Robust geostatistical prediction of trace elements across France

Saby, Nicolas P.A.; Marchant, B. P.; Lark, R. M.; Jolivet, Claudy; Arrouays, Dominique

doi:10.1016/j.geoderma.2011.03.001

Cited by 59 publications

(43 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Metal(loid) contents in topsoil are influenced by the type, the intensity, and the history of anthropogenic activities and occupation (Matschullat et al, 2000). The contribution of atmospheric deposition in topsoil is traceable to very high values of the coefficient of variation (Hovmand et al, 2008;Thorpe and Harrison, 2008;Saby et al, 2011). The low values of the coefficient of variation of metal(loid) contents in soils of the Manaus region, 'Yellow' Haplic Ferralsols (FR-ha 'a'), 'Yellow' Acrisols Alumic (AC-au 'a'), and Albic Podzols (PZ-ab) ( Table 4), suggest that possible atmospheric deposition of pollutants was not detected.…”

Section: Discussionmentioning

confidence: 99%

Geochemical Signature of Amazon Tropical Rainforest Soils

Souza

Fontes

Gilkes

et al. 2018

Rev. Bras. Ciênc. Solo

View full text Add to dashboard Cite

Evaluating soil geochemical diversity in the Amazon Basin has been a challenge largely because most study sites have been at the edge of the basin and it is difficult to get samples in such a region. Here we show that even among the most weathered soils, physicochemical soil properties express lithology. Our results are based on topsoil samples collected from different locations in minimally disturbed areas in the state of Amazonas, Brazil. Soil properties were measured using methods which are suitable for highly developed soils. The Chemical Index Alteration and Weathering Index of Parker was calculated based on the content of metal(loid)s in soils determined by X-ray fluorescence. Descriptive statistics, Pearson correlation, and Principal Component Analysis (PCA) were performed on data. In general, Amazon rainforest soils are more deeply weathered than soils in other Brazilian biomes and tropical rainforests in Asia and Africa. The high coefficient of variation of metal(loid) contents express pedogenesis and parent material diversity. Correlation analysis indicated that the tri-pentavalent elements are strongly associated with Al and Fe contents in the topsoil. In contrast, mono-divalent elements are correlated with sand and silt fractions. According to PCA, five soil groups with defined geochemical compositions and degrees of weathering could be identified: i) acidic sandy podzolized soils; ii) acidic loamy ferralitic soils with the highest content of tri-pentavalent ions; iii) acidic clayey kaolinitic soils with low metal(loid) contents; iv) acidic loamy kaolinitic soils with low metal(loid) contents; and v) silty neutral 2:1 clay soils. This study is the first effort to analyze the geochemical diversity in Amazon rainforest soils. These data are extremely valuable in determining the geochemical background for these soil types and this region. Geochemical variability can be predicted to some extent by lithology and pedogenesis, which can be applied to define the sampling required in future studies.

show abstract

Section: Discussionmentioning

confidence: 99%

Geochemical Signature of Amazon Tropical Rainforest Soils

Souza

Fontes

Gilkes

et al. 2018

Rev. Bras. Ciênc. Solo

View full text Add to dashboard Cite

show abstract

“…Therefore when this assumption is implausible, Box-Cox and logarithmic transforms are applied to the observations. To deal with outliers, we applied robust geostatistics as described in Saby et al (2011) and developed in the following paragraphs.…”

Section: Robust Geostatistical Modellingmentioning

confidence: 99%

“…Such an understanding is required by policy makers. Maps of some trace elements in soils across France have been derived from punctual data using robust geostatistics (Marchant et al, 2010;Saby et al, 2011). Similar methods were used by Marchant et al (2011) to map soil metal contamination in an industrial region of Wales.…”

Section: Introductionmentioning

confidence: 99%

“…We suspected these factors would not encompass the entire variation of the properties as spatial correlations exist and as additional predictors, for example accounting for punctual anomalies, were not available. Therefore the adopted methodology coupled 1) regression trees to understand factors that affect the properties with 2) robust geostatistics presented by Saby et al (2011) to obtain maps by adding the fixed effects derived from regression trees to simple kriging predictions on winsorized data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Mapping soil Pb stocks and availability in mainland France combining regression trees with robust geostatistics

Lacarce¹,

Saby²,

Martín³

et al. 2012

Geoderma

View full text Add to dashboard Cite

“…Copper salts are used as fungicides in vineyards, and they contribute to the large concentrations of Cu in the soil (e.g. Pietrzak and McPhail 2004;Martín et al 2007;Saby et al 2011). More diffuse atmospheric pollution can be attributed to traffic, industrial emissions, street dust, and the burning of fossil fuels (de Miguel et al 1999).…”

Section: Introductionmentioning

confidence: 99%

Spatial variation of trace elements in the peri-urban soil of Madrid

Cueva¹,

Marchant

Quintana³

et al. 2013

J Soils Sediments

View full text Add to dashboard Cite

Purpose The peri-urban region to the south east of Madrid contains a mixture of housing, manufacturing industry and farming, some of which disperse metals, in particular cadmium, copper, lead, and zinc, into the soil. We have mapped the concentrations of these elements and identified the major influences on their distributions. Material and methods We sampled the topsoil at 125 sites across 1,050 km 2 of peri-urban land to the south east of the city on two grids, one nested inside the other. At each site, we measured the current contents of the four trace elements in the soil. We used robust geostatistical methods to model the complex spatial distributions of the data as mixtures of fixed and random effects. The empirical best linear unbiased predictor was used to map the elements. Site descriptors (lithology, land cover, cultivation, relief, erosion, and stoniness) were then included as covariates to identify significant effects on trace element concentrations. Results and discussion The complex spatial distributions of the elements seem to arise from several sources. The concentrations generally increase from southeast to northwest, i.e., with increasing proximity to Madrid itself, the main potential source of pollution. This pattern is clear for lead and similar for copper and zinc, though with "hot spots" at or near industrial sites. The spatial pattern of cadmium is more complex and depends on varied lithology, industry, and land use such as irrigation and cultivation. In general, the concentrations of the four elements appear to decrease with increases in stoniness and erosion, and to be largest on the valley floors. Conclusions Robust geostatistical methods enabled us to analyze and map the complex patterns of spatial variation of trace elements in a peri-urban region of Madrid. They show that distance to the city center, lithology, manufacturing industry, and cultivation all play their parts in loading the soil with lead, copper, zinc, and cadmium. In the event, none of the metals has yet exceeded the legislative thresholds, but some concentrations are already substantially greater than would arise from natural sources, especially closest to Madrid itself.

show abstract

Robust geostatistical prediction of trace elements across France

Cited by 59 publications

References 28 publications

Geochemical Signature of Amazon Tropical Rainforest Soils

Geochemical Signature of Amazon Tropical Rainforest Soils

Mapping soil Pb stocks and availability in mainland France combining regression trees with robust geostatistics

Spatial variation of trace elements in the peri-urban soil of Madrid

Contact Info

Product

Resources

About