.. Trout Habitat, Abundance, and Fishing Opportunities in Fenced
vs Unfenced Riparian Habitat along Sheep Creek, Colorado 1
Robert J. Stuber 2
Abstract.--Fencing was used to protect 40 hectares of
riparian stream habitat along 2.5 km of Sheep Creek. Colorado.
from adverse impacts due to heavy streamside recreation use and
cattle grazing. Fish habitat within the fenced area was
narrower. deeper. had less streambank alteration. and better
streamside vegetation than comparable unfenced sections.
Estimated trout standing crop was twice as great, and
proportional stock density (PSD) was higher than in unfenced
sections. There was a higher proportion of nongame fish present
in unfenced sections. Projected fishing opportunities within the
fenced sections were double those estimated for a comparable
length of unfenced habitat along the same stream.
INTRODUCTION
have resulted in degradation. FenCing was used
to protect riparian stream habitat along Sheep
Creek, Colorado, from adverse impacts due to
heavy streamside recreational use (e.g .•
recreati ona 1 vehi c1es. campi ng, etc.) and cattle
grazing. Trout habitat characteristics and
abundance are compared in an ongoing evaluation
between fenced and unfenced sections of stream.
The integrity of riparian/stream ecosystems
is extremely important from the perspective of
fisheries, since the quality of eXisting fish
habitat is often directly related to the overall
condition of the riparian habitat. This is
especially true for many medium and small size
coldwater streams, as the smaller the stream,
the more important the riparian zone and the
influence it has (Raleigh 1979). Well developed
riparian vegetation provides a number of benefits
for salmonids, including cover (Raleigh 1982),
streambank stabilization (McCluskey et a1. 1983),
shading for stream temperature regulation (Reiser
and Bjornn 1979; Raleigh 1982), and a source of
allochthonous food input (Meehan et a1. 1977;
Raleigh 1982). Maintaining this integrity is
very important, especially in light of increas­
ing angling use on many of these streams.
The purpose of this paper is to discuss
differences in these parameters, along with
projected potential fishing opportunities in ~e
fenced porti on of Sheep Creek ver$US a comparable
1ength of unfenced habi tat along the same stream.
METHODS
Study Area
Sheep Creek is a sma 11 (4-5 mwi dth) stream
on the Arapaho and Roosevelt National Foresh
within the South Platte River basin in
northeastern Colorado. El evati on of the section
under consideration is approx~mate1y 2,500 m. 1
Low flow is about 0.2 - 0.3 m /sec. It is a c­
stream type accordi ng to a Fores t Servi ce str~am
classification procedure (Rosgen 1985). GradIent
is 1.0 - 1.5% and sinuosity is 1.5 - 2.0.
Dominant channel material is cobble with a
mixture of small boulders and coarse gra~e1: ::
channel is moderately confined. Soils wIthIn
valley bottom are predominante1y coarse textured,
with stable high alluvial terraces.
Multiple land use practices often result in
fish habitat degradation within the riparian/
stream ecosystem. Man has dramatically reduced
the quantity and quality of natural riparian
ecosystems by intensively developing them for
other uses. This development has resulted in
losses of natural vegetation to the detriment of
fish and wildlife and associated recreation
(Swift 1984). Fencing riparian habitat is one
technique which has been employed to protect or
improve fish habitat where conflicting land uses
1Paper presented at the symposium, Riparian
Ecosystems and Their Management: Reconciling
Conf~icting Uses, April 16-18, Tucson, Arizona.
Fisheries Biologist. USDA Forest Service,
Arapaho and Roosevelt National Forests, Fort
Collins. Colorado.
A total of 2.5 km of stream on National to
Forest land was originally fenced in 1~56
protect 40 hectares of riparian/stream habItat
from catt1 e grazi ng impacts. The fences were ars
maintained periodically; however, in recent ye
310 had fallen i nto a state of di srepai r
fencesebuilt in 1982-83. Both t he ori gina l
weret~on and repai r were a cooperative
b l the col or ado Divi sio n of Wil dl i f e and
s~rvice. The fen ced area is divided in to
1 sures (1.9 and 0. 6 km of stream,
i ~elY) , with 1. 3 km betwee~ , and 4. 0 km
ced hab i tat above t hem (Flg . 1).
impac ts on t he unfenced ri pari an habitat
y st re amsi de rec re at iona l use and catt le
'It should be noted t he l and adj acent
. Creek outside t he fence d areas i s i n
ownershi p.
no ngame fis h); ( 2) trout stan di ng crop (kg/ha);
and, (3) tro ut species pr opor t i onal stock densit .
( PSD), an index of pote nt ia l fi shing quality based
on le ngt hs of fish (And erson 1980) . Only fish
~ 150 mm in length were ut il i zed in these
• estima tes. Data from th e individual station s
were pooled (fenced vs. unfenc ed) for the
compara tive estimates. Popul at io n estimate s fo r
trout were made using the Seber and Le Cren
(1967) two cap tu re me t hod. Length s and weights
of a ll captured fish we re taken and trout
standi ng crop estimates were made. Stoc k an d
quality lengths for the trout spec ies PSD
calcul ations were taken from Anderson (1 980).
Fishing Opportunities
Projected potential fishing op portunities
within f enced ar eas were compared wi th th ose f rom
unf enced sections of the stream. Projections
were based upon a fact (known angling use per km
of st ream in Colorado) combined with an
assumption (s tream fishery value can be eq uat ed
to opportu nities for angling use). The rationale
behind this approach is pres ente d below.
2.0 km
The Colorado Division of Wil dlife (1983)
reported that the 13, 600 km of co ldwater stream
supported 3.8 million fish in g- day s in 1980 (280
fishing-d ays per km). According t o the US DA
Forest Service (1980), t hr ee fishing-d ays equal
one recreational visitor day (RVD). Therefore,
1.0 km of stream supported 93 RVD's for col dwater
fishing in 1980. The Division of Wild life als o
assi gns one of six fishery values to a st ream
(None, Poor, Below Average, Ave r age, Above
Ave rage, Ex cell ent). It was ass ume d t hat a
stream with an "Average" f i shery va lue would
support 93 RVD' s (per km per year ) of coldwate r
fishin g , and that th e r emaining fish ery val ues
wou l d support a proportionate amount of projected
use (Table 1).
ci ng was used
al ong Sheep
cts due to
(e. g.,
c.) and cattle
st ies and
ng evaluation
ns of stream.
o discuss
lo ng with
un ities in the
us a comparable
he same streaJI.
Figure l.--Locat i on of fence d (F) and unfenced
(U ) sections of Sheep Creek, Colorado
Habitat Sampling
Habitat mea suremen ts were take n at two
125 mrepresen t ati ve st at ion s i n 1984, one
Insi de and one outs i de t he f enced a re a . The
unfenced sampl in g stat i on was between the two
enclosures. Fiv e stream/riparian habitat
parameters we re measured : stream wi dth, depth,
tt.o.mh;on k alt er at ion ( % eroding banks),
streamban k vegetati ve stab i 1i ty (% veg etation),
and streamsi de cover (domi nant veg etation type).
The width/depth r ati o was also calcul ated. It
~s felt th at thes e six characteristics woul d
yield a comp arative assessment of the overall
fish habitat condition between the two areas.
Vegetat ive stabil ity and st r eams ide cover were
rated from 1 (worst) to 4 (best). Measurements
~d applicable ratings were made accordin g to
procedures outlined by Pl atts et al. (1 983).
Proj ected f ishing opportuni t ies withi n f enced
vs . unfe nc ed areas were obtained by arbi trarily
assigning a fishery value to each respect i ve area
based on tro ut standing c~op and PS D es timates .
Total opport unities wi th i n t he f enc ed section of
stream and a comparable length of unfenced stream
were standardized by multiply ing the respective
projected opportu ni t ies per km of s tream b~
2.5 km (length of fen ced sect ion of stream) .
wi dth) stream
na 1 Forests
n in
of the section
ly 2, 500 m.
Fish Popula t ion Sampl ing
. It is a C·l
Servi ce stream
Th ree stations were sampled in late
985) . Gradien! October , 1983 (one ins i de vs. two outs i de) ; the
- 2.0.
ons~t.of win te r wea ther co nditions precluded
e with a
ij ddltl0nal sampli ng.
Five stations were sampled
se gravel. The n late September, 1984 (three inside vs. t wo
oil s withi n the outside ). All statio ns were 125 m in l ent h, and
oa rs e textured, s~m p1ing wa s don e wi th a generator powe red
e,ectrof is hi ng unit. The habit at stations
samp ~ed in 1984 were two of the fish populat ion
National statlons samp led in bot h 1983 and 1984.
ced in 1956 to ream habitat
c Three fish popul atio n characteristic s we re
fen ces were
between f enced and unfenced areas:
rl
in recent yea tjPared
I relative spec ies occurren ce (bot h t rout and
Table 1.-- Proj ected RVD's (per km of stream per
year) assoc iated with f i shery va lue of stream
in Colorado.
3 11 Fis hery Value
RVD 's
None Poor Be low Average
Ave rage
Abo ve Ave rage
Excell en t
31
62
93
124
155
o
/.
Table 2.--Comparative fish habitat characteristics between fenced
and unfenced sections of Sheep Creek, Colorado, 1984.
Fenced
Unfenced
Average width em)
3.7 5.5
Average depth em)
0.2 0.1
18.5 55.0
Width:Depth
Streambank alteration
(% eroding banks)
Moderate
(26-50)
Major
(51-75)
Streambank stability
(% vegetation)
Good
(50-79)
Fair
(25-49)
4 (Excellent)
(Shrubs)
2 (Fair)
(Grass/Forbs)
Streamside cover (rating)
(dominant vegetation type)
RESULTS
Table 3.--Re1ative abundance (%) of trout and
nongame species captured in fenced vs.
unfenced sections of Sheep Creek, Colorado
1983-84 (average value from both years).
Fi sh Habitat
The stream was generally wider and
shallower in unfenced areas as there was a
significant difference in the average width
(P = 0.0002) and depth (P = 0.0006) between the
two areas. This resulted in a much lower
width/depth ratio within the fenced area (Table
2). Also, there was more streambank alteration,
less streambank vegetative stability, and lower
quality streamside cover in the unfenced area.
Fenced
Unfenced
96.5
85.0
Brown
91.5
72.5
Brook
4.0
9.5
Rainbow
1.0
3.0
3.5
15.0
Trout
Fish Population
Nongame (Longnose Sucker)
On a relative basis there were signifi­
cantly more game fish (trout) present in the
fenced sections, and the number of nongame fish
(longnose suckers; Catostom~s catostomus) was
higher in unfenced areas (X analysis;
P = 0.0001). Brown trout (Sa1mo trutta) were
the predominant species captured in both fenced
and unfenced areas. Brook trout (Sa1ve1inus
fontinalis) and rainbow trout (Sa1mo rairdneri)
were also captured within both areas Table 3).
All captured fish were from wild populations, as
no stocking takes place.
I
(KG/HA)
186
-
132
There was a significant difference
(P = 0.04) in estimated trout standing crop in
1983 between fenced vs. unfenced areas as
estimated standing crop was 96% higher
(91.0 k9/ha greater) within fenced areas
(Fig. 2). Estimated trout standing crop was
127% higher (74 kg/hal within fenced areas in
1984, although this was not significantly
different (P = 0.08) from unfenced areas.
95
58
~
U
Proportional stock density (PSD) values
for brown trout within the fenced areas were 8.3
and 5.1 in 1983 and 1984, respectively, compared
with a value of zero within ·the unfenced areas
in both years. In other words, 8.3 and 5.1%
1983
F
U
F
1984
Figure 2.--Comparative average trout st)and~ng
crop estimates between unfenced (U ~n lorado.
fenced (F) sections of Sheep Creek. 0
312 tock size (~ 200 mm ) brown tro ut were of
of t~e \ ize (325 mm or l arge r) within the
qua l l~Y rea~ in 1983 and 1984, r espect i ve l y , and
fe nce are no qua 1i t y s ize brown trout i n t he
~ere w~ areas in eithe r yea r . The la rgest f is h
unfenced in both 1983 and 1984 were br own trout
captu~~e fenced areas (500 and 375 mm , r espec ­
f~oml )
Ther e were no qual i ty si ze r ainbow or
tl" ~ / t ; out captured withi n either fenc ed or
bro ed a rea sin both yea rs .
un fenc
.
Fish ing Oppo rtun it ies
trout and
ed vs.
, Colorado,
years ).
-
(1 97 9) f elt t hat t he best fluvia l t rout habitat
is as sociated wi th a high st anding crop. Trout
standing crop est imates were 96.5% higher
( 1983- 84 avera ge ) within the fenced are as of
Sheep Cree k. Grea ter trout abundance within
prot ect ed ripar ian st ream habitat has been
reported in other s imilar studies . Gunder son
(1968 ) repo r ted t ha t brown t rout standing crop wa s
31 % greater in an ungr azed vs . an ad j acent graz ed
sectio n of Rock Creek , Montana . In a subseq uent
study, Marcuson ( 1977) r eporte d th at standing
crop i nside the ungraz ed area ha d i ncreased to
3.4 times that f ound i n t he grazed section of
th is same stream. Van Vel son ( 1979) f ound t hat
the fish populati on of Ot te r Cr eek , Neb ra ska,
changed from 88% nongame fi sh to 97% trout aft er
4. 8 km we re fenc ed to excl ude l i ves t ock.
Fishing oppo rtunities within t he f enced
and a comparab le leng t h of unfe nc ed st r eam
are: proj ected to be 310 vs. 155 RVD' s per yea r,
werpect i velY (e .g., 310 RVD's = 124 RVD's/
res
) Th ese proJec
" t"10n s were
km/year x 2. 5 km , etc..
based on t he as sumpti ons that the f is he ry va lue
ithin the fe nced area was "Above Average" (124
; VD's /km/yea\,, ; Table 1) wh ereas it was "Bel ow
Average" (62 RVD' s/k~/yea r) in "the unfenced
area. These assumpt10ns were 1n t urn based on
~e facts t ha t estimated trout st and i ng crop was
twice as grea t "in" fe nced a r e a ~ ~nd th at t here
were qual ity f1s hln g opportumtl es for brown
trout (based on PSD values) within th e fe nced
area, wh ereas none were pres ent in th e unf enced
Pla tt s (1982) felt tha t t hese studi es were
somewhat biased, as t he re was no p r e~trea tm ent
data, and it could not conclu s i vely be prove n
that differe nces in reported trou t abu ndan ce were
not just a natural occurrence . The pr ese nt study
at Sheep Creek lacks pr e- fe ncing fis h popu l ation
and habitat data; however, t he f ence d and
unfenced sections have si mil ar channe l typ e ,
substrate, gradie nt , flow reg i me, and
geomorphology. These s i mi la r ities r edu ce th e
possibility that th e differences were j ust a
natural occurrence and it can 5afe ly be assumed
that pro te ction has re su l ted in superior ha bi tat
conditions and hi gh er t r ou t standing crop.
Unfenced
85.0
72 .5
9.5
3.0
15.0
F
84
standing
·U ) and
!k, Colorado.
DISCUSSION
In addition to providing bette r hab i tat
conditions, it sh ould be noted th at t he well
developed will ow stan ds al ong the f enced sec tions
of Sheep Creek pr obably of f er some prote ction for
trout fr om ang le r s. Fi shabi l i ty i s more
difficult th an i n adjacen t unfence d sections and
this fact may al s o co ntr i bu t e t o t he higher
estima ted standing crop .
Protect i on of the ri parian stream
ecosystem by f enc i ng r esulted i n su perior fish
habitat co nd it io ns . Fenced are as had a nar rowe r
stream width , greater dept h, an d a lower
width/depth rat io . De pt h is import ant in
provi di ng a combi nation of pools, cover and
instream movement areas for trout (Raleigh
1982) . Lower wi dth / de pth rat i os are associated
wi th bet t er fish habitat (Behnke an d Za rn 1976;
Pl atts 1981 ). The more stable st r eambank s
(lowe r % ero di ng banks) and good streamb ank
vegetati ve stabil ity wit hi n f enced areas provi de
~otec tio n fr om erosion and subseque nt siltati on
within t he stream. The predomi nan ce of
well-dev el oped shrubs (willows) with in the
~nc ed area co ntribute more streamside cover
than t he gras s/forbs which pr edomina t e i n
unfenced areas . Platts (1 974) found t ha t
streams bordered by shrubs had hi gher f i sh
stand ing cr ops t ha n si milar sized streams wit h
other vegetat ion type borders. It appea r s that
~avy streamsi de recre at ion use and cattle
g~z i ng ha ve resulted in adverse imoacts to t he
stream/riparia n hab itat in the unfe nced sec t ions
of stream, whi ch wa s evi denced by t he res ults of
the compara t i ve habitat sampl in g (i . e . , wider,
shallower st re am , more s treambank alte rat ion,
etc.) . It has been demonstrated i n numero us
~llier stu di es th at protected secti ons of st r eam
aVe superior fi sh habitat co nditi ons. See
Platts (1982 ) and Pl atts an d Wags ta ff (1 984 ) f or
a synopsis of the resu lt s of t he se st udi es.
In conclusion, fen ci ng f or the pr otectio n of
the ripa r ian/stream ecosystem at Sheep Creek has
resulte d in super ior fis h habitat con ditions , 96 .5%
hig he r estimated trout standing crop , higher PSD
values, and twice the pr oj ected pot ential fishing
opportunities th an in adjacent unfen ced areas.
ACKNOWLEDGMENTS
The author grateful ly acknowl edges
John Bustos of th e Redfea th er Ranger Di stric t for
his coord ina tion of fe nce r econst r ucti on and
main tenance . Th an ks are als o given t o
St eve Pu ttma nn and Doug Kr ieger of the Colorado
Division of Wildl ife for coo r dinati ng the fish
populati on sampl ing . The autho r i s grateful to
th e followin q in div i duals who have assi ste d with
various phases of thi s study: John Englebert ,
Nancy Ryke, De l Lobb, Nancy Do bbs , and
Pat Beauc hamp. Rod Van Vel son of t he Division of
Wil dlife pr ov i de d a co nst ructive r eview of the
man uscrip t . Fin al l y, sincere gratitude is
expressed t o Li nda McLees and Gayle Porter for
typ i ng , and Hen ry Salas for graphic
il l us trati ons.
w' h ~nc re ased estimated trou t standin g crop
1t 1 n t he f enced a r ea wa s t he re su 1t of
sUperio r habitat conditi ons. Binns and Eis erman
313 'LITERATURE CITED Platts, W. S. 1982. Livestock and riparian
fi s hery i nteracti ons: What a re the fact',
pp. 507-515 in Trans. 47th N. American s.
Wil d1 ife andNa tura 1 Resource Conference
Wildlife Management Institute, Washingto~
D.C.
Anderson, R. O. 1980. Proportional stock
density (PSD) and relative weight (W ):
Interpretive indices for fish populations
and communities. pp. 27-33 in Practical
Fisheries Management:. more with less in
the 1980's, Workshop Proceedings. Gloss,
S. and B. Shupp, eds. New York Chapter,
American Fisheries Society.
I
I
Platts, W. S., W. F. Megahan, and G. W. Minsh 1 •
1983. Methods for evaluating stream, a 1,
riparian, and biotic conditions. USDA ~r
Service Gen. Tech. Rep. INT-138.
est
Behnke, R. J. and M. Zarn. 1976. Biology and
management of threatened and endangered
western trout. USDA Forest Service, Gen.
Tech. Rep. RM-28.
Platts, W. S. and F. J. Wagstaff. 1984. FenCi
to control 1ivestock grazing on riparian n~
habitats along streams: is it a viable
alternative? N~ Am. J. Fish. Manage.
4:266-272.
Binns, N. A. and F. M. Eiserman. 1979.
Quantification of fluvial trout habitat
in Wyoming. Trans. Am. Fish. Soc.
108:215-218.
t
Raleigh, R. F. 1979. Closing remarks.
pp. 89-90 in Proc. of Forum - grazing and
riparian/stream ecosystems, O. B. Cope, ed,
Trout Unlimited, Vienna, Virginia.
'
Colorado Division of Wildlife. 1983. Today's
strategy .•. tomorrow's wildlife. A
Comprehensive Management Plan for
Colorado's Wildlife. Third ed., Colo.
Div. Wi1dl., DOW-M-D-I-83, Denver. 96 p.
Raleigh, R. F. 1982. Habitat suitability index
mode1s : Brook trout. USDI Fi sh and Wi 1d1ife
Service. FWS/OBS-82/10.24. 42 p.
Gunderson, D. R. 1968. Floodplain use as
related to stream morphology and fish
populations. J. Wildl. Manage. 32(3):
507-514.
Reiser, D. W. and T. C. Bjornn. 1979. Habitat requirements of anadromous salmonids in Influence of forest and rangeland management on anadromous fish habitat in western North America. USDA Forest Service Gen. Tech. Rep, PNW-96. Marcuson, P. E. 1977. The effect of cattle
grazing on brown trout in Rock Creek,
Montana. Spec. Proj. Rep. F-20-R-21-11-a.
Montana Dept. Fish and Game, Helena. 26 p.
Rosgen, D. 1985. A stream classification Q~a
in Proc. of Ri pa ri an ecosystems and their management, Tucson, Arizona, April 16-18, 1985. USDA Forest Service, Rocky Mountain Forest and Range Experiment Station Gener~ Technical Report series (in preparation). McCluskey, D. C., J. Brown, D. Bornholdt, D. A.
Duff, and A. H. Winhard. 1983. Wi 11 ow
Planting for Riparian Habitat Improvement.
Tech. Note 363, USDI Bureau Land Mgmt.,
Denver, Colo. 21 p.
Meehan, W. R., F. J. Swanson, and J. R. Sedell.
1977. Influences of riparian vegetation
on aquatic ecosystems with particular
reference to sa1monid fishes and their
food supply. pp. 137-145 in Proc.
Symposium on Importance, Preservation, and
Management of Riparian Habitat. USDA Forest
Service Gen. Tech. Rep. RM-43.
Seber, G. A. and E. D. LeCren. 1967. Estimatin~
population parameters from catches large relative to the population. J. Animal Ecology 36(3):631-643. Platts, W. S. 1974. Geomorphic and aquatic
conditions influencing sa1monids and
stream classification - with application to
ecosystem management. USDA SEAM Program
Billings, Montana 199 p.
USDA Forest Service. 1980. Resource planning act assessment, 1980. USnA Forest Servia, Washington, D.C. Swift, B. L. 1984. Status of riparian ecosystems in the United States. Water Resources Bulletin 20(2):223-228. Van Ve1son, R. 1979. Effects of livestock grazing upon rainbow trout in Otter Creek. pp. 53-55 in Proc. of Forum - grazing and riparian/stream ecosystems., O. B. Cope, ed, Trout Unlimited, Vienna, Virginia. Platts, W. S. 1981. Effects of sheep grazing
on riparian - stream environment. USDA
Forest Service Res. Note INT-307.
314