The osmotic potential of soil water in plant/soil systems

Abstract

Various techniques for measuring the osmotic potential of water in sand and loam at a range of soil water contents were examined. Results were inconsistent and variable when osmotic potential was derived by subtracting matric potential from total potential. Osmotic potential measurements on soil solution extruded at pressure through membranes were also unsatisfactory, probably due to salt sieving in the soil and/or at the membrane. Determining osmotic potential by linear dilution of an extract of 0.5 g g-1 soil solution can be criticized on several grounds, though the results presented for these soils seemed reasonable. The measurement of osmotic potential with in situ salinity sensors worked well in the loam but not in the sand. Measurements of the osmotic potential of displaced soil solutions were satisfactory for both soils. We concluded that the displacement technique was the most suitable for calibrating soil osmotic potential against soil water content, because it was simple, inexpensive in materials and time, and probably the least subject to error. The osmotic potential of soil dried by evaporation alone through a range of water contents was the same as that of soil dried by transpiration via lupins at two transpiration rates and via radiata pine. We concluded that the osmotic potential of the bulk soil in closed pots was independent of the activity of plants over the time scale of these experiments.