Summary A reconnaissance survey was undertaken on soil near mine tailings to investigate variation in the content of copper, chromium and uranium. A nested sampling design was used. The data showed significant relations between the content of copper and uranium in the soil and its organic matter content, and a significant spatial trend in uranium content with distance from the tailings. Soil pH was not significantly related to any of the metals. The variance components associated with different scales of the sample design had large confidence intervals, but it was possible to show that the random variation was spatially dependent for all spatial models, whether for variation around a constant mean, or with a mean given by a linear effect of organic matter or distance to the tailings. For copper, we showed that a fractal or multifractal random model, with equal variance components for scales in a logarithmic progression, could be rejected for the model of variation around the fixed mean. The inclusion of organic matter as an explanatory factor meant that the fractal model could no longer be rejected, suggesting that the effect of organic matter results in spatial variation that is not scale invariant. It was shown, taking uranium as a case study, that further spatially nested sampling to estimate scale‐dependent variance components, or to test a non‐fractal model with adequate power, would require in the order of 200–250 samples in total. Highlights Sampling was undertaken to investigate spatial variation of metal content in soil near mine tailings. Chromium and uranium were related to soil organic matter content; uranium showed a spatial trend. Spatial variation was scale dependent, variation of copper was not scale‐invariant. Characterizing random spatial variation requires substantial sample effort.
Abstract. An estimated variogram of a soil property can be used to support a rational choice of sampling intensity for geostatistical mapping. However, it is known that estimated variograms are subject to uncertainty. In this paper we address two practical questions. First, how can we make a robust decision on sampling intensity, given the uncertainty in the variogram? Second, what are the costs incurred in terms of oversampling because of uncertainty in the variogram model used to plan sampling? To achieve this we show how samples of the posterior distribution of variogram parameters, from a computational Bayesian analysis, can be used to characterize the effects of variogram parameter uncertainty on sampling decisions. We show how one can select a sample intensity so that a target value of the kriging variance is not exceeded with some specified probability. This will lead to oversampling, relative to the sampling intensity that would be specified if there were no uncertainty in the variogram parameters. One can estimate the magnitude of this oversampling by treating the tolerable grid spacing for the final sample as a random variable, given the target kriging variance and the posterior sample values. We illustrate these concepts with some data on total uranium content in a relatively sparse sample of soil from agricultural land near mine tailings in the Copperbelt Province of Zambia.
Abstract. An estimated variogram of a soil property can be used to support a rational choice of sampling intensity for geostatistical mapping. However, it is known that estimated variograms are subject to uncertainty. In this paper we address two practical questions. First, how can we make a decision on sampling intensity which is robust, given the uncertainty in the variogram? Second, what are the costs incurred in terms of oversampling because of uncertainty in the variogram model used to plan sampling? To achieve this we show how samples of the posterior distribution of variogram parameters, from a computational Bayesian analysis, can be used to characterize the effects of variogram parameter uncertainty on sampling decisions. We show how one can select a sample intensity so that a target value of the kriging variance is not exceeded with some specified probability. This will lead to oversampling, relative to the sampling intensity that would be specified if there were no uncertainty in the variogram parameters. One can estimate the magnitude of this oversampling by treating the tolerable grid spacing for the final sample as a random variable, given the target kriging variance and the posterior sample values. We illustrate these concepts with some data on total uranium content in a relatively sparse sample of soil from agricultural land near mine tailings in the Copperbelt Province of Zambia.
We report results from an evaluation of the levels of heavy metals, i.e., copper (Cu), cadmium (Cd), lead (Pb), nickel (Ni), manganese (Mn), chromium (Cr), and iron (Fe) in sediment and tilapia fish samples from a wide stretch of the Kafue river of Zambia. In sediment samples, the highest Pb and Fe concentrations were recorded at Hippo Dam, i.e., 36.2 ± 0.1 mg/kg dw and 733 ± 37 mg/kg dw at Kafue Town, respectively. Other notably high metal concentrations in sediment were Cr at Kafue Bridge (42.5 ± 0.1 mg/kg dw [dw]), Cu at Mpongwe (233 ± 5 mg/kg dw), and Mn at Kafue Town (133 ± 1 mg/kg dw); it was highest at Ithezi Tezhi Dam at 166 ± 1 mg/kg d. Three fish species, i.e., three-spot bream Tilapia andersonii, red-breasted bream T. rendalli, and nile tilapia Oreochromis niloticus were evaluated for levels of the seven metals. The concentrations of the metals in these fish species afforded estimation of the biota sediment-accumulation factor, which is the ratio of the concentration of the metal in liver to that in the sediment. The coefficients of condition (K) values, which give an indication of the health of the fish, were also estimated. The K values ranged from 2.5 ± 0.5 to 5.1 ± 0.6 in all of the three fish species. Partial least squares analysis showed that heavy metals are generally sequestered evenly in all of the parts of all of the three fish species except for elevated levels of Mn, Cd, and Pb in the liver samples.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.