Laserfiche WebLink
provides period-of-record statistics for stations with multiple complete years. Statistics on stations that <br />have records from 1950 through 1995 are presented in Table 2. <br />Table 2 also provides information about persistence and trends. Persistence was determined by <br />calculating the annual serial correlation for a station. Only serial correlations that were significant at <br />the probability P 0.05 level (meaning that the probability of no true correlation is less than 5%) are <br />reported in Table 2. As expected, persistence becomes more apparent as one moves downstream. This <br />is because downstream stations have more "memory," through reservoir releases and return flows, of <br />previous years' streamflow conditions. Upstream stations will primarily reflect current climate <br />conditions, and will only show persistence if the climate shows persistence, while downstream stations <br />are increasingly influenced by previous years' human activities. None of the long-term stations showed <br />persistence beyond 1 year for the 1950 through 1995 period. Two of the long-term stations on the Rio <br />Grande showed persistence. Long-term stations on the Conejos River did not show persistence. <br />The other time series statistic, trend, was determined by calculating the correlation between annual <br />streamflow and year. Positive correlations indicate increasing trends and negative correlations indicate <br />decreasing trends. Only trends that were significant at the P 0.05 level are reported in Table 2. The <br />degree of correlation is indicated by the correlation coefficient, which can range from -1.0 to +1.0. <br />Three long-term stations on the Rio Grande main stem and one station on the Conejos River show <br />significant, increasing trends over the 1950-1995 period. Again, trends appear in downstream stations <br />rather than upstream stations. This could be caused by recent wet years, increased return flows from <br />non-tributary groundwater extraction, and tighter administration toward meeting the Rio Grande <br />Compact. <br />Figure 2 illustrates the positive trend in Rio Grande discharge at Lobatos over the period 1950 through <br />1995. This positive trend (+0.42) is likely due to drought conditions during the 1950s, above average <br />precipitation in many years since 1960 (particularly since 1980), and strict Rio Grande Compact <br />administration by Colorado beginning in the late 1960s. The longer term trend (1899-1995) at this <br />same station is actually significantly negative (-0.35), probably reflecting the overall increase in <br />consumptive use as lands came under production. <br />Rio Grande near Lobatos, CO <br />1,000,000 <br />~ 900,000 <br />~ 800,000 <br />700,000 <br />~ 600,000 <br />L 500,000 <br />c~ 400,000 <br />3 300,000 <br />~ 200,000 <br />Q 100,000 <br />0 <br />h~ h~ h~ h~ X0`1' ~~ ~~ 1~ 1~` 1~ ~~ ~~ ~~ ~~ ~`1' ~~ <br />Figure 2. Trend in Annual Streamflows at Lobatos <br />C:Acdss\Task2-3.doc Identify Study Period April 16, 1999 Page 3 of 18 <br />