Spatial methods to analyze the relationship between Spanish soil properties and cadmium content

In this study, concentrations of cadmium using 3778 samples encompassing the total size Spain (about 505 km2) were investgated. Two novel spatial methods namely Moran eigenvector spatially varying coefficient (MESVC) and filtered unconditional quantile regression (SF-UQR) employed with aim avoiding problem local collinearity which is prevalent in models. Additionally, coefficients applied to assess influence soil properties together texture on variations cadmium. It was indicated that overall level low compared found around world. particular, values Cd varied between 0.01 2.00 mgkg?1, median 0.23 mgkg?1. The residual standard error adjusted R2 produced by MESVC 0.16 0.69, respectively are better than 0.21 0.39 yielded SF-UQR model. Both these models outperformed geographically weighted (GWR) performance also traditional method kriging. For instance, terms willmott index (d) root mean squared relative (RMSRE), had superior equal 0.612 0.275 0.399 0.379 obtained for ordinary GWR demonstrated CaCO3, sand, silt clay a negligible whereas, EC largest contribution followed SOM pH.