mg/kg. Mercury contents in stream sediments range from 0.065 to 1.4 mg/kg, with a mean of 0.64 mg/kg. The highest determined value in soils was found in the area around the former roasting furnace, where the ore was processed. Increased Hg concentrations were also found on the mine waste dump (108 mg/kg). Mercury contents in soils generally decrease with soil profile depth and with the distance from the mine and from the roasting furnace location. Mercury also appears in higher concentrations along the road that runs through the valley, which results from the use of Hg-bearing ore residues in road construction. The average enrichment factor (EF) of Hg in topsoil with respect to subsoil is 3.3. Calculated enrichment factors show higher values also for Cd (3.2), Pb (2.7), Ca (2.4) and P (1.9). The average enrichment factor of Hg in topsoil with regard to the established Slovenian soil averages (EF_slo) is 19. EF_slo of other determined chemical elements do not exceed 3.0.     

念青唐古拉山扎当冰川冰储量估算及冰下地形特征分析

冰川体积估算对水资源以及冰川变化研究具有重要的意义. 但是实测的冰川厚度数据十分稀少,限制了冰川体积的估算. 2011年5月对念青唐古拉山北坡扎当冰川进行了雷达测厚工作,获取了该冰川的厚度分布状况. 基于该冰川的厚度数据,测量点的GPS数据,1970年的地形图和2010年Landsat TM影像,在ArcGIS技术的支持下,采用简单Kriging插值方法对冰川非测厚区域的厚度进行了插值计算,绘制出了冰川厚度等值线图并估算了冰川的冰储量. 结果表明：冰川最大厚度出现于海拔约5 748 m靠近主流线的位置,最大冰厚度为108 m,冰川平均厚度为38.1 m,2010年冰川面积为1.73 km²,扎当冰川的冰储量为0.066 km³. 将扎当冰川表面DEM与冰川厚度分布图相结合,绘制出了该冰川的冰床地形图. 结果显示,在冰川厚度大的区域,冰床地形呈现近V字形分布,这与其相对平缓的冰面地形形成明显对比;同时,在冰表地形较陡区域,冰川厚度不大,冰床地形呈现U形分布.     

Comparison of FFNN and ANFIS models for estimating groundwater level

Prediction of water level is an important task for groundwater planning and management when the water balance consistently tends toward negative values. In Maheshwaram watershed situated in the Ranga Reddy District of Andhra Pradesh, groundwater is overexploited, and groundwater resources management requires complete understanding of the dynamic nature of groundwater flow. Yet, the dynamic nature of groundwater flow is continually changing in response to human and climatic stresses, and the groundwater system is too intricate, involving many nonlinear and uncertain factors. Artificial neural network (ANN) models are introduced into groundwater science as a powerful, flexible, statistical modeling technique to address complex pattern recognition problems. This study presents the comparison of two methods, i.e., feed-forward neural network (FFNN) trained with Levenberg–Marquardt (LM) algorithm compared with a fuzzy logic adaptive network-based fuzzy inference system (ANFIS) model for better accuracy of the estimation of the groundwater levels of the Maheshwaram watershed. The statistical indices used in the analysis were the root mean square error (RMSE), regression coefficient (R ²) and error variation (EV).The results show that FFNN-LM and ANFIS models provide better accuracy (RMSE = 4.45 and 4.94, respectively, R ² is 93% for both models) for estimating groundwater levels well in advance for the above location.