Western North American Naturalist Volume 62 Number 3 Article 18 7-30-2002 Full Issue, Vol. 62 No. 3 Follow this and additional works at: https://scholarsarchive.byu.edu/wnan Recommended Citation (2002) "Full Issue, Vol. 62 No. 3," Western North American Naturalist: Vol. 62 : No. 3 , Article 18. Available at: https://scholarsarchive.byu.edu/wnan/vol62/iss3/18 This Full Issue is brought to you for free and open access by the Western North American Naturalist Publications at BYU ScholarsArchive. It has been accepted for inclusion in Western North American Naturalist by an authorized editor of BYU ScholarsArchive. For more information, please contact [email protected], [email protected]. Western North American Naturalist 62(3), © 2002, pp. 257–265 A CLASSIFICATION OF AQUATIC PLANT COMMUNITIES WITHIN THE NORTHERN ROCKY MOUNTAINS John R. Pierce1 and Mark E. Jensen2 ABSTRACT.—A synecological study of aquatic macrophyte plant communities was conducted across northern Idaho and western Montana during the summers of 1997, 1998, and 1999. A total of 111 natural and man-made water bodies were sampled based on a stratification of environmental variables thought to influence plant species distribution (i.e., elevation, landform, geology, and water body size). Plant species foliar cover data were used to develop a hierarchical, floristic-based community type classification with TWINSPAN and DECORANA software. Six planmergent (conspicu- ous portion of vegetative plant body on the water surface) and 24 submergent (vegetative plant body found primarily underwater) community types were identified. Multivariate analysis indicated that all community types displayed signif- icant differences in plant species composition, and the Sorenson’s floristic similarity between communities averaged 10% for planmergent and 8% for submergent types. Canonical correspondence analysis was used to inspect relation- ships between abiotic factors and plant species abundance. Results of this analysis indicated some relationships between species distributions and abiotic factors; however, chance introduction of plant species to water bodies is a process con- sidered to be equally important to the presence of the community types described. Key words: aquatic plant communities, aquatic macrophyte vegetation, ecological classification, synecological study, vegetation classification. Aquatic plant communities are widely dis- aquatic plant community classification has been tributed throughout the Northern Rockies, largely descriptive, with little supporting field occurring in both natural and man-made water data (Schuyler 1984, Sawyer and Keeler-Wolf bodies of glaciated landscapes such as lakes, 1995). The primary objective of this paper is ponds, reservoirs, and low-velocity streams. to provide a quantitative classification of aquatic The ecological value of these communities is macrophyte plant communities for the United twofold. First, they enhance a variety of pro- States portion of the Northern Rockies. The cesses in aquatic ecosystems such as oxygen classification scheme presented is hierarchical production, substrate stabilization, nutrient and designed to nest within the national wet- cycling, improved water quality, and phyto- land and deep-water habitat system first pro- plankton reduction (Nichols 1986, Scheffer posed by Cowardin et al. (1979) and later 1998, Jurgens and Jeppeson 1997). Addition- modified for the Northern Rockies by Rabe ally, these communities provide habitat for and Chadde (1994). Specifically, our aquatic aquatic fauna, nesting sites for waterfowl, and plant community classification describes im- forage for large ungulates such as moose portant types at the subclass and dominant (Scheffer 1998, Lodge et al. 1997, Nichols 1986, levels of these previous classification schemes. Fraser et al. 1982). Our classification provides a much needed tool Despite the importance of aquatic plant for future land management activities within communities to ecosystem function and species lacustrine and riverine systems. habitat, they have received relatively little attention in the scientific literature. To date METHODS most research concerning mesic plant commu- Study Sites nities within the Northern Rockies has focused primarily on wetland and riparian types We sampled 111 water bodies or sites across (Hansen et al. 1995); however, no classification northern Idaho and western Montana during the system exists for aquatic plant communities. summers of 1997, 1998, and 1999 (Fig. 1). Site Even for the western United States, work on selection was based on “subjective sampling 1737 Locust, Missoula, MT 59802. 2USDA Forest Service, Northern Region, Missoula, MT 59807. Corresponding author. 257 258 WESTERN NORTH AMERICAN NATURALIST [Volume 62 in the identification of vascular plant species included Fassett (1940), Hitchcock and Cron- quist (1973), Dorn (1984), Borman et al. (1997), and Douglas et al. (1994), in decreasing order of usage. We identified mosses and algae by the taxonomies of Lawton (1971) and Prescott (1978), respectively. Macrophyte plants were defined in this study as all vascular and moss species, as well as 2 genera of algae (i.e., Chara spp. and Nitella spp.). Voucher speci- mens were collected for each plant species within a macroplot with representative sam- Fig. 1. Location of sampled water bodies used in devel- ples of each species deposited at the MRC in oping a classification of aquatic plant communities for the Northern Rocky Mountains. Missoula, MT. Abiotic information collected at each macro- plot included elevation; minimum and maxi- without preconceived bias” as described by mum water depth; water temperature, pH, Mueller-Dombois and Ellenberg (1974). Our and conductivity within 15 cm of the substrate objective was to describe representative aquatic bottom; and total nitrogen, organic carbon, plant communities for the study area. Accord- and available phosphorus (USDA NRCS 1996) ingly, we stratified our sample to ensure that within a 15-cm-deep surface core of the water we characterized the following environmental body substrate. factors that might influence species distribu- Data Analysis tions: elevation (sites ranged from 628 to 2667 m), landform (valley bottoms to cirque basins), In a previous hierarchical classification of geology, and water body size. Six water bodies aquatic plant communities, Schuyler (1984) with less than 3-m visibility were visited but identified 3 life form groupings: pleustophytes not sampled because we felt that the chance (free floating plants such as Lemna sp.), emer- of missing a plant species or misreading asso- gents (terrestrial wetland communities), and ciated measurements was too great in such benthophytes (plants with their basal portion situations. in or on the water body substrate). Of these life forms, we sampled only benthophytes in Field Data this study. Schuyler further broke down ben- Sampling at each site was conducted along thophytes into 2 growth form types or associa- a water depth gradient that began where tions: submergents (vegetative plant body shoreline terrestrial vegetation ended and pro- parts found primarily underwater) and plan- ceeded to a depth where aquatic plant species mergents (conspicuous portion of vegetative disappeared and the substrate was unvege- plant body on the water surface). In our classi- tated. Sampling of the aquatic plant communi- fication we first split sample data into submer- ties present along these transects was made gent and planmergent groupings. For each within a 405-m2 macroplot following ECO- grouping we then analyzed data concerning DATA vegetation sampling procedures routine- macroplot species presence and abundance ly used by the USDA Forest Service, North- using TWINSPAN and DECORANA software ern Region (USDA, Forest Service 1992). We (Hill 1979a, 1979b). sampled 169 macroplots in the 111 water bodies Two-Way INdicator SPecies ANalysis studied. The number of macroplots sampled (TWINSPAN) was used to develop initial along each water body transect ranged from 1 aquatic plant community classifications for the to 3, depending on the number of communi- macroplots based on species lists and associ- ties present at different depths. SCUBA equip- ated foliar cover values. TWINSPAN is a poly- ment was used extensively in field sampling. thetic, divisive, hierarchical classification tech- Vegetation information collected at each nique similar to the Braun-Blanquet classifica- macroplot included a complete list of all mac- tion method that emphasizes indicator species rophyte plants present, as well as their aver- and the production of an arranged species- age height and foliar cover. Taxonomies used sample data matrix (Gauch 1982). Macroplots 2002] AQUATIC PLANT COMMUNITIES 259 were tentatively assigned to a community type on species richness and abundance using based on their TWINSPAN cluster assign- Sorenson’s similarity coefficient. Results of our ment. Each macroplot was then inspected to MANOVA analysis indicated that all plant see if it contained abundances of indicator communities within the submergent and plan- plant species similar to those of other samples mergent growth form groupings differed sig- in its assigned community type. Macroplots nificantly from each other in average vegeta- displaying low similarity to other samples in tion composition, as interpreted at a 95% con- their community type were reassigned to a fidence level. Additionally, these communities different community type when appropriate. displayed low