Roger Rosentreler
Department of Botany
Universiry of Monrana
N{issoula,Montana 59812
The Zonalion of Mosses and Lichens
Along the Salmon River in ldaho
Abstract
Q)mmunitics of mrxses rncl lichens lxlon high rvater mark o. th€ Salmon River jn Idaho are
v i c r v e d a s d i s r i n c t a s s o c i a t i o n sT. h e s c z o n c s m a v b e a s m u c h a s s e v e r a lm c t c r s i n i l c p t b . t s a s e do n
d r e s p e c i e sp r e s e n t a n L l o n f l u c t u a r i n g r v a t e r l e v c l s , f o u r z o n c s a r c f o u n d . R e s u f t s f r o m p a i r e d
t r a n s t c t s i n d j c a t c t h a t s t . b i l i i r o f t h e s u b s t r a t e ,f o r c c o f t h e c u r r c n t , a n d d i s t a n c e a b o v c t h e l o r v
s ' a t e r l c l c L d c t c r r n i n c s p c c i c sd o m i n a n c e . C o m p a r e d r o t h c h c a d r v e t e r st,h e l o w e r r e a c h e so f t h c
r i v e r h a r c m o r c i a i o r . r b l e c , , n r t i r i o n sf o r g r o w t h o f m o s s e sa n d l i c h e n s a n c l f o r c l e v e l o p m c n to f
u ( ) n a t i o np a e r n s .
The Salmon Rivcr
is unique because ir is thc largcst river in the wesfern Uoited
that Lloes nor her,e reguLator) clams. Its headrvarers are ptorected by Wilderness
\Vild
an<l Scenic Rircr
clcsi!niltions. The
Salmon
Rivcr
otiginates
States
and
Jn the graniric
Idah,l Bntlrolith in central Llaho. and dre lo\\'cr sccrion fjo\rs over basalr. The zonation
dcscribe.l in fhis paper occlrrs on borh granitc ancl basalt aloog rhe course of the river.
T,ichcns lncl other fofms of vcgerarion can bc use.l for flood frequency analysis and
for calcuhting river chxnnel capacity (Cregory 1976). Lichens may also be useful
in *'rter quality stlclies (Harvksworth 197.1). The zonation of plants along e river
can be use(l for rhe preclicrion of dischatge from channel geometry ar uogauged sires
(Cregory 19i61. lurthermore, absenceof iichens frorn suitable habitats may indicate
- i r r r . l r . r n n e I r . r r r r b . r n . ue r m o d i l i c l i o o .
Crytogem zonations along streams have been reported by Hale (1!J0), Glime
( 1 9 1 0 ) , P e n t c c o s t( 1 9 7 7 ) , r n d C r a $ ' ( 1 9 7 6 ) . T h e s e z o n e sw e r c n o b r o a d e rt h a n 1 0
cnr per zonc antl less chan,10 cn'r total (Cras' 1976)- ln comparison,zonarion along the
Salrnon Rivet is higher and therefore more significant for use in managemeflr.Since
the Salmon Rivcr flucfuatcs as much as severalmerers in depth. these broad zones mly
bc useful inclicatorsof florxl frequency and flood levels (Fig 11.
'Ihc
purpose of rhis stLr(lyis to dcscribe thc lichen ancl moss associationsfound
l|rn.g rhc S:rlmon River anrl rc relate differeoces in speciescomposition to hyrlraulic
.heracteristicsof rhe channel end ro subsrrlte stabiiity. Also of iorerest is rhe rr-r1 in
lhich planr xss{)cierions
changtd over the river's courseStudy Area
Thc stucll rrca incluclcrl the NIain and the l\'IiLidLeFork ot rhe Srlmon Ilivcr. klalto.
llSA. Thc llfi,:lcllcFork tas srudierl flrm f)agrer frLls ( 116l rr) tl-,t nsrre.Lntro its
aontlllence Nitir thc Norrh Fork ol the Srlmon The mein Saln)on River $as studicd
t r o r n i t s c o n f l L r c n csei r l t t h c N I i r L l l eF o I k r o i t s n r o u r h i t r h c S n a k c R i ! e r ( l l i m L
SIring snosmelr in thc rrountainoLrshcerltercrs rcsulrs in mrrximurn fLrrr's. antl
10t
N n r r h s c . r 5 L r e n c eV. o i < x . N ,
I
lqRI
1.
- An overview of vegetation zones on the Lower reaches of thc Snlmon Rive' in ldnho:
Fisurc
'--'Z o " - . r B . N o r m a l l t o o d Z o n e ; C H i g h F L o o dZ o n e ; D E x t r c m e F l o o d
A.-i;;-Vut..
Zone.
fot Vhitebird, ldaho
minimum flows occut in wi[ter (December) Measurements
(/
discharge,
320 mr|sec (11'290
year
average
follos's:
(lorver main Salmon) are as
(130,000
17
c{s)
m'isec
June 1974; minicfs);maximum recordeddischarge3,6130
(US (ieological
1932
(1,t80
December
11
cfs)
sec
mn
mum recordeddischarge.44.7
Survey1976).
The topographyo{ the srudy atea varies from sreephills to a narrow calyon with
vertical w;lls. The riuer channel is usually conlrolle(l by bedrock and thlrs is nalrow
and rarelymeanders.
The surroun.lingvegetationvaries from mesic forestsin the upper reachesaod
in rhe loner re^chesand oo south'facing
on north-facingsloles to xeric grasslands
Pteltlo /lgd rrrcazzerli(Douglas fir) and
in
the
ate
types
slopes.The foiest habitat
(Steeleal '?/' 1981) Shrub steppecon(Ponderosa
series
Pinat pontletov
Pioe)
areasitheseincludeArtenitia' Pattl:titt'
and
rocky
sloPes
oPeo
munitiesoccuron sreeP
Paf/tL ! are dre tiP^rian shrubsand
^nd
Alnat
Rhat
Arttelancbier,
^nd Crrcoctlr1a!.
have the PoteotlaLto suPPott
grasslands
The
dtainages.
and
rrees in the side draws
(fescue) habitat ryPe series
FeJlxctl
(Blneb;nch
ar'i
wheatgtass)
Agro2lroll lpicdtl./Dt
( T i s d a l e1 9 7 9 ) .
Methods
of ren pairecltnnsectsat intervalsof approximately70 knl along
Samplingconsisted
Paired transectsalloweclcompatisonof speciesresPonse\to
tiver.
ti.r" cn,rrr" of the
The Zooation of Mossesaod Lichens
109
Fisurc 2. Effect of current force in the normal fhd zone on the Salmon River in Idaho: A. Thc
upstream side of the rocks which rcccivc thc stronger curreot are co'tered by Dermato
carpan tetic ldtum, B. The protcctcd cddy side of thc rocks are co\eted. by Sn lelia
aqtuaticd.
hydraulic conditions unaltered by elevation or by other changing environmental condirions along rhe river's length. At each sire, individual transects were selected ro
include vatiation in srream hydraulic conditions. For example, a transect pair might
have ooc transecr in a clitect current and a second traosect ar the same iapid oo a
similar rock type but in an eddy curreot. River rocks that n'ere altered by man, with
a knorvn disturbance date, s'ere noted but s'ere not used for traosect sices.
Sampling rvas clone in a low rvater period, allo*'ing easy accessto seasonally flooded
iueas. Line rransecls ran perpendicular ro the shoreline from *'atet levei through the
high flo*l level, as marked by the occutrence of strictly terrestrial vegetatio!. For each
decimetet point along rhe traosect line, species preseoce \\'as recorded. The intervals
betq'een points \l'ere oarrow enough to cletect such small-scale topographic changes
as thc rop or side of a small boulcler.
At each clecimeter point on the transects, the following data were
l) location in rhe currenr,2) rock type, l) degree of siltarion,4) percent slope, and
5) species.Species that occurrccl in fe*er than four transectsaie not rePorred herer
nor ^re occisional vascular plants occurting beneath sample points. Instead, the neatest
non-vascular species Nere recorded if they \i'ete present *'ithio 1 dmi. Percent cover
*'as calculatedby speciespresencear sampied dm points over the total dm points in a
tlansect_
Species of l/errlc,tria were not separated in the sampling- Barhuh rtbiginota aod
110
Ros€nrlere!
G m.mia montdna are often difficult to desring,uishin the field without sporophytes,
and flood waters physicallybatter the leaf awns, so rhey wefe counted togethe! in the
transect data.
Nomenclature of lichens and mossesfollorvs lists of Hale and Culberson (1970)
and Crum et al, (.1973). Nomenclature of vascular plants follows that of Hitchcock
aod Cronquist (1973). Voucher specimensfor the cryptogamsare at MONTU.
Photogtaphswete taken at various sites along the river. Since these may be useful
lor comparisonin funher studiesof the area, 18 of the photographsare oo file at the
Idaho Historical SocietyLibrary, Boise,Idaho.
Resultsand Discussion
The results from transect pairs placed in coltrasting current forces indicate the importanceof curreotto communitycomposition(Fig.2 aod Table 1). As current force
1. 'I'he averrse Dercenl rreqnenc! (\tiih 9;% conlidcnce inlervrl)
flood zone from t0 Daired iransecls conlrasiins siream bydraulic
T-\Bl,Il
Scorleria aquali(a
Del'nralocarpon
reliculrlum
Bare
ro.k
(n,) species Dres.nl)
or sl,ecies jn thc n.rmal
condillons.
9.7% t ',t.3
2 8 . 7 %t 1 0 . 4
1 0 . 0 %= 1 , 0 . 1
2 1 . 5 7 a) : 1 4 . 2
decreases,species domioance shifts from Vertucaria spp. to Dertnalocarpo tet;cdatlh/l
aocl finally to Scodetid dq&atictt,
Several disrinct zones of vegetatioo colresponding to recurreot flood patterns were
found along the Salmon River: low water zone, normai flood zone, high flood zone,
and exlreme high flood zooe. The characteristic species in each zone are as follows
I Fig. .l) :
1. Io*' water zone-V erruc.tritt spp. and aigae.
2. Normal flood zone Dernlatoc.trpo1? telic/.ialam at ScoaLer;a aq dtitn.
3. High fiood zone-Barbtrld. r' b;ginosa ar'd G tnnia tnontana.
of vegetatioo or occupied by terrestrial
4. Extreme high flood zone-balren
vegetation.
For vascular plants associatedwith seasonalflooding along rhe Salmon Rjver see
Appendix A.
Irregular and recurreot clisturbancesate a part of many climax commuoities; for
example, fires ale a part of some climax Ponderosa pinc communities (Steele et al.
191J1).The moss and lichen zones are aPParendy a.lapted to and maintaine(l by different flood frequencies and durations. These communities do not appear ro be successional states leacliog to other vegetation types! nor do they accumulate soil. Two
old photographs from the early 1900s (60-52.869 and 60-175.19 in rhe Idaho Historical
Society Library) sho$'the presenceof dreseve.qetationalzones,proving their prolonged
existence.
Zones on smooth vertical {alls *'ere disrinct at a resolution of 1 dm. Hoq'ever,
zones oo broken topography tlid not have distioct divisions. These irregularities can
be explained by physical configuration ancl are comparable to lhe vegetatiooal zones
The
nation of Mosses and Lichens
Htgn
velocity
CURRENT
(:'''t\\)\
-\
aqualrca
'
\
.
NO
\...
\
.
.
\\
Barrerr
v&cLly
Slabte
Figure l.
S U B STRATE
Unstab{e
Hypotherical model for the effect of cutrent velocity and substrate stability in controlling
species composition in the normal flood zone.
fouod on a mountainside with broken topography. Sampling at decimeter intervals
recorded irreguiarities rhat caused some noise in the zonation patterns; a larger sampling
interval might avoid this. Nevertheless, what appears as noise io the results may be
explained at a finer level of analysis as substrate stability, orientation of microsites to
rhe current, aocl sheltering by other rocLs.
Low Water Zone
The lorl waret zone cootaioing l/etacaria spp., crustose lichens, q'as poorly sampled
in transecrs because much of the zone was submerged. Veracaria is adapted to and
occurs on submerged rocks in many regioos of the world ( Swinscov' 1968).
N o r m a lF l o o d Z o n e
The normal flow zone conaited Detmatoc.trpon reticrLatam, an umbilicate lich€n,
and ScoaLeria 4q atica, ^ moss. Species dominance is apparently cletermined by subsrrate srability, curreot fofce, aod height above the waler level.
I12
Rosentreter
HtoHFtooozoNE
Barbula rubtovwsa
Grimmrt
Height Crn)
nwr$onrL
NORMALFLOODZONE
5
Dermafocarpon r€ticulotum
Scoulerio oquotico
Water level
\&lIucorl
LOWWATERZONE
-spp./ bare rock
Figure 4. Sp€ciescomposition*'ithin zonesand actualheights (:- 1 dm) of eachzone taken from
t6e averageof t*'o eddy transectsat Telcher Crcek on the lower reachesof thc Salmoo
Rivcr in Idaho.
Sites rvith increasingly stable subsrrate had a higher ptoponion ol S. dq .altc.1,
less of D. reticalahht, ancl leasr of Verrzcaria sPp. (Fig. l). Unstable substtates were
often in the same location as strong cutten!; in such cases the factors controiling the
vegelalion were not clearly distinguishable. An unstable substrate may be a soft, easilv
The Zonation of Mossesaod Lichens
111
eroded rock type (Swinscow 1968 ) ot ^ hard rock type thar moves as part of the bed
load. River banks must contain stable rock material to support a moss or lichen flora
("a rolling srofle gathers no moss"). Bed load formulae (Hebertsol 1969, Einstein
1950) are based on the principle that the capacity of the stream to transport sediment
along its bed varies directly with the difference between the shear srfess actiog on the
bed particies and the critical shear sttess required for ioitiarion of particle motion. The
bed load is related to the amount of $-ater, the stream's profile, and the hydraulic
gradient; therefore, rhe grearer the bed load capacity, the larger the r<xk which can
be transported (U.S. Bureau of Reclamation 1973).
Colonization by aquatic lichens is siow (Swioscow 19613). For example, large
boulders (1 m3), *'hich were moved in 1930 for road coostruction. show oo moss or
lichen flora colonization non', although a rich flora is adjacent to rhe boulders.
Curreot velociry controls the vegetational distribution in several ways. Increased
current velocity is direcriy related to an increased ability to carry greater suspended
loads. The type, size, speed, and amount of the suspended load io a river appears ro
detennioe distribution of rnosses aod lichens in its channel. For example, rhe Grand
Canyon of the Colorado River carries large amounts of course sand and silt io its
suspended load rnd has no moss or lichen flora (personal observation), n'hereas a river
such as the Salmon, with a moderate suspended load, can suppon ao aquatic moss and
lichen flora.
The structure of the river chanoel creates differences in culrent force and, in con,
sequeoce, differences io species dominance. Results ftom rhe traosect pairs indicate
that, as curreor increased, the vegemtioo became strucruraly more compacr and tougher
in texture. Transects located in stronger currents consistenly contained D. reticulat m,
rvhich has a more compact growrh habit and is apparently tough-textured, rather than
the soft-textured S. ,lq/./.1t;ca(Fig. 2 and Table 1). Cornpact growth habirs were less
susceptibie to desruction by hydraulic pressure. The textural differeoce may reflect
selection by the abrasive forces. The less abrasive rhe forces occurring at a site, the
softer the texture of the species rhan can occupy rhese sites. Therefore, the force of
current, size, speed, and amount of sediment load withio a river's cross-seciol determines the species composition at any given sire.
l.ligh Flood Zone
The high flood zone contained !\!'o mosses, Barbala ntbiginosa and Grnttma ntontana.
This zone occurs along rhe liver's course more consistenrly than any other zone. Height
irbove mean water level rvas rhe only facror that appeared to relate to the high flood
zone. Given a long periotl without a hich flood, relrestrial vegetarion probably could
temporatily invadc this zone.
E x l r e m eH i g h F l o o d Z o n e
The extreme high flood zone is an area rhar usually supporrs terrestrial velletation.
lo 1911, an exrremely high flood occurred v.hich clearly marks this zooe (Fig. 1).
At presenr,this zone is predominaotly barren of vegetarion. The aquatic mossesancl
lichens in the other chree flood zones are strongly attached to the rock substrate.
Therefore. these species resisted the desrrucrive hydraulic and abrasive forces of the
extrernc high floo<1, rvhereas terrestrial species were clislodged. The age of terrestrial
I f .i
Roseotreter
speciespreselt io the extreme high flood zone could be used to date the txcurrence
of the last previousextremehigh flood (Cregory 1975).
Patrernsof presenceand absenceof vegetatioo suggestcooditions required for the
survival of aquatic moss and lichen communities. These requitemeots are a stable
substrare,Iack of sttong abrasive forces, fluctuating water levels, and high dissolved
carbon dioxide levels ( Ruttner 1963, Gessner1950).
The Salmon River is ideally suited for an aquatic moss and lichen flora zonation.
Much of the streamchannelis composeclof solid stablebedrockor latge stableboulders;
the abtasive forces are moderateaod decreasein areasnot io the direct culrent: the
Salmon is uodarnmedand naturally exhibic a large recutrent seasonalfluctuarion: and
it is rapid throughout its length, *'hich results in an abundant supPly of dissolved
catboo dioxide. Mossesare unable to photosyothesizeusilg only bicarbonateioos as a
carbon source (Rr:toer 1963, Hynes 1970). Therefore, they are confined to water
where thele is an adequatesupply of dissolved carbon dioxide Rapids also Protect
streamsidevegetarionby restricting destructivewinter ice floes to the midstream
and LowerReaches
of ZonationbelweenHeadwaters
Comparison
The headwarershad an averagetotal height (a 95 Percent confideoce interval) ot
srreamsidelichen and mosszonesof 1.9 m ( L .3 m), comparedto that of ovef 8.0 m
( ! 1.1 m) for the lower reaches.Headwater transectshad an averageof only 80
percent (l: 32) cover compared to 99 percent ( ! 2) cover in the lower reaches.
These differencesmay be explaioed as follows. First, rhe headwatershave a mote
opeo canyontopography with less stable rocks. The river gradient is steeperand thus
the suspendedload cao be of a larger size and speedthan in the lower reachesFlooding
of the headn'atersis of short duration becauseof the small size and oarrow elevational
range of its drainage. By comparison,the lower feachesof the river dtain a much
larger area and a broader elevationalrange,resulting in more frequent floods of longer
duration. The aquatic moss aod iicheo flora in the lower rEachesis therefore consistentliy flooded for a looger period of time than those in the headwaterreaches
Sioce the climate is wetter in the uPPer reaches,there are more seePageareasthat
are not dependenton flooding for moisture.In comparison,the lack of moisture in the
xeric climate of rhe lower reachesmakes rnore of the flora dependentuPoo flooding.
These differencesmay accountio part for the greater developmentof zonation patterns
in the lower reaches.
'fhe
Barbala'high Ilood
lnformation from this stLrdycould be used in sevetalways.
zone could be used as an iodicator of channei capacityancl of high flood level at ungualiedsites.At any given site and date,the water level could be comparedto the vege
rarionai zones,thereby evaluatingthe flood state. The frequencyof high fiood that derermin€srhe BubaLa'hi,ghflood zonecould, in a future study,be calculatedby correlating
the Bath a level to flood levels at a gauged site. Speciescompositlun ar enl given
and the forceof the culrent at that
site coulcltell us abourthe stabilityof that substrate
site.
Acknowledgments
Institution,JackStanfordof rhe Universityol
I chankMasonHale of chc Smirhsooian
MontanaBiolosicaLStation,and BruceMcCunefor their commentson this Paper'
The Zonarion of Mosses.rnd Lrchen.
l1t
Litelature
Cited
Craw, R. C. 1976. Streamside bryophyte zonations. New Zealand Journal of Botaoy 14: 1!,28.
Crum, H. A.. \tr. C. Steere. and I' E. Anderson. 1973. A new list of mosses of North Amcricn north
of Mexico. The Bryolosist 76: 8t-130.
Einstein, H. A. 1950. The Bedload Iunctions for Sediment Transporrarion in Open Cllannel Flows.
USDA S^rl Clnren,:t.cn Senre Technical Bullerin No. I026.
Gessner, F. 1950. Dic okologischc Bedeutung der Stromungscheindigkeit lliessender Gewasser und
jhre Messung auf Kleinstem Raum. Archivs fiir HydrobioloSie 41: 159-165.
Glirne, J. M. 1970. Zonation of bryophytes in the headwaters of a New Hampshire stream. Rhodora
72:276"279.
Gregory, K. L. 1976. Lichcns and determination of river channel capacity. Earth Surface Process 1:
2/3-285.
Hale, M. E. 1950. The lichens of Aton forest, Connecticut. The Bryologist 53. 181-213.
-.
and $/. L. Culberson. 1970. A fourth checklisr of the lichens of the continental Unired
S t a r e sa n d C a D a d a .T h e B r y o l o g i s t 1 ) : 4 9 9 . 5 4 3 .
Hawksworth, D. L. 1979. Lichens aod indicators of environmental chaoges. Environmeotal Change
6: 180-386.
Heb€rtson, J. c. 1969. A critjcal review of conventional bed load formulae. Journal of Hydrology
a: L26.
Hitchcock, C. L., and A. CrooquGt. 1973. Flora of the Pacific Nonhwest. University of Washingtoo
Press. Seattle.
Hynes, H. B N- 1970. The lcology of Runaing \faters. University of Toronto Press, Toronto,
Peotecost, A. 1977. A compaiison of two mountain streams in Gwynedd. lichenologist 9: 107-111.
Runner, F. 1961. Fundamentals of limnology. University of Toronto Press, Toronto, Canada (logl i s h r r a n s l .) .
Steele, R., R. D. Pfister, R. A. Ryker. aod J. A. Kittams. 1981. Forest Habitat Typcs of Central
Idaho- Intermountain Forest and Range Experiment Station Technical Report INT-114,
Ogden, Utah.
Swinsco*', T. D. V. 1968. Pyrenocarpus lichens: thirteen freshwater species of Verracaria 1n thc
B ' i r i s h l s l e : . L I ( l , e n o l o g i s4t \ 4 - r r
Tisdale, E. \v. i979. A Preliminary Classification of Snake River C?nyon Grasslands jn ldaho.
Forcst lTildlife and Raoge lxp€riment Station Notes No. J2, Moscow, IdahoUn;ted States Bureau of Rerlamation. 1911. Desien of Smali Dams. A \flater Resource Technical
Publi.arion- Sctood Edrrion.
United States Geoiogical Survey. 1976. Vater Resources Data for Idaho rVater Year 1975. rVater
Dara R€port, lD'75, National Tcrhnical Ioformation Service, Springfield, Virginia.
Appendax
A
vascular
plants associrted
Corldallii casoana
Iloilecrlheon leiireri
lrlomisin Undlelnntr
lvirh seasonal flooding alorg
riverbank
\Lood\Yorn
variegaied
horseiail
rive.bank
\Lildrrc
the Salhon
lliler.
(head\raters
only) sill
(headn'aters ody) sjlr
roc\ or sanaly areas
rocky silled:rreas
Il.lui!elrm
nrieFalum
walcr seeps (Dartial shadc)
'n silt
Hlr-\ trilobalr
llelllotus oilictnalis
GrlJ(lrrhizr l(,tidola
I ello{
( hh!olst\
l.ana(rlrn
solden hair! aster
Yillosr
rtrlsrris
('0tr1zi .rnrrltnsi\
116
Roseffreaer
rockt
silied arcas
dislurbed
s$cet
and lariahle
clov.rr
(lorer
reaches)
sand
rounded river rocks
(lower reacbes) deeD sill
Llt]'
Panlcum
E{:rtbntriantrm
scribner
witchg.ass
boJtres
Chcnolodlum
Alocynum
androEremilolium
Zanthium strumr
um
variable to alry sites
(lover
reaches) sand
rocky, sand/sill areas
(lower reaches) lariable
(lo$'er reaches)
sand
ReceioedMarcb 14, 198i
Accepted.fot prbkcation June 25, 1983
The Zonationof Mosses.rnd Li,hen'
11t