Laserfiche WebLink
<br />. <br /> <br />. <br /> <br />. <br /> <br />- <br /> <br />Fo~this analysis, we chose a 5 percent risk level, coinciding with <br />the probable minimum forecast error terms. We assume that we will <br />operate this year according to the March 24-month study through March <br />1988. This means that we expect the forecast of 7.2 MAF to hold for <br />this year and that we will have an average runoff next year. Z-water <br />releases will be made in the summer and fall of 1987 and flood control <br />releases will ~ occur in the winter of 1988, all as shown in the <br />March 24-month study. To have an event with a 5 percent chance of <br />occuring, we then assumed that the forecast totally went to hell in <br />April 1988, dropping to a lower decile runoff. The difference between <br />an average runoff (8.2 MAF April-July) and the lower decile runoff <br />(4.6 MAF April-July) is exactly equal to the probable minimum error <br />term of 3.6 MAF on January 1 (5 percent chance of exceedence). <br /> <br />Using this lower decile inflow for April 1988 through December 1988, <br />we operated all of the Colorado River Basin resevoirs down past Lake <br />Mead. True, there are some funny things happening this next year at <br />Blue Mesa and Navajo with their refilling, but we thought that it <br />would be best to deal with the real reservoir conditions this next <br />year. Lake Powell's operation was adjusted to recover to the January <br />target of 22.6 MAF and Lake Mead's operation was cut back to <br />downstream requirements. The attached sheets are included so a <br />comparison can be drawn between Powell and Mead, again noting that no <br />adjustments were made in the operation until April 1988 so that we can <br />see the effects of a 'worse-case' scenario. Realistically, we would <br />adjust operations monthly to best match the current hydrology with the <br />system management goals. <br /> <br />,.~ <br /> <br />