Primary consumer and detritivore communities (Diptera: Ephydridae) in newly restored and constructed wetlands

2004·6 MB·

by  Bruce A Steinly

PROC. ENTOMOL. SOC. WASH.
106(2), 2004, pp. 460-471
PRIMARY CONSUMER AND DETRITIVORE COMMUNITIES
(DIPTERA: EPHYDRIDAE) IN NEWLY
RESTORED AND CONSTRUCTED WETLANDS
Bruce A. Steinly
OH
Miami University, Zoology Department, 212 Pearson Hall, Oxford, 45056, U.S.A.
(e-mail: [email protected])
—
Abstract. Although maximum diversity of higher Diptera has been reported in the
interface between wetlands and other ecosystems, few studies have documented successful
restoration of wetland insects within these ecotones. The consistent collection of shore
flies (Diptera: Ephydridae) supports the hypothesis that shore flies rapidly colonize newly
restored and constructed wetlands at Miami Trace and Winton Woods county Parks, Ham-
ilton County, Ohio. Rapid colonization of shore flies exhibited a temporal shift that is
associated with aquatic vegetation colonization, growth, and maturation. Species com-
position of shore-fly communities suggests that a minimum of five general nutrient sources
were exploited in newly established wetlands, and shore flies may be an essential part of
A
the establishment of complex food webs. comparison of Sorenson and diversity indices
suggests that shore-fly communities in constructed and restored Ohio wetlands were dis-
tinct species assemblages. Significant differences among shore-fly communities are attri-
buted to variation in species abundance. Although restored and constructed wetlands were
flooded during 1998, variation in species abundance and monthly species accumulation
suggests that the development of ecosystems within each wetland was asynchronous.
Key Words: restored wetlands, Diptera, Ephydridae, colonization, community diversity,
food webs
In many regions of the United States, the essary for reproduction and rapid growth
restoration of wetlands has been initiated to (Driver et al. 1974). Decreased availability
provide habitat for vertebrate species. The of food has been associated with duckling
success of bird, amphibian, and mammal mortality (Johnson et al. 1992, King and
recolonization in wetlands is directly relat- Brazner 1999), and duckling brood avoid-
ed to the type, quality, and abundance of ance and abandonment of wetlands that
food resources within wetland ecosystems have small numbers of invertebrates (Coo-
(Wilson 1987). One of the major sources of per and Anderson 1996). Although the im-
food for waterfowl, shore birds, and fish portance of insect resources to waterfowl
species is insects (Martin and Uhler 1939, development has been recognized, an un-
Zahl 1967, Clarke 1976, Murkin and Batt derstanding of the contributions of insects
1987). During egg laying and brood rear- to wetland food webs is limited (Rosenberg
ing, adult and juvenile ducks consume in- and Danks 1987, Batzer and Wissinger
vertebrate food (including insects) (Krapu 1996, Hansen and Castelle 2000). As Batz-
and Swanson 1975, Bataille and Baldassar- er and Wissinger (1996) stated, many of the
re 1993) to obtain the protein that is nec- previous experiments and assumptions con-
VOLUME NUMBER
106. 2 461
cerning wetland insect ecology require re- newly flooded decaying organic materials
evaluation to improve the management of (Layton and Voshell 1991). However, as an
insect resources as waterfowl food. Foun- area matures, plant detritus decreases and
dations for regulating insect and waterfowl production of unicellular and multi-cellular
communities are in the initial stages of de- algae and macrophytes increases (Layton
velopment (Batzer and Wissinger 1996). and Voshell 1991) resulting in increased
Previously, a majority of wetland inves- primary consumer abundance and species
tigations have focused on plant communi- richness. Although Batzer and Wissinger
ties, soils, hydrology, chemical processes, (1996) did not monitor primary consumer
benthic invertebrates, and vertebrate popu- communities, their review suesests that de-
lations, such as waterfowl (Sharitz and tritivore communities are not affected by
Batzer 1999). Although maximum diversity macrophyte and algal growth. They suggest
of higher Diptera is at the interface between that detritivore populations either remain
mature wetlands and other ecosystems constant or increase where multi-cellular
(Deonier 1965, LaSalle and Rozas 1991, plants were cut and debris removed.
Marshall 1994, Scheiring and Foote 1973, In this paper, evidence is presented to
Steinly 1986, Thier and Foote 1980, Keiper document the initial colonization of newly
et al. 2002), and the greatest production of restored and constructed wetlands by shore-
insect biomass is found in sparse emergent fly species that have been associated with
vegetation zones (Voigts 1976, Grains the consumption of detritus and primary
1980, Kaminiski and Prince 1981, Mc- production. Temporal shifts in the richness
Lauglin and Harris 1990), only a few in- and abundance of primary consumers and
vestigations have focused on the successful detritivores i.e., shore flies are associated
restoration of wetland macroinvertebrates with the growth of macrophytes and matu-
that are found in the water column and/or ration of newly established wetlands.
sediments (Danell and Sjoberg 1982, Flor- Shore-fly diversity, relative abundance,
ida Department of Environmental Protec- richness, and similarity values from these
tion 1994, Brown et al. 1997). The attrac- wetlands are compared. Diversity values are
tion of shore flies to artificial pools and oth- compared with a /-test to identify signifi-
er habitats in a constructed wastewater cant differences in shore-fly communities.
treatment wetland at 2 days post-flooding
Materials and Methods
and the establishment of shore-fly popula-
tions suggest that colonization by these spe- During 1997, the Hamilton County Park
cies was not incidental (Keiper and Walton District completed the restoration of a sin-
2002, Keiper et al. 2002). Although two in- gle wetland at Miami Trace, and the con-
vestigations have focused on Diptera that struction of two wetlands at Winton Woods,
inhabit soil and consume decaying vegeta- Hamilton County, Ohio. At Miami Trace,
tion in restored Florida (Streever et al. the renewed flooding represents a restora-
1996) and Washington (Hansen and Cas- tion of a wetland that was surveyed and de-
A
telle 2000) wetlands, studies of primary scribed before the settlement of Ohio.
consumer richness or abundance have not single wetland was leveed at Miami Trace
been reported. County Park without basin alteration, and
Layton and Voshell ( 1991 ) suggested that is located northwest of New Haven and ap-
an increase in habitat diversity would most proximately, 0.55 km east (39°17.3'N,
likely lead to increases in invertebrate di- 84°44.8'W) of a larger restored wetland
versity within wetland ecosystems. It is complex. Before wetland restoration, the
common for detritus-feeding insects (i.e., area was covered with old-field vegetation
insects that feed on decaying plant and an- and was well drained. Post-restoration veg-
imal tissues) to dominate areas that contain etation consisted of scattered stands of T\-
462 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON
pha latifoUa L. and patches of filamentous habitat) (Scheiring and Foote 1973; Re-
algae that were surrounded by unvegetated gensburg 1976; Steinly 1978, 1986, 1990,
mud shores. The Winton Woods County 2001). One sweep sample is the combina-
Park (Winton Woods, Ohio) sites consist of tion of a back and forth movement of the
two basins on the northern edge of the park net through a habitat. Sweep net sampling
that are named Mallard (39°15.9'N, is a quick and inexpensive means of sam-
84°31.rW) and Heron (39°15.9'N, pling diverse communities of invertebrates
84°30.9'W) Wetlands. These wetlands were (Murkin et al. 1983, Cheal et al. 1993). The
constructed to control precipitation run-off size of sampling areas and the number of
from the suiTounding watershed and to pro- net sweeps were equal among habitats,
vide habitat for vertebrate species. Mallard Plant debris was removed immediately
Wetland was bordered on the south by a from the samples and stored temporarily in
steep grass shore while the wetland interior petri dishes. Mounted and unpinned speci-
contained scattered stands of T. latifolia mens were identified, labeled and counted,
mixed with Spargcmium americanum Nut- Mounted and vialed voucher specimens are
tall, Alisma subcordatuni Raf., and Carex. deposited in the Miami University Insect
Heron Wetland had a similar grass shore on Collection.
the south side that was well drained. The The percentage relative abundance is cal-
shallow slope of the northern shore of Her- culated with the formula R. A. = Ai/N X
on Wetland promoted the retention of soil 100, where R. A. is the percentage relative
water that sustained the dense growth of abundance. A, is equal to the abundance of
N
sedges and scattered Sagittaria latifolia each species, and is the total number of
Willd. on the shoreline. Emergent Ranun- shore flies within the wetland. The percent-
culiisflabellaris Raf. occupied two thirds of age ranges are characterized as follows: 1-
the wetland pool. Although trace amounts 2% rare (r), >2-8% occasional (occ), >8-
of precipitation fell during July—September, 14% common (c), > 14-25% abundant (a),
the Winton Woods sites held water at a fair- and >25-100% very abundant (va) (Deon-
ly constant level until mid August. During ier 1965; Scheiring and Foote 1973; Deon-
late August through September, areas of ier and Regensburg 1978; Steinly and
limestone sediment in both Winton Woods Deonier 1980; Steinly 1984, 1986, 1990).
wetlands were exposed that was covered The Shannon-Wiener diversity index
with sparse plant debris. As the season pro- (H') was calculated for all wetlands because
gressed, larger areas of sediment were ex- it incorporates both species richness and
posed in Mallard Wetland and these mud abundance (Scheiring 1974). Shore-fly di-
shores were colonized by widely scattered versity (H') values for each wetland were
clumps of Setaria glauca Kuntze (yellow compared with a /-test to identify differenc-
foxtail), grass (-es), and Sp. americanum, es in shore-fly communities (Zar 1984). Di-
while Carex sp. and Sa. latifolia grew on versity is calculated by: H' = —S Pi logmPj
newly exposed mud shore at Heron Wet- where p, is n,/N, n, is the number of indi-
land. Visual evidence of leaf mining con- viduals of the ith species of the habitat be-
firmed that pioneering grasses were colo- ing considered, and N is the total number
nized by dipteran primary consumers. of individuals per habitat. H' is essentially
Bi-weekly collections of shore flies were dimensionless and usually not affected by
initiated at Miami Trace and Winton Woods sample size (N) (Olive and Dambach 1973,
in May and continued through October Scheiring 1974). Newly restored wetland
1998. Wetlands were subdivided into habi- habitats of comparable area were sampled
tats that were characterized by substrate and for approximately the same amount of time
vegetation types and sampled with a mod- and probable differences in richness and
ified aerial sweep net (150 net sweeps per abundance of shore-fly species reflect bio-
VOLUME NUMBER
106. 2 463
o
0>3 o
U C3 ON — \C o
o
iri
72
O
s^!
x»
5 o
464 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON
logical differences among wetlands (Scheir- ble 4) did not contain shore-fly specimens
ing 1974). (Fig. 1). Although initial collection of shore
The community composition of each flies from Mallard and Heron wetlands
wetland was compared by means of the So- started in June and July, respectively, spe-
renson Index of Similarity (I). Similarity is cies accumulation patterns are similar (Fig.
calculated with the formula 1 = 2 C/(A + 1). Shore-fly species accumulation in the
B), where I is the index of similarity, C is Miami Trace wetland started in May and
the number of species shared, A is the num- approached asymptote during September
ber of species in habitat A, and B is number (Fig. 1).
of species in habitat B (Scheiring and In the Mallard and Heron wetlands, the
Deonier 1979, Steinly 1984). The Sorenson abundance of Allotrichoma simplex, Poly-
index ranges from when there is no sim- trichophora orbitalis, and Discocerina ob-
ilarity (no species shared) between habitats scurella, increased during August and Sep-
to 1 when there is complete similarity. tember (Table 2 and 3), respectively, while
A. simplex and Hydrellia griseola appeared
Results in May collections and abundance contin-
During May through October of 1998, ued to increase through July in the restored
4,818 shore flies were collected from the wetlands at Miami Trace (Table 4). Al-
newly restored and/or constructed wetlands though a single Po. orbitalis adult was col-
within Miami Trace and Winton Woods lected in May in the restored wetland, ad-
County Parks, respectively (Table 1). Col- ditional specimens were not captured until
lections from Mallard and Heron wetlands July (Table 4). A. simplex and Po. orbitalis
at Winton Woods, and the restored wetlands consume decaying organic matter (Foote
at Miami Trace contained 31, 26, and 21 and Eastin 1971, Foote 1995).
(Table 1) species of shore fly within sam- The abundance of two leaf-mining spe-
ples of 3,307, 949, and 562 specimens, re- cies, Hydrellia griseola and Hydrellia tibi-
spectively (Tables 1-4). Of the total num- alis, and T. atra increased dramatically dur-
ber of shore flies, detritivores accounted for ing July through August, and September in
approximately 78, 49, and 60% of the ephy- Miami Trace and Winton Woods wetlands,
drid communities within Mallard, Heron, respectively (Tables 2-4). Also, appreciable
and Miami Trace wetlands, respectively numbers of Paralimna punctipennis and
(Table 1). Approximately 86.7%, i.e., 4,177 Scatella stagnalis were found in Mallard
of the total number of shore flies collected and Heron wetlands (Tables 2 and 3) during
during the season from all wetlands con- September. Sc. stagnalis, and Pa. puncti-
sisted of Allotrichoma simplex (Loew) pennis feed on cyanobacteria and diatoms,
{1.1%), Discocerina obscurella (Fallen) respectively (Foote 1995). Hyadina albov-
(48.8%), Hydrellia griseola (Fallen) enosa Coquillett, Hyadina binotata (Cres-
(9.2%), Hydrellia tibialis Cresson (4.5%), son), Hyadina pruinosa (Cresson), and Hy-
Paralimna punctipennis (Wiedmann) drellia formosa Loew were rare (r) species
(3.5%), Polytrichophora orbitalis (Loew) (Table 1) and collected early in the season
(4.6%), Scatella stagnalis (Fallen) (4.82%) from grass shore habitat at Heron Wetland
and Typopsilopa atra Loew (3.1%). The (Table 3). Although relatively large num-
abundance of these species ranged from oc- bers of Parydra (Table 5) and Notiphila
casional (occ) to very abundant (va) (Table species were routinely collected in aquatic
1). All other species were rare (r) and were habitats and mature wetlands in Ohio
represented by fewer than 100 individuals (Steinly 1978, Todd and Foote 1987, Lar-
(Tables 1—4). May samples from the Heron son and Foote 1997), respectively, Notiphi-
and Mallard wetlands (Tables 2 and 3) and la adusta Mathis, Notiphila loewi Cresson,
October collections from Miami Trace (Ta- Notiphila pauroura Mathis, Notiphila
VOLUME NUMBER
106, 2 465
Table 2. Shore-fly community of Mallard Wetland at Winton Woods County Park.
466 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON
Table 3. Shore-fly community of Heron Wetland at Winton Woods County Park.
July Aug. Sept. Oct. ToI;il
AIlotrichoma simplex 41 2 44
Diachaeta caudata 3 3
Discocerina brunneonitens 1 1
Discocerina obscurella 39 155 196
Hyadina albovenosa 34 1 35
Hyadiiui binotata 10 2 12
Hyadina pniiiiosa 1 1
Hydrelliaformosa 6 1 7
Hydrellia griseola 4 82 16 102
Hydrellia tibialis 47 47
Lytogaster excavata 3 4 1 8
Nostima sciitellaris 2 5 8
Notipliila adusta 8 9
Notiphila loewi 5 6
Notiphila paiiroitra 22 22
Notiphila phaeopsis 2 18 20
Paralimna pimctipennis 3 70 79
Parydra aquila 5 5
Parydra breviceps 10 13
Parydra quadrituhercidata 3 3
Philygria debilis 1
Polytrichophora orbitalis 30 36
Ptilomyia enigma 3 3
Scatella stagnalis 145 2 147
Typsilopa atra 13 9 58 45 125
Zerosftavipes 14 2 16
Total = 71 26 613 239 949
were not adversely affected by low quan- creased when macrophytes were cut and de-
tities of decaying organic matter. Although bris removed.
shore-fly primary consumers were collected The early colonization of restored wet-
from Miami Trace and Mallard wetlands, land ecosystems by large numbers of A.
approximately 60 to 78% of the specimens simplex, D. obscurella, and Po. orbitalis
collected were detritivores, respectively suggests these species are important com-
(Table 1). This investigation confirms that ponents in the early development of viable
shore-fly primary consumers and detriti- food chains that sustain diverse predaceous
vores rapidly colonize restored and con- invertebrate and/or vertebrate communities.
structed wetlands in Ohio. Although Scheir- During late summer, increases in the pop-
ing and Deonier (1979) proposed that re- ulations of Hydrellia griseola, Hydrellia
source quantity increased the richness and tibialis (i.e., leaf miners), T. atra, and a di-
abundance of well-adapted ephydrid spe- atom feeder, Paralimna punctipennis are as-
cies in transient habitats, the low detritus sociated with increased density and growth
quantities, and abundance of detritivores in of macrophyte and diatom food resources.
newly restored wetlands supports Batzer Typopsilopa spp. have been associated with
and Wissinger's (1996) contention that de- the consumption of decaying tissue on dam-
tritivore colonization was not affected by aged monocot stems (Keiper et al. 2001).
the growth of macrophytes and algae, and The increase of micro- and macrophyte pro-
insect abundance remained constant or in- duction represents a significant augmenta-
VOLUME NUMBER
106. 2 467
Table 4. Shore-fly community of a new wetland at Miami Trace County Park.
May June Jul\ Au'j Sept.
Allotrichoma simplex
468 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON
Table 6. Sorenson Index of Similarity values for a newly created and two constructed wetlands in Ohio.
Mallard Welland New Wetland
Winton Woods Miami Trace
Heron Wetland Winton Woods 0.81* 0.51*
Mallard Wetland Winton Woods 0.65*
* Comparison of community diversity values P < 0.001.
ami Trace complex of restored wetlands flies, wetland production is accessible to
(Steinly, unpub. data), only a few adults predaceous invertebrates (e.g., Ochthera
were collected from the restored wetland anatoUkos Clausen) and vertebrates. The
(Table and the low richness and abun- abundance of insect consumer populations
1 )
dance of Notiphila spp. are attributed to the has been linked to waterfowl reproductive
paucity of emergent vegetation in the re- success that is dependent on the quantity
stored wetland. and quality of insect protein (Driver et al.
The abundance of detritivores and leaf- 1974). In some instances early emergence
miners, diatom consumers, i.e.. Pa. pimcti- of the Chironomidae (Diptera) provides a
pennis, and Sc. stagnalis, a possible sec- large quantity of protein (McLaughlin and
ondary consumer of damaged monocots Harris 1990), while waterfowl utilize shore
stems, T. atra, and a cyanobacteria feeder flies and other families of Diptera for food
suggests that at least five main nutrient later in the season.
sources were exploited by shore flies during Superficially, the significant difference in
the early stages of wetland restoration. Dur- H' values and the high Sorenson similarity
ing the initial stages of wetland restoration, value for the Mallard and Heron richness
the abundance of detritivores, leafminers, a comparison are contra indicators. Although
secondary consumer, an algivorous species, a Sorenson index of 0.81 and monthly spe-
and consumers of cyanobacteria (e.g., Nos- cies accumulation (Fig. 1) suggest that
tima and Hyadina species) suggest that the shore-fly community richness values in
rudiments of viable food chains were pre- Mallard and Heron wetlands are compara-
sent in late spring and early summer within ble, the significant difference in H' values
the Mallard and Heron wetlands. Early col- (P < .0001, t = -22.77, df = 2540) sug-
lection, the increase of shore-fly richness gests that these wetlands harbored unique
and abundance (Tables 2-4) during the species assemblages. The difference in
summer months, and the utilization of five shore-fly assemblages is attributed to ex-
different nutrient resources, i.e., algae, mac- treme variation of individual species abun-
rophytes, damaged monocot stems, detritus, dance within each wetland (Table 1). Al-
and cyanobacteria suggest that shore-fly though, the relatively high index of simi-
colonization provides a foundation for the larity (Table 6) suggests that biological and
development of diverse food chains and physical conditions were comparable in
food webs. The successful restoration and Mallard and Heron wetlands, the variation
maturation of wetland food chains is de- in shore-fly species abundance and the dif-
pendent on establishment and growth of ference in the abundance of detritivores and
aquatic micro- and macrophyte and microbe primary consumers suggest that ecosystem
communities, accumulation of detritus, and development and/or maturation within each
concurrent colonization and establishment wetland was asynchronous.
of resident primary consumers and detriti- Comparison of Miami Trace to Mallard
vores, i.e., shore flies that move nutrients and Heron Wetlands yielded low Sorenson
into food webs. Once detritus, bacterial, and indices (Table 6), differences in species ac-
plant nutrients are assimilated by shore cumulation patterns, and H' values that
VOLUME NUMBER
106. 2 469
35 n
Miami Trace Wetland
- Mallard Wetland
30
- - Heron Wetland
§ 25
s
S
3 20
<
a
1/3
o
JS
!Z3
May June July August September October
Fig. 1. Shore-fly species accumulation in newly constructed and restored wetlands.
were significantly different (P < 0.001, t = communities in restored wetlands is want-
9.38, df = 922 and t = -5.68, df = 863, ing. The scarcity of information is attribut-
respectively). These differences suggest ed to the daunting task of identifying large
that the Miami Trace species assemblage numbers of invertebrate species, and spe-
was unique. The difference in shore-fly cies interactions within and between food
communities is attributed to local precipi- webs, and the lack of sampling in shoreline
tation frequency and quantity, and disparate habitats (Keiper et al. 2002). Additionally,
physical and biological conditions that are the number of food chain interactions and
unique to restored and/or constructed wet- food web dynamics may vary from one
lands. Further, the difference in shore-fly geographic region to another.
communities at Miami Trace as compared The Ephydridae are a trophically diverse
to the Winton Woods wetlands is confirmed family that provides a unique opportunity
by early colonization of shore flies in May to study the movement of wetland produc-
vs. June and July and early development of tion into food webs because shore-fly spe-
species accumulation asymptote (Fig. in cies are routinely collected and abundant in
1 )
the restored wetlands. wetlands. Without the colonization of wet-
The importance of insect primary con- lands by shore flies and/or other insect fam-
sumers and detritivores within food chains ilies that contain primary consumers and
has been acknowledged (Batzer and Wis- detritivores, the movement of primary pro-
singer 1996, Hansen and Castelle 2000), duction and detritus into food chains and
but a comprehensive understanding of in- development of food webs may be limited.
sect herbivore, detritivore. and predator Design and management of wetlands that