Relationship of Map Unit Variability to Shrink–Swell Indicators

Abstract

A study was conducted in the Culpeper Basin in northern Virginia to quantify soil shrink–swell indicators, to relate the indicators to soil properties, and to partition variability into map unit components. Five delineations in each of five map units were selected for study. Three profiles were sampled within each delineation to complete a nested sampling scheme. Three map units of smectitic, high shrink–swell soils with phase names of Haymarket silt loam, Jackland silt loam, and Waxpool loam, all 0 to 2% slopes, were selected. Also selected was a vermiculitic, high shrink–swell soil map unit named Kelly silt loam, 0 to 2% slopes, and a kaolinitic, a low shrink–swell soil map unit named Davidson clay loam, 2 to 6% slopes, eroded. The Bt horizons of Haymarket, Jackland, Waxpool, and Kelly were clayey and had high cation‐exchange capacities (CECs), liquid limits, and plasticity indices (PIs). Soils in the Davidson map unit also had high clay contents but had lower swell indices, liquid limits, PIs, and CECs than the other four map units. Shrink–swell indices and related soil properties exhibited high variability. However, the variability was partitioned within the delineations of each map unit. Each delineation within an individual map unit consisted of the same variability as map units with similar names. Swell indices in all five map units were correlated with liquid limit, clay content, and CEC. However, individual map units exhibited differing relationships. Liquid limit and clay content were the best predictors of swell index in the Haymarket, Jackland, and Waxpool map units, whereas clay content was the best predictor in the Kelly and Davidson map units. Cation‐exchange capacity was weakly correlated with swell index in all five map units.