Abstract Upland residual soils formed from mica gneiss and schist in the Piedmont and Blue Ridge provinces of North Carolina represent developmental sequences where particle‐size class, profile development, and mineralogy are related to landscape position and slope. In this study the distribution of total and free iron in 10 Hapludult and Dystrochrept pedons is examined and interpreted with respect to degree of profile development. Primary mineral alteration and B horizon free iron/total iron ratios vary nearly systematically with particle‐size class and solum thickness. The latter two parameters were found to be good indicators of degree of profile development and hence of relative soil age. The free iron/total iron ratio was found to be a particularly sensitive indicator of relative soil development. The ratio reflects the degree of primary mineral alteration, the amount of clay formed, as well as the pedogenic process of illuviation. The B horizon ratios are much better indicators of relative soil development than those from either C or A horizons. Ratios of free iron/total iron in A horizons are influenced by downslope movement of surficial material, the low extractability of iron oxides of large particle size, and clay content differences. Free iron/total iron ratios of C horizons are dependent on sampling depth, the amount of clay formed, and the degree of illuviation that has occurred below the solum. All soils had oxidic ratios (%Fe 2 O 3 + %gibbsite/%clay) > 0.2; however, all but one was excluded from the oxidic mineralogy class due to having >40% mica. A simple free iron/total iron ratio in B horizons, which is systematically related to the degree of weathering of primary minerals and the degree of morphological soil development, better serves as soil family criteria than the present % Fe 2 O 3 + % gibbsite/% clay.
Abstract A detailed study of the relationships between landform, surficial geology, and variability of loess thickness on uplands was conducted at two 2.32‐ha study sites in northern Delaware. The purpose of the study was to determine what factors are responsible for the short‐distance variability in loess thickness observed in this region. Thickness of loess and paleotopography of the subjacent Pensauken (Columbia) Formation were measured using ground‐penetrating radar at 1326 grid points within each plot. Present topography was measured using a transit at 676 points in one plot and 1326 points in the second plot. Surface grid diagrams and isopleth maps show a lack of relationship between loess thickness and landform and relief except along ephemeral drainages. Over short distances, the thickness and even the occurrence of loess is dependent on the topography of the paleosurface. The loess was deposited upon an uneven erosional surface and conforms on a gross scale to that paleosurface but has filled the concavities and low points, thus producing a more regular present topography. Modern denudation of the present land surface occasionally exposes the coarser‐textured Pensauken sediments at the surface over high points in the paleolandscape. The irregular erosional paleosurface has resulted in a high degree of short‐distance variability in loess thickness that is difficult to predict except where headslope dissection has been severe enough to produce gullying and the formation of ephemeral drainages. These drainages closely conform to the occurrence of paleodrainages. The high degree of variability in loess thickness and occurrence presents mapping difficulties that need to be addressed in design, naming, and description of soil mapping units.
Abstract Map units named for the Matapeake series (fine‐silty, mixed, mesic Typic Hapludults) comprise one‐third of the Coastal Plain in the soil survey of northern Delaware. The concept was one of well‐drained soils formed in thin loess over coarse‐textured fluvial sediments. This study uses characterization data from pedons within the loess‐blanketed terrain of northern Delaware to assess the taxonomy and composition of Matapeake map units. All pedons are Alfisols rather than Ultisols, and this status has not been influenced by liming. Fine‐loamy rather than fine‐silty soils typically occur where loess is <80 cm thick. Much soil variability exists on interfluves where thin loess discontinuously blankets the irregular surface of the sandy Pensauken Formation. This results in an inability to map specific consociations, even for large‐scale, detailed surveys. The existing soil survey used specific map units — consociations named for phases of soil series — that do not adequately depict the diversity on mappable landscapes. The Matapeake and associated series used in this area are too narrowly defined in ranges of characteristics, are too general in geographic setting, have loess thickness limits that are not concordant with particle‐size classes, and leave gaps in the lithologic‐hydrologic continuum. They therefore provide little utility for organizing knowledge, understanding relationships, or as objects for research. Soil survey in this area requires more generalized map units, greater emphasis on stratigraphy and geomorphology and their interaction to influence hydrology and lithology, and less emphasis on artificial conceptual groupings at the lowest taxonomic category, the series.
Abstract In Soil Taxonomy , mineralogy classes of pedons that have a fragmental, sandy, sandy‐skeletal, loamy, or loamy‐skeletal particle‐size class are defined by weight percent minerals in the coarse silt and larger fractions of the control section. However, a footnote to the key to mineralogy classes states that percentages by weight are estimated from grain counts of one or two of the dominant separates from a conventional mechanical analysis. The relationship of grain count number to volumetric percentage depends upon grain shape, size and cleavage, and the counting method used, making weight estimations, or even volumetric estimations impracticable. Revisions in the key to mineralogy classes are suggested because presently the key requires determination of grain size and shape relationships as well as the identification of minerals and their densities where often there is a gradation of mineral compositions and properties. According to the estimates used in this study, a line count of 80 to 85% micaceous minerals is needed to classify a pedon as micaceous; however, field classification of micaceous occurred at around 70% line count micaceous grains. This paper discusses limitations of grain counts for quantitative mineralogical analysis of the coarser fine‐earth fractions, presents a case for the revision of the key to mineralogy classes through use of actual mineralogic data, and recommends adoption of the line count method of determining composition for the purpose of mineralogy class placement.
Abstract Mobility of nutrients and heavy metals from mineralization of sewage sludge in acid soils is an environmental concern. Chemical properties of a forested acid sandy soil on flat terrain were examined at 6 and 30 months after ceasing multiple spray applications of sewage effluent. Total dry sludge applied was 5 and 26 Mg ha −1 on the treated 3‐ by 15‐m portion of 9‐ by 15 m plots. Less than 1.1 mg L −1 of either NO ‐ 3 ‐N or PO 3‐ 4 ‐P was found at the 24‐ to 90‐cm depth in soil solution extracted from the untreated bed adjacent to the planting area during 144 days. Treated areas were sharply defined by the growth of indigenous dog fennel ( Eupatorium sp.) which was the dominant species. Mean soil pH was below 4.7 at 6 months and 5.1 at 30 months. Treated interbed Cd, Cu, and NH + 4 ‐N were higher with depth at 30 months than 6 months while the reverse held for PO 3‐ 4 ‐P. Untreated bed or interbed soils had lower soil test values than corresponding treated soils. Much higher metal values were found at the 7‐cm depth than deeper. Differences in metal extracted from treated and untreated soils were readily distinguished by DA (Mehlich I), DTPA‐TEA, and EDTA‐AA soil test methods. All three methods extracted from 25 to 75% of the metals applied in the sludge. Lack of more downward movement of N, P, and metals was attributed to slow mineralization of the sewage sludge.
Abstract Biotite transformations in 10 pedons representing a developmental sequence of soils formed from coarse‐grained regional metamorphic rocks in the Piedmont and lower elevations of the Blue Ridge provinces of North Carolina are reported. Kaolinite is common to abundant in sand fractions of six of the pedons and is found to some extent in nine of the 10 pedons. Most of the kaolinite is derived from the pseudomorphic (isomorphic and largely isovolumetric) alteration of biotite. Biotite first alters to interstratified biotite‐vermiculite in which the vermiculite is hydroxy‐Al interlayered. The kaolinization of biotite extends throughout the grain, the pseudomorphs retaining the morphology of the biotite precursor but having the optical and structural properties of kaolinite. The koalinite pseudomorphs can be considered as single crystals because unit cells within the individual grains are all aligned crystallographically. A transformation model is proposed that accounts for most of the properties observed in weathering biotite. The degree of kaolinization of biotite increases with nearness to the soil surface and with increased degree of soil development. In the saprolite of the shallowest Dystrochrept biotite is relatively unweathered. In the sola of deeper Dystrochrepts and fine‐loamy Hapludults most of the biotite is kaolinized. Sand‐sized kaolinite pseudomorphs are less elastic and tenacious than fresh biotite and begin to physically disintegrate in the upper sola of fine‐loamy soils under the influence of processes of pedoturbation. In deeply developed clayey soils, kaolinite pseudomorphs may be absent from the sola. The increased clay content of clayey soils over that of fine‐loamy soils is largely a result of claysized kaolinite added from the comminution of sand‐sized kaolinite pseudomorphs of biotite.