Page 1
J. N. Am. Benthol. Soc., 2009, 28(1):80–92
? 2009 by The North American Benthological Society
DOI: 10.1899/07-122.1
Published online: 18 November 2008
BRIDGES
This article is the last to be published under the original BRIDGES model (bridging basic and applied science). We
thank past Bridges Co-editors M. Barbour, M. Gurtz, and N. Aumen for a job well done.
BRIDGES is a recurring feature of J-NABS intended to provide a forum for the interchange of ideas and information
relevant to J-NABS readers, but beyond the usual scope of a scientific paper. Guest editors and authors may contribute
2–4 short papers (,3000 words each) around a topic that bridges from aquatic ecology to other disciplines, e.g.,
political science, economics, education, chemistry, or other biological sciences. Papers may be complementary or take
alternative viewpoints. Authors with ideas for topics should contact Associate Editors Ashley Moerke and Leska Fore.
Ashley Moerke, amoerke@lssu.edu
Leska Fore, leska@seanet.com
Co-editors
Analysis of functional traits in reconfigured channels: implications
for the bioassessment and disturbance of river restoration
Desiree D. Tullos1
Biological and Ecological Engineering Department, Oregon State University, Corvallis, Oregon 97330 USA
David L. Penrose2AND Gregory D. Jennings3
Department of Biological and Agricultural Engineering, North Carolina State University, Raleigh,
North Carolina 27695 USA
W. Gregory Cope4
Department of Environmental and Molecular Toxicology, North Carolina State University, Raleigh,
North Carolina 27695 USA
Abstract.
ecological responses to channel reconfiguration have not been rigorously assessed. We compared physical-
habitat variables, taxonomic and functional-trait diversities, taxonomic composition, and functional-trait
abundances between 24 pairs of upstream (control) and downstream reconfigured (restored) reaches in 3
catchment land uses (urban, agricultural, rural) across the North Carolina Piedmont. We asked how
environmental filters and functional species traits might provide insight to biological responses to
restoration. Taxonomic and functional-trait differences between control and restored reaches suggest that
restoration affected aquatic assemblages only in agricultural and rural catchments. Our results highlight 2
important aspects of channel reconfiguration as a restoration practice. First, responses to restoration differ
between agricultural/rural and urban catchments, possibly because of modified hydrological regimes
caused by urbanization. Second, we find evidence that channel reconfiguration disturbs food and habitat
resources in stream ecosystems. Taxa sensitive to disturbance were characteristic of control reaches, whereas
insensitive taxa were characteristic of restored reaches. Abundances of traits related to reproduction
(voltinism, development, synchronization of emergence, adult life span), mobility (occurrence in drift,
maximum crawling rate, swimming ability), and use of resources (trophic and habitat preferences) differed
significantly between control and recently restored reaches. Our results suggest that taxa in restored habitats
are environmentally selected for traits favored in disturbed environments. Our work suggests how
functional-trait approaches could benefit the practice of river restoration when used to target restoration
activities and to develop informed expectations regarding recovery following restoration activities.
Channel reconfiguration is a popular but controversial approach to river restoration, and
Key words:
sessment, river restoration.
benthic macroinvertebrate, functional traits, channel reconfiguration, disturbance, bioas-
River restoration often is undertaken to repair
degradation caused by development of water and
natural resources. It encompasses many objectives and
1E-mail addresses: tullosd@engr.orst.edu
2dave_penrose@ncsu.edu
3jennings@ncsu.edu
4greg_cope@ncsu.edu
80
Page 2
outcomes (e.g., enhance water quality, manage ripar-
ian zones, improve instream habitat, fish passage,
bank stabilization; Bernhardt et al. 2005). A common
goal of river restoration is to recreate geomorphically
stable reaches with high habitat availability and
diversity (FISRWG 1998). This approach is based on
an implicit positive correlation between habitat diver-
sity and taxonomic and functional diversity of stream
assemblages, as described by the habitat templet
theory (Southwood 1977, 1988).
Channel reconfiguration is a widely practiced
restoration activity based on natural channel design
(Skidmore et al. 2001), intended to restore channel
geomorphology (Shields et al. 2003) and habitat
complexity (FISRWG 1998) in systems simplified by
development. The objectives of channel reconfigura-
tion projects are often to increase instream habitat
quality and biological diversity, but how well projects
achieve these objectives is rarely evaluated (Palmer et
al. 2005), in part because of poorly defined objectives
and a lack of comprehensive monitoring (Bernhardt et
al. 2005). Lack of assessment leads to uncertainty
about outcomes of river restorations because it
maintains the currently inadequate understanding of
short- and long-term ecosystem consequences of
channel reconfiguration and the currently limited
knowledge of the processes and mechanisms that
control those consequences.
An effort to improve understanding of the processes
and consequences of river restoration is underway as
practitioners move toward more ecologically based
restoration designs. Analysis of functional traits offers
one approach toward this effort to better understand
ecosystem functions, processes, and health (Loreau
1998, Chapin et al. 2000, Diaz and Cabido 2001, Loreau
et al. 2001, Mouillot et al. 2006). Analysis of functional
traits has already been applied to characterize various
disturbances in stream ecosystems (Statzner et al.
1997), including wastewater treatment facilities (Char-
vet et al. 1998) and catchment urbanization (Dole ´dec et
al. 1999, Gayraud et al. 2003).
Channel configurations often restructure habitat
stability, food resources, and the thermal regime (Poff
et al. 2006), creating conditions that favor only those
species possessing functional traits suited to the
constructed environment. In this sense, restoration
activities that modify local environmental filters could
determine which species from the regional pool would
occur locally (Tonn et al. 1990, Weiher and Keddy 1995,
Diaz et al. 1999). For example, removal of riparian
vegetation and exposure of the streambed during
channel reconfiguration could influence food resources
and thermal regime (Sweeney et al. 2004), and increase
dominance of grazing, warm eurythermal taxa in
restored relative to control reaches. Creation of an
entirely new channel could influence habitat stability
and select for mobile taxa that reproduce rapidly.
Beneficial outcomes of restoration, including reduced
erosion of streambeds and banks, creation of scour
pools, and construction of habitat structures, might
increase habitat diversity and stability and generate
conditions favorable for taxa that prefer stable
substrates. These examples suggest that functional
traits and environmental filters could benefit the field
of river restoration by identifying activities and design
strategies (e.g., passive vs active approaches, building
habitats vs building habitat processes) that actively
encourage or discourage specific species or ecological
functions in the restored habitat.
Also relevant to understanding restoration conse-
quences is the concept of ecological disturbance (sensu
Townsend and Hildrew 1994), a discrete event that
leads to replacement of individuals by members of the
same or different taxa. Whether channel reconfigura-
tion is a disturbance that filters taxa suited to modified
and disturbed environments in the first years of post-
construction recovery is unclear. Furthermore, prevail-
ing environmental conditions, including variability
and stability of habitats, can influence ecological
responses to disturbances (Poff and Ward 1990).
Catchment land uses influence stream hydrology and
water quality and filter organisms on the basis of
habitat stability (Temperton et. al 2004), but whether
ecological responses to restoration differ among land
uses is currently unclear.
These concepts of restoration as filters and distur-
bances are important for effective assessment of stream
restorations. Few projects are monitored for longer
than 5 y, so short-term effects of disturbance and
recovery associated with construction of new channels
and subsequent adjustments of the river might unduly
influence evaluation of the success of the project. To
begin investigating these concepts in the context of
channel reconfiguration, we evaluated differences
between paired upstream (control) and downstream
reconfigured (restored) channels across 3 catchment
landuse types. We assessed effects of restoration based
on differences in physical habitat, taxonomic and
functional diversity, community composition, and
functional-trait abundances between control and re-
stored reaches. We asked: 1) Do taxa in newly
reconfigured reaches reflect a disturbance effect of
the restoration practice? 2) Does catchment land use
influence the presence and type of response observed?
3) Does the bioassessment approach used influence the
presence and type of response observed? We use
results from these analyses to discuss progress
2009]81BIOASSESSMENT OF CHANNEL RECONFIGURATION
Page 3
necessary for effective monitoring of stream restora-
tions.
Methods
Study sites
We first identified all stream restoration projects
under the regulation of state, federal, and regional
permitting agencies in the North Carolina Piedmont.
We filtered sites by limiting catchment areas to ?13
km2to minimize environmental and community
variation associated with increasing catchment size.
We restricted sites to those at which comprehensive
reconfiguration of the channel was performed. Resto-
ration consisted of reconstructing channel pattern,
profile, and dimension by cutting an entirely new
channel or floodplain. Designs often included replant-
ing of native vegetation and inclusion of various
habitat structures (e.g., root wads, pool-scouring cross-
vanes). Time since completion of construction ranged
from 1 to 4 y prior to sampling. We anticipated that
recovery at the study reaches might be incomplete
when we sampled (,4 y after project completion)
because time to recovery after restoration is not well
understood (Fuchs and Statzner 1990). Thus, we use
restored simply in reference to reaches where reconfig-
uration activities had occurred.
We used ArcGIS 9.1 (ESRI, Redlands, California) to
delineate catchments of each site and analyzed
dominant land use in each catchment based on the
1996 EarthSat Land Use Land Cover from BasinPro 8
(California Geographic Information Association, Ra-
leigh, North Carolina). We adopted the US Geological
Survey landuse categories developed by Anderson et
al. (1976) and classified landuse polygons into 1 of 5
relevant categories (urban and built-up, agriculture,
brush or transitional between open and forest, forest,
barren). We classified each site as urban, agricultural,
or rural based on the greatest percentage of areal
coverage at the site. We defined rural polygons as
brush or transitional between open and forest, forest,
and barren lands. We used 8 urban (71–94% urban land
use in catchment), 8 agricultural (53–99% agricultural),
and 8 rural (57–99% rural) sites in our study.
All sites are in the Piedmont ecoregion of North
Carolina, bounded by the Appalachian Mountains to
the west and the low-elevation coastal plains to the
east. Streams in the Piedmont are characterized by
moderate slopes (0.07–2.67%) and igneous to meta-
morphic geology. Condition of the riparian area varies
from mature to immature deciduous forest, herba-
ceous cover, lawn grass, and urban–suburban devel-
opment. Minimal irrigation is required in the humid,
warm-temperate climate of the North Carolina Pied-
mont. Thus, the effect of agriculture on the hydrologic
regime is less than in areas in the western US, where
irrigated agriculture accounts for
withdrawn and ;90% of water use (Gollehon and
Quinby 2000).
1/
3 of the water
Study design
At each site, we paired the restored reach with a
control reach immediately upstream of the restoration
project. This upstream–downstream balanced, blocked
(8 reach pairs/landuse type) study design assumes
that catchment conditions of control and restored
reaches are similar. Several control reaches did not fit
the definition of reference or least-disturbed condition
(Whittier et al. 2007) because of current or historical
activities in the catchment. The control reach repre-
sented either: 1) the initial degraded condition from
which the restored reach was expected to improve
(urban catchments) or 2) a less disturbed and more
desirable condition toward which restored reaches
were expected to improve (rural and agricultural
catchments). In urban control reaches, both local- and
catchment-scale processes drive channel simplification
(e.g., channelization, straightening, armoring, loss of
floodplain access, removal of riparian vegetation, loss
of instream habitat complexity, increased frequency,
timing, and magnitude of runoff). Thus, we expected
indicators of habitat quality and biodiversity to be
better at restored than at control reaches for the urban
sites. Rural and agricultural control reaches are largely
unmodified and restoration activities in restored
reaches address reach-scale simplification caused by
local impacts (e.g., cattle access, realignment). Thus,
we expected indicators of habitat quality and biodi-
versity to be similar between restored and control
reaches for the rural and agricultural sites if restoration
activities established the degree of habitat complexity
found in the control reaches.
Site assessment and sampling
We visited each pair of reaches, in random order,
once during the summer (June–August) in either 2003
or 2004. During the site visit, we assessed channel
conditions and sampled benthic macroinvertebrates.
Channel features and habitat.—We defined 60-m-long
restored and control study reaches at each site. In each
reach, we surveyed 2 riffle cross-sections and a
longitudinal profile and measured 24 variables to
characterize channel form and features. We calculated
average bankfull width, depth, and area to describe
the form of the channels. We characterized the habitat
based on the fine (sand or finer) and coarse (D90, the
grain size for which 90% of the bed material grains are
82 [Volume 28D. D. TULLOS ET AL.
Page 4
finer) fractions of the sediment matrix from pebble
counts (Harrelson et al. 1994). We defined channel
complexity within the active channel by the number of
pools, defined as 1.53 riffle depth, and the volume of
large woody debris (LWD) occurring along the
longitudinal profile (volume of LWD/60 m). We used
the Bank Erosion Hazard Index (Rosgen 2001) to
characterize bank stability. We dried and weighed
organic material trapped in the kick net (see Benthic
macroinvertebrates below) as a surrogate for coarse
particulate organic matter availability and storage. We
also estimated Rapid Bioassessment Protocol (RBP)
scores for each of the reaches with the protocol for
high-gradient streams (Lazorchak et al. 1998, Barbour
et al. 1999). This set of 11 RBP variables characterizes
habitat features structured by channel substrates and
morphology, streambed and bank characteristics, and
riparian vegetation.
Benthic macroinvertebrates.—We collected benthic
macroinvertebrates using the North Carolina Division
of Water Quality Qual 5 method (NCDENR 2006). This
protocol prescribes sampling insects from 1 riffle (kick
net, 107 cm 3 114 cm, 500-lm mesh), margin habitats
(D-frame dip net, 900-lm mesh), rock/log habitat
(fine-mesh wash), and 1 leaf pack, with a ;15-min
visual search of larger habitats. We sampled control
and restored reaches once on the same day at each site
and moved from downstream (restored) to upstream
(control). We picked insects by hand from samples,
composited samples within each reach, and preserved
them in 95% ethanol in the field. We identified insects
in the laboratory to species when possible.
We used a majority rule approach and best
professional judgment to assign primary functional
traits (Merritt and Cummins 1996, Poff et al. 2006) to
each genus based on the most abundant and common
species in the genus. We analyzed all traits listed in
Poff et al. (2006), including those identified as
phylogenetically constrained (Poff et al. 2006). We
did not assign fuzzy traits (Chevenet et al. 1994,
Dole ´dec et al. 2006) because we lacked sufficient
information on secondary taxon traits.
Statistical analyses
Diversity and habitat characteristics of control and
restored reaches.—We calculated Shannon diversity
(Shannon and Weaver 1949) based on number of
genera and on number of trait states (e.g., rare,
common, or abundant in drift) in each reach. We used
paired t-tests (Excel 2003; Microsoft, Seattle, Wash-
ington) to evaluate whether habitat variables and
Shannon genus and functional-trait diversities differed
between paired reaches within each landuse group. We
evaluated significance at a ¼ 0.05.
Taxonomic composition of control and restored reaches.—
We evaluated differences in assemblage composition
between control and restored reaches within each
landuse group with blocked Multi-Response Permu-
tation Procedures (MRBP; Mielke 1984) (PC-ORD,
version 5/b; MjM Software, Gleneden Beach, Oregon).
We log10(xþ1)-transformed taxon relative abundances
and used median alignment within blocks (control vs
restored) and squared Euclidean distance (Mielke and
Berry 2001, McCune and Grace 2002). We applied
indicator taxon analysis (PC-ORD) when MRBP
returned a significant difference between blocks.
Indicator taxon analysis (Dufre ˆne and Legendre
1997) combines abundance and frequency-of-occur-
rence information for each taxon in each group (control
or restored) into a single indicator value (IV), for which
100 is the highest possible value. We evaluated
statistical significance of each indicator value with a
Monte Carlo randomization method (1000 permuta-
tions).
Functional traits of control and restored reaches.—We
used the same functional traits as for analysis of
functional-trait diversity and calculated relative abun-
dances of all functional-trait states at each site by
multiplying the site 3 relative abundance of each genus
matrix by the presence of each functional-trait state 3
genus matrix. We used the relative abundances of each
functional-trait state (e.g., semivoltine, univoltine,
multivoltine) at a reach to examine the functional
composition of benthic communities in control and
restored reaches.
We used a hierarchical agglomerative cluster anal-
ysis (Sørenson distance, flexible b¼?0.25; PC-ORD) by
land use to cluster sites on the basis of dissimilarities in
trait relative abundances. We used the resultant
dendrograms as a visual tool to determine whether
clusters reflected control and restored reach classifica-
tions, a result that would indicate a functional-trait
response to restoration activities within landuse
groups. Last, we used paired t-tests to evaluate
whether relative abundances of individual traits
differed between control and restored reaches within
landuse groups.
Results
Diversity and habitat characteristics of control and restored
reaches
Habitat characteristics.—Many variables describing
habitat complexity and stability, biodiversity, and
functional traits did not differ between control and
restored reaches. Percent vegetation cover was signif-
2009]83BIOASSESSMENT OF CHANNEL RECONFIGURATION
Page 5
icantly higher at control than at restored reaches, but
no other habitat characteristic differed significantly
between control and restored reaches across landuse
groups (Table 1). In rural catchments, variables
associated with habitat complexity (number of
pools/60 m, % vegetation cover, and epifaunal
substrate/available cover) were significantly lower in
restored than in control reaches, but no other
significant differences in habitat characteristics were
detected between control and restored reaches (Table
1). In urban catchments, one variable associated with
area of habitat (channel flow status) was significantly
greater in restored than control reaches, but other
variables associated with habitat complexity (number
of pools/60 m, % vegetation cover) were significantly
lower in restored than in control reaches (Table 1). In
agricultural catchments, variables associated with
habitat complexity (% vegetation cover, epifaunal
substrate/available cover, RBP score) were significant-
ly higher in control than in restored reaches, whereas
habitat area (bankfull width) was significantly lower in
control than in restored reaches (Table 1).
TABLE 1.
and downstream (restored) reaches of 8 channel reconfiguration sites in rural, agricultural, and urban catchments in North Carolina
(24 sites total). p-values are from paired t-tests comparing control and restored reaches within landuse types. Bold font indicates
significant (a ¼ 0.05) differences. LWD ¼ large woody debris, BEHI ¼ Bank Erosion Hazard Index, CPOM ¼ coarse particulate
organic matter, D90 ¼ grain size for which 90% of sampled material is finer, RBP ¼ Rapid Bioassessment Protocol.
Mean values of physical-habitat and diversity (genus and functional-trait states) variables at paired upstream (control)
Variables
Rural Agricultural Urban
Control Restored
p
Control Restored
p
ControlRestored
p
Environmental
No. pools/60 m
Bankfull width (m)
Bankfull depth (m)
Bankfull area (m2)
Slope (%)
Bank:height ratio
Width:depth ratio
Volume of LWD (m3/60 m)
BEHI
CPOM (g/kick-net sample)
D90 (mm)
% sand and fines
% vegetation cover
Epifaunal substrate/available covera
Embeddedness
Variability in velocity/depth regimeb
Sediment depositionc
Channel flow statusd
Channel alteratione
Frequency of riffles (or bends)
Left bank stabilityf
Right bank stabilityf
Vegetative protectiong
Riparian vegetation width (m)
Diversity
RBP score
Shannon genus diversity
Shannon functional-trait state diversity
0.021
6.1
0.6
12.9
0.006
1.6
11.3
0.11
11.73
2.80
46.56
60
53
13
14
12
12
14
14
16
7
7
13
17
0.011
6.3
0.4
8.4
0.007
1.7
20.9
0.01
17.11
4.55
64.33
54
21
5
13
13
11
14
14
16
8
7
12
15
0.000
0.86
0.17
0.28
0.19
0.29
0.10
0.24
0.10
0.32
0.20
0.25
0.03
0.01
0.44
0.59
0.43
0.68
1.00
0.66
0.53
0.86
0.71
0.27
0.015
4.1
0.5
7.0
0.012
1.2
10.6
0.01
17.24
3.54
57.50
55
66
16
13
12
10
14
15
18
6
5
14
14
0.011
5.7
0.5
9.9
0.008
1.6
12.9
0.00
21.47
6.48
251.52
52
11
5
12
11
9
13
12
17
7
7
13
11
0.176
0.05
0.87
0.30
0.44
0.33
0.53
0.09
0.65
0.23
0.17
1.00
0.00
0.00
0.21
0.46
0.49
0.53
0.21
0.43
0.28
0.16
0.68
0.25
0.023
6.0
0.5
10.1
0.006
1.6
13.6
0.01
15.82
1.43
66.68
55
52
7
12
9
11
12
11
14
6
6
10
9
0.013
5.8
0.6
12.1
0.006
1.4
10.8
0.00
24.11
5.59
64.89
56
23
4
12
11
13
14
13
16
7
7
12
9
0.008
0.80
0.38
0.53
0.95
0.35
0.33
0.23
0.09
0.15
0.95
0.99
0.03
0.18
0.52
0.09
0.14
0.01
0.21
0.18
0.33
0.41
0.41
0.85
1341230.30
0.69
0.45
1371180.03
0.28
0.74
102 1120.25
0.01
0.09
2.34
49.11
2.28
50.00
1.98
52.56
1.74
50.67
1.36
38.33
1.77
41.44
asensu Lazorchak et al. 1998; area and variety of hard surfaces, including rocks and snags
bsensu Lazorchak et al. 1998; availability of 4 velocity–depth conditions: 1) slow–deep, 2) slow–shallow, 3) fast–deep, and 4)
fast–shallow
csensu Lazorchak et al. 1998; presence of depositional features that indicate sediment accumulation in an unstable channel
dsensu Lazorchak et al. 1998; the amount of useable substrate as defined by the degree to which the channel is filled with water
esensu Lazorchak et al. 1998; presence of large-scale anthropogenic changes (e.g., riprap, channelization) that simplify and
reduce availability of habitat
fsensu Lazorchak et al. 1998; presence of or potential for bank failure, as evidenced by crumbling, unvegetated banks, exposed
tree roots, and exposed soil
gsensu Lazorchak et al. 1998; extent to which bank is covered by vegetation
84[Volume 28 D. D. TULLOS ET AL.
Page 6
Taxonomic diversity.—Shannon genus diversity was
significantly higher at restored than at control reaches
in urban catchments (Table 1). Shannon genus diver-
sity did not differ significantly between control and
restored reaches in rural or agricultural catchments.
Mean values of taxonomic diversity were lower at the
restored sites than at the upstream sites in the rural
(upstream ¼ 2.34, restored ¼ 2.28) and agricultural
(upstream ¼ 1.98, restored ¼ 1.74) settings, but these
differences were not statistically significant. Shannon
functional-trait diversity did not differ significantly
between control and restored reaches in urban, rural,
or agricultural catchments.
Taxonomic composition of control and restored reaches
Taxonomic composition of assemblages differed
between control and restored reaches in rural and
agricultural catchments, but not in urban catchments
(Table 2). Effect size, indicated by the A-statistic, was
higher in agricultural than in rural catchments,
indicating a greater difference in taxonomic composi-
tion between control and restored communities in the
agricultural catchment than in the rural catchments. In
rural catchments, relative abundances of Chironomi-
dae and Baetidae were greater in restored than in
control reaches, whereas relative abundance of Hydro-
psychidae was lower in restored than in control
reaches (Fig. 1A). In agricultural catchments, relative
abundance of Chironomidae did not differ between
control and restored reaches, but relative abundances
of Baetidae and Hydropsychidae were higher in
restored than in control reaches (Fig. 1B). These taxa
appear to have replaced Philopotamidae and Hepta-
geniidae at restored reaches in agricultural catchments.
The relative abundances of opportunistic taxa, such as
TABLE 2.
dures analysis of taxonomic composition differences between
paired upstream (control) and downstream (restored) reach-
es of 8 channel reconfiguration sites in rural, agricultural,
and urban catchments in North Carolina (24 sites total). The
A-statistic describes the strength of taxonomic composition
differences between control and restored sites. Bold font
indicates significant (a ¼ 0.05) differences.
Results from multi-response permutation proce-
ResultRuralAgriculturalUrban
A-statistic
p
0.070
0.050
0.090
0.022
0.049
0.155
FIG. 1.
reaches of 8 channel reconfiguration sites in rural (A), agricultural (B), and urban (C) catchments in North Carolina (24 sites total).
Chiron. ¼ Chironomidae, Hydropt. ¼ Hydroptilidae, Simul. ¼ Simuliidae.
Relative abundances of benthic macroinvertebrate families in paired upstream (control) and downstream (restored)
2009] 85BIOASSESSMENT OF CHANNEL RECONFIGURATION
Page 7
Chironomidae, Baetidae, Hydroptilidae, and Simulii-
dae, were high in both control and restored reaches in
urban catchments compared to in rural and agricul-
tural catchments, and illustrate the loss of taxonomic
diversity within urban catchments (Fig. 1C). In urban
catchments, relative abundances of these 4 families
differed by ,5 percentage points between control and
restored reaches, and suggest that restoration activities
had little effect on taxonomic composition at reaches in
urban catchments.
Some of these patterns in family composition also
were reflected in the indicator species analysis. For
example, Chironomus was a good indicator (IV ¼ 72, a
¼ 0.04) for restored reaches in rural catchments,
whereas Baetis was a relatively good indicator (IV ¼
57, a ¼ 0.04) for restored reaches in agricultural
catchments. In contrast, Helichus (IV ¼ 63, a ¼ 0.02)
was a reliable indicator for control reaches in rural
catchments, and Dixa was a reliable indicator of
control reaches in agricultural (IV ¼ 60, a ¼ 0.03) and
rural (IV ¼ 67, a ¼ 0.009) catchments.
Functional traits of control and restored reaches
Control and restored reaches tended to group
separately in cluster dendrograms in rural (Fig. 2A)
and agricultural (Fig. 2B) catchments, but not in urban
(Fig. 2C) catchments. This suggests that functional
traits of macroinvertebrates differed between control
and restored reaches in agricultural and rural catch-
ments, but not in urban catchments.
Relative abundances of functional traits differed
between control and restored reaches within rural and
agricultural catchments, but not within urban catch-
ments (Table 3). In rural catchments, relative abun-
dances of fast seasonal development, very short adult
life spans, abundant occurrence in drift, very low
maximum crawling rate, medium size at maturity, and
collector-gatherer trophic habit were significantly
FIG. 2.
upstream (control) and downstream (restored) reaches of 8 channel reconfiguration sites in rural (A), agricultural (B), and urban (C)
catchments in North Carolina (24 sites total).
Dendrograms (Sørenson distance) for functional-trait relative abundances of macroinvertebrate assemblages in paired
86[Volume 28 D. D. TULLOS ET AL.
Page 8
greater in restored than in control reaches, whereas
relative abundances of slow seasonal development,
rare occurrence in drift, low crawling rate, small size at
maturity, sprawling habit, and shredder trophic habit
were significantly greater in control than in restored
reaches. In agricultural catchments, relative abundanc-
es of multivoltinism, poorly synchronized emergence,
abundant occurrence in drift, strong swimming ability,
swimming habit, and collector-gatherer trophic habit
were significantly greater in restored than in control
reaches, whereas univoltinism, well-synchronized
emergence, rare occurrence in drift, weak swimming
ability, clinging habit, and herbivore trophic habit were
significantly greater in control than in restored reaches
(Table 3). These differences in trait abundances in the
agricultural and rural catchments generally suggest
that an effect of restoration is to favor organisms with
rapid population turnover, a tendency to drift, and
habitat preferences that require greater swimming
than crawling and clinging abilities. Collector-gather-
ers were abundant in restored reaches in both rural
and agricultural catchments and appear to have
replaced shredders in agricultural catchments and
herbivores in rural catchments.
In rural and agricultural catchments, differences in
relative abundances of functional traits between
control and restored reaches were generally consistent
with the results of the indicator taxon analysis. In rural
catchments, Helichus and Dixa were reliable indicators
of control reaches. These taxa, characterized by slower
development, longer adult life spans, and rarer drift
occurrence, generally are not found in heavily dis-
turbed and impaired habitats. In contrast, Chironomus
was a reliable indicator of restored reaches. This taxon
is a common colonizer of recently disturbed habitats,
and its presence suggests that restored reaches favor
taxa associated with a disturbed environment. In
agricultural catchments, Dixa was a good indicator of
control reaches, whereas Baetis was a characteristic
taxon in restored reaches. Baetis has a short adult life
span, multivoltine reproduction, and abundant occur-
rence in drift, and tends to be an early recolonizer
following disturbance (Vieira et al. 2006).
TABLE 3.
reconfiguration sites in rural and agricultural catchments in North Carolina (16 sites total). Only significant (a ¼ 0.05) differences
between the upstream and restored sites are shown. Trait assignments are based on Poff et al. (2006). D indicates direction of change
in relative abundance at restored reaches relative to control reaches. þ indicates higher relative abundance at restored reaches, ?
indicates lower relative abundance at restored reaches.
Functional-trait relative abundances at paired upstream (control) and downstream (restored) reaches of 8 channel
Functional traitTrait state
Rural Agricultural
ControlRestored
p
D
Control Restored
p
D
Life history
VoltinismUnivoltine
Multivoltine
Fast seasonal
Slow seasonal
Poor
Well
Very short
0.64
0.26
0.48
0.43
0.02
0.01
?
þ
Development0.40
0.58
0.62
0.36
0.03
0.03
þ
?
Synchronization of emergence 0.29
0.71
0.40
0.59
0.00
0.00
þ
?
Adult life span
Mobility
Occurrence in drift
0.430.650.01
þ
?
þ
þ
?
Rare0.29
0.23
0.26
0.53
0.19
0.47
0.46
0.371
0.03
0.02
0.04
0.05
0.36
0.28
0.21
0.42
0.05
0.03
?
þ
Abundant
Very low
Low
Strong
Weak
Maximum crawling rate
Swimming ability0.17
0.34
0.30
0.21
0.00
0.01
þ
?
Morphology
Size at maturitySmall
Medium
0.51
0.34
0.32
0.57
0.04
0.02
?
þ
?
Ecology
HabitSprawl
Cling
Swim
0.080.040.01
0.58
0.18
0.33
0.44
0.30
0.45
0.02
0.00
0.05
?
þ
þ
?
Trophic Collector-gatherer
Shredder
Herbivore
0.31
0.07
0.48
0.02
0.05
0.00
þ
?
0.190.080.05
2009]87BIOASSESSMENT OF CHANNEL RECONFIGURATION
Page 9
Discussion
Channel reconfiguration as a disturbance filter
We used functional traits to interpret how restoration
modified local filters on community assembly during
channel recovery. In general, both indicator species
analysis and functional-trait differences indicated that
channel reconfiguration in rural and agricultural
catchments produced conditions favorable for organ-
isms characterized by traits, such as multivoltinism,
short adult life span, and rapid reproduction, that
confer resistance or resilience to disturbance (Townsend
et al. 1997b, Diaz et al. 2008). These results suggest that
channel reconfiguration acts as a filter against organ-
isms that require stable habitats in which to complete
their life cycles in systems (nonurban) where distur-
bance regimes have not already been modified greatly.
Poff et al. (2006) showed that some functional-trait
states are correlated and might be constrained such
that the state of one trait influences the state of another.
Such traits are poor indicators of environmental
change, emphasizing the value of phylogenetically
unconstrained (labile) traits for understanding mech-
anisms behind community responses to a changing
environment. Subsequent work showed that phyloge-
netically unconstrained traits are more reliable than
constrained traits for evaluating the effects of environ-
mental filters (Cavender-Bares et al. 2004). Several of
the functional traits that differed between control and
restored sites in our study were identified by Poff et al.
(2006) as phylogenetically unconstrained (e.g., trophic
habit, habitat preference, occurrence in drift, maxi-
mum crawling rate, and voltinism). Thus, we are
confident that the differences we detected in functional
traits reflect differences in environmental filters be-
tween control and restored reaches.
Broad environmental change in aquatic habitats can
impose 3 primary selective filters: thermal regime,
food resources, and habitat stability (Poff et al. 2006).
We do not know whether channel reconfiguration
changed thermal regimes at our sites because we did
not measure temperature. Further, we observed no
differences in functional traits related to thermal
preferences between control and restored reaches in
any landuse group. The absence of such differences in
rural and agricultural catchments could be interpreted
as a benefit of the channel reconfiguration (i.e.,
restoration of a natural thermal regime in reconfigured
channels), but we also observed no differences in
urban catchments where divergence from control
would be evidence of a benefit. Thus, any conclusions
regarding the influence of channel reconfiguration on
the thermal regime are not strongly supported.
We can interpret differences in functional traits
related to food resources and habitat stability filters
with more confidence. Channel reconfiguration led to
differences in taxonomic composition and functional-
trait abundances between control and restored reaches
in rural and agricultural catchments. Individuals in
restored reaches were highly mobile, able to reproduce
and grow rapidly, and opportunistic in habitat and
food preferences. These characteristics are favorable in
disturbed environments. Thus, channel reconfiguration
appears to be a disturbance that imposes food-resource
and habitat-stability filters in stream ecosystems.
Other lines of evidence in our study suggest that
channel reconfiguration imposed a disturbance on
aquatic communities following construction. Tolerant
taxa appear to replace intolerant taxa in restored
reaches in rural and agricultural catchments (Fig. 1A,
B). Indicator species analysis suggests a disturbance
signal associated with channel reconfiguration. Sensi-
tive taxa (Dixa and Helichus) were characteristic of
control reaches, whereas tolerant taxa (Chironomus and
Baetis; Brigham et al. 1982, Townsend and Hildrew
1994, Merritt and Cummins 1996, Statzner et al. 1997,
Lamouroux et al. 2004) were characteristic of restored
reaches.
Landscape context in evaluating restoration impacts
Catchment land use strongly influenced the channel
reconfiguration signal in our study. In rural and
agricultural catchments, causes for channel reconfigu-
ration tended to be local (e.g., bank hardening,
damage caused by livestock grazing) rather than
catchment-wide. The signal in restored reaches indi-
cated that channel reconfiguration was a disturbance.
Thus, the effects of channel reconfiguration in rela-
tively undisturbed catchments might be positive in the
long term, but they appear to be detrimental over the
short term.
Ecosystem responses to disturbance often are
determined by the variability and severity of estab-
lished disturbance regimes (Poff and Ward 1990).
Biological assemblages in relatively undisturbed envi-
ronments should respond to anthropogenic distur-
bances (Holling 1973, Connell and Sousa 1983, Resh et
al. 1988). In contrast, taxa able to persist in modified
and highly variable environments respond more
modestly to disturbances than do taxa in natural and
more stable environments (Levins 1968, Holling 1973).
The dominant taxa and functional traits at our urban
sites, both upstream and restored, reflected existing
pressure from the urban watershed; i.e., taxa were
behaviorally and physiologically flexible with oppor-
tunistic life-history traits (Pianka 1970, Poff and Ward
1990). Thus, the biological response to channel
88[Volume 28D. D. TULLOS ET AL.
Page 10
reconfiguration in urban reaches might have been
minimal because the biota already had been filtered by
the effects of urbanization (Paul and Meyer 2001,
Konrad and Booth 2005). Thus, the limited improve-
ment in habitat and biological diversity following
restoration of urban reaches probably is related to
catchment-wide stressors that dominate local-scale
filtering processes. This result has implications for
the success of restoration activities in urban streams
where modified hydrology and water quality at the
catchment scale might reduce the effectiveness of
instream restoration (Walsh et al. 2005).
Community context in evaluating restoration impacts
Biotic and abiotic responses to channel reconfigura-
tion often were inconsistent with each other. For
example, in rural catchments, habitat conditions and
biodiversity were similar between control and restored
reaches and indicated improved conditions in restored
reaches. However, taxonomic and functional-trait
composition differed between control and restored
reaches in ways that suggested differences in habitat
and food resource filters (i.e., a restoration disturbance
signal). We see 2 possible explanations for the apparent
decoupling of biotic and abiotic responses. 1) Physical-
habitat assessments and biodiversity measures might
be insensitive indicators of processes driving habitat
stability and ecosystem functioning when used as a
static snapshot of general condition. 2) Physical-habitat
assessments and biodiversity measures might respond
faster than taxonomic and functional-trait composition
to restoration. Thus, assemblages might still have been
recovering from the disturbance associated with
project construction, and recolonization by taxa
requiring more stable habitat and food resources
might not have been complete. It would be useful to
know how long disturbance-related taxa and func-
tional traits persist following construction and whether
increased habitat diversity and stability of restored
channels will increase the relative abundances of
habitat specialists (Poff and Allan 1995) and traits
unrelated to disturbance (Townsend et al. 1997a).
However, continued long-term monitoring is needed
to address these questions.
Our study emphasizes the value of using several
lines of evidence when evaluating the effectiveness of
channel reconfiguration. Compositional and functional
differences suggest that channel reconfiguration is a
disturbance filter, at least for the years immediately
following construction. This result is important be-
cause an initial disturbance signal could influence the
ability of evaluators to determine the ecological
success of a restoration. Moreover, this disturbance
signal might provide justification for restoration
designers to consider more passive strategies for
restoration in areas where abandoning an existing
channel is not necessary. However, a disturbance
signal was not evident in standard habitat assessment
and biodiversity measures. These summary statistics
are valuable in reach-wide assessments of general
ecosystem condition, but they might be inadequate for
evaluation of responses to restoration (Tullos et al.
2006). Moreover, representing community diversity by
summary measures is not always appropriate (Ludwig
and Reynolds 1988). In contrast, functional analysis of
invertebrate traits can be used to distinguish between
reference and impacted sites (Charvet et al. 1998,
Dole ´dec et al. 1999, Statzner et al. 2001) and offers
mechanistic explanations of effects based on filtering
of species attributes by environmental change (Ri-
chards et al. 1997). Furthermore, functional-trait
analysis may be applied to monitoring restoration
activities to identify ways to modify environmental
filters driving degradation at local and catchment
scales. Assessments based on invertebrates provide
specific and causal knowledge (Lake et al. 2007), and
monitoring the effectiveness of restoration projects is a
learning opportunity (Muotka et al. 2002, Moerke et al.
2004). Therefore, we recommend using functional
traits in addition to other measures of physical and
biological change to gain insight into mechanisms and
processes driving restoration outcomes.
Limitations and applications
We recognize that our statistical approaches were
simplistic investigations of complex ecosystems. We
considered traits individually and disregarded the
complex analytical and theoretical work necessary to
untangle interactions and tradeoffs among traits (Roff
1992, Stearns 1992). We did not fuzzy-code our data,
and we analyzed all traits, regardless of their
evolutionary lability. Moreover, no prerestoration data
were available for any of our 24 sites. Nevertheless,
our results strongly suggest that channel reconfigura-
tion imposes a disturbance on aquatic communities
during the years immediately following construction,
that catchment land use filters biological responses to
restoration activities, and that substantial value exists
in using multiple approaches to monitor the effective-
ness of restoration activities.
Our results support application of functional-trait
analysis in planning, implementing, and evaluating
restoration activities. We emphasize the importance of
thorough and mechanistic evaluations of pre- and
post-project condition of benthic macroinvertebrate
assemblages. Functional-trait analyses offer much-
2009]89BIOASSESSMENT OF CHANNEL RECONFIGURATION
Page 11
needed mechanistic explanations of the consequences
of simplification and degradation that lead to the need
for restoration. If we view channel reconfiguration as a
modifier of local environmental filters and understand
functional traits associated with those filters, we can:
1) use preproject monitoring to identify missing
ecosystem functions to define and target restoration
objectives and approaches, 2) develop restoration
designs to modify specific environmental filters to
maximize the functionality and diversity of the river,
and 3) improve postimplementation assessment with
informed expectations for recovery duration and
trajectory.
Acknowledgements
Financial support for this research was provided the
US Environmental Protection Agency, North Carolina
Department of Environment and Natural Resources,
North Carolina Water Resources Research Institute,
and US Department of Agriculture Cooperative State
Research, Education, and Extension Service. This
manuscript benefited greatly from comments from
Laura Morrison, LeRoy Poff, and an anonymous
referee.
Literature Cited
ANDERSON, J. R., E. E. HARDY, J. T. ROACH, AND R. E. WITMER.
1976. A land use and land cover classification system for
use with remote sensor data. US Geological Survey
Paper 964. US Geological Survey, Washington, DC.
BARBOUR, M. T., J. GERRITSEN, B. D. SNYDER, AND J. B. STRIBLING.
1999. Rapid bioassessment protocols for use in streams
and wadeable rivers: periphyton, benthic macroinverte-
brates and fish. 2ndedition. EPA 841-B-99–002. Office of
Water, US Environmental Protection Agency, Washing-
ton, DC.
BERNHARDT, E. S., M. A. PALMER, J. D. ALLAN, G. ALEXANDER, K.
BARNAS, S. BROOKS, J. CARR, S. CLAYTON, C. DAHM, J.
FOLLSTAD-SHAH, D. GALAT, S. GLOSS, P. GOODWIN, D. HART,
B. HASSETT, R. JENKINSON, S. KATZ, G. M. KONDOLF, P. S.
LAKE, R. LAVE, J. L. MEYER, AND T. K. O’DON. 2005.
Synthesizing U.S. river restoration. Science 308:636–637.
BRIGHAM, A. R., W. U. BRIGHAM, AND A. GNILKA (EDITORS). 1982.
Aquatic insects and oligochaetes of North and South
Carolina. Midwest Aquatic Enterprises, Mahomet, Illi-
nois.
CAVENDER-BARES, J., D. D. ACKERLY, D. A. BAUM, AND F. A.
BAZZAZ. 2004. Phylogenetic oversdispersion in Floridian
oak communities. American Naturalist 163:823–843.
CHAPIN, F. S., E. S. ZAVALETA, V. T. EVINER, R. L. NAYLOR, P. M.
VITOUSEK, H. L. REYNOLDS, D. U. HOOPER, S. LAVOREL, O. E.
SALA, S. E. HOBBIE, M. C. MACK, AND S. DIAZ. 2000.
Consequences of changing biodiversity. Nature 405:234–
242.
CHARVET, S., A. KOSMALA, AND B. STATZNER. 1998. Bioassess-
ment through biological traits of benthic macroinverte-
brates: perspectives for a general tool in stream
management. Archiv fu ¨r Hydrobiologie 142:415–432.
CHEVENET, F., S. DOLE´DEC, AND D. CHESSEL. 1994. A fuzzy
coding approach for the analysis of long-term ecological
data. Freshwater Biology 31:295–309.
CONNELL, J. H., AND W. P. SOUSA. 1983. On the evidence
needed to judge ecological stability or persistence.
American Naturalist 121:789–824.
DIAZ, A. M., M. L. S. ALONSO, AND M. S. V. GUTIERREZ. 2008.
Biological traits of stream macroinvertebrates from a
semi-arid catchment: patterns along complex environ-
mental gradients. Freshwater Biology 53:1–21.
DIAZ, S., AND M. CABIDO. 2001. Vive la diffe ´rence: plant
functional diversity matters to ecosystem processes.
Trends in Ecology and Evolution 16:646–655.
DIAZ, S., M. CABIDO, AND F. CASANOVES. 1999. Functional
implications of trait environment linkages in plant
communities. Pages 338–362 in P. Weiher and P. Keddy
(editors). Ecological assembly rules: perspectives, ad-
vances, and retreats. Cambridge University Press, Cam-
bridge, UK.
DOLE´DEC, S., N. PHILLIPS, M. SCARSBROOK, R. H. RILEY, AND C. R.
TOWNSEND. 2006. Comparison of structural and functional
approaches to determining landuse effects of grassland
stream invertebrate communities Journal of the North
American Benthological Society 25:44–60.
DOLE´DEC, S., B. STATZNER, AND M. BOURNARD. 1999. Species
traits for future bioassessment across ecoregions: pat-
terns along a human-impacted river. Freshwater Biology
42:737–758.
DUFREˆNE, M., AND P. LEGENDRE. 1997. Species assemblages and
indicator species: the need for a flexible asymmetrical
approach. Ecological Monographs 67:345–366.
FISRWG (FEDERAL INTERAGENCY STREAM RESTORATION WORKING
GROUP). 1998. Stream corridor restoration: principles,
processes, and practices. (Available from: http://www.
nrcs.usda.gov/Technical/stream_restoration/newtofc.
htm)
FUCHS, U., AND B. STATZNER. 1990. Time scales for the recovery
potential of river communities after restoration: lessons
to be learned from smaller streams. Regulated Rivers:
Research and Management 5:77–87.
GAYRAUD, S., B. STATZNER, P. BADY, A. HAYBACHP, F. SHOLL, P.
USSEGLIO-POLATERA, AND M. BACCHI. 2003. Invertebrate
traits for the bioassessment of large European rivers—an
initial assessment of alternative metrics. Freshwater
Biology 48:2045–2064.
GOLLEHON, N.,
AND W. QUINBY. 2000. Irrigation in the
American west: area, water and economic activity. Water
Resources Development 16:187–195.
HARRELSON, C. C., C. L. RAWLINS, AND J. P. POTYONDY. 1994.
Stream channel reference sites: an illustrated guide to
field technique. General Technical Report RM-245. Rocky
Mountain Forest and Range Experiment Station, Forest
Service, US Department of Agriculture, Fort Collins,
Colorado.
HOLLING, C. S. 1973. Resilience and stability of ecological
90[Volume 28D. D. TULLOS ET AL.
Page 12
systems. Annual Review of Ecology and Systematics 4:1–
23.
KONRAD, C., AND D. B. BOOTH. 2005. Hydrologic changes in
urban streams and their ecological significance. Pages
157–177 in L. R. Brown, R. H. Gray, R. M. Hughes, and
M. Meador (editors). Effects of urbanization on stream
ecosystems. American Fisheries Society, Bethesda, Mary-
land.
LAKE, P. S., N. BOND, AND P. REICH. 2007. Linking ecological
theory with stream restoration. Freshwater Biology 52:
597–615.
LAMOUROUX, N., S. DOLE´DEC, AND S. GAYRAUD. 2004. Biological
traits of stream macroinvertebrate communities: effects
of microhabitat, reach, and basin filters. Journal of the
North American Benthological Society 23:449–466.
LAZORCHAK, J. M., D. J. KLEMM, AND D. V. PECK (EDITORS). 1998.
Environmental Monitoring and Assessment Program—
Surface Waters: field operations and methods for
measuring the ecological condition of wadeable streams.
EPA/620/R-94/004F. US Environmental Protection
Agency, Washington, DC.
LEVINS, R. 1968. Evolution in changing environments.
Princeton University Press, Princeton, New Jersey.
LOREAU, M. 1998. Biodiversity and ecosystem functioning: a
mechanistic model. Proceedings of the National Acade-
my of Sciences of the United States of America 95:5632–
5636.
LOREAU, M., S. NAEEM, P. INCHAUSTI, J. BENGTSSON, J. P. GRIME, A.
HECTOR, D. U. HOOPER, M. A. HUSTON, D. RAFFAELLI, B.
SCHMID, D. TILMAN, AND D. A. WARDLE. 2001. Biodiversity
and ecosystem functioning: current knowledge and
future challenges. Science 294:804–808.
LUDWIG, J. A., AND J. F. REYNOLDS. 1988. Statistical ecology: a
primer on methods and computing. John Wiley and
Sons, New York.
MCCUNE, B.,
AND J. GRACE. 2002. Analysis of ecological
communities. MJM Press, Gleneden Beach, Oregon.
MERRITT, R. W., AND K. W. CUMMINS (EDITORS). 1996. Aquatic
insects of North America. 3rdedition. Kendall/Hunt,
Dubuque, Iowa.
MIELKE, P. W. 1984. Meteorological applications of permuta-
tion techniques based on distance functions. Pages 813–
830 in P. R. Krishnaiah and P. K. Sen (editors). Handbook
of statistics. Volume 4. North-Holland Publishers, Am-
sterdam, The Netherlands.
MIELKE, P. W., AND K. J. BERRY. 2001. Permutation methods: a
distance function approach. Springer-Verlag, New York.
MOERKE, A., K. GERARD, J. LATIMORE, R. HELLENTHAL, AND G.
LAMBERTI. 2004. Restoration of an Indiana, USA, stream:
bridging the gap between basic and applied lotic
ecology. Journal of the North American Benthological
Society 23:647–660.
MOUILLOT, D., S. SPATHARIS, S. REIZOPOULOU, T. LAUGIER, L.
SABETTA, A. BASSET, AND T. DO CHI. 2006. Alternatives to
taxonomic-based approaches to assess changes in tran-
sitional water communities. Aquatic Conservation: Ma-
rine and Freshwater Ecosystems 16:469–482.
MUOTKA, T., R. PAAVOLA, A. HAAPALA, M. NOVIKMEC, AND P.
LAASONEN. 2002. Long-term recovery of stream habitat
structure and benthic invertebrate communities from in-
stream restoration. Biological Conservation 105:243–253.
NCDENR (NORTH CAROLINA DEPARTMENT OF ENVIRONMENT AND
NATURAL RESOURCES). 2006. Standard operating proce-
dures for benthic macroinvertebrates. Biological Assess-
ment Unit, Division of Water Quality, North Carolina
Department of Environment and Natural Resources,
Raleigh, North Carolina. (Available from: http://h2o.
enr.state.nc.us/esb/BAUwww/benthossop.pdf)
PALMER, M., E. BERNHARDT, J. ALLAN, P. LAKE, G. ALEXANDER, S.
BROOKS, J. CARR, S. CLAYTON, C. DAHM, J. FOLLSTAD SHAH, D.
GALAT, S. GLOSS, P. GOODWIN, D. HART, B. HASSETT, R.
JENKINSON, M. KONDOLF, R. LAVE, J. MEYER, T. O’DONNELL,
L. PAGANO, P. SRIVASTAVA, AND E. SUDDUTH. 2005. Standards
for ecologically successful river restoration. Journal of
Applied Ecology 42:208–217.
PAUL, M. J., AND J. L. MEYER. 2001. Streams in the urban
landscape. Annual Review of Ecology and Systematics
32:333–365.
PIANKA, E. R. 1970. On r- and K-selection. American
Naturalist 104:592–597.
POFF, N. L. 1997. Landscape filters and species traits: towards
mechanistic understanding and prediction in stream
ecology. Journal of the North American Benthological
Society 16:391–409.
POFF, N. L., AND J. D. ALLAN. 1995. Functional organization of
stream fish assemblages in relation to hydrologic
variability. Ecology 76:606–627.
POFF, N. L., J. D. OLDEN, N. K. M. VIEIRA, D. S. FINN, M. P.
SIMMONS, AND B. C. KONDRATIEFF. 2006. Functional trait
niches of North American lotic insects: trait-based
ecological applications in light of phylogenetic relation-
ships. Journal of the North American Benthological
Society 25:730–755.
POFF, N. L., AND J. V. WARD. 1990. The physical habitat
template of lotic systems: recovery in the context of
historical pattern of spatio-temporal heterogeneity. En-
vironmental Management 14:629–646.
RESH, V. H., A. V. BROWN, A. P. COVICH, M. E. GURTZ, H. W. LI,
G. W. MINSHALL, S. R. REICE, A. L. SHELDON, J. B. WALLACE,
AND R. WISSMAR. 1988. The role of disturbance in stream
ecology. Journal of the North American Benthological
Society 7:433–455.
RICHARDS, C., R. HARO, L. JOHNSON,
Catchment and reach-scale properties as as indicators of
macroinvertebrate species traits. Freshwater Biology 37:
219–230.
ROFF, D. A. 1992. The evolution of life histories. Theory and
analysis. Chapman and Hall, New York.
ROSGEN, D. L. 2001. A practical method of computing
streambank erosion rate. Pages 9–15 in Proceedings of
the 7thFederal Interagency Sedimentation Conference.
Volume 2. March 25–29, 2001. US Inter-agency Commit-
tee on Water Resources, Subcommittee on Sedimenta-
tion, Reno, Nevada.
SHANNON, C., AND W. WEAVER. 1949. The mathematical theory
of communication. University of Illinois Press, Urbana,
Illinois.
SHIELDS, F. D., R. R. COPELAND, P. C. KLINGEMAN, M. W. DOYLE,
AND G. HOST. 1997.
2009] 91BIOASSESSMENT OF CHANNEL RECONFIGURATION
Page 13
AND A. SIMON. 2003. Design for stream restoration.
Journal of Hydraulic Engineering 129:575–584.
SKIDMORE, P. B., F. D. SHIELDS, M. W. DOYLE, AND D. E. MILLER.
2001. Categorization of approaches to natural channel
design. Proceedings of the ASCE Wetlands Engineering
and River Restoration Conference, Reno, Nevada.
(Available from: https://fs.ogm.utah.gov/pub/
MINES/AMR_Related/NAAMLP/StrmRest/Skidmore.
pdf)
SOUTHWOOD, T. R. E. 1977. Habitat, the templet for ecological
strategies? Journal of Animal Ecology 46:337–365.
SOUTHWOOD, T. R. E. 1988. Tactics, strategies, and templates.
Oikos 52:3–18.
STATZNER, B., A. G. HILDREW, AND V. H. RESH. 2001. Species
traits and environmental constraints: entomological
research and the history of ecological theory. Annual
Review of Entomology 46:291–316.
STATZNER, B., K. HOPPENHAUS, M. ARENS, AND P. RICHOUX. 1997.
Reproductive traits, habitat use and templet theory: a
synthesis of world-wide data on aquatic insects. Fresh-
water Biology 38:109–135.
STEARNS, S. C. 1992. The evolution of life histories. Oxford
University Press, Oxford, UK.
SWEENEY, B., T. L. BOTT, J. K. JACKSON, L. A. KAPLAN, J. D.
NEWBOLD, L. J. STANDLEY, W. C. HESSION, AND R. J. HORWITZ.
2004. Riparian deforestation, stream narrowing, and loss
of stream ecosystem services. Proceedings of the Na-
tional Academy of Sciences of the United States of
America 101:14132–14137.
TEMPERTON, V. K., R. J. HOBBS, T. NUTTLE,
(EDITORS). 2004. Assembly rules and restoration ecology:
bridging the gap between theory and practice. Island
Press, Washington, DC.
TONN, W. M., J. J. MAGNUSON, M. RASK, AND J. TOIVONEN. 1990.
Intercontinental comparison of small-lake fish assem-
blages: the balance between local and regional processes.
American Naturalist 136:345–375.
TOWNSEND, C., AND A. HILDREW. 1994. Species traits in relation
AND S. HALLE
to a habitat template for river systems. Freshwater
Biology 31:265–275.
TOWNSEND, C. R., S. DOLE´DEC, AND M. R. SCARSBROOK. 1997a.
Species traits in relation to temporal and spatial
heterogeneity in streams: a test of the habitat templet
theory. Freshwater Biology 37:367–387.
TOWNSEND, C. R., M. R. SCARSBROOK, AND S. DOLE´DEC. 1997b.
Quantifying disturbance in streams: alternative mea-
sures of disturbance in relation to macroinvertebrate
species traits and species richness. Journal of the North
American Benthological Society 16:531–544.
TULLOS, D., D. PENROSE, AND G. JENNINGS. 2006. Development
and application of a bioindicator for benthic habitat
enhancement in the North Carolina piedmont. Ecological
Engineering 27:228–241.
VIEIRA, N. K. M., N. L. POFF, D. M. CARLISLE, S. R. MOULTON, M.
L. KOSKI, AND B. C. KONDRATIEFF. 2006. A database of lotic
invertebrate traits for North America. U.S. Geological
Survey Data Series 187. US Geological Survey, Reston,
Virginia. (Available from: http://pubs.water.usgs.gov/
ds187)
WALSH, C. J., T. D. FLETCHER, AND A. R. LADSON. 2005. Stream
restoration in urban catchments through redesigning
stormwater systems: looking to the catchment to save the
stream. Journal of the North American Benthological
Society 24:690–705.
WEIHER, E., AND P. KEDDY. 1995. Assembly rules, null models,
and trait dispersion: new questions from old patterns.
Oikos 74:159–164.
WHITTIER, T. R., J. L. STODDARD, D. P. LARSEN, AND A. T. HERLIHY.
2007. Selecting reference sites for stream biological
assessments: best professional judgment or objective
criteria. Journal of the North American Benthological
Society 26:349–360.
Received: 26 October 2007
Accepted: 4 August 2008
92[Volume 28D. D. TULLOS ET AL.
Download full-text