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

Table 3. Trend analysis for monthly precipitation data from Pagosa Springs (1939-1966) with tree growth
data from station SJS-1 for 1939-1970.
Month Trend Coeff. Index Para ~ Ratio Total N Missin~
July (P) .51 1. 38 18 8 2.25 28 1
Aug (P) .25 .62 19 7 2.71 28 1
Sept (P) -.37 .46 12 12 1.00 28 3
Oct (P) .17 .58 17 7 2.43 28 3
Nov (P) .41 1.49 16 10 1.60 28 1
Dec (P) .91 4.02 22 4 5.50 28 1
January .68 2.03 18 7 2.57 28 1
February .41 2.05 14 12 1.17 28 1
March .62 2.02 17 8 2.13 28 1
April -.38 1.02 8 18 .44 28 0
May -.07 .95 11 12 .92 28 4
June .58 2.00 17 9 1.89 28 0
July .06 1.55 11 15 .73 28 1
August -.73 .36 8 18 .44 28 1
September .30 3.12 9 15 .60 28 3
apara11e1 trend.
b trend.
Opposit~

Table 4. Means and standard deviation of tree ring widths of each study site chronology during the
periods for which there are temperature and precipitation data. Correlation coefficients
between the tree ring chronologies for the same periods.

Durango 1896-1970
Mean S.D. SJS-1
(mn) (mn) r1
1.39 .34 1.0
1.33 .54
2.15 1.84
Pagosa Springs 1942-1970
Mean S.D. SJS-1
(mn) (mn) r1
1.33 .18 1.0
.82 .26
1. 53 .38
Va11ecito Dam 1944-1970
Mean S.D. SJS-1
(mn) (mn) r1
1.33 .19 1.0
.79 .22
1.50 .35

Tree-ring
Site

SJS-1
SJS-2
SJS-3

SJS-1

SJS-2

SJS-3

SJS-1
SJS-2
SJS-3

SJS-2
r2

SJS-3
r3

.62
1.0

.18
.17

1.0

SJS-2
r2

.37
1.0

SJS-3
....:L
.40
.75
1.0

SJS-2
r2

SJS-3
....:L
.37
.70
1.0

.37
1.0

As a result of the trend analysis, data from two
weather stations were selected for the first attempt
of detailed analyses. These stations were Durango
and Pagosa Springs. These data were used in multiple
stepwise regressions. One regression was run for
Durango (1895-1965) with 22 independent variables,
and a second regression was run for Pagosa Springs
(1940-1965) with 3) independent variables. These
initial regressions were done in cooperation with
Bradley and Barry (1973). (See Barry and
Bradley, this vol. p.43 ).

In the first analysis it became evident that Durango
was too distant from the coring area to be used to
account for growth response variance. However, it

I
~

is worth mentioning that the two main variables were
June mean maximum temperature and precipitation for
the preceding December. These together accounted for
only 18 percent of the variance of tree growth.

In the regression with Pagosa Springs data, four
main variables were indicated, accounting for 71
percent of the total growth response variance.
These variables were precipitation for the preceding
December, June mean maximum temperature, August mean
maximum temperature and April mean temperature.

In both regressions, precipitation for the December
prior to the year of growth layer formation is a
principle variable. In both instances the partial
correlation coefficient is positive. This indicates

74