METHODS OF MEASURING WATER POTENTIAL 



Nonpsychrometric Methods 



Several techniques are available for determining the water potential in the soil- 

 plant system, but some of them have a rather limited application; others are undesirable 

 in view of recent advances in water relations technology. Techniques for measuring soil 

 water potential that are still in wide use but of limited applicability include: tensi- 

 ometers capable of measuring matric potential only between and -1 bar; freezing point 

 depression; electrical resistance units with a sensitive range for matric potential of 

 -0.5 to -15 bars; and the pressure membrane or plate technique for inferring matric 

 potential from water content. Of course, these techniques can be valuable depending 

 upon the specific parameter of soil water being investigated, but inference of soil 

 water potential from them can lead to considerable error. Of particular danger is the 

 inference of water potential from soil water content. The relationships between free 

 energy and water content are different for each kind of soil (and perhaps between 

 samples of the same kind of soil), changes in temperature, and because of hysteresis. 

 These techniques and their probable errors are discussed in detail by Taylor and others 

 (1961) and Slatyer (1967) . 



Many techniques are based on direct measurements of plant water content still in 

 wide use for evaluating the water status in plant tissue. Although Barrs (1968) believes 

 there is some doubt concerning whether water content or water potential is the more 

 important parameter affecting plant growth, there is little argument that water 

 absorption, water transfer from cell to cell, and transpiration losses result from water 

 potential gradients. This question is considered, together with a thorough analysis of 

 methods of determination of water deficits in plant tissues, by Barrs (1968). 



The Thermocouple Psychrometer Method 



Spanner (Peltier) and Richards and Ogata (Wet-Loop) Psy chrome ter s : In recent years 

 considerable research has produced instrumentation capable of sensing the vapor pressure 

 of water in a system. Equations (1) and (2) show that vapor pressure is a sensitive 

 indicator of the water potential; most of the other component potentials also could be 

 measured from vapor pressure. However, since the relative vapor pressure of soil water 

 and plant tissue within the range of usual physiological interest (0 to -75 bars) lies 

 very close to the saturated vapor pressure (95 to 100 percent) , the method must be 

 capable of detecting very small changes in vapor pressure. Spanner (1951) first 

 demonstrated that sufficiently sensitive measurements of the relative vapor pressure of 

 water in this very narrow range of interest can be made with small sensitive thermo- 

 couples. This led to a great deal of research on the development and use of thermocouple 

 psychrometers for measurements of water potential. This method offers great sensitivity 

 and accuracy and can be used either in the laboratory with very small samples or in the 

 field over extended periods. 



The Spanner psychrometer consists of a small thermocouple sealed either in a small 

 chamber or other housing (figure 1). The thermocouple is usually constructed of 0.001- 

 inch diameter chromel-constantan (although Spanner originally used bismuth-bismuth-5 

 percent tin) wires with a small welded bead at the junction; these thermocouple wires 

 are attached to copper lead wires connected to a galvanometer, recorder, or a microvolt- 

 meter. In laboratory models the thermocouple is suspended directly over a soil or leaf 

 sample all of which are inside a sealed chamber, and the chamber is then immersed in a 

 water bath to maintain a constant temperature. Under these isothermal conditions, the 

 vapor pressure of the atmosphere above the sample and around the thermocouple will come 

 into equilibrium with the water potential of the sample, usually within a few hours. 

 After vapor and temperature equilibria are achieved, the water potential of the sample 

 can be determined. A small amount of water is condensed on the thermocouple junction 



5 



