Great Salt Lake Pumping Project/Operation

The Pumps Start
Section 73-23-5(l) and(2) of SB 150 of the 1988 General Session of the Utah State Legislature requested the Division of Water Resources to evaluate the first-year operation of the West Desert Pumping Project and report this information to the Energy, Natural Resources and Agricultural Interim Comn-dttee. The report, Evaluation of West Desert Pumping Project, Senate Bill 150, October 1988, analyzed actual performance and evaluated the amounts of water pumped and the effect on the level of the Great Salt Lake, natural gas consumption, evaporation rates in the West Pond and performance of the West Pond. It also defined the operational range and limitations of the pumping project as an emergency flood control project, including the physical constraints imposed on the operation of the pumping project by elevations of the Great Salt Lake and its limitations.

Data were gathered continuously, beginning with project startup in April 1987, and the requested evaluation covered the first 17 months of operation. Pumping data were compiled from monthly pumping operations and performance reports. Actual pumping volumes were determined from hour meter readings reported by the Pumping Plant operator, Dresser-Rand, Inc. Pumping rates were established by flow tests completed by Eckhoff, Watson and Preator Engineers (EWP) and the USGS in August 1987.

Areas of the West Pond were developed from satellite imagery by BYU Professor Woodruff Miller. These areas were similar to area/volume curves developed by the Division of Water Resources, and both sources were utilized to develop area/volume curves used in the operational computer models. Water was returned to the Great Salt Lake from the West Pond by return flow over the Newfoundland weir or removed by AMAX Magnesium for its brine pond system near Knolls, Utah. This information was reported by Bingham Engineering and AMAX Magnesium.

Climatological data, including temperature, humidity, solar radiation, wind speed and wind direction, were collected from seven remote weather stations located in the project area by the Utah State Climatologist's Office, Utah State University. Analyzed information was then employed by an evaporation model developed by EWP which simulated operations of the entire system. The model was capable of hindcasting or forecasting the project's performance, and it provided water and salt budgets for the project using climatological and pumping data.
return to top of page

Pumping
During the first 17 months of operation, April 10, 1987, through August 3 1, 1988, the project removed more than 1.75 million acre-feet of water from the Great Salt Lake. This lowered the water surface of the Great Salt Lake approximately 14.5 inches and caused the lake to recede by approximately 50,000 acres of shoreline. Over 1.4 million acre-feet of water was removed during the first year of operation, representing about 40 percent of the total lake level decline.

Net evaporation is the actual evaporation less precipitation. As of August 31, 1988, the West Pond contained approximately 400,000 acrefeet of water. Net evaporation, therefore, was approximately 1,350,000 acre-feet. Almost 660,000 acre-feet of water had been evaporated through the first year of operating the pumping project. Net evaporation becames the principal means of removing water from the lake. Relatively minor amounts of water returned to the Great Salt Lake from the West Pond.

The first pump went on line on April 10, 1987, with the water surface elevation of the Great Salt Lake near its historic peak. The second pump started up on May 4, 1987, and the third pump joined the battle on June 3, 1987. The pumps were on approximately 80 percent of the time during the first nine months of the project, April through December 1987. During that period the Pumping Plant averaged 122,000 acre-feet of water a month, or 2,025 cubic feet per second (cfs). With below normal inflow, the lake reached its annual peak in February 1988. As the lake level receded, wind tide effect on the intake brine became a major factor in plant operations. Several times a steady southern wind moved water away from the intake canal and caused pumping to be suspended. During the first eight months of 1988, the pumps operated approximately 50 percent of the time. The pumps removed an average of 83,000 acre-feet, or 1,375 cfs, of water per month from the lake.

Evaluating the first year operation of the pumping project against original design criteria was not realistic because design scenarios were based on the West Pond filling during the winter months when evaporation was low and the natural evaporation process working on a large body of water during the summer. The West Pond did not fill, however, until November 1987 when the peak evaporation season was over. Evaluation of the actual performance of the pumping project instead was prorated to an annual figure and compared to anticipated performance. Original design anticipated annual pumping of 1.45 million acre-feet of lake water into the West Pond, evaporating 1 million acre-feet, and returning 100,000 acre-feet to the lake from the West Pond. Prorating the project performance since startup yields an actual performance of - pumping 1.25 million acre-feet of water into the West Pond, evaporating 950,000 acre-feet, and returning only minor amounts to the lake. Comparing these values, it was concluded that the project's actual operation was consistent with anticipated design intent.
return to top of page

Gas Consumption
Monthly volumes of natural gas in million BTUs consumed by the three 3,500 h.p. engines driving the vertical pumps during the first operational year were reported by Mountain Fuel Supply. Gas consumption at the Pumping Plant varied directly with the volume of water pumped, but actual gas consumption was consistent with design specifications. Generally, with each engine operating at maximum speed, the plant consumed 43,000 million BTUs or 39,300 cubic feet of natural gas per month. At minimum speeds the plant consumed 30,500 million BTUs or 28,000 cubic feet per month. Gas consumption generally improved with operator adjustments and fine-tuned engines.
return to top of page

Evaporation
Evaporation rates for the West Pond were estimated using climatological data as a base in the systematic model and comparing the predicted results with observed performance of the West Pond. Normal evaporation was calculated using the translated normal (30-year data base) actual weather data as experienced on the West Desert. Actual pond evaporation was determined from actual freshwater evaporation with allowances for salinity of the brine. As the pond became more saline, evaporation rates reduced, and vice versa. The actual evaporation rates entered in the computer model produced results which closely emulated the actual observed performance of the West Pond.

The first nine months of the pumping operation in 1987, actual evaporation in the West Pond fell below normal rates in all but two months. During the first eight months of 1988, evaporation rates were approximately 10 percent above normal.

The pumping project was evaporating water at a greater rate than anticipated due to climatological factors, i.e., higher than normal wind velocities, and higher than normal temperatures.
return to top of page

West Pond
The surface area and growth of the West Pond was tracked monthly using satellite imagery. The West Pond filled to its operation water surface level range after five months (November 1987) of pumping. From November 1987 through June 1988, the pond operated at a water surface elevation range of 4215.5 to 4216.5 (covering 255,000 to 315,000 acres). Then the West Pond volume declined due to higher than normal evaporation rates and lower than anticipated pumping rates.

A sampling program of the West Pond indicated the pond was well mixed and not normally stratified with light and dense brines. Brine sampling also verified computer model predications of anticipated densities (salinity) throughout the pond.

As anticipated in the project design, from April through August of a given evaporation season, evaporation generally exceeded inflow from the pumps. This created a net reduction in the pond's surface elevation, area and volume. The remainder of the year's inflow from the pumps was greater than evaporation. The pond reacted as predicted by the specialized evaporation model for this project.
return to top of page

Operational Range and Limitations of the West Desert Pumping Project
The operational range of the West Desert Pumping Project is directly linked to the water surface elevation of the Great Salt Lake. The Pumping Plant was designed to operate down to a lake water surface elevation of 4205. Due to limitations of the Intake Canal, the project now can only be operated to elevation 4207. The upper limit at which the Pumping Plant could operate is water surface elevation 4216.5. At this elevation, the gallery way in the Pumping Plant would be flooded.

Other constraints alter the way the project operates. Design features allowed pumping down to elevation 4205, but the U.S. Air Force permit limits pumping to around elevation 4208. At a water surface elevation of about 4217, water would naturally flow into the West Pond area and submerge the Newfoundland Weir. The West Pond would then be an extension of the Great Salt Lake and pumping would be unnecessary.

Limitations of the pumping plant to act as an emergency flood control project are dependant upon the elevation of the Great Salt Lake when pumps start up, the time of year when pumping begins, the extent of the hydrologic cycle being experienced at the time, and the prediction of future hydrologic cycles.
return to top of page

Public Tours to Hogup
Possibly the best seat in town on July 20-22, 1987, was on a state-chartered tour bus to the West Desert Pumping Project.

Public interest in seeing the huge pumps in action and the river of water flowing from the Great Salt Lake into the western desert were hosted by personnel from the Division of Water Resources. Interest was spawned by the flooding, political controversy over spending $58 million for the project, and the highly publicized cross-country truck journey of the huge engines. And perhaps the mystery of the site of the Pumping Plant was an attraction, because it is a part of the state rarely seen by the general public.

Almost 1,500 people were shuttled free the first day along the 10-mile Southern Pacific Transportation Company causeway from Lakeside to the Pumping Plant site at Hogup Ridge. Many people waited up to two hours during a rainstorm that persisted most of the first day for a seat on one of three buses. Two buses were used the next day, and six on the last day.

Permission to conduct the tours was negotiated with the U.S. Air Force and SPTC, because most of the area is military or railroad property.

The first half of the 20-mile round-trip bus ride featured a quick explanation of causeway construction and the flow of water through the project, identification of lake islands and other landmarks, and bits of lake history. Visitors to the plant were first shown photos and drawings in the plant's entrance area about the project's construction and operation. They were then shown to the control room where plant personnel explained plant operation. Ear plugs were passed out and people were escorted on a walk-through of the noisy engine room. Outside again, the tour took in the pumping plant's forebay, an explanation of the natural gas pipeline and operating system, and a trip across the plant's outlet area with a view of the four-mile long canal to the West Desert. The return bus trip was all questions and answers.

A second round of tours was held August 19-2 1. Visitors were charged $2 a person for the tour, and reservations were requested. Final tours, on a weekend, were held September 12-13. Reservations and a $2 fee were also required for the Saturday and Sunday tours.

An estimated 4,700 people saw the pumps during the three tour sessions. In addition, numerous tours were conducted during the project's pumping period for students on school district buses.

Response to the tours was gratifying to the governor and division engineers and managers. Typical comments were:

State officials believed response to the tours was overwhelmingly positive.
return to top of page

Early Problems
Project engineers anticipated problems with the pumping project once it got underway, because it had been constructed and put into operation so quickly. Happily, only three problems briefly shut down some or all of the pumping operation.

The first was a burned out lower bearing in the large pump of Unit I during the first week after startup. The pumping project's sophisticated sensor system detected the problem before major damage occurred. The damaged bearing was quickly and easily replaced. The bearing cost an estimated $50,000, but it was covered by a manufacturer's warranty.

Nearly six months into the full pumping operation, a cooling water pump failed, shutting down the pumping operation for several days. The lake brine apparently corroded the pump's shaft near a coupler and the pump erupted from its base. The cooling water pump served the three pumping units, distributing lake water to engine heat exchangers and to cool engine oil and turbochargers.

It was determined the coupler and shaft were made from different grades of stainless steel, and the heavy brine caused crevice corrosion on the lower grade 1.5-inch shaft. A replacement cooling water pump with a 316 stainless steel shaft and coupler was installed.

The third problem was a phenomenon called Glauber salt, or mirabilite, that thinly coated the pumps' impellers and section bells during the winter and shut down the pumping project from December 14, 1987, to February 2, 1988, and at approximately the same time in 1988-89. The coating wasn't uniform and caused almost imperceptible vibrations that were caught by sensors in the Pumping Plant. The turning impeller was a natural attractor.

Officials at Great Salt Lake Minerals Company, who had encountered Glauber's salt during their operations, explained the Glauber's salt is a colorless crystalline sulfate of sodium that apparently can form on objects in the Great Salt Lake when the lake brine is 28 to 30 degrees, and can occur anytime between mid-September to February. Although divers removed most of the coating from the pump impellers with hot water shortly after the first occurrence, project managers determined nothing could be done to prevent the occurrence of the Glauber salt.
return to top of page