WMOD00277 - Laserfiche WebLink

Home Browse Search

,~l30 020 Vi ,86-, BI-,85 "5 ~ 10 E " z ,91-,95 RANGE Figure 57. --Snow water equivalent index vs. seasonal runoff, Colorado River Basin. No. of stations with indicated correlation coefficients in idal of 70 stations. 3. Trends in Streamflow and Streamflow B:ecords During the Past Sixty Years A limited study of snow course and streamflow records show that averages have remained rather stable over the past 25 to 35 years. However, the years of low runoff tend to occur consecutively. It is not unusual to have a five-year or longer average which is 100/0; or sometimes even 20% more or less than a 30-year average. Further, for the few runoff records which are available, the average for the period from about 1906 or 1910 to 1925 is 120 to 1400/0 greater than for the past 20, 30 or 40 years. Figure 58 shows the consecutive five- year averages for the period 1906-1967 for the Roaring Fork at Glenwood Springs and the five-year averages for the period 1911-1967 for the Frase,r near Winter Park. Both stations have been corrected I for measured trans -mountain diversions. The drop in flow since the 1910-1920 period of the Roaring Fork is most dramatic. Present use of water within this Basin represents only some 50/0 of the total water produced. This amount has probably increased slightly since the 1910-1920 period, but is not significant. FigurEs 59 and 60 show accumulated runoff for the Roaring Fork at Glenwood Springs and the Fraser near Winter Park in their respective records. Both stations show a definite break in trends of runoff in the middle 19201s, but are reasonably consistent for periods before and aft~r that date. It is believed that further investigations would show' similar trends for streams in the south- west with sufficient records. At the moment, no explanation is available to Ec'xplain the change. It also appears there is a slight trend toward a decline in streamflow during the past 35 years, but it is not substantial. Some years ago an interagency group made an analysis of,a large number of streamflow, precipitation and snow course records. The studies were based on the question "At what period of record will the following year be closest to the average for the period?" On the average this period turned out , to be 1 7 years with about half of the total being between 15 and 20 years. The maximum record considered was 30 years, a common base for USGS and USWB at that time. Both the lack of records and the desirability of having data represent present condi- tions indicates that data based on the most recent periods is preferable for determining means or deviations that are most likely to occur during the periods immediately ahead. 4. The Use of Snow Course Data for Evaluation of Weather Modification Snow courses are utilized to provide an index of snow water accumulation in an area. They have only limited application for verification of weather modification activities. ,The measurement procedure itself is rather gross, but perhaps more accurate than other methods except daily measure- ments of snowboards. There is some variation in melting among the years up La April 1, eopeci8.lly 8.t exposed locations under 10,000 feet. Variation in melting aiter April 1 is so extreme as to preclude use of snow courses to check weather modification results after than date. The use of snow course data is limited further for cloud seeding done on a random basis. This becomes particularly serious when the high probability that both increases and decreases in precipitation have resulted during experimental- type seeding. Snow courses will tend to show even minor changes in patterns between areas over the years. Several combinations of April 1 snow water at upwind snow courses in comparison with those at Wolf Creek Pass have been investigated. These courses are located in the western end of the San Juan Range on the headwaters of the Dolores, Animas and Rio Grande. Correlation coefficients between snow courses in the upwind area and those at Wolf Creek Pass varied from 0.85 to 0.93. An e~ploratory-type investigation of snow course data has been carried out for seeded years at Climax and at Wolf Creek. For this exploration-type investigation at Climax, the perio'd 1960-69 was considered to have been seeded. Actually only about one-third of the precipitation days were seeded and an analysis of precipitation data, presented in V,B above, indicate that both increases and decreases have occurred. The 1939- 59 period was used for non-seeded years, although some commercial seeding was carried out in this area on some of those years. There tended to be more snow accumulated in Fremont Pass area in the 1960-69 period th:m during the years before 1960 on April 1 relative to surrounding snow courses. The Fremont Pass area tended to have slightly less snowfall relative to other areas up to March 1. The snow accumulation during March was high in comparison to surrounding areas so that the overall accumulation to April 1 was somewhat greater than would be expected. If tTue, this result might be expected from the model presented in V,A above if there were more cases with cold 500 mb temperatures before March (these would cause decreases) and more cases w:ifh 500 mb tempera1ures greater 1han -20 during 69