Get documents about "The Impact of Insecticide Treatment on Abundance
Document Sample
Limnologica28 (2) LIMNOLOGICA
(1993)93-106
i; by Gustav Fischer Verlag Jena
Department of Entomology, University of Georgia, Athens. Georgia, USA;
Institute of Ecology, University of Georgia. Athens, Georgia. USA
The Impact of Insecticide Treatment on Abundance, Biomass
and Production of Litterbag Fauna in a Headwater Stream:
A Study of Pretreatment, Treatment and Recovery
K. CHUNG, J. B. WALLACE & J. W. GRUBAUGH
With 5 Figures
Key words: Macroinvertebrates: Disturbance: Recovery; Recolonization:
Organic matter processing: Streams.
Summary
The insecticide methoxychlor was applied seasonally for three source populations. The results show that community recovery
years to one of two small headwater streams (Catchment 54[C54]) from short-term toxic chemical pulses can be quite rapid compared
at the Coweeta Hydrologic Laboratory. North Carolina. Inver- with long-term physical or chronic disturbance.
tebrate fauna colonizing litterbags and litter processing rates in
the treatment and the reference stream (C55) were examined prior
to and during treatment and for two recovery years. During Introduction
treatment, leaf processing rates in C54 were very low (g50%
that of the average of pretreatment years) and invertebrate fauna Pesticides are among the major anthropogenic distur-
in C54 was dominated by large numbers of small collector- bances to stream ecosystems (WIEDERHOLM 1984). Pe-
gatherers (primarily non-insects) and predators, whereas insect
sticides are often applied directly to streams and lakes to
shredders were virtually eliminated. During the first recovery year,
populations of non-insect taxa in C54 remained high and many reduce noxious species such as the larvae of black flies
insect taxa, especially those with long life cycles, remained absent and other nuisance insects of aquatic origin as well as
or rare. Reappearance of the insect community, especially insect certain fishes (WALLACE & HYNES 1981; MUIRHEAD-THOM-
shredders, during the second year recovery was accompanied by SON 1987). Pesticides applied for pest control in agricultural
restoration of leaf processing rates in C54. or forested areas can also contaminate streams (MuiR-
During the second year of recovery, taxa richness, abundance, HEAD-THOMSON 1987). In either case, pesticides alter the
standing stock biomass. and functional group structure of inver- structure and function of stream ecosystems by reducing
tebrate communities colonizing litterbags in C54 became similar species diversity, modifying food chains, and changing
to those of the reference stream: however, large differences existed patterns of energy flow and nutrient cycling (HURLBERT
between dominant taxa in treated and untreated streams. Follo- 1975; PIMENTEL & EDWARDS 1982).
wing cessation of treatments, taxa having vagile aerial adults and Recovery in aquatic systems has received less study than
those surviving insecticide treatments dominated the litterbag
the effects of disturbances. Risk assessment typically has
communities in the recovering stream (e.g., Lepidostoma andTipu-
la in the shredder functional feeding group; Lanthus and Cerato- concentrated on the probabilities of exposure and effects
pogonidae in the predator functional feeding group). This pattern but rarely on the relative rate of recovery following
was similar to recovery process observed several years earlier in disturbances. However, recovery is of comparable impor-
an adjacent stream that received a similar insecticide manipula- tance, since those communities which can rebound rapidly
tion. The close agreement in the recovery process of macroinver- from disturbances are at less risk than those requiring
tebrate communities in these streams appeared to result from the extended recovery periods (YouNT&NiEMi 1990a). Fac-
similarity of disturbance and the proximity of numerous headwa- tors which influence recovery of invertebrate communities
ter streams within the Coweeta basin, which can provide stable following disturbance include: (1) the proximity of source
7 timnoloeica 23 2 Limnologica23(1993)2 93
populations; (2) conditions of the habitat following distur- Study Sites
bance; (3) the timing of disturbance relative to the life
history stage of organisms; (4) the presence of survivors This study was conducted at the Coweeta Hydroiogic Labora-
from disturbance; (5) vagility of organisms; and, (6) tory (CHL), a 1626-ha drainage basin located in the Nanta-
stream size and relative location in the drainage network hala Mountain range of western North Carolina, within the
(headwater vs. larger downstream areas) (see CAIRNS & Blue Ridge Physiographic Province, latitude 35C03' N. longitude
DICKSON 1977; GUSHING & GAINES 1989; WALLACE 1990). 83°25'W (SWANK & CROSSLEY 1988). Generally, precipitation
increases with elevation along the east-west axis of Coweeta
Recovery differs for different types of disturbances. Physi-
valley, and is distributed fairly evenly throughout the year, with
cal disturbances to catchments (e.g., logging and road late summer and fall being the drier months (SwiFT et al.
building) affect stream biota by altering energy inputs 1988).
and/or habitat quality (WEBSTER et al. 1983). Generally,
recovery of stream fauna from such catchment distur-
bances is linked to the recovery of the surrounding Table 1. Physical parameters of study streams at Coweeta Hy-
terrestrial component (YouNT&NiEMi 1990b). drologic Laboratory. N.C.. Elevations were measured at the
YOUNT&NIEMI (1990b) noted that it is often difficult gauging flumes.
to separate effects and recovery from normal variation
since few studies have contrasted recovery rates estimated C54 CSS
using both before and after data on the same site. Data
from disturbed and undisturbed sites are needed to judge Area (ha) 5.5 7.5
the validity of this procedure (YouNT & NIEMI 1990b). Elevation (m a.s.l.) 841 810
This study is part of a project which used an insecticide Channel
to examine and quantify the role of invertebrates in organic Length (m) 282 170
matter dynamics in headwater streams at the Coweeta Bankful area (m 2 ) 443 373
Hydrologic Laboratory, North Carolina (WALLACE et al. Gradient (cm/ml 33 20
1982; CUFFNEY et al. 1990). Insecticide treatments caused Annual Discharge (m 3 )
massive invertebrate drift (WALLACE et al. 1989). sub- 1985 29190 30158
stantial decrease in both leaf litter processing rates (CUFF- 1988 23036 21974
NEY et al. 1990) and concentration of fine paniculate 1989 62499 74058
organic matter (FPOM) (CUFFNEY et al. 1990; WALLACE 1990 64802 79803
et al. 199la). During the years of insecticide treatments,
Annual degree-days
invertebrate community structure shifted from one domi- 1985 4608 4695
nated by insect taxa to one dominated by non-insect 1988 4182 4181
taxa (LUGTHART & WALLACE 1992; WHILES & WALLACE 1989 4250 4331
1992). 1990 4507 4632
This study encompasses a 6-year period in which the
macroinvertebrate fauna associated with litterbags was
examined prior to and during treatment and for two
recovery years. The objectives of this study are: 1) to Two first-order streams used in this study drain Catchments
examine the community structure of invertebrate fauna 54 (C54) and 55 (C55). They are similar in their southern aspect,
altitude, drainage area, discharge, and thermal regime (Table 1).
colonizing litterbags in insecticide treated and reference
Dominant riparian vegetation includes red maple (Acer rubrutn
streams; 2) assess the impact of insecticide treatment on L.), rhododendron (Rhododendron maxima L.), tulip popular
abundance, biomass, and secondary production of major (Liriodendron tulipifera L.), red oak (Ouercus rubra L.), and white
taxa and functional feeding groups; 3) follow the recovery oak (Quercus alba L.). Both streams are heavily shaded by
of litterbag fauna from insecticide treatments; 4) to com- understory rhododendron.
pare litterbag fauna in treated and reference streams; and C54 received three years (1986— 1988) of seasonal insecticide
5) to compare results with those of a similar study (WAL- treatments. This study was conducted during pretreatment (1985).
LACE et al. 1986) conducted in an adjacent catchment 8 the third year of treatment (1988). and two recovery years (1989
years before the present study. The six year study period and 1990) following the cessation of treatment. Treatments
includes both the driest and wettest years for 57-year period resulted in massive drift of insect taxa and a shift in benthic
community structure from one dominated by insect taxa to one
of record (U.S. Forest Service. Coweeta Hydrologic Labo-
dominated by non-insect taxa (LUGTHART & WALLACE 1992).
ratory, data records). Thus, the study provides a rare During the treatment period of C54. litter processing rates were
opportunity to examine the impact of natural (extreme reduced relative to untreated streams ( CUFFNEY et al. 1990) and
discharges) and anthropogenic disturbances (insecticide annual export of fine paniculate organic matter (FPOM) de-
treatment) on the litterbag communities in reference and creased, while export of coarse paniculate organic matter in-
treatment streams. creased (WALLACE et al. 1991 a).
94 Limnologica23(1993)2
Materials and Methods AFDM using taxon specific length-weight regressions (HuRYN
1986. and unpubl.). For salamanders, the larval stages were
1. Leaf litter processing determined by the presence of gills, and snout-vent length was
measured for each specimen.
Red maple and rhododendron leaves were collected at CHL during For most insect taxa. annual production was estimated by the
mid-October prior to each year of study. About 15 g (air dry size-frequency method (HAMILTON 1969). and corrected for cohort
weight) of each species were placed into separate 20 x 35 cm plastic production interval (CPI) (BENKE 1979). CPIs were estimated
mesh bags (mesh size: ca. 5mm). During 16 — 21 December in from a series of histograms representing the percentage distribu-
1984 and in 1987 — 1989, 60 pairs of litterbags (each pair consists tion of each size class in each taxa (CHUNG unpubl.), or were
of one red maple bag and one rhododendron bag) were placed taken from LUGTHART & WALLACE (1992). We used the method
in C54 and CSS. Five pairs of litterbags were collected randomly of KRUEGER & MARTIN (1980) to determine 95% confidence
from each stream at approximately monthly intervals during the intervals for size-frequency production estimates. For chironomid
first half of the year and at six-week intervals during the last half production, non-Tanypodinate Chironomidae were treated as one
of the year. category (gatherer chironomids). and the instantaneous growth
To estimate the handling loss, ten breakage bags of each leaf method (HURYN & WALLACE 1986: HURYN 1990) was used for
species were prepared for each stream. Breakage bags were estimating production. For some taxa. production was estimated
handled identically to the other bags except they were returned by multiplying annual standing stock biomass by production/bio-
to the laboratory and weighed to determine the percent loss by mass (P/B) ratio. The P/B was assumed to be 5 for those taxa
handling. Initial air dry weights of litterbags were then corrected possessing a CPI of ca. 365 days (BENKE 1984). For example,
based on this handling loss. oligochaetes were assumed to have a P/B = 5: however, estimates
Ash free dry mass (AFDM) was obtained by washing litter in of their production may be conservative (see BRINKHURST & COOK
the bag to remove organic and inorganic deposits and macroinver- 1980). For Copepods, a P B of 18 was used (O'DoHERTY 1985).
tebrates. drying (60 CC for 5 days), weighing, ashing (500 °C for For salamanders, the lengths of the larval periods ( BRUCE
approximately 12 h) and reweighing. Exponential processing rates 1988a. b, 1989) were used.
(day" 1 ) were calculated by regressing Ln (% AFDM remaining) Cluster analysis was used to compare the faunal assemblages
on time in days (WALLACE et al. 1982). Leaf processing rates were among years and between streams. Q-mode analysis was conduc-
calculated only from sets of 5 litterbags whose average AFDMs ted on the annual abundance, standing stock biomass, and
exceeded 5% of their initial values. We used leaf processing for production of 22 important taxa in all functional groups. The
the weight loss of leaf litter by any means (physical, microbial chord distance measure with the flexible strategy (P = -0.25)
and animal, etc.). was used for clusters (L.UDWIG & REYNOLDS 1988).
3. Pesticide treatment
2. Macroinvertebrates in red maple
litterbags A 25% emulsifiable concentrate of methoxychlor, (1,1,1-tri-
chloro-2.2-bis[p-methoxyphenyl] ethane) was applied seasonally
Sediments, detritus smaller than 5 x 5 mm. and invertebrates were to C54 during December 1985 through October 1988 at the rate
washed from red maple litterbags and retained on a 125 um- of 10 ppm based on discharge at the flume. Treatment consisted
opening sieve. Invertebrates attached to the bags were removed of spraying the entire stream including stream margins, seeps, leaf
by hand. Invertebrates and other materials were stored in a 6 — 8% bags, and debris dams from the flume to headwater spring seep
formalin solution containing a small amount of Phloxine B dye with two hand sprayers. The initial 4 h treatment (December 1985)
to facilitate sorting macroinvertebrates from debris (MASON & YE- was followed by subsequent seasonal treatments of 2 h every
VICH 1967). Samples were processed through nested 1 mm- and three months. For more details about the treatments and meth-
125 urn-opening sieves. All animals retained on the 1 mm sieve oxychlor residues in the sediments, see WALLACE et al. (1989,
were removed. The sample retained on the 125um sieve was 1991b).
subsampled (1/4 to 1/64 of the original sample) using a sample Seasonal and annual abundances and standing stock bio-
splitter (WATERS 1969) before removing animals. Macroinver- mass were estimated for each taxon. Summer and fall were
tebrates in subsamples were removed by hand with the aid of a combined because of less frequent sampling. Winter was
dissecting microscope ( l O x magnification). Due to time con- represented by January, February and March, spring by April.
straints, macroinvertebrates associated with rhododendron litter- May and June, and summer-fall by all months from July through
bags were not analyzed. December.
Taxonomic and functional feeding group assignments followed Since most litterbags were collected during the year following
that of MERRITT & CUMMINS (1984), or other studies of the benthic litterbag placement (December), 1984—1985 was designated
fauna in CHL (HURYN & WALLACE 1987; LUGTHART & WALLACE as 1985; 1987-1988 = 1988: 1988-1989 = 1989: and 1989 to
1992). Shredders and predators follow the terminology of MER- 1990 = 1990. For C54. 1985 was the pretreatment year and 1988
RITT & CUMMINS (1984). For collector-gatherers and collector- represents the last (third) year of treatment. Some first year (1986)
filterers. we use gatherers and filterers. respectively, throughout effects of treatment on litterbag fauna have been described
this paper. elsewhere (CuFFNEY et al. 1990). The first and second year of
Body lengths of all macroinvertenrates were measured to the recovery of C54 following the cessation of treatment are 1989
nearest mm under a dissecting microscope, then converted to and 1990, respectively.
Limnologica23(1993) 2 95
Results abundances were 64% and 148%, respectively, of the
pretreatment level. Three dipteran taxa (gatherer chirono-
1. Leaf litter processing mids. Tanypodinae. and Ceratopogonidae) dominated
insect abundance for the entire study period, with gatherer
Leaf processing rates in C54 increased rapidly after the chironomids the most abundant (68 — 82% of total in-
cessation of insecticide treatment (Table 2). During 1988 sects). The relative contributions of Tanypodinae and
and 1989, red maple processing rates were 43% and 56%. Ceratopogonidae peaked in 1989, but dropped to pretreat-
respectively, of the average of pretreatment years reported ment levels in 1990. Absolute numbers of Tanypodinae
by CUFFNEY et al. (1990). During the second year of also peaked in 1989, while Ceratopogonidae continued to
recovery (1990), red maple processing rates in C54 were increase in 1990. By 1990, Lepidostoma. a trichopteran
>1.3x greater than the average of pretreatment years shredder, increased to represent 8.5% of all insects.
(CUFFNEY et al. 1990).
The increase in processing rates in C 54 was greater for
rhododendron than for red maple (Table 2). In 1988, | Insects Q Non-insects
rhododendron processing rates were about a 50% of the
average of pretreatment years reported by CUFFNEY et al. C54: 1985
(1990). In 1989. rhododendron processing rate in C54 was C55:1985
similar to the pre-treatment average, and by 1990, 1.7x
of pretreatment average. to
CD C54: 1988 ](Tre)
C55: 1988
Table 2. Red maple and rhododendron leaflitter processing rates CO
K {% • day" ') based on exponential model Ln(Y) = Ln(A) + K.X.
CO C54: 1989
where x is elapsed time in days and Y is the percentage of original £ C55: 1989
dry mass remaining. Values in parentheses are time (days) to 95% oo
loss (1985 data from CUFFNEY et al. 1990). 1988 is the third C54: 1990
treatment vear for C54.
C55: 1990
Year Years without treatment Treated
2000 4000 6000 8000
C54 C55 Individuals/Litterbag
Red maple Fig. 1. Abundances of insects and non-insect macroinvertebrates
1985 -0.0118 (254) -0.0103 (291) in litterbags (individuals • bag' 1 ) in C54 and C55. 1988 is the
1988 Treated -0.0080 (374) -0.0057 ( 2 ) third treatment vear for C54.
58
1989 -0.0074 (403) -0.0075 40
(0)
1990 -0.0174 (172) -0.0090 (332)
Standing stock biomass: The overall annual standing stock
Rhododendron biomass of litterbag fauna in both streams was similar
1985 -0.0030 (985) -0.0046 (651) during the pretreatment year (Fig. 2). Furthermore, stand-
1988 Treated -0.0060 (501) -0.0021 (1454) ing stock biomass in C55 was similar among years. During
1989 -0.0047 (636) -0.0030 (1008) the treatment year (1988), standing stock biomass in C54
1990 -0.0079 (378) -0.0055 (548) reduced to 40% of the pretreatment level; however, it
rebounded to 79% in the first year of recovery and
surpassed the pretreatment level by 140% in the second
year of recovery.
2. Macroinvertebrates Insect biomass in C54 during the treatment year was
in red maple litterbags only 18% of the pretreatment level and dominated by
odonates (Cordulegaster and Lanthus) which represented
Abundance: The abundances of insects and non-insect 68% of insect biomass (Fig. 2). In the first year of recovery
macroinvertebrates were similar in both streams during (1989) insect biomass in C 54 was 62% of the pretreatment
the pretreatment year (1985) (Fig. 1). Non-insect macroin- level and significantly lower than that of C 55 for same year
vertebrates markedly increased in both streams in 1988. (Fig. 2). In the first recovery year, dominance of odonates
and peaked during the first year of recovery (1989) in was reduced to 49% of all insect biomass while several
C54. Insecticide treatment during 1988 reduced insect dipteran (gatherer chironomids, Ceratopogonidae and
abundance in C54 to 24% of the pretreatment level. Tipula) and trichopteran (Dolophilodes and Lepidostoma)
Recovery of insects in C54 during subsequent years was taxa attained biomass similar to or higher than those of
rapid; in first (1989) and second (1990) years of recovery pretreatment levels (Table 3). In the second year of reco-
96 Limnoloaica23(1993)2
f£] Insects n Non-insect invertes. B Shredder ] Predator
• Salamanders 0 Gatherer Filterer
C54: 1985 ' = - '-•-" I I C54: 1985
C55: 1985 .": •- IB C55: 1985
CO C54: 1988 | | (Ire) CO C54: 1988
> C55: 1988 : CD
TJ I • C55: 1988
C
CO T3
c
C54: 1989 (0 C54: 1989
CO i I
<u C55: 1989 ! • to C55: 1989
55 £
55 C54: 1990
C54: 1990 1 •
C55: 1990 1 I C55: 1990 Y///////M
0 20 40 60 80 100 0 200 400 600
Biomass (mg AFDM) / Litterbag Functional group production
(mg AFDM / Litterbag)
Fig. 3. Annual standing stock biomass and secondary production
C54: 1985 of functional feeding groups (mg AFDM • bag~'). 1988 is the third
C55: 1985 treatment year of C54.
C54: 1988 1 1 (Tre) recovery (1989), secondary production in C54 rebounded
C55: 1988 to 94% of the pretreatment level: however, insect produc-
c
CO tion was still low (59% of the pretreatment level) and was
E C54: 1989
CO C55:1989 less than half that of C55 for the same year. By the second
year of recovery (1990), secondary production in C54
W
C54: 1990 reached 175% of the pretreatment level and became similar
to that of C55 for the same year (Fig. 2). Also, production
C55: 1990 1 of insects in C54 in 1990 slightly surpassed C55 and was
1.6x that of pretreatment levels.
0 100 200 300 400 500 600
Production (mg AFDM) / Litterbag
Functional group production: In both streams prior to
Fig. 2. Annual standing stock biomass and secondary production treatment, secondary production within litterbags was
of insects, non-insect macroinvertebrates. and salamanders (mg distributed primarily among three functional feeding
AFDM • bag" ') in C54 and C55. 1988 is the third treatment year groups: gatherers, predators and shredders (Fig. 3). While
for C54. functional group production in C55 showed only moderate
annual fluctuation, that in C54 changed greatly with the
insecticide treatment. The shredder functional group was
very (1990), insect biomass in C54 exceeded the pretreat- most affected by insecticide treatment. In C54 during
ment level, and was similar to that of C55 for the same treatment, shredder production represented < 1% of total
year (Fig. 2). Some taxa (Cordulegaster, Dolophilodes, production: gatherer and predator groups represented 71
Lepidostoma and Tipula) browed very high biomass during and 28%, respectively, of the total. Shredders and predators
the second year of recovery (Table 3). For example. increased in C54 during the first year of recovery (1989),
Cordulegaster biomass in 1990 was > 9 x of the pretreat- and gatherer production decreased to <60% of total
ment level. production; however, shredder and filterer production
remained low in 1989 but increased substantially in 1990
(Fig. 3).
Production: During the pretreatment year (1985), overall
secondary production within litterbags in C55 was higher • Shredders: Shredder production in C54 in 1985 was
than that of C54 (Fig. 2). From 1988 through 1990. 81% that of C55 (Table 4), but their percentage contribu-
secondary production in C55 ranged from 109% to 136% tion to total production was similar. Shredder production
that of 1985. In C54, litterbag production during treatment in C54 during 1988 was about 1 mg AFDM • bag' 1 , or
was reduced to 53% of the pretreatment level, due to 1% that of C55, 31% of C55 in 1989, but recovered fully
decreases in insect populations (Fig. 2). In the first year of in 1990, attaining 96% that of C 55. In spite of low shredder
Limnologica 23 (1993) 2 97
Table 3. Annual standing stock biomass in mg AFDM bag ' (with 1 s.e.l of dominant taxa of functional feeding groups in C54 and
C55. 1988 is the third treatment year for C54.
Insect 1985 1988 1989 1990
order1'
C54 n = 45 n = 45 n = 50 n = 50
• Filterer
Diplectrona metaqui T <0.1 (<0.1) 0.0 0.1 (0.1) 0.4 (0.2)
Diplectrona modesta T 0.7 (0.2) 0.0 0.1 (0.1) 1.0 (0.3)
Dolophilodes T 0.3 (0.1) 0.0 0.3 (0.1) 1.7 (0.9)
Total 1.3 (0.2) 0.0 0.5 (0.2) 3.3 (1.0)
• Gatherer
Paraleplophlebia E 1.0 (0.3) 0.0 0.1 (0.0) 0.2 (0.1)
Chironomidae* D 4.1 (0.6) 1.9 (0.2) 5.2 (0.6) 6.5 (0.7)
Copepoda 0.9 (0.1) ^ *) (0.3) 5.3 (0.4) 4.1 (0.4)
Oligochaeta 1.9 (0.3) 6.0 (1.0) 6.0 (1.0) 4.4 (1.2)
Other gatherers 0.8 1.0 0.8 0.5
Total 8.7 (0.9) 12.1 (1.3) 17.4 (1.4) 15.7 (1.7)
• Predator
Cordulegaster 0 0.6 (0.2) 2.0 (0.7) 3.3 (0.9) 6.0 (2.1)
Lanthus O 9.5 (2.3) 4.6 (1.2) 12.9 (2.1) 11.0 (1.7)
Beloneuria P 1.1 (0.4) 0.0 0.0 (0.0) 0.3 (0.3)
Ceratopogonidae D 3.3 (0.5) 0.5 (0.2) 3.7 (0.9) 9.9 (1.3)
Tanypodinae D 0.3 (0.1) 0.3 (0.1) 1.1 (0.1) 0.7 (0.1)
Turbellaria 1.0 (0.2) 3.8 (0.6) 1.2 (0.3) 1.4 (0.3)
Other predators 11.5 0.6 1.9 9.7
Total 27.3 (3.4) 11.8 (1.6) 24.1 (2.8) 39.0 (4.5)
• Shredder
Leuctra P 2.4 (0.5) <0.1 (<0. 1) 1.2 (0.3) 0.9 (0.3)
Peltoperlidae P 11.2 (2.6) <0.1 (<0.1) <0.1 (<0.1) 1.7 (0.7)
Fattigia T 2.3 (0.8) 0.0 0.0 <0.1 (<0.1)
Lepidostoma T 1.3 (0.3) 0.1 (<0.1) 1.6 (0.3) 10.5 (1.8)
Pycnopsyche T 4.4 (1.6) 0.0 0.0 2.2 (0.8)
Tipula D 1.7 (0.5) <0.1 ( <0.1) 2.4 (1.1) 9.1 (2.6)
Other shredders 0.3 <0.1 <0.1 0.1
Total 23.6 (3.1) 0.2 (0.1) 5.2 (1.2) 24.5 (3.8)
CSS n = 44 n = 45 n = 50 n = 50
• Filterer
Diplectrona metaqui T 0.0 <0.1 (<0.1) 0.0 <0.1 (<0.1)
Diplectrona modesta T 1.5 (0.3) 1.5 (0.4) 2.1 (0.5) 3.0 (0.5)
Dolophilodes T 0.8 (0.2) 2.1 (0.5) 1.3 (0.3) 1.4 (0.3)
Total 2.3 (0.4) 3.7 (0.7) 3.5 (0.6) 4.4 (0.6)
• Gatherer
Paraleplophlebia E 1.9 (0.3) 4.3 (0.9) 2.3 (0.4) 2.7 (0.6)
Chironomidae* D 4.3 (0.5) 6.7 (0.9) 5.7 (0.7) 6.7 (0.7)
Copepoda 1.3 (0.2) 4.3 (0.5) 2.9 (0.3) 3.3 (0.3)
Oligochaeta 0.9 (0.2) 1.8 (0.6) 2.9 (0.9) 2.8 (0.9)
Other gatherers 1.4 1.0 0.9 1.9
Total 9.8 (0.9) 18.1 (1.8) 14.7 (1.2) 17.4 (1.6)
• Predator
Cordulegaster O 0.7 (0.4) 0.9 (0.2) 0.6 (0.1) 0.8 (0.5)
Lanthus 0 5.3 (1.0) 9.4 (1.9) 7.9 (1.7) 5.3 (1.0)
98 Limnologica 23 (1993) 2
Table 3. (continued)
Insect 1985 1988 1989 1990
order''
CSS n = 44 n = 45 n = 50 n = 50
Betoneuria P 4.1 (1.6) 2.7 (0.8) 2.6 (0.8) 5.0 (1.2)
Ceratopogoniade D 4.2 (0.8) 5.5 (0.5) 6.6 (1.1) 7.1 (1.7)
Tanypodinae D 0.5 (0.1) 1.0 (0.2) 0.7 (0.1) 0.8 (0.1)
Turbellaria 1.4 (0.2) 1.3 (0.4) 0.7 (0.2) 0.6 (0.1)
Other predators 11.7 14.7 14.1 11.1
Total 27.9 (3.4) 35.5 (3.7) 33.2 (4.3) 30.7 (3.5)
• Shredder
Leuctra P 4.0 (0.7) 5.1 (0.9) 6.7 (1.1) 8.1 (1.3)
Peltoperlidae P 16.8 (4.6) 2.6 (0.6) 3.2 (0.7) 9.6 (1.9)
Fattigia T 2.0 (0.5) 3.1 (0.7) 1.4 (0.4) 0.9 (0.3)
Lepidostoma T 1.4 (0.2) 1.1 (0.2) 1.6 (0.3) 1.6 (0.3)
Pycnopsyche T 1.6 (0.5) 0.8 (0.3) 3.2 (0.9) 4.8 (1.2)
Tipula D 2.5 (0.6) 2.8 (0.9) 3.3 (1.0) 3.5 (1.0)
Other shredders O.i 0.3 0.6 0.6
Total 28.4 (5.1) 15.8 (1.5) 20.0 (2.2) 29.1 (2.7)
": D: Diptera: E: Ephemeroptera: O: Odonata: P: Plecoptera: T: Trichoptera
*: Chironomidae exclusive of Tanypodinae
population during the treatment year, three taxa {Leuctra. in both streams in 1988. due to increased abundances of
Lepidostoma and Tipula) showed strong return during the copepods. Gatherer production in C54, primarily non-
first year of recovery (1989) and represented 98% of total insects, increased continuously and exceeded that of C55
shredder production in C 54 (Table 4). In 1990, production by 1989. Production of insect gatherers, rebounded sharply
of Lepidostoma and of Tipula was high, and represented in C54 during 1990 and exceed that of C55.
>88% of total shredder production in C54. Other major Chironomids were the only insects that contributed
shredder taxa, Leuctra, Peltoperlidae and Pycnopsyche, significantly to gatherer production in C54 in 1988 and
displayed low production relative to 1985. In a similar 1989. Gatherer chironomids in C54 represented 22% of
insecticide manipulation on an adjacent stream (C53), total gatherer production in 1988 and 24% in 1989, and
shredder production in the second year of recovery of C 53 >54% in 1990. Non-insect gatherers in C54 contributed
during 1982 was also dominated by Lepidostoma and only 28% to gatherer production during pretreatment
Tipulaf\vhich represented 91 % of total shredder production versus 78% and 72% in 1988 and 1989. This was due
(CuFFNEY et al. 1990). primarily to increases in copepods and oligochaetes.
During the second year of recovery (1990), production of
• Filterers: During pretreatment, filterer production in copepods and oligochaetes decreased in C54. In C55,
C54 represented a minor portion of litterbag production production of insect gatherers exceeded that of non-insects,
and was 43% that in C55 for the same year. In C55 during as non-insects in C55 represented 35 — 43% of gatherer
1988 through 1990. filterer production ranged from 159% production.
to 185% of 1985 level. In C54, filterers were not present
in bags during treatment; however, filterer production • Predators: In 1985, predator production was similar in
rebounded rapidly. During the first year of recovery both streams (Table 5). With some annual variations in
(1989), Diplectrona metaqui and Dolophilodes production the production of individual taxa, total predator produc-
was similar to that of 1985 (Table 5). In 1990, production tion in C55 was more consistent among years than that of
of these two taxa was 11 — 19 x greater than in 1985. and C54. Ceratopogonidae and Lanthus were the most
Dolophilodes was especially important to overall insect productive predators in litterbags of C55. Following
production. D. modesta, a dominant taxon before in- treannent, predator production in C54 declined to about
secticide treatments, showed relatively low production in a half of that of pretreatment (1985), approximated that of
1989. however, its production in 1990 was 2 x that of 1985. pretreatment during the first year of recovery, and was 1.4 x
pretreatment during the second year recovery, exceeding
• Gatherers: In 1985, gatherer production in both streams that of C55. Lanthus and Cordulegaster dominated
was lowest for all years (Table 5). Gatherers were higher predator production in C54 during treatment. Ceratopo-
Limnologica23(1993) 2 99
Table4. Estimates of annual secondary production of shredder group insects in C54 and C55. in mg AFDM -bag ' (±95%
confidence interval). 1988 is the third treatment vear for C54.
Inse. . 1985 1988 1989 1990
order
C54
Leucira P 19.6 ± 4.7 0.0 6.3 ± 2.0 6.5 ± 2.8
Peltoperlidae P 34.3 ± 7.4 0.1 0.0 5.0 ± 3.4
Fanigia T 5.4 ± 2.1 0.0 0.0 0.0
Lepidostoma T 12.5 + 3.2 0.7 ± 0.3 19.1 ± 6.7 99.8 + 36.4
Pvcnopsvche T 26.2 ± 12.1 0.0 0.0 6.3 ± 3.3
Tipula D 12.2 ± 4.7 0.1 8.3 ± 3.9 40.3 ± 19.2
Others* 0.9 0.4 0.6 0.3
Total 111.1 1.3 34.3 158.2
CSS
Leucira P 29.1 ± 4.5 36.2 ± 8.9 40.9 ± 7.5 39.5 + 9.5
Peltoperlidae P 66.2 + 12.2 13.9 + 5.3 14.2 ± 4.8 49.6 ± 11.5
Fanigia T 5.1 ± 2.0 6.1 ± 2.0 2.9 ± 1.3 1.9 ± 1.0
Lepidoswma T 12.7 ± 3.1 16.2 ± 6.5 17.4 ± 5.6 15.2 + 5.0
Pycnopsyche T 6.7 + 3.8 3.9 ± 1.7 18.1 + 6.5 26.7 ± 9.5
Tipula D 16.5 + 5.4 16.2 ± 7.1 13.3 + 6.4 27.1 ± 12.1
Others* 0.9 1.7 2.8 4.2
Total 137.1 93.9 109.6 164.2
" : D: Diptera; P: Plecoptera: T: Trichoptera
*: Psiloireta (T), Limonia (D), Molophilus (D)
gonidae and Tanypodinae increased in C 54 following the spring. During the summer-fall of 1989, all filterer and
cessation of treatment (Table 5). Ceratopogonidae produc- most gatherer insects, as well as two predators, Beloneuria
tion increased to 38% of total predator production (highest and Rhvacophila reappeared. During winter of 1990,
among predator taxa) during the first year of recovery (1989) Pycnopsyche, Fattigia and some predatory Plecoptera
and represented 50% of all predator production during recolonized. Most taxa present during pretreatment (1985)
second year recovery. Other predatory insect taxa in C54 reappeared by the end of 1990. Abundances within
exhibited low production in 1990; however, most of these litterbags also showed a seasonal pattern (Fig. 4B).
taxa showed high abundances as early instars in the last However, since copepods dominated abundances, this
half of 1990. Therefore, compared with previous years, pattern largely reflected seasonal changes in their numbers.
their projected recovery during 1991 should have been Except for the very dry year of 1988, biomass in the
strong. In contrast, some predatory taxa, e.g. Alloperla. untreated stream tended to be highest during winter and
did not reappear by 1990. spring when most insect taxa were in their mid to late
Production of the non-insect predators in C54, Acari instars (Fig. 4C). Biomass decreased during the summer-
and Turbellaria, was highest during treatment (1988) fall when leaf bags contained less leaf material, few late
(>43% of predator production) and declined rapidly instars and many earlier instars. During the summer-fall
following the cessation of treatment. Their contribution period of 1989, biomass in C54 exceeded that of the spring.
to predator production decreased to 10.4% and to 5.0% This was attributable to recolonization by larger bodied
for 1989 and 1990, respectively. taxa of insects during 1989. During winter, 1990. two
shredders (Lepidostoma and Tipula) and two odonates
Community structure: The decline in the number of taxa (Lanthus and Cordulegaster) were the most important
in C 54 clearly demonstrated the treatment effects and the contributors to biomass in C54. By spring, 1990, biomass
seasonality of recolonization (Fig. 4A). The number of in C54 approximated that of C55.
taxa in C54 increased during the summer-fall period of Based on cluster analysis, litterbag fauna of C54 and
1988; however, the last treatment in late-October, 1988, C55 was very similar during pretreatment (Fig. 5). For
eliminated many of the recolonizing insect taxa, resulting abundance, biomass, and production of individual taxa,
in reduced taxa abundance during the winter of 1989. C54 was similar in 1988 and 1989, but dissimilar to other
These taxa recolonized during 1989. Dolophilodes. several years. During 1988 and 1989 litterbags were dominated
dipteran predators, and Peltoperlidae reappeared by by a few small insects (e.g., gatherer chironomids) and
100 Limnologica 23 (1993) 2
Table 5. Estimates of annual secondary production in mg. AFDM • bag ' I ±95% confidence interval! of dominant taxa of functional
feeding groups other than shredders. 1988 is the third treatment year for C54.
Insect 1985 1988 1989 1990
order
C54
• Filterer
Diplectrona metaqui T 0.1 + 0.3 0.0 0.4 + 0.4 1.9 ± 1.4
Diplectrona modesta T 4.2 ± 1.8 0.0 0.4 ± 0.3 8.0 + 4.0
Dolophilodes T 1.9 ± 1.0 0.0 1.9 + 0.8 21.6 ± 11.1
Total 6.2 0.0 2.7 32.5
• Gatherer
Paraleptophlebia E 4.3 ± 1.5 0.0 0.1 ± 0.2 2.5 ± 3.5
Chironomidae* D 64.2 26.1 43.4 122.2
Copepoda 17.8 58.2 94.5 72.1
Oligochaeta 9.6 29.8 29.8 21.9
Other gatherers 3.3 7.2 13.5 6.0
Total 99.2 121.3 181.3 224.7
• Predator
Cordulegaster O 0.7 + 0.4 2.3 ± 1.2 5.3 + 2.5 7.2 + 2.5
Lantkus O 31.6 ± 8.1 18.6 ± 6.1 26.7 ± 6.3 28.0 ± 8.1
Beloneuria P 2.7 + 1.5 0.0 0.0 0.1 ± 0.3
Ceratopogonidae D 22.1 ± 5.0 4.0 ± 2.9 32.9 + 8.2 75.9 ± 15.7
Tanypodinae D 2.1 ± 0.4 2.4 ± 0.9 8.4+ 1.5 6.0 ± 1.3
Turbellaria 4.8 19.0 6.1 7.0
Other predators 41.9 3.0 7.3 26.6
Total 105.9 49.3 86.7 150.8
CSS
• Filterer
Diplecirona metaqui T 0.0 <0.1 + <0.1 0.0 <0.1 + <0.1
Diplectrona modesta T 8.8 + 3.7 10.4 ± 4.0 20.5 ± 9.1 15.2 + 3.6
Dolophilodes T 9.0 ± 6.2 18.4 ± 7.6 13.0 ± 4.2 17.5 + 8.3
Total 18.2 29.0 33.8 32.8
• Gatherer
Paraleptophlebia E 10.8 ± 2.7 24.7 ± 8.1 14.5 ± 3.4 17.1 + 6.4
Chironomidae* D 87.9 103.5 75.0 109.9
Copepoda 22.8 77.1 52.3 59.3
Oligochaeta 4.7 9.2 14.7 14.0
Other gatherers 10.5 4.7 4.9 12.3
Total 136.7 219.2 161.4 212.6
• Predator
Cordulegaster O 1.0 + 0.6 0.8 ± 0.3 0.6 ± 0.2 1.0 ± 0.5
Lanthus O 16.9 + 7.5 30.1 + 9.2 17.0 + 4.8 13.5 + 3.9
Beloneuria P 9.5 ± 3.7 4.9+ 1.9 7.5 ± 2.7 12.1 ± 3.4
Ceratopogonidae D 24.8 ± 5.2 36.6 ± 7.8 43.2 + 10.9 42.1 ± 13.2
Tanypodinae D 3.5 + 0.8 7.7 ± 1.9 5.8 ± 1.0 4.4 + 1.1
Turbellaria 6.9 6.7 3.7 2.8
Other predators 36.3 56.6 42.6 44.5
Total 98.9 143.4 120.4 120.4
": D: Diptera: E: Ephemeroptera; O: Odonata; P: Plecoptera; T: Trichoptera
*: Chironomidae exclusive of Tanypodinae
Limnologica23(1993) 2 101
C54 C55 (A) by number
.—054:1985
(A). 60, l_C55:1985
50. 055:1990
.055:1989
40- 55:1988
054:1990
30-
C54:1989
20- •Co 54:1988(Tre)
10 (B) by biomass
10000 -055:1989
(B).
-055:1988
8000 - -055:1985
6000 -
-054:1990
I2 4000 .
z :> -054:1988(1 re)
1 2000 .
0 (C) by production .055:1990
(C). 120 -i -055:1989
100- -055:1988
.055:1985
tn li- 80-
ra < -054:1985
"c en 60- -054:1990
40- -054:1989
U) 00
-054:1988(Tre)
20-
I"
!§ E 1.0 0.5
(U O
co 5 H- H—I I I I I I I I
W S S W S S W S SWSS
F F F F
Chord distance
Fig. 5. Dendrogram resulting from cluster analysis of litterbag
Fig. 4. Changes in litterbag communities in C54 and CSS, (A): data of individual taxa from C54 and CSS. (A): by number of
seasonal total number of taxa ( ----- : 95% confidence interval individuals (B): by annual standing stock biomass (C): by
for seasonal total number of taxa in CSS). (B): seasonal mean production estimates. Clustering was performed by using the chord
number of individuals/bag. (C): seasonal mean standing stock distance measure and flexible strategy (P = —0.25). 1988 is the
biomass (mg AFDM • bag" 1 ). 1988 is the third treatment year third treatment year for C54.
for C54. W: winter, S: spring, SF: summer-fall.
non-insects. By 1990, community structure of C54 was obtained in an earlier manipulation of an adjacent stream
similar to that of C55. Based on abundances and (C53) (WALLACE et al. 1986). Macroinvertebrates in
production, community structure of C54 during 1990 was rhododendron bags were not analyzed; however, increased
similar to that of C55, whereas biomass was more similar rhododendron processing during recovery is probably
to C54in 1988 and 1989. attributable to the feeding behavior of the dominant
shredder. Lepidostoma. in C54. In a laboratory study of
the feeding behavior of Lepidostoma. WHILES (unpublished
Discussion data) found Lepidostoma consumed more rhododendron
than red maple.
Leaf processing
The changes in litter processing rates in C54 coincided Leaf bag macroinvertebrate community
with changes in litterbag fauna, especially insect shredders.
Following treatment, processing rates of rhododendron During treatment invertebrate population in litterbags
increased more than red maple. Similar results were were dominated by non-insects. In contrast, insects domi-
102 Limnologica23(1993) 2
(WALLACE et al. 1991 b). Although FPOM can be generated with disturbances that alter the physical environment or
by mechanisms other than shredder feeding (ANDERSON energy resources, the insecticide treatment was short-
& SEDELL 1979), results of this study indicate that lived.
invertebrates play an important role in FPOM genera-
tion.
Recovery of C54 in the present study was similar to the Acknowledgements
earlier study of C53 (WALLACE et al. 1986). A period of
8 years separated the two studies and streams experienced We thank B. CANNAMELA. Drs. T. CUFFNEY. A. D. HURYN. and
severe fluctuations in discharge during the 8-year interval. G. J. LUGTHART for assistance with field and laboratory work.
The similarity in recovery of both streams is undoubtedly We also thank G. B. CUNNINGHAM. Dr. W. T. SWANK and the
due to the similarity in source populations of nearby staff of the Coweeta Hydrologic Laboratory for assistance and
streams. A number of studies have found year to use of their facilities. This paper benefitted from the comments
year changes in populations (e.g.. TOWNSEND & SCHOFIELD of Dr. JUDY MEYER. The research was supported by grants
1987); however, in C55 extreme fluctuations in precipita- BSR83-16082 and BSR87-18005 from the National Science
Foundation.
tion did not result in large population shifts within
litterbags. Also, extreme precipitation did not influence
similarity of stream recovery processes in C53 and C54.
Within the Coweeta basin, diverse stream habitats in References
proximity provide an abundant array of source popula-
tions for recolonization. In contrast, in endorheic spring ANDERSON. N. H. & SEDELL, J. R. (1979): Detritus processing by
streams of cold desert regions, where streams are small macroinvertebrates in stream ecosystems. Ann. Rev. Entomol.
and widely separated, recolonization is unpredictable (e.g., 24: 351-377.
GUSHING & GAINES 1989). BENKE. A. C. (1979): A modification of the Hynes method for
estimating secondary production with particular significance
Given the small size of disturbed area, the recovery of for multivoltine populations. Limnol. Oceanogr. 24:168 — 171.
C 54 was relatively slow. However, recovery in other studies — (1984): Secondary production of aquatic insects. In: V. H.
was rapid where immediate source populations were pro- RESH & D. M. ROSENBERG (eds.). The ecology of aquatic in-
vided. For example, recovery of benthic fauna completely sects, pp. 289-322. New York.
eradicated by a chemical spill over several hundred kilo- BRINKHURST, R. O. & COOK, D. G. (eds.) (1980): Aquatic Oligo-
meters in the Rhine River occurred within a year, although chaete Biology. New York.
the pre-accident ecosystem was not diverse (CAPEL et al. BRUCE. R. C. (1988a): An ecological life table for the salamander
1988). Recovery was also rapid in a warm desert stream, Eurycea wilderae. Copeia 1988: 15 — 26.
— (1988b): Life history variation in the salamander Desmo-
where unpredictable and catastrophic floods eliminated gnathus quadramaculalus. Herpetologica 44: 218 — 227.
most of macro invertebrate fauna. FISHER et al. (1982) and — (1989): Life history of the salamander Desmognathus montico-
GRAY & FISHER (1981) attributed fast recovery of desert la. with a comparison of the larval periods of D. monticola
streams to the aerial colonization by long lived adults, or and D. ochrophaeus. Herpetologica 45: 144—155.
by continuous emergence throughout a year. Similar rapid CAPEL. P.O.. GIGER. W., REICHERT, P. & WANNER. O. (1988):
recovery of macroinvertebrate assemblages in other warm Accidental input of pesticides into the Rhine River. Environ.
desert streams has been observed following flash flooding Sci. & Technol. 22:992-997.
(MEFFE & MINCKLEY 1987). Thus, presence of source po- CAIRNS, J.. Jr. & DICKSON, K. L. (1977): Recovery of streams
from spills of hazardous materials. In: J. CAIRNS, Jr.. K. L.
pulations for recolonization is a very important factor to DICKSON & E. HERRICKS (eds.). Recovery of restoration of
the rapid recovery of a disturbed stream. damaged ecosystems, pp. 24—42. Charlottesville.
In evaluating a disturbance, both immediate and CUFFNEY. T. F.. WALLACE. J. B. & LUGTHART. G. J. (1990): Ex-
long-term impacts should be considered. Methoxychlor perimental evidence quantifying the role of benthic inver-
treatment did not alter the physical environment or energy tebrates in organic matter dynamics of headwater streams.
resources, especially allochthonous inputs and associated Freshwater Biol. 23: 281-299.
microbes (CLTFFNEY et al. 1990). In contrast, nearby — — & WEBSTER. J. R. (1984): Pesticide manipulation of a
low-order streams that received catchment-wide physical headwater stream: invertebrate responses and their sig-
disturbances such as logging, altered both the physical nificance for ecosystem process. J. N. Am. Benthol. Soc. 3:
153-171.
environment and energy inputs in a stream (WEBSTER et
CUSHING. C. E. & GAINES. W. L. (1989): Thoughts on recoloniza-
al. 1983). WALLACE et al. (1988) observed that 5 years tion of endorheic cold desert spring-streams. J. N. Am.
after logging of a nearby catchment, there was no evidence Benthol. Soc. 8: 277-287.
that the stream fauna of the logged catchment was FISHER. S. G.. GRAY. L. J.. GRIMM. N. B. & BUSCH. D. E. (1982):
converging, taxonomically or functionally, toward that of a Temporal succession in a desert stream ecosystem following
stream draining an unlogged catchment. Thus, compared flash flooding. Ecol. Monogr. 52: 93-110.
104 Limnologica23(1993)2
nated litterbags during the pretreatment and second reco- recolonization by downstream drift (TOWNSEND & MIL-
very years. Within a year following treatment, insects DREW 1976), or upstream movement by aquatic stages
became dominant, although production of non-insects (S6DERSTROM 1987), was unlikely. Therefore, aerial adult
remained high during this period. Production of non- insects or within-stream survivors are the most likely route
insects decreased during the second year of recovery. The of recolonization in C54 during the treatment and initial
paucity of insect predators may be responsible for the high recovery periods.
production of non-insects during the first year of recovery. The odonates, Lanihus and Cordulegaster. obviously
Non-insect domination of macroinvertebrate communities survived and oviposited during the periods of repeated
during treatment was also observed by CUFFNEY et al. pesticide treatments. These taxa have 2 — 3 year life cycles
(1984, 1990) for litterbags and LUGTHART & WALLACE in the study streams and all size classes were present during
(1992) for benthic samples. the third year of treatment. High survivorship of odonates
Population level and community composition at a site under repeated methoxychlor treatments was also observed
can change with fluctuating discharge (!VERSEN et al. 1978; by CUFFNEY et al. (1984) and LUGTHART & WALLACE
PEARSON 1984; McELRAVYet al. 1989; RAE 1990). During (1992).
high discharge, loss of stream biota is common (McELRAVY The lengths of life cycles also influence rate of recovery.
et al. 1989: RAE 1990), while drought induces movement Gatherer chironomids recolonized the treated stream re-
and concentration of macroinvertebrates into smaller peatedly during the three-year period of seasonal treat-
wetted areas (L.ARIMORE et al. 1959; KAMLER & RIEDEL ments (WALLACE et al. 1991 c) and the population increased
1960). Production of macroinvertebrates in this study was rapidly during the first year of recovery. Adult chironomids
moderately affected by discharge regime. For example, are good fliers (OLIVER 1971) with relatively short life
Coweeta experienced severe drought in 1988. while 1989 cycles (HuRYN 1990). Thus, their early recolonization is
was the wettest year of the 57-year record (cf. Table 1), facilitated by: 1) aerial oviposition from nearby source
followed by 1990. another wet year. Increased production areas and. 2) adults emerging within the treated area
in C55 during in 1988, compared with 1985. may be between periods of consecutive treatments. In contrast.
attributable to taxa concentrating in a reduced wetted area recolonization of insects with long life cycles (uni- or
with declining discharge (LARIMORE et al. 1959). This semi-voltine) and/or poor flying ability was slow. These
would increase population density per unit wetted area include some plecopteran (e.g., Beloneuria, Peltoperlidae)
(or litterbags in this study). With the exception of shred- and large bodied trichopteran (e.g., Fattigia and Pycno-
ders, all functional groups showed increased production psyche) taxa.
between 1985 and 1988 in C55. Although change in the The timing of last treatment (late October, 1988) was
production of each shredder taxon was moderate, Pelto- important to the recolonization of certain taxa. Pycno-
perlidae showed a marked decrease in 1988 (Table 4). It psyche reappeared in the second year of recovery, which
is not clear why Peltoperlidae behaved differently from indicated they hatched in autumn of 1989. WALLACE et al.
other shredders, but it may be partly due to a reduction (1991c) observed a large number of early instar larvae of
in suitable habitat for early instars (moss cover on bed rock) Pycnopsyche in drift samples during the last treatment in
during the continued dry period. Although multiple factors autumn 1988. If the last treatment had been applied before
probably contributed to the low production in litterbags Pycnopsyche hatched in autumn, this taxon would have
in C55 during 1989, expansion of the wetted stream area reappeared during the first year recovery.
during this "wet" year compared with 1988 may have Recovery of taxonomic richness was largely completed
diminished invertebrate density per unit area during 1989. within two years following the cessation of treatment.
However, production in litterbags was very high in 1990, However, dominant taxa in the second year of recovery
another very wet year, despite the increase in wetted differed from those of untreated streams. Despite these
channel area. During 1990, the largest storm in a 6-year differences, functional group recovery, based on abun-
period occurred in mid-February removed much of the dances, biomass and production, occurred prior to taxo-
detritus from the stream bed (personal observation). nomic recovery, due to strong recolonization by a few
When C55 receded to normal flow, detritus was unavaila- members of each functional group. In an earlier treatment
ble to stream macroinvertebrates, making litterbags a of C53 at CHL, functional recovery from insecticide
concentrated source of litter for stream macroinver- treatment, measured as leaf processing rates, was com-
tebrates. pleted within two years following the cessation of treatment
WILLIAMS & HYNES (1976) identified four principal (WALLACE et al. 1986). Functional recovery of C54 during
recolonization pathways for stream benthos: aerial move- the present study was also reflected in the annual export of
ment, downstream drift, upstream movements and vertical FPOM. During the treatment years, annual FPOM export
movement within the stream substrata. C54 was treated per unit discharge was lower in C54 than untreated streams.
from the upstream source to a flume which effectively During recovery, FPOM concentration and export in C54
isolated the treated area from downstream reaches, thus. increased, and was similar to that of untreated streams
Limnologica 23 (1993) 2 103
GRAY. L. J. & FISHER. S. G. (1981): Postflood recoionization PEARSON. R. G. (1984): Temporal changes in the composition and
pathways of macroinvertebrates in a lowland Sonoran desert abundance of the macro-invertebrate communities of the River
stream. Am. Midi. Nat. 106: 249-257. Hull. Arch. Hydrobiol. 100: 273-298.
HAMILTON. A. L. (1969): On estimating annual production. PIMENTEL, D. & EDWARDS. C. A. (1982): Pesticides and eco-
Limnol. Oceanogr. 14: 771 -782. systems. Bioscience 32: 595-599.
HURLBERT. S. H. (1975): Secondary effects of pesticides on RAE. J. G. (1990): The effect of naturally varying discharge rates
aquatic ecosystems. Residue Rev. 57: 81 — 149. on the species richness of lotic midges. Hydrobiologica (The
HURYN. A. D. (1986): Secondary production of the macroinver- Hague) 196: 209-216.
tebrate community of a high-elevation stream in the southern SODERSTROM. O. (1987): Upstream movements of invertebrates in
Appalachian mountains. Dissertation. The University of running waters — a review. Arch. Hydrobiol. Ill: 197 — 208.
Georgia. Athens. SWANK. W. T. & CROSSLEY. D. A. (1988): Introduction and site
— (1990): Growth and voltinism of lotic midge larvae: patterns description. In: W. T. SWANK & D. A. CROSSLEY (eds.). Forest
across an Appalachian Mountain basin. Limnol. Oceanogr. hydrology and ecology at Coweeta (Ecol. Stud. 66), pp. 3—16.
35: 339-351. New York.
— & WALLACE. J. B. (1986): A method for obtaining in situ SWIFT. L. W. J.. CUNNINGHAM. G. B. & DOUGLASS. J. E. (1988):
growth rates of larval Chironomidae (Diptera) and its Climatology and hydrology. In: W. T. SWANK & D. A.
application to studies of secondary production. Limnol. CROSSLEY (eds.). Forest hydrology and ecology at Coweeta
Oceanogr. 31: 216-222. (Ecol. Stud. 66). pp. 35 — 55. New York.
(1987): Local geomorphology as a determinant of
TOWNSEND. C. R. & HILDREW. A. G. (1976): Field experiments
macrofaunal production in a mountain stream. Ecology 68:
on the drifting, colonization and continuous redistribution of
1932-1942.
stream benthos. J. Anim. Ecol. 45: 759 — 772.
IVERSEN. T. M.. WIBERG-LARSEN. P.. BIRKHOLMHANSEN. S. &
HANSEN. F. S. (1978): The effects of partial and total drought - & SCHOFIELD. K. (1987): Persistence of stream invertebrate
on the macroinvertebrate communities of three small streams. communities in relation to environmental variability. J. Anim.
Hydrobiologia (The Hague) 60: 235-242. Ecol. 56: 597-613.
KAMLER. E. & RIEDEL. W. (1960): The effect of drought on the WALLACE. J. B. (1990): Recovery of lotic macroinvertebrate
fauna Ephemeroptera. Plecoptera and Trichoptera of a community from disturbance. Environ. Manage. 14: 605 to
mountain stream. Pol. Arch. Hydrobiol. 8: 87 — 94. 620.
KRUEGER. C. C. & MARTIN. F. B. (1980): Computation of confi- - CUFFNEY. T. F.. GOLDOWITZ. B. S. CHUNG. K. & LUGTHART.
dence intervals for the size-frequency (Hynes) method of G. J. (1991a): Long-term studies of the influence of
estimating secondary production. Limnol. Oceanogr. 25: invertebrate manipulations and drought on paniculate organic
773-777. matter export from headwater streams. Verh. Internal. Verein.
LARIMORE. R. W.. CHILDERS. W. F. & HECKROTTE. C. (1959): De- Limnol. 24: 1676-1680.
struction and re-establishment of stream fish and invertebrates WEBSTER. J. R.. LUGTHART, G. J.. CHUNG. K. & GOLDO-
affected by drought. Trans. Amer. Fish. Soc. 88: 261 —285. WITZ. B. S. (1991b): Export of fine organic particles from
LUDWIG. J. A. & REYNOLDS, J. F. (1988): Statistical ecology. New headwater streams: effects of season, extreme discharge, and
York. invertebrate manipulation. Limnol. Oceanogr. 36: 670 — 682.
LUGTHART. G. J. & WALLACE. L B. (1992): Effects of disturbance - GURTZ. M. E. & SMITH-CUFFNEY. F. (1988): Long-term com-
on benthic functional structure and production in mountain parisons of insect abundances in disturbed and undisturbed
streams. J. N. Am. Benthol. Soc. 11: 138-164. Appalachian headwater streams. Verh. Internal. Verein.
MASON. W. T. & YEVICH. P. P. (1967): The use of Phloxine Band Limnol. 23: 1224-1231.
Rose Bengal stains to facilitate sorting benthic samples. Trans. - HURYN, A. D. & LUGTHART. G. J. (1991c): Colonization of a
Amer. Microsc. Soc. 86: 221 —222. headwater stream during three years of seasonal insecticide
MCELRAVY. E. P., LAMBERTI. G. A. & RESH. V. H. (1989): Year- applications. Hydrobiologica (The Hague) 211: 65 — 76.
to-year variation in the aquatic macroinvertebrate fauna of - LUGTHART. G. J.. CUFFNEY. T. F. & SCHURR, G. A. (1989):
northern California stream. J. N. Am. Benthoi. Soc. 8: 51—63. The impact of repeated insecticidal treatments on drift and
MEFFE. G. K. & MINCKLEY. W. L. (1987): Persistence and sta- benthos of a headwater stream. Hydrobiologia (The Hague)
bility of fish and invertebrate assemblages in a repeatedly 179: 135-147.
disturbed Sonoran Desert stream. Am. Midi. Nat. 117: - VOGEL, D. S. & CUFFNEY. T. F. (1986): Recovery of a headwa-
177-191. ter stream from an insecticide-induced community distur-
MERRITT. R. W. & CUMMINS. K. W. (eds.). (1984): An Introduc- bance. J. N. Am. Benthol. Soc. 5: 115-126.
tion to the Aquatic Insects of North America. Dubuque. - WEBSTER, J. R. & CUFFNEY. T. F. 81982): Stream detritus dy-
MUIRHEAD-THOMSON. R. C. (1987): Pesticide impact on stream namics: regulation by invertebrate consumers. Oecologia 53:
fauna with special reference to macroinvertebrates. Cam- 197-200.
bridge. WALLACE. R. R. & HYNES. H. B. N. (1981): The effect of chemical
O'DOHERTY. E. C. (1985): Stream-dwelling copepods: their life treatments against blackfly larvae on the fauna of running
history and ecological significance. Limnol. Oceanogr. 30: waters. In: M. LAIRD (ed.). Blackflies: the future for biological
554-564. methods in integrated control, pp. 237 — 258. New York.
OLIVER. D. R. (1971): Life history of the Chironomidae. Ann. WATERS. T. F. (1969): Subsampling for dividing large samples of
Rev. Entomol. 16: 211-230. stream invertebrate drift. Limnol. Oceanogr. 14: 813 — 815.
Limnologica 23 (1993) 2 105
WEBSTER. J. R., GURTZ. M. E.. HAINS. J. J.. MEYER. J. L.. SWANK. WILLIAMS. D. D. & HYNES. H. B. N. (1976): The recolonization
W. T.. WAIDE.J. B.& WALLACE. J. B. (1983): Stability of mechanisms of stream benthos. Oikos 27: 265 — 272.
stream ecosystems. In: J. R. BARNES & G. W. MINSHALL YOUNT. J. D. & NIEMI. G. L (1990a): Preface. Environ. Manage.
(eds.). Stream ecology: application and testing of general 14: 515-516.
ecological theory, pp. 355 — 396. New York. (1990b): Recovery of lotic communities and ecosystems
WHILES. M. R. & WALLACE.:. B. (1992): First-year benthic reco- from disturbance — A narrative review of case studies. Ibid.:
very of a headwater stream following a 3-year insecticide- 547-569.
induced disturbance. Freshwater Biol. 28: 81—92.
WIEDERHOLM, T. (1984): Responses of aquatic insects to en- Received: 21. 1. 1993
vironmental pollution. In: V. H. RESH & D. M. ROSENBERG
(eds.). The ecology of aquatic insects, pp. 508 — 557. New Address for correspondence: Dr. K. CHUNG, Department of
York. Entomology, University of Georgia, Athens, Ga-30602, USA.
106 Limnologica23(1993)2
Related docs
