|dc.description.abstract||Three kinds of soils and four kinds of water were used to prepare laboratory soil-water mesocosms. Soil-water mesocosms were prepared containing 5-cm deep layers of soil and 21 L of water from each of the four sources. Water pH, conductivity, major cations and alkalinity were measured at intervals for 200 days. The Coastal Plain and Blackland Prairie soils caused greater increase in intensities of variables in rain and stream water than did the Piedmont Plateau soils, while conductivity, calcium, sodium, and total alkalinity in rain and stream water increased more in mesocosms with Blackland Prairie soil; magnesium increased more in the mesocosms with Coastal Plain soil. In saline water and seawater mesocosms, the Blackland Priarie soil caused significant changes in all measured variables in well water except pH and conductivity. In sea water, only calcium, magnesium and potassium in the mesocosm water were significantly different from the original source water.
A shaker experiment revealed equilibrium concentration of water quality variables increased as the ratio of soil to water increased. Concentrations of nutrients increased rapidly for some variables for the first 2 days. However, attainment of equilibrium might require more than 9 days. Findings of this study suggest that water quality of the soil-water systems can be predicted from source water chemical composition and soil
properties. However, because of the wide variation in soil characteristics and in the quality of source water, neither the mesocosm technique nor the shaker procedure seems to provide a highly reliable way of predicting water quality in a pond.||en