Laserfiche WebLink
<br />001835 <br /> <br />and Havasu were treated as river reaches. Data errors were subse- <br />quently detected and operating criteria changed. <br /> <br />Effect of Reservoir Mixing <br /> <br />The procedure used to compute water quality in the reservoirs is <br />based on the complete mixing model that Hendrick[6, Section 7.4, <br />p 150] found useful for reservoirs with a limited history or with <br />changing operating criteria. Details of the program methods are <br />given in the User's Hanual.[2, Section 3.2] <br /> <br />Under normal operating conditions, monthly releases from large <br />reservoirs are a small percentage of the total volume in storage <br />at the start of the month. If the reservoir does not go dry dur- <br />ing the month, the reservoir concentration at the start of the month <br />is used to compute salt masses leaving the reservoir by releases. <br />For Lakes Powell and Mead, monthly releases are roughly 5 percent <br />of the starting contents. <br /> <br />During evolution of realistic operating criteria, early base run <br />attempts resulted in Lake Havasu getting very dry. The normal com- <br />putation procedure resulted in large fluctuations in reservoir salin- <br />ity. For example, an average content of 84,000 acre-feet resulted <br />in a time-weighted concentration of 20,500 ppm with a standard devi- <br />ation of 67,300 ppm and a flow-weighted concentration of 3,550 ppm. <br />These results emphasized the limitations of the normal procedure. <br /> <br />Under the condition of relatively large releases, outflow concentra- <br />tions during the month would be expected to reflect the current month's <br />inflows~ For this situation, the program uses a linear mix of the <br />starting salt mass and water volume with the inflow of salt and water <br />during the month. <br /> <br />Defining: <br /> <br />FACTOR = release volume/starting content (i.e. surface + bank <br />storage) <br /> <br />Several runs were made to investigate the effect of using different <br />limiting values of FACTOR to determine which computational procedure <br />should be used. When the computed value for a given month is equal <br />to or greater than the limiting value, the large release procedure <br />is employed. <br /> <br />Results for three limiting values are given in Table VIII for the <br />Imperial Dam station. The flow-weighted concentration is seen to <br />be identical, the time-weighted concentration to be lower with a <br />reduced limit, and the standard deviation reduced by 15 percent. <br />A limiting value of 0.25 was adopted for program use. <br /> <br />25 <br />