SILICATE SOURCES OF THICK, RELICT SOILS ON THE EDWARDS PLATEAU, CENTRAL TEXAS

Thick soils on isolated uplands of the Edwards Plateau in central Texas are interpreted to be relicts of a formerly more extensive Pleistocene soil cover. The lower Cretaceous Glen Rose Limestone and the overlying Segovia and Fort Terrett members of the Edwards Limestone are the main units exposed on the plateau, but the relict, thick soils are most commonly found on the Edwards Limestone members. The occurrence of the thick soils on pure limestone indicates by mass balance that the thick soils did not form from the weathering of the underlying limestone bedrock. Instead, these thick soils may have formed from silicate material derived from eolian dust or from a stratigraphically higher, more clay-rich unit that has subsequently been eroded.

We studied the distribution of the thick, relict soils with elevation and lithology, as well as the grain size, clay-sized mineralogy, and the neodymium isotopic composition of a thick, upland soil in Kerr County. The lower horizons of a thick, relict soil profile in this area are more clay-rich than the surficial horizons and the modern thin soils. Therefore, we suggest that dust is not a significant source of silicate material to the thick soils. This interpretation is further supported by mineralogical differences between the surficial and lower horizons of the thick soil profile and between the modern, thin and relict, thick soils.

We propose that the silicate material in the relict, thick soils was derived from overlying, clay-rich strata such as the locally-eroded Del Rio Clay. Evidence for this interpretation includes: 1) the thick soil silicates have a Nd isotopic composition that is similar to the Del Rio Clay and different from the modern thin soil, 2) the Del Rio Clay and the lower horizons of the thick soil have a similar texture that is different from the textures of surficial horizons of the thick soil and the modern thin soil, and 3) independent of elevation, the thick soils occur dominantly over the Edwards Limestone and are relatively absent on the Glen Rose Limestone. This deduction is consistent with previous findings of Rabenhorst and Wilding (1986) that were based on the mineralogy, quartz grain morphology, and grain size distribution of the modern soils, insoluble limestone residues, and dusts and has implications for understanding soil-forming processes in karst terranes.