Study on Spatial Variability of Soil Nutrients at Multi-scale in a Hilly Region
Liu Hong-bin
Abstract
Liu Hong-bin
Abstract
Spatial variability and accuracy of soil organic matter,available phosphorus and available potassium in the top soil at multi-scales were studied in a hilly region using geostatistics and spatial analysis modules in ArcGIS.The results showed that the spatial variability of soil organic matter was the weakest and that of available phosphorus was the strongest at the same scale.Typical semivarigram structures existed at different scales.At large and meso-scales,the nuggets of soil nutrients varied from 0.001 to 0.207,indicating that the analyzed soil nutrients had a strong spatial autocorrelation.At the small scale,the nugget value of available phosphorus was 0.394,showing that it had an obvious spatial autocorrelation.The results of cross-validation suggested that the prediction precision was different at different scales.At the same scale,the prediction of soil organic matter and available potassium presented higher accuracies.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Spatial variability and accuracy of soil organic matter,available phosphorus and available potassium in the top soil at multi-scales were studied in a hilly region using geostatistics and spatial analysis modules in ArcGIS.The results showed that the spatial variability of soil organic matter was the weakest and that of available phosphorus was the strongest at the same scale.Typical semivarigram structures existed at different scales.At large and meso-scales,the nuggets of soil nutrients varied from 0.001 to 0.207,indicating that the analyzed soil nutrients had a strong spatial autocorrelation.At the small scale,the nugget value of available phosphorus was 0.394,showing that it had an obvious spatial autocorrelation.The results of cross-validation suggested that the prediction precision was different at different scales.At the same scale,the prediction of soil organic matter and available potassium presented higher accuracies.
Key concepts: Geostatistics, Spatial variability, Environmental science, Soil science, Soil nutrients, Phosphorus, Nutrient, Organic matter