chart is proportional to the concentration of each 

 component in the original water sample. 



The saturometer, unlike the other two methods, 

 is designed to measure total gas pressure in situ 

 and consists of a length of semipermeable silastic 

 tubing which is attached to a pressure gauge. The 

 tubing is inserted in the river and after gas equili- 

 bration across the tubing is achieved, the gas 

 pressure is measured directly. 



While the Van Slyke and gas chromatographic 

 methods of measuring dissolved gas levels are 

 appropriate for small-scale research studies, they 

 are not suitable for long-term monitoring of dis- 

 solved gas levels. Water samples must be collected, 

 returned quickly and carefully to the laboratory, 

 and analyzed by highly trained individuals at a 

 relatively sophisticated laboratory. Routine moni- 

 toring at a large number of locations using either 

 of these methods would be both inconvenient and 

 expensive. 



The saturometer, however, has been designed 

 primarily for routine monitoring. It is manually 

 operated, relatively inexpensive and designed for 

 use by individuals with limited training. However, 

 investigators are somewhat skeptical about the 

 quality of the data acquired using this device. This 

 skepticism results primarily from the variability of 

 the results obtained using different devices. This 

 may be due in part to allowing an insufficient 

 amount of time for this system to come to equilib- 

 rium. Fifteen to 30 min is commonly required, with 

 the last small change in pressure being the most 

 time consuming. High pressure leaks are also fre- 

 quent occurrences and are difficult to detect and 

 eliminate. If these problems are satisfactorily 

 solved, this instrument may be the most useful of 

 the currently available devices for routine monitor- 

 ing. The three methods described above are manual 

 techniques which do not lend themselves to 

 unattended operation as would be required to inter- 

 face with automatic data collection equipment. No 

 device capable of unattended monitoring of dis- 

 solved gases is currently available. Should such a 

 device become available, it would certainly be the 

 most convenient and perhaps the most cost effective 

 way of maintaining a large monitoring program. As 

 a result of this need, the Seattle District, Corps of 

 Engineers, asked CRREL to investigate the possi- 

 bility of configuring a monitoring device capable of 

 unattended operation and incorporation in a data 

 collection network. 



EXPERIMENTAL APPROACH 



Basically, two different approaches can be 

 envisioned with regard to constructing an 

 unattended monitoring system. One could either 

 measure total gas content of the water in situ or the 



gas could be removed quantitatively from the water 

 and analyzed in the gaseous state. While the first 

 approach would be ideal, the number of applicable 

 methods are quite limited. Since any device used 

 must be capable of measuring both oxygen and 

 nitrogen the in situ dissolved oxygen electrodes in 

 common use are not sufficient. In addition the 

 possibility of monitoring dissolved nitrogen in a 

 similar manner is very remote. In fact, the only 

 current in situ approach which merits consideration 

 for unattended operation is one which physically 

 measures total gas pressure, like the saturometer. 



On the other hand, if the gases were removed 

 from the water matrix, many current state-of-the-art 

 chemical detection techniques are applicable. These 

 include many common techniques such as gas 

 chromatography, mass spectroscopy, and optical 

 spectroscopy, as well as various less expensive 

 specific gas detectors. However, the use of any of 

 these methods requires a satisfactory solution to 

 two problems: 1) River water must be supplied on 

 a continuous basis to an external device while sam- 

 ple integrity is maintained with respect to total gas 

 concentration; 2) Dissolved gases must be quanti- 

 tatively stripped from the continuously supplied 

 water sample. 



Faced with two alternatives, a limited time 

 schedule and a small budget, a decision was required 

 on whether to pursue the in situ or gas phase 

 monitoring approach. While the in situ method of 

 directly sensing total gas pressure as in the 

 saturometer has some obvious advantages, the 

 problems identified with manual operation of the 

 saturometer would be even more significant for 

 unattended use. While these problems are potentially 

 solvable, the more novel second alternative, with a 

 wider choice of detectors, seemed to have a higher 

 probability of success and was thus chosen for addi- 

 tional study. With this in mind, the FY-74 feasibility 

 study was directed toward developing prototype 

 systems to solve the two problems previously men- 

 tioned. Although time did not allow for the develop- 

 ment of a prototype detector system, it was hoped 

 that a specific detector, which was both applicable 

 and relatively inexpensive, could be identified for 

 future development. 



SYSTEM CONFIGURATION 



Pumping System 



Initial discussions with Corps personnel at 

 Libby Dam, Montana, indicated that significant 

 problems had been encountered in attempting to 

 pump river water supersaturated with dissolved 

 gases. Bubble formation within the pump lines was 

 so intense that the water supplied became efferves- 

 cent. After discussions with CRREL engineers, a 



702 Jenkins 



