The Holocene Phytolith Record from Morgan Playa in the Rolling Plains of Texas
Fredlund, Glen G., Bousman, C. Britt, Boyd, Douglas K., Plains Anthropologist
Phytolith analyses of playa-fill sediments from Morgan Playa yield strong evidence for Holocene vegetation and climate change. Morgan Playa is located in the Rolling Plains of Garza County, north-central Texas. Backhoe excavation of the small (about 3.2 hectares) basin exposed 250 cm of Holocene age sediments. A strongly developed Early Holocene paleosol occurs at the base (205-250 cm) of the profile. Dark gray-brown, clayey playa sediments compose the upper 205 cm. A single bulk radiocarbon sample from 170-180 cm dating to 6840 +/- 140 yrs BP (Beta-48207) provides limited chronological control. Compositional changes in grass phytolith assemblages are used to define four bio-stratigraphic zones. Phytolith assemblages from the lowest zone, correlated with a paleosol, represent a mixed grass prairie. Climatic estimates derived from this zone suggest an annual temperature near present levels but significantly greater (+100 mm) annual precipitation than today. A transitional zone (145-205 cm) shows increase in warm-season short grasses (grama and buffalo grasses) and Aristida (three-awn). This change is associated with increasing aridity during the Middle Holocene. Warm-season short-grass phytoliths dominate the bulk of the profile (15-145 cm). Climatic estimates for this zone include both lower annual precipitation (-50 mm) and significantly higher (+3 deg C) annual temperatures than today. Reduction in short-grass phytolith percentages and an increase in phytolith percentages from cool-season grasses characterize the upper IS cm of deposit. It is unclear whether this upper zone is the result of climate change or historic land use. The Morgan Playa record demonstrates the utility of phytolith analysis for inferring paleoclimatic change in a region where pollen analysis is unfruitful and unreliable.
Keywords: opal phytoliths; paleoclimate; playas; Rolling Plains
Holocene vegetation and climate variations in the Great Plains remain poorly documented. Quantitative analysis of opal phytolith assemblages provides a tool for assessing vegetation and climate change. Pollen analysis has provided some information on vegetation change in the southern Great Plains (Bryant and Holloway 1985; Fredlund 1995; Hall and Valastro 1995). However, taxonomic limitations of pollen identification to the family level for the most important non-arboreal plants of the region, including the grasses (Poaceae), reduce the sensitivity of the method. In addition, regional pollen records are often discontinuous and may produce spurious interpretations resulting from problematic pollen preservation (Hall 1981; Holliday 1987; Hall and Valastro 1995). The resulting regional knowledge gap in the Great Plains is apparent in mapped summaries of North American fossil data (COHMAP 1988). Geomorphology, geoarchaeology, and related studies have partially filled this vacuum (e.g., Holliday 1989, 1995; Ferring 1990, 1995; Mandel 1992; Meltzer 1995). These studies infer climate change indirectly from the sedimentary, soil, and associated evidence preserved in a landscape. These geomorphic-paleoclimatic approaches to climatic reconstruction lack the more direct link to vegetation change provided by fossil pollen and phytoliths. Quantitative phytolith analysis is an important addition to other methods for inferring climate change in the southern Great Plains. It should be viewed as a complement to, rather than substitute for, other paleoenvironmental evidence.
Opal phytoliths are formed as hydrated silica, taken up from soil solution and deposited within and between the cells of living plants. The decay of the plant tissues bearing silicate casts of cells releases them at the soil surface (Piperno 1988). These silicate bodies are often preserved in situations where pollen and macrofossils are not. Grass phytolith forms are generally representative of grass taxonomy (Twiss et al. 1969; Rovner 1971). Grass phytolith assemblages provide evidence for differences in regional grassland composition because of this strong, albeit imperfect, taxonomic relationship (Fredlund and Tieszen 1994). …