Soil‐landscape Relationships and Soil Variability in North Central Florida

Abstract

AbstractMorphological and selected chemical and physical properties of 12 pedons on three landscape positions (summit, shoulder, and backslope) were determined to evaluate how soil properties and landscape positions are related in north central Florida. A multistage nested design was employed to compare the variance of selected soil properties within and between landscape positions. Soil properties in the study area were related to landscape position. Soil‐landscape relationships were enhanced by the nonuniformity of the parent material. The parent material of the soils studied on the backslope was the clayey‐phosphatic Hawthorn Formation. This geologic formation also influenced the genesis of the soils on the shoulder but to a lesser extent. The soils on the summit were developed from sandy marine sediments. Classification problems were encountered for some of the soils on the shoulder and all of the soils studied on the backslope. The principal differences and trends across the landscape were decreases in A horizon thickness, matrix colors with chroma >2, sand percentage, pH, and organic C content below the A horizon downslope. Mottles and matrix colors with chroma ≤2, silt and clay percentages, and total P contents increased downslope. Clay and total P depth distributions were genetically and statistically related and highly significant (0.01 level) when studied for the whole landscape (rs = 0.77) and for the soils (rs = 0.81) on the backslope. Substantial increases in total P and clay contents were associated with the Hawthorn Formation. Several statistical analyses were performed. Kolmogorov D test showed nonnormal distribution of the data. The x2 test indicated a highly significant (0.01 level) dependence between soil properties and landscape position. Several relationships were masked when the Spearman correlation (rs) analysis was performed using data for the whole landscape. When rs was low (<0.50) for the whole landscape, rs values computed according to landscape position increased. The results of the Kruskal‐Wallis test indicated significant differences (0.01 level) for each soil property, except organic C, in at least two of the three landscapes. In a nested design >50% of the variability for pH in KCl, total P, sand, silt, and clay was due to differences between landscapes. For organic C content, >70% of its variability was due to differences between horizons.