Ecosystems (2019) 22: 629–642

https://doi.org/10.1007/s10021-018-0292-7
 2018 Springer Science+Business Media, LLC, part of Springer Nature

A Global Assessment of the Effects

of Eucalyptus Plantations on Stream
Ecosystem Functioning
Verónica Ferreira,1* Luz Boyero,2,3,4 Clementina Calvo,5 Francisco Correa,6
Ricardo Figueroa,7 José Francisco Gonçalves Jr.,8 Guillermo Goyenola,5
Manuel A. S. Graça,1 Luiz U. Hepp,9 Samuel Kariuki,10
Anahı́ López-Rodrı́guez,5 Néstor Mazzeo,5 Charles M’Erimba,10
Silvia Monroy,2 Alessandra Peil,8 Jesús Pozo,2 Renan Rezende,11
and Franco Teixeira-de-Mello5

1
Department of Life Sciences, MARE – Marine and Environmental Sciences Centre, University of Coimbra, 3004-517 Coimbra,
Portugal; 2Department of Plant Biology and Ecology, Faculty of Science and Technology, University of the Basque Country (UPV/
EHU), P.O. Box 644, 48080 Bilbao, Spain; 3IKERBASQUE, Basque Foundation for Science, 48013 Bilbao, Spain; 4College of Science
and Engineering, James Cook University, Townsville, QLD 4811, Australia; 5Grupo de Ecologı́a y Rehabilitación de Sistemas
Acuáticos, Departamento de Ecologı́a y Gestión Ambiental, Centro Universitario de la Región Este, Universidad de la República,
Montevideo, Uruguay; 6Department of Environmental Science, Faculty of Natural Resource, Catholic University of Temuco, Temuco,
Chile; 7Departamento de Sistemas Acuáticos, Facultad de Ciencias Ambientales y Centro EULA-Chile, Universidad de Concepción,
Concepción, Chile; 8AquaRiparia/Departamento de Ecologia, Instituto de Ciências Biológicas, Universidade de Brası́lia – UnB,
Campus Universitário Darcy Ribeiro, Asa Norte, Brası́lia, DF 70910-900, Brazil; 9Departamento de Ciências Biológicas, Universidade
Regional Integrada do Alto Uruguai e das Missões, Erechim, Rio Grande do Sul 99709-910, Brazil; 10Department of Biological
Sciences, Egerton University, Njoro, Kenya; 11Program of Postgraduate in Environmental Sciences, Community University of the
Chapecó Region, Chapecó, SC CEP 89809-000, Brazil

ABSTRACT
Forest change is a major environmental problem effects on stream functioning have been addressed
worldwide. Forest streams, with their large aqua- mostly in the Iberian Peninsula, which limits gen-
tic–terrestrial interface and strong dependence on eralization to other regions. We assessed the effect
terrestrially derived organic matter, are highly of eucalyptus plantations on total (microbial
sensitive to forest changes. Fast-wood plantations decomposers and macroinvertebrates; in coarse
can be particularly threatening if they markedly mesh bags) and microbial-driven (in fine mesh
differ from native forests. Eucalyptus plantations, in bags) leaf litter decomposition by comparing
particular, cover large areas worldwide (> 20 streams flowing through native forests and euca-
million ha, mostly from 35S to 35N), but their lyptus plantations in seven regions in the Iberian
Peninsula, Central Africa and South America. We
found an overall significant inhibition of total litter
Received 21 April 2018; accepted 2 August 2018; decomposition by 23%. The effect did not signifi-
published online 21 August 2018
cantly differ across regions, although a significant
Electronic supplementary material: The online version of this article
inhibition was found for Spain (- 41%), South
(https://doi.org/10.1007/s10021-018-0292-7) contains supplementary Brazil (- 31%) and Uruguay (- 36%) (Portugal
material, which is available to authorized users. had a marginally nonsignificant inhibition by 50%)
Author Contributions VF designed the study, analyzed the data and
wrote the original manuscript; all authors performed local experiments
but not for other regions, suggesting that the effects
and commented on previous versions of the manuscript. of plantations in temperate climates are mediated
*Corresponding author; e-mail: veronica@ci.uc.pt through effects on macroinvertebrate communities.

629
630 V. Ferreira and others

Contrarily, the overall effect for microbial-driven strongly affect streams due to their large aquatic–
litter decomposition was non-significant, but it terrestrial interface and strong dependence on
significantly differed across regions with a signifi- riparian litter (Vannote and others 1980; Wallace
cant stimulation in Central Brazil (110%) and Ur- and others 1997).
uguay (32%), and nonsignificant effects for other Eucalyptus trees have high water demands
regions (Kenya had a marginally nonsignificant (Calder 2002), leading to decreases in runoff and in
inhibition by 48%), suggesting that functional aquifer levels (Lara and others 2009). Also, soils in
redundancy among microbial communities is not eucalyptus plantations are highly hydrophobic
general and effects can occur if plantations induce (Abelho and Graça 1996; Fernández and others
changes in nutrient availability, solar irradiation or 2006), likely due to the accumulation of oils re-
litter characteristics. leased from the leaves, hindering rain water pen-
etration and the replenishment of groundwater
Key words: effect size; forest change; leaf litter reservoirs. Thus, stream hydrology is generally al-
decomposition; mesh bags; meta-analysis; response tered, with reduced water flow over the year,
ratio; stream functioning. spates during heavy rain events and drought during
the dry season, especially in arid regions (Scott and
Lesch 1997; Lara and others 2009; Cordero-Rivera
and others 2017). Streams flowing through plan-
HIGHLIGHTS tations may also have altered light and temperature
regimes, if single-stemmed eucalyptus trees replace
wide-canopy deciduous trees.
 Effects of eucalyptus plantations on stream
Eucalyptus plantations may also promote chan-
functioning were assessed in seven regions.
ges in stream substrates and habitats due to the
 Effects of plantations on litter decomposition
deposition of bark and branches (Pozo and others
depended on decomposer communities and
1997; Molinero and Pozo 2004; Santiago and oth-
region.
ers 2012) and fine sediments (Siegloch and others
 Inhibition of decomposition was mostly mediated
2014). Additionally, streams flowing through
by effects on macroinvertebrates.
eucalyptus plantations may endure changes in lit-
ter fall phenology, diversity, quantity and quality
when these plantations replace diverse native for-
INTRODUCTION
ests composed of trees with traits markedly differ-
Total forest area has decreased worldwide, while ent from those of eucalyptus (Abelho and Graça
the area occupied by tree plantations has increased 1996; Pozo and others 1997; Graça and others
(Payn and others 2015). In 2000, approximately 90 2002; Molinero and Pozo 2004, 2006). These
million ha were occupied by large-scale monocul- changes in environmental characteristics may af-
tures that are intensively managed to produce high fect aquatic biota and ecosystem processes, which
volumes of wood at a fast rate (Gerber 2011). Such may jeopardize streams ability to provide ecosys-
plantations are mostly used to supply the paper tem services (for example, good water quality,
industry in developing countries and have been food) (Graça and others 2002; Ferreira and others
promoted for mitigating climate change by trapping 2016).
atmospheric carbon (Bäckstrand and Lövbrand One fundamental ecosystem process in forest
2006; Zomer and others 2008). Thus, the increased streams is litter decomposition (Vannote and others
paper demands and the opportunities created by 1980; Wallace and others 1997), which is mostly a
the carbon sequestration market will likely increase biological process carried out by microbial decom-
the pace of land conversion into fast-wood plan- posers (mostly aquatic hyphomycetes) and
tations. macroinvertebrate shredders (Hieber and Gessner
Eucalyptus plantations occupy large areas world- 2002). Microbes colonize the litter soon after it has
wide (ca. 20 million ha in 2009; Iglesias-Trabado leached secondary compounds with antimicrobial
and Wilstermann 2009), mostly between 35S and activity (for example, polyphenols), and their
35N. In the Iberian Peninsula, these plantations activities lead to the mineralization of organic car-
occupy 1.5 million ha (IFN3 2007; ICNF 2013) and bon, incorporation of carbon and nutrients into
are the predominant type of industrial fast-wood microbial biomass and litter softening (Hieber and
plantation in the southern hemisphere (Iglesias- Gessner 2002; Gulis and Suberkropp 2003; Gulis
Trabado and Wilstermann 2009). The replacement and others 2006). These changes increase litter
of native forests by eucalyptus plantations may palatability to shredders who incorporate it into
 Eucalyptus Plantations and Stream Functioning 631

secondary production (Hieber and Gessner 2002; litter decomposition in streams in regions with high
Cornut and others 2010). The relative role of mi- precipitation and where streams do not dry out (for
crobes and macroinvertebrates on litter decompo- example, in tropical afromontane forests), in re-
sition depends on environmental characteristics gions with riparian vegetation dominated by ever-
with cool-water streams (for example, at higher green species with recalcitrant litter (for example,
latitude or elevation) generally bearing higher in the Brazilian savannah), or in regions where
species richness and density of shredders, which macroinvertebrates are not important decomposers
contribute significantly to litter decomposition, and litter decomposition is mostly microbial-driven
than warm-water streams where litter decomposi- (for example, some tropical regions). Several
tion can be mostly microbial-driven (Dobson and studies have compared the decomposition of native
others 2002, 2004; Gonçalves and others 2006; leaf litter with that of different eucalyptus species
Boyero and others 2011a, b). Litter traits also (Eucalyptus globulus: Canhoto and Graça 1996; Eu-
moderate the role of macroinvertebrates on calyptus grandis: Gonçalves and others 2012a; Eu-
decomposition since these generally prefer soft and calyptus camaldulensis: Gonçalves and others 2012b;
nutrient-rich litter over more recalcitrant litter that Eucalyptus cloeziana: Rezende and others 2014), but
is mostly decomposed by microbes (Canhoto and these do not allow assessing the full effects of forest
Graça 1995; Hieber and Gessner 2002; Gonçalves change (which go beyond those mediated by
and others 2007; Casotti and others 2014; Kiffer changes in litter characteristics) on ecosystem
and others 2018). processes.
According to a recent meta-analysis, total (that We present the first global assessment of how
is, macroinvertebrate + microbial-driven) litter litter decomposition is affected by eucalyptus
decomposition is reduced in streams flowing plantations, including seven regions distributed
through eucalyptus plantations compared with across the Iberian Peninsula, Central Africa and
streams in native deciduous forests in the Iberian South America. In this collaborative, coordinated
Peninsula (Ferreira and others 2016). This is likely study, we tested the following hypotheses: (1)
due to the lower taxa richness, density or biomass Eucalyptus plantations have an effect on litter
of benthic macroinvertebrates generally found in decomposition, and the effect depends (2) on the
the former than the latter streams (Abelho and type of decomposer community involved (total:
Graça 1996; Larrañaga and others 2009a, 2009b; macroinvertebrates + microbes or microbes only),
Ferreira and others 2015; Cordero-Rivera and and (3) on the region.
others 2017). The same meta-analysis reported that
microbial-driven litter decomposition is less af-
MATERIALS AND METHODS
fected by forest change than total litter decompo-
sition (Ferreira and others 2016). Although species Study Sites
richness and evenness of aquatic hyphomycetes are
Three streams flowing through native forests
generally lower in streams flowing through plan-
(henceforth, native streams) and three streams
tations than in streams in native deciduous forests
flowing through eucalyptus plantations (hence-
in Portugal, and community structure differs be-
forth, eucalyptus streams) were selected in each of
tween stream types in both Portugal and Spain
seven regions: Spain, Portugal, Kenya, Central
(Chauvet and others 1997; Bärlocher and Graça
Brazil, South Brazil, Uruguay and Chile (Table 1).
2002; Ferreira and others 2006), the weak response
Eucalyptus and native streams were selected to
of microbial-driven litter decomposition to forest
represent low-order streams within each forest type
change suggests high functional redundancy in
in each region and were paired based on size,
microbial communities between stream types
substrate and geomorphology (three stream pairs
(Bärlocher and Graça 2002; Ferreira and others
per region). Climate, soil, type of native forest and
2006).
eucalyptus species in plantations varied across re-
Importantly, however, most studies addressing
gions (Table 1). Differences in water temperature,
the effects of eucalyptus plantations on stream
electrical conductivity, pH and dissolved nutrients
functioning have been carried out in the Iberian
across regions reflected differences in climate,
Peninsula (reviewed by Graça and others 2002;
geology and vegetation (Table 2). Within each re-
Ferreira and others 2016), which limits general-
gion, streams had similar geomorphology and
ization to regions with distinct climate, vegetation
bedrock; native streams were surrounded by simi-
and contribution of macroinvertebrates to litter
lar vegetation. Streams were small (first–third or-
decomposition. For instance, it is uncertain what
der; generally < 5 m wide and < 50 cm deep),
would the effect of eucalyptus plantations be on
632 V. Ferreira and others

Table 1. Location, Climate, Soil and Forest in Seven Regions

Country Latitude Longitude Climate1 Soil2 Native Eucalyptus species
() () forest

Spain 43.21 to - 3.27 to Cfb: warm tem- Umbrisols Deciduous Eucalyptus globulus
43.35 - 3.23 perate, fully (UM) broadleaf
humid with
warm sum-
mers
Portugal 40.07 to - 8.31 to Csb: warm tem- Umbrisols Deciduous Eucalyptus globulus
40.52 - 8.20 perate, warm (UM) broadleaf
and dry sum-
mers
Kenya 0.41 to 37.34 to As: equatorial, Luvisols, Cam- Tropical Eucalyptus globulus or Eucalyptus
0.47 37.42 dry summers bisol (LV) afromon- grandis
tane
Central - 15.79 - 48.07 to Aw: equatorial, Ferralsols, Ac- Brazilian Mixture of Eucalyptus grandis,
Brazil to - - 47.94 dry winters risols, Niti- savannah Eucalyptus saligna, Eucalyptus
15.73 sols (FR) alba, Eucalyptus urophyla,
Corymbia citriodora (previously
Eucalyptus citriodora)
South - 27.72 - 52.33 to Cfa: warm tem- Ferralsols, Ac- Araucaria Eucalyptus grandis
Brazil to - - 52.06 perate, fully risols, Niti- rainforest
27.45 humid with sols (FR)
hot summers
Uruguay - 34.65 - 55.31 to Cfa: warm tem- Phaeozems Subtropical Mixture of Eucalyptus globulus and
to - - 54.99 perate, fully (PH) and seasonal Eucalyptus viminalis
34.18 humid with Kas- semi-de-
hot summers tanozems, ciduous
Solonetz
(KS)
Chile - 37.69 - 73.36 to Cfb: warm tem- Cambisol (CM) Coastal Eucalyptus globulus
to - - 73.24 perate, fully deciduous
37.51 humid with
warm sum-
mers

Location of streams, climate, soil, type of native forest and eucalyptus species used in plantations in seven regions.
1
World maps of Köppen–Geiger climate classification: http://koeppen-geiger.vu-wien.ac.at (accessed on October 2017); 2World soil resources map: ftp://ftp.fao.org/agl/agll/fa
omwsr/wsavcl.jpg (accessed on October 2017).

with no sign of human impact (except for euca- poses (for example, Boyero and others 2011a, b;
lyptus plantations in eucalyptus streams) (Fig- Ferreira and others 2012; Woodward and others
ure 1). 2012) because it is a highly palatable leaf species to
shredders (Graça and others 2001; Azevedo-Pereira
Leaf Litter Decomposition and others 2006; Graça and Cressa 2010). Batches
of 2.30–2.80 g of air-dry leaves were weighed,
Alder (Alnus glutinosa (L.) Gaertn.) leaves were
sprayed with distilled water to render them soft and
collected from the forest floor just after abscission
avoid breakage during handling and enclosed in
in autumn 2014, in Coimbra, central Portugal.
fine mesh (FM; 0.5-mm mesh) and coarse mesh
Leaves were air-dried at room temperature in the
(CM; 10-mm mesh) bags ( 12 9 15 cm). Two
dark, stored in paper boxes and posted to all teams.
mesh sizes were used to allow assessing microbial-
A. glutinosa is a broadleaf deciduous tree species
driven litter decomposition (FM bags) and litter
typical of riparian forests in Europe (Graça and
decomposition driven by both microbes and
Poquet 2014), but it has been used in decomposi-
macroinvertebrates (CM bags). Twelve litter bags of
tion experiments worldwide for comparative pur-
Table 2. Mean Water Characteristics for Streams in Seven Regions
Regions Stream type Channel Water Discharge Temperature Conductivity pH Dissolved DIN (lg L-1) SRP (lg L-1)
width (m) depth (m) (L s-1) (C) (lS cm-1) oxygen (%)

Spain Native 0.9–1.6 0.1–0.1 34.0–44.0 9.9–11.5 83–101 7.3–7.6 96–98 337–661 4–6
Eucalyptus 0.6–1.6 0.1–0.2 11.8–54.5 10.7–11.9 73–167 7.1–7.8 94–99 145–307 5–7
Portugal Native 1.4–2.1 0.1–0.2 16.6–137.0 8.0–8.4 27–36 na na 69–355 4–6
Eucalyptus 0.7–1.7 0.1–0.1 22.9–49.2 8.1–10.1 34–44 na na 161–401 3–5
Kenya Native 2.4–11.9 0.1–0.3 20.9–944.6 16.5–19.0 33–41 7.7–8 96–102 34–45 18–30
Eucalyptus 1.3–3.6 0.1–0.4 149.7–367.6 16.6–18.8 34–43 7.5–7.9 90–100 39–51 5–7
Central Brazil Native 1.1–1.5 0.2–0.3 83.4–219.7 18.9–20.0 4–5 5.2–5.8 65–87 bdl bdl
Eucalyptus 0.6–2.0 0.2–0.7 26.7–254.1 16.2–21.6 4–5 4.8–5.4 57–69 bdl bdl
South Brazil Native 1.9–2.4 0.1–0.2 36.7–63.0 19.2–20.1 67–76 6.2–6.5 91–101 2–3 84–98
Eucalyptus 1.1–2.2 0.1–0.5 28.4–151.3 19.4–21.3 59–92 6.1–6.6 86–97 6–22 80–173
Uruguay Native 2.4–5.2 0.3–0.3 0.5–8.6 22.2–28.4 297–558 8.4–8.7 112–143 170–219 5–8
Eucalyptus 2.4–4.5 0.3–0.5 1.8–5.1 15.7–24.9 119–276 7.7–8.1 67–128 104–225 2–3
Chile Native 4.9–10.8 0.2–0.3 295.7–936.4 13.0–19.9 22–37 6.1–6.4 89–98 26–143 5–28
Eucalyptus 2.3–11.9 0.1–0.2 24.2–1060.0 12.8–13.9 32–39 6.4–6.5 84–89 10–51 23–88

Mean water characteristics (min–max) for streams (n = 3) flowing through native forests and eucalyptus plantations in seven regions. Water temperature was assessed with dataloggers or field probes, electrical conductivity, pH and
dissolved oxygen were assessed with field probes, and dissolved nutrients were assessed by ion chromatography (generally, n = 2–4; water temperature in Spain, n = 31; channel width, depth and discharge in Kenya: n = 12).
DIN dissolved inorganic nitrogen, SRP soluble reactive phosphorus, na not assessed, bdl below detection limit (100 lg L-1).
Eucalyptus Plantations and Stream Functioning
633
634 V. Ferreira and others

Figure 1. Examples of streams flowing through native forests (N) and eucalyptus plantations (E).

each mesh size were deployed along a 50-m reach ‡ 48 h) and weighed for the determination of
in each stream. remaining dry mass (DM). Decomposition rates
Additionally, sets of six litter bags from each (k,/d) were estimated by fitting a negative expo-
mesh size were prepared as described above, taken nential model to the fraction of mass remaining
to the field on day 0 and submerged for about over time; DMt/DMi = e-kt, where DMt is the DM
10 min in one stream per region. Litter bags were remaining at time t (d), DMi is the initial DM and
taken back to the laboratory, and litter was oven- k is the decomposition rate on a daily basis. Since
dried (60C, ‡ 48 h) and weighed for the determi- streams differed in water temperature, decompo-
nation of initial oven-dry mass that allowed com- sition rates were also estimated on a degree-day
puting an initial air-dry mass to initial oven-dry basis (k,/dd) by replacing time in the model above
mass conversion factor taking into account han- by the cumulative thermal sum at the sampling
dling losses. This conversion factor was used to date, and these values were used for the analyses.
estimate the initial oven-dry mass of the samples.
Replicate litter bags (n = 4) were sampled on
three occasions over the incubation period that Data Analysis
lasted for 21–60 days, between November 2014
The overall effect of eucalyptus plantations on
and February 2016. Maximum incubation period
stream water and alder leaf litter decomposition,
and sampling dates varied across regions, but not
and the effects of region and mesh size on the
across streams within regions, and were defined
magnitude and direction of the response of these
to allow at least 50% mass loss in coarse mesh
variables to plantations were assessed using meta-
bags. Litter bags were retrieved, enclosed indi-
analytic techniques. These are most often used in
vidually in plastic bags and taken to the labora-
the context of systematic reviews, but can also be
tory in an ice chest. Leaves were rinsed with
applied to multisite studies (Rustad and others
running tap water over a 0.5-mm mesh sieve to
2001; Walker and others 2006; Gurevich 2013).
retain small leaf fragments, oven-dried (60C,
 Eucalyptus Plantations and Stream Functioning 635

The response of water characteristics and alder 26%) and DIN concentration was significantly
leaf litter decomposition to the replacement of na- lower in eucalyptus than in native streams in Spain
tive forests by eucalyptus plantations (effect size) (by 58%). Discharge and oxygen saturation did not
was estimated as the response ratio (R), which is significantly differ between eucalyptus and native
given by the ratio of the variable of interest (for streams at the regional or overall level.
example, litter decomposition rate) in the euca- Overall water temperature, conductivity, pH and
lyptus stream (kE) to the variable of interest in the soluble reactive phosphorus (SRP) did not signifi-
paired native stream (kN) (R = kE/kN); R = 1 indi- cantly differ between eucalyptus and native
cates no effect of eucalyptus plantations, R < 1 streams, but there were significant differences be-
indicates an increase and R > 1 indicates a de- tween stream types in some regions. Temperature
crease in the variable of interest in eucalyptus was significantly higher in eucalyptus than in na-
streams (Hedges and others 1999). Values were ln- tive streams in Portugal (by 18%) and lower in
transformed (lnR) for the analyses. The variance Uruguay (by 21%); conductivity was significantly
associated with the effect size (VlnR), which is lower in eucalyptus streams in Chile (by 8%) and
needed to weigh each effect size by its precision, in Uruguay (by 54%); pH was significantly higher
was estimated from the standard deviation and in eucalyptus streams in South Brazil (by 4%) and
sample size associated with each mean value. lower in Uruguay (by 7%); and SRP concentration
The overall response of the variable of interest to was significantly lower in eucalyptus streams in
eucalyptus plantations was determined using a Kenya (by 74%) and in Uruguay (by 73%) (Table 2
random-effects model of meta-analysis; between- and Table S1).
study variance was estimated by the restricted
maximum likelihood method (Borenstein and Leaf Litter Decomposition
others 2009). The effects of mesh size (that is, type
Alder leaf litter decomposition rates varied 15-fold
of decomposer community) and region on the
in FM bags across regions, from 0.00024/dd in na-
magnitude and direction of the response of the
tive streams in Central Brazil to 0.0037/dd in native
variable of interest to eucalyptus plantations were
streams in Kenya, and sixfold in CM bags, from
assessed by subgroup analyses. For the interpreta-
0.0009/dd in native streams in Central Brazil to
tion of the results, R values were converted into
0.0049/dd in native streams in Portugal, being
percentage of change with 0% indicating no effect
generally higher in CM than in FM bags (kCM/kFM,
of eucalyptus plantations, negative values indicat-
range: 1.0–3.9, mean: 2.2) (Figure 2).
ing percentage decrease and positive values indi-
The overall response of litter decomposition rates
cating percentage increase of the variable of
to eucalyptus plantations was an apparent decrease
interest in eucalyptus streams. A significant effect
by 11%, which was non-significant (95% CL: - 24
existed if the 95% CL did not include 0, and effects
to 5%) (Figure 3), but the percentage of total
were significantly different if their 95% CLs did not
variation explained by between-study variation
overlap (QM, p < 0.050; Borenstein and others
was high (I2 = 93%), suggesting that differences in
2009). The percentage of total variability that was
mesh size and region may moderate the response of
due to between-study variation rather than sam-
litter decomposition to forest change. When re-
pling error (I2) was also calculated (Borenstein and
sponses were considered per mesh size, no signifi-
others 2009). Analyses were performed with
cant effect was found for FM bags (95% CL: - 16 to
OpenMEE software (Wallace and others 2017).
29%), whereas an overall significant decrease by
23% (95% CL: - 39 to - 3%) was found for CM
RESULTS bags (Figure 3), although the difference between
mesh sizes was non-significant (QM = 3.586, df = 1,
Stream Characteristics
p = 0.058). The percentage of total variation ex-
Overall channel width was significantly lower in plained by between-study variation was, however,
eucalyptus than in native streams, although dif- high for both FM and CM bags (I2 = 93% and 92%,
ferences between stream types were consistently respectively).
non-significant within each region (Table 2 and The response of litter decomposition to planta-
Table S1). Overall water depth and dissolved inor- tions significantly differed across regions for FM
ganic nitrogen (DIN) did not significantly differ bags (QM = 25.238, df = 6, p < 0.001), with a sig-
between stream types, although eucalyptus streams nificant increase in litter decomposition in euca-
were significantly deeper than native streams in lyptus streams by 110% (95% CL: 11–300%) for
Uruguay (by 40%) and shallower in Portugal (by Central Brazil and by 32% (95% CL: 13–54%) for
636 V. Ferreira and others

Figure 2. Litter decomposition rates (mean ± standard error, n = 3) in fine mesh and coarse mesh bags incubated in
native (N) and eucalyptus (E) streams in seven regions.

Uruguay and nonsignificant effects for the litter decomposition by affecting shredder com-
remaining regions (for Kenya the effect was mar- munities. Microbial-driven litter decomposition
ginally non-significant: mean decrease by 48%; may also respond to eucalyptus plantations, espe-
95% CL: - 73 to 0.2%) (Figure 3). cially if these are accompanied by changes in dis-
For CM bags, the response of litter decomposition solved nutrient availability, litter characteristics or
to plantations did not significantly differ across re- solar irradiation. The present study builds up on the
gions (QM = 7.498, df = 6, p = 0.277), although a meta-analysis by Ferreira and others (2016) by
significant decrease was found for Spain (- 41%; showing that, indeed, results from studies in the
95% CL: - 52 to - 27%), South Brazil (- 31%; Iberian Peninsula cannot be extrapolated to other
95% CL: - 41 to - 19%) and Uruguay (- 36%; regions since there are regions where total litter
95% CL: - 49 to - 20%), while no significant ef- decomposition was not overall affected by euca-
fects were found for the remaining regions (for lyptus plantations and there are regions where
Portugal the effect was marginally non-significant: microbial-driven litter decomposition was affected
mean inhibition by 50%; 95% CL: - 75 to 3%) by plantations, contrarily to what has been ob-
(Figure 3). There was a high variation in some re- served for the Iberian Peninsula.
gions (for example, Central Brazil for FM bags and There was an overall reduction of total litter
Kenya, Central Brazil and Chile for CM bags; Fig- decomposition (that is, in CM bags) in streams
ure 3) suggesting that local (stream) conditions flowing through eucalyptus plantations, whereas
moderate the response of litter decomposition to there were no overall effects on microbial-driven
forest change. litter decomposition (that is, in FM bags). This
agrees with results from the largest previous study
comparing leaf litter decomposition between native
DISCUSSION
and eucalyptus streams in the Iberian Peninsula (5
Our first global assessment of the effects of euca- stream pairs in each of Portugal and Spain; Ferreira
lyptus plantations on stream ecosystem functioning and others 2015), which reported a mean decrease
shows that such effects depend on the type of by 42% for litter decomposition in CM bags in
decomposer communities (microbes only or eucalyptus streams, while litter decomposition in
microbes + macroinvertebrates) and vary among FM bags was less affected (mean decrease by 7%).
different regions across the distribution area of Also, a meta-analysis of 10 primary studies assess-
eucalyptus plantations. The observed variation in ing the effects of eucalyptus plantations on leaf
the response of litter decomposition to eucalyptus litter decomposition in Iberian streams showed a
plantations suggests that we cannot make gener- stronger decrease of litter decomposition in CM
alizations across regions. In regions where shred- bags (mean decrease by 36%) than in FM bags
ders are abundant (as indicated in previous (mean decrease by 7%) (Ferreira and others 2016).
studies), eucalyptus plantations seem to inhibit Thus, our present results are consistent with those
 Eucalyptus Plantations and Stream Functioning 637

crease in species with long life cycles, large body

sizes and feeding on coarse particulate organic
matter in eucalyptus streams (Larrañaga and others
2009a). These changes can be partially explained
by increased intermittency in eucalyptus streams. It
has been shown that the density of benthic
macroinvertebrates (and of shredders in particular)
decreases with decreasing flow permanence and
increasing duration and frequency of drying
events, leading to a reduction of litter decomposi-
tion rates (Datry and others 2011).
Changes on benthic macroinvertebrate commu-
nities in eucalyptus streams may also be due to
changes in the characteristics of litter inputs (Lar-
rañaga and others 2014). For instance, streams
flowing through mixed native deciduous forests in
the Iberian Peninsula receive most of the litter fall
in autumn and winter; this diverse litter input is
composed of litter species with distinct character-
istics, ranging from soft, nutrient-rich to tough,
nutrient-poor litter (Abelho and Graça 1996; Pozo
and others 1997; Molinero and Pozo 2004; some of
these streams were used in the present study). On
the contrary, streams flowing through E. globulus
plantations receive litter throughout the year, with
a peak in summer (Abelho and Graça 1996; Pozo
and others 1997; Molinero and Pozo 2004). This
accumulation of eucalyptus litter in the benthos at
a time of low flow may contribute to the formation
Figure 3. Response of litter decomposition to eucalyptus of pools with detrimental conditions for aquatic
plantations (% change), overall, per mesh size and per live, that is, high concentration of toxic leaf lea-
region. The dashed line (0%) indicates no effect of chates (Canhoto and Laranjeira 2007; Canhoto and
eucalyptus plantations, positive values indicate
others 2013; Gama and others 2014). Also, euca-
percentage increase, and negative values indicate
percentage decrease of litter decomposition in
lyptus leaves have high concentration of secondary
eucalyptus streams. Significant effects exist when the compounds with antimicrobial activity (for exam-
95% CL does not include 0 (black symbols), and effects ple, polyphenols and essential oils) and a tick waxy
significantly differ between levels within each moderator cuticle (Molinero and others 1996; Pozo and others
if their 95% CLs do not overlap. CM coarse mesh, FM fine 1997; Canhoto and Graça 1999; Molinero and Pozo
mesh. 2006; Hepp and others 2009), which makes them a
poor resource for shredders (Canhoto and Graça
reported previously for the Iberian Peninsula, 1995, 1999). The sub-optimal eucalyptus resource,
which suggested that the effects of eucalyptus when compared with some native litter species,
plantations on litter decomposition are mediated may lead to reduced growth and survival of
through effects on shredders. shredders (Canhoto and Graça 1995; Correa-Ara-
In fact, taxon richness, density or biomass of neda and others 2015; Kiffer and others 2018),
benthic macroinvertebrates (including shredders) which in the long term may result in depleted
are generally lower in streams in E. globulus plan- macroinvertebrate communities.
tations than in streams in native deciduous forests Additionally, changes in benthic macroinverte-
in the Iberian Peninsula (Abelho and Graça 1996; brate communities may be driven by modifications
Larrañaga and others 2009a, b; Ferreira and others in fish communities induced by forest change. In-
2015; Cordero-Rivera and others 2017; some of creases in solar exposition (for example, by
these streams were used in the present study). Also, replacing wide canopy species by single-stemmed
changes in shredders trait distribution have been eucalyptus trees) can lead to increases in fish
reported when native deciduous forests are re- densities (Teixeira-de-Mello and others 2015) that
placed by E. globulus plantations, including a de- may prey on macroinvertebrates and reduce their
638 V. Ferreira and others

densities in eucalyptus streams (Mcintosh and with eucalyptus streams in other regions. The
others 2005). Given that fish species richness and strong stimulation of litter decomposition in some
densities are generally higher in subtropical than in eucalyptus streams in Chile could result from less
temperate regions (Teixeira-de-Mello and others recalcitrant litter in these streams than in native
2012), effects of forest change on fish communities streams where some of the endemic tree species
might be stronger in streams in the former regions. contribute with litter that is tough and has high
Although the effect of eucalyptus plantations on lignin and secondary compounds concentrations
total litter decomposition did not significantly differ (Muñoz-Concha and others 2004). This variation
across regions, it was significant or nearly signifi- in the effects of eucalyptus plantations on leaf litter
cant only for regions with temperate climate decomposition reinforces the suggestion that effects
(Spain, Portugal, South Brazil and Uruguay). This are stronger for streams where shredders are well
suggests that changes in macroinvertebrate com- represented, but may depend on local factors (Pozo
munities induced by eucalyptus plantations re- and others 1998; see above).
ported for temperate climates cannot be Litter decomposition in fine mesh bags was not
extrapolated to other climatic zones. The highly overall affected by eucalyptus plantations, but
variable effect of eucalyptus plantations on total there were significant differences across regions.
litter decomposition in Kenya, Central Brazil and Microbial decomposers generally differ between
Chile (for example, varying from strong decrease, streams in E. globulus plantations and streams in
to no effect, to strong increase within a single re- native deciduous forests in the Iberian Peninsula
gion), which resulted in nonsignificant effects of (Chauvet and others 1997; Bärlocher and Graça
eucalyptus plantations at the regional level, may be 2002; Ferreira and others 2006; some of these
due to variation in local factors as discussed below. streams were used in the present study), but this is
Shredder abundance is low in some Kenyan not generally translated into altered microbial-dri-
streams (Dobson and others 2002), and thus, litter ven litter decomposition (Ferreira and others 2016;
decomposition could be primarily mediated by this study), suggesting high functional redundancy
microbes in these streams (Dobson and others in microbial communities in those areas. Contrarily
2004), while shredders contribute substantially to to what has been observed in the Iberian Peninsula,
litter decomposition in other streams (Boyero and microbial-driven litter decomposition was signifi-
others 2015). A negligible contribution of shredders cantly stimulated in eucalyptus streams in Central
to litter decomposition has been found also in some Brazil and Uruguay and nearly significantly inhib-
tropical streams in Central and South America, ited in Kenya, which suggests that functional
including the Brazilian savannah (Gonçalves and redundancy in microbial decomposer communities
others 2006, 2007; Camacho and others 2009; is not universal. In Kenya, microbial-driven litter
Boyero and others 2015), whereas shredder abun- decomposition was lower in eucalyptus streams
dance and richness seem to increase with elevation that had significantly lower soluble reactive phos-
in different tropical regions (Camacho and others phorus concentrations than native streams, which
2009; Yule and others 2009). This variation in may have hindered microbial activity and litter
shredder abundance in tropical streams may ex- decomposition in the former streams (Gulis and
plain why litter decomposition is inhibited by others 2006). In Central Brazil, microbial-driven
eucalyptus plantations in some streams whereas no litter decomposition was higher in eucalyptus than
effect is observed in others. Additionally, the higher in native streams likely due to the better quality of
litter decomposition in one of the eucalyptus eucalyptus than native leaves as mentioned above
streams in Kenya may be attributed to the much (Gomes and others 2018). In Uruguay, microbial-
higher discharge in this stream when compared driven litter decomposition was also higher in some
with the native stream (368 vs. 21 L/s). Also, the eucalyptus than in native streams, which may be
higher litter decomposition in eucalyptus streams partially explained by the fact that some eucalyptus
in Central Brazil could be due to eucalyptus leaves streams were exposed to solar irradiation since
being of higher nutritional quality than native plantations were not present in the stream banks
leaves (Gomes and others 2018), which may result (Figure 1). Exposure to solar irradiation, by allow-
in higher microbial inoculum potential in euca- ing the development of primary producers, may
lyptus streams. Eucalyptus plantations in Central have facilitated a priming effect by which microbial
Brazil were also more diverse than in other regions decomposers would benefit from labile carbon
(Table 1), which may have ensured a higher produced by algae and decompose more recalci-
diversity of substrates for aquatic communities in trant substrates (that is, litter) at faster rates
eucalyptus streams in this region when compared (Danger and others 2013; Kuehn and others 2014).
 Eucalyptus Plantations and Stream Functioning 639

Our results show that, even if eucalyptus streams #421632/2016-0 and Grant #305203/2017-7).
are surrounded by trees, providing shade and or- Additional financial support was provided by the
ganic matter, they endure multiple changes that Basque Government (Grant No. IT-302-10), the
result in altered ecosystem functioning, which may Water Research Centre for Agriculture and Mining
jeopardize their ability to provide ecosystem ser- (CHRIAM 1513001) and Initiation Project
vices. The ratio of litter decomposition rates at 11170390 from National Commission for Scientific
impacted and reference sites (kimp/kref), which is and Technological Research of the Chilean
equivalent to the response ratio used here (R = kE/ Government, the Fundação de Apoio à Pesquisa do
kN), was proposed as a metric to assess functional Distrito Federal (FAP-DF/Brazil; Edital 03/
stream integrity (Gessner and Chauvet 2002). We 2015—No. 193.000.870/2015), and the CNPq/
found that total litter decomposition (in CM bags) Brazil (Public call MCTI/CNPq No. 14/
should be addressed for an accurate evaluation of 2013—Universal Proc.: 471767/2013-1; CT-Hidro/
eucalyptus effects on ecosystem functioning, Climatic Changes/Water Resources/CNPq Proc.
whereas microbial-driven litter decomposition (in 403949/2013-0; Fellowship PQ No. 302957/2014-
FM bags) is preferable when shredders are rare or 6). Comments by two anonymous reviewers and
absent and physical abrasion by current and sedi- the subject-matter editor are also acknowledged.
ments is a possibility. The use of litter decomposi-
tion as a bioassessment tool has several advantages
compared with several methods based on structural Compliance with Ethical Standards
characteristics of stream ecosystems (Bonada and
others 2006). First, it is a fundamental ecosystem Conflict of interest The authors declare
process, and thus, it allows assessing stream func- that they have no conflict of interest.
tional integrity, which is intimately linked with the
ability of the system to provide services. Secondly, REFERENCES
it is not based on taxonomic identification, and Abelho M, Graça MAS. 1996. Effects of eucalyptus afforestation
thus, it allows cross-regional comparisons of forest on leaf litter dynamics and macroinvertebrate community
change effects on streams. In addition, it is easy to structure of streams in Central Portugal. Hydrobiologia
standardize and implement, even in comparison 324:195–204.
with other stream processes (von Schiller and Azevedo-Pereira HVS, Graça MAS, González JM. 2006. Life
history of Lepidostoma hirtum in an Iberian stream and its role
others 2017). Thus, the response ratio is a useful
in organic matter processing. Hydrobiologia 559:183–92.
approach to address forest change effects on stream
Bäckstrand K, Lövbrand E. 2006. Planting trees to mitigate cli-
ecosystem functioning (Ferreira and others 2015, mate change: contested discourses of ecological moderniza-
2016). This is especially relevant at a time of in- tion, green governmentality and civic environmentalism.
tense forest change with the increase in the area Glob Environ Polit 6:50–75.
occupied by commercial tree plantations, changes Bärlocher F, Graça MAS. 2002. Exotic riparian vegetation lowers
in forest composition due to climate change (for fungal diversity but not leaf decomposition in Portuguese
streams. Freshw Biol 47:1123–35.
example, increase in drought-tolerant and decrease
Bonada N, Prat N, Resh VH, Statzner B. 2006. Developments in
in drought-sensitive species in a scenario of
aquatic insect biomonitoring: a comparative analysis of recent
increasing aridity) and invasion of native forests by approaches. Annu Rev Entomol 51:495–523.
exotic species (Kominoski and others 2013). Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. 2009.
Introduction to meta-analysis. Chichester: Wiley. p 421.
ACKNOWLEDGEMENTS Boyero L, Pearson RG, Dudgeon D, Graça MAS, Gessner MO,
Albarino RJ, Ferreira V, Yule CM, Boulton AJ, Arunachalam
This study was financed by the Portuguese Foun- M, Callisto M, Chauvet E, Ramirez A, Chara J, Moretti MS,
dation for Science and Technology (FCT), through Goncalves JF Jr, Helson JE, Chará-Serna AM, Encalada AC,
the strategic Project UID/MAR/04292/2013 granted Davies JN, Lamothe S, Cornejo A, Li AOY, Buria LM, Vil-
to MARE and through financial support given to VF lanueva VD, Zuniga MC, Pringle CM. 2011a. Global distribu-
tion of a key trophic guild contrasts with common latitudinal
(SFRH/BPD/76482/2011; IF/00129/2014). FTM
diversity patterns. Ecology 92:1839–48.
was supported by the Sistema Nacional de Investi-
Boyero L, Pearson RG, Gessner MO, Barmuta LA, Ferreira V,
gación-Agencia Nacional de Investigación e Inno- Graça MAS, Dudgeon D, Boulton AJ, Callisto M, Chauvet E,
vación (SNI-ANII) and Programa de Desarrollo de Helson JE, Bruder A, Albariño RJ, Yule CM, Arunachalam M,
las Ciencias Básicas (PEDECIBA)-Geociencias. LUH Davies JN, Figueroa R, Flecker AS, Ramı́rez A, Death RG,
received financial support from the Conselho Na- Iwata T, Mathooko JM, Mathuriau C, Gonçalves JF Jr, Moretti
MS, Jingut T, Lamothe S, M’Erimba C, Ratnarajah L,
cional de Desenvolvimento Cientı́fico e Tec-
Schindler MH, Castela J, Buria LM, Cornejo A, Villanueva VD,
nológico (CNPq/Brazil; Edital CNPq/Universal West DC. 2011b. A global experiment suggests climate
640 V. Ferreira and others

warming will not accelerate litter decomposition in streams Dobson M, Mathooko JM, Ndegwa FK, M’Erimba C. 2004. Leaf
but might reduce carbon sequestration. Ecol Lett 14:289–94. litter processing rates in a Kenyan highland stream, the Njoro
Boyero L, Pearson RG, Gessner MO, Dudgeon D, Ramı́rez A, River. Hydrobiologia 519:207–10.
Yule CM, Callisto M, Pringle CM, Encalada AC, Arunachalam Fernández C, Vega JA, Gras JM, Fonturbel T. 2006. Changes in
M, Mathooko J, Helson JE, Rincón J, Brider A, Cornejo A, water yield after a sequence of perturbations and forest
Flecker AS, Mathuriau C, M’Erimba C, Gonçalves JF Jr, management practices in an Eucalyptus globulus Labill. water-
Moretti M, Jinggut T. 2015. Leaf-litter breakdown in tropical shed in Northern Spain. For Ecol Manag 234:275–81.
streams: is variability the norm? Freshw Sci 34:759–69. Ferreira V, Elosegi A, Gulis V, Pozo J, Graça MAS. 2006. Euca-
Calder IR. 2002. Eucalyptus, water and the environment. In: lyptus plantations affect fungal communities associated with
Coppen JJW, Ed. Eucalyptus. The genus Eucalyptus. London: leaf litter decomposition in Iberian streams. Archiv für
Taylor and Francis. p 36–51. Hydrobiologie 166:467–90.
Camacho R, Boyero L, Cornejo A, Ibáñez A, Pearson RG. 2009. Ferreira V, Encalada AC, Graça MAS. 2012. Effects of litter
Local variation in shredder distribution can explain their diversity on decomposition and biological colonization of
oversight in tropical streams. Biotropica 41:625–32. submerged litter in temperate and tropical streams. Freshw Sci
Canhoto C, Graça MAS. 1995. Food value of introduced eucalypt 31:945–62.
leaves for a Mediterranean stream detritivore: Tipula lateralis. Ferreira V, Koricheva J, Pozo J, Graça MAS. 2016. A meta-
Freshw Biol 34:209–14. analysis on the effects of changes in the composition of native
Canhoto C, Graça MAS. 1996. Decomposition of Eucalyptus forests on litter decomposition in streams. For Ecol Manag
globulus leaves and three native leaf species (Alnus glutinosa, 364:27–38.
Castanea sativa and Quercus faginea) in a Portuguese low order Ferreira V, Larrañaga A, Gulis V, Basaguren A, Elosegi A, Graça
stream. Hydrobiologia 333:79–85. MAS, Pozo J. 2015. The effects of eucalypt plantations on
Canhoto C, Graça MAS. 1999. Leaf barriers to fungal coloniza- plant litter decomposition and macroinvertebrate communi-
tion and shredders (Tipula lateralis) consumption of decom- ties in Iberian streams. For Ecol Manag 335:129–38.
posing Eucalyptus globulus. Microb Ecol 37:163–72. Gama M, Guilhermino L, Canhoto C. 2014. Comparison of three
Canhoto C, Laranjeira C. 2007. Leachates of Eucalyptus globulus shredders response to acute stress induced by eucalyptus leaf
in intermittent streams affect water parameters and inverte- leachates and copper: single and combined exposure at two
brates. Int Rev Hydrobiol 92:173–82. distinct temperatures. Ann de Limnol Int J Limnol 50:97–107.
Canhoto C, Calapez R, Gonçalves AL, Moreira-Santos M. 2013. Gerber J-F. 2011. Conflicts over industrial tree plantations in the
Effects of Eucalyptus leachates and oxygen on leaf-litter pro- South: who, how and why? Glob Environ Change 21:165–76.
cessing by fungi and stream invertebrates. Freshw Sci 32:411– Gessner MO, Chauvet E. 2002. A case for using litter breakdown
24. to assess functional stream integrity. Ecol Appl 12:498–510.
Casotti CG, Kiffer WP Jr, Moretti MS. 2014. Leaf traits induce Gomes PP, Ferreira V, Tonin AM, Medeiros AO, Júnior JF Jr.
the feeding preference of a shredder of the genus Triplectides 2018. Combined effects of dissolved nutrients and oxygen on
Kolenati, 1859 (Trichoptera) in an Atlantic Forest stream, plant litter decomposition and associated fungal communities.
Brazil: a test with native and exotic leaves. Aquat Insects Microb Ecol 75:854–62.
36:43–52. Gonçalves JF Jr, Graça MAS, Callisto M. 2006. Leaf-litter
Chauvet E, Fabre E, Elosegui A, Pozo J. 1997. The impact of breakdown in 3 streams in temperate, Mediterranean, and
eucalypt on the leaf-associated aquatic hyphomycetes in tropical Cerrado climates. J N Am Benthol Soc 25:344–55.
Spanish streams. Can J Bot 75:880–7. Gonçalves JF Jr, Graça MAS, Callisto M. 2007. Litter decompo-
Cordero-Rivera A, Álvarez AM, Álvarez M. 2017. Eucalypt sition in a Cerrado savannah stream is retarded by leaf
plantations reduce the diversity of macroinvertebrates in small toughness, low dissolved nutrients and a low density of
forested streams. Anim Biodivers Conserv 40:87–97. shredders. Freshw Biol 52:1440–51.
Cornut J, Elger A, Lambrigot D, Marmonier P, Chauvet E. 2010. Gonçalves JF Jr, Rezende RS, França J, Callisto M. 2012a.
Early stages of leaf decomposition are mediated by aquatic Invertebrate colonisation during leaf processing of native,
fungi in the hyporheic zone of woodland streams. Freshw Biol exotic and artificial detritus in a tropical stream. Mar Freshw
55:2541–56. Res 63:428–39.
Correa-Araneda F, Boyero L, Figueroa R, Sánchez C, Abdala R, Gonçalves JF Jr, Rezende RS, Martins NM, Gregório RS. 2012b.
Ruiz-Garcı́a A, Graça MAS. 2015. Joint effects of climate Leaf breakdown in an Atlantic Rain Forest stream. Aust Ecol
warming and exotic litter (Eucalyptus globulus Labill.) on 37:807–15.
stream detritivore fitness and litter breakdown. Aquat Sci Graça MAS, Cressa C. 2010. Leaf quality of some tropical and
77:197–205. temperate tree species as food resource for stream shredders.
Danger M, Cornut J, Chauvet E, Chavez P, Elger A, Lecerf A. Int Rev Hydrobiol 95:27–41.
2013. Benthic algae stimulate leaf litter decomposition in Graça MAS, Cressa C, Gessner MO, Feio MJ, Callies KA, Barrios
detritus-based headwater streams: a case of aquatic priming C. 2001. Food quality, feeding preferences, survival and
effect? Ecology 94:1604–13. growth of shredders from temperate and tropical streams.
Datry T, Corti R, Claret C, Philippe M. 2011. Flow intermittence Freshw Biol 46:947–57.
controls leaf litter breakdown in a French temporary alluvial Graça MAS, Poquet JM. 2014. Do climate and soil influence
river: the ‘‘drying memory’’. Aquat Sci 73:471–83. phenotypic variability in leaf litter, microbial decomposition
Dobson M, Magana A, Mathooko JM, Ndegwa FK. 2002. and shredder consumption? Oecologia 174:1021–32.
Detritivores in Kenyan highland streams: more evidence for Graça MAS, Pozo J, Canhoto C, Elosegi A. 2002. Effects of Eu-
the paucity of shredders in the tropics? Freshw Biol 47:909– calyptus plantations on detritus, decomposers, and detritivores
19. in streams. Sci World J 2:1173–85.
 Eucalyptus Plantations and Stream Functioning 641

Gulis V, Ferreira V, Graça MAS. 2006. Stimulation of leaf litter Molinero J, Pozo J, González E. 1996. Litter breakdown in
decomposition and associated fungi and invertebrates by streams of the Agüera catchment: influence of dissolved
moderate eutrophication: implications for stream assessment. nutrients and land use. Freshw Biol 36:745–56.
Freshw Biol 51:1655–69. Molinero J, Pozo J. 2004. Impact of a eucalyptus (Eucalyptus
Gulis V, Suberkropp K. 2003. Leaf litter decomposition and globulus Labill.) plantation on the nutrient content and
microbial activity in nutrient-enriched and unaltered reaches dynamics of coarse particulate organic matter (CPOM) in a
of a headwater stream. Freshw Biol 48:123–34. small stream. Hydrobiologia 528:143–65.
Gurevich J. 2013. Meta-analysis of results from multisite studies. Molinero J, Pozo J. 2006. Organic matter, nitrogen and phos-
In: Koricheva J, Gurevich J, Mengersen K, Eds. Handbook of phorus fluxes associated with leaf litter in two small streams
meta-analysis in ecology and evolution. Princeton: Princeton with different riparian vegetation: a budget approach. Archiv
University Press. p 313–20. für Hydrobiologie 166:363–85.
Hedges LV, Gurevitch J, Curtis PS. 1999. The meta-analysis of Muñoz-Concha D, Voguel H, Razmilic I. 2004. Variación de
response ratios in experimental ecology. Ecology 80:1150–6. compuestos quı́micos en hojas de poblaciones de Drimys spp.
Hepp LU, Delanora R, Trevisan A. 2009. Compostos secundários (Magnoliophyta: Winteraceae) en Chile. Revista Chilena de
durante a decomposição foliar de espécies arbóreas em um Historia Natural 77:43–50.
riacho do sul do Brasil. Acta Botanica Brasilica 23:407–13. Payn T, Carnus JM, Freer-Smith P, Kimberley M, Kollert W, Liu
Hieber M, Gessner MO. 2002. Contribution of stream detriti- S, Orazio C, Rodriguez L, Silva LN, Wingfield MJ. 2015.
vores, fungi, and bacteria to leaf breakdown based on biomass Changes in planted forests and future global implications. For
estimates. Ecology 83:1026–38. Ecol Manag 352:57–67.
ICNF. 2013. IFN6 – Áreas dos usos do solo e das espécies flo- Pozo J, Basaguren A, Elosegui A, Molinero J, Fabre E, Chauvet
restais de Portugal continental. Resultados preliminares. Lis- E. 1998. Afforestation with Eucalyptus globulus and leaf litter
bon: Instituto de Conservação da Natureza e das Florestas. p decomposition in streams of northern Spain. Hydrobiologia
34. 373(374):101–9.
IFN3. 2007. Tercer inventario forestal nacional. Madrid: Minis- Pozo J, González E, Dı́ez JR, Molinero J, Elosegui A. 1997. In-
terio de Agricultura, Alimentación y Medio Ambiente. puts of particulate organic matter to streams with different
Iglesias-Trabado G, Wilstermann D. 2009. Eucalyptus Univer- riparian vegetation. J N Am Benthol Soc 16:602–11.
salis. Global cultivated eucalypt forests map 2009. GIT For- Rezende RS, Petrucio MM, Gonçalves JF Jr. 2014. The effects of
estry Consulting’s Eucalyptologics: Information resources on spatial scale on breakdown of leaves in a tropical watershed.
Eucalyptus cultivation worldwide. Available at http://www.g PlosOne 9:e97072.
it.forestry.com/. Rustad LE, Campbell JL, Marion GM, Norby RJ, Mitchell MJ,
Kiffer WP Jr, Mendes F, Casotti CG, Costa LC, Moretti MS. 2018. Hartley AE, Cornelissen JHC, Gurevitch J, GCTE-NEWS.
Exotic Eucalyptus leaves are preferred over tougher native 2001. A meta-analysis of the response of soil respiration, net
species but affect the growth and survival of shredders in an nitrogen mineralization, and aboveground plant growth to
Atlantic Forest stream (Brazil). PLoS ONE 13:e0190743. experimental ecosystem warming. Oecologia 126:543–62.
Kominoski JS, Shah JJF, Canhoto C, Fischer DG, Giling DP, Santiago J, Molinero J, Pozo J. 2012. Impact of timber harvesting
González E, Griffiths NA, Larrañaga A, LeRoy CJ, Mineau on litterfall inputs and benthic coarse particulate organic
MM, McElarney YR. 2013. Forecasting functional implications matter (CPOM) storage in a small stream draining a euca-
of global changes in riparian plant communities. Front Ecol lyptus plantation. For Ecol Manage 262:1146–56.
Environ 11:423–32. Scott DF, Lesch W. 1997. Streamflow responses to afforestation
Kuehn KA, Francoeur SN, Findlay RH, Neely RK. 2014. Priming with Eucalyptus grandis and Pinus patula and to felling in the
in the microbial landscape: periphytic algal stimulation of Mokobulaan experimental catchments, South Africa. J Hydrol
litter-associated microbial decomposers. Ecology 95:749–62. 199:360–77.
Lara A, Little C, Urrutia R, McPhee J, Álvarez-Garretón C, Siegloch AE, Suriano M, Spies M, Fonseca-Gessner A. 2014.
Oyarzún C, Soto D, Donoso P, Nahuelhual L, Pino M, Aris- Effect of land use on mayfly assemblages structure in
mendi I. 2009. Assessment of ecosystem services as an Neotropical headwater streams. Anais da Academia Brasileira
opportunity for the conservation and management of native de Ciencias 86:1735–47.
forests in Chile. For Ecol Manag 258:415–24. Teixeira-de-Mello F, Meerhoff M, Baattrup-Pedersen A, Mai-
Larrañaga A, Basaguren A, Elosegi A, Pozo J. 2009a. Impacts of gaard T, Kristensen PB, Andersen TK, Clemente JM, Fosalba
Eucalyptus globulus plantations on Atlantic streams: changes in C, Kristensen EA, Masdeu M, Riis T, Mazzeo N, Jeppesen E.
invertebrate density and shredders traits. Fundam Appl Lim- 2012. Community structure of fish in lowland streams differ
nol/Archiv für Hydrobiologie 175:151–60. substantially between subtropical and temperate climates.
Larrañaga A, Basaguren A, Pozo J. 2009b. Impacts of Eucalyptus Hydrobiologia 684:143–60.
globulus plantations on physiology and population densities of Teixeira-de-Mello F, Meerhoff M, González-Bergonzoni I, Kris-
invertebrates inhabiting Iberian Atlantic streams. Int Rev tensen EA, Baattrup-Pedersen A, Jeppesen E. 2015. Influence
Hydrobiol 94:497–511. of riparian forests on fish assemblages in temperate lowland
Larrañaga S, Larrañaga A, Basaguren A, Elosegi A, Pozo J. 2014. streams. Environ Biol Fishes 99:133–44.
Effects of exotic eucalypt plantations on organic matter pro- Vannote RL, Minshall GW, Cummins KW, Sedell JR, Cushing
cessing in Iberian streams. Int Rev Hydrobiol 99:363–72. CE. 1980. The river continuum concept. Can J Fish Aquat Sci
Mcintosh AR, Greig HS, Mcmurtrie SA, Nystrøm PER, Winter- 37:130–7.
bourn MJ. 2005. Top-down and bottom-up influences on von Schiller D, Acuña V, Aristi I, Arroita M, Basaguren A, Bellin
populations of a stream detritivore. Freshw Biol 50:1206–18. A, Boyero L, Butturini A, Ginebreda A, Kalogianni E, Larra-
ñaga A, Majone B, Martı́nez A, Monroy S, Muñoz I, Paunovic
642 V. Ferreira and others

M, Pereda O, Petrovic M, Pozo J, Rodrı́guez-Mozaz S, Rivas D, source software for meta-analysis in ecology and evolutionary
Sabater S, Sabater F, Skoulikidis N, Solagaistua L, Vardakas L, biology. Methods Ecol Evol 8:941–7.
Elosegi A. 2017. River ecosystem processes: a synthesis of Woodward G, Gessner MO, Giller PS, Gulis V, Hladyz S, Lecerf
approaches, criteria of use and sensitivity to environmental A, Malmqvist B, McKie BG, Tiegs SD, Cariss H, Dobson M,
stressors. Sci Total Environ 595–597:465–80. Elosegi A, Ferreira V, Graça MAS, Fleituch T, Lacoursière JO,
Walker MD, Wahren CH, Hollister RD, Henry GHR, Ahlquist LE, Nistorescu M, Pozo J, Risnoveanu G, Schindler M, Vadineanu
Alatalo JM, Bret-Harte MS, Calef MP, Callaghan TV, Carroll A, Vought LBM, Chauvet E. 2012. Continental-scale effects of
AB, Epstein HE, Jónsdóttir IS, Klein JA, Magnússon B, Molau nutrient pollution on stream ecosystem functioning. Science
U, Oberbauer SF, Rewa SP, Robinson CH, Shaver GR, Suding 336:1438–40.
KN, Thompson CC, Tolvanen A, Totland Ø, Turner PL, Yule CM, Leong MY, Liew KC, Ratnarajah L, Schmidt K, Wong
Tweedie CE, Webber PJ, Wookey PA. 2006. Plant community HM, Pearson RG, Boyero L. 2009. Shredders in Malaysia:
responses to experimental warming across the tundra biome. abundance and richness are higher in cool upland tropical
Proc Natl Acad Sci 103:1342–6. streams. J N Am Benthol Soc 28:404–15.
Wallace JB, Eggert SL, Meyer JL, Webster JR. 1997. Multiple Zomer RJ, Trabucco A, Bossio DA, Verchot LV. 2008. Climate
trophic levels of a forest stream linked to terrestrial litter in- change mitigation: a spatial analysis of global land suitability
puts. Science 277:102–4. for clean development mechanism afforestation and refor-
Wallace BC, Lajeunesse MJ, Dietz G, Dahabreh IJ, Trikalinos TA, estation. Agr Ecosyst Environ 126:67–80.
Schmid CH, Gurevitch J. 2017. OpenMEE: intuitive, open-