WMOD00071

• Applicant: Western Weather Consultants
• Project Name: Vail Pass
• Date: 11/1/1984
• Weather Modification - Doc Type: Application

EFfECTS OF VARYI~G SNQWPACK O~ S~ALL ~I
Roger ~" Slee?~r~!_ Albert A. Speccer. and Harold ~. Steinhoff

ASSTii.ACT

Nest boxes, live traps, and kill traps were used to
collect data on ch1pmunks {Eutaroias minimus and E:--
quaar1v1tlal1S), deer mice (Peromyseus manieulat~s).
red-backed voles (Clethrionomvs ~apperi), Microtus spp.
(~icrotus meatanus and ~. lon~icaudus) and pocket go-
Ehers (Thornolllvs talpoides) in the San Juan Mountains
of southIJestern Colorado. 'A l.ate snow tree date
corresponaea to a sn~It 1n attainment of breeding cem-
pet.ent.e to a period lat.er in the su:mmer for nIl species
except pocket gophers, for which there were no "breeding
data. Deer ciee stayed sexually active until a later
date, follo~ing this delay in onset of breeding. Delay
1n breeding activity was also found ~ichin the same
year on north aspects as compared to south aspects for
MIcrotus spp. Onset of breed1ng ~as related to snOW-
melt date and 1n1t1ation of plant growth." Microtus
8pp. changed food habits from old growth of herbaceous
plants and.baY~ of shrubs in late winteY to green
plants as soon as they were available. Both mark-
recapture live trapping and kill trap census lines
shoved a marked decline in deer mouse and chipmunk
population densities after winters of heavy snowfall.
Deer mouse populations on Missionary Ridge (Y) were
most strongly related to varying snowpack (X). as
described by Y - -0.05 X + 5.08. Population changes
of the other species in relation to sno~fal1 were not
detected. Lov density. combined with the delay in
breeding prevented the deer mouse population from
regaining a high density level in the first summer
after a ~inter of deep snow. Chipmunk populations were
more resilient and recovered within one summer. Deer
mice were the only regular users of nest boxes. Aver-
age deer mouse litter size declined fro~ 5.5 ~ 0.7 at
birth to 3.3 t 1.0 young at weaning time. Data were
insufficient to compare litter survival in summers
following below versus above average snowpacks.

INTRODUCTION

The objective of the small mammal project ~as to in-
vestiRate effects ot varY1n~ snowfall on asoects of
the population dynamtcs which relate to size of small
mammal populatIons. These aspects may show responses
which demonstrate not only changes in populations. but
the more basic reasons for these changes. Small
mammal numbers fluctuate consider~bly, but there are
environmental reasons for these variations. and snow
may be an important factor. The six Jobs in this
project were aimed at sensitive and investigatable
points of small mammal population dynamics.

Small mammals, although seldom seen, are a part of the
montane ecosystem. Small rodents are food for the
carnivores, but the small rodents are more important
as ,onsumers of primary (plant) production. This con-
sumption of primary production has been estimated at
one percent (Gordzinski et a1. 1966), 1.5 to 2.8
percent (Hansson 1971). and 3 to 47 percent of the
potential food supply (Crodzinski 1971). Generally.
small rodents have little impact on primary production
through the amount consumed. but specific food habits
may conflict with man's interests. Cranivorous (seed
eating) rodents may binder nautral reforestation, and
herbivorous rodents may eat bark from seedlings,
shrubs. and tre.es "in sufficient quantities to girdle
and k.111 these plants. The obvious detrimental affects
of small animals ~ere emphasized in the past. but more

recent-research has brought forth beneficial functions
of small ma~~ls. Grant (1974) found that grassland
small mammals had a significant positive effect on the
quantity of nitrogen in the top soil layer, and this
was the most likely mechan~sm by ~hich small ma~als
may influence primary production. Grant (1974) fur-
ther stated that the t~o major pathways for reintro-
duetioD of material into the biological cycle, physical
introduction of previously unavailable soil organic
matter and decomposition of fresh organic matter. are
both directly influenced by small ~mmals. Thus, any
effects of varying sno~fall on small mammals could
influence other componentS of the ecosystem.

. ;

Increased snowfall may affect small mammals more than
migratory birds and large mammals, because small mam-
mals are non-mi rato and are 1n the same area circum-
annually. Large mammals can QOve to elevations e ow
the effective cloud seeding area which begins at about
2700 m. and only the smaller, less mobile. m&~l6 re-
main in the area of effective cloud seeding. Mammals
that weigh less than about 200 g live beneath the snow
(Pruit 1958) in the space formed ae the snow and
ground interface. This subnivean environment is
characterized by fairly constant temperatures and
saturated air (Pruit 1957). The subnivean space is
formed by the melting of tbe snow at the ground-sno~
interface or by snOW being supported on vegetation
(Coul1anos and Johnels 196)). .

-See+lon.$

0... jtfej

.j

Broad Si~nificance of Results

If deer mou~ populations are related to varying snow-
pack as indicated by the equation on page 444, in
association with Table 9, then a 30 percent iner~ase
in snowpack. on Missionary Ridge could result in a
populat~on decline ot ~l percent in aD average year.
In a liRhter 5no~ year and at a lower elevation the
effect would be less, perhaps a little 8S 25 percent
in a light snowpack year. The population theoretic~
ally could decline to zero if the snowpack were aug-
mented by 30 percent 1n an otherwise heavy snow year
at high elevations.

Deer mice occur naturally at elevations to4267 In
(14,000 feet) and are presumably genetically adapted
there to heavier snowpacks. Ihereto~.any extrrpa~e~
local. population would presumably be replaced by one
better adjusted to heavier snowpacks and the void
Would be of short duration. The new population would
probably exist at ~ lower population level, however,as
indicated by data in Table 9. Also. these predictions
are based on the assumption of a linear relation
bet~een deer mouse populations and snowpack depth.
During the years of study no snowpacks occurred at 90
to IJS percent of dvera~e. Therefore the linear re-
relationship might not be an adequate description. It
could be a .curvilinear. or a threshold response at a
certain level of snowpack.

The implication for deer mouse activities which inter-
fere ~ith ~n. such as girdling of tree seedlings and
competit~on tor torage with livestock and big game, -
.is that these would be lessened as a result ot snowpack
augmentation. The forester could predict that the -
year atter a heavy sno~fall would be a ~ood one to
plant trees.
In Steinhoff, H. ~.. and J. D. Ives (Eds). 1976. Ecological impacts of snowpack augmentation in the San Juan
Mountains. Colorado. Final Report, San Juan Ecology Project. Colorado State Univ. Publ., Fort Collins.
Present address: Dept. of Fishery and ~ildlife Biology, Colorado State Vniv.. Fort Collins.
437

11
;-2 '
!-'

-11-

A-8d-S