Content uploaded by Max Mallen-Cooper
All content in this area was uploaded by Max Mallen-Cooper on May 08, 2022
Content may be subject to copyright.
Global Ecol Biogeogr. 2022;00:1–12.
Received: 7 December 2021
Revised: 12 April 2022
Accepted: 19 April 2022
DOI : 10.1111/geb .1352 2
1,2 |1|1|1,2 |
1Ecology and Evolution Research
Centre, School of Biological, Earth and
Environmental Sciences, University of
New South Wales, Sydney, New South
2Centre for Ecosystem Science, School
of Biological, Earth and Environmental
Sciences, University of New South Wales,
Sydney, New South Wales, Australia
Max Mallen- Cooper, Ecology and
Evolution Research Centre, School of
Biological, Earth and Environmental
Sciences, University of New South Wales,
Sydney, NSW 2052, Australia.
Naia Morueta- Holme
Eucalypts have a widespread global distribution owing to their popularity for
agroforestry and as environmental plantings. Despite an abundance of site- specific
evidence that eucalypts modify soils and soil processes, we lack a quantitative syn-
thesis of their overall effects at the global scale. This limits our capacity to assess the
likely impacts of future introductions in any given region of the world.
Eucalyptus, Angophora and Corymbia.
We used a systematic search to derive a database of empirical data from
227 studies across 33 countries (neffect size = 2,806) and tested three predictions about
the effects of eucalypts on soil proper ties and whethe r thes e effe c ts varied with plan-
tation age and soil depth.
Compared with (non- eucalypt) native vegetation, eucalypts significantly
reduced soil moisture, microbial abundance, nitrogen, cations and anions. Relative
reductions in soil microbes and ions were stronger in older eucalypt plantations. A
comparison of eucalypts with (non- eucalypt) silvicultural and agropastoral systems
revealed similar effects on most soil properties, although eucalypts tended to reduce
potassium and enhance carbon to a greater extent than other managed systems. We
found no consistent effects of eucalypts on soil pH.
Our study provides the first extensive global meta- analysis of the
effects of eucalypts on soil properties and processes and demonstrates that effects
are highly dependent on the community with which they are compared (i.e., natural
or managed). In general, our findings reinforce the widely held belief that eucalypts
deplete soil nutrients and dominate water resources. Understanding how eucalypts
affect soils allows us to assess their global suitability for agroforestry, soil rehabilitation
and soil carbon enhancement, while considering the potential environmental costs.
acidity, carbon, eucalypt, meta- analysis, nitrogen, nutrients, soil function
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in
any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2022 The Authors. Global Ecology and Biogeography published by John Wiley & Sons Ltd.
M ALLEN- COO PER Et AL .
Increasing globalization is associated with the spread of many
plants and animals beyond their natural ranges (Hobbs, 2000).
This trend has occurred naturally or by accident (e.g., zebra mus-
sels; Ricciardi, 2003) or intentionally, whereby a species is actively
encouraged in order to meet an agricultural, economic or social
objective (e.g., cane toads; Easteal, 1981). The impacts of intro-
duced species on local socio- ecological systems are highly context
and organism dependent and range from potentially beneficial
(Walther et al., 2009) to environmentally or economically detri-
mental (Bradshaw et al., 2021). Eucalypts, which include Eucalyptus,
Angophora and Corymbia spp., have been hugely successful at
establishing beyond their natural range (Stanturf et al., 2013). They
occur naturally in Australia and the drier areas of Papua New Guinea
and were originally introduced to Europe as ornamental plants be-
fore their rapid expansion around the globe, largely attributable to
their value as timber, pulp, firewood, charcoal and other products
(Stanturf et al., 2013; Turnbull, 1999; Williams & Brooker, 1997).
Eucalypt plantations now cover c. 19 million hectares across the
Americas, Africa, Europe, the Middle East and Asia (Iglesias- Trabado
& Wilstermann, 2008; Stanturf et al., 2013; Turnbull, 1999), repre-
senting c. 6.5% of planted forests globally (FAO, 2020).
Eucalypts are highly prized in the silvicultural industry owing
to their pest resistance, rapid growth and timber production under
a wide range of soil and climatic conditions (Eldridge et al., 1993;
Zaiton et al., 2020). They are also valued for their ability to control
erosion, reduce waterlogging and rehabilitate saline and sodic soils
(Jagger & Pender, 2003; Mishra et al., 2003; Teketay, 2000; Zohar
et al., 2008), and have also been promoted widely for their eco-
nomic benefits to rural smallholders (Jagger & Pender, 2003; Jaleta
et al., 2016). Yet, despite these positive benefits, there is increasing
anecdotal and empirical evidence that, beyond their native range,
eucalypts have substantial negative effects on soils and ecological
processes. These negative effects range from soil nutrient (Zhang
et al., 2015) and groundwater (Christina et al., 2017) depletion to
soil surface hydrophobicity (Burch et al., 1989) and acidification
(Jobbágy & Jackson, 2003; Rhoades & Binkley, 1996) and allelo-
pathic effects on native vegetation (Chu et al., 20 14; del Moral &
Muller, 1970; Zhang & Fu, 2009). The extent to which eucalypts af-
fect soils and soil function, however, is not well understood across
their entire range, largely because these effects are strongly con-
text dependent and therefore likely to vary markedly with soil and
climatic conditions, tree or plantation age and, potentially, species
identity. For example, in pasture systems that were afforested with
eucalypts, Turner and Lambert (2000) found a consistent decrease
in soil carbon (C) in Australia, whereas Lima et al. (2006) reported
the opposite pattern in Brazil. The lack of a global consensus on the
effects of eucalypts on soils makes it difficult to characterize their
global suitability for agroforestry or rehabilitating degraded soils or
their potential to provide solutions to hydrological problems asso-
ciated with soil salinity and waterlogging. Understanding these soil
effects is particularly important when considering the potential for
eucalypts outside their native range to increase the likelihood of en-
vironmental and ecological damage or reductions in soil functions.
Here, we examine the impacts of eucalypts on soil processes
and properties using a synthetic database derived from 227 stud-
ies published world- wide between 1986 and 2021. Our study fo-
cused on four main hypotheses. First, we expected that eucalypts
would reduce soil N, P, K, moisture, pH, cations and microbial ac-
tivity relative to both native vegetation and agrosilvicultural sys-
tems. Many tree taxa resorb N, P and K from the leaves back into
the tree before leaf abscission, resulting in little return to the soil via
leaf litter decomposition, yet eucalypts have a particularly high re-
translocation efficiency, resorbing on average 56% of N and 98% of
P (Killingbeck, 1996; Saur et al., 2000). For this reason, we predicted
soil beneath eucalypts to be relatively low in N, P and K, although
certain management practices, such as fertilization, have the po-
tential to offset this effect (Gonçalves et al., 2004). Likewise, soil
moisture would be expected to be lower beneath eucalypts because
of their high requirement for water (Madalcho et al., 2019), prolif-
eration of hydrophobic plant residues (Burch et al., 1989) and great
ab i lity to ex trac t gro und w ater th r ough de ep ro ot syste ms, of ten low -
ering water table depths by several metres (Christina et al., 2017 ).
There are numerous reports of eucalypts acidifying the soil (Leite
et al., 2010; Rhoades & Binkley, 1996; Soumare et al., 2016), through
a combination of mechanisms including cation redistribution, root
respiration and litter leachates (Jobbágy & Jackson, 2003), and this
acidification is likely to drive a reduction in microbial activity (Curtin
et al., 1998; Iovieno et al., 2 010). Furthermore, the retention of cat-
ions in living eucalypt biomass is likely to deplete cations in the soil,
although some might be returned to the topsoil via decomposition of
leaf litter (Jobbágy & Jackson, 20 03).
Our second hypothesis was that soils under eucalypts would be,
on average, more C rich than other agrosilvicultural systems owing
to higher rates of litter production (Demessie et al., 2012; Paul
et al., 2002; Sangha et al., 2006), yet C poor relative to native veg-
etation with leaf litter that tends to be decomposed more readily
(Bernhard- Reversat & Schwartz, 1997; Castro- Díez et al., 2012).
Third, we predicted that certain effects would vary with the
maturity of the eucalypt plantation. Older eucalypts have greater
canopy coverage and produce more litter than younger eucalypts,
resulting in stronger effects on soil C and rainsplash erosion (Chen
et al., 2013; Sun et al., 2018; Zou & Bashkin, 1998). The growth rate
of eucalypts is also much greater for younger trees, leading to inten-
sified nutrient depletion and water extraction (Forrester et al., 2010).
Fourth, we expected that soil effects relating to C and nutrients
(N, P, K, other cations and anions) and their recycling (microbes and
enzymes) would be highest in the uppermost soil profile, where lit-
ter enters the soil and microbial activity is typically highest (Blume
et al., 2002; Fang et al., 2005).
Despite the fact that eucalypts have an extensive global distribu-
tion and are highly prized commercially, there has been no quantita-
tive global synthesis of their effects on specific soil properties. Some
effe cts of euc aly pts on broad eco sys tem ser vices (e.g ., soil for matio n)
were examined in a synthesis by Castro- Díez et al. (2021), yet this
MALLEN- COOPER Et AL .
study spanned several tree genera and, consequently, had limited
data on eucalypt soil effects (neffec t size = 141). Here, we apply the
latest rigorous meta- analytical techniques (Nakagawa et al., 2017;
Noble et al., 2017) to an extensive global dataset focused solely on
eucalypts sensu lato (neffect size = 2,806) an d thei r soi l effe c t s. Th e large
extent of our data enables us to produce a more powerful, nuanced
and globally meaningful synthesis of the effects of eucalypts on indi-
vidual soil proper ties, such as soil carbon and phosphorus. Our meta-
analysis also offers two new extensions to the current understanding
of eucalypt effects; that is, the importance of plantation age and soil
depth as drivers of variation in these effects, and comparisons not
only with other silvicultural systems, but also with agropastoral sys-
tems and native vegetation. These advances in knowledge will pro-
vide important information to guide policy- makers and practitioners
in pursuit of improved environmental outcomes in areas where euca-
lypt plantings continue outside their native range.
We searched two electronic databases (Web of Science and Scopus)
on 25 August 2021 for published studies, using the following search
terms: ((eucalypt*) NEAR/5 (soil* OR microbe* OR plant* OR animal*
OR invertebrate*)). Search terms were sought in the title, abstract
and keywords, and all studies up to the search date were included.
The “NEAR/5” term and its Scopus equivalent (“PRE/5”) were used
to limit the number of studies to a feasible amount, while main-
taining reproducibility. Note that additional search terms and un-
restricted distance between terms could have captured additional
relevant studies; therefore, our search strategy could be considered
systematic but not entirely comprehensive. Although a systematic
and reproducible approach is essential for minimizing bias, true com-
prehensiveness is not necessary in meta- analysis and can be very
difficult to achieve in large fields (Nakagawa et al., 2017 ).
Our search yielded 5,731 results from Web of Science and 3,948
results from Scopus. One study conducted by the authors that was
not identified in this search was also added. A total of 3,933 du-
plicates between search- engine results were removed, resulting
in 5,747 total records for screening. This list was refined manually,
based on titles and abstracts, to remove items that focused on un-
related fields (e.g., genetics, modelling, remote sensing, plant phys-
iology), resulting in 1,449 studies remaining for full- text screening.
With a large number of studies remaining, we decided to narrow
our focus solely to soil characteristics, thereby excluding the ani-
mal, plant and macrofungus studies that we initially sought in the
keyword search. We also chose to focus on eucalypt plantations
and exclude native eucalypt forests, which differ markedly in their
land- management history. After full- text screening (including re-
moval of animal, plant and macrofungus studies), 227 studies re-
mained for data extraction (a list of the data sources is provided in
the Supporting Information Appendix S1). A PRISMA checklist and
diagram are included in the Supporting Information and detail the
screening process and adherence to best practice reporting guide-
lines (Table S1; Figure S1; Page et al., 2021). It is worth noting that
additional search terms (e.g., “edaphic”, “bacteria”) have the potential
to capture additional relevant studies; therefore, our review is un-
likely to be wholly comprehensive.
Assessing the effects of eucalypts requires a comparison with a
eucalypt- free community. One option is to adopt a repeated meas-
ures approach to compare soil properties before and after eucalypt
afforestation (e.g., Epron et al., 2009). An alternative is to compare
eucalypts with nearby eucalypt- free communities that experience
similar climatic conditions and are likely to share a similar soil-
forming history (e.g., Guedes et al., 2016). Our initial searches re-
vealed that the latter approach, requiring no long- term commitment,
was far more common and would therefore yield a larger dataset
with which to test our predictions. Accordingly, we decided to use
non- eucalypt communities (e.g., pine plantations, native vegetation,
pastures, croplands) as the “control” in our analyses.
For each reported soil property (Table 1), we extracted the mean
value, sample size and a measure of the variance (SD, SE or confidence
in ter val), wh ere pr ovi ded, fo r the eucalypt co mmun ity and co ntr ol co m-
munity. Given that some studies took replicate measurements within
sites and among sites, we chose to consider within- site replicates as
pseudoreplicates and extracted variance and sample size only when
they were reported at the site level. Owing to differences in what was
considered a “site”, we followed the definition of the authors of each
study. Generally, sites were defined as separate, non- contiguous com-
munities, but in some cases, multiple sites were contained within the
same community, although separated by several kilometres. If stud-
ies reported only within- site replicates, we recorded no variance and
a conservative sample size of one. When a study reported repeated
measures over time, we extracted only the most recent value to main-
tain the independence of effect sizes. Values not reported in the text
were extracted from figures using ImageJ software, v.1.53e (Abràmoff
et al., 2004). Where no measure of variance (at the site level) was re-
ported in the text, we used imputation to estimate these values (see
Section 2.3 below) after compilation of the final dataset.
We also retrieved the following additional information for each
soil property record: control context (i.e., silviculture, native vegeta-
tion or agropastoral), age of eucalypt plantation, soil depth, and the
identities of the eucalypt and control species where known. Owing
to the wide variation in soil profiles globally, we extracted soil depth
as a binary variable (upper or lower) according to the discretion of
the original authors. In the vast majority of studies, the upper soil
profile corresponded to the top 10 cm, while the lower soil profile
generally corresponded to anywhere from 10 to 50 cm. When soil
properties from more than two different soil depths were reported,
we extracted only the uppermost and lowermost values to minimize
the risk of non- independence.
M ALLEN- COO PER Et AL .
The wide range of reported soil properties were grouped into
14 main properties: C, N, P, K, anions, cations, conductivity, density,
enzymes, erosion, microbes, temperature and moisture (Table 1).
We used the natural log response ratio (lnRR) as the effect size in
our meta- analysis (Hedges et al., 1999), calculated as lnRR = log(x̄ eu
calypt/x̄control) where x̄ is the mean value of a soil property beneath
eucalypts or the control community. The sampling error variance in
lnRR, the inverse of which was used to weight effect sizes in our meta-
analytical models, was calculated according to the formulas presented
by Hedges et al. (1999). The lnRR was chosen because it is simple to
interpret and largely unaffected by non- independent samples (Noble
et al., 2017 ). The lnRR is positive when the magnitude of a soil prop-
erty is greater beneath eucalypts relative to the control community,
and vice versa. Values of zero in the means of specific soil properties
occurred in 0.1% of cases and were managed using single imputa-
tion (Lajeunesse, 2013; Nakagawa, 2015). To do this, we set zeroes in
means recorded in eucalypt plantations to the value that would pro-
duce the lowest lnRR value in the dataset, and vice versa for means
recorded in control communities. The lnRRs of some original prop-
erties were coined (multiplied by minus one; Table 1), meaning that
all properties within a group were expected to respond in the same
direction. For example, within the soil density group, soil porosity was
mu l tipli ed by minus on e to ma tch bulk de nsi t y, su ch that greate r values
corresponded to reduced porosity. We used imputation to derive val-
ues of standard deviation that were not reported for 43% of eucalypt
means and 48% of control means. Standard deviations are used in the
weighting of effect sizes, which, if associated with high variance, are
downweighted in the meta- analytical models. The type of imputation
we chose uses th e relationship between the log10- transformed means
and log10- transformed standard deviations to back- calculate suitable
values (Lajeunesse, 2013). In our case, the relationship used in impu-
tation was strong (R2 = .81), implying that imputation is likely to be
reliable for this dataset despite large amounts of missing data (but see
section 2.4. Publication bias and sensitivity analyses).
A large proportion of the included studies (55%) compared the
same eucalypt plantations with multiple plant communities. To man-
age this non- independence in treatment means, we divided the anal-
yses into three separate comparisons: eucalypt versus silvicultural
plantations (e.g., Pinus), eucalypt versus native vegetation and euca-
lypt versus agropastoral systems (e.g., cropland, managed pasture).
By separating many of the shared eucalypt (treatment) means among
the three analyses, we were able to use 929 (33%) additional effect
sizes than if we had included all comparisons in the same model.
To manage shared control means (e.g., native forests compared
with two sets of eucalypt plantations), we constructed variance–
covariance matrices for each comparison, in which the diagonals
represented the sampling error variance in lnRR and the off- diagonal
cells represented the covariance resulting from shared controls
(Noble et al., 2017). In other words, off- diagonal cells were set to
zero unless there was a shared control, in which case the covariances
were calculated according to the method of Lajeunesse (2011).
We used study identity and eucalypt species as random factors
in the intercept models and meta- regressions. Within this model
structure, study identity accounted for the variance explained by
Original soil properties within each broad group (with number of effect sizes in parentheses; all abbreviations are standard
Anions Carbonate (4); chloride (4); S (14); sulphate (4)
Cations Al (49); B (8); bases (9); base saturation (26); Ca (117); cation exchange capacity (83); cations (9); Co (1); Cu (11); Fe (19); Hg
(1); Mg (114); Mn (13); Na (42); Na absorption ratio (4); Ni (1); Pb (1); V (1); Zn (12)
Conductivity Electrical conductivity (64); salinity (4)
Soil C C fractions (20); C stock (53); C (113); organic C (227); organic matter (67)
Soil density Bulk density (211); compaction (4); hardness (14); porositya (61)
Soil enzymes Aminopeptidase (2); cellobiosidase (7); chitinase (2); dehydrogenase (3); glucosidase (14); N- acetylglutamate (1); peroxidase
(4); phenoloxidase (4); phosphatase (14)
Soil erosion Sediment production (2); erosion (14); stabilitya (15)
Soil K K (152)
Soil microbes Bacterial abundance (1); microbial activity (6); microbial biomass (110); microbial composition (13); microbial respiration
(24); microbial richness (15); soil respiration (7)
Soil N Ammonium (16); N (242); N fixation (4); N lossa (2); N mineralization (12); N stock (34); nitrate (17); nitrification (10);
organic N (6)
Soil P P (184); P leachinga (24)
Soil pH pH (269)
Soil temperature Temperature (27)
Soil moisture Evaporationa (5); infiltration (9); moisture (112); repellencya (4); runoffa (9); water- holding capacit y (25)
aAttribute was coined (multiplied by minus one).
MALLEN- COOPER Et AL .
study- specific phenomena, such as methodology or multiple mea-
surements of soil properties. Three moderators were used in meta-
regression: soil property (categorical, 14 levels), eucalypt plantation
age (continuous, log10- transformed before analysis) and soil depth
(categorical, two levels). Age and soil depth were structured as an
interaction with soil property in separate meta- regression models
because each was associated with a different and reduced subset
of effect sizes. If soil property, age and soil depth were included in
the same model (i.e., as a three- way interaction), the number of ef-
fect sizes would be reduced by 69% (n = 1,937 fewer effect sizes)
and the interpretation of results would become highly complex. All
estimated coefficients are presented as true values rather than rel-
ative to a reference group. The native vegetation comparison con-
tained the highest number of effect sizes (ntotal = 1,239, nage = 718
and nsoil = 690; Supporting Information Table S2), followed by the
silvicultural comparison (ntotal = 913, nage = 451 and nsoil = 365;
Supporting Information Table S2) and agropastoral comparison
(ntotal = 654, nage = 429 and nsoil = 445; Supporting Information Table
S2). Thus, the total number of effect sizes used in our analyses was
neffect size = 2,806. All models were conduc ted using the rma.mv fu nc-
tion in the metafor R pa c kage v.3.0 - 2 (Viech tbaue r, 2010), an d we cal-
culated a marginal R2 for each meta- regression that represented the
variance explained by moderators (Nakagawa & Schielzeth, 2013).
The overall meta- analytical mean, derived from the intercept model,
is virtually meaningless in ecological meta- analyses, where differ-
ent groups, such as soil properties, often have opposing effects and
where heterogeneity (I2; Higgins & Thompson, 2002) is typically high
(O'Dea et al., 2021). We therefore use the intercept model only to
evaluate the heterogeneity among effect sizes, which, when high,
can reduce statistical power and therefore make estimates more
conservative (Valentine et al., 2010). We considered an effect signif-
icant when the 95% confidence interval did not cross zero.
Public ation bia s was assesse d using a modifie d version of Egge r re gres-
sion (Sterne & Egger, 2006), trim- and- fill tests (Duval & Tweedie, 2000)
and a visual assessment of the funnel plot of precision (inverse standard
error) of lnRRs against the meta- analytical residuals (sensu Naka gaw a &
Santos, 2012), which were extracted using the MCMCglmm R package
v.2.32 (Hadfield, 2010; Hadfield & Nakagawa, 2010). We performed
three types of sensitivity analysis to test the robustness of estimated
effects. First, we removed the five studies contributing the highest
number of effect sizes in each data subset. Second, we removed ex-
treme values of lnRR and sampling error variance. Third, we ran un-
weighted meta- regressions, which are unbiased but less precise than
weighted models (Morrissey, 2016; Nakagawa & Lagisz, 2016), using
the lmer function in the lme4 package v.1.1- 27.1 (Bates et al., 2015).
Effects were considered robust if they remained quantitatively similar
to the original analysis, allowing for qualitative differences in signifi-
cance owing to loss of statistical power.
The 227 studies in our database were located across five continents,
in both tropical and sub- tropical ecosystems (Figure 1), with more
than half of the studies conducted in Brazil (74 studies), China (38
studies) and India (16 studies). A total of 13 studies, comprising 105
Global map (latitude and longitude) showing locations of included studies, coloured by continent
M ALLEN- COO PER Et AL .
effe ct size s, were con duc ted on eu cal ypt pla nt ations within their na-
tive range. The earliest published study was in 1986, but most were
published after 2000, with a consistent increase in the number of
publications to the present day (Supporting Information Figure S2a).
The median age of the eucalypt plantations in our analyses was
11 years. The most represented species were Eucalyptus grandis,
Eucalyptus globulus, Eucalyptus urophylla, Eucalyptus camaldulensis,
Eucalyptus saligna and Eucalyptus tereticornis, six of the “big nine”
cultivated eucalypt species (Stanturf et al., 2013), but 22% of studies
did not identify eucalypts to species level (Supporting Information
About 3– 6% of the variation in eucalypt effects was explained by
soil property (shown by R2 values in Supporting Information Table
S3), with significant effects for soil moisture, microbes, anions, cati-
ons, conductivity, C, N an d K (Figure 2; Supporting Information Table
S4). Most of these effects represented declines when eucalypts
were compared with plantations, native vegetation or agropastoral
systems, although soil C increased in one comparison. Furthermore,
eucalypt species identity explained very little of the variance in soil
effects (<6% of the variance in all but one meta- regression model;
Supporting Information Table S3), and we found high heterogene-
ity in our intercept models (I2
total >.98; Supporting Information Table
Eucalypt effects, however, varied markedly depending on the
community with which they were compared (i.e., silvicultural, native
vegetation or agropastoral). Soil moisture and soil microbes were
significantly lower (by an average of 23% and 47%, respectively)
under eucalypts relative to native vegetation, but there were no
significant differences when eucalypts were compared with man-
aged systems (Figure 2; Supporting Information Table S4). Likewise,
(a, b, c) Main effects of eucalypts on soil properties and (d, e, f) interaction slopes with eucalypt plantation age (−ve = negative,
+ve = positive) when compared with (a, d) silvicultural systems, (b, e) native vegetation and (c, f) agropastoral systems; confidence intervals
are represented by a black bar extending from each estimated mean. Significant results are highlighted in colour. Sample sizes are indicated
by numbers along the vertical axis (results based on <10 effect sizes are excluded from the figure owing to unreliability). Raw effect sizes
(within the −1.5 to +1.5 range) are shown as background points
MALLEN- COOPER Et AL .
eucalypts significantly reduced soil N relative to both silvicultural
systems and native vegetation. Soils beneath eucalypts had signifi-
cantly lower amounts of K, but only when compared with managed
systems (i.e., not native vegetation). The only positive effect of eu-
calypts (soil C) was significant only when eucalypts were compared
with agropastoral systems.
Several effects varied with the age of eucalypt plantations. For ex-
ample, the reductions in soil cations and microbes beneath euca-
lypt s (relative to native vegetation) intensifi ed over tim e (Figures 2
and 3; Supporting Information Table S4). In contrast, positive
effects of eucalypts on soil C, relative to agropastoral systems,
increased with plantation age (Figure 2; Supporting Information
Finally, our meta- regression models also indicated that some ef-
fects were significant only at a particular soil depth (upper or lower;
Figure 4). For example, when compared with agropastoral systems,
eucalypts enhanced soil cations and carbon, but only in the upper
soil profile. In contrast, negative effects of eucalypts on soil P and K,
relative to agropastoral systems, were significant only in the subsoil
We found some evidence of publication bias in the silvicultural
(Egger: z = −3.13, p = .002; trim- and- fill: four missing studies de-
tected) and agropastoral (Egger: z = −3.16, p = . 0 0 2 ; t r i m - a n d - f i l l :
two missing studies detected) data subsets, though these re-
sults were driven by a few extreme outlying values (Supporting
Information Figure S4). Meta- regression results were quan-
titatively similar to those obtained from sensitivity analyses
(Supporting Information Figures S5– S9), although some changed
in significance as power was lost.
Our global meta- analysis revealed that eucalypts have mostly nega-
tive effects on soil properties and associated processes in compari-
son to other plantation forests, native vegetation and agropastoral
systems. However, the effects of eucalypts were highly dependent
on the community with which they were compared. Reductions in
soil microbes, moisture and cations in eucalypt soils were evident
only when compared with native vegetation, but not managed
Bubble plot showing the model- predicted relationship between eucalypt plantation age (natural logarithm of years) and the
natural logarithmic response ratio (lnRR) of (a) soil cation effects and (b) soil microbe effects, in comparison to native vegetation (colours
indicate negative or positive effect sizes; point size is proportional to relative weight in the model; and the line represents the modelled
relationship ±95% confidence interval in grey)
M ALLEN- COO PER Et AL .
vegetation (silvicultural and agropastoral systems). Conversely,
the effect of eucalypts in reducing soil K was significant only rela-
tive to managed vegetation. Soil C was enhanced only relative to
agropastoral systems, and this effect was stronger in older eucalypt
plantations. Despite these effects, eucalypt species identity did not
explain a substantial amount of variance in soil effects, indicating
that these effects are, at most, weakly controlled by differences in
species traits, such as root chemistry and leaf morphology (Senior
et al., 2016; Zaiton et al., 2020). Although soil P was generally but
not significantly reduced by eucalypts, these effects strengthened
significantly with increasing eucalypt plantation age, as did the nega-
tive effects of eucalypts on cations and microbes. There was also
some evidence that the effects of eucalypts on soil C were strong-
est in the topsoil. Overall, our findings provide robust empirical evi-
dence that eucalypts in plantation settings exert a strong influence
on a range of soil properties and functions globally.
We found strong evidence that eucalypts modify soil hydrology
relative to native vegetation. It is well known that eucalypts exploit
both ground water and moisture from the upper vadose zone (Engel
et al., 2005; Madalcho et al., 2019), even under low water poten-
tials (Thorburn & Walker, 1994). Despite the tendency of eucalypts
to scavenge moisture from the uppermost soil layers and to inter-
cept rainfall (Livesley et al., 2014), they also have the potential to
enhance soil moisture by conducting water from deeper to surface
layers (hydraulic lift; Brooksbank et al., 20 11), increasing infiltration
(Eldridge & Freudenberger, 2005), conducting rainfall via stem flow,
and reducing evaporation through shading and wind buffering (Bosi
et al., 2020). The net effects of eucalypts on soil moisture there-
fore depend on the balance of water- enhancing and water- reducing
processes. Eucalypt stands had significantly lower soil moisture than
native vegetation, suggesting that water- reducing processes (pri-
marily water uptake) are stronger and/or water- enhancing processes
(e.g., macropore creation, canopy shading) are weaker in eucalypt
systems. The finding that eucalypt plantations did not differ signifi-
cantly in their soil moisture compared with non- eucalypt silvicul-
tural plantations is likely to reflect similarities in water use (Benyon
& Doody, 2015; White et al., 2021) or land management, because
harvesting machinery can lead to soil compaction and reduced infil-
tration (Greacen & Sands, 1980).
The soil nutrient results generally align with the broader narra-
tive of nutrient depletion beneath eucalypts (Jagger & Pender, 2003;
Madalcho et al., 2019; Zaiton et al., 2020). The great ability of eu-
calypts to retranslocate and conserve N, P and K during leaf senes-
cence is likely to be driving the observed reductions in soil N and K
(Killingbeck, 1996; Saur et al., 2000). Equally plausible is that the par-
ticular chemical composition of eucalypt litter, characterized by high
concentrations of polyphenols and lignified compounds (del Moral
& Muller, 1970), acts to inhibit nitrification and thus reduce soil ni-
trate (Castro- Díez et al., 2012). The latter explanation is supported by
our finding that eucalypts reduce microbial activity relative to native
Effects of eucalypts on soil properties, partitioned among soil depths, relative to (a) silvicultural systems, (b) native vegetation
and (c) agropastoral systems. Results significantly different from zero are indicated by a black plus sign. No properties were significantly
different among soil depths. Background points represent raw effect sizes, and numbers along the vertical axis indicate sample sizes
MALLEN- COOPER Et AL .
vegetation, although this finding could also be explained by low-
quality eucalypt litter (Bini et al., 2013), differences in canopy struc-
ture resulting in lower soil temperature and moisture (Kara et al., 2008;
Wang et al., 2020), and other effects on soil moisture, such as high
water uptake (White et al., 2021). We also found that eucalypt plan-
tations had higher soil C relative to agropastoral systems, and this
effect was strongest in more mature plantations. Soil C is known to
decline initially with eucalypt plantings (Cook et al., 2016) and then to
increase gradually with age, although residual litter from a previous
land use can persist for several years and obscure this pattern (Epron
et al., 2009; Paul et al., 20 02). Eucalypts typically produce more litter
than pastures (Paul et al., 2002; Sangha et al., 2006), particularly in
more mature stands, resulting in higher C returned to the soil.
We found that eucalypts did not have a consistent effect on soil
pH, contrary to our hypothesis and much empirical evidence (e.g.,
Leite et al., 2010; Rhoades & Binkley, 1996; Soumare et al., 2016).
Soil acidity can arise from several sources, yet the evidence from the
study by Jobbágy and Jackson (2003) suggests a dominant effect of
cation redistribution, whereby base cations are largely relocated from
the main rooting zone to the surface, via absorption and subsequent
litter fall, leading to higher acidit y with increasing depth. However, we
found no strong evidence that pH was reduced by eucalypts on av-
erage nor that cation effects were stratified by soil depth. There are
several possible explan ations for our result s: (1) sou rces of acid ity ten d
to be similar across agrosilvicultural systems and native vegetation; (2)
variation in soil pH is controlled predominantly by other factors, such
as climate (Hong et al., 2019); and/or (3) ecophysiological processes
governing plant effects on pH are highly variable among ecosystems.
There are several important caveats to consider when inter-
preting the results of our study. First, as with most meta- analyses,
our findings represent average effects that are underlaid by a large
amount of variation (e.g., Figure 2). Consequently, our results do not
preclude neutral or opposite effects in certain ecosystem condi-
tions. Second, there was some evidence of publication bias in par-
ticular data subsets, which could inflate the significance of certain
results. It is also worth noting that a surprisingly large number of
studie s did not measure replicate communities (e.g. , eucal ypt plant a-
tion, native Cerrado savanna, managed pasture), although mean val-
ues were produced from pseudoreplicates located within the same
community. However, we still found clear and unambiguous effects
of eu calypts on soil prop ert ies, despite th e fac t that our analyses en-
compassed a large number of studies, often poorly replicated, from
markedly different environmental contexts, years, seasonal condi-
tions and eucalypt species. Determining the extent to which climate,
soil type and other environmental factors control variation in the
effects of eucalypts on soil is a worthy topic of further investigation.
Our findings have immediate practical implications, allowing
managers to predict the likely outcomes arising from land- use
transitions. For example, when converting natural vegetation
to a eucalypt plantation, there is likely to be a reduction in soil
N, moisture, cations and microbial functioning, with the last two
strengthening as plantation age increases. In the case of a eucalypt
plantation replacing a different plantation (e.g., Pinus), few changes
are likely to occur beyond reductions in soil N and K. When pas-
tures or croplands are converted to eucalypt plantations, there is
likely to be an increase in soil C, particularly as plantations become
more mature, which aligns with the general model of afforestation
(Paul et al., 2002). The implications of increasing soil C might have
benefits for carbon abatement programmes. However, it is impor-
tant to note that our analysis considere d only soil effects, and posi-
tive effects of eucalypt afforestation on soil C are likely to trade
off against negative effects on biodiversity and other ecosystem
properties (Phifer et al., 2017; Saccol et al., 2017).
Our study provides new evidence of the effects of eucalypts on soils
at the global scale, relative to both natural and managed ecosystems.
Overall, our synthesis suggest s a multitude of negative outcomes for
soils and microbial functioning when eucalypt soils are compared
with soils from native (non- eucalypt) plant communities, implying
that eucalypt plantations would be a poor substitute for native eco-
systems and their ecological processes. Nevertheless, our synthesis
indicates that eucalypts might have a role in increasing soil carbon in
managed landscapes. Several past studies have recommended that
degraded agricultural lands or wastelands be converted to eucalypt
plantations (e.g., Jagger & Pender, 2003; Liang et al., 2016), thereby
balancing socio- economic benefits (reviewed by Jaleta et al., 2016
and Madalcho et al., 2019) and environmental outcomes, which in
this context would be largely positive. Our findings generally sup-
port this recommendation. Another recent study indicates that
there might also be a role for eucalypts as an intermediate phase in
the reforestation of agricultural land, providing rapid canopy cover,
enhanced soil C and a source of revenue while intercropped native
plants can regenerate (Brancalion et al., 2020). Altogether, our re-
sults provide a basis for reconciling such trade- offs across different
ecosystems world- wide, allowing policy- makers and land managers
to assess the net environmental and economic benefits of eucalypts
and avoid potentially detrimental effects on ecological functioning.
We are grateful for the indomitable statistical advice of Shinichi
Nakagawa and the feedback of the handling editor and two anony-
mous reviewers. Open access publishing facilitated by University of
New South Wales, as part of the Wiley - University of New South
Wales agreement via the Council of Australian University Librarians.
D.J.E. conceived the study. M.M.- C. led the design of the extraction
protocol with support from D.J.E., J.A. and Z.A.X. All authors con-
tributed equally to data extraction. Data curation was conducted by
M ALLEN- COO PER Et AL .
G.M.C., M.M.- C., J.A. and Z.A.X. Data analysis was led by M.M.- C.
with support from J.A. and Z.A.X. Data presentation was led by
M.M.- C. and Z.A.X., with support from B.W. The first draft of the
manuscript was led by M.M.- C. with support from D.J.E., J.A. and
B.W., and all authors contributed to edits therein.
All data and code are available on the Open Science Framework at
Max Mallen- Cooper https://orcid.org/0000-0002-8799-8728
Joe Atkinson https://orcid.org/0000-0001-9232-4421
Zoe A. Xirocostas https://orcid.org/0000-0001-7103-5153
Baptiste Wijas https://orcid.org/0000-0001-7895-083X
Frederick A. Dadzie https://orcid.org/0000-0001-6130-9907
David J. Eldridge https://orcid.org/0000-0002-2191-486X
Abràmoff, M. D., Magalhães, P. J., & Ram, S. J. (2004). Image processing
with ImageJ. Biophotonics International, 11, 36– 42.
Bates, D., Machler, M., Bolker, B., & Walker, S. (2015). Fitting linear
mixed- effects models using lme4. Journal of Statistica l Software, 67,
1 – 4 8 . https://doi.org/10.18637/ jss.v067.i01
Benyon, R. G., & Doody, T. M. (2015). Comparison of interception, forest
floor evaporation and transpiration in Pinus radiata and Eucalyptus
globulus plantations. Hydrological Processes, 29, 1173– 1187. ht t p s : //
Bernhard- Reversat, F., & Schwartz, D. (1997). Change in lignin con-
tent during litter decomposition in tropical forest soils (Congo):
Comparison of exotic plantations and native stands. Comptes
Rendus de l'Académie des Sciences- Series IIA- Earth and Planetary
Science, 325, 427– 432.
Bini, D., dos Santos, C. A., Bouillet, J.- P., de Morais Goncalves, J. L., &
Cardoso, E. J. B. N. (2013). Eucalyptus grandis and Acacia mangium
in monoculture and intercropped plantations: Evolution of soil and
litter microbial and chemical attributes during early stages of plant
development. Applied Soil Ecology, 63, 57– 66.
Blume, E., Bischoff, M., Reichert, J. M., Moorman, T., Konopka, A., &
Turco, R. F. (2002). Surface and subsurface microbial biomass,
community structure and metabolic activity as a function of soil
depth and season. Applied Soil Ecology, 20, 171– 181. ht tps: //doi.
o r g / 1 0 . 1 0 1 6 / S 0 9 2 9 - 1 3 9 3 ( 0 2 ) 0 0 0 2 5 - 2
Bosi, C., Pezzopane, J. R . M., & Sentelhas, P. C. (2020). Soil water avail-
ability in a full sun pasture and in a silvopastoral system with euca-
lyptus. Agroforestry Systems, 94, 429– 440. https://doi.org/10.10 07/
s 1 0 4 5 7 - 0 1 9 - 0 0 4 0 2 - 7
Bradshaw, C. J. A., Hoskins, A. J., Haubrock, P. J., Cuthbert, R. N., Diagne,
C., Leroy, B., Andrews, L., Page, B., Cassey, P., Sheppard, A. W., &
Courchamp, F. (2021). Detailed assessment of the reported eco-
nomic costs of invasive species in Australia. NeoBiota, 67, 511– 550.
Brancalion, P. H., Amazonas, N. T., Chazdon, R. L., van Melis, J.,
Rodrigues, R. R., Silva, C. C., Sorrini, T. B., & Holl, K. D. (2020).
Exotic eucalypts: From demonized trees to allies of tropical for-
est restoration? Journal of Applied Ecology, 57, 55– 66. ht t p s : //d o i .
org /10.1111/1365- 266 4.13513
Brooksbank, K., Veneklaas, E. J., White, D. A., & Carter, J. L. (2011).
The fate of hydraulically redistributed water in a semi- arid zone
eucalyptus species. Tree Physiology, 31, 649– 658. h t t p s : //d o i .
Burch, G. J., Moore, I. D., & Burns, J. (1989). Soil hydrophobic effects
on infiltration and catchment runoff. Hydrological Processes, 3, 211–
222. https://doi.org/10.1002/hyp.33600 30302
Castro- Díez, P., Alonso, Á., Saldaña- López, A., & Granda, E. (2021).
Effects of widespread non- native trees on regulating ecosystem
services. Science of the Total Environment, 778, 14 6141. ht t p s ://d o i .
Castro- Díez, P., Fierro- Brunnenmeister, N., González- Muñoz, N., &
Gallardo, A. (2012). Effects of exotic and native tree leaf litter on
soil properties of two contrasting sites in the Iberian Peninsula.
Plant and Soil, 350, 179– 191. https://doi.org/10.1007/s1110
4 - 0 1 1 - 0 8 9 3 - 9
Chen, F., Zheng, H., Zhang, K., Ouyang, Z., Wu, Y., Shi, Q., & Li, H. (2013).
Non- linear impacts of eucalyptus plantation stand age on soil mi-
crobial metabolic diversity. Journal of Soils and Sediment s, 13, 887–
894. h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / s 1 1 3 6 8 - 0 1 3 - 0 6 6 9 - 3
Christina, M., Nouvellon, Y., Laclau, J.- P., Stape, J. L., Bouillet, J.- P.,
Lambais, G. R., & le Maire, G. (2017). Importance of deep water
uptake in tropical eucalypt forest. Functional Ecology, 31, 509– 519.
https://doi.or g/10.1111/1365- 2435.12727
Chu, C., Mortimer, P. E., Wang, H., Wang, Y., Liu, X., & Yu, S. (2014).
Allelopathic effects of eucalyptus on native and introduced tree
species. Forest Ecology and Management, 323, 79– 84. ht tps://doi.
Cook, R. L., Binkley, D., & Stape, J. L. (2016). Eucalyptus plantation effects
on soil carbon after 20 years and three rotations in Brazil. Forest
Ecology and Management, 359, 92– 98. https://doi.org /10.1016/j.
Curti n, D., Ca mpb ell , C . A ., & Jali l, A. (1998) . Ef fec ts of acid ity on min eral-
ization: PH- dependence of organic matter mineralization in weakly
acidic soils. Soil Biology and Biochemistry, 30, 57– 64.
del Moral, R., & Muller, C. H. (1970). The allelopathic effects of Eucalyptus
camaldulensis. The American Midland Naturalist, 83, 254– 282.
Duval, S., & Tweedie, R. (2000). Trim and fill: A simple funnel-
plot– based method of testing and adjusting for publication
bias in meta- analysis. Biometrics, 56, 455– 463. ht t p s : //doi.
Easteal, S. (1981). The history of introductions of Bufo marinus
(Amphibia: Anura); a natural experiment in evolution. Biological
Journal of the Linnean Society, 16, 93– 113. https://doi.org/10.1111/
j . 1 0 9 5 - 8 3 1 2 . 1 9 8 1 . t b 0 1 6 4 5 . x
Eldridge, D. J., & Freudenberger, D. (2005). Ecosystem wicks: Woodland
trees enhance water infiltration in a fragmented agricultural land-
scape in eastern Australia. Austral Eco logy, 30, 336– 347. ht t p s : //d o i .
org /10.1111/j.1442- 9993.2005.01478.x
Eldridge, S. R., Thorburn, P. J., McEwan, K. L., & Hatton, T. J. (1993).
Health and structure of eucalyptus communities on Chowilla and
Monoman Islands of the river Murray floodplain, South Australia.
https://publi catio ns.csiro.au/rpr/pub?list=BRO&pid=proci
t e : b 2 4 a b 6 2 5 - 2 7 a 7 - 4 3 a 6 - 8 6 d 8 - 3 4 b b 8 8 b 4 7 0 a 8
Engel, V., Jobbágy, E. G., Stieglitz, M., Williams, M., & Jackson, R. B.
(2005). Hydrological consequences of eucalyptus afforestation
in the argentine pampas. Water Resources Research, 41, W10409.
Epron, D., Marsden, C., Thongo M'Bou, A ., Saint- André, L., d'Annunzio,
R., & Nouvellon, Y. (2009). Soil carbon dynamics following affor-
estation of a tropical savannah with eucalyptus in Congo. Plant and
Soil, 323, 309– 322. h t t p s : // d o i . o r g / 1 0 . 1 0 0 7 / s 1 11 0 4 - 0 0 9 - 9 9 3 9 - 7
Fang, C., Smith, P., Smith, J. U., & Moncrieff, J. B. (2005). Incorporating
microorganisms as decomposers into models to simulate soil or-
ganic matter decomposition. Geoderma, 129, 139– 146. h t t p s : //doi.
MALLEN- COOPER Et AL .
FAO (2020). Global Forest Resources Assessment: Main report. FAO.
Forrester, D. I., Collopy, J. J., & Morris, J. D. (2010). Transpiration along
an age series of Eucalyptus globulus plantations in southeastern
Australia. Forest Ecology and Management, 259, 1754– 1760. h t t p s : //
Gonçalves, J. L. M., Stape, J. L., Laclau, J. P., Smethurst, P., & Gava, J. L.
(2004). Silvicultural effects on the productivity and wood quality of
eucalypt plantations. Forest Ecology and Management, 193, 45– 61.
Greacen, E. L., & Sands, R. (1980). Compaction of forest soils: A review.
Australian Journal of Soil Research, 18 , 163– 189.
Guedes, B. S., Olsson, B. A., & Karltun, E. (2016). Effects of 34- year- old Pinus
taeda and Eucalyptus grandis plantations on soil carbon and nutrient
status in former miombo forest soils. Global Ecology and Conservation,
8, 190– 202. https://doi.org/10.1016/j.gecco.2016.09.005
Hadfield, J. D. (2010). MCMC methods for multi- response generalized lin-
ear mixed models: The MCMCglmm R package. Journal of Statistical
Software , 33, 1– 22. https://doi.org/10.18637/ jss.v033.i02
Hadfield, J. D., & Nakagawa, S. (2010). General quantitative genetic
methods for comparative biology: Phylogenies, taxonomies
and multi- trait models for continuous and categorical charac-
ters. Journal of Evolutionary Biology, 23, 494– 508. h t t p s : //d o i .
org /10.1111/j.1420 - 9101.2009.01915. x
Hedges, L. V., Gurevitch, J., & Curtis, P. S. (1999). The meta- analysis of
response ratios in experimental ecolog y. Ecology, 80, 1150– 1156.
Higgins, J. P. T., & Thompson, S. G. (2002). Quantifying heterogeneity in
a meta- analysis. Statistics in Medicine, 21, 1539– 1558. ht t p s : //d o i .
Hobbs, R. J. (2000). Invasive species in a changing world. Island Press.
Hong, S., Gan, P., & Chen, A. (2019). Environmental controls on soil
pH in planted forest and its response to nitrogen deposition.
Environmental Research, 172, 159– 165.
Iglesias- Trabado, G., & Wilstermann, D. (2008). Eucalyptus universa-
lis. Global cultivated eucalypt forests map 20 08. GIT Forestr y
Consulting's EUCALYPTOLOGICS: Information Resources on
Eucalyptus Cultivation Worldwide http://www.Git- Fores try.com
Iovieno, P., Alfani, A., & Bååth, E. (2010). Soil microbial community
structure and biomass as affected by Pinus pinea plantation in two
Mediterranean areas. Applied Soil Ecology, 45, 56– 63.
Jagger, P., & Pender, J. (2003). The role of trees for sustainable man-
agement of less- favored lands: The case of eucalyptus in Ethiopia.
Forest Policy and Economics, 5, 83– 95. https://doi.org/10.1016/
s 1 3 8 9 - 9 3 4 1 ( 0 1 ) 0 0 0 7 8 - 8
Jaleta, D., Mbilinyi, B., Mahoo, H., & Lemenih, M. (2016). Eucalyptus
expansion as relieving and provocative tree in Ethiopia. Jo urnal of
Agricu lture and Ecology Re search International, 6, 1– 12. h t tps: //doi.
Jobbágy, E. G., & Jackson, R. B. (2003). Patterns and mechanisms
of soil acidification in the conversion of grasslands to forests.
Biogeochemistry, 64, 205– 229. ht tps://doi.org/10.1023/A:10249
Kara, Ö., Bolat, İ., Çakıroğlu, K., & Öztürk, M. (2008). Plant canopy ef-
fects on litter accumulation and soil microbial biomass in two tem-
perate forests. Biology and Fertility of Soils, 45, 193– 198.
Killin gbeck, K. T. (1996). Nutrie nts in sene sced leave s: Keys to the search
for potential resorption and resorption proficiency. Ecology, 77,
1716– 1727. https://doi.org/10.2307/2265777
Lajeunesse, M. J. (2011). On the meta- analysis of response ratios for
studies with correlated and multi- group designs. Ecology, 92, 2049–
2055. https://doi.org/10.1890/11- 0423.1
Lajeunesse, M. J. (2013). 13. Recovering missing or par tial data from
studies: A survey of conversions and imputations for meta- analysis
(Vol. 13, pp. 195– 206). Princeton University Press. ht t p s ://d o i .
o r g / 1 0 . 1 5 1 5 / 9 7 8 1 4 0 0 8 4 6 1 8 4 - 0 1 5
Leite, F. P., Silva, I. R., Novais, R. F., De Barros, N. F., & Neves, J. C. L.
(2010). Alterations of soil chemical properties by Eucalyptus culti-
vation in five regions in the Rio Doce Valley. Revista Brasileira de
Ciência do Solo, 34, 821– 831.
Liang, J., Reynolds, T., Wassie, A., Collins, C., & Wubalem, A. (2016).
Effects of exotic Eucalyptus spp. plantations on soil proper ties in
and around sacred natural sites in the northern Ethiopian high-
lands. AIMS Agriculture and Food, 1, 175– 193.
Lima, A. M., Silva, I. R., Neves, J. C., Novais, R. F., Barros, N. F., Mendonça,
E. S., Smyth, T. J., Moreira, M. S., & Leite, F. P. (2006). Soil organic
carbon dynamics following afforestation of degraded pastures with
eucalyptus in southeastern Brazil. Forest Ecology and Management,
235, 219– 231.
Livesley, S. J., Baudinette, B., & Glover, D. (2014). Rainfall interception
and stem flow by eucalypt street trees – The impacts of canopy
density and bark type. Urban Forestry & Urban Greening, 13, 192–
Madalcho, A. B., Lemma, B., Mena, M. M., & Badesso, B. B. (2019). Is the
expansion of eucalyptus tree a curse or an opportunity? Implications
from a dispute on the trees ecological and economic impact in
Ethiopia: A review. Journal of Ecology and The Natural Enviro nment,
11, 75– 83. https://doi.org/10.5897/JENE2 019.0765
Mishra, A., Sharma, S. D., & Khan, G. H. (2003). Improvement in physical
and chemical properties of sodic soil by 3, 6 and 9 years old planta-
tion of Eucalyptus tereticornis: Biorejuvenation of sodic soil. Forest
Ecology and Management, 18 4, 115– 124. https://doi.org/10.1016/
S 0 3 7 8 - 1 1 2 7 ( 0 3 ) 0 0 2 1 3 - 5
Morrissey, M. B. (2016). Meta- analysis of magnitudes, differences and
variation in evolutionary parameters. Journal of Evolution ary Biology,
29, 1882– 1904. https://doi.org/10.1111/jeb.12950
Nakagawa, S. (2015). Missing data: Mechanisms, methods, and messages.
In G. A . Fox, S. Negrete- Yankelevich, & V. J. Sosa (Eds.), Ecological
Statistics. Oxford University Press. https://doi.org/10.1093/acpro
f:oso/97801 99672 547.003.0005
Nakagawa, S., & Lagisz, M. (2016). Visualizing unbiased and biased un-
weighted meta- analyses. Journal of Evolutionary Biology, 29, 1914–
1916. https://doi.org/10.1111/je b.12945
Nakagawa, S., & Santos, E. S. A. (2012). Methodological issues and ad-
vances in biological meta- analysis. Evolutionar y Ecology, 26, 1253–
1274 . h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / s 1 0 6 8 2 - 0 1 2 - 9 5 5 5 - 5
Nakagawa, S., & Schielzeth, H. (2013). A general and simple method
for obtaining R2 from generalized linear mixed- effects mod-
els. Methods in Ecology an d Evolution, 4, 133– 142. ht t p s : //d oi .
org /10.1111/j.2041- 210 x. 2012.00261.x
Noble, D. W. A., Lagisz, M., O'dea, R . E., & Nakagawa, S. (2017).
Nonindependence and sensitivity analyses in ecological and evo-
lutionary meta- analyses. Molecular Ecology, 26, 2410– 2425. ht t p s : //
doi .org/10.1111/mec.14 031
Nakagawa, S., Noble, D. W., Senior, A. M., & Lagisz, M. (2017). Meta-
evaluation of meta- analysis: Ten appraisal questions for biologists.
BMC Biology, 15, 1– 14.
O'Dea, R. E., Lagisz, M., Jennions, M. D., Koricheva, J., Noble, D. W. A.,
Parker, T. H., Gurevitch, J., Page, M. J., Stewart, G., Moher, D., &
Nakagawa, S. (2021). Preferred reporting items for systematic re-
views and meta- analyses in ecology and evolutionary biology: A
PRISMA extension. Biological Reviews, 96, 1695– 1722. h t t p s : //d o i .
org /10.1111/b rv.1272 1
Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T.
C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan,
S. E., Chou, R., Glanville, J., Grimshaw, J. M., Hróbjartsson, A.,
Lalu, M. M., Li, T., Loder, E. W., Mayo- Wilson, E., McDonald, S., …
Moher, D. (2021). The PRISMA 2020 statement: An updated guide-
line for repor ting systematic reviews. BMJ, 372, n71. ht tps: //doi.
Paul, K. I., Polglase, P. J., Nyakuengama, J. G., & Khanna, P. K. (2002).
Change in soil carbon following afforestation. Forest Ecology
M ALLEN- COO PER Et AL .
and Management, 16 8, 241– 257. https://doi.org/10.1016/S0378
- 1 1 2 7 ( 0 1 ) 0 0 7 4 0 - X
Phifer, C. C., Knowlton, J. L., Webster, C. R ., Flaspohler, D. J., &
Licata, J. A. (2017). Bird community responses to afforested
Eucalyptus plantations in the Argentine pampas. Biodiversity and
Conservation, 26, 3073– 3101. https://doi.org/10.1007/s1053
1 - 0 1 6 - 1 1 2 6 - 6
Rhoades, C., & Binkley, D. (1996). Factors influencing decline
in soil pH in Hawaiian eucalyptus and Albizia plantations.
Forest Ecology and Management, 80, 47– 56. ht tps://doi.
o r g / 1 0 . 1 0 1 6 / 0 3 7 8 - 1 1 2 7 ( 9 5 ) 0 3 6 4 6 - 6
Ricciardi, A. (2003). Predicting the impacts of an introduced species
from its invasion history: An empirical approach applied to zebra
mussel invasions. Freshwater Biology, 48, 972– 981. ht t p s ://d o i .
Saccol, S. S. A., Bolzan, A. M. R., & dos Santos, T. G. (2017). In the shadow
of trees: Does eucalyptus afforestation reduce herpetofaunal diver-
sity in southern Brazil? South American Journal of Herpetology, 12,
4 2 – 5 6 . h t t p s : / / d o i . o r g / 1 0 . 2 9 9 4 / S A J H - D - 1 6 - 0 0 0 2 8 . 1
Sangha, K . K., Jalota, R. K., & Midmore, D. J. (2006). Litter production,
decomposition and nutrient release in cleared and uncleared pas-
ture systems of Central Queensland, Australia. Journal of Tropical
Ecology, 22, 177– 189.
Sa ur, E. , Nam bia r, E. K. S. , & Fife, D. N. (2000). Fo lia r nut rie nt retr anslo ca-
tion in Eucalyptus globulus. Tree Physiology, 20, 1105– 1112. ht t p s ://
Senior, J. K., Potts, B. M., Davies, N. W., Wooliver, R. C., Schweitzer, J. A.,
Bailey, J. K., & O'Reilly- Wapstra, J. M. (2016). Phylogeny explains
variation in the root chemistr y of eucalyptus species. Journal of
Chemical Ecology, 42, 1086– 1097. https://doi.org/10.1007/s1088
6 - 0 1 6 - 0 7 5 0 - 7
Soumare, A., Sall, S. N., Sanon, A., Cissoko, M., Hafidi, M., Ndoye, I., &
Duponnois, R. (2016). Changes in soil pH, polyphenol content and
microbial community mediated by Eucalyptus camaldulensis. Applied
Ecology and Environmental Research, 14 , 1– 19.
Stantur f, J. A., Vance, E. D., Fox, T. R., & Kirst, M. (2013). Eucalyptus
beyond its native range: Environmental issues in exotic bioen-
ergy plantations. International Journal of Forestry Research, 2013,
Sterne, J. A . C., & Egger, M. (2006). Regression methods to detect pub-
lication and other bias in meta- analysis. In H. R. Rothstein, A. J.
Sutton, & M. Borenstein (Eds.), Publication bias in meta- analysis:
Prevention, assessment and adjustments. John Wiley & Sons.
Sun, J., Yu, X., Wang, H., Jia, G., Zhao, Y., Tu, Z., Deng, W., Jia, J., &
Chen, J. (2018). Effects of forest structure on hydrological pro-
cesses in China. Journal of Hydrolog y, 561, 187– 199. h t t p s : //d o i .
Teketay, D. (2000). Facts and experiences on eucalypts in Ethiopia and
elsewhere: Ground for making wise and informed decision. Walia,
21, 25– 46.
Thorburn, P. J., & Walker, G. R. (1994). Variations in stream water uptake
by Eucalyptus camaldulensis with differing access to stream water.
Oecologia, 100, 293– 301.
Turnbull, J. W. (1999). Eucalypt plantations. New Forests, 17, 37– 52.
Turn er, J., & Lamber t, M. (2 000). Change in orga ni c carbo n in for est plan-
tation soils in eastern Australia. Forest Ecology and Management,
133, 231– 247.
Valentine, J. C., Pigott, T. D., & Rothstein, H. R. (2010). How many stud-
ies do you need?: A primer on statistical power for meta- analysis.
Journal of Educational and Behavioral Statistics, 35, 215– 247. h t t p s : //
doi.org/10.3102/10769 98609 346961
Viechtbauer, W. (2010). Conducting meta- analyses in R with the meta-
for package. Journal of Statistical Software, 36, 1– 48. ht t p s : //d o i .
Walther, G.- R., Roques, A., Hulme, P. E., Sykes, M. T., Pyšek, P.,
Kühn, I., Zobel, M., Bacher, S., Botta- Dukát, Z., Bugmann, H.,
Czúcz, B., Dauber, J., Hickler, T., Jarošík, V., Kenis, M., Klotz, S.,
Minchin, D., Moora, M., Nent wig, W., … Settele, J. (2009). Alien
species in a warmer world: Risks and opportunities. Tren ds in
Ecology & Evolution, 24, 686– 693. https://doi.org/10.1016/j.
Wang, G., Huang, W., Zhou, G., Mayes, M. A., & Zhou, J. (2020). Modeling
the processes of soil moisture in regulating microbial and carbon-
nitrogen cycling. Journal of Hydrology, 585, 124777.
White, D. A., Silberstein, R. P., Balocchi- Contreras, F., Quiroga, J. J.,
Meason, D. F., Palma, J. H. N., & Ramírez de Arellano, P. (2021).
Growth, water use, and water use efficiency of Eucalyptus globu-
lus and Pinus radiata plantations compared with natural stands of
roble- Hualo forest in the coastal mountains of Central Chile. Forest
Ecology and Management, 501, 119676. https://doi.org/10.1016/j.
Williams, J. E., & Brooker, M. I. H. (1997). Eucalyptus: An introduction. In
J. E. Williams & J. C. Z. Woinarski (Eds.), Eucalypt ecology: Individuals
to ecosystems. Cambridge University Press.
Zaiton, S., Sheriza, M. R., Ainishifaa, R., Alfred, K., & Norfaryanti, K.
(2020). Eucalyptus in Malaysia: Review on environmental impacts.
Journal of Landscape Ecology, 13, 79– 94. https://doi.or g/10.2478/
jleco l- 2020- 0011
Zhang, C., & Fu, S. (2009). Allelopathic effects of eucalyptus and the
establishment of mixed stands of eucalyptus and native species.
Forest Ecology and Management, 258, 1391– 1396. ht t p s : //doi.
Zhang, K., Zheng, H., Chen, F. L., Ouyang, Z. Y., Wang, Y., Wu, Y. F., Lan, J.,
Fu, M., & Xiang, X. W. (2015). Changes in soil quality after convert-
ing Pinus to eucalyptus plantations in southern China. Solid Earth, 6,
115– 123. h t t p s : / /d o i . o r g / 1 0 . 5 1 9 4 / s e - 6 - 1 1 5 - 2 0 1 5
Zohar, Y., Gafni, A., Morris, J., & Shalhevet, S. (2008). Eucalyptus plan-
tations in Israel: An assessment of economic and environmental
viability. New Forests, 36 , 135– 157. https://doi.org/10.1007/s1105
6 - 0 0 8 - 9 0 8 9 - 4
Zou, X., & Bashkin, M. (1998). Soil carbon accretion and earthworm re-
covery following revegetation in abandoned sugarcane fields. Soil
Biology and Biochemistry, 30, 825– 830. https://doi.org/10.1016/
S 0 0 3 8 - 0 7 1 7 ( 9 7 ) 0 0 1 5 5 - 7
Our research team (the Eucanerds) assembled during a coro-
navirus disease 2019 lockdown, owing to a shared interest in
broad- scale patterns of terrestrial ecosystem functioning and
a desire to stay connected and collaborative in a rapidly digi-
Additional supporting information may be found in the online
version of the article at the publisher’s website.
Mallen- Cooper, M., Atkinson, J.,
Xirocostas, Z. A., Wijas, B., Chiarenza, G. M., Dadzie, F. A.,
Eldridge, D. J. (2022). Global synthesis reveals strong
multifaceted effects of eucalypts on soils. Global Ecology and
Biogeography, 00, 1–12. https://doi.org /10.1111/geb.13522