WSP12246

• File Number: 8210.310
• Description: Colorado River Basin Organizations and Entities - Colorado River Water Conservation District
• State: CO
• Basin: Western Slope
• Date: 1/1/1996
• Author: David L Butler
• Title: Trend Analysis of Selected Water-Quality Data Associate With Salinity-Control Projects in the Grand Valley in the Lower Gunnison River Basin and at Meeker Dome Western Colorado
• Water Supply Pro - Doc Type: Report/Study

,,('"
n'\~'.)
Table 3. Monotonic tr.iJ~ij'(monthly dissolved-solids loads for gaging stations 09095500 Colorado River near Cameo,
09152500 Gunnison River near Grand Junction, and 09163500 Colorado River near the Colorado-Utah State line, water
years 1970-93

IP~riods ace in water years; slopes are in Ions per month per year; percent is lhe slope expressed as percent change per year; p value is the significance level
oflM test; SL. significance lnels, whkh art: HS is highly significant, p is less than or equal to 0.01; S is significant, p is grealerthan 0.01 and less
th,an or equal Ie 0.05; MS i!; m:ll"ginl!!!Y significant p !5 grr.'aler than 0.05 and less than or equal to 0.10; and NS is not significant. p is greater th.m 0.10;
<.Iess than}

Station

Period

09095500

t970-93
1980-93
1986-93

09152500

1970-93
t980-93
1986-93

09163500

t970-93
1980-93
1986-93

Unadjusted monthly load Flow-adjusted monthly load
Slope Percent p value SL Slope Percent p value SL
-686 --{).56 <O.OOt HS -333 --{).27 <0.001 HS
-1,280 -1.03 <.001 HS -687 -.55 .00t HS
-5.440 --4.69 <.001 HS -3,840 -3.3t <.OOt HS

.002 HS --451 -.43 <.001 HS
.074 MS 317 .30 .201 NS
.125 NS 332 .J4 .'::J'li. N'
<.001 HS -1.680 -.61 <.001 HS
.001 HS -1.080 -.39 .009 HS
<.001 HS --4,480 -1.74 <.OOt HS

-660
-866
-2,210

-.63
-.83
-2.26

-2,180
-2,930
-t2,100

-.80
-1.06
--4.72

Annual Dissolved-Solids Loads

None of the trends in the unadjusted annual
dissolved-solids loads were significant (table 4).
The flow-adjusted annual loads for the Gunnison
River at station 09152500 and for the Colorado
River at station 09]63500 had significant or highly
significant downward trends for water years 1970-93.
The LOWESS smooth curve in figure 8 indicates
that the "ow-adjusted annual loads decreased at
station 09]63500 during the 1970's and again after
1986. The trend slope of -{j2,3oo lonslyr for 1986-93
for station 09163500 was highly significant (table 4).
The "ow-adjusted annual loads for 1986-93 at
station 09095500 also were decreasing, and the lrend
was highly significant. The LOWESS smooth curve
for station 09152500 (fig. 8) indicates that most of the
trend in flow-adjusted annual loads for 1970-93 in the
Gunnison River occurred before 1980. The trends in
flow-adjusted annual loads for 1980-93 and 1986-93
(table 4) for station 09152500 were not significant.

There were fewer significant trends in annual
dissolved-solids loads than in monthly dissolved-solids
loads. One problem with using annual data is that the
sample sizes are decreased compared to data sets com-
posed of monthly data. The p value for hypothesis test-
ing is affected by sample size. For a given magnitude

(trend magnitude for trend tests) and variance, p values
tend to increase as the sample size decreases (Helsel
and Hirsch, 1992); therefore, it becomes more difficult
to reject the null hypothesis (of no trend) as sample size
becomes smaller. Although the trend tests on the "ow-
adjusted annual loads for water years] 986-93 are
based on only eight values, the magnitude of the slopcs'
for the two Colorado River stations (09095500 and
09163500 in table 4) were sufficiently large to have
highly significant p values.

Seasonal Dissolved-Solids Loads

Seasonal dissolved-solids loads represcnt the
total load for selected periods, or seasons, within a
year. Seasonal loads were computed by summing~l"
monthly loads of the individual months in a season and
then were Ireated as annual values for performing trend
tests using linear-regression analysis. The trend results
would reflect only the trend slopes and p values for
loads for a particular season and would be independeru.
from the remainder of the year. Flow-adjusted trends
were determined by using the average daily stream-
flows for each seasonal period of each year in the flow-
adjustment models. Trends were analyzed for two sea-
sonal periods; August through October (late irrigation-
season effects) and November through February (low-
flow effects).

18 Trend Analysis 01 Selected Water-Quality Data Assoclsted With Sallnlty-Conlrol Projecls In the Grand Va/lay,
In the Lower Gunnison River Basin, and at Meeker Dome, Western Colorado