Sediments from Soap Lake, an alkaline-saline lake on the arid Columbia Plateau, demonstrate postdepositional changes in response to a measured osmotic potential gradient through the sediment. The primary mechanism for osmotic readjustment is redistribution of water and dissolved ions through the sediment, altering scdimcnt water content and pore water salinity. This redistribution provides a mechanism for mobilizing ions from exchange sites and from minerals to form pore waters of altered composition, The historical record of aquatic ecosystems preserved in lake sediments provides information about the evolution of ecosystems and their reaction to environmental change. Microfossils are commonly used as indicators of a lake’s history, but in saline lakes, the biota is often limited, its members may exhibit a wide range of tolerance, and their preservation may be poor. In very saline lakes, a wealth of authigenic minerals can be used for environmental interpretation. In some lakes, however, the water is too saline to use microfossils, yet too dilute to exhibit many authigenic minerals; in such lakes the major record is in the salinity of the pore water. Sediments are multiphase systems; redistribution of ions can occur not only among the phases present in a sediment stratum but also between strata. Because of the likelihood of these postdepositional alterations, direct paleolimnological interpretation based on chemical and mineralogical characteristics of the sediment is difficult. The purpose of this study was to evaluate the usefulness of chemical characteristics of saline sediment for paleolimnological interpretation as well as the diagenetic mechanisms altering the saline sediment after deposition. l Supported by National Research Council of Canada Fellowships and National Science Foundation Grant GB 3567, 2 Present address: Division of Life Sciences, Scarborough College, University of Toronto, West Hill, Ontario. In saline sediments, an important diagenetic mechanism is the movement of water and dissolved ions up and down the sediment column. Because the rates are too slow to measure in the field, this movement has to be inferred indirectly. One approach is to measure sediment water potential. Sediment water potential ( @kACd) is defined as the difference between the chemical potential of water in the sediment ( JLC~) and the chemical potential of pure free water at the same temperature and pressure ( ~~0)) divided by the partial molal volume of water (V*,) to convert !Pksed to units of pressure (Kramer 1969) : !a scd = (j-b /-ho> /En. Since the chemical potential of a system is a measure of its capacity to do work, the sediment water potential gradient defines the direction in which water can move spontaneously: water will move from sediment with a high sediment water potential to sediment with a lower potential. The sediment water potential expression can be written as !I! scd RTln ’ =v*W& [ 1 ,o where e is the vapor pressure of water in the sediment and e” the vapor pressure of pure free water at the same temperature and pressure ( Kramer 1969). Thus, .by measuring the vapor pressure of water in the sediment, we can measure the water potential of the sediment and define the LIMNOLOGY AND OCEANOGRAPHY 403 MAY 1973, V. H(3) 404 MARK A. MANTUANI potential gradients that control the direction of movement of water in the sediment, This concept has been used to study the dynamics of salt and water movement in the soil-plant system, and it seems natural to apply it to readjusting wet, unconsolidated saline sediment. It emphasizes the colligative properties of the system rather than simply focusing on the vertical distribution of a few ions, Finally, sediment water potential allows integration of the various mechanisms that can cause water movement, i. e. !?? sea = *k,n + $I + *k, + *k,, where ek,, is matric or capillary potential, *, is pressure potential, !lXk, is osmotic or solute potential, and 9, is gravitational potential. Since many of these components are most easily expressed in terms of pressure units, it is useful to have +\ksed also expressed as pressure, rather than as the activity of water. Soap Lake, an alkaline-saline lake in the Lower Grand Coulee, Washington, lies in the Columbia Basin, in the rainshadow of the Cascade Mountains. Annual rainfall is less than 25 cm per year, and the potential evaporation is greater than the precipitation throughout the year ( Gilkeson 1962). The Grand Coulee was formed by the Columbia River when it was diverted from its regular channel by ice dams formed during the Wisconsin glaciation; upon retreat of the glacier at the end of the Pinedale glaciation, the last substage of the Wisconsin, the Columbia river reoccupied its old channel, leaving a series of lakes in the coulee (Richmond et al. 1965). The level of Soap Lake, the southernmost lake in the series, dropped below its outlet 11,000 years ago, and the lake rapidly became saline. In the last 20 years alteration of the hydrology of this basin by the Columbia Irrigation Project has significantly diluted Soap Lake (Friedman and Redfield 1971). I chose the sediments of Soap Lake for this study because they should record both its evolution from a dilute periglacial lake to an alkaline-saline lake and the recent dilution.