Mineral Potential Mapping Using Bayesian Learning for Multilayer Perceptrons

Multilayer perceptrons (MLPs) can be used to discover a function which can be used to map from a set of input variables onto a value representing the conditional probability of mineralization. The standard approach to training MLPs is gradient descent, in which the error between the network output and the target output is reduced in each iteration of the training algorithm. In order to prevent overfitting, a split-sample validation procedure is used, in which the data is partitioned into two sets: a training set, which is used for weight optimization, and a validation set, which is used to optimize various parameters that can be used to prevent overfitting. One of the problems with this approach is that the resulting maps can display significant variability which stems from (i) the (randomly initialized) starting weights and (ii) the particular training/validation set partition (also determined randomly). This problem is especially pertinent on mineral potential mapping tasks, in which the number of deposit cells is a very small proportion of the total number of cells in the study area. In contrast to gradient descent methods, Bayesian learning techniques do not find a single weight vector; rather, they infer the posterior distribution of the weights given the data. Predictions are then made by integrating over this distribution. An important advantage of the Bayesian approach is that the optimization of parameters such as the weight decay regularization coefficient can be performed using training data alone, thus avoiding the noise introduced through split-sample validation. This paper reports results of applying Bayesian learning techniques to the production of maps representing gold mineralization potential over the Castlemaine region of Victoria, Australia. Maps produced using the Bayesian approach display significantly less variability than those produced using gradient descent training. They are also more reliable at predicting the presence of unknown deposits.     

Bayesian Modelling of Spatial Data Using Markov Random Fields, With Application to Elemental Composition of Forest Soil

Spatial datasets are common in the environmental sciences. In this study we suggest a hierarchical model for a spatial stochastic field. The main focus of this article is to approximate a stochastic field with a Gaussian Markov Random Field (GMRF) to exploit computational advantages of the Markov field, concerning predictions, etc. The variation of the stochastic field is modelled as a linear trend plus microvariation in the form of a GMRF defined on a lattice. To estimate model parameters we adopt a Bayesian perspective, and use Monte Carlo integration with samples from Markov Chain simulations. Our methods does not demand lattice, or near-lattice data, but are developed for a general spatial data-set, leaving the lattice to be specified by the modeller. The model selection problem that comes with the artificial grid is in this article addressed with cross-validation, but we also suggest other alternatives. From the application of the methods to a data set of elemental composition of forest soil, we obtained predictive distributions at arbitrary locations as well as estimates of model parameters.     

Exposure and bioavailability of arsenic in contaminated soils from the La Parrilla mine, Spain

Arsenic derived from mining activity may contaminate water, soil and plant ecosystems resulting in human health and ecotoxicological risks. In this study, exposure assessment of arsenic (As) in soil, spoil, pondwater and plants collected from the areas contaminated by mine tailings and spoils in and around the La Parrilla mine, Caceres province, Spain, was carried out using AAS method. Water solubility, bioavailability and soil–plant transfer coefficients of As and phytoremediation potential of plants were determined. Arsenic concentrations varied from 148 to 2,540 mg/kg in soils of site 1 and from 610 to 1,285 mg/kg in site 2 exceeding the guideline limit for agricultural soil (50 mg/kg). Arsenic concentrations in pond waters varied from 8.8 to 101.4 μg/l. High concentrations of water-soluble As in the soils that ranged from 0.10 to 4.71 mg/kg in site 1 and from 0.46 to 4.75 mg/kg in site 2 exceeded the maximum permitted level of water-soluble As (0.04 mg/kg) in agricultural soils. Arsenic concentrations varied from 0.8 to 149.5 mg/kg dry wt in the plants of site 1 and from 2.0 to 10.0 mg/kg in the plants of site 2. Arsenic concentrations in plants increased in the approximate order: Retama sphaerocarpa < Pteridium aquilinum < Erica australis < Juncus effusus < Phalaris caerulescens < Spergula arvensis in site 1. The soil–plant transfer coefficients for As ranged from 0.001 to 0.21 in site 1 and from 0.004 to 0.016 in site 2. The bioconcentration factor based on water-soluble As of soil varied from 3.2 to 593.9 in the plants of site 1 whereas it varied from 2.1 to 20.7 in the plants of site 2. To our knowledge, this is the first study in Europe to report that the fern species P. aquilinum accumulates extremely low contents of As in its fronds despite high As levels in the soils. Therefore, the S. arvensis, P. caerulescens and J. effusus plant species grown in this area might be used to partly remove the bioavailable toxic As for the purpose of minimization of mining impacts until hypothetical hyperaccumulating and/or transgenic plants could be transplanted for the phytoremediation of As contaminated soils.     

Bayesian analysis of marked stress release models for time-dependent hazard assessment in the western Gulf of Corinth

We consider point processes defined on the space–time domain which model physical processes characterized qualitatively by the gradual increase over time in some energy until a threshold is reached, after which, an event causing the loss of energy occurs. The risk function will, therefore, increase piecewise with sudden drops in correspondence to each event. This kind of behaviour is described by Reid's theory of elastic rebound in the earthquake generating process where the quantity that is accumulated is the strain energy or stress due to the relative movement of tectonic plates. The complexity and the intrinsic randomness of the phenomenon call for probabilistic models; in particular the stochastic translation of Reid's theory is given by stress release models. In this article we use such models to assess the time-dependent seismic hazard of the seismogenic zone of the Corinthos Gulf. For each event we consider the occurrence time and the magnitude, which is modelled by a probability distribution depending on the stress level present in the region at any instant. Hence we are dealing here with a marked point process. We perform the Bayesian analysis of this model by applying the stochastic simulation methods based on the generation of Markov chains, the so called Markov chain Monte Carlo (MCMC) methods, which allow one to reconcile the model's complexity with the computational burden of the inferential procedure. Stress release and Poisson models are compared on the basis of the Bayes factor.     

U–U–Th disequilibria and timescale of sedimentary transfers in rivers: Clues from the Gangetic plain rivers

This study uses ²³⁸U–²³⁴U–²³⁰Th disequilibria in river sediments in order to constrain the transfer times of sediments in alluvial plains of rivers from Himalaya and the Gangetic plain. From the observed distributions we infer sediment transfer times of about 100 ka in the Gangetic plain for rivers taking their source in the Himalayan chain, and longer transfer of about 160–250 ka for foothill-fed rivers. This difference is probably related to the difference in the sediment transport dynamics of these two types of rivers.     

Unsupervised Change Detection in Multitemporal SAR Images Using MRF Models

An unsupervised change-detection method that considers the spatial contextual information in a log-ratio difference image generated from multitemporal SAR images is proposed. A Markov random filed (MRF) model is particularly employed to exploit statistical spatial correlation of intensity levels among neighboring pixels. Under the assumption of the independency of pixels and mixed Gaussian distribution in the log-ratio difference image, a stochastic and iterative EM-MPM change-detection algorithm based on an MRF model is developed. The EM-MPM algorithm is based on a maximiser of posterior marginals (MPM) algorithm for image segmentation and an expectation-maximum (EM) algorithm for parameter estimation in a completely automatic way. The experiment results obtained on multitemporal ERS-2 SAR images show the effectiveness of the proposed method.     

Test application of Bayesian Programming: Adaptive water quality management under uncertainty

A new method, Bayesian Programming (BP), developed by Harrison [Harrison KW. Multi-stage decision-making under uncertainty and stochasticity: Bayesian Programming. Adv Water Resour, submitted for publication] is tested on a case study involving optimal adaptive management of a river basin. The case study considers anew the process of permitting pulp mills on the Athabasca River in Alberta, Canada. The problem has characteristics common to many environmental management problems. There is uncertainty in the water quality response to pollutant loadings that will not be completely resolved with monitoring and the resolution of this uncertainty is impeded by the stochastic behavior of the water quality system. A two-stage adaptive management process is optimized with BP. Based on monitoring data collected after implementation of the first-stage decision, the uncertainties are updated prior to the second decision stage using Bayesian analysis. The worth of this two-stage adaptive management approach to this problem and the worth of monitoring are evaluated. Conclusions are drawn on the general practicality of BP for adaptive management. Potential strategies are outlined for extending the BP approach to secure further benefits of adaptive management.     

黄河流域县域碳排放的时空格局演变及空间效应机制

利用空间面板模型、空间自相关分析和以区域背景与最近邻状况为空间滞后的空间马尔科夫链对2000—2017年黄河流域县域碳排放时空格局与空间效应进行分析,结果表明：① 2000年以来黄河流域碳排放量激增,由山东全域和陕甘宁蒙交界的高值区向外圈层与轴向扩张,形成东高西低碳排放格局;② 存在“俱乐部趋同”现象,高碳排放县集聚于山东全域和陕甘宁蒙交界,低碳排放县集聚于西南部;2000年与2017年对比发现县域碳排放类型稳定性强,较高碳排放变为较低碳排放的县集中在东南部区域,而相反方向转变的县集中在内蒙古;③ 高碳溢出效应与低碳锁定效应是塑造时空格局的重要作用力,前者作用力更强;区域背景增强了“俱乐部趋同”与被包围异常值趋同,作用力强于最近邻状况,不显著区域内碳排放类型转变概率提高。④ 空间面板模型结果显示年轻人口结构、大经济规模、二产为主产业结构、高生活水平和高公共支出促进了碳排放量增加与空间效应作用,其中经济规模与产业结构是重要驱动因素。     

基于投入产出表的中国水资源消耗结构路径分析

为了解各部门在生产链中水资源的具体消耗路径和在消耗路径中所发挥的作用以及维持生产链中水资源消耗的关键部门,论文基于2017年投入产出表和相关气象数据,在农业用水中加入绿水资源消耗量,利用投入产出模型和结构路径分析(structural path analysis, SPA)法分析中国42个部门水资源消耗的总量和结构,以及水资源在生产链中消耗的具体路径。结果表明：① 城镇居民消费和进口的水足迹最高,其次为农村居民消费,大多数水足迹较高的部门,都是以居民消费为主,进口水资源弥补了国内部分生产用水的消耗,但水资源短缺的形势仍然严峻;② 生产链前5个层级的水足迹消耗占比为95%左右,其中部门内部消耗占比最大,水足迹随层级数的增加而下降,水资源反馈循环程度较弱;③ 从部门生产链看,部门内部的水足迹占比最大,部门流入和流出水足迹差值较大,部门之间普遍存在单向依赖特征,关联网络不够稳定;④ 从最终消费角度看,农业部门内部水足迹占比最大,是生产链中重要的衔接结点,多样化的消费需求增加上游部门的水资源压力。研究为调整节水政策和提升水资源配置与管理能力提供了一定的理论基础。