scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Partitioning vegetation response to anthropogenic stress to develop multi-taxa wetland indicators.

TL;DR: H hierarchical partitioning was used to evaluate the independent influence of geomorphology, geography, and anthropogenic stress on common wetland plants of the U.S. Great Lakes coast and multi-taxa models indicating wetland condition were developed, making them easier to implement than existing indicators.
Abstract: Emergent plants can be suitable indicators of anthropogenic stress in coastal wetlands if their responses to natural environmental variation can be parsed from their responses to human activities in and around wetlands. We used hierarchical partitioning to evaluate the independent influence of geomorphology, geography, and anthropogenic stress on common wetland plants of the U.S. Great Lakes coast and developed multi-taxa models indicating wetland condition. A seven-taxon model predicted condition relative to watershed-derived anthropogenic stress, and a four-taxon model predicted condition relative to within-wetland anthropogenic stressors that modified hydrology. The Great Lake on which the wetlands occurred explained an average of about half the variation in species cover, and subdividing the data by lake allowed us to remove that source of variation. We developed lake-specific multi-taxa models for all of the Great Lakes except Lake Ontario, which had no plant species with significant independent effects of anthropogenic stress. Plant responses were both positive (increasing cover with stress) and negative (decreasing cover with stress), and plant taxa incorporated into the lake-specific models differed by Great Lake. The resulting models require information on only a few taxa, rather than all plant species within a wetland, making them easier to implement than existing indicators.

Summary (4 min read)

INTRODUCTION

  • Coastal wetlands are the focal point of much human activity, both direct and indirect, that threatens their condition and existence.
  • Some plants decrease in abundance with anthropogenic stress, some increase with anthropogenic stress, and some are insensitive to anthropogenic stress.
  • Structures such as roads, dikes, and ditches within wetlands have more direct effects on wetland vegetation than do other anthropogenic modifications.
  • The objective of this research was to develop vegetation-based indicators of coastal wetland condition utilizing only plants that respond (either positively or negatively) to anthropogenic stressors, distinct from naturally occurring environmental factors that influence CAROL A. JOHNSTON ET AL.984 Ecological Applications Vol. 18, No. 4 plant distribution and abundance.

METHODS

  • Site selection and vegetation sampling Ninety wetlands along the U.S. Great Lakes coasts were selected using an objective, stratified random statistical design representing the entire range of existing anthropogenic stress (Danz et al. 2007).
  • Wetlands were also selected to represent three hydrogeomorphic types based on wetland vulnerability to hydrologic forces (Keough et al. 1999): open-coast wetlands, riverinfluenced wetlands, and protected wetlands.
  • Plant sampling was conducted in 1 3 1 m quadrats distributed along randomly placed transects within areas of emergent herbaceous wetland vegetation in the study wetlands selected.
  • Transect endpoint coordinates were uploaded into a handheld global positioning system (GPS) prior to field campaigns, but the location of individual quadrats was determined and recorded in the field.
  • Within each quadrat all vascular plant species were identified to the lowest taxonomic division possible by trained botanists who were tested to ensure consistency of visual observations.

Site environmental characterization

  • All wetlands were categorized according to three nonanthropogenic environmental characteristics: hydrogeomorphic type, ecoprovince, and the Great Lake on which they occurred.
  • The CSI is a generalized stress gradient for watersheds draining to the U.S. Great Lakes coast developed by Danz and coworkers (2007), derived from a GIS database of 149 variables related to five types of anthropogenic stress: agriculture, atmospheric deposition, land cover, human population, and point source pollution.
  • The authors employed the HMI in this study as a measure of within-wetland anthropogenic stress because of the demonstrated importance of dikes in altering Great Lakes coastal wetland vegetation (Sherman et al.
  • A total of 119 plant taxa were evaluated .
  • The HP results were evaluated using one-way analysis of variance to determine statistically significant differences in mean I values associated with the variance components across environmental variables.

Measures of anthropogenic stress

  • The CSI values for the wetlands studied ranged from 0.7 (least disturbed) for the Mismer Bay wetland on northern Lake Huron to 3.6 (most disturbed) for the Huron River wetland on the south shore of Lake Erie.
  • The large standard errors of the CSI averages for Lakes Huron and Michigan reflected large disturbance gradients from very disturbed in the south to relatively pristine in the north portions of those lakes.
  • The HMI ranged from 0 to 111 m/ha, with the three highest values associated with diked wetlands of western Lake Erie.

Basin-wide analysis

  • For the 49 taxa that occurred in 20% or more of the study wetlands basin-wide, the sum of I values for all five environmental characteristics ranged from 0.422 for TYinv to only 0.057 for the submergent Ceratophyllum demersum (CEDE4), the latter value indicating that CEDE4 abundance was poorly predicted by any of the five environmental variables (Fig. 1).
  • The six taxa that preferred the southerly EBF ecoprovince were more invasive (Phragmites australis [PHAU7], TYinv) and weedy (e.g., Boehmeria cylindrica [BOCY], Leersia oryzoides [LEOR], Urtica dioica [URDI]) than those preferring the LMF ecoprovince (Table 2).
  • The independent effect of CSI was relatively small (mean I ¼ 12.3%), and all but two of the 12 taxa with significant independent effects of CSI were also significantly affected by the ‘‘lake’’ or ‘‘ecoprovince’’ variables (Fig. 1).
  • Life-form had a significant effect on plant responsiveness to watershed anthropogenic stress when species were separated into two groups: (1) submerged and floating aquatic plants (i.e., CEDE4, CHVU, Elodea canadensis [ELCA7], LEMI3, Nymphaea odorata [NYOD], SPPO, Utricularia macrorhiza [UTMA]) vs. (2) emergents and shrubs (i.e., the other taxa listed in Fig. 1).

Lake-specific analysis

  • Separately analyzing the data for each of the five Great Lakes increased the average I values for the independent effect of CSI and HMI (Fig. 2).
  • In Lake Erie, only four and three taxa of the 30 taxa tested exhibited significant independent effects with CSI and HMI, respectively (Tables 3 and 4; see Plate 1).
  • Four of the seven species that exhibited significant independent effects of ecoprovince on Lake Michigan were similarly affected in the basin-wide analysis, but on Lake Huron only Carex stricta (CAST8) and Schoenoplectus tabernaemontani (SCTA2) exhibited significant independent effects of ecoprovince at both the basin and lake scales (Table 2).
  • Saginaw Bay wetlands are subject to more stress, particularly from agricultural sources, so that the average CSI value for the six Saginaw Bay wetlands (x̄ ¼ 2.7, SE ¼ 0.08) was more than twice that of the eight northern wetlands on Lake Huron (x̄ ¼ 1.2, SE ¼ 0.10).
  • Of the 58 taxa that occurred in at least 30% of Lake Huron wetlands, 14 exhibited significant independent effects with CSI but only three were related to HMI (Tables 3 and 4).

BOCY EBF* EBF*

  • Eastern broadleaf forest; LMF, Laurentian mixed forest.
  • Species insufficiently frequent to be considered in hierarchical partitioning (HP) analysis at this scale.
  • June 2008 989MULTI-TAXA WETLAND VEGETATION INDICATORS URDI) also exhibited significant effects of CSI (.

Multi-taxa models

  • Stepwise multiple regressions using only those species with significant independent effects of CSI or HMI yielded highly significant multi-taxa models (Table 5).
  • A regression between computed FQI values for the 90 study wetlands vs. actual CSI values yielded a significant fit (R2¼ 0.560, P , 0.001) that was quite comparable to that obtained using the seven-taxon basin-wide model (Fig. 3B).
  • The cover of SCPU10, PHAU7, and Populus deltoides (PODE3) seedlings increased with increasing CSI (positive model coefficients); all three species occurred primarily in Saginaw Bay.
  • The Lake Erie CSI model utilized three of the four species identified as having significant independent effects of CSI (Tables 3 and 5).
  • On Lake Erie, the one-species HMI model distinguished the Hickory Island (HMI¼ 107 m/ha) and Winous Point (HMI¼ 97 m/ha) wetlands because they had the greatest POCO14 cover, but the other eight Lake Erie wetlands had predicted values at or close to the formula’s intercept value of 33 because they were largely devoid of.

DISCUSSION

  • Using individual plant species to indicate environmental condition is a goal that ecologists have sought for decades.
  • In 1950, Eville Gorham (1950) wrote, ‘‘The question of how far different organisms may ‘indicate’ environmental conditions is one of great interest to ecologists, although the fact that so many of the organic and environmental factors concerned may vary independently suggests that close correlations of species to single habitat factors will be rather rare.
  • PHAR3 benefits from nutrients (Mason and Miltimore 1970, Dean and Clark 1972, Kline and Broersma 1983, Maurer and Zedler 2002), sedimentation (Werner and Zedler 2002), and hydrological disturbance (Kellogg et al.
  • The three species that increased with increasing landscape stress in the Lake Superior CSI model, CALA16, Utricularia macrorhiza (UTMA), and CIBU, were not intuitive.
  • Inclusion of CAST8 in the Lake Superior HMI model as a species promoted by within-wetland structures was also initially counterintuitive because CAST8 is negatively impacted by stormwater runoff (Werner and Zedler 2002), but other studies corroborated an increase in CAST8 abundance associated with diked wetlands (Jorgensen and Nauman 1994, Ellison and Bedford 1995, Stanley et al. 2005).

Biogeographical differences

  • The set of plants that exhibited significant independent effects with CSI or HMI generally differed by Great Lake.
  • In some cases, differences among the Great Lakes in the independent responses of plant taxa to stress may be due to the geographic distribution of stressors, rather than the geographic distribution of plants.
  • Such non-linear responses have been previously documented in stream and wetland biota (King et al. 2005, Lougheed et al. 2007).
  • Recognition that individual species can behave differently as a function of biogeography is a concept that is built into commonly used ranking systems.
  • The states of Minnesota, Wisconsin, Illinois, Indiana, Michigan, and Ohio have all developed separate C value lists which assign similar but not identical C values to plant taxa.

Utility of models

  • Ecological indicators should capture the complexities of the ecosystem yet remain simple enough to be easily and routinely monitored (Jackson et al. 2000, Dale and Beyeler 2001).
  • The authors believe that the models presented here are scientifically rigorous, yet require less implementation effort than the currently used FQI, because users need only distinguish the few taxa that are in the models rather than all plants present in a wetland.
  • Sedges are notoriously difficult to identify to species.
  • The models developed here are robust indicators based on conditions that occurred during the sampling period (2001–2003), but their temporal variability is not known.
  • Most of the taxa included in the models are perennials, and many have extensive rhizome systems and other growth forms (e.g., tussocks, floating mats) that would allow them to persist in an area despite water level fluctuations.

ACKNOWLEDGMENTS

  • This research has been supported by a grant from the United States Environmental Protection Agency’s Science to Achieve Results (STAR) Estuarine and Great Lakes program through funding to the Great Lakes Environmental Indicators (GLEI) Project, U.S. EPA Agreement EPA/R-828675.
  • It has not been subjected to the Agency’s required peer and policy review and therefore does not necessarily reflect the views of the Agency, and no official endorsement should be inferred.
  • Michael Aho, Kathy Bailey, Aaron Boers, Spencer Cronk, Charlene Johnson, and Laura Ladwig provided field assistance, and Terry Brown assisted with data analysis.

LITERATURE CITED

  • Seasonal allocation of biomass and nitrogen in four Carex species from mesotrophic and eutrophic fens as affected by nitrogen supply.
  • Journal of Applied Ecology 41:824–835. H. John Heinz III Center for Science, Economics and the Environment.
  • Methods for evaluating wetland condition: using vegetation to assess environmental conditions in wetlands.
  • Cornell University Press, Ithaca, New York, USA.

Did you find this useful? Give us your feedback

Figures (9)

Content maybe subject to copyright    Report

Ecological Applications, 18(4), 2008, pp. 983–1001
Ó 2008 by the Ecological Society of America
PARTITIONING VEGETATION RESPONSE TO ANTHROPOGENIC STRESS
TO DEVELOP MULTI-TAXA WETLAND INDICATORS
CAROL A. JOHNSTON,
1,5
DANA M. GHIOCA,
1,6
MIRELA TULBURE,
1
BARBARA L. BEDFORD,
2
MICHAEL BOURDAGHS,
3,7
CHRISTIN B. FRIESWYK,
4,8
LYNN VACCARO,
2
AND JOY B. ZEDLER
4
1
Department of Biology, South Dakota State University, Brookings, South Dakota 57007 USA
2
Department of Natural Resources, Cornell University, Ithaca, New York 14853 USA
3
Natural Resources Research Institute, University of Minnesota, Duluth, Minnesota 55811 USA
4
Department of Botany, University of Wisconsin, Madison, Wisconsin 53703 USA
Abstract. Emergent plants can be suitable indicators of anthropogenic stress in coastal
wetlands if their responses to natural environmental variation can be parsed from their
responses to human activities in and around wetlands. We used hierarchical partitioning to
evaluate the independent influence of geomorphology, geography, and anthropogenic stress
on common wetland plants of the U.S. Great Lakes coast and developed multi-taxa models
indicating wetland condition. A seven-taxon model predicted condition relative to watershed-
derived anthropogenic stress, and a four-taxon model predicted condition relative to within-
wetland anthropogenic stressors that modified hydrology. The Great Lake on which the
wetlands occurred explained an average of about half the variation in species cover, and
subdividing the data by lake allowed us to remove that source of variation. We developed
lake-specific multi-taxa models for all of the Great Lakes except Lake Ontario, which had no
plant species with significant independent effects of anthropogenic stress. Plant responses were
both positive (increasing cover with stress) and negative (decreasing cover with stress), and
plant taxa incorporated into the lake-specific models differed by Great Lake. The resulting
models require information on only a few taxa, rather than all plant species within a wetland,
making them easier to implement than existing indicators.
Key words: Carex; dikes; emergent; fen; floristic quality; Great Lakes; hydrologic modification;
invasive plants; land use; lemnids; marsh; Typha.
INTRODUCTION
Coastal wetlands are the focal point of much human
activity, both direct and indirect, that threatens their
condition and existence. Environmental managers in
coastal regions and elsewhere seek indicators that can
rank wetlands by their condition or provide early
warning of environmental degradation. Plants are
valued as potential indicator species because their
responses to anthropogenic disturbance are less ephem-
eral than the chemical measurements that are typically
used to characterize environmental quality. Plants also
have the advantage over many faunal indicators in that
they remain in place, simplifying sampling and increas-
ing the likelihood that the biotic indicator is spatially
coincident with in situ stressors. However, not all plants
respond equally to human disturbance. Some plants
decrease in abundance with anthropogenic stress, some
increase with anthropogenic stress, and some are
insensitive to anthropogenic stress. The best (i.e., most
parsimonious) vegetation-based indicator of coastal
wetland condition would utilize information provided
by the plants that are most responsive to anthropogenic
stress, ignoring the rest. Unfortunately, such relation-
ships are not known for all plants. We evaluated plant-
environment relationships using a geographically exten-
sive database for U.S. coastal wetlands of the Lau-
rentian Great Lakes, and constructed indicator formulae
that estimate coastal wetland condition based on the
strength of those relationships.
A multi-taxa vegetation-based indicator that is
commonly used to evaluate wetlands in the Great Lakes
region is the floristic quality index (FQI), also known as
the floristic quality assessment (FQA) or the floristic
quality assessment index (FQAI) (National Research
Council 2001, U.S. Environmental Protection Agency
2002). To compute the index, all plant species within a
wetland are identified, and their average ‘coefficient of
conservatism’ (C value) is computed and multiplied by
the square root of the total number of species within the
area censused (Swink and Wilhelm 1994). C value is a
number from 0 to 10 indicating a plant’s fidelity to
‘remnant natural plant communities,’’ a value of 10
signifying a plant that almost certainly comes from an
Manuscript received 24 July 2007; revised 20 November
2007; accepted 27 November 2007. Corresponding Editor: D. S.
Schimel.
5
E-mail: carol.johnston@sdstate.edu
6
Present address: Departmen t of Biology, Frostburg State
University, Frostburg, Maryland 21532 USA.
7
Present address: Environmental Analysis and Outcomes
Division, Minnesota Pollution Control Agency, 520 Lafa-
yette Road North, St. Paul, Minnesota 55155 USA.
8
Present address: Cleveland Botanical Garden, 11030 East
Boulevard, Cleveland, Ohio 44106 U SA.
983

undegraded natural plant community. The FQI has been
tested and widely adopted as a biological indicator of
human influence, and C values lists have been compiled
by nine U.S. states and one Canadian province (Lopez
and Fennessy 2002, Bourdaghs et al. 2006). The FQI
requires that all plant species be tallied, even though
some species may provide more information than others
about anthropogenic stress, or the lack thereof. Addi-
tionally, the assignment of C values to plant species,
typically done on a state-by-state basis by botanists
knowledgeable of local flora, can be somewhat subjec-
tive (Bourdaghs et al. 2006). The main practical problem
with using FQI, however, is the botanical expertise
needed to identify to species all plants within a wetland.
Thus, the availability of wetland indicators based on
only a few taxa would greatly reduce the effort and cost
of assessing wetland condition by environmental man-
agers.
Initial development of coast al we tland indicators
using field data requires characterization of sources of
anthropogenic stress as well as identification of the
vegetation present. Anthropogenic stressors may be
activities that have occurred within the boundaries of
the wetlands themselves, directly affecting wetland
vegetation and function, or they may result from off-
site activities in the landscape adjacent to the wetland,
indirectly affecting coastal wetlands through fluxes of
water, nutrients, and contaminants (Johnston 1994). An
index of cumulative anthropogenic stress was developed
for watersheds on the U.S. side of the Great Lakes basin
by Danz and coworkers (2007) and was verified to be
related to coastal wetland water quality (Morrice et al.
2007) and biotic characteristics (Reavie et al. 2006,
Brazner et al. 2007). Other studies have also confirmed
that wetland vegetation is affected by upslope agricul-
ture and urbanization (Findlay and Houlahan 1997,
Mensing et al. 1998, Galatowitsch et al. 2000, Lougheed
et al. 2001). The mechanism by which these land uses
affect wetland vegetation is often nutrient enrichment of
runoff (Craft et al. 2007).
Structures such as roads, dikes, and ditches within
wetlands have more direct effects on wetland vegetation
than do other anthropogenic modifications. Dikes are
intentionally constructed in Great Lakes coastal wet-
lands as a means of managing vegetation by controlling
water levels, often with the purpose of maintaining
waterfowl habitat (Sherman et al. 1996, Kroll et al. 1997,
Sanzone and McElroy 1998, Thiet 2002). Diking of
Great Lakes coastal wetlands modifies natural hydro-
logic regimes, leading to nutrient-rich aquatic environ-
ments that are vulnerable to plant invasion (Herrick and
Wolf 2005). Dikes are also used to construct fly-ash
disposal and cooling ponds in wetlands near power
plants, which may raise water levels in adjacent undiked
wetlands via seepage (Wilcox et al. 1985, Ellison and
Bedford 1995). Roads have a variety of ecological effects
on aquatic ecosystems, including disruption of the
physical environment, alteration of the chemical envi-
ronment, and the spread of exotic species (Wilcox 1989,
Forman and Alexander 1998, Trombulak and Frissell
2000). When roads bisect wetlands, they can interrupt
the flow of water, changing the depth and duration of
flooding on both the upslope and downslope sides,
thereby alterin g wetland vegetation (Findlay and
Bourdages 2000). Plant community alterations can also
be caused by deicing road salts (Wilcox 1986). Ditches in
highly impacted landscapes reduce native perennial
importance (Galatowitsch et al. 2000).
Plants used as environmental indicators should be
sensitive to human activities independent of their
sensitivity to other environmental variables. However,
identification of significant plant-environment relation-
ships using field data is hampered by multiple co-
occurring conditions that may obscure, amplify, or
dampen the effects of each other. For example, climate
directly influences plant distribution, but climate also
affects the distribution of anthropogenic practices that
can indirectly affect plant distribution, such as the
presence of agriculture that releases excess nutrients to
coastal wetlands. It is difficult to distinguish the effects
of human activities from the effects of natural environ-
mental variation when the two covary, particularly for
such a large region as the 5900-km U.S. shoreline of the
Great Lakes, which intercepts a range of climatic,
geologic, hydrologic, and disturbance conditions.
We used hierarchical partitioning (HP) as a statistical
tool to identify plant taxa with independent responses to
anthropogenic stressors (Chevan and Sutherland 1991,
Christensen 1992), implemented as a software module of
the public domain statistical program, R (Mac Nally
and Walsh 2004). HP employs goodness-of-fit measures
(i.e., R
2
for multiple regression) for each of the 2
K
possible models for K predictor variables, partitioning
the measures so that the total independent contribution
of a given predictor variable is estimated (Mac Nally
2002). For each predictor variable, explanatory power is
ultimately segregated into independent effects, I, and
effects that cannot be unambiguously associated with
that single variable but are due to joint effects with other
variables, J. The utility of HP has been demonstrated in
habitat studies for a variety of fauna, including
butterflies (Luoto et al. 2006), birds (Bennett et al.
2004, Heikkinen et al. 2004, Betts et al. 2006), fish (Pont
et al. 2005), amphipods (Walsh et al. 2004), diatoms
(Newall and Walsh 2005), and rocky intertidal organ-
isms (Lindegarth and Gamfeldt 2005, Arenas et al.
2006), but HP has been used less often for plant habitat
studies (Ku
¨
hn et al. 2004, Yao et al. 2006). HP was
previously used for an analysis of Great Lakes fauna
and flora by Brazner et al. (2007), which included five of
the wetland plant taxa considered here.
The objective of this research was to develop
vegetation-based indicators of coastal wetland condition
utilizing only plants that respond (either positively or
negatively) to anthropogenic stressors, distinct from
naturally occurring environmental factors that influence
CAROL A. JOHNSTON ET AL.984
Ecological Applications
Vol. 18, No. 4

plant distribution and abundance. Specific goals of the
research were to (1) identify Great Lakes coastal
wetland p lant taxa with a signific ant response to
anthropogenic stress independent of other influences
and (2) relate those responsive plant taxa to measures of
anthropogenic stress to form ulate vegetation-based
environmen tal ind icators forGreatLakescoastal
wetlands based on those relationships.
M
ETHODS
Site selection and vegetation sampling
Ninety wetlands along the U.S. Great Lakes coasts
were selected using an objective, s tratified random
statistical design representing the entire range of existing
anthropogenic stress (Danz et al. 2007). Wetlands were
divided among two ecological provinces (Keys et al.
1995), the eastern broadleaf forest (EBF) ecoprovince (n
¼35 wetlands) and the more northerly Laurentian mixed
forest (LMF) ecoprovince (n ¼ 55 wetlands). Wetlands
were also selected to represent three hydrogeomorphic
types based on wetland vulnerability to hydrologic
forces (Keough et al. 1999): open-coast wetlands, river-
influenced wetlands, and protected wetlands. Open-
coast wetlands (n ¼ 27) contained emergent plants
growing out of littoral lakebed that was relatively
exposed to wave action. River-influenced wetlands (n ¼
35) bordered a river at its confluence with one of the
Great Lakes, receiving hydrologic inputs from upstream
as well as from the lake. Protected wetlands (n ¼ 28)
were hydrologically connected with the Great Lakes, but
were protected from the full force of wave action by
their location behind a sand spit, barrier beach,
bayhead, or dike. Specific study wetlands and their
assigned geomorphic types are listed in Johnston et al.
(2007).
Plant sampling was conducted in 1 3 1 m quadrats
distributed along randomly placed transects within areas
of emergent herbaceous wetland vegetation in the study
wetlands selected. Transects were established with a
geographic information system (GI S) prior to field
campaigns, using a program called Sample (Quantitative
Decisions, Rosemont, Pennsylvania, USA) to randomize
transect placement (Johnston et al. 2007). Transects
were placed in areas mapped by national and state
wetland inventories as emergent wetland vegetation.
Each transect intersected a randomly selected point
generated by the Sample program, and was oriented
along the water depth gradient extending from the
deepwater edge of wetland emergents to the upland–
wetland boundary, or to a shrub-dominated wetland
zone, if present. Transect length and target number of
sample quadrats were determined in proportion to the
size of the wetland to be sampled (20 quadrats/60 ha,
minimum transect length ¼ 40 m, minimum quadrats/
wetland ¼ 8). Transect endpoint coordinates were
uploaded into a handheld global positioning system
(GPS) prior to field campaigns, but the location of
individual quadrats was determined and recorded in the
field.
To establish quadrat locations in the field, each
transect was first divided into 20-m segments, and then
a13 1 m quadrat was randomly located within each
segment using a random numbers table. Within each
quadrat all vascular plant species were identified to the
lowest taxonomic division possible by trained botanists
who were tested to ensure consistency of visual
observations. Although transects were terminated at
the edges of zones dominated by tall shrubs or
sub mergent aquatic vegetation (SAV), short-stature
bog shrubs and isolated shrub or SAV patches occurring
within a predominantly emergent plant matrix were
included in the sampling. Large, identifiable nonvascular
species such as Chara vulgaris and Sphagnum spp. were
also given cover estimates. Plants were identified using
published t axonomic manuals (e.g., Fassett 1957,
Chadde 1998), but the Interagency Taxonomic Infor-
mation System was used as the ultimate taxonomic
authority (taxonomic information available online).
9
Percent cover was estimated visually for each taxon
according to modified Braun-Blanquet cover class
ranges (American Society for Testing and Materials
1997): ,1%,1% to , 5%,5% to ,25%,25% to ,50%,
50% to ,75%, and 75% to 100%. Quadrat cover classes
for each taxon observed were converted to the midpoint
percent cover of each class using the algebraic midpoints
of the six cover class ranges (0.5%, 3.0%, 37.5%, 62.5%,
87.5%), and values for all quadrats at a wetland were
averaged by taxon. Vegetation sampling was conducted
from 2001 to 2003, and was restricted to the months of
July and August to ensure that most of the vegetation
could be identified and peak annual growth was
observed.
Site environmental characterization
All wetlands were categorized according to three non-
anthropogenic environmental characteristics: hydrogeo-
morphic type, ecoprovince, and the Great Lake on
which they occurred. The number of wetlands studied
on each of the five Great Lakes varied due to differences
in sho reline length and coastal wetland abundance
(Table 1). Two indices of anthropogenic stress were
also obtained for each wetland, the cumulative stress
index (CSI) and the hydrologic modification index
(HMI), of which the former is a watershed-scale stressor
index and the latter is a within-wetland stressor index.
The CSI is a generalized stress gradient for watersheds
draining to the U.S. Great Lakes coast developed by
Danz and cowor kers (2007), derived from a GIS
database of 149 variables related to five types of
anthropogenic stress: agriculture, atmospheric deposi-
tion, land cover, human population, and point source
pollution. Principal components (PCs), normalized to
9
hhttp://itis.govi
June 2008 985MULTI-TAXA WETLAND VEGETATION INDICATORS

range between zero and one, were used to integrate the
information within each of the five categories of stress
variables. The CSI is the sum of the normalized first
principal component (PC) from each of the five stressor
categories, with values ranging from 0.33 (least stress) to
4.03 (most stress) for the 762 watersheds of the U.S.
Great Lakes coast. The CSI has been demonstrated to
be strongly related to characteristics of diatom, bird,
fish, amphibian, and coastal wetland vegetation com-
munities (Reavie et al. 2006, Brazner et al. 2007, Danz et
al. 2007). The constituent PCs of the CSI were also
significantly related to coastal water chemistry (total P,
total N, dissolved inorganic N, total suspended solids,
and Cl
) measured in 98 Great Lakes coastal wetlands
(Morrice et al. 2007). We adopted the CSI as an
independent measure of watershed-scale anthropogenic
stress based on its proven utility in prior studies, and did
not attempt to evaluate or modify it.
The HMI was initially developed by Bourdaghs and
coworkers (2006) to test the FQI and its variants. The
HMI was computed by mapping linear features that
likely disrupt the natural flow and fluctuation of water in
wetlands (e.g., road beds, dikes, ditches), as interpreted
from contemporary digital orthophotoquads displayed
in an ArcView 3.3 GIS (ESRI, Redlands, California,
USA). The summed length of these features (in meters)
was then divided by the area (in hectares) of the wetland
complex to calculate the HMI. Bourdaghs and cowork-
ers (2006) tested the HMI in Great Lakes coastal
wetlands of the LMF ecoprovince, and found that FQI
significantly decreased with increasing HMI in open-
coast wetlands; they did not test the HMI in the EBF
ecoprovince. We employed the HMI in this study as a
measure of within-wetland anthropogenic stress because
of the demonstrated importance of dikes in altering
Great Lakes coastal wetland vegetation (Sherman et al.
1996, Kroll et al. 1997, Thiet 2002, Herrick and Wolf
2005).
Plant-based indices of environmental condition
We used HP to evaluate the independent influence of
geomorphology, geography (ecoprovince, Great Lake),
and anthropogenic stress effects on percent cover of
selected plant taxa. HP analyses were conducted for the
combined dataset from all 90 study wetlands (‘‘basin-
wide’’), as well as the separate datasets for each of the
five Great Lakes. Candidate plant taxa for hierarchical
partitioning were restricted to those that could be
identified to species, with the exception of the genus
Sphagnum, several free-floating aquatic genera (Wolffia,
Azolla, Riccia), and ‘invasive Typha (TYinv), which
included both Typha angustifolia and Typha 3 glauca.
Candidate plant taxa were also required to be relatively
common, occurring in at least 20% of all wetlands (i.e.,
18 of the 90 study sites) to be considered for basin-wide
indicators, and occurring in at least 30% of wetlands on
an individual Great Lake to be considered for lake-
specific indicators. A total of 119 plant taxa were
evaluated (Appendix A). With the exception of TYinv
and CHVU (Chara vulgaris), all taxa symbols follow the
Plants Database (available online).
10
We used the hier.part package (Walsh and Mac Nally
2004) in the statistical software R version 2.1.0 (R
Development Core Team 2004) for each response, using
three environmental variables and two anthropogenic
stressors in the initial analysis: (1) wetland geomorphic
type (three levels), (2) Great Lake (five levels), (3)
ecoprovince (two levels), (4) the CSI (continuous
variable), and (5) the HMI (continuous variable). We
modeled the response of taxa percent cover as following
a pseudo-binomial distribution of residuals and specified
a generalized linear model in hier.part. R
2
was used as
the measure of fit. Taxa that independently explained a
larger proportion of variance than would be expected by
chance were identified by comparing their observed
value of independent contribution to variance (I)toa
population of ‘‘I values from 1000 randomizations of
the data matrix. Significance was accepted at the upper
90% confidence level (Z score 1.28) for lake-specific
HP analyses, and at the 95% confidence limit for the
basin-wide analysis (Z score 1.65; Mac Nally 2002).
The HP results were evaluated using one-way analysis of
variance (ANOVA) to determine statistically significant
differences in mean I values associated with the variance
components across environmental variables.
TABLE 1. Summary of wetland site characteristics for each Great Lake and basin-wide.
Lake or basin
No. ecoprovince sites No. geomorphology sites
CSI per site HMI per siteLMF EBF P R C
Superior 26 0 8 14 4 1.39
a
6 0.06 14.9
a
6 3.5
Huron 8 6 2 1 11 1.86
a
6 0.21 12.9
a
6 3.3
Michigan 21 6 5 11 11 2.42
b
6 0.12 9.1
a
6 2.3
Erie 0 10 8 2 0 3.22
c
6 0.09 47.1
b
6 13.2
Ontario 0 13 4 7 2 2.84
bc
6 0.04 6.4
a
6 2.8
Basin 55 35 27 35 28 2.19 6 0.08 15.2 6 2.3
Notes: Abbreviations are: LMF, Laurentian mixed forest; EBF, eastern broadleaf forest; P, protected wetlands; R, river-
influenced wetlands; C, open-coast wetlands; CSI, cumulative stress index; HMI, hydrologic modification index. Lake means within
a column followed by the same lowercase letter are not significantly different (P . 0.05).
Values are the mean of the index 6 SE.
10
hhttp://plants.usda.govi
CAROL A. JOHNSTON ET AL.986
Ecological Applications
Vol. 18, No. 4

To develop the final multi-taxa indices, we generated
predictive models of anthropogenic stress (CSI, HMI)
using stepwise regression in SAS 9.1.3 (SAS Institute
2004), where the predictive variables entered in the
modeling process were all taxa that had significant
responses to the CSI or the HMI in HP analysis.
Predictive models were developed for the basin-wide
dataset (i.e., all 90 wetlands), as well as for individual
Great Lakes.
R
ESULTS
Measures of anthropogenic stress
The CSI values for the wetlands studied ranged from
0.7 (least disturbed) for the Mismer Bay wetland on
northern Lake Huron to 3.6 (most disturbed) for the
Huron River wetland on the south shore of Lake Erie.
Average CSI values were significantly greater in the EBF
ecoprovince than in the LMF ecoprovince (Wilcoxon, W
¼170, P , 0.001), and also varied significantly by Great
Lake (F
4,85
¼ 34.9, P , 0.001: Table 1). The large
standard errors of the CSI averages for Lakes Huron
and Michigan reflected large disturbance gradients from
very disturbed in the south to relatively pristine in the
north portions of those lakes. There was no significant
effect of geomorphology on average CSI (F
2,87
¼0.19, P
¼ 0.83).
The HMI ranged from 0 to 111 m/ha, with the three
highest values associated with diked wetlands of western
Lake Erie. Twenty of the 90 wetlands contained no
structures (HMI ¼ 0). Mean HMI values were signifi-
cantly greater on Lake Erie than on the other four Great
Lakes (F
4,85
¼8.6, P , 0.001; Table 1), but did not vary
significantly by ecoprovince (Wilcoxon, W ¼ 850, P ¼
0.35). There was no significant effect of geomorphology
on average HMI (F
2,87
¼ 2.74, P ¼ 0.07), although the
five wetlands with HMI values .50 m/ha were all of the
‘protected’ geomorphic type.
Basin-wide analysis
For the 49 taxa that occurred in 20% or more of the
study wetlands basin-wide, the sum of I values for all
five environmental characteristics ranged from 0.422 for
TYinv to only 0.057 for the submergent Ceratophyllum
demersum (CEDE4), the latter value indicating that
CEDE4 abundance was poorly predicted by any of the
five environmental variables (Fig. 1). ‘Lake’ had the
strongest influence among the five environmental
variables (mean I ¼ 51.6%), meaning that plants tended
to be geographically aggregated within one of the five
Great Lakes. For example, Thelypteris palustris (THPA)
occurred primarily on Lake Ontario (lake I ¼ 79.3%),
and its cover was poorly predicted by any of the other
environmental variables. ‘‘Lake’ exhibited a significant
independent contribution to variance for 25 of the 49
candidate taxa in the basin-wide analysis (Fig. 1).
Geomorphology had the next greatest influence on
taxa abundance (mean I ¼ 18.0%), but its independent
effect was significant for only 10 species in the basin-
wide analysis (Fig. 1). Four species preferred open-coast
wetlands (Carex comosa [CACO8], Eleocharis erythro-
poda [ELER], Eupatorium perfoliatum [EUPE3], Juncus
nodosus [JUNO2]), four species preferred river-influ-
enced wetlands (Carex lacustris [CALA16], Comarum
palustre [COPA28], Scutellaria galericulata [SCGA],
Salix discolor [SADI]), and two species preferred
protected wetlands (Carex lasiocarpa [CALAA] and
Myrica gale [MYGA]).
There was a small independent effect of ecoprovince
when wetlands were analyzed basin-wide (mean I ¼
10.8%). The six taxa that preferred the southerly EBF
ecoprovince were more invasive (Phragmites australis
[PHAU7], TYinv) and weedy (e.g., Boehmeria cylindrica
[BOCY], Leersia oryzoides [LEOR], Urtica dioica
[URDI]) than those preferring the LMF ecoprovince
(Table 2).
The independent effect of CSI was relatively small
(mean I ¼ 12.3
%), and all but two of the 12 taxa with
significant independent effects of CSI were also signif-
icantly affected by the ‘lake’ or ‘‘ecoprovince’ variables
(Fig. 1). The independent effect of HMI was very small
(mean I ¼ 7.4%). The four species with significant
independent effects of HMI included two invasives
(Phalaris arundinacea [PHAR3] and PHAU7) and two
free-floating plants (Lemna minor [LEMI3] and Spiro-
dela polyrrhiza [SPPO]), which were very abundant in
diked wetlands of western Lake Erie.
Life-form had a significant effect on plant responsive-
ness to watershed anthropogenic stress when species
were separated into two groups: (1) submerged and
floating aquatic plants (i.e., CEDE4, CHVU, Elodea
canadensis [ELCA7], LEMI3, Nymphaea odorata
[NYOD], SPPO, Utricularia macrorhiza [UTMA]) vs.
(2) emergents and shrubs (i.e., the other taxa listed in
Fig. 1). The average I value for the independent effect of
CSI was significantly lower (Wilcoxon, W ¼ 251, P ¼
0.003) for submerged and floating aquatic plant species
(
¯
x ¼ 0.007, SE ¼ 0.005) than it was for emergent and
shrub species (
¯
x ¼ 0.032, SE ¼ 0.002) and none of the
submerged and floating aquatic plant species exhibited a
significant independent response to CSI. The average I
value for the independent effect of HMI was not
significantly different between these two life-form
groups (Wilcoxon, W ¼ 103, P ¼ 0.21).
Lake-specific analysis
Separately analyzing the data for each of the five
Great Lakes increased the average I values for the
independent effect of CSI and HMI (Fig. 2). CSI mean I
values for Lakes Huron and Erie were about four times
greater than the CSI mean I value basin-wide (Fig. 2A).
The HMI mean I value for Lake Erie (the lake with the
greatest number of diked wetlands) was nine times
greater than the HMI mean I value basin-wide, and
HMI mean I values for Lakes Superior and Michigan
were 4.4 times greater (Fig. 2B).
June 2008 987MULTI-TAXA WETLAND VEGETATION INDICATORS

Citations
More filters
Journal ArticleDOI
TL;DR: Some indicators of restoration progress followed increasing trajectories and achieved or surpassed levels equivalent to high-quality reference sites within five years, whereas others appeared destined to either not reach equivalency or to take much longer than mitigation wetlands are typically monitored.
Abstract: Temporal trends in attributes of restored ecosystems have been described conceptually as restoration trajectories. Measures describing the maturity or ecological integrity of a restoration site are often assumed to follow monotonically increasing trajectories over time and to eventually reach an asymptote representative of a reference ecosystem. This assumption of simple, predictable restoration trajectories underpins federal and state policies in the United States that mandate wetland restoration as compensation for wetlands damaged during development. We evaluated the validity of this assumption by tracking changes in 11 indicators of floristic integrity, often used to determine legal compliance, in 29 mitigation wetlands. Each indicator was expressed as a percentile relative to the distribution of that indicator among >100 naturally occurring reference wetlands. Nonlinear regression was used to fit two alternative restoration trajectories to data from each site: an asymptotic (negative exponential) inc...

163 citations

Journal ArticleDOI
01 Nov 2010-Ecology
TL;DR: The results suggest that the anthropogenic structuring of urban areas has disrupted the spatial organization of environmental variables and inhibited the development of biotic spatial processes, resulting in urban community composition mainly influenced by local environmental variables.
Abstract: Spatially organized distribution patterns of species and communities are shaped by both autogenic processes (neutral mechanism theory) and exogenous processes (niche theory). In the latter, environmental variables that are themselves spatially organized induce spatial structure in the response variables. The relative importance of these processes has not yet been investigated in urban habitats. We compared the variance explained by purely spatial, spatially structured environmental, and purely environmental components for the community composition of spiders (Araneae), bees (Apidae), and birds (Aves) at 96 locations in three Swiss cities. Environmental variables (topography, climate, land cover, urban green management) were measured on four different radii around sampling points (,10 m, 50 m, 250 m, 1000 m), while Moran's eigenvector maps (MEMs) acted as spatial variables. All three taxonomic groups showed weak spatial structure. Spider communities reacted to very fine- scaled environmental changes of lawn and meadow management and climate. Bird community composition was determined by woody plants as well as solar radiation at all radii, the scale of the influence varying among species. Bee communities were weakly explained by isolated variables only. Our results suggest that the anthropogenic structuring of urban areas has disrupted the spatial organization of environmental variables and inhibited the development of biotic spatial processes. The near absence of spatial structure may therefore be a feature typical of urban species assemblages, resulting in urban community composition mainly influenced by local environmental variables. Urban environments represent a close-knit mosaic of habitats that are regularly disturbed. Species communities in urban areas are far from equilibrium. Our analysis also suggests that urban communities need to be considered as being in constant change to adapt to disturbances and changes imposed by human activities.

128 citations

Journal ArticleDOI

100 citations

Journal ArticleDOI
01 Nov 1940-Nature
TL;DR: Fassett as discussed by the authors presented a Manual of Aquatic Plants for the identification of aquatic plants in sterile as well as in flowering or fruiting conditions, and the species included in the Manual are those occurring in the region from Minnesota to Missouri and eastward to the Gulf of St. Lawrence and Virginia.
Abstract: NOTWITHSTANDING the relative uniformity of their environment, aquatic plants exhibit a remarkable variety of form and constitute a very distinct biological group comprising a considerable number of quite unrelated families. An extensive literature already exists relating to the special features of their morphology and ecology; but apart from monographs on certain genera, the group as a whole has not been analysed systematically with the view of simplifying identification. For various reasons the recognition of aquatic species is not always easy and the present work by Prof. Fassett is intended to make as simple as possible the identification of aquatic plants in sterile as well as in flowering or fruiting conditions. The species included in the Manual are those occurring in the region from Minnesota to Missouri and eastward to the Gulf of St. Lawrence and Virginia. Though the area is thus restricted, many plants are dealt with which are familiar to botanists outside America. A Manual of Aquatic Plants By Prof. Norman C. Fassett. Pp. vii + 382. (New York and London: McGraw-Hill Book Co., Inc., 1940.) 26s.

86 citations

Journal ArticleDOI
18 May 2010-Wetlands
TL;DR: The invasion and expansion of the non-native Phragmites australis in Great Lakes coastal wetlands is of increasing concern, but quantitative studies of the extent, rate, and causes of invasion have been lacking as mentioned in this paper.
Abstract: The invasion and expansion of the non-native Phragmites australis in Great Lakes coastal wetlands is of increasing concern, but quantitative studies of the extent, rate, and causes of invasion have been lacking. Here we revisited 307 plots in 14 wetlands along the Great Lakes coast in 2005 that had previously been sampled for vegetation in 2001–2003. During the 2–4 years between sample events, Phragmites occurred in 101 plots. Genetic analysis revealed that none of the Phragmites samples collected at the 14 wetlands belonged to the native genotype. Decreases in water depth and bare soil area were associated with the greatest increases in Phragmites cover. Phragmites invasion was greater on Lakes Michigan, Huron, and Erie than it was on Lake Ontario, and occurred predominantly on sandy substrates. Soil water concentrations of NO3-N, NH3-N, and soluble reactive P did not differ significantly between plots with and without Phragmites. Monitoring coastal wetlands where water level has dropped and controlling Phragmites at early stages of invasion are essential for maintaining healthy Great Lakes coastal wetlands of high species diversity and wildlife habitat. This becomes important as water levels in the Great Lakes have reached extreme lows and are expected to decline with future climate change.

77 citations


Cites background from "Partitioning vegetation response to..."

  • ...…Minchinton and Bertness 2003), the presence of exposed mineral soil (Tulbure et al. 2007), and alterations of the hydrologic regime by roads, dikes, and ditches ( McNabb and Batterson 1991; Bart and Hartman 2000, 2003;Herrick and Wolf 2005; Maheu-Giroux and de Blois 2007; Johnston et al. 2008)....

    [...]

  • ...2007), and alterations of the hydrologic regime by roads, dikes, and ditches ( McNabb and Batterson 1991; Bart and Hartman 2000, 2003;Herrick and Wolf 2005; Maheu-Giroux and de Blois 2007; Johnston et al. 2008)....

    [...]

References
More filters
Journal ArticleDOI
TL;DR: Road density and network structure are informative landscape ecology assays and Australia has huge road-reserve networks of native vegetation, whereas the Dutch have tunnels and overpasses perforating road barriers to enhance ecological flows.
Abstract: A huge road network with vehicles ramifies across the land, representing a surprising frontier of ecology. Species-rich roadsides are conduits for few species. Roadkills are a premier mortality source, yet except for local spots, rates rarely limit population size. Road avoidance, especially due to traffic noise, has a greater ecological impact. The still-more-important barrier effect subdivides populations, with demographic and probably genetic consequences. Road networks crossing landscapes cause local hydrologic and erosion effects, whereas stream networks and distant valleys receive major peak-flow and sediment impacts. Chemical effects mainly occur near roads. Road networks interrupt horizontal ecological flows, alter landscape spatial pattern, and therefore inhibit important interior species. Thus, road density and network structure are informative landscape ecology assays. Australia has huge road-reserve networks of native vegetation, whereas the Dutch have tunnels and overpasses perforating road barriers to enhance ecological flows. Based on road-effect zones, an estimated 15‐20% of the United States is ecologically impacted by roads.

2,949 citations


"Partitioning vegetation response to..." refers background in this paper

  • ...Key words: Carex; dikes; emergent; fen; floristic quality; Great Lakes; hydrologic modification; invasive plants; land use; lemnids; marsh; Typha....

    [...]

  • ...Roads have a variety of ecological effects on aquatic ecosystems, including disruption of the physical environment, alteration of the chemical envi- ronment, and the spread of exotic species (Wilcox 1989, Forman and Alexander 1998, Trombulak and Frissell 2000)....

    [...]

Journal ArticleDOI
TL;DR: Roads are a widespread and increasing feature of most landscapes. as mentioned in this paper reviewed the scientific liter- ature on the ecological effects of roads and found support for the general conclusion that they are associated with negative effects on biotic integrity in both terrestrial and aquatic ecosystems.
Abstract: Roads are a widespread and increasing feature of most landscapes. We reviewed the scientific liter- ature on the ecological effects of roads and found support for the general conclusion that they are associated with negative effects on biotic integrity in both terrestrial and aquatic ecosystems. Roads of all kinds have seven general effects: mortality from road construction, mortality from collision with vehicles, modification of animal behavior, alteration of the physical environment, alteration of the chemical environment, spread of exotics, and increased use of areas by humans. Road construction kills sessile and slow-moving organisms, injures organisms adjacent to a road, and alters physical conditions beneath a road. Vehicle collisions affect the demography of many species, both vertebrates and invertebrates; mitigation measures to reduce roadkill have been only partly successful. Roads alter animal behavior by causing changes in home ranges, move- ment, reproductive success, escape response, and physiological state. Roads change soil density, temperature, soil water content, light levels, dust, surface waters, patterns of runoff, and sedimentation, as well as adding heavy metals (especially lead), salts, organic molecules, ozone, and nutrients to roadside environments. Roads promote the dispersal of exotic species by altering habitats, stressing native species, and providing movement corridors. Roads also promote increased hunting, fishing, passive harassment of animals, and landscape modifications. Not all species and ecosystems are equally affected by roads, but overall the pres- ence of roads is highly correlated with changes in species composition, population sizes, and hydrologic and geomorphic processes that shape aquatic and riparian systems. More experimental research is needed to com- plement post-hoc correlative studies. Our review underscores the importance to conservation of avoiding con- struction of new roads in roadless or sparsely roaded areas and of removal or restoration of existing roads to benefit both terrestrial and aquatic biota.

2,506 citations


"Partitioning vegetation response to..." refers background in this paper

  • ...Key words: Carex; dikes; emergent; fen; floristic quality; Great Lakes; hydrologic modification; invasive plants; land use; lemnids; marsh; Typha....

    [...]

  • ...Roads have a variety of ecological effects on aquatic ecosystems, including disruption of the physical environment, alteration of the chemical envi- ronment, and the spread of exotic species (Wilcox 1989, Forman and Alexander 1998, Trombulak and Frissell 2000)....

    [...]

Journal ArticleDOI
TL;DR: Ecological indicators need to capture the complexities of the ecosystem yet remain simple enough to be easily and routinely monitored.

1,251 citations


"Partitioning vegetation response to..." refers background in this paper

  • ...Ecological indicators should capture the complexities of the ecosystem yet remain simple enough to be easily and routinely monitored (Jackson et al. 2000, Dale and Beyeler 2001)....

    [...]

01 Jan 2002
TL;DR: Results indicate that an introduction of a non-native strain of Phragmites has occurred, and the introduced type has displaced native types as well as expanded to regions previously not known to have PhragMites.
Abstract: iological invasions threaten species and ecosystems world-wide (1). An estimated 50,000 exotic species have beenintroduced to the United States, of which 5,000 are plant speciesthat have escaped and now exist in natural environments (2).Both the actual number of invaders and the impacts of thesespecies may be underestimated because of the presence ofcryptic invaders, or species that cannot be easily classified asnative or introduced (3). Over the past decade, use of PCR-basedmolecular techniques have revealed repeated occurrences ofsuch invasions in marine ecosystems (4–6), and studies havedemonstrated both genetic and physiological differences be-tween invading and native populations (6). Given that crypticinvaders typically are unrecognized or are mistaken for nativespecies, knowledge of historical trends in geographic distributionand population genetic structure in cases of suspected introduc-tions are of particular interest when trying to reconstruct theinvasionhistoryofaspecies.Insuchcases,museumorherbariumspecimens are an invaluable resource for reconstructing popu-lation history.Common reed,

942 citations


"Partitioning vegetation response to..." refers background in this paper

  • ...PHAU7 has both native and nonnative genotypes that differ in their aggressiveness (Galatowitsch et al. 1999, Saltonstall 2002)....

    [...]

Journal ArticleDOI
TL;DR: The distribution and abundance of Phragmites australis in North America has increased dramatically over the past 150 years as mentioned in this paper, and the hypothesis that a non-native strain of Phagmites is responsible for the observed spread is tested.
Abstract: Cryptic invasions are a largely unrecognized type of biological invasion that lead to underestimation of the total numbers and impacts of invaders because of the difficulty in detecting them. The distribution and abundance of Phragmites australis in North America has increased dramatically over the past 150 years. This research tests the hypothesis that a non-native strain of Phragmites is responsible for the observed spread. Two noncoding chloroplast DNA regions were sequenced for samples collected worldwide, throughout the range of Phragmites. Modern North American populations were compared with historical ones from herbarium collections. Results indicate that an introduction has occurred, and the introduced type has displaced native types as well as expanded to regions previously not known to have Phragmites. Native types apparently have disappeared from New England and, while still present, may be threatened in other parts of North America.

925 citations

Frequently Asked Questions (1)
Q1. What contributions have the authors mentioned in the paper "Partitioning vegetation response to anthropogenic stress to develop multi-taxa wetland indicators" ?

Schimel et al. this paper used hierarchical partitioning to evaluate the independent influence of geomorphology, geography, and anthropogenic stress on common wetland plants of the U.S. Great Lakes coast and developed multi-taxa models indicating wetland condition.