AQUATIC CONSERVATION: MARINE AND FRESHWATER ECOSYSTEMS

Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)

Published online in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/aqc.511

Leaf litter breakdown in Patagonian streams: native versus

exotic trees and the effect of invertebrate size

* O* and E. G. BALSEIRO
R. J. ALBARIN
Lab. Limnolog!ıa, Centro Regional Universitario Bariloche, Universidad Nacional del Comahue. Unidad Postal
Universidad, Bariloche, Argentina

ABSTRACT
1. Nothofagus native forest in the Southern Andes is being progressively substituted by forestation
with rapid growth of the exotic trees, mainly species of Pinaceae. The effect on stream processing
dynamics is explored through in situ experiments.
2. The effects of leaf litter quality and macroinvertebrate size on in situ litter breakdown were
analysed. Experiments were run in litter bags which allowed access of macroinvertebrate fauna in
streams running through a dense forest of the deciduous Nothofagus pumilio.
3. In Experiment 1, the decay rates of N. pumilio leaves and Pinus ponderosa needles were
measured during an autumn-winter period. N. pumilio decayed twice as fast as P. ponderosa
(P50.01). Shredders fed only on N. pumilio leaves. The total abundance of macroinvertebrates
colonizing both treatments was similar; however, the biomass was higher in the N. pumilio treatment.
Large shredders were only found colonizing N. pumilio leaves. Since no decay due to shredders was
observed in P. ponderosa, the presence of macroinvertebrates in these litter bags was related to refuge
and feeding on FPOM-biofilm resources.
4. In Experiment 2, N. pumilio leaf litter was exposed in order to allow (open bags) or restrict
(closed bags) access of invertebrates. The invertebrate assemblage in open bags showed the similar
pattern observed for N. pumilio in the first experiment. Gathering-collectors were generally smaller
and dominant in number while shredder biomass was higher in open bags as a result of high
individual biomass. N. pumilio decayed faster when the whole size spectrum of macroinvertebrates
colonized the bags (P50.01). However, feeding signs of small shredders were observed in closed
bags, therefore their role on leaf litter breakdown in Andean streams should not be neglected.
5. The combined results of both experiments lead to the conclusion that the whole litter
processing mechanism would be affected as a consequence of the substitution of native forest by
exotic pine forestation.
Copyright # 2002 John Wiley & Sons, Ltd.

KEY WORDS: litter decay; streams; deciduous Nothofagus pumilio; shredders; Andes; Pinus ponderosa

*Correspondence to: Lab. Limnolog!ıa, Centro Regional Universitario Bariloche, Universidad Nacional de Comahue. Unidad Postal
Universidad, 8400 Bariloche, Argentina. E-mail: [email protected]

Copyright # 2002 John Wiley & Sons, Ltd. Received 30 April 2001
Accepted 5 January 2002
182 R. J. ALBARIÑO AND E. G. BALSEIRO

1. INTRODUCTION

Native forests in Patagonia cover the headwaters of all the Pacific and Atlantic basins. They extend on both
sides of the western Andes range between 398S and 558S. As in other areas of the world, land use has
gradually changed the native woodland. Agriculture, and wood extraction for heating for the paper and
construction industries have been important historically mainly in the Western Andes (Chile) (Donoso and
Lara, 1996). However, forestry has become increasingly relevant from the 1950s, particularly by the
replacement of woodland with rapid growth species such as pines and eucalyptus (Lara et al., 1996). The
Pacific native forests have been severely altered, reaching up to 60% of woodland substitution with
monocultures of exotic species in several regions of Southern Chile between 1978 and 1987 (Lara et al.,
1996). On the other hand, the Atlantic slope, with about 2,600 km2 of native woodland, has suffered
considerably less impact. In Argentina, however, forestry is promoted and grant-aided by the Federal
Government of Argentina (Postler et al., 1999). Up to 1998, the Andes eastern slope (Argentina) was
forested with about 500 km2 of pine plantations and it was estimated that this would reach about 800 km2 at
the present (Schlichter and Laclau, 1998). Including both the Andean (woodland) and the extra-Andean
(ecotone woodland-steppe) areas of Argentina, these authors mentioned that between 7000 and 20 000 km2
are suitable for pine introduction.
Forest substitution in the Andes of Patagonia has mainly produced ecological consequences in the same
terrestrial ecosystem, such as lowering biodiversity both of plants and vertebrates (Lara et al., 1996; Muru! a,
1996; Rozzi et al., 1996). Because riparian vegetation is linked to instream processes, it is expected that its
alteration cause changes in aquatic systems (Cummins et al., 1989). These changes may influence food
resources for stream communities by altering the input timing, standing crop or food quality of detritus
(Sedell et al., Cummins et al., 1989; Murphy and Giller, 2000). Thus, it has been suggested that debris
processing in streams has been affected by the introduction of Eucalyptus in Portugal (Abelho and Graca,
1996) and by extensive plantations of Pinus radiata in Australia (Campbell, 1994). In-low order forested -
streams, allochthonous organic matter is a major source of food and refuge for aquatic invertebrates
(Smock et al., 1989; Prochazka et al., 1991; Dobson, 1994). Invertebrates inhabiting these organic patches
are mainly detritivores belonging to the shredder and gathering-collector guilds (Cummins et al., 1989) and
both groups are predicted to coexist in headwaters (Vannote et al., 1980). Biological processes may account
for up to 70% loss in leaf litter mass in a stream patch (Cummins, 1996) with shredder feeding activity as a
significant contributory factor (Petersen and Cummins, 1974; Cuffney et al., 1990; Stewart, 1992; Robinson
et al., 1998). Coarse detritus-feeders are generally large invertebrates adapted to shred and feed on large
plant debris. However, smaller invertebrates may also be important in the breakdown of coarse particulate
organic matter (CPOM) (Short et al., 1980; Graça et al., 1993; Dieterich and Anderson, 1995).
In Andean canopied headwaters, the Nothofagus temperate forest is the most important source of organic
matter (Modenutti et al., 1998). The genus Nothofagus is endemic to the Southern Hemisphere (Allan, 1982)
and deciduous species are restricted to South America, where six tree species develop in the temperate
forests. Among these, N. pumilio has the broadest distribution, from 398 to 558S forming a low temperature
treeline at higher elevations (Barrera et al., 2000), a feature indicated as distinctive of the Southern Andes
(Wardle, 1998). Until now, Nothofagus decay has only been studied in New Zealand evergreen tree species
(Davis and Winterbourn, 1977; Rounick and Winterbourn, 1983; Linklater, 1995) and recently in Southern
Argentina (Albarin* o and Balseiro, 1998, 2001). Over the last 30 years, exotic species of Pinaceae have been
introduced into the Atlantic slope of the Southern Andes (Schlichter and Laclau, 1998). However, the
impact of such activity on aquatic systems has not yet been evaluated. Although Fagaceae have been
reported to have very low decay rates, Pinaceae attain the lowest known values (Webster and Benfield,
1986). Therefore, substitution of native woodlands with Pinaceae forest would affect instream benthic
CPOM-dynamics by altering both trophic assemblages and litter decay rates, since shredders would be less
able to process such new debris.

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
 LEAF LITTER BREAKDOWN IN PATAGONIAN STREAMS 183

Using in situ experiments, this study attempts to analyse the effect of substituting Nothofagus pumilio
litter by Pinus ponderosa needles in low order South Andes streams. A subsequent experiment explores the
effect of excluding large shredders on Nothofagus pumilio leaf litter decay.

STUDY AREA

The experiments were carried out in the upper section of the N * ireco catchment (418080 S; 718170 W).
This basin drains an area densely covered by a forest of N. pumilio (called ‘lenga’) a species of deciduous
beech that dominates the vegetation and forms the highest mountain belt of the temperate forest up
to the timberline in the Patagonian Andes. The basin has a pluvio-nival hydrological regime with
high discharge in autumn (rain) and spring (snowmelt) and low discharge in late summer. N * ireco stream
discharges range from 2300 L/s
1 (mean maximum } November) to 230 L/s
1 (mean minimum }
March).
Two different experiments were carried out, Experiment 1 in the main channel of N * ireco stream and
Experiment 2 in an unnamed tributary called hereafter A1. Experiments were performed in riffle areas of
both streams where the substrate was dominated by boulders and cobbles. Chemical characteristics were
similar: conductivity 48.8 mS cm
1, pH neutral to slightly basic (7.7), and dissolved oxygen concentration
always at saturation levels. The main differences between both streams are order and width since N * ireco
stream is a second-order stream and 4 m wide, while A1 is a first-order stream and 1 m wide.

MATERIALS AND METHODS

Experiment 1
This experiment was conducted in N * ireco stream during autumn–winter 1995. Litter bags were used to
measure processing rates and macroinvertebrate colonization for leaf litter of both a native and an exotic
tree species. Nothofagus pumilio leaves range from 20 to 40 mm in length and bags are more adequate than
other methods when used with small leaves (Davis and Winterbourn, 1977). Litter bags of 18  12  2 cm
were constructed combining two mesh sizes 10 mm (15% of bag surface) and 1 mm (remaining 85%) } and
placed in the substrate with the large mesh size upstream. Thus, it was assumed that large
macroinvertebrates could gain access to leaves and also that leaf fragments (>1 mm) produced by the
breakdown processes would not be washed out. Physical abrasion was assumed to be negligible because of
the physical protection provided by bags. Therefore, the loss of leaf mass inside the bags was attributed
only to leaching, microbial decomposition and macroinvertebrate activities. Undamaged freshly fallen
leaves of N. pumilio were collected from the riparian zone, and needles of P. ponderosa were obtained from
a pine plantation nearby, both collections being made on 14 May 1995 (Southern autumn). Two treatments
were performed: one with N. pumilio leaves (Np) and another with P. ponderosa needles (Pp). Each bag was
filled with fresh leaves of one species equivalent to 4.00 g DM (dry mass) weighed to the nearest 0.01 g. The
amount of leaf litter was 50% greater than the estimated CPOM mass m
2 for this stream (R. Albarin* o,
personal observation). Leaves used in experiments were not artificially dried to avoid overestimation of
decay rates (Gessner and Dobson, 1993). Leaf dry mass in the mesh bags was estimated by the following
regressions:
DMNp1 ¼ 0:50 FMNp1 þ 0:19 ðr2adj ¼ 0:99; n ¼ 14Þ

DMPp ¼ 0:84FMPp þ 0:03ðr2adj ¼ 0:99; n ¼ 14Þ

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
184 R. J. ALBARIÑO AND E. G. BALSEIRO

where FM is the fresh mass of leaf or needle. Dry mass was determined by drying leaf material at 608C for
48 h and weighing to the nearest 0.01 g. During that time, fresh leaves were stored in a refrigerator at 58C in
dark conditions. Stored leaves lost about 5% of their initial fresh mass (probably due to water loss) and this
factor was included in the estimation of fresh mass in the bags.
On 17 May 1995 (t0), 20 bags of each leaf species were introduced in an 80 m-reach of N * ireco stream.
Each bag was ballasted with a small cobble inside and placed randomly in one of four riffle areas. Five bags
(i.e. five replicates) of each treatment were removed after 2, 16, 42 and 118 days of exposure (t2, t16, t42 and
t118) and returned to the laboratory.
In the laboratory, macroinvertebrates were separated from remaining leaf material by rinsing. Leaf
remaining dry mass (LRDM) for each bag was obtained by drying at 608C for 48 h. Macroinvertebrates
were taxonomically identified and assigned to their corresponding functional feeding groups (FFGs)
according to Merritt and Cummins (1996): shredder, gathering-collector, filtering-collector, scraper and
predator. Taxa were assigned to each FFG by a combination of methods (mouthpart morphology, gut
content analysis and/or feeding experiments) (Albarin* o and Valverde, 1998; Albarin* o, 1999; D!ıaz
Villanueva and Albarin* o, 1999; Albarin* o, 2001). Some taxa were assigned using the classification of Merritt
and Cummins (1996) and related literature. In the present analysis only detritivore groups associated with
leaf litter } shredders and gathering-collectors } were considered. Following Cummins and Merritt (1996),
early instars of shredders were assumed to be gathering-collectors. Macroinvertebrates were quantified both
in number and biomass and relative values were referred to as grams of LRDM for each experimental unit.
Biomass was obtained by drying organisms at 608C for 24 h and weighing to the nearest 0.01 mg.
An exponential decay model was estimated in both treatments for LRDM versus time:
Wt ¼ W0 e
kt
where Wt is the LRDM at a given time t, W0 is the initial leaf DM, k is the decay rate and t is time in days.
Based on this model t50% was calculated, i.e. the time when the first 50% of CPOM would decay.

Experiment 2
This experiment was conducted in stream A1 during spring–summer 1995–1996. Similar bags to those in
Experiment 1 were used in two treatments in order to test the effect of shredder size on leaf litter breakdown
of N. pumilio. The open bag treatment (OB) consisted of composite bags as in Experiment 1, which allowed
large shredders to colonize. Closed bag treatment (CB) consisted of similar plastic bags but constructed
entirely of 1 mm mesh to restrict the access of large invertebrates.
N. pumilio leaves were collected from stream surroundings on 13 November 1995; therefore they had
overwintered on the ground during the rainy and snowy period. Each bag of both treatments was filled with
the equivalent of 3.50 g DM of leaves, and then ballasted with a cobble. Dry mass of leaf material was
estimated by the following regression equation established using the same procedure as in Experiment 1:
DMNp2 ¼ 0:87 FMNp2 -0:02ðr2adj ¼ 0:99; n ¼ 12Þ
Thirty bags of OB and 15 of CB were placed in the stream on 17 November 1995 (t0). Five bags (i.e. five
replicates) of OB were removed at 7, 20, 32, 47, 89 and 135 days, and five bags of CB (replicates) were
removed after 20, 47 and 135 days of exposure. The same procedure of Experiment 1 was followed to
obtain LRDM, number and biomass of macroinvertebrates from each experimental unit. An exponential
model for each treatment was estimated for LRDM versus. time and t50% was calculated.
In order to test for statistical differences in decay rates (k) between treatments in each experiment, leaf
dry mass data were log-transformed and a regression model against time was obtained. Regression
coefficients were compared using the method of Sokal and Rohlf (1981) and difference was tested at 1%
significance.

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
 LEAF LITTER BREAKDOWN IN PATAGONIAN STREAMS 185

RESULTS

Experiment 1
Water temperature was 3.48C (  0.4; mean  SE). Maximum and minimum temperatures were 4.58C
(May) and 1.08C (June), respectively. At day 2, mass loss due to leaching was 3.8% for N. pumilio and was
not detectable for P. ponderosa. Feeding signs by macroinvertebrates were observed on N. pumilio leaves at
t42 indicating that feeding activity had started after t16. In contrast, needles showed no evidence of having
been fed on throughout the experiment. The decay rate coefficient (k) of N. pumilio was significantly higher
than that of P. ponderosa (P50.01) (Table 1, Figure 1(a)). The estimated t50% for P. ponderosa was twice
that of N. pumilio leaves (Table 1). At day 118, leaves and needles had lost 34% and 19% of their initial
DM (Figure 1(a)). Although bags were placed with the large mesh size upstream, no coarse litter, not
included originally in the bags, was found; this was especially evident in the pine bags, where any other litter
can be easily recognized.
Macroinvertebrates gradually colonized the litter bags and maximum mean densities (49 and 56 ind. g
1
LRDM for Np and Pp, respectively) were registered at day 118 (Figure 1(b)). However, macroinvertebrate
biomass was higher at day 42 for Np (7.4 mg DM ind. g
1 LRDM) and at day 118 for Pp (5.1 mg
DM ind. g
1 LRDM) (Figure 1(c)). Shredders colonized both treatments in a similar numerical pattern,
corresponding to 5% and 10% of total macroinvertebrates in Np and Pp, respectively. Shredders were
mainly represented by larvae of three plecopteran families: Austroperlidae (K. kuscheli), Gripopterygidae
(Antarctoperla michaelseni (Klapalek)), Notonemouridae (Austronemoura sp. and Udamocercia uramifera
(Aubert)), and one trichopteran family: Sericostomatidae (Parasericostoma cristatum Flint). Although the
shredder density was around 1 ind. g
1 of LRDM in both treatments (Figure 1(d)), the biomass showed a
different pattern. In Np, shredder biomass represented up to 53–83% of total macroinvertebrate biomass
(t42–t16, respectively). Maximum biomass was 3.9 mg g
1 LRDM at t42 (53%). In Pp, shredder biomass was
considerably lower than in the other treatment, ranging from 0.2 to 1.0 mg g
1 LRDM (t2–t118) accounting
for 19% of total macroinvertebrate biomass. During the experiment, gathering-collectors increased both in
number and biomass (Figure 1(f) and (g)) in both treatments. In terms of density, they represented the
major component of the macroinvertebrate assemblage (Figure 1(b)–(f)), since a large number of
oligochaetes, chironomids and ephemeropterans colonized the bags. The relative abundance of gathering-
collectors to total macroinvertebrates ranged from 83% to 88% (Np) and 62% to 87% (Pp). In terms of
biomass, this fraction ranged from 9% to 35% (Np) and from 19% to 35% (Pp).

* ireco basin
Table 1. Breakdown estimates of leaf litter for the two experiments in N

Treatment Stream temperature (8C) Decay rate (d
1)  SE t50% (d)a N r2adj Fsb df Fab 0.01
Experiment 1 F(1, 30)
Nothofagus pumilio 0.0033  0.0003 212 20 0.90
1.0–4.5 21.53 36 7.56
Pinus ponderosa 0.0017  0.0002 408 20 0.78

Experiment 2c F(1, 40)

Open bags 0.0074  0.0007 94 30 0.89
5.5–13.5 21.15 41 7.31
Closed bags 0.0053  0.0001 131 15 0.99
a
t50% is the time when the first 50% of CPOM would decay.
b
Fs and Fa 0.01 } statistic and critical values for the regression coefficient’s comparison method.
c
conducted with N. pumilio leaves.

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
186 R. J. ALBARIÑO AND E. G. BALSEIRO

Figure 1. Leaf litter mass loss and macroinvertebrate colonization for Nothofagus pumilio and Pinus ponderosa treatments
(Experiment 1). LRDM=leaf remaining dry mass  1 SE.

Individual mean biomass of shredders was always higher than that of gathering-collectors in both
treatments (Figure 2(a)). Shredders colonizing Np presented the highest individual mean biomass, with
maximum values at t16 and t42 (about 3 mg ind.
1). Gathering-collector individual biomass was rather
constant throughout the experiment, and it slightly increased towards the end of the experiment (0.1 mg
ind
1 at t118) (Figure 2(a)).

Experiment 2
Mean water temperature was 8.78C. (  0.8 SE). Lower temperatures were recorded in November (5.58C)
and March (7.58C) while higher ones were measured in December (13.58C). During the first 7 days, mass
loss, only measured in open bags, accounted for 6% of initial leaf litter. Subsequently, leaf fragmentation
gradually increased towards the end of the experiment. Although this process was apparently higher in
open bags, shredder feeding signs on leaves were also identified in closed bags. At day 135, leaf material had
lost 69% (OB) and 51% (CB) of their initial DM (Figure 3(a)).
Decay rate of N. pumilio leaves was significantly higher in OB than in CB (P50.01) (Table 1,
Figure 3(a)). The estimated t50% in OB treatment was almost one half of that for CB (Table 1).
Total macroinvertebrate abundance showed an increase towards the end of the experiment (Figure 3(b))
with a maximum of 322 ind. g
1 LRDM in open bags and 97 ind. g
1 LRDM in closed ones. Biomass was
markedly higher in OB, ranging from 18.6 to 38.8 mg g
1 LRDM (Figure 3(c)), while in CB it ranged
from 1.4 to 7.1 mg g
1 LRDM. The closed treatment successfully prevented the access of large

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
 LEAF LITTER BREAKDOWN IN PATAGONIAN STREAMS 187

Figure 2. Individual biomass of shredders and gathering-collectors colonizing leaf bags in Experiment 1 (a) and Experiment 2 (b).
Np=N. pumilio, Pp=Pinus ponderosa, OBT=open bag treatment and CBT=closed bag treatment. Mean  1 SE.

Figure 3. Mass loss and macroinvertebrate colonization of Nothofagus pumilio leaf litter for open and closed bag treatments
(Experiment 2). LRDM=leaf remaining dry mass  1 SE.

macroinvertebrates to the leaf litter inside the bags. This difference was observed for both shredders and
gathering-collectors, in terms of abundance and biomass (Figure 3(b)–(g)).
Shredders represented by Plecoptera, Diptera and Trichoptera larvae colonized both treatments, but they
were more abundant in OB (7–10 ind. g
1 LRDM) (Figure 3(d)). In open bags, shredder biomass increased

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
188 R. J. ALBARIÑO AND E. G. BALSEIRO

to 29.2 mg g
1 LRDM at day 89, when litter had lost 47% of its initial dry mass (Figure 3(e)). In closed bag
treatments, shredder density and biomass showed a tendency to increase. At day 135, both values were
2.0 ind. g
1 LRDM and 0.8 mg g
1 LRDM respectively (Figure 3(d) and (e)). Although in open bags, the
relative density of shredders to total macroinvertebrates was low (3–25%), in terms of biomass they
accounted for 70–93%, except at t135 when they decreased to 40%. Mean individual biomass of shredders
decreased substantially during the experiment from 8.0 (t7) to 1.1 mg ind.
1 (t135) in OB (Figure 2(b)). In
closed treatments, lower values were found throughout the experiment, up to 0.5 mg ind.
1 (Figure 2(b)). In
contrast to the pattern registered in OB, mean biomass per shredder increased 18% in CB. Large shredders
(intermediate and large-sized instars of K. kuscheli, trichopteran Monocosmoecus sp., and the dipteran
Tipula sp.) only colonized the open bags. On the other hand, closed bags were colonized by small shredders
such as the plecopteran Antarctoperla michaelseni and Parasericostoma cristatum, corresponding with the
presence of shredding signs on leaves. Many leaves were found shredded and leaf fragmentation increased
remarkably throughout the experiment as in open bags, indicating that smaller invertebrates were able to
bring about loss in CPOM mass.
Gathering-collectors were represented by Oligochaeta, Ephemeroptera, Diptera and early instars of
shredder taxa. In open bags, abundance increased throughout the experiment, reaching 291 ind. g
1 LRDM
(Figure 3(f)), although a slight decrease at day 89 was observed (Figure 3(f)). The relative abundance of
gathering-collectors was always high, varying between 63% and 92% in open bags and 85% and 87% in
closed ones. Individual mean biomass was always low in both treatments (Figure 2(b)). Due to their smaller
size, the contribution to total biomass was low (4–35% of total macroinvertebrates) in spite of their high
densities. The mesh size also limited the access of this feeding group, since density and biomass were lower
in closed than in open bags (Figure 3(f) and (g)).

DISCUSSION

In the autumn-winter experiment (Experiment 1), both CPOM decay rate and invertebrate colonization
differed between the two treatments (Figure 1). Although the native Nothofagus pumilio and the exotic
Pinus ponderosa litter showed very low leaching, N. pumilio leaves were eaten by shredders and decayed
faster than P. ponderosa needles. Conifer needles have highly refractory organic compounds (Sedell et al.,
1975; Gregory et al., 1991) and Pinaceae have the lowest decay rates known (Webster and Benfield, 1986).
Cuticular thickness and presence of resins delay microbial conditioning and thus its availability for
detritivores (B.arlocher et al., 1978; Michaelides and Kendrick, 1978). In another study with litter 1 month
pre-conditioned, Albarin* o and Balseiro (2001) found P. ponderosa leaf toughness to be nine times higher
than that of N. pumilio leaves. Fagaceae also have low decay rates (Webster and Benfield, 1986) and N.
pumilio mass loss obtained in the winter experiment (3.48C, mean temperature) was 34% after 118 days.
However, P. ponderosa decay rate was one half of that of N. pumilio (Figure 1(a)) and needles did not show
any sign of shredder feeding during 4 months indicating that the failure of shredders to consume such tough
litter contributed to this difference.
The colonization patterns in the two treatments were similar in terms of total individual numbers, but
different in terms of total biomass (Figure 1). These discrepancies were due to the size of individuals that
colonized the litter inside the bags (Figure 2a). Large shredders such as the plecopteran Klapopteryx
kuscheli and the trichopteran Parasericostoma cristatum were only found in the Nothofagus bags and not in
the Pinus ones. The dissimilar colonization displayed by large shredders would reflect a preference for N.
pumilio. Although leaf litter in streams is primarily a food source for shredders (Minshall, 1984), terrestrial
debris may also be very important as a refuge and substratum for invertebrates (Reice, 1980). For example,
Dobson (1994) found that the number, biomass and diversity of invertebrates colonizing litter bags
decreased in high quality litter, because shredder consumption depleted the resource. In the present study

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
 LEAF LITTER BREAKDOWN IN PATAGONIAN STREAMS 189

the decay of N. pumilio was not high enough to limit the resource and thereby cause a decrease in shredder
numbers. On the other hand, Pinus needles were also colonized by a high number of macroinvertebrates,
especially dominated by gathering-collectors. Since pine needles are tough and have low nitrogen content
(Friberg and Jacobsen, 1994; Albarin* o and Balseiro, 2001), it seems that this low quality leaf pack acted as
a natural filter both by trapping fine particulate matter and being a suitable surface for biofilm
development. In consequence, throughout the experiment pine needles were more important as a refuge and
substratum for both functional feeding groups studied than as a food resource itself.
The second experiment investigated the relationship between invertebrate size and the decay rates of N.
pumilio leaf litter. The mesh size of open bags allowed large shredders such as K. kuscheli, Monocosmoecus
sp. and Tipula sp. to colonize and consume the leaves, while only small macroinvertebrates gained access to
the closed bags. The invertebrate assemblage in open bags showed a pattern similar to that observed for
Nothofagus treatment in the first experiment. Gathering-collectors were constantly dominant in number
and represented by small specimens while shredder biomass was dominant as a result of its high individual
biomass (Figure 2(b)). Both total shredder density and biomass were low in the closed bags compared with
the open ones (Figure 3(d), and 3(e)). Open bags, with a broader size spectrum of shredders, had higher
decay rates. Rounick and Winterbourn (1983) and Stewart (1992) found similar results in shredder-rich
streams (New Zealand and South Africa, respectively) when they conducted comparable enclosure/
exclosure leaf litter breakdown experiments. Also Campbell et al. (1994) obtained identical results in an
Australian stream. Those studies registered higher differences between treatments than in these experiments
because of the very small mesh size used in their ‘closed bag’ treatment (180–300 mm). The relatively large
mesh size (1000 mm) used in the closed bags of the present study made them more permeable to small
invertebrate colonization and this may have been the reason for the lower differences observed. During the
135 day period, leaf litter in closed bags decayed by 51%. On the assumption that 6% of mass loss at day 7
resulted from leaching and microbial activity, the remaining mass loss (49%) during the experiment should
be related both to microbes and small shredders. However, with these treatments, it is not possible to
differentiate microbial activity from that of small detritivores. The fact that shredder feeding signs and leaf
fragmentation were conspicuous and increased throughout the experiment suggests that small shredders are
likely to have some influence in leaf decay. It was assumed that physical leaf fragmentation did not occur
because the leaf bags acted as shelter against water and suspended sediment abrasion (Campbell et al.,
1994). Short et al. (1980) pointed out that small shredders can produce similar results to large ones. Early
instars are known to display higher consumption rates per body mass than late ones (Scriber and Slansky,
1981). In a previous experimental study on N. pumilio breakdown, Albarin* o and Balseiro (1998) found that
smaller larvae of K. kuscheli consumed significantly more leaf litter than large ones, achieving a
consumption rate of 0.24% body dry mass day
1. However, shredder biomass in open bags was
significantly higher than that in closed ones, and this resulted in higher breakdown activity. Although
shredder size was directly related with decay rate, the activity of small shredders in Andean headwaters
cannot be neglected.
The south mountain beech Nothofagus pumilio forms pure stands that characterize the upper belt of
Andean-patagonian headwaters (Hildebrand-Vogel et al., 1990). Therefore, the input and breakdown
timing of N. pumilio leaf litter might be key elements of ecosystem functioning. These experiments showed
that the use of exotic Pinaceae with different leaf characteristics resulted in changes to invertebrate
colonization and feeding patterns influencing leaf litter decay. Although pine needles may be consumed by
aquatic invertebrates (Grafius and Anderson, 1980; Friberg and Jacobsen, 1999), they would only be
accessible later than native deciduous leaves. Moreover, the longer leaf litter conditioning time required is
likely to lead to a higher probability of downstream transport by seasonal floods. Thus, pine needles would
be more prone to export than to in situ processing. In consequence, pine detritus entering Andean streams
as the main food source may delay or disrupt shredder life cycles. Friberg and Jacobsen (1994) mentioned a
similar situation for Fagus woodland replacement with pine plantations in Sweden. Furthermore, the

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
190 R. J. ALBARIÑO AND E. G. BALSEIRO

timing of litter input is different between the two tree species. N. pumilio is a deciduous tree with an autumn
pulse of leaf fall, while pines produce litter all year round with higher amounts in summer (Bray and
Gorham, 1964). This difference in litter pulses may also have great effect on shredder populations, although
this was not investigated in these experiments.
The fact that in experiment 1, large shredders were found only in Nothofagus bags, and not in pine ones,
would indicate that large shredders are not attracted by this litter. Invertebrate movements in searching
suitable food patches would consume not only energy and time, which might have been allocated to
secondary production, but would also increase predation losses. Both consequences have detrimental effects
on large shredder populations. Experiment 2 demonstrated that large shredders are a key to the leaf litter
decay of Nothofagus leaves. If these large shredders were disadvantaged due to an increment of pine needles
in the coarse organic matter standing crop, the whole litter processing mechanism would be affected, with
consequences for the entire stream biota. The estimate of suitable areas in Patagonia for pine forestation
made by Schlichter and Laclau (1998) suggests that great impacts on aquatic ecosystems can be expected as
a consequence of changing litter quality and input timing.

ACKNOWLEDGEMENTS

We thank Dr. T.Moulton and Dr. M.Dobson whose comments helped us to improve an early version of this
manuscript, and Drs. B.E.Modenutti and C.P.Queimalin* os for helpful discussion. This study was supported by a
FONCYT grant PICT 01-06035, CONICET PIP 0739/98.

REFERENCES
Abelho M, Graca MAS. 1996. Effects of eucalyptus afforestation on leaf litter dynamics and macroinvertebrate
-
community structure of streams in Central Portugal. Hydrobiologia 324: 195–204.
*
Albarino RJ. 1999. Din!amica del procesamiento de la materia org!anica particulada gruesa por el macrozoobentos en
arroyos andinos. Doctoral thesis, Universidad Nacional del Comahue. Bariloche, Argentina.
Albarin* o RJ. 2001. The food habits and mouthpart morphology of a South Andes population of Klapopteryx kuscheli
(Plecoptera: Austroperlidae). Aquatic Insects 23: 171–181.
Albarin* o RJ, Balseiro EG. 1998. Larval size and leaf conditioning in the breakdown of Nothofagus pumilio leaves by
Klapopteryx kuscheli (Insecta, Plecoptera) in a South Andean stream. International Review of Hydrobiology 83: 397–
404.
Albarin* o RJ, Balseiro EG. 2001. Food quality, larval consumption and growth of Klapopteryx kuscheli (Plecoptera)
from a South Andes stream. Journal of Freshwater Ecology 16: 517–526.
Albarin* o RJ, Valverde AC. 1998. H!abito alimentario del estado larval de Parasericostoma cristatum (Trichoptera:
Sericostomatidae). Revista Sociedad Entomolo!gica Argentina 57: 131–135.
Allan HH. 1982. Flora of New Zealand, vol. 1. Manaaki Whenua Press: Wellington, New Zealand.
B.arlocher F, Kendrick BK, Michaelides J. 1978. Colonization and conditioning of Pinus resinosa needles by aquatic
hyphomycetes. Archiv fu.r Hydrobiologie 81: 462–474.
Barrera MD, Frangi JL, Richter LL, Perdomo MH, Pinedo LB. 2000. Structural and functional changes in Nothofagus
pumilio forest along a latitudinal gradient in Tierra del Fuego, Argentina. Journal of Vegetation Science 11: 179–188.
Bray JR, Gorham E. 1964. Litter production in forests of the world. In Advances in Ecological Research. Cragg JB (ed).
Academic Press, London; 102–157.
Campbell IC. 1994. Ecology of Australian detritus-based streams}a note in response to Graca. Limnologica 24: 291–
293. -
Campbell IC, Enierga GM, Fuchshuber L. 1994. The influence of pack size, and position, leaf type, and shredder access
on the processing rate of Atherosperma moschatum leaves in an Australian cool temperate rainforest stream.
Internationale Revue Gesamt Hydrobiologie 79: 557–568.
Cuffney, TF, Wallace JB, Lugthart GJ. 1990. Experimental evidence quantifying the role of benthic invertebrates in
organic matter dynamics of headwater streams. Freshwater Biology 23: 281–299.
Cummins KW. 1996. Invertebrates. In River Biota, Petts G, Calow P (eds). Blackwell Science: Oxford; 75–91.
Cummins KW, Merritt RW. 1996. Ecology and distribution of aquatic insects. In An Introduction to the Aquatic Insects
of North America, Merritt RW, Cummins KW (eds). Kendall/Hunt Publishing Company, Dubuque, IA; 74–86.

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
 LEAF LITTER BREAKDOWN IN PATAGONIAN STREAMS 191

Cummins KW, Wilzbach MA, Gates DM, Perry JB, Taliaferro WB. 1989. Shredders and riparian vegetation.
BioScience 39: 24–30.
Davis SF, Winterbourn MJ. 1977. Breakdown and colonization of Nothofagus leaves in a New Zealand stream. Oikos
28: 250–255.
D!ıaz Villanueva V, Albarin* o RJ. 1999. Feeding habit of Notoperla archiplatae (Plecoptera) larvae in a North Patagonia
Andean stream, Argentina. Hydrobiologia 412: 43–52.
Dieterich M, Anderson NH. 1995. Life cycles and food habits of mayflies and stoneflies from temporary streams in
western Oregon. Freshwater Biology 34: 47–60.
Dobson M. 1994. Microhabitat as a determinant of diversity: stream invertebrates colonizing leaf packs. Freshwater
Biology 32: 565–572.
Donoso C, Lara A. 1996. Utilizacio! n de los bosques nativos en Chile: pasado, presente y futuro. In Ecolog!ıa de Los
Bosques Nativos de Chile, Armesto J, Villagr!an C, Arroyo MK (eds). Editorial Universitaria SA: Santiago de Chile;
363–387.
Friberg N, Jacobsen D. 1994. Feeding plasticity of two detritivore-shredders. Freshwater Biology 32: 133–142.
Friberg N, Jacobsen D. 1999. Variation in growth of the detritivore-shredder Sericostoma personatum (Trichoptera).
Freshwater Biology 42: 625–635.
Gessner MO, Dobson M. 1993. Colonisation of fresh and dried leaf litter by lotic macroinvertebrates. Archiv fu.r
Hydrobiologie 127: 141–149.
Graca MAS, Maltby L, Calow P. 1993. Importance of fungi in the diet of Gammarus pulex and Asellus aquaticus I:
feeding strategies. Oecologia 93:139–144.
Grafius E, Anderson NH. 1980. Population dynamics and role of two species of Lepidostoma (Trichoptera:
Lepidostomatidae) in an Oregon coniferous forest stream. Ecology 61: 808–816.
Gregory SV, Swanson FJ, McKee WA, Cummins KW. 1991. An ecosystem perspective of riparian zones. BioScience 8:
540–551.
Hildebrand-Vogel R, Godoy R, Vogel A. 1990. Subantarctic-Andean Nothofagus pumilio forests. Vegetatio
89: 55–68.
Lara A, Donoso C, Aravena JC. 1996. La conservacio! n del bosque nativo de Chile: problemas y desaf!ıos. In Ecolog!ıa de
Los Bosques Nativos de Chile, Armesto J, Villagr!an C, Arroyo MK (eds). Editorial Universitaria SA: Santiago de
Chile, 335–362.
Linklater W. 1995. Breakdown and detritivore colonisation of leaves in three New Zealand streams. Hydrobiologia 306:
241–250.
Merritt RW, Cummins KW. 1996. An Introduction to the Aquatic Insects of North America. Kendall/Hunt Publishing
Company, Dubuque, IA.
Michaelides J, Kendrick B. 1978. An investigation of factors retarding colonization of conifer needles by amphibious
hyphomycetes in streams. Mycologia 70: 419–430.
Minshall GW, 1984. Aquatic insect-substratum. In The Ecology of Aquatic Insects, Resh VH, Rosenberg DM (eds).
Praeger Publishers, New York, 358–400.
Modenutti BE, Balseiro EG, Queimalin* os CP, An* o! n Su!arez DA, Di!eguez MC, Albarin* o RJ. 1998. Structure and
dynamics of food webs in Andean lakes. Lakes and Reservoirs: Research and Management 3: 179–186.
Murphy JF, Giller PS. 2000. Seasonal dynamics of macroinvertebrate assemblages in the benthos and associated with
detritus packs in two low-order streams with different riparian vegetation. Freshwater Biology 43: 617–631.
Muru! a R. 1996. Comunidades de mam!ıferos del bosque templado de Chile. In Ecolog!ıa de los bosques nativos de Chile,
Armesto J, Villagr!an C, Arroyo MK (eds). Editorial Universitaria SA: Santiago de Chile; 113–133.
Petersen RC, Cummins KW. 1974. Leaf processing in a woodland stream. Freshwater Biology 4: 343–368.
Postler V, Urretavizcaya F, Grosfeld J, Ramilo E, Caracotche S, Co! rdoba M, Cerutti T, Finster G, Arrechea D,
Manfredi R. 1999. Plan estrate!gico de manejo de la reserva forestal cuartel Lago Epuye!n.
Prochazka K, Stewart BA, Davies BR. 1991. Leaf litter retention and its implications for shredder distribution in two
headwater streams. Archiv fu.r Hydrobiologie 120: 315–325.
Reice SR. 1980. The role of substratum in benthic macroinvertebrate microdistribution and litter decomposition in a
woodland stream. Ecology 61: 580–590.
Robinson CT, Gessner MO, Ward JV. 1998. Leaf breakdown and associated macroinvertebrates in alpine glacial
streams. Freshwater Biology 40: 215–228.
Rounick JS, Winterbourn MJ. 1983. Leaf processing in two contrasting beech forest streams: effects of physical and
biotic factors on litter breakdown. Archiv fu.r Hydrobiologie 96: 448–74.
Rozzi R, Mart!ınez D, Willson M, Sabag C. 1996. Avifauna de los bosques templados de Sudam!erica. In Ecolog!ıa de Los
Bosques Nativos de Chile, Armesto J, Villagr!an C, Arroyo MK (eds). Editorial Universitaria SA: Santiago de Chile;
135–152.

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)
192 R. J. ALBARIÑO AND E. G. BALSEIRO

Schlichter T, Laclau P. 1998. Ecotono estepa-bosque y plantaciones forestales en la Patagonia norte. Ecolog!ıa Austral 8:
285–296.
Scriber JM, Slansky F. 1981. The nutritional ecology of immature insects. Annual Review of Entomology 26: 183–211.
Sedell JR, Triska FJ, Triska NS. 1975. The processing of conifer and hardwood leaves in two coniferous forest streams:
I. Weight loss and associated invertebrates. Verhandlungen der Internationalen Vereinigung fu.r theoretiche und
angewandte Limnologie 19: 1617–1627.
Sedell JR, Triska FJ, Hall JD, Anderson NH, Lyford JH. 1974. Sources and fates of organic inputs in coniferous forest
streams. Coniferous Forest Biome Bulletin 5: 57–69.
Short RA, Canton SP, Ward JV. 1980. Detrital processing and associated macroinvertebrates in a Colorado Mountain
stream. Ecology 61: 727–732.
Smock LA, Metzler GM, Gladden JE. 1989. Role of debris dams in the structure and functioning of low-gradient
headwater streams. Ecology 70: 764–775.
Sokal RR, Rohlf FJ. 1981. Biometry. Freeman and Co., New York.
Stewart BA. 1992. The effect of invertebrates on leaf decomposition rates in two small woodland streams in southern
Africa. Archiv fu.r Hydrobiologie. 124: 19–33.
Vannote RL, Minshall GW, Cummins KW, Sedell JR, Cushing CE. 1980. The river continuum concept. Canadian
Journal of Fisheries and Aquatic Sciences, 37, 130–137.
Wardle P. 1998. Comparison of alpine timberline in New Zealand and the Southern Andes. Royal Society New Zealand
Miscellaneous Publication 48: 69–90.
Webster JR, Benfield EF. 1986. Vascular plant breakdown in freshwater ecosystems. Annual Review of Ecology and
Systematics 17: 567–594.

Copyright # 2002 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 12: 181–192 (2002)