Hypolimnetic Injection - 1995-96

The Student: Ed Cimbalik

The Teacher: Clinton A. Kennedy

Awards: Westinghouse Scholarship Semi-Finalist - Top 300 in the Nation, and Intermountain Junior Science and Humanities Symposia - one of 18 selected official research paper presenters - won Honorable Mention, 6th place overall

The Effectiveness of Hypolimnetic Injection at Enhancing Salmonidae sp. Trout Habitat In Ultra Eutrophic Reservoirs That Experience Hypolimnion Anoxia


The purpose of this project was to research the effectiveness of using hypolimnetic injection to increase cold water salmonidae sp. habitat in Cascade Reservoir, an ultra eutrophic reservoir that has been experiencing hypolimnion anoxia due to annual blooms of two types of toxic cyanobacteria, anabana and microsystis. Thermonstratification prevents a new oxygen from mixing throughout the hypolimnion where it is then accessible by the trout. With the full support of the Idaho Department of Environmental Quality and a grant of $6,545.00 we were able to set up a hypolimnetic injection model in Cascade Reservoir. By injecting cold- oxygentated water from a small mountain stream, into the hypolimnion of the reservoir, through a four inch corrugated pipe, we were able to detect small changes in temperature readings. This indicated that hypolimnetic injection, if built on a large scale under the right conditions, would prove to be an effective way in which to increase cold water salmonidae sp. habitat in ultra eutrophic reservoirs.


The purpose of this project was to research the effectiveness of using hypolimnetic injection to increase cold water salrnonidae sp. habitat and to generate information for local agencies to determine if it is economically worthwhile to implement hypolimnetic injection on a larger scale. Hypolimnetic injection is the infilsion of cold-oxygenated wafer into the hypolimnion of a eutraphic body of water which has become warm and anoxic. The eutrophic lake we studied was Cascade Reservoir on the North Fork of the Payette River in central Idaho. Cascade Reservoir is a 17 mile long, 4 mile wide, man-made, irrigation impoundment that in 1989 produced over 880,000 pounds of fish and provided a million hours of fishing time. Recently, this reservoir has been experiencing many problems associated with cultural eutrophication. Massive fish kills have been occurring regularly, culminating in the almost complete elimination of usable salmonidae habitat in the summer of 1993.

The cold oxygenated water sources we chose to use were small mountain streams coming off the west side of the reservoir' s watershed. The west side of the reservoir is so shallow that the water from these streams is not available for trout habitat due to the warming and depletion of oxygen that occurs before the streams reach deep water.

The project involved laying 500 feet of 4 inch corrugated pipe down the stream bed and another 1000 feet of 4 inch corrugated pipe into the reservoir where it could be effectively utilized by the trout. This system, if running et maximum efficiency, would deliver 187,000 gallons of water per day or approximately 260 gallons per minute(.58 cubic feet per second). Over the two month period in which the project was operating, over 11,220,000 gallons (35 acre feet) of water


Cascade Reservoir, located in central Idaho, is a 26,500 acre draw-down impoundment designed to meet agricultural, flood control, and hydro-power needs. A southern Idaho irrigation district contracts out approximately 200-300 thousand acre-feet of water from Cascade Reservoir every year. Not only does the lake serve ranchers and farmers, it is also one of the top trout fishing lakes in the state and is vital to the tourist based economy of the community. The reservoir is filled yearly by snow runoff from the West Mountain Range which rapidly ascends 5,000 feet above the reservoir. The problem is that for the past few years the process of cultural eutrophication has been destroying the reservoir.

Excess nutrients in the water have led to the proliferation of two types of toxic cyanobacteria, anabaena and microcystys, commonly referred to as blue-green algae. With these two types of bacteria thriving in the ultra-eutrophic reservoir, anoxic conditions have been prevalent dunng the snmmer months. An excessive amount of phosphorus and other nutrients entering the reservoir from all six of the sub watersheds results in uncontrolled growth of the algae. The anoxic conditions in the reservoir occur when the algae die and sink to the bottom of the reservoir. Here in the hypolimnion the bacteria break down the detritus (algae) through the process of decomposition, using up much of the oxygen that trout need to survive. This is extremely harmful to the cold water species which inhabit the lake because of thermostratification, the density layering of water. Thermostratification prevents new oxygen from mixing and reaching the hypolimnion. Trout need water that is 4-7 °C and that is the water which makes up the hypolimnion. With the lake having a large surface to volume ratio and an average depth of only 25 feet, the water in the reservoir's surface is quickly warmed above the temperature tolerance level for trout. Combine this rapid thermostratification with the decrease in dissolved oxygen levels and Usable Trout Habitat (UTH) is rapidly diminished. (See appendix )

During the summer of 1993 when low DO levels in the hypolimnion and high temperatures in the epilimnion forced trout out of their usual habitat, the trout then migrated to the streams lining the west side of the reservoir. The shallowness of the creek mouths kept the fish from accessing the cold water located in the creeks. Fisherman noticed that a large number of fish were living and dying around these creeks. Throughout the summer months the Idaho Department of Fish and Game trapped over 5,000 trout that couldn't make it up the streams and relocated them to a nearby reservoir. The trout were trying to escape almost certain death in the nearly anoxic lake yet they couldn't reach the only available cold-oxygenated water.

It was now obvious that if the fish were to be saved the cold-oxygenated water from the streams would have to be delivered to them. Hypolimnetic Injection seemed to be a feasible way of injecting cold-oxygenated water into the hypolimnion of the reservoir where it would increase cold water habitat. In conjunction with Tom Lance, a local leading expert on Hypolimnetic Injection from the Soil Conservation Of fice, and Jeff McLaughlin, an engineer from the Bureau of Reclamaticn, I started to compile information regarding hypolimnetic injection, anoxic lakes, and thermostratification. Through the internet and library search I was also able to contact people and access libraries around the world I was able to talk with professors, teachers, technologists, engineers, and computer scientists from Oregon, Washington, Idaho, Utah, and even Canada about the physics behind hypolimnetic injection. With hypolimnetic injection being such a new idea it was necessary to use the computer to gather information that was so widespread throughout the world.

With a knowledge of the physics behind hypolimnetic injection it was now time to look at possible sites with Jenny Fisher, a hydrologist from the Boise National Forest Cascade Ranger District. She took us to several streams that were located on U.S. Forest Service land and provided us with all of the stream flow data that had been gathered over the past few years. This data was needed to calculate the amount of pressure and volume that would be flowing through the pipe. To determine the length and drop of the streams we used a clinometer and tape measure, which wasn't easy considering the vegetation on the banks was well developed. We also met with Dewey Worth and looked over many topographical maps of the reservoir in order to find which creek would have the shortest distance to the low water mark. Based on the length of the stream, 500 feet from the road to the reservoir, the amount of drop obtainable over this distance, 27 feet, the low water marks in the reservoir, 1000 feet off shore, and where the land was located, we were able to develop an ideal location for the testing of hypolimnetic injection. With the site on U.S. Forest Service land, the lake being Bureau of Reclamation land, and the Army Corps of Engineers dealing with construction on or in the reservoir, it was obvious that many permits would be required. With the majority of the permits coming from the U.S. Forest Service, Jenny Fisher agreed to help us in filing all of them. The waiving of the NEPA Study permit for research purposes was a big help. Other permits included a 404 permit from the Army Corps of Engineers, Endangered Species permits from the U.S. Fish and Wildlife Service, water rights permits from the Idaho Board of Water Resources, ancient human evidence and Indian artifacts permits from The Archeological Society, and a notice to the public printed in the local newspaper. Upon receiving conformation that the permits would be accepted, we designed several different options for the pipe structure. With the help of local engineers we were able to determine that four inch corrugated pipe would be the cheapest and most effective pipe for this study. We knew that we would run the pipe down the creek bed and out past the low water mark, yet we had no idea how to dispense the water once we got to the end of the pipe. We contemplated the idea of either leaving the pipe open at the end or creating some sort of branch system that would dispense the water over a broader area.(See appendix 2-3, pages 12-13) The branch system seemed as if it would distribute water over a larger area yet the practicality of building and monitoring these didn't seem reasonable. The benefits of using a straight pipe with the end left open seemed to outnumber the disadvantages; it would be easy to construct and monitor and we would probably be able to notice changes in temperature and DO levels to a greater extent. So after careful examination we chose to go with the straight pipe instead of a branched system. Now that we had a pipe system the reservoir, and sunk, work went slow. Even with the 26 volunteers who showed up to help it took upward of eight hours to complete the installation. In order to fill the pipe with water we had to run it in a snake like fashion back and forth through the chest deep water on the shore line. The pipe needed to filled with water before it sunk so that no air would be trapped and restrain water flow. The next step was to hook one end of the pipe and slowly pull it out into the reservoir while the rest of us swam alongside and guided the pipe. Once the pipe was stretched fairly straight, 50 cinder blocks had to be loaded in the boats and dropped along the pipe to serve as anchors. When everything was finally connected, we hooked a buoy to the end in the lake and the other end to a check dam, at this point the water was now flowing through the pipe. The next day when we came to check on the pipe we noticed that a few sections of the pipe were floating, indicating that there was still some air trapped in the pipe. To release the air we poked a few small holes in the pipe and it sank. Later we found out from the diver that not all of the air was out of the pipe. A small section was raised one to two feet off of the bottom, reducing water flow through the pipe. Now all we had to do was wait for the hydrolab to arrive and monitoring could begin. On August 15 the hydrolab arrived and immediately we met with Dewey Worth in order to calibrate the new monitoring equipment.

With the new equipment calibrated we headed to the field, actually the lake, to see if the new water injected into the lake was creating a plume. Testing proved to be very difficult, we had to hold the boat in one spot long enough to run the tests down every meter, on an average it proved to be about six tests per spot. This took a tremendous amount of time; the hydrolab took approximately one minute to adjust every time a reading was taken. We usually tested at eight to ten locations on the monitoring chart and two to three random locations. Six tests at every one of these locations calculates to roughly 60-100 tests and if every test takes a minute or two to run that is an average of two to three hours of testing every time. (Remember this is still summer vacation time!) That total does not include driving time to and from the boat ramp, the time it takes to get the boat positioned right, and the time to write down all of the numbers.

In spite of time constraints, our first real problem did not come until August 26 when we went to monitor and the buoy marking the end of the pipe was missing. How could we detect minute changes in water quality when we didn't even know where to look? We tried dropping a ping pong ball down the pipe, however, it stuck somewhere along the way end never came out. It was time to throw on the masks, snorkels, and fins and start swimming. We were able to follow the pipe out into the reservoir for about 500 feet, then the water got too cold and deep to dive in. Luckily one of my friends at school knew how to scuba-dive, and was able to finish the diving for us, find the end of the pipe, and tie a new marker on. So on the 29 of August we were finally able to go out and monitor the test site again. This time took longer than usual because of wind. Even with anchors on both ends of the boat it proved a challenge to hold the boat in one place with the wind blowing. Once again disaster struck, the buoy was missing again. Again we enlisted the help of the diver in finding the end of the pipe. Instead of simply tying rope on the end of the pipe we used a chain and padlock. The buoy wouldn't turn up missing ever again, in fact we had to cut the pipe to get it off. From here on out monitoring went well. It seemed as though every time we went out to monitor there was anywhere from ten to thirty fishing boats anchored around our buoy. Occasionally while we were testing we would find a spot where the conductivity of the water would jump from 30 to the high 60, indicating that the water from the creek was making a small change. The only problem with monitoring was having to see and smell the lake. The algae blooms got a late start this year due to the extremely wet and windy summer months, but by late September to early October the algae blooms were in full swing and the whole lake looked like a pot of pea soup and smelled like something far worse.

Analysis of Results

Once all of the monitoring was completed the data was organized into data tables and spreadsheets, about 30 pages worth. The first thing I did was get in contact with Jeff McLaughlin, the engineer from the Bureau of Reclamation, and find out exactly how to analyze all of the data. Jeff recommended separating the data and looking mainly at the data gathered right around the mouth of the pipe and the control data from random points on the lake. By just comparing these two sets of data it was hard to tel1 if hypolimnetic injection had any real effect on the water quality in the hypolimnion. Jeffs' next suggestion was to graph the two and compare the graphs.(See appendix 6-7, pages 16-17) This made a world of difference and for once it appeared as if hypoli- mnetic injection did affect the hypolimnion of the reservoir.

If you examine the graphs you will notice that in the graph with the data plotted from the random points on the lake (page 16) the lines are almost straight up and down. It looks as if there is no real temperature change and obviously no thermostratification. Now if you look at the graph with the data plotted from the end of the pipe (page 17) you will notice that during the first meter, 6 meters to 5 meters, the lines on all but two are at angles, then they straighten out. This is what indicates that a change did take place; it shows that the temperature did fluctuate a little bit during that frst meter. We concluded hypolimnetic injection was successful on a very small scale. The reasons for such a small scale change can be attributed to the fact that no thermal stratification occurred. Had thermostratification occurred the cold-oxygenated water that was being injected into the hypolimnion of the reservoir wouldn't have been able to diffuse throughout all of the layers. Instead it would have remained in the hypolimnion, adding both cold water and oxygen. Another reason why hypolimnetic injection made such a small change was due to the fact that this was only a study project built on a small scale. If one wanted to obtain more substantial results then a greater volume of water would have to be injected into the reservoir. Starting next month I intend to make formal presentations to the U.S. Forest Service, Bureau of Reclamation, Idaho Department of Environmental Quality, Idaho Board of Water Resources, Idaho Department of Fish and Game, and The Army Corps of Engineers concerning the results of my hypolimnetic injection study. There is ample room for a larger scale project on the west side of the reservoir. Many of the streams on that side of the reservoir are very large and flow year round, creating a perfect water supply for hypolimnetic injection.

After two years and countless hours of studying hypolimnetic injection I now feel that it is safe to say that, if built on a large scale under the right conditions, hypolimnetic injection will prove effective in increasing cold water salmonidae sp. habitat. I can only hope that the city of Cascade and the agencies will realize that there is a proven, economical way to save the trout of Cascade Reservoir.

This web site brought to you by the Advanced Biology Class at Cascade High School

Back to the top of this page
Advanced Biology Home Page