Discussion
Pheidole megacephala invasion strongly influenced rates of carbon
fixation in A. drepanolobium , consistent with other studies
demonstrating that invasive species can alter fundamental ecosystem
processes (Hooper et al. 2005; Morales et al. 2017). These
long-term negative effects of P. megacephala on carbon fixation,
despite an initial positive effect of invasion on leaf carbon fixation
rates, highlight that some ecologically relevant effects of invasive
species can lag behind their initial appearance in a habitat (Crooks
2005; Simberloff 2011). The change in sign of the net effect of P.
megacephala on host tree carbon fixation across different temporal
scales illuminates the value of long-term studies examining the effects
of invasive species on ecosystem processes (sensu Strayer et al.2006).
Our results from Transition sites suggest that P. megacephalatriggers short-term benefits for newly invaded trees, which may enable
newly invaded trees to maintain similar wet season photosynthetic rates
before and immediately after invasion, despite a decline of
photosynthetic rates in their neighboring control trees in that same
time period. The loss of nectivorous mutualists in recently invaded
areas likely removes a carbohydrate sink for A. drepanolobium(Stanton & Palmer 2011), freeing resources to support leaf growth,
photosynthetic upregulation, and other metabolic processes (Wiley &
Helliker 2012; Glanz-Idan & Wolf 2020). Increased carbohydrate
availability may support costly molecular and biochemical mechanisms
that mitigate heat-related damage to photosynthetic apparatuses (Mathuret al. 2014). High maximum daily temperatures (28-31°C), which
can cause substantial interannual changes in phenology of other
deciduous tree species (e.g., Muraoka et al. 2010) were recorded
in the January-March dry season that occurred between 2017 and 2018 wet
season measurements (Caylor et al. 2020). This may explain why
Control trees showed a large interannual decline in photosynthetic
rates, while Transition trees had consistent photosynthetic rates in
2017 (before invasion) and 2018 (after invasion). These benefits of
invasion were apparently negated in the subsequent drought period,
perhaps because mixed feeders that forage on woody species during the
dry season (Illius & O’Connor 1999) began to target invaded trees.
These results suggest that maintenance of photosynthesis may only be a
sustainable strategy to support growth under low intensity herbivory
(sensu Gadd et al. 2001) in favorable abiotic conditions.
Contrasting with short-term results, we found that A.
drepanolobium had markedly lower canopy photosynthetic capacity in
long-term Invaded areas vs. Uninvaded areas. In savannas that have been
invaded for > 5 years, a 2-meter-tall tree has a canopy
photosynthetic capacity of only ca . 31% of the carbon fixation
per hour as a comparable uninvaded tree during the wet season, a primary
growing period for many African acacias (Gourlay 1995). This difference
is partially driven by leaf photosynthetic rates that are ca.13% lower during the wet season for invaded trees, but is greatly
magnified by canopy leaf areas that are ca. 65% lower for trees
in long-term Invaded areas. The P. megacephala -driven decline in
photosynthetic capacity for invaded A. drepanolobium is similar
in magnitude to declines in photosynthesis in North American hardwood
forests subjected to defoliation by non-native insects (Kurz et
al. 2008; Albani et al. 2010; Clark et al. 2010).
Results from our herbivore and ant exclusion experiment suggest that
vertebrate herbivory is the primary driver of changes in the leaf and
canopy photosynthesis and transpiration for trees in Invaded habitats.
Large herbivores, particularly elephants, suppress the canopy size of
many woody plants in Sub-Saharan Africa (Pellew 1983; Biggs & Jacobs
2002; Goheen & Palmer 2010), and King and Caylor (2010) showed that
large herbivores can suppress leaf photosynthetic rates for A.
drepanolobium trees when occupied by a native ant (Crematogaster
sjostedti ) that ineffectively repels herbivores (Martins 2010). Our
results similarly show that P. megacephala invasion triggers
large changes in leaf and canopy photosynthetic capacity when
co-occurring with large herbivores. Reduced leaf area in invaded trees
is the strongest driver of canopy photosynthesis decline, but herbivores
can also induce changes to leaf structure and function in invaded areas.
For example, A. drepanolobium occupied by less defensive native
mutualists increase leaf phenol concentrations in ca . 1-m-tall
branches in response to herbivory (Ward & Young 2002), which could be
occurring on invaded trees at our study sites as well. The tradeoff
between leaf phenolic concentrations and photosynthetic rates is well
conserved (e.g., Ishida et al. 2008; Sumbele et al. 2012)
and plastic (e.g., Keenan & Niinemets 2016 and references therein)
across plant species, and so we may expect to see similar changes in
photosynthetic rates for other plants that experience intense herbivory
after ant invasion.
The occurrence of invasive ants in long-term invaded trees directly
impacted leaf physiology, though to a lesser degree than the occurrence
of large herbivores in invaded areas. This may be driven by root damage
by nesting workers or through facultative interactions with other insect
pests in the canopy. While we did not explicitly examine these
relationships here, P. megacephala tends lycaenid caterpillars
(pers. observation T. Palmer; a widespread association in Australia,
Eastwood & Fraser 1999) and appeared to tolerate cerambycid larvae
infestations (pers. observation P. Milligan) in the canopy. Infestations
of phloem-feeding insects can affect photosynthetic rates and water use
efficiency of many plant taxa (Cockfield et al. 1987; Meyer &
Whitlow 1992; Haavik et al. 2008; Golan et al. 2015). We
did not quantify the size of P. megacephala nests around tree
roots in this field experiment, but the excavation of nest cavities
around roots has been observed in A. drepanolobium saplings
(Milligan et al. in preparation). Moreover, resource-limited
invaded trees may also produce fewer fine root hairs (a key factor for
water uptake in clay dominated soils, Lambers et al . 2008 ) or
reduce the production and activation of water channels in cell membranes
(e.g., aquaporins). Nesting by invasive ants near can affect plant water
status and soil water content (Moutinho et al. 2003), and water
status influences photosynthesis, transpiration, and solute management
of other African acacias (Szarek & Woodhouse 1978; Kebbas et al.2015). Thus, the simple process of nest excavation around plant roots
may enable invasive ants to affect host plant physiology in this and
other systems.
Leaf and canopy gas exchange differed between long-term invaded and
uninvaded trees but leaf water potential ranges did not, which
altogether may be attributable to differences in leaf water management.Acacia drepanolobium in long-term invaded habitats may close
stomata to minimize water loss, which could result in similar leaf water
potential ranges at a cost to photosynthesis (Farquhar & Sharkey 1982).
Additionally, A. drepanolobium in Uninvaded habitats may have
more carbohydrates to support the management of solute concentrations in
leaves, which can allow high gas exchanges rates and capacities without
inducing a change in leaf water potential range (Inoue et al.2017; Zhang et al. 2019). These osmotic adjustments can be
metabolically costly but would also allow large increases in carbon
fixation, and thus may create net photosynthetic benefits at the canopy
level for uninvaded trees. These proposed mechanisms occur in tissues
that can have high turnover in acacias (Jha & Mohapatra 2010) and so
may be reversible if herbivore pressure is reduced. Recent climate
change models predict that the East African region will shift to wetter
climates with less severe droughts over the next century (Shongweet al. 2011; Haile et al. 2020), and physiological
adjustments by A. drepanolobium to these climatic changes may
differ in invaded and uninvaded savannas.
Both herbivore exclusion and P. megacephala removal positively
impacted host tree physiology in invaded habitats, but we did not find a
positive effect of herbivore exclusion or a negative effect of C.
mimosae removal in uninvaded habitats. These effects would be expected,
because C. mimosae strongly repels herbivores that can reduce
tree performance. The mechanisms underlying this pattern are not clear,
but one potential explanation relates to the idea of “associational
defense” (Barbosa et al. 2009), where plants with weak defenses gain
protection by proximity to well-defended neighbors. Our C.
mimosae -removal trees in unfenced areas were typically adjacent
(ca. 5 m) to well-defended A. drepanolobium trees occupied
by mature colonies of defensive ants, which may have reduced browsing on
removal plants to comparable levels seen both on C.
mimosae -occupied plants in open areas, and host plants within herbivore
exclosures. Coverdale et al. (2018) found evidence for
associational defenses at another Laikipia conservancy between otherAcacia species that deploy inducible defense (i.e. spines) and
their understory plant neighbors, though our study suggests that the
same principle applies to ant-defended trees and their conspecific
neighbors.
Our study demonstrates how an invasive ant can interact with vertebrate
herbivores to limit carbon fixation, and in turn contribute to a carbon
“deficit” in a foundational plant. Such deficits have been shown to
initiate plant performance declines in many systems. For example, both
McDowell (2011) and Wiley and Helliker (2012) discuss the process of
carbon starvation, where plant mortality rises due to large declines in
carbon fixation or strong increases in metabolic costs. Allen et
al. (2010) primarily attribute observed contemporary global increases
in plant mortality to climatic shifts, while McDowell (2011) describes
how invasive insects can increase mortality before the tree has even
exhausted its resources. As demonstrated in this study, invasive insects
can also strongly affect carbon fixation via system-specific
interactions with vertebrate herbivores that may be difficult to
generalize for broad carbon starvation hypotheses.
Some aspects of our observed decline in carbon fixation are
context-specific, but others are clearly paralleled in other systems. In
dry conditions, herbivores are often more selective for high quality
forage both in this system (Veblen 2008) and other savannas (e.g.,
Roques et al. 2001; Kos et al. 2012; Abraham et al.2019), and herbivory thus becomes frequent and intense for plants likeA. drepanolobium with leaves containing high amounts of crude
protein and minerals (Rubanza et al. 2007). Chronic herbivory
imposes cumulative respiration costs for both undefended A.
drepanolobium (regrowing lost foliage, Gadd et al. 2001;
producing defense metabolites, Ward & Young 2002) and for plants in
many other systems (e.g., Kozlov & Zvereva 2017; Wilson et al.2018). Inducible responses to herbivory (osmotic adjustment, Freelandet al. 1985; tannin/saponin production, Sharpe et al.1986; spine production, Young & Okello 1998) can consume a large
portion of the tree’s carbohydrate budget and contribute to a decline in
photosynthetic rates (as well as metabolism, regeneration, and chemical
defense; reviewed by Wiley & Helliker 2012).
Our results also contribute to our understanding of the A.
drepanolobium -ant model mutualism and ant-plant mutualisms more
broadly. The results from our recently invaded Transition sites support
the argument that ant-plants prioritize resources for ant symbionts
despite the cost of these allocation decisions for other biological
processes. Ant mutualists may impose costs to host plants in a variety
of ways, including consuming plant-provisioned food bodies (O’Dowd 1980;
Heil et al. 1997; Stanton & Palmer 2011), disrupting pollinator
visitation (Ness 2006; Villamil et al. 2020), or causing floral
castration (Stanton et al. 1999; Gaume et al. 2005).
However, ant-plant partnerships typically yield long-term net benefits
for the plant by reducing herbivore damage (Chamberlain & Holland 2009)
or even increasing competitiveness against other plants (Fiala et
al. 1989). Our study adds to this literature, demonstrating that native
ant associates impose significant metabolic costs to host plants, while
yielding positive net effects on photosynthesis across longer time
scales owing to effective herbivore protection.
The long-term loss of photosynthetic capacity for ant-plants in invaded
habitat reduces A. drepanolobium ’s carbohydrate pool, which may
affect other ecosystem processes to which this foundational tree
contributes. For example, nitrogen is a limiting resource for plant
productivity in black cotton savannas, and A. drepanolobiumimports nitrogen into these systems through N-fixation (Fox-Dobbset al. 2010). If host plants must reduce their photosynthate
allocation to roots in invaded habitats, this could in turn reduce both
N-fixing symbiont activity and soil respiration, similar to effects seen
in a large-scale girdling experiment in a boreal forest by Högberget al. (2001). Potentially compounding this effect, elephants may
reduce tree cover within invaded savannas over the longer term, further
reducing N inputs into the system, as has been shown in other areas of
East Africa where removal of acacia species reduces both the total
content and mineralization of C and N in soils (Glaser et al.2001). Finally, Acacia drepanolobium has density-dependent
effects on the productivity of understory plants (Riginos et al.2009), and thus the carbon fixation performance of invaded trees may
also be linked to understory productivity. By increasing the mortality
(Riginos et al. 2015) and decreasing the performance of this
monodominant tree species, P. megacephala invasion may
fundamentally alter carbon cycling and connected ecosystem processes in
these savannas.
Acknowledgements: We thank the Kenyan government
(NACOSTI/P/18/4376/9459) for their permission to conduct this work.
Gabriella Mizell, Nelly Maiyo, Jackson Ekadeli, Gilbert Buseinei, Isaac
Kipkoech, and John Lemosiany provided excellent field assistance. Mpala
Research Centre administration and the Ol Pejeta Conservancy management
team (particularly Samuel Mutisya and Benard Gituku) provided
substantial logistical support. This research was supported by a
University of Florida International Center RADS grant to PDM, a
Smithsonian ForestGeo grant to PDM, a National Geographic Society Young
Explorer grant to PDM, and a National Science Foundation grant (NSF DEB
1556905) to TMP, CR and JRG.