Forest Ecology and Management 256 (2008) 1124–1132

Contents lists available at ScienceDirect

Forest Ecology and Management

j ou r n al h o m e pa g e : w ww . e l s e v i e r . c o m / l o c a t e / f o r e c o

Influence of forest riparian vegetation on abundance and biomass of nocturnal

flying insects
Holly K. Ober a,b,*, John P. Hayes a,1
a
Department of Forest Science, Oregon State University, Corvallis, OR 97331, United States
b
Department of Fisheries and Wildlife, Oregon State University, Corvallis, OR 97331, United States

A R T I C L E I N F O
A B S T R A C T

Article history:
Received 31 December 2007
Despite widespread recognition of linkages between vegetation and insects, understanding of the
Received in revised form 5 June ecological mechanisms underlying these relationships is limited. Better comprehension of relationships
2008 Accepted 9 June 2008 linking abundance and biomass of insects to vegetation would increase accuracy of predictions of the
effects of forest management activities on insect communities. This knowledge could also be pivotal
Keywords: to understanding predator–prey dynamics linked to insect populations. We sampled nocturnal flying
Forest management insects and measured vegetation characteristics in 34 stream reaches in conifer-dominated forests of the
Insects Oregon Coast Range in the Pacific Northwest of the United States. We considered five a priori
Pacific Northwest hypotheses (resource quality, resource diversity, resource abundance, resource concentration, and
Riparian stream cover hypotheses) that could explain mechanisms underlying associations between riparian
Vegetation composition
vegetation and nocturnal flying insects, and used an information-theoretic approach to determine the
relative strength of evidence for each. The resource quality hypothesis, which predicts that
abundance and biomass of insects increases with cover of deciduous vegetation, explained substantial
variation for nearly every order of insect investigated, whereas the remaining hypotheses explained
relatively little. Abundance and biomass of insects had stronger associations with characteristics of
canopy trees than with characteristics of shrub or understory trees, suggesting that deciduous trees
are an important habitat element for nocturnal flying insects in these areas. Resource managers
planning riparian vegetation management in conifer-dominated forests should be aware that
manipulation of the cover of deciduous trees in riparian areas could have a large impact on these
insects and their vertebrate predators. By providing information on forest canopy composition,
remote sensing may offer a low-cost tool for identifying areas with high abundance and biomass of
insects during conservation planning.
© 2008 Elsevier B.V. All rights reserved.

1. Introduction
indirectly influencing food resources for predaceous and parasitic
species (Schowalter, 1985; Berryman, 1986). However, much of the
Recognition of the diverse and important ecosystem services
research examining relationships between plants and insects in
provided by insects (Kim, 1993; Miller, 1993; Losey and Vaughan,
forested ecosystems to date has focused on devising silvicultural
2006) and increased awareness of the large proportional con-
practices to cope with insect pests (Gadgil and Bain, 1999; Muzika
tribution of insects to faunal biodiversity (Ehrlich and Wilson,
and Liebold, 2000; Kelty, 2006), while relatively little attention has
1991) have fostered appreciation for the roles insects play in
been given to increasing understanding of the underlying
maintaining forest ecosystem health (Taylor and Doran, 2001;
mechanisms controlling abundance and biomass of entire insect
Langor and Spence, 2006). Human activities in forests can
communities in seemingly healthy forests. Predictions of the
drastically alter insect communities by directly influencing food
effects of riparian forest management activities on insect com-
resources for phytophagous and detritivorous species, and
munities are limited in accuracy because these relationships have
thereby
not been well studied.
Vegetation characteristics such as species richness (Haddad
et al., 2001), species diversity (Knops et al., 1999), structural
* Corresponding author. Present address: Department of Wildlife Ecology and
Conservation, North Florida Research and Education Center, University of Florida,
complexity (Araujo et al., 2006), biomass (Polis et al., 1997; Knops
Quincy, FL 32351, United States. Tel.: +1 850 875 7150; fax: +1 850 875 7188. et al., 1999; Haddad et al., 2001), and composition (Schowalter
E-mail address: Holly.Ober@ufl.edu (H.K. Ober). et al., 1986; Mattson and Scriber, 1987) influence patterns of
Present address: Department of Wildlife Ecology and Conservation, University abundance of insects. Despite ample evidence of strong linkages
1

of Florida, Gainesville, FL 32611, United States.

0378-1127/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.foreco.2008.06.010
 H.K. Ober, J.P. Hayes / Forest Ecology and Management 256 (2008) 1124–
1
between plants and insect abundance in particular circumstances,
insects, allowing for greater abundance and biomass of insects
unified principles underlying these relationships across ecosys-
(Lawton, 1983; Haddad et al., 2001).
tems have not emerged (Lewinsohn et al., 2005). Analyses that
The resource abundance hypothesis predicts that abundance and
enable evaluation of the relative strength of associations between
biomass of insects within a stream reach will increase as the
vegetation characteristics and abundance of insects in a variety
cover of riparian vegetation bordering that stream reach
of settings would improve understanding of the relative
increases. This hypothesis is founded on observations that plants
importance of different mechanistic hypotheses, helping settle
can determine the carrying capacity for phytophagous insects
the debate over which vegetative factors are most influential
through influences on food availability (Price, 1992) and on the
(Lewinsohn et al., 2005).
number of eggs female insects are able to oviposit (Dempster,
The primary objectives of our research were to determine (1)
1983). As a consequence, increased vegetative cover may allow
which mechanistic hypotheses best explain associations between
for increased abundance and biomass of insects (Polis et al.,
forest riparian vegetation characteristics and abundance and
1997; Knops et al., 1999; Haddad et al., 2001).
biomass of common orders of nocturnal flying insects, (2)
The resource quality hypothesis predicts that abundance and
whether the same hypotheses consistently explain the most
biomass of insects within a stream reach will increase as the
variation among different orders of insects, (3) whether results
cover of deciduous foliage bordering that stream reach increases.
differ when abundance or biomass are considered as response
Species-specific factors such as foliar nutrient content and
variables, and
secondary defense compound concentrations influence the palat-
(4) which vegetation stratum (shrubs, understory trees, or canopy
ability and quality of food that vegetative species provide to
trees) is responsible for driving these associations. A more
phytophagous and detritivorous insects (Schowalter et al., 1986;
thorough ecological understanding of these relationships should
Ohgushi, 1992). Deciduous foliage generally has higher concen-
facilitate more accurate predictions regarding responses of
trations of nitrogen (a limiting nutrient for many herbivorous
insects to forest riparian vegetation manipulations, which may
insects) and lower concentrations of secondary defense com-
also be useful in predicting habitat use of vertebrates reliant on
pounds (inhibitory compounds of many herbivorous insects) than
insects for food (Ober and Hayes, 2008).
coniferous foliage, making deciduous foliage a higher quality food
resource for many insects (Mattson and Scriber, 1987; Friberg
1.1. Research approach
and Jacobsen, 1994).
The resource concentration hypothesis predicts that
Plants often exert inordinately strong bottom-up effects in
abundance and biomass of insects will peak in stream reaches
terrestrial food webs, determining many attributes of local insect
bordered by vegetation heavily dominated by coniferous or
communities (Price, 2002; Gruner, 2004). We anticipated strong
heavily dominated by deciduous vegetation and will decrease as
associations between local vegetation and the abundance and
vegetation composi- tion tends towards an even mixture of the
biomass of insects because of the prevalence of phytophagy among
two vegetation types. Optimal foraging theory predicts that
insects (Strong et al., 1984), and because the diet specificity and
residence time within a habitat patch will vary as a function of
limited scale of movements typical of phytophagous species
rate of energy gain within the patch relative to the rate outside
should cause insects to remain in close proximity to their host
the patch; organisms should stay in a patch as long as it is more
plants (Ehrlich and Raven, 1964; Smith and Remington, 1996).
profitable for them to stay than it is for them to travel to and
Furthermore, we anticipated strong associations between local
forage in a different patch (Charnov, 1976; Stephens and Krebs,
vegetation and abundance and biomass of parasitic and predac-
1986). Accordingly, phytophagous and detritivorous insects
eous insects due to the likely associations spanning three trophic
should spend long periods of time in patches containing high
levels, with the distribution of the vegetation that supports the
densities of plants that provide food (Feeny, 1976; Andow, 1991;
phytophagous species ultimately determining the distribution of
Long et al., 2003). Many species of insects consume more than one
the insects at higher trophic levels (Price et al., 1980; Siemann,
species of closely related plants, but few are capable of
1998; Fraser et al., 2007). We reasoned that quantity, quality, or
consuming taxonomically distant species (Holloway and Hebert,
distribution of vegetation would determine abundance and
1979; Jaenike, 1990; Robinson et al., 2000). Therefore, we
biomass of insects. We considered five a priori hypotheses that
expect areas dominated by either coniferous or by deciduous
could explain mechanisms underlying the relationships between
plant species to receive more use from insects than areas with
vegetation and insects, used these hypotheses to formulate
plants of both types mixed together.
candidate models (Table 1), and evaluated the strength of The stream cover hypothesis predicts that abundance and
evidence for each using an information-theoretic approach biomass of insects within a stream reach will increase as canopy
(Burnham and Anderson, 2002). cover directly over the stream channel increases. Solar radiation
The resource diversity hypothesis predicts that abundance and
influences both abiotic (e.g., temperature and humidity) and
biomass of insects within a stream reach will increase as the biotic (e.g., plant growth) factors. The riparian canopy intercepts
vegetative species richness bordering that stream reach increases. sunlight, increasing humidity in the understory and moderating
This hypothesis is based on the proposition that increases in temperatures of both air and water below. Gaps in the forest
plant species richness result in increases in potential niche canopy allow sunlight to penetrate, increasing daily and seasonal
space for

Table 1
Relationships between riparian vegetation and number of captures or biomass of nocturnal insects predicted by a priori hypotheses and the linear models associated with
each

Hypothesis Predicted relationship Model structure

Resource diversity Positive effect of vegetative species richness b0 + b1 (vegetative species richness)
Resource abundance Positive effect of vegetative cover b0 + b1 (percent vegetative cover)
Resource quality Positive effect of deciduous vegetative cover b0 + b1 (percent deciduous cover)
Resource concentration Positive effect of homogeneity of vegetation b0 + b1 (j0.5 — deciduous BA/total BAj)
Stream channel cover Positive effect of composition canopy cover over the b0 + b1 (canopy cover over stream)
stream channel
a
BA: basal area (m2).
 1 H.K. Ober, J.P. Hayes / Forest Ecology and Management 256 (2008) 1124–
variation in microclimatic conditions, which can reduce the
abundance, lifespan, and reproductive success of adult insects of
both terrestrial (Shure and Phillips, 1991) and aquatic origin
(Collier and Smith, 2000). Although gaps in the forest canopy
can promote growth of aquatic plants that in turn benefit some
insects of aquatic origin (Hawkins et al., 1982; Bilby and Bisson,
1992), we expect the negative impacts of microclimatic variation
resulting from reduced canopy cover to result in an overridingly
positive relationship between stream channel cover and
abundance and biomass of adult insects (Shure and Phillips,
1991; Peng et al., 1992).
2. Methods
2.1. Study area
We conducted research in the Oregon Coast Range in the Pacific
Northwest of the United States, an area characterized by a
maritime climate with wet, mild winters and cool, dry summers
(Franklin and Dyrness, 1973). The terrain is steep and rugged, with
Fig. 1. Location of 34 stream reaches where nocturnal flying insects were sampled in the Oregon Coast Range of western Oregon, USA.
a dense network of streams throughout. Elevation ranges from
sea level to 1250 m. Common riparian overstory species are
Douglas- fir (Pseudotsuga menziesii), western hemlock (Tsuga
heterophylla), red alder (Alnus rubra), big-leaf maple (Acer
macrophyllum), and western redcedar (Thuja plicata). The
understory tree layer is dominated by vine maple (Acer
circinatum) and hazelnut (Corylus cornuta), and the shrub layer
by salmonberry (Rubus spectabilis) and swordfern (Polystichum
munitum).
The northern, southern, western, and eastern boundaries of
our study region were the Columbia River, the southern border
of the Coos River sub-basin, the Pacific Ocean, and the Willamette
Valley, delineating a region approximately 300 km north to south
(Fig. 1). Within the boundaries of 17 watersheds located
throughout this region we selected one conifer-dominated
stream reach (canopy
cover <50% deciduous trees) and one deciduous-dominated
stream reach (canopy cover >50% deciduous trees). To locate
these stream reaches we generated random UTM coordinates and
determined the nearest point on a second- or third-order stream
for each point using 1:24,000 scale topographic quadrangle
maps. Points were eliminated from consideration if they (1) were
not on
 H.K. Ober, J.P. Hayes / Forest Ecology and Management 256 (2008) 1124–
1
either public property or private property to which we had been trees with <2 cm DBH) and understory trees (trees with 2–5 cm
granted access, (2) were located either along a stream already
selected or within a 2 km radius of a point already selected, (3)
had vegetation completely obstructing the airspace directly
over the
stream, (4) were <3 m or >7 m wide, or (5) were <0.5 m deep (as
determined by averaging the depth at 3 random points taken
along a transect perpendicular to stream flow). To reduce
logistical constraints, we divided the study area into subregions
and worked in the central subregion in 2002, the southern
subregion in 2003, and the northern subregion in 2004.

2.2. Insect sampling

We captured nocturnal insects using Universal Black Light

Traps (Bioquip Inc., Rancho Dominguez, California, USA) with
12- W fluorescent black light tubes, powered with 12-V marine gel-
cell batteries. We placed one trap per site 0.5 m off the ground
within 2 m of the stream edge, in a location that maximized
visibility from all directions. A ‘‘no-pest strip’’ (Hotshot,
Newport Beach, California, USA) was placed in the bottom of
each trap as a killing agent. Although not all species of insects
are equally attracted to black light traps, we used these traps
because we believe they are most suitable for catching nocturnal
aerial insects, they are useful for making relative comparisons of
insect abundance among sites, and they are known to have a
relatively small capture radius in forested areas, congruent with
the scale of our vegetation sampling (Ricketts et al., 2002;
Spalding and Parsons, 2004). To minimize the influence weather
conditions may have on insect captures, we did not sample on
nights with precipitation or on nights where
minimum temperatures were <5 8C. Wind speeds were minimal
in
riparian areas on all nights.
To account for temporal variability within a summer, we visited
all stream reaches once during each 2-week period between mid-
June and early September, for a total of five visits to each stream
reach. We sampled two of the stream reaches from a single
watershed each year to assess inter-annual variability in number
of insects captured. Each night, we began trapping within 30 min
of sunset and ceased 0–30 min after sunrise. In the morning,
insects were removed from traps, placed in labeled plastic
boxes, and
stored at —10 8C until analysis.
In the laboratory, we sorted each insect to order and measured
its body length from the anterior of the head to the posterior of
the last abdominal segment (antennae and cerci excluded) to the
nearest mm using a dissecting microscope. Directly weighing
insects often gives inaccurate results due to changes in biomass
as storage time increases and as carbohydrates are lost from
specimens during defrosting (Burgherr and Meter, 1997; Benke
et al., 1999). Therefore, we estimated biomass using length-mass
equations derived from insects in forested ecosystems of the
Pacific Northwest, USA (Judy Li, personal communication; Sabo
et al., 2002). For those orders not included in these studies
(Hemiptera, Lepidoptera, and Neuroptera) we used equations from
other regions to estimate biomass (Sample et al., 1993; Benke et al.,
1999).

2.3. Vegetation sampling

At each location where we deployed insect traps, we sampled

vegetation throughout a 30 m × 60 m plot (Fig. 2). We estimated
species richness, cover, and composition of vegetation in
30 m × 30 m plots on each side of the stream and
estimated canopy cover directly over the stream.
We recorded number of species of woody shrubs (defined as
woody plants with multiple stems arising from a single base or
1 H.K. Ober, J.P. Hayes / Forest Ecology and Management 256 (2008) 1124–
reach, and summed all three estimates to generate an index of
total woody

Fig. 2. Vegetation sampling plot. We identified and measured DBH of all trees
>5 cm DBH within the entire 30 m × 60 m plot. Within each of the eighteen 10
m × 2 m subplots, we recorded each species of shrub and understory tree,
estimated percent
cover of shrubs and understory trees, and estimated height of shrubs. At the
18 locations indicated by gray circles we measured canopy cover.

DBH) in eighteen 2 m × 10 m subplots located uniformly

through- out each 30 m × 60 m plot (Fig. 2) for our index of
species richness of shrubs and understory trees. Western
swordfern (Polystichum
munitum) was recorded as a shrub because it is a perennially
erect plant that contributes substantial biomass year-round
throughout the study region. Vine maple (Acer circinatum) was
always recorded as an understory tree, on account of the
stature of this
species. We recorded number of species of trees (DBH >5 cm)
throughout the entire 30 m × 60 m plot to estimate richness
of
canopy trees. We summed number of species of shrubs,
understory trees, and canopy trees (counting those species that
occurred in >1 vegetation stratum only once) to derive an
index of richness of all
woody species.
We visually estimated percent cover of woody shrubs and
understory trees in each of the eighteen 2 m × 10 m subplots
and estimated canopy tree cover from the center of each
subplot using
a moosehorn densiometer (Garrison, 1949). Each estimate was
made independently by two observers and averaged. We used
the mean of measurements from all 18 subplots as an index of
shrub, understory tree, and canopy tree cover for the stream
 H.K. Ober, J.P. Hayes / Forest Ecology and Management 256 (2008) 1124–
vegetative cover. We also derived an estimate of vegetative
cover of deciduous species by using this approach, limiting
measure- ments to deciduous species.
We measured DBH of each tree >5 cm DBH within each
30 m × 60 m plot, calculated the basal area of all coniferous and
deciduous trees separately, and used the ratio of basal area of
deciduous trees relative to that of all trees as an index of
vegetation composition of the stream reach. A value near 0
describes a stream reach where most trees are coniferous and a
value near 1 describes a stream reach where most trees are
deciduous. We then subtracted this ratio from 0.5 and took the
absolute value of this difference, resulting in a value ranging
between 0 and 0.5. Values near zero indicate a stream reach
contains an even mix of coniferous and deciduous vegetation
(low resource concentration), and values near 0.5 indicate either
a stream reach contains nearly exclusively coniferous trees or
nearly exclusively deciduous trees (high resource concentration).
We used a moosehorn densiometer to determine percent cover of
trees in the canopy directly over the stream channel at seven
points located at 5 m intervals along the 30 m stream reach. We
used the mean of these values as an index of canopy cover over
the stream.
2.4. Statistical analyses
We analyzed data for insects in the six most commonly
captured orders: Coleoptera, Diptera, Ephemeroptera, Hymenop-
tera, Lepidoptera, and Trichoptera. We used the total number of
individuals captured of a given order per stream reach per night
and the biomass of all individuals captured of a given order per
stream reach per night as response variables in all subsequent
analyses.
We developed statistical models to represent each of our five
hypotheses. For three of these hypotheses (resource diversity,
resource abundance, and resource quality), the possibility existed
that all three vegetation strata (shrubs, understory trees, canopy
trees) combined or one vegetation stratum alone was responsible
for driving associations between vegetation characteristics and
insects. To improve our understanding of which component of
riparian vegetation was most influential to insect abundance and
biomass, we first determined the relative ability of variables
pertaining to shrubs, understory trees, canopy trees, and the sum
of all three combined to explain variability in number of captures
and biomass of each order of insect for each of these three
hypotheses. We used the model receiving the most support in
this preliminary model selection step in subsequent analyses
evaluat- ing the degree of support from the data for each of the
five hypotheses relative to one another.
In this second step we evaluated the relative strength of
models representing each of the five original hypotheses (using
the most
1
relevant vegetation stratum determined in step one for the
resource diversity, resource abundance, and resource quality
hypotheses) and the null model (intercept only model) in
describing the observed patterns of number of captures and
biomass of each order of insect. We did not include quadratic
terms in our analyses because we had no biological reasons to
expect such relationships, the presence of such trends would be
difficult to detect given our sample size, and the number of
potential models would be excessively large if we included both
linear and non-linear trends for each hypothesis. All response
variables were ln transformed to more closely meet assumptions
of statistical models.
We used model selection to rank models according to their
likelihood, given the data (Burnham and Anderson, 2002). We
found no evidence of overdispersion and therefore used Akaike’s
Information Criteria corrected for small sample sizes (AIC c) to rank
models. We placed all models in rank order according to the
difference between the AICc score of each model and the lowest
AICc score of all models for that order (DAICc), and considered all
models with DAICc ≥ 2.0 to have substantial empirical support.
We used Akaike weights, wi, to evaluate the likelihood of each
model
relative to the others. All analyses were conducted in SAS (v9.1)
using PROC REG.
We tallied the number of times a model representing each of
the five hypotheses had DAICc ≥ 2.0 (omitting models with which
the null model was strongly competing, as these were not truly
supported by the data) to obtain an indication of which
hypotheses best explained number of captures and biomass of the
nocturnal flying insect community. This approach was preferable
to a single analysis with all orders combined because it provided
equal weighting for each order of insect.
3. Results
We collected 133,536 individual insects from 14 orders over
190 trap nights. On average, we captured 733 insects per stream
reach per night. The orders of insects most commonly captured
were Diptera (68.2% of individuals), Lepidoptera (13.3%),
Trichoptera (11.6%), Ephemeroptera (3.5%), Coleoptera (1.2%),
and Hymenoptera (0.7%) (Table 2); these six orders comprised
98.4% of all individuals captured. We found no statistically
significant inter-annual differences for number of captures of any
of these six most common orders of insects in stream reaches
repeatedly measured each year (p > 0.35 for all ANOVA
comparisons among years for each order of insect at each stream
reach), so we assumed the effect of year was negligible and
pooled data across years for subsequent analyses. Vegetation
characteristics at capture sites covered a wide range of variability
in forest conditions (Table 3).

Table 2
Number of individuals and biomass per stream reach per night of common orders of nocturnal insects captured in black light traps in riparian areas of the Oregon Coast Range,
summers 2002–2004

Order No. of individuals Biomass (mg)

X¯ S.E. Range X¯ S.E. Range

Coleoptera 6.7 1.01 0–98 62.8 10.99 0–1216.1

Diptera 517.6 83.76 7–9969 184.2 16.99 3.9–1239.0
Ephemeroptera 26.7 5.98 0–657 13.3 3.92 0–587.4
Hemiptera 0.2 0.04 0–3 0.1 0.07 0–15.0
Homoptera 2.4 0.37 0–29 0.8 0.29 0–37.8
Hymenoptera 4.5 0.47 0–40 5.2 0.70 0–60.6
Lepidoptera 85.1 12.92 1–1551 1778.9 145.04 6.2–9669.1
Neuroptera 0.1 0.04 0–4 4.7 1.71 0–138.7
Plecoptera 3.5 0.93 0–135 22.5 5.84 0–869.1
Trichoptera 78.8 7.69 1–605 146.7 14.59 0.6–1064.8
 1 H.K. Ober, J.P. Hayes / Forest Ecology and Management 256 (2008) 1124–

Table 3
Coleoptera, and Hymenoptera) had particularly strong associations
Range of values for characteristics of woody vegetation in riparian areas of the
Oregon Coast Range, summers 2002–2004 with canopy foliage. All eight of the models that received
substantial empirical support for these orders of insects had
either deciduous
Vegetation canopy cover orMinimum
characteristic stream channel canopy cover
Maximum as explanatory variables, and each of these models had relatively high
Range

Akaike weights,
Shrub indicating strong support
species richness 6 for these 21
models relative15to the others evaluated. Associations between vegetation strata and insects
withUnderstory
aquatic origins (Ephemeroptera0 and Trichoptera)
species richness 11 were less11
evident (Table 6).
Canopy species richness 2 6 4
Total woody species richness 10 21 11
4. Discussion
Shrub cover 26 94 68
Understory cover 0 55 55
Canopy cover 50 99 49 4.1. Relative importance of deciduous vegetation
Total woody vegetation cover 116 200 84
Deciduous shrub cover 18 83 64 The resource quality hypothesis received considerably more
Deciduous understory cover 0 42 42
Deciduous canopy cover 3 97 94
support from our data than any of the other hypotheses proposed,
Total deciduous cover 42 180 138 and it explained substantial variation in number of captures and
Concentration 0.003 0.5 0.497 biomass of nearly every order of insect investigated. These six
Each
Stream of the
channel five
canopy hypotheses3 received
cover substantial
100 empirical
97 orders of insects likely comprised the majority of the nocturnal
support (DAICc ≥ 2.0) for at least one of the two response variable flying insect community within riparian areas during summer
investigated, but support was not evenly distributed among throughout the study area, as they represented >98% of all insects
hypotheses (Tables 4 and 5). When we tallied the number of captured in black light traps set in randomly selected locations
times each hypothesis received substantial support, the resource throughout the region, collected throughout the course of three
abundance, resource diversity, and resource concentration summers. Consequently, we interpret these findings as an
hypotheses received one tally each, the stream channel cover indication that deciduous vegetation is an important habitat
hypothesis received two tallies, and the resource quality hypoth- component for the nocturnal flying insect community in this
esis received 10 tallies. Models pertaining to the resource quality ecosystem.
hypothesis explained 10–28% of the variation for either number of The resource quality hypothesis predicts that abundance and
captures or biomass of every order of insect investigated. For biomass of insects increase with amount of deciduous vegeta-
each order of insect, results were similar when either number of tive cover. Deciduous foliage is generally more palatable than
captures or biomass was used as the response variable. coniferous foliage; deciduous leaves have higher mineral
Each vegetative stratum investigated was relevant to number content, lower lignin content, and fewer resins relative to
of captures or biomass of insects, but each stratum was not coniferous foliage (Triska et al., 1975; Feeny, 1976; Mattson and
equally represented among strongly supported models. Of the 15 Scriber, 1987). Furthermore, nitrogen, the most limiting nutrient
models that received substantial empirical support, one model for growth and survival of many insects (Hodar et al., 2002),
pertained to the shrub stratum, one to the understory tree tends to occur in higher concentrations in deciduous foliage
stratum, two to all strata combined, and the remaining 11 to the than in conifer needles (Mattson and Scriber, 1987; Barbehenn
canopy tree stratum (Table 6). The three orders of terrestrial et al., 1999). Red alder, the dominant deciduous tree species in
insects (Lepidoptera,

Table 4
Candidate models relating number of captures per order of insect to vegetation characteristics in riparian areas of the Oregon Coast Range, with the distance from the most
parsimonious model (DAICc) and the Akaike weight (wi) associated with each

Coleoptera Diptera Ephemeroptera Hymenoptera Lepidoptera Trichoptera

DAICc wi DAICc wi DAICc wi DAICc wi DAICc wi DAICc wi

Resource diversity 6.7 0.02 0.0 0.27 0.0 0.34 8.7 0.01 3.4 0.08 5.9 0.04
Resource abundance 5.1 0.05 1.5 0.13 0.8 0.23 7.2 0.02 2.4 0.13 3.9 0.10
Resource quality 0.0 0.61 0.1 0.26 0.6 0.25 0.0 0.76 0.0 0.43 0.0 0.72
Resource concentration 2.1 0.22 1.4 0.13 2.9 0.08 11.5 0.00 4.1 0.05 4.7 0.07
Stream cover 5.4 0.04 3.0 0.06 4.8 0.03 2.7 0.20 1.8 0.17 6.9 0.02
Null 4.7 0.06 1.1 0.15 3.2 0.07 9.8 0.01 2.2 0.14 5.1 0.06

Models receiving substantial empirical support (DAICc ≤ 2.0) are in bold.

Table 5
Candidate models relating biomass per order of insect to vegetation characteristics in riparian areas of the Oregon Coast Range, with the distance from the most parsimonious
model (DAICc) and Akaike weight (wi) associated with each

Coleoptera Diptera Ephemeroptera Hymenoptera Lepidoptera Trichoptera

DAICc wi DAICc wi DAICc wi DAICc wi DAICc wi DAICc wi

Resource diversity 3.0 0.08 4.9 0.04 0.0 0.29 5.2 0.05 10.8 0.00 7.1 0.02
Resource abundance 3.2 0.08 5.4 0.03 1.8 0.12 4.5 0.07 9.1 0.01 4.8 0.07
Resource quality 0.0 0.38 0.1 0.40 0.4 0.24 0.0 0.67 0.0 0.91 0.0 0.79
Resource concentration 5.3 0.03 0.0 0.42 1.4 0.15 7.0 0.02 11.6 0.00 5.4 0.05
Stream cover 0.3 0.33 5.5 0.03 3.6 0.05 3.1 0.14 5.2 0.07 8.0 0.01
Null 2.7 0.10 3.1 0.09 1.4 0.15 5.3 0.05 9.2 0.01 5.7 0.05

Models receiving substantial empirical support (DAICc ≤ 2.0) are in bold.

 H.K. Ober, J.P. Hayes / Forest Ecology and Management 256 (2008) 1124–
Table 6
Explanatory variables and proportion of variance explained in each of the models that received substantial empirical support (DAICc ≤ 2.0) relating nocturnal flying insects to
riparian vegetation in the Oregon Coast Range
Metric Order
No. of captures Coleoptera
Ephemeroptera
Hymenoptera
Lepidoptera
Trichoptera
Biomass Coleoptera
Diptera
Hymenoptera
Lepidoptera
Trichoptera
No results are reported for instances where the null model received substantial empirical support.
a
All regression coefficients were positive, with the exception of the coefficient relating shrub species richness to Ephemeroptera.
riparian areas in the region, has foliage with unusually high
nitrogen content, making it a valuable food resource for many
leaf-feeding insects (Iversen, 1974). Given the prominence of
herbivores and detritivores in the orders of insects we
examined, this superior palatability of deciduous leaves in
general and of red alder leaves in particular was likely the
driving force underlying the strong associations we found
between insects and deciduous cover. Previous work has shown
that streams in the Pacific Northwest bordered by deciduous
trees tend to contain greater biomass of benthic invertebrates
(Piccolo and Wipfli, 2002), invertebrate drift (Romero et al.,
2005), and terrestrially derived invertebrates (Allen et al., 2003)
relative to streams bordered by coniferous trees. Our results
indicate that riparian areas with high densities of deciduous
trees contain greater abundance and biomass of the adult life
stages of nocturnal terrestrial and aquatic emergent insects as
well.
Although Ephemeroptera had an association with deciduous
canopy cover, the strength of this relationship was weaker
than that for terrestrial insects. In fact, associations between
Ephemeroptera and all the factors we measured were rela-
tively weak. This is likely due to the fact that most larval
Ephemeroptera are scrapers or collector-gatherers, consuming
either algae or fine detrital particles (Cummins and Klug, 1979).
None of the factors we measured reflected the presence of
algae, and due to downstream drift of small detrital particles,
the riparian vegetation conditions of a given stream reach may
not provide an accurate indication of food availability for
collector-gatherers.
Investigations of insect-plant associations spanning three
trophic levels, from insect predators and parasitoids through their
prey to vegetation, are necessarily complex (Price et al., 1980;
Siemann, 1998). If habitat selection by predators and parasitoids
is driven by abundance of prey, then the same factors that
influence habitat selection of potential prey should ultimately
determine habitat selection by predators and parasitoids, leading
to congruence in strength of associations between vegetative
factors and insects from all functional feeding groups (bottom-up
control, Hunter and Price, 1992). The similarity in associations
we found between insects and vegetative factors among all
orders of insects investigated suggests that this was the case.
Associations spanning from insect predators and parasitoids to
composition and structure of forest vegetation been reported on
other continents (Jukes et al., 2002; Fraser et al., 2007).
1
Explanatory variable in modela Proportion of variance explained
Deciduous canopy cover 0.17
Shrub species richness 0.13
Deciduous canopy cover 0.10
Understory tree cover 0.11
Deciduous canopy cover 0.28
Deciduous canopy cover 0.11
Stream channel cover 0.07
Total deciduous cover 0.18
Deciduous canopy cover 0.14
Stream channel cover 0.13
Vegetation concentration 0.15
Deciduous canopy cover 0.14
Deciduous canopy cover 0.20
Deciduous canopy cover 0.28
Total deciduous cover 0.20
4.2. Relative importance of the canopy stratum
Across all orders of insects examined, 11 of the final 15
models that received substantial support had explanatory
variables pertaining to the canopy, indicating that abundance and
biomass of nocturnal flying insects had stronger linkages with
canopy cover than with shrub or understory tree cover. The three
orders of terrestrial insects (Lepidoptera, Coleoptera, and
Hymenoptera) had particularly strong associations with canopy
cover. This may be the result of the host-specificity characteristic
of many insect species (Jaenike, 1990), and the greater
abundance of canopy foliage relative to understory and shrub
foliage. Specializing on plant species that occur in high densities
across wide geographic distributions (such as the species that
dominate the canopy of temperate forests) may be advantageous
for phytophages and detritivores because this behavior ensures a
more stable food supply across time and space than would
specialization on understory or shrub species that tend to have
more localized distributions (Ehrlich and Raven, 1964). A similar
mechanism may govern the abundance and biomass of predators
and parasitoids, whereby dependence on prey supported by
canopy trees occurring in high densities across wide geographic
distributions is preferable to reliance on prey associated with
understory or shrub species with more localized distributions.
Associations between Trichoptera and deciduous cover differed
from those between terrestrial insects and deciduous cover in that
Trichopteran associations encompassed all vegetative strata
rather than being more closely tied to the canopy stratum in
particular. This may be due to the fact that many Trichoptera are
shredders, typically selecting leaf litter on the basis of leaf
conditioning stage rather than on the basis of plant species
(Cummins and Klug, 1979). This selection criterion may make the
stratum from which terrestrial leaf litter originated a less
influential factor for most Trichoptera than it is for most
terrestrial insects.
The importance of the canopy in determining abundance and
biomass of insects relative to other vegetative stratum has
implications for conservation planning. Remotely sensed data
can provide reliable information on forest canopy composition at
fine- scale resolutions much more quickly and at lower costs than
field surveys (Martin et al., 1998; Hill and Thomson, 2005). Thus,
remote sensing of riparian vegetation could quickly and cheaply
facilitate theidentificationof areaslikely
tohavehighabundanceandbiomass of nocturnal flying insects, and
these areas could be given priority in limiting disturbance during
forest management activities.
1 H.K. Ober, J.P. Hayes / Forest Ecology and Management 256 (2008) 1124–
4.3. Scope and limitations
Although our study clearly indicates that deciduous riparian
vegetation explains variation in the abundance and biomass of all
orders of insects investigated, it is important to recognize the
scope and limitations of this study. Black light traps sample
nocturnal insects only, and they do not attract all orders of
nocturnal insects equally. Thus, diurnal orders were not sampled
at all, and relatively sedentary orders of nocturnal insects such as
Plecoptera and Homoptera were not well represented. We
recommend that future studies sample insect communities more
comprehensively to ascertain whether the same patterns hold
true across the entire riparian insect community.
5. Conclusions
Abundance and biomass of nearly all orders of insect
investigated were associated with cover of deciduous trees.
Deciduous vegetation is an important habitat element for many
other terrestrial and aquatic organisms in riparian areas of
coniferous forests, such as small mammals, amphibians, reptiles,
fish, and spiders (McComb et al., 1993; Gomez and Anthony, 1996;
Whiles and Wallace, 1997; Wipfli, 1997; Piccolo and Wipfli, 2002;
Allen et al., 2003; Premdas, 2004). Results suggest that heightened
insect abundance or biomass may be the underlying cause for
these associations previously noted between deciduous
vegetation and insect predators. Thus, maintaining or increasing
the cover of deciduous trees in riparian areas of coniferous
forests may confer benefits not only to insects, but also to those
vertebrates dependent upon them for food. These findings
parallel those from other regions (Angelstam and Mikusinski,
1994; Jokimaki et al., 1998; Wipfli and Musslewhite, 2004),
suggesting that deciduous vegetation may be universally
important for insects and their vertebrate predators in conifer-
dominated forests.
As concern grows over the influence of forest management
practices on biodiversity (Christensen et al., 1996; Simberloff,
1999; Spiecker, 2003), so too does the need to understand
mechanisms controlling biodiversity. Results indicate that abun-
dance and biomass of nocturnal flying insects in riparian areas of
conifer-dominated forests is strongly tied to cover of deciduous
trees, suggesting that manipulation of deciduous trees in riparian
areas represents a means by which resource managers can
potentially alter the abundance of both insects and their
predators. Conservation guidelines for riparian areas in conifer-
dominated forests should include recommendations regarding
the manipula- tion of vegetation composition.
Acknowledgements
A. Baker, J. Boland, N. Duncum, A. Fedoruk, T. Fox, A. Hendrickson,
J. Jones, F. Nogash, and A. Sjollema assisted with data collection, and
J. Boland, N. Duncum, and P. Giampaoli assisted with insect
sorting. We thank E. Bruna, J. Li, and anonymous reviewers for
providing comments that improved the document. This research
was supported by the Cooperative Forest Ecosystem Research
(CFER) program with funding from the USGS Forest and
Rangeland Ecosystem Science Center (FRESC), the US Forest
Service, and Simpson Timber Company and Weyerhaeuser
Company.
References
Allen, J.D., Wipfli, M.S., Caouette, J.P., Prussian, A., Rodgers, J., 2003. Influence of
streamside vegetation on inputs of terrestrial invertebrates to salmonid food-
webs. Can. J. Fish. Aquat. Sci. 60, 309–320.
Andow, D.A., 1991. Vegetational diversity and arthropod population response.
Annu. Rev. Entomol. 36, 561–586.
Angelstam, P., Mikusinski, G., 1994. Woodpecker assemblages in natural and
managed boreal and hemiboreal forest—a review. Annu. Zool. Fennici 31,
157–172.
Araujo, A.P.A., De Paula, J.D., Carneiro, M.A.A., Schoereder, J.H., 2006. Effects of host
plant architecture on colonization by galling insects. Aust. Ecol. 31, 343–348.
Barbehenn, R.V., Reese, J.C., Hagen, K.S., 1999. The food of insects. In: Huffaker, C.B.,
Gutierrez, A.P. (Eds.), Ecological Entomology. second edition. Wiley, New York,
pp. 83–113.
Benke, A.C., Huryn, A.D., Smock, L.A., Wallace, J.B., 1999. Length–mass relationships
for freshwater macroinvertebrates in North America with particular reference
to the southeastern United States. J. N. Am. Benthol. Soc. 18, 308–343.
Berryman, A.A., 1986. Forest Insects: Principles and Practice of Population Manage-
ment (Population Ecology). Plenum Press, New York, 294 pp.
Bilby, R.E., Bisson, P.A., 1992. Allochthonous versus autochthonous organic matter
contributions to the trophic support of fish populations in clear-cut and old-
growth forested streams. Can. J. Fish. Aquat. Sci. 49, 540–551.
Burgherr, P., Meter, E.I., 1997. Regression analysis of linear body dimensions vs. dry
mass in stream macroinvertebrates. Arch. Hydrobiol. 139, 101–112.
Burnham, K.P., Anderson, D.R., 2002. Model Selection and Multi-Model Inference: a
Practical Information-Theoretic Approach, second edition. Springer-Verlag,
New York, 496 pp.
Charnov, E.L., 1976. Optimal foraging, the marginal value theorem. Theor. Pop. Biol.
9, 129–136.
Christensen, N.L., Bartuska, A.M., Carpenter, S., D’Antonio, C., Francis, R., Franklin,
J.F., MacMahon, J.A., Noss, R.F., Parsons, D.J., Peterson, C.H., Turner, M.G.,
Woodman- see, R.G., 1996. The report of the Ecological Society of America
committee on the scientific basis for ecosystem management. Ecol. Appl. 6, 665–
691.
Collier, K.J., Smith, B.J., 2000. Interactions of adult stoneflies (Plecoptera) with
riparian zones. I. Effects of air temperature and humidity on longevity. Aquat.
Insects 22, 275–284.
Cummins, K.W., Klug, M.J., 1979. Feeding ecology of stream invertebrates. Annu.
Rev. Ecol. Syst. 10, 147–172.
Dempster, J.P., 1983. The natural control of populations of butterflies and moths.
Biol. Rev. 58, 461–481.
Ehrlich, P.R., Raven, P.H., 1964. Butterflies and plants: a study in coevolution.
Evolution 18, 586–608.
Ehrlich, P.R., Wilson, E.O., 1991. Biodiversity studies: science and policy. Science
253, 758–762.
Feeny, P., 1976. Plant apparency and chemical defense. Recent Adv. Phytochem.
10, 1–40.
Franklin, J.F., Dyrness, C.T., 1973. Natural Vegetation of Oregon and Washington.
General Technical Report PNW-8. USDA Forest Service, Pacific Northwest
Experimental Station, Portland, Oregon.
Fraser, S.E.M., Dytham, C., Mayhew, P.J., 2007. Determinants of parasitoid abun-
dance and diversity in woodland habitats. J. Appl. Ecol. 44, 352–361.
Friberg, N., Jacobsen, D., 1994. Feeding plasticity of two detritivore shredders.
Freshwater Biol. 32, 133–142.
Gadgil, P.D., Bain, J., 1999. Vulnerability of planted forest to biotic and abiotic
disturbances. New Forests 17, 227–238.
Garrison, G.A., 1949. Uses and modifications for the ‘‘moosehorn’’ crown closure
estimator. J. For. 47, 733–735.
Gomez, D.M., Anthony, R.G., 1996. Amphibian and reptile abundance in riparian and
upslope habitats of five forest types in western Oregon. Northwest Sci. 70,
109– 118.
Gruner, D.S., 2004. Attenuation of top-down and bottom-up forces in a complex
terrestrial community. Ecology 85, 3010–3022.
Haddad, N.M., Tilman, D., Haarstad, J., Ritchie, M., Knops, J.M.H., 2001. Contrasting
effects of plant richness and composition on insect communities: a field
experiment. Am. Nat. 158, 17–35.
Hawkins, C.P., Murphy, M.L., Anderson, N.H., 1982. Effects of canopy, substrate
composition, and gradient on the structure of macroinvertebrate communities
in Cascade Range streams of Oregon. Ecology 63, 1840–1856.
Hill, R.A., Thomson, A.G., 2005. Mapping woodland species composition and struc-
ture using airborne spectral and LiDAR data. Int. J. Remote Sensing 26, 3763–
3779.
Hodar, J.A., Zamora, R., Castro, J., 2002. Host utilization by moth and larval survival
of pine processionary caterpillar Thaumetopoea pityocampa in relation to food
quality in three Pinus species. Ecol. Entomol. 27, 292–301.
Holloway, J.D., Hebert, P.D.N., 1979. Ecological and taxonomic trends in macro-
lepidopteran host plant selection. Biol. J. Linn. Soc. 11, 229–251.
Hunter, M.D., Price, P.W., 1992. Playing chutes and ladders: heterogeneity and the
relative roles of bottom up and top down forces in natural communities.
Ecology 73, 724–732.
Iversen, T.M., 1974. Ingestion and growth in Sericostoma personatum
(Trichoptera) in relation to the nitrogen content of ingested leaves. Oikos 25,
278–282.
Jaenike, J., 1990. Host specialization in phytophagous insects. Annu. Rev. Ecol. Syst.
21, 243–273.
Jokimaki, J., Huhta, E., Itamies, J., Rahko, P., 1998. Distribution of arthropods in
relation to forest patch size, edge, and stand characteristics. Can. J. For. Res. 28,
1068–1072.
Jukes, M.R., Ferris, R., Peace, A.J., 2002. The influence of stand structure and
composition on diversity of canopy Coleoptera in coniferous plantations in
Britain. For. Ecol. Manage. 163, 27–41.
Kelty, M.J., 2006. The role of species mixtures in plantation forestry. For. Ecol.
Manage. 233, 195–204.
 H.K. Ober, J.P. Hayes / Forest Ecology and Management 256 (2008) 1124–
1
Kim, K.C., 1993. Biodiversity, conservation, and inventory: why insects matter.
actions between insect herbivores and natural enemies. Annu. Rev. Ecol.
Biodiv. Cons. 2, 191–214.
Syst. 11, 41–65.
Knops, J.M.H., Tilman, D., Haddad, N.M., Naeem, S., Mitchell, C.E., Haarstad, J.,
Ricketts, T.H., Daily, G.C., Ehrlich, P.R., 2002. Does butterfly diversity predict moth
Ritchie, M.E., Howe, K.M., Reich, P.B., Siemann, E., Groth, J., 1999. Effects of
diversity? Testing a popular indicator taxon at local scales. Biol. Cons. 103,
plant species richness on invasion dynamics, disease outbreaks, insect abun-
361– 370.
dances and diversity. Ecol. Lett. 2, 286–293.
Robinson, G.S., Ackerly, P.R., Kitching, I.J., Beccaloni, G.W., Hernandez, L.M.,
Langor, D.W., Spence, J.R., 2006. Arthropods as ecological indicators of sustainability
2000. Host Plants of the Moth and Butterfly Caterpillars of America north of
in Canadian forests. For. Chron. 82, 344–350.
Mexico, vol. 69. American Entomological Institute, Gainesville, FL.
Lawton, J.H., 1983. Plant architecture and the diversity of phytophagous insects.
Romero, N., Gresswell, R.E., Li, J.L., 2005. Changing patterns in coastal cutthroat
Annu. Rev. Entomol. 28, 23–39.
trout (Oncorynchus clarki clarki) diet and prey in a gradient of deciduous
Lewinsohn, T.M., Novotny, V., Basset, Y., 2005. Insects on plants: diversity of
canopies. Can. J. Fish. Aquat. Sci. 62, 1797–1807.
herbivore assemblages revisited. Annu. Rev. Ecol. Syst. 36, 597–620.
Sabo, J.L., Bastow, J.L., Power, M.E., 2002. Length–mass relationships for adult
Long, Z.T., Mohler, C.L., Carson, W.P., 2003. Extending the resource concentration
aquatic and terrestrial invertebrates in a California watershed. J. N. Am. Benthol.
hypothesis to plant communities: effects of litter and herbivores. Ecology 84, 652–
Soc. 21, 336–343.
665.
Sample, B.E., Cooper, R.J., Greer, R.D., Whitmore, R.C., 1993. Estimation of insect
Losey, J.E., Vaughan, M., 2006. The economic value of ecological services provided
biomass by length and width. Am. Midl. Nat. 129, 234–240.
by insects. Bioscience 56, 311–323.
Schowalter, T.D., 1985. Adaptations of insects to disturbance. In: Pickett, S.T.A.,White,
Martin, M.E., Newman, S.D., Congalton, R.G., 1998. Determining forest species
P.S. (Eds.), The Ecology of Natural Disturbance and Patch Dynamics. Academic
composition using high spectral resolution remote sensing data. Remote Sen-
Press, Orlando, FL, pp. 235–252.
sing Environ. 65, 249–254.
Schowalter, T.D., Hargrove, W.W., Crossley Jr., D.A., 1986. Herbivory in forested
Mattson, W.J., Scriber, J.M., 1987. Nutritional ecology of insect folivores of
ecosystems. Annu. Rev. Entomol. 31, 177–196.
woody plants: nitrogen, water, fiber, and mineral considerations. In: Slansky,
Jr., F., Rodriguez, J. (Eds.), Nutritional Ecology of Insects, Mites, Spiders, and Siemann, E., 1998. Experimental tests of effects of plant productivity and diversity
Related Invertebrates. Wiley, New York, pp. 105–146. on grassland arthropod diversity. Ecology 79, 2057–2070.
McComb, W.C., McGarigal, K., Anthony, R.G., 1993. Small mammal and amphibian Shure, D.J., Phillips, D.L., 1991. Patch size of forest openings and arthropod popula-
abundance in streamside and upslope habitats of mature Douglas-fir stands, tions. Oecologia 86, 325–334.
Simberloff, D., 1999. The role of science in the preservation of forest biodiversity.
western Oregon. Northwest Sci. 67, 7–15.
For. Ecol. Manage. 115, 101–111.
Miller, J.C., 1993. Insect natural history, multi-species interactions and biodiversity
Smith, H.R., Remington, C.L., 1996. Food specificity in interspecies competition:
in ecosystems. Biodiv. Cons. 2, 233–241.
comparisons between terrestrial vertebrates and arthropods. Bioscience 46, 436–
Muzika, R.M., Liebold, A.M., 2000. A critique of silvicultural approaches to
447.
managing defoliating insects in North America. Agric. For. Entomol. 2, 97–105.
Spalding, A., Parsons, M., 2004. Light trap transects—a field method for ascertaining
Ober, H.K., Hayes, J.P., 2008. Influence of vegetation on bat use of riparian areas at
the habitat preferences of night-flying Lepidoptera, using Mythimna turca
multiple spatial scales. J. Wildl. Manage. 72, 396–404.
(Linnaeus 1761) (Lepidoptera: Noctuidae) as an example. J. Insect Cons. 8,
Ohgushi, T., 1992. Resource limitation on insect herbivore populations. In: Hunter,
185–190.
M.D., Ohgushi, T., Price, P.W. (Eds.), Effects of Resource Distribution on
Spiecker, H., 2003. Silvicultural management in maintaining biodiversity and
Animal– Plant Interactions. Academic Press, San Diego, CA, pp. 199–241.
resistance of forest in Europe—temperate zone. J. Environ. Manage. 67, 55–65.
Peng, R.K., Fletcher, C.R., Sutton, S.L., 1992. The effect of microclimate on flying
Stephens, D.W., Krebs, J.R., 1986. Foraging Theory. Princeton University Press, NJ,
dipterans. Int. J. Biometeorol. 36, 69–76.
262 pp.
Piccolo, J.L., Wipfli, M.S., 2002. Does red alder (Alnus rubra) in upland riparian
Strong, D.R., Lawton, J.H., Southwood, R., 1984. Insects on Plants. Blackwell Scien-
forests elevate macroinvertebrate and detritus export from headwater streams to
tific, Oxford, United Kingdom, 313 pp.
downstream habitats in southeastern Alaska? Can. J. Fish. Aquat. Sci. 59, 503–
Taylor, R.J., Doran, N., 2001. Use of terrestrial invertebrates as indicators of the
513.
ecological sustainability of forest management under the Montreal Process. J.
Polis, G.A., Hurd, S.D., Jackson, C.T., Pinero, F.S., 1997. El Nino effects on the dynamics
Insect Cons. 5, 221–231.
and control of an island ecosystem in the Gulf of California. Ecology 78, 1884–
Triska, F.J., Sedell, J.R., Buckley, B., 1975. The processing of conifer and hardwood
1897.
leaves in two coniferous forest streams. II. Biochemical and nutrient changes.
Premdas, S., 2004. The influence of canopy type and seasons on adult aquatic
Vereinigung Theoretische Angew. Limnol. 19, 1628–1639.
insect emergence and riparian spiders in the Oregon Coast Range. M.S. Thesis.
Whiles, M.R., Wallace, J.B., 1997. Leaf litter decomposition and
Oregon State University, Corvallis, OR.
macroinvertebrate communities in headwater streams draining pine and
Price, P.W., 1992. Plant resources as the mechanistic basis for insect herbivore
hardwood catchments. Hydrobiology 353, 107–119.
population dynamics. In: Hunter, M.D., Ohgushi, T., Price, P.W. (Eds.), Effects of
Wipfli, M.S., 1997. Terrestrial invertebrates as salmonid prey and nitrogen sources
Resource Distribution on Animal–Plant Interactions. Academic Press, San Diego,
in streams: contrasting old-growth and young-growth riparian forests in south-
CA, pp. 139–173.
eastern Alaska, USA. Can. J. Fish. Aquat. Sci. 54, 1259–1269.
Price, P.W., 2002. Resource-driven terrestrial interaction webs. Ecol. Res. 17, 241–247.
Wipfli, M.S., Musslewhite, J., 2004. Density of red alder (Alnus rubra) in headwaters
Price, P.W., Bouton, C.E., Gross, P., McPheron, B.A., Thompson, J.N., Weis, A.E.,
influences invertebrate and detritus subsidies to downstream fish habitats in
1980. Interactions among three trophic levels: Influence of plants on inter-
Alaska. Hydrobiology 520, 153–163.