Laserfiche WebLink
<br />o <br />r- <br />.. .., <br /> <br />r, <br />~- <br /> <br />~-~) <br /> <br />-' <br /> <br />.' <br /> <br />CHAPTER II I <br /> <br />RIVER BUDGET RESULTS <br /> <br />budgets presented in this report lie within the range of thE! USGS <br /> <br />estimates. <br /> <br />Trends <br /> <br />The results were analyzed for trends by performing a linear correlation <br /> <br />of annual sal t pickup as a func t ion of the year. Then the s lope of the <br /> <br />regression line was tested to see if it was statistically di.fferent from <br /> <br />zero..!/ Table 16 and Figure 11 display the results. It ca~, be seen that <br /> <br />both Cisco budgets showed decreasing trends in pickup and both State Line <br /> <br />budgets showed increasing trends but only the unadjusted Cisco budget can be <br /> <br />interpreted as having a statistically significant decreasing trend. <br /> <br />There are a number of factors which could cause trends. There are three <br /> <br />~ possible downward influences on salt pickup. First, with the beginning of <br /> <br />Ute Water Service, the need for providing winter water via the irrigation <br /> <br />canal system was eliminated, which would have a downward influence on the <br /> <br />pickup. <br /> <br />Prior to Ute Water (1965), both the Government Highline Canal and the <br /> <br />Grand Valley Canal made winter runs for farmers to fill livestock ponds. The <br /> <br />Government Highline Canal made two to three runs of about 5 to 6 days each, <br /> <br />and the Grand Valley Canal made two runs of about 10-15 days each. After <br /> <br />1965, the Government Highline Canal has only made occasional winter runs. <br /> <br />1/ For a discussion of this method for testing the statistical <br />significance of regression coefficients, see Yerjevich, Vujica, Probability <br />and Statistics in Hydrology, Water Resources, Publications, Fort Collins, <br />Colorado, 1972, pp. 254-257. <br /> <br />. <br /> <br />1II-4 <br />