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

First, long-period climatic data outside the seeding
area, yet relatively close to it, are very scarce
in the San Juan area. Second, the partial correlation
coefficients in any regression equation have standard
errors of varying magnitudes. In the case of
variables giving high explanation, the standard error
is relatively low. However, as new variables are
added, the overall explanation may go up but the
standard error on each partial correlation coefficient
in the regression equation also increases until it
is generally larger than the coefficient itself.
In other words, trying to explain tree growth by a
~arge number of variables is questionable for the
error margin is too large. On the other hand, to
use fewer variables with lower standard errors
reduces the overall percentage explanation so
prediction becomes similarly restricted.

Finally, we return to our original comments that,
given a correlation of tree growth and climatic data
for a set period, any extrapolation either backwards
or forwards in time assumes a constant relationship
between the controlling factors and the tree's
response; as has been pointed out, this is not
necessarily the case.

ANALYSIS OF SYNOPTIC TYPES

A classification of weather types for the western
U.S., developed by members of the California
Institute of Technology in the early 1940's,
was obtained from North American Weather Consul-
tants, Santa Barbara. The classification is of
3 day weather types and covers the period 1928-1958.
The classification was developed to aid forecasting
procedures and recognizes approximately 16 character-
istic sequences of synoptic events which form the
basis of particular weather types. Each type is
identified principally by the positions of the semi-
permanent .elements of circulation, such as the Pacific
Anticyclone, the Aieutian cyclone, and the trajec-
tories of polar outbreaks and cyclone paths, and
produces characteristic surface conditions over the
western U.S. (Figure 27). As an example, Elliott
(1951) notes for type Bn-c "the meridional character-
istic of the flow becomes quite mar~ed with a well
developed trough aloft at about 100 W. Frequently,
a strong dynamic cyclone remains trapped for
several days in the southern Great Basin. The polar
outbreak...flows westward through mountain passes as
well as southward along the eastern front of the
Rockies and produces much-bel ow-normal temperatures
throughout most of the western U.S....Precipitation
is likely to be excessive near the mean trough."

Initially, the weather types were grouped into two
major units: zonal types (i.e. those in which the
principal component of the circulation is W - E);
and meridional types (i.e. those in which the
principal component of the circulation is N - S).
The percentage frequency of these two groups on an
annual basis, 1928-1958, is shown in Figure 28 .
There is clearly a decrease in zonal types and an
increase in meridional types up to approximately
1951 when a reversal of the two trends occurs.

The frequencies of meridional and zonal types have
been further investigated on a seasonal basis.
Only the winter months will be discussed here
(Figure 29). A more detailed analysis is pre-
sented by Bradley (1976).

300

'0100
.;
z

MERIDIONA~S ______ /'\ /\

/ -" V \ A

'\ ZONA~TYP;.S ^ ^*I \.~
~ '-' ""J V \/"'-*

50

o
1928 30 32 34 36 38 40 42 44 46 48 50 52 54 56 1958

Figure 28. Synoptic Weather Types of North America;
frequency of zonal and meridional types.
Starred line indicates the minimum esti-
mates due to absence of data.

120

100

'040
.;
z
20

..........\
I \ I
\ I \ I
\1 \ I
\ I " \I
'. I ZONAL TYPES
V

o
1928 30 32 34 36 38 40 42 44 46 48 50 52 54 1956

Figure 29. Synoptic Weather Types of North America;
winter frequencies of zonal types
(dashed line) and meridional types
(solid line). Years indicate beginning
of winter seasons.

Zonal types show a marked fall in frequency at
approximately 1941-1942. The mean seasonal fre-
quency from 1928-1929 to 1941-1942 is 72 days out of
151. For the period 1942-1943 to 1957-1958 the mean
is 61 days. Further, the variability of the
occurrence increases markedly in the latter period.

For meridional types, the analysis shows the
complementary discontinuity in the frequency of
types around 1941-1942, the mean for the earlier
period being 79 days and for the latter period being
90 days. Thus, the ratio of zonal types to
meridional types during winter months is generally
about 1:1 in the earlier period and 1.1.5 in the
late period. In anyone season, marked deviations
from these values will result in extreme weather
situations.

Further work using the California Institute of
Technology classification was undertaken in relation
to daily climatic data for Durango and Silverton.
Average climatic characteristics associated with each
type on a monthly and seasonal basis were computed
for each station for the period 1928-58. On this
basis it was hoped that observed fluctuations in the
climate of southwestern Colorado could be accounted
for synoptically.

Precipitation

-Winter: Fifteen types occurred during the 30
winters examined; four of these types accounted for
approximately 60 percent of precipitation at Durango
and Silverton, though they occurred on only 47 per-
cent of days (Table 7). Types EL and E occur when
an extensive ridge of high pressure fro~ the north
causes a southerly shift in storm tracks and hence
precipitation. Type A occurs when a strong ridge

59