WMOD00593

• Contract/Permit #: 14-06-D-7052
• Title: Ecological Impacts of Snowpack Augmentation in the San Juan Mountains, Colorado
• Date: 3/1/1976
• State: CO
• Weather Modification - Doc Type: Report

30

~25
c
o
020
.~
>
15
'0

~IO
:~
~ 5
o
u

1914-38

191 -42 1922-46 192 -50 1930-54 1934-58 1938-62 1942-66

Figure 9. Ames - change in annual coefficient of
variation (1914-38) to (1944-68);
25 year overlapping periods.

80

70

50

c
'"
Z30
"-

20 1894-1904
10

1924-1934

1946-1957
.....------~

1895 \905

25

55

1965

35

45

15

Figure 10.

Winter precipitation (Oct. - March) as a
percentage of annual total (Durango,
1895-1968). Dashed lines denote
drought periods for "Pacific border
areas" (after Thomas 1959).

east of the Continental Divide (circles on Figure 11)
do not show a close relationship like those
west of the Divide and are hence not considered in
the regression analysis. The maximum station ele-
vation used in this analysis is approximately
3000 meters (9800 feet) and hence the relationship
above this level is merely extrapolation. Clearly,
there must be a point above which precipitation does
not increase linearly with height and, given enough
data for higher elevations, a curvilinear relation-
ship would no doubt be apparent.

An interesting change in the precipitation-elevation
relationship is apparent when two periods of data
for the same set of stations are examined. For
twelve stations in the San Juan Mountains, precipi-
tation increased 12.37 cm (4.87 inches) per 304.8 m
(1000 feet) for the period 1940-1949 and only 8 cm
(3.15 inches) per 304.8 m (1000 feet) for the period
1950-1960. This change in the relationship was
due principally to the mean annual precipitation at
higher elevations decreasing more between the tw~
periods than at stations at lower elevations. Such
changes have important implications for ecological
and dendrochronological studies.

Running Mean Analysis

Precipitation amounts on an annual and seasonal
basis were examined by means of 9 year weighted
running means (moving averages) for nine Weather
Bureau stations. The stations chosen all have
records going back to 1914 or earlier and one,

30
,

Precipitation (em)
50 60
,

80

40
,

70

110

Y" 0.3081x - 4.63

3250

100

R" 0.7414 (<0.1% significance I

I
I
I

3000

~90

o
.,
2750 Q:;
::;;

'0
o
~
.;; 80
5
J:

.S

c
2500.~

o
>
.,
UJ
2250

c
o
; 70
~
UJ

2000

60

1750

14 16 18 20 22 24 26 28 30 32
Precipitation (Inches)

Figure 11.

Precipitation-elevation relationship
(1951-1960) for San Juan area, west of
the Continental Divide. Circles denote
stations east of the Divide. Wolf Creek
Pass (elevation: 9425 feet), which has
a mean annual precipitation total of
42.55 inches, plots to the right of
the graph.

Durango, has a good record back to 1896. Nineteenth
century data analysis is discussed below in the
following section.

Nine year weighted running means are calculated thus:

(Value for T) =

1 (T-4)+2(T-3)+4(T-2)+6(T-1)+8T+6(T+1)+4(T+2)+2(T+3)
+1 (T+4)

where T is the value plotted for each overlapping
9 year period. Running means weighted in this way
smooth over the higher frequency (short-term)
fluctuations in a record and render the lower
frequency (longer-term) fluctuations most clearly.
Winter (November to March, included), spring (April,
May), summer (June, July, August) and fall
(September, October) values were computed using a
program developed during the course of this study.
This program greatly facilitates computations and
allows output on line printer, Calcomp plotter, or
microfilm. The following station records were
examined: Ames, Cascade, Durango, Fort Lewis,
Ignacio, Rico, Silverton, Telluride, and Trout Lake.
Of the resulting graphs, two station records,
Durango and Silverton, have been chosen as repre-
sentative and are illustrated in Figures 12 and 13.
As one might expect, the station records are not
identical, but on a seasonal basis the following
features are clear (years mentioned, unless in
parentheses, are the mid-point of a 9 year
period):

51