BME prediction of continuous geographical properties using auxiliary variables

Using auxiliary information to improve the prediction accuracy of soil properties in a physically meaningful and technically efficient manner has been widely recognized in pedometrics. In this paper, we explored a novel technique to effectively integrate sampling data and auxiliary environmental information, including continuous and categorical variables, within the framework of the Bayesian maximum entropy (BME) theory. Soil samples and observed auxiliary variables were combined to generate probability distributions of the predicted soil variable at unsampled points. These probability distributions served as soft data of the BME theory at the unsampled locations, and, together with the hard data (sample points) were used in spatial BME prediction. To gain practical insight, the proposed approach was implemented in a real-world case study involving a dataset of soil total nitrogen (TN) contents in the Shayang County of the Hubei Province (China). Five terrain indices, soil types, and soil texture were used as auxiliary variables to generate soft data. Spatial distribution of soil total nitrogen was predicted by BME, regression kriging (RK) with auxiliary variables, and ordinary kriging (OK). The results of the prediction techniques were compared in terms of the Pearson correlation coefficient (r), mean error (ME), and root mean squared error (RMSE). These results showed that the BME predictions were less biased and more accurate than those of the kriging techniques. In sum, the present work extended the BME approach to implement certain kinds of auxiliary information in a rigorous and efficient manner. Our findings showed that the BME prediction technique involving the transformation of variables into soft data can improve prediction accuracy considerably, compared to other techniques currently in use, like RK and OK.     

Conditional bias-penalized kriging (CBPK)

Simple and ordinary kriging, or SK and OK, respectively, represent the best linear unbiased estimator in the unconditional sense in that they minimize the unconditional (on the unknown truth) error variance and are unbiased in the unconditional mean. However, because the above properties hold only in the unconditional sense, kriging estimates are generally subject to conditional biases that, depending on the application, may be unacceptably large. For example, when used for precipitation estimation using rain gauge data, kriging tends to significantly underestimate large precipitation and, albeit less consequentially, overestimate small precipitation. In this work, we describe an extremely simple extension to SK or OK, referred to herein as conditional bias-penalized kriging (CBPK), which minimizes conditional bias in addition to unconditional error variance. For comparative evaluation of CBPK, we carried out numerical experiments in which normal and lognormal random fields of varying spatial correlation scale and rain gauge network density are synthetically generated, and the kriging estimates are cross-validated. For generalization and potential application in other optimal estimation techniques, we also derive CBPK in the framework of classical optimal linear estimation theory.     

UXO target area identification with hidden Markov models

Site characterization activities at potential unexploded ordnance (UXO) sites rely on sparse sampling collected as geophysical surveys along strip transects. From these samples, the locations of target areas, those regions on the site where the geophysical anomaly density is significantly above the background density, must be identified. A target area detection approach using a hidden Markov model (HMM) is developed here. HMM’s use stationary transition probabilities from one state to another for steps between adjacent locations as well as the probability of any particular observation occurring given each possible underlying state. The approach developed here identifies the transition probabilities directly from the conceptual site model (CSM) created as part of the UXO site characterization process. A series of simulations examine the ability of the HMM approach to simultaneously determine the target area locations within each transect and to estimate the unknown anomaly intensity within the identified target area. The HMM results are compared to those obtained using a simpler target detection approach that considers the background anomaly density to be defined by a Poisson distribution and each location to be independent of any adjacent location. Results show that the HMM approach is capable of accurately identifying the target locations with limited false positive identifications when both the background and target are intensities are known. The HMM approach is relatively robust to changes in the initial estimate of the target anomaly intensity and is capable of identifying target locations and the corresponding target anomaly intensity when this intensity is approximately 60% higher than the background intensity at intensities that are representative of actual field sites. Application to data collected from a wide area assessment field site show that the HMM approach identifies the area of the site with elevated anomaly intensity with few false positives. This field site application also shows that the HMM results are relatively robust to changes in the transect width.     

Improving kriging of groundwater level data using nonlinear normalizing transformations—a field application

Abstract

The present research study investigates the application of nonlinear normalizing data transformations in conjunction with ordinary kriging (OK) for the accurate prediction of groundwater level spatial variability in a sparsely-gauged basin. We investigate three established normalizing methods, Gaussian anamorphosis, trans-Gaussian kriging and the Box-Cox method to improve the estimation accuracy. The first two are applied for the first time to groundwater level data. All three methods improve the mean absolute prediction error compared to the application of OK to the non-transformed data. In addition, a modified Box-Cox transformation is proposed and applied to normalize the hydraulic heads. The modified Box-Cox transformation in conjunction with OK is found to be the optimal spatial model based on leave-one-out cross-validation. The recently established Spartan semivariogram family provides the optimal model fit to the transformed data. Finally, we present maps of the groundwater level and the kriging variance based on the optimal spatial model.

Editor D. Koutsoyiannis; Associate editor A. Montanari

Citation Varouchakis, E.A., Hristopoulos, D.T., and Karatzas, G.P., 2012. Improving kriging of groundwater level data using nonlinear normalizing transformations—a field application. Hydrological Sciences Journal, 57 (7), 1404–1419.     

The use of secondary information in geostatistical target area identification

The identification and characterization of target areas at former bombing ranges is the first step in investigating these sites for residual unexploded ordnance. Traditionally, magnetometer surveys along transects are used in identifying areas with high densities of magnetic anomalies, which are likely former target areas. Combining magnetometer survey data with other data sources may reduce the level of survey data required for site characterization, increasing characterization efficiency. Here, several techniques for incorporating secondary information into kriging estimates of magnetic anomaly density are investigated for a former bombing range located near Pueblo, Colorado. In particular, kriging with external drift, collocated ordinary cokriging, and simple kriging with local means (SKLM) are used to incorporate information from a secondary variable. The secondary variable consists of a grid of crater density values derived from a topographic light detection and ranging (LIDAR) analysis. The craters, which are clearly identifiable in the LIDAR data, were generated through munitions use at the site and are therefore related to the target locations. The results from this study indicate that the inclusion of the secondary information in the kriging estimates does benefit target area characterization and provides a means of elucidating target area details from only limited magnetometer transect data. For the Pueblo site, the use of SKLM with the crater density as a secondary variable and only limited magnetometer transect data, provided results comparable to those obtained from using much larger magnetometer transect data sets.     

A kriging approach based on Aitchison geometry for the characterization of particle-size curves in heterogeneous aquifers

We consider the problem of predicting the spatial field of particle-size curves (PSCs) from a sample observed at a finite set of locations within an alluvial aquifer near the city of Tübingen, Germany. We interpret PSCs as cumulative distribution functions and their derivatives as probability density functions. We thus (a) embed the available data into an infinite-dimensional Hilbert Space of compositional functions endowed with the Aitchison geometry and (b) develop new geostatistical methods for the analysis of spatially dependent functional compositional data. This approach enables one to provide predictions at unsampled locations for these types of data, which are commonly available in hydrogeological applications, together with a quantification of the associated uncertainty. The proposed functional compositional kriging (FCK) predictor is tested on a one-dimensional application relying on a set of 60 PSCs collected along a 5-m deep borehole at the test site. The quality of FCK predictions of PSCs is evaluated through leave-one-out cross-validation on the available data, smoothed by means of Bernstein Polynomials. A comparison of estimates of hydraulic conductivity obtained via our FCK approach against those rendered by classical kriging of effective particle diameters (i.e., quantiles of the PSCs) is provided. Unlike traditional approaches, our method fully exploits the functional form of PSCs and enables one to project the complete information content embedded in the PSC to unsampled locations in the system.     

Statistical algorithms accounting for background density in the detection of UXO target areas at DoD munitions sites

Statistically defensible methods are presented for developing geophysical detector sampling plans and analyzing data for munitions response sites where unexploded ordnance (UXO) may exist. Detection methods for identifying areas of elevated anomaly density from background density are shown. Additionally, methods are described which aid in the choice of transect pattern and spacing to assure with degree of confidence that a target area (TA) of specific size, shape, and anomaly density will be identified using the detection methods. Methods for evaluating the sensitivity of designs to variation in certain parameters are also discussed. Methods presented have been incorporated into the Visual Sample Plan (VSP) software (free at ) and demonstrated at multiple sites in the United States. Application examples from actual transect designs and surveys from the previous two years are demonstrated.     

Spatial prediction of river channel topography by kriging

Topographic information is fundamental to geomorphic inquiry, and spatial prediction of bed elevation from irregular survey data is an important component of many reach‐scale studies. Kriging is a geostatistical technique for obtaining these predictions along with measures of their reliability, and this paper outlines a specialized framework intended for application to river channels. Our modular approach includes an algorithm for transforming the coordinates of data and prediction locations to a channel‐centered coordinate system, several different methods of representing the trend component of topographic variation and search strategies that incorporate geomorphic information to determine which survey data are used to make a prediction at a specific location. For example, a relationship between curvature and the lateral position of maximum depth can be used to include cross‐sectional asymmetry in a two‐dimensional trend surface model, and topographic breaklines can be used to restrict which data are retained in a local neighborhood around each prediction location. Using survey data from a restored gravel‐bed river, we demonstrate how transformation to the channel‐centered coordinate system facilitates interpretation of the variogram, a statistical model of reach‐scale spatial structure used in kriging, and how the choice of a trend model affects the variogram of the residuals from that trend. Similarly, we show how decomposing kriging predictions into their trend and residual components can yield useful information on channel morphology. Cross‐validation analyses involving different data configurations and kriging variants indicate that kriging is quite robust and that survey density is the primary control on the accuracy of bed elevation predictions. The root mean‐square error of these predictions is directly proportional to the spacing between surveyed cross‐sections, even in a reconfigured channel with a relatively simple morphology; sophisticated methods of spatial prediction are no substitute for field data. Copyright © 2007 John Wiley & Sons, Ltd.     

Predicting ordinary kriging errors caused by surface roughness and dissectivity

The magnitude of kriging errors varies in accordance with the surface properties. The purpose of this paper is to determine the association of ordinary kriging (OK) estimated errors with the local variability of surface roughness, and to analyse the suitability of probabilistic models for predicting the magnitude of OK errors from surface parameters. This task includes determining the terrain parameters in order to explain the variation in the magnitude of OK errors. The results of this research indicate that the higher order regression models, with complex interaction terms, were able to explain 95 per cent of the variation in the OK error magnitude using the least number of predictors. In addition, the results underscore the importance of the role of the local diversity of relief properties in increasing or decreasing the magnitude of interpolation errors. The newly developed dissectivity parameters provide useful information for terrain analysis. Our study also provides constructive guides to understanding the local variation of interpolation errors and their dependence on surface dissectivity. Copyright © 2005 John Wiley & Sons, Ltd.     

Multigaussian kriging for point-support estimation: incorporating constraints on the sum of the kriging weights

In the geostatistical analysis of regionalized data, the practitioner may not be interested in mapping the unsampled values of the variable that has been monitored, but in assessing the risk that these values exceed or fall short of a regulatory threshold. This kind of concern is part of the more general problem of estimating a transfer function of the variable under study. In this paper, we focus on the multigaussian model, for which the regionalized variable can be represented (up to a nonlinear transformation) by a Gaussian random field. Two cases are analyzed, depending on whether the mean of this Gaussian field is considered known or not, which lead to the simple and ordinary multigaussian kriging estimators respectively. Although both of these estimators are theoretically unbiased, the latter may be preferred to the former for practical applications since it is robust to a misspecification of the mean value over the domain of interest and also to local fluctuations around this mean value. An advantage of multigaussian kriging over other nonlinear geostatistical methods such as indicator and disjunctive kriging is that it makes use of the multivariate distribution of the available data and does not produce order relation violations. The use of expansions into Hermite polynomials provides three additional results: first, an expression of the multigaussian kriging estimators in terms of series that can be calculated without numerical integration; second, an expression of the associated estimation variances; third, the derivation of a disjunctive-type estimator that minimizes the variance of the error when the mean is unknown.     

Copula-based geostatistical modeling of continuous and discrete data including covariates

It is common in geostatistics to use the variogram to describe the spatial dependence structure and to use kriging as the spatial prediction methodology. Both methods are sensitive to outlying observations and are strongly influenced by the marginal distribution of the underlying random field. Hence, they lead to unreliable results when applied to extreme value or multimodal data. As an alternative to traditional spatial modeling and interpolation we consider the use of copula functions. This paper extends existing copula-based geostatistical models. We show how location dependent covariates e.g. a spatial trend can be accounted for in spatial copula models. Furthermore, we introduce geostatistical copula-based models that are able to deal with random fields having discrete marginal distributions. We propose three different copula-based spatial interpolation methods. By exploiting the relationship between bivariate copulas and indicator covariances, we present indicator kriging and disjunctive kriging. As a second method we present simple kriging of the rank-transformed data. The third method is a plug-in prediction and generalizes the frequently applied trans-Gaussian kriging. Finally, we report on the results obtained for the so-called Helicopter data set which contains extreme radioactivity measurements.     

Comparison of ordinary kriging and artificial neural network for spatial mapping of arsenic contamination of groundwater

In this technical note, we investigate the hypothesis that ‘non-linearity matters in the spatial mapping of complex patterns of groundwater arsenic contamination’. The spatial mapping pertained to data-driven techniques of spatial interpolation based on sampling data at finite locations. Using the well known example of extensive groundwater contamination by arsenic in Bangladesh, we find that the use of a highly non-linear pattern learning technique in the form of an artificial neural network (ANN) can yield more accurate results under the same set of constraints when compared to the ordinary kriging method. One ANN and a variogram model were used to represent the spatial structure of arsenic contamination for the whole country. The probability for successful detection of a well as safe or unsafe was found to be atleast 15% larger than that by kriging under the country-wide scenario. The probability of false hopes, which is a serious issue in public health monitoring was found to be significantly lower (by more than 10%) than that by kriging.     

Wavelet analysis residual kriging vs. neural network residual kriging

 This paper deals with the problem of spatial data mapping. A new method based on wavelet interpolation and geostatistical prediction (kriging) is proposed. The method – wavelet analysis residual kriging (WARK) – is developed in order to assess the problems rising for highly variable data in presence of spatial trends. In these cases stationary prediction models have very limited application. Wavelet analysis is used to model large-scale structures and kriging of the remaining residuals focuses on small-scale peculiarities. WARK is able to model spatial pattern which features multiscale structure. In the present work WARK is applied to the rainfall data and the results of validation are compared with the ones obtained from neural network residual kriging (NNRK). NNRK is also a residual-based method, which uses artificial neural network to model large-scale non-linear trends. The comparison of the results demonstrates the high quality performance of WARK in predicting hot spots, reproducing global statistical characteristics of the distribution and spatial correlation structure.     

Interpolations of groundwater table elevation in dissected uplands

The variable elevation of the groundwater table in the St. Louis area was estimated using multiple linear regression (MLR), ordinary kriging, and cokriging as part of a regional program seeking to assess liquefaction potential. Surface water features were used to determine the minimum water table for MLR and supplement the principal variables for ordinary kriging and cokriging. By evaluating the known depth to the water and the minimum water table elevation, the MLR analysis approximates the groundwater elevation for a contiguous hydrologic system. Ordinary kriging and cokriging estimate values in unsampled areas by calculating the spatial relationships between the unsampled and sampled locations. In this study, ordinary kriging did not incorporate topographic variations as an independent variable, while cokriging included topography as a supporting covariable. Cross validation suggests that cokriging provides a more reliable estimate at known data points with less uncertainty than the other methods. Profiles extending through the dissected uplands terrain suggest that: (1) the groundwater table generated by MLR mimics the ground surface and elicits a exaggerated interpolation of groundwater elevation; (2) the groundwater table estimated by ordinary kriging tends to ignore local topography and exhibits oversmoothing of the actual undulations in the water table; and (3) cokriging appears to give the realistic water surface, which rises and falls in proportion to the overlying topography. The authors concluded that cokriging provided the most realistic estimate of the groundwater surface, which is the key variable in assessing soil liquefaction potential in unconsolidated sediments.     

Spatial interpolation methods for nonstationary plume data

Plume interpolation consists of estimating contaminant concentrations at unsampled locations using the available contaminant data surrounding those locations. The goal of ground water plume interpolation is to maximize the accuracy in estimating the spatial distribution of the contaminant plume given the data limitations associated with sparse monitoring networks with irregular geometries. Beyond data limitations, contaminant plume interpolation is a difficult task because contaminant concentration fields are highly heterogeneous, anisotropic, and nonstationary phenomena. This study provides a comprehensive performance analysis of six interpolation methods for scatter-point concentration data, ranging in complexity from intrinsic kriging based on intrinsic random function theory to a traditional implementation of inverse-distance weighting. High resolution simulation data of perchloroethylene (PCE) contamination in a highly heterogeneous alluvial aquifer were used to generate three test cases, which vary in the size and complexity of their contaminant plumes as well as the number of data available to support interpolation. Overall, the variability of PCE samples and preferential sampling controlled how well each of the interpolation schemes performed. Quantile kriging was the most robust of the interpolation methods, showing the least bias from both of these factors. This study provides guidance to practitioners balancing opposing theoretical perspectives, ease-of-implementation, and effectiveness when choosing a plume interpolation method.     

Regional flood frequency analysis using residual kriging in physiographical space

In this article, an approach using residual kriging (RK) in physiographical space is proposed for regional flood frequency analysis. The physiographical space is constructed using physiographical/climatic characteristics of gauging basins by means of canonical correlation analysis (CCA). This approach is a modified version of the original method, based on ordinary kriging (OK). It is intended to handle effectively any possible spatial trends within the hydrological variables over the physiographical space. In this approach, the trend is first quantified and removed from the hydrological variable by a quadratic spatial regression. OK is therefore applied to the regression residual values. The final estimated value of a specific quantile at an ungauged station is the sum of the spatial regression estimate and the kriged residual. To evaluate the performance of the proposed method, a cross‐validation procedure is applied. Results of the proposed method indicate that RK in CCA physiographical space leads to more efficient estimates of regional flood quantiles when compared to the original approach and to a straightforward regression‐based estimator. Copyright © 2010 John Wiley & Sons, Ltd.     

Mass Discharge in a Tracer Plume: Evaluation of the Theissen Polygon Method

A tracer plume was created within a thin aquifer by injection for 299 d of two adjacent “sub‐plumes” to represent one type of plume heterogeneity encountered in practice. The plume was monitored by snapshot sampling of transects of fully screened wells. The mass injection rate and total mass injected were known. Using all wells in each transect (0.77 m well spacing, 1.4 points/m² sampling density), the Theissen Polygon Method (TPM) yielded apparently accurate mass discharge (M_d) estimates at three transects for 12 snapshots. When applied to hypothetical sparser transects using subsets of the wells with average spacing and sampling density from 1.55 to 5.39 m and 0.70 to 0.20 points/m², respectively, the TPM accuracy depended on well spacing and location of the wells in the hypothesized transect with respect to the sub‐plumes. Potential error was relatively low when the well spacing was less than the widths of the sub‐plumes (>0.35 points/m²). Potential error increased for well spacing similar to or greater than the sub‐plume widths, or when less than 1% of the plume area was sampled. For low density sampling of laterally heterogeneous plumes, small changes in groundwater flow direction can lead to wide fluctuations in M_d estimates by the TPM. However, sampling conducted when flow is known or likely to be in a preferred direction can potentially allow more useful comparisons of M_d over multiyear time frames, such as required for performance evaluation of natural attenuation or engineered remediation systems.     

Moving window kriging with geographically weighted variograms

This study adds to our ability to predict the unknown by empirically assessing the performance of a novel geostatistical-nonparametric hybrid technique to provide accurate predictions of the value of an attribute together with locally-relevant measures of prediction confidence, at point locations for a single realisation spatial process. The nonstationary variogram technique employed generalises a moving window kriging (MWK) model where classic variogram (CV) estimators are replaced with information-rich, geographically weighted variogram (GWV) estimators. The GWVs are constructed using kernel smoothing. The resultant and novel MWK–GWV model is compared with a standard MWK model (MWK–CV), a standard nonlinear model (Box–Cox kriging, BCK) and a standard linear model (simple kriging, SK), using four example datasets. Exploratory local analyses suggest that each dataset may benefit from a MWK application. This expectation was broadly confirmed once the models were applied. Model performance results indicate much promise in the MWK–GWV model. Situations where a MWK model is preferred to a BCK model and where a MWK–GWV model is preferred to a MWK–CV model are discussed with respect to model performance, parameterisation and complexity; and with respect to sample scale, information and heterogeneity.     

Geostatistical prediction of a local geometric geoid - kriging and cokriging with the use of EGM2008 geopotential model

We investigate prediction abilities of different variants of kriging and different combinations of data in a local geometric (GNSS/leveling based) geoid modeling. In order to generate local geoid models, we have used GNSS/leveling data and EGM2008 geopotential model. EGM2008 has been used twofold. Firstly, it was used as a basic long wave-length trend to be removed from geoid undulation data to generate a residual field of geoid heights modeled later by kriging (remove-restore technique). Secondly, EGM2008-based undulations were used as a secondary variable in a cokriging prediction procedure (as pseudo-observations). Besides the use of EGM2008, the kriging-based local geometric geoid models were generated only on the basis of raw undulations data. Kriging itself was used in two variants, i.e. ordinary kriging and universal kriging for univariate and bivariate cases (cokriging). The quality of kriging-based prediction for all its variants and all data combinations have been investigated on one fixed validation dataset consisting of 86 points and three training data sets characterized by a different density of sampling. Results of this study indicate that incorporation of EGM08 as a long wave-length trend in kriging prediction procedure outperforms cokriging strategy based on incorporation of EGM08 as a secondary spatially correlated variable.     

Regularized kriging as a generalization of simple, universal, and bayesian kriging

In this article the properties of regularized kriging (RK) are studied. RK is obtained as a result of relaxing the universal kriging (UK) non-bias condition by using the support vectors methodology. More specifically, we demonstrate how RK is a continuum of solutions in function of the regularizing parameter, which includes as particular and extreme cases, simple kriging (SK) and UK, and as an intermediate case, Bayesian kriging (BK). Likewise, expressions are obtained for the mean, variance and mean squared error (MSE), as also the expression for the corresponding estimator of the coefficients of the mean. Finally, we investigate the relationship between RK and the support vector machines. By means of simulations we compare the MSE for RK with those for BK and UK, for different association models, for different levels of noise, and for differently sized mean coefficients. The RK results prove to be an improvement on the UK and BK results, and, moreover, these improvements are proportionally greater for greater levels of noise.