Premium
Spatial Variability for One Type of Phosphate Mine Land in Central Florida
Author(s) -
Bos J.,
Collins M. E.,
Gensheimer G. J.,
Brown R. B.
Publication year - 1984
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1984.03615995004800050033x
Subject(s) - transect , variogram , sampling (signal processing) , soil ph , soil test , spatial variability , geostatistics , soil science , overburden , environmental science , soil water , chemistry , mathematics , geology , kriging , mining engineering , statistics , oceanography , filter (signal processing) , computer science , computer vision
Soil‐plant interaction problems are frequently encountered in the reclaimed phosphate minelands of central Florida. To solve some of these problems, a knowledge of the soil properties and their spatial variability must be acquired. An optimum sampling design for reclaimed mine lands can only be constructed after this information is known. Therefore, selected soil properties of one type of reclaimed phosphate mine land (overburden over sandtailings) in central Florida were investigated. A 6 × 21 sampling grid with 10‐m interval sampling points was constructed and soil samples (0–25 cm) were taken in four directions along selected transects. The results indicated that mean values were very high for total P (8.3 g/kg), extractable P (0.9 g/kg), and extractable Ca (2.3 g/kg) by the double acid method. The double acid method of extracting Ca, Mg, K, and Na yielded higher mean contents than the procedure using NH 4 OAc at pH 7. Correlation coefficients were highly significant between both extractable Ca and Mg and total extractable bases. Higher r values were obtained using KCl‐pH than H 2 O‐pH in studying the relationships of pH to the other soil properties except for total P. Semivariograms were calculated for the different soil properties studied in the four sampling directions. A composite semivariogram of all four directions was also constructed. Except for CEC and relative elevation, spatial correlation of all properties was encountered at sampling distances < 2.6 to 7.1 m. At a lag distance of six peak values occurred in several sampling directions for many of the soil properties studied. This indicates similar optimum sampling distances for the soil properties of interest.