The water potential data were plotted in figure 10, and a regression analysis 

 yielded an equation for a simple linear relationship between the two kinds of sensors. 

 This relationship between sensors shows a definite lag in the response of the ceramic 

 psychrometers as compared to the response of the screen psychrometers . At relatively 

 high water potentials (above -2 bars) the ceramic psychrometers indicated a water 

 potential of bars, but the screen psychrometers were indicating values somewhat 

 lower. As the water potential continued to decrease, the difference in the water 

 potential values indicated by the two types of psychrometers increased. Within the 

 growing range for most plants (-0.5 to -3.0 bars), the ceramic psychrometers showed a 

 lag of from 1 to 2 bars behind the screen psychrometers. The magnitude of this lag 

 increased until at -30 bars the difference in the indicated soil water potential was 

 about 8 bars . 



These water potential data clearly show that once liquid contact is lost between 

 the soil and the psychrometer , the ceramic cup offers a significant resistance to vapor 

 exchange. Also, the magnitude of the resistance apparently increases as the ceramic 

 cup dries. The rate of soil drying in this experiment was somewhat greater than might 

 be expected under natural conditions, with the soil water potential decreasing from 

 to less than -30 bars over a 3-day period. It may be reasonable to assume that if the 

 rate of drying was slower, the ceramic psychrometers would then be capable of maintain- 

 ing a closer equilibrium with the true energy status of water in the soil. However, it 

 should be noted that if the sensor offers a resistance to vapor exchange under rapidly 

 changing water potential conditions, its use under natural conditions would be subject 

 to error and question. The magnitude of this error would be a function of the resist- 

 ance to vapor exchange imposed by the ceramic cup and the rate at which the soil water 

 potential was changing. Therefore, from the standpoint of psychrometer response to a 

 changing water potential, and the ease of construction, the screen psychrometer would 

 seem to offer several obvious advantages over the ceramic cup psychrometer. 



DISCUSSION 

 Evaluation of Thermocouple Psychrometers 



Under conditions where the free energy status of water is changing rapidly, it is 

 apparent that ceramic cup psychrometers lead to spurious estimates of water potential. 

 This clearly points out the need to evaluate the response characteristics of instruments 

 under the environmental conditions for which they are to be used. For instance, althoug 

 the rate of soil drying under natural conditions may usually be somewhat less than re- 

 ported for the data in figure 10, there are conditions where the soil water potential 

 can be expected to decrease more rapidly. A decrease in soil water potential of over 

 30 bars in a few days is not unusual in the soil surface of high-elevation harsh sites 

 where high radiation loads , steep vapor-pressure gradients, and high windspeeds are com- 

 mon. If ceramic cup psychrometers are used under natural conditions where a high rate 

 of decreasing water potential can reasonably be expected, then erroneous data should be 

 suspected, especially where water transfer occurs only in the vapor phase. Although the 

 response-capability limits of screen psychrometers have not been determined, these in- 

 struments are free of the sources of error that affect the ceramic cup psychrometers. 



The foregoing discussion suggests that there are several significant joints to be 

 considered in the evaluation of a thermocouple psychrometer which include: (1) the 

 capability of distinguishing small differences in vapor pressure; (2) a linear response 

 to water potential over the range of interest; (3) ease of construction and calibration; 

 (4) a stable calibration with relatively long periods of use; and (5) the ability to 

 respond to rapidly changing water potential conditions. The screen cup thermocouple 

 psychrometer discussed above appears to meet all of these criteria. 



18 



