Diss. ETH No. 16988

Variation in growth responses among and within native and invasive juvenile trees in Seychelles A dissertation submitted to the SWISS FEDERAL INSTITUTE OF TECHNOLOGY ZURICH for the degree of

Doctor of Natural Sciences presented by

Eva-Maria Rita Schumacher Dipl. Umwelt-Natw. ETH born October 10, 1973 citizen of Vilters-Wangs SG Accepted on the recommendation of Prof. Dr. Peter J. Edwards, examiner Dr. Hansjörg Dietz, co-examiner Dr. Karl Fleischmann, co-examiner Dr. Luc Gigord, co-examiner 2007

Contents

Summary

1

Zusammenfassung

3

General Introduction

7

Chapter 1 Influence of light and nutrient conditions on seedling growth of native and invasive trees in the Seychelles 23 Chapter 2 Influence of drought and shade on seedling growth of native and invasive trees in the Seychelles

45

Chapter 3 The role of forest gaps in woody plant invasions into secondary forests in the Seychelles

65

General Conclusions

89

Acknowledgements

95

Curriculum Vitae

97

Summary 1 Many recent studies have investigated the factors that make a small proportion of introduced plant species invasive in a new area. Most of these have focused on species invasiveness and habitat invasibility, especially in productive habitats. In contrast, studies on the invasiveness of plants in very nutrient poor habitats, including many tropical forests, are few. In this thesis, invasive tree species threatening tropical secondary rainforests on nutrient poor soils in the Seychelles were investigated. The main aims of the study were: (i) to evaluate the invasiveness of woody invasive plants in forests under varying light, nutrient, and water conditions, and (ii) to derive conclusions relevant for invasive species management and forest rehabilitation. 2 A common garden (pot) experiment performed with seedlings of six invasive and seven native tree species under varying light and nutrient levels showed that the successful invaders of closed-canopy secondary forests are either stress-tolerant invaders (e.g. Psidium cattleianum, Syzygium jambos) which respond similarly to native species in terms of growth, or fast-growing invaders with particular adaptations to nutrient-poor soils (e.g. Alstonia macrophylla). In contrast, the more typical, fast-growing alien species with no particular adaptations to nutrient-poor soils (e.g. Psidium guajava, Tabebuia pallida) seem to be restricted to relatively nutrient-rich sites in the lowlands. 3 A pot experiment performed with seedlings of five invasive and five native tree species grown under varying light and water levels (ambient watered and drought stressed) showed that even moderately severe dry spells did not cause a strong reduction of relative growth rates in either native or invasive species. Our results indicate that fast-growing invasive species may be able to cope with moderate stress through high phenotypic plasticity in growth allocation patterns. As such plastic responses may not be possible under conditions of severe or multiple stresses, the invasive species may be disadvantaged in such situations relative to slow growing, stress-tolerant native species. 

4 A field experiment was performed with seedlings of five invasive and four native species transplanted into understorey and gap plots in a secondary mid-altitude forest on Mahé island. Three main points emerged from the results. First, only a fraction of the invasive trees in Seychelles have the growth characteristics that allow them to take advantage of forest gaps. However, those species may grow considerably faster in gaps than most native species. Second, even fairly small disturbances to the canopy may facilitate the spread of invasive species. Light levels in the understorey of secondary Cinnamomum verum forest are relatively high (c. 10% of ambient), and at these levels seedlings respond strongly to even small increases in light. Third, for seedling growth in Cinnamomum verum secondary forests, the effect of forest gaps on belowground root competition is as important as the effect on light availability. 5 Overall, this study confirms that the alien tree flora of the granitic islands of the Seychelles includes the ‘classic’ plant invaders that can grow fast by exploiting pulses of increased resource levels. However, it also demonstrates that some successful invaders are adapted to effective nutrient-uptake on nutrient-poor soils while others are stresstolerant species with characteristics similar to that of native or endemic species. This finding has a number of important implications for forest management. First, weed risk assessment systems need to be aware of the particular risks that species adapted to low light and low nutrient environments may pose. Second, even in very nutrient poor tropical forests, only deeply shaded and undisturbed natural vegetation seems to be relatively resistant to all types of plant invaders. Third, multiple stresses (low light, low nutrient availability, drought stress and possibly other negative influences such as herbivory) may be an effective barrier against the great majority of invasive trees. Therefore, habitat management in such secondary forests has to be developed on the basis of a multi-dimensional concept of environmental stress effects.



Zusammenfassung 1 In den letzten Jahrzehnten hat sich ein umfassendes Forschungsinteresse für sogenannte invasive Pflanzen entwickelt. Invasive Pflanzen sind Pflanzenarten, welche durch den Menschen in ein neues Gebiet eingeführt wurden und sich dort schnell ausbreiten. Interessanterweise wird nur ein kleiner Anteil dieser gebietsfremden Arten invasiv, d.h. breitet sich schnell aus. Viele Studien haben die Eigenschaften von erfolgreichen invasiven Arten (invasiveness) oder von stark von Invasionen betroffenen Habitaten (invasibility) untersucht. Der Fokus lag dabei jedoch hauptsächlich auf Habitaten mit hoher Ressourcenverfügbarkeit. Im Gegensatz dazu gibt es nur wenige Studien zu den Eigenschaften von invasiven Pflanzen in nährstoffarmen Ökosystemen (zum Beispiel tropische Wälder). In dieser Dissertation wurden gebietsfremde Gehölzpflanzen untersucht, welche sich auf den ozeanischen Inseln der Seychellen in tropische Regenwälder auf nährstoffarmen Böden ausbreiten. Die Hauptziele der Studie waren: (i) die Analyse der Wachstumseigenschaften von invasiven Gehölzpflanzen unter verschiedenen Licht-, Wasser- und Nährstoffbedingungen, und (ii) die Entwicklung von Empfehlungen für die Kontrolle von invasiven Gehölzpflanzen und die Renaturierung von tropischen Wäldern in den Seychellen. 2 In einem Topfexperiment wurde die Reaktion von Jungpflanzen sechs invasiver und sieben einheimischer Gehölzpflanzenarten auf Unterschiede in der Verfügbarkeit von Licht und Nährstoffen untersucht. Dabei konnte gezeigt werden, dass erfolgreiche invasive Gehölze in den Sekundärwäldern der Seychellen entweder stresstolerante Arten sind (z.B. Psidium cattleianum, Syzygium jambos), die sich bezüglich ihrer Wachstumsraten ähnlich wie einheimische Pflanzen verhalten, oder schnell wachsende Arten, welche speziell an nährstoffarme Böden angepasst sind (z.B. Alstonia macrophylla). Im Gegensatz dazu sind die typischen, schnell wachsenden invasiven Pflanzen ohne spezielle Anpassung an nährstoffarme Böden (z.B. Psidium guajava, Tabebuia pallida) fast ausschliesslich in den relativ nährstoffreichen Küstengebieten invasiv. 

3 In einem Topfexperiment mit fünf invasiven und fünf einheimischen Gehölzpflanzenarten, welche unter verschiedenen Licht- und Wasserbedingungen (ausreichend gewässert bzw. trockengestresst) wuchsen, führte Austrocknung des Substrats bis zum Welkpunkt der Pflanzen weder bei den invasiven noch bei den ein– heimischen Arten zu einer starken Reduktion der Wachstumsraten. Meine Ergebnisse deuten darauf hin, dass schnell wachsende invasive Arten durch hohe phänotypische Plastizität der Biomasseallokation eine Wachstumsbeeinträchtigung durch mittelstarken Wasserstress teilweise kompensieren können. Dieses Reaktionsmuster ist aber unter sehr starkem oder mehrfachem Umweltstress (z.B. wenig Licht plus Trockenstress) nicht möglich. Unter solchen Bedingungen ist daher zu erwarten, dass die langsam wachsenden einheimischen Arten gegenüber den invasiven Arten im Vorteil sind. 4 Im Rahmen eines Feldexperiments in einem Sekundärwald auf mittlerer Höhenstufe auf der Insel Mahé wurde das Wachstum von Jungpflanzen von fünf invasiven und

vier einheimischen Arten in Waldlichtungen und im ungestörten Waldunterwuchs verglichen. Diese Untersuchung führte zu drei wichtigen Schlussfolgerungen. Erstens, nur ein Teil der invasiven Gehölzarten der Seychellen profitiert im Vergleich zu den einheimischen Arten von Waldlichtungen. Diese Arten wachsen in Lichtungen jedoch deutlich schneller als die meisten einheimischen Gehölze. Zweitens, bereits geringe Störungen können in diesen vom der invasiven Baumart Cinnamomum verum dominierten Sekundärwäldern die Wahrscheinlichkeit von Pflanzeninvasionen erhöhen. Die Lichtverfügbarkeit ist in diesen Wäldern vergleichsweise hoch (c. 10% des totalen Sonnenlichts erreicht den Waldboden), und in diesem Bereich der Lichtverfügbarkeit reagieren viele tropische Jungpflanzen sehr sensitiv auf eine Erhöhung des Lichtangebots. Drittens, für das Wachstum von Jungpflanzen in Sekundärwäldern, die von in C. verum dominiert werden, ist in Waldlichtungen eine reduzierte Wurzelkonkurrenz durch ausgewachsene Bäume von gleicher Bedeutung wie eine erhöhte Lichtverfügbarkeit. 5 Insgesamt bestätigt diese Studie, dass die invasive Gehölzflora der Seychellen schnell wachsende Arten umfasst, welche plastisch auf erhöhtes Licht- und Nährstoffangebot reagieren können. Die Studie zeigt jedoch auch, dass einige erfolgreiche invasive Pflanzen speziell an nährstoffarme Böden angepasst sind, oder eine hohe Toleranz von Umweltstress aufweisen und sich ähnlich wie einheimische Pflanzen verhalten. Diese Resultate sind aus mehreren Gründen für den Naturschutz in tropischen Wäldern auf den Seychellen relevant. Erstens, Früherkennungssysteme von potentiell problematischen gebietsfremden Arten (Risikoanalyse) müssen in Zukunft das



besondere Risiko von an nährstoffarme Böden angepassten oder schattentoleranten Arten besser berücksichtigen. Zweitens, sogar auf sehr nährstoffarmen Böden sind nur sehr schattige und wenig gestörte tropische Wälder wirklich resistent gegenüber Pflanzeninvasionen. Drittens, nur mehrfacher Umweltstress (wenig Licht, nährstoffarme Böden, Trockenstress und andere mögliche Faktoren wie Befall von Schädlingen) scheint effektiv gegen eine Invasion fast aller gebietsfremden Baumarten zu wirken. Deshalb sollten sich Renaturierungsprojekte in diesen Wäldern an einem multi-dimensionalen Konzept von Umweltstresseinflüssen orientieren.



General Introduction The issue of biological invasions

Invasive species are species that have been introduced by humans to a geographical region where they are not naturally present (alien species) and that are spreading into natural areas by themselves (Richardson et al. 2000, Falk-Petersen et al. 2006). The issue of biological invasions is of top priority, both in basic ecological research (e.g. Williamson 1996, Mack et al. 2000, Mooney et al. 2005, Dietz and Edwards 2006) and in nature conservation (e.g. McNeely et al. 2001, Millennium Ecosystem Assessment 2005). Invasions of alien species can result in massive economic and ecological costs (e.g. Mack et al. 2000, Millennium Ecosystem Assessment 2005, Pimentel et al. 2005). Together with other drivers of global change, biological invasions may bring about severe changes in ecosystem properties and processes (Vitousek et al. 1997, Dukes and Mooney 1999, Levine et al. 2003, Dukes and Mooney 2004), contribute to the extinction of species (e.g. Millennium Ecosystem Assessment 2005), and contribute to a homogenization of global biodiversity (McKinney and Lockwood 1999, Olden et al. 2004). The study of plant invasions has advanced the science of ecology, leading, for instance, to a better understanding of the impact of individual species on ecosystem processes (Vitousek et al. 1987), the role of herbivory in plant competition (Keane and Crawley 2002), the role of soil feedbacks for plant communities (Callaway and Aschehoug 2000, Kironomos 2002), and the potential for rapid evolution in plants (Lee 2002).

A conceptual model of invasion processes

The classic model of invasion biology states that three factors need to be integrated in order to understand and predict colonization processes by invasive species: the traits of the species (invasiveness), the vulnerability of the habitat to invasions (invasibility) and the number of propagules (e.g. seeds, cuttings) dispersed to a particular site (propagule pressure) (Williamson 1996, Lonsdale 1999). Correspondingly, in the focus of this study are traits that may make alien tree species invasive on the tropical oceanic islands of the Seychelles (Schumacher et al. 2003). The study is complementary to another study of the same system that investigated the roles of habitat invasibility and propagule pressure (Kueffer et al. 2003, Kueffer 2006). 

General Introduction

Invasiveness: the search for a common set of traits in invasive species Only a small fraction of introduced species becomes invasive (Williamson 1996), and much research has been devoted to understanding why this is the case (invasiveness, Rejmanek 1996, Kolar and Lodge 2001, Grotkopp et al. 2002, Daehler 2003, Hamilton et al. 2005). The search for general characteristics shared by successful plant invaders has not revealed any simple ‚formula’ for success. Nevertheless, some ecological attributes seem to be more often associated with invading species than others. These include: large plant size, high seed output, early maturity, high phenotypic plasticity, rapid growth, climatic matching between areas and (pre-)adaptation to anthropogenic site conditions and disturbances (see Williamson 1996, Kolar and Lodge 2001, Grotkopp et al. 2002, Daehler 2003, Rejmanek et al. 2005, Richardson and Pysek 2006). Based on these patterns, it has been generalised that many invasive plant species tend to be opportunists that rapidly take advantage of increased resource availability (Davis et al. 2000, Huston 2004), an idea that can be traced back to the very beginnings of invasion biology (Baker 1974). According to Grime’s concept of plant strategies (Grime 2001), many invasive plants can thus be classified as ruderals; and in line with Grime (2001), it has been hypothesised that these invaders are therefore not well adapted to environmental stress, especially drought (Alpert et al. 2000). However, while there are many studies that show the benefit of increased resource levels for invasive plants (e.g. Daehler 2003), data on stress tolerance of plant invaders is relatively scarce and ambiguous (e.g. Milchunas and Lauenroth 1995, Alpert et al. 2000, Stohlgren et al. 2001, Stratton and Goldstein 2001, Daehler 2003, Brock and Galen 2005, Pratt and Black 2006).

Limits of the current understanding of plant invasiveness The notion that a typical plant invader is a species that is able to exploit pulses of unused resources is supported by a recent, comprehensive review (Daehler 2003) showing that invasive plant species are not generally superior to native species, but rather that competition hierarchies between invasive and native species change along resource gradients, with invasive species general outcompeting native species under high resource availability but native species possessing a competitive advantage under low resource availability. In this sense, our modern understanding extends earlier models that tried to identify traits of ‘super-weeds’ that characterise an invasive species independently of habitat conditions (Baker 1974). However, this more recent understanding suggests that there may also be invasive species that are adapted to habitats of low resource availability and/or high environmental stress (e.g. Dietz and Edwards 2006). In fact, so far invasion biology has probably been biased 

General Introduction towards studying only the more apparent phases and types of plant invasions in highly disturbed habitats (primary invasions), while concomitant or later invasions into more natural and undisturbed habitat with lower resource levels have been neglected (secondary invasions) (e.g. Dietz and Edwards 2006). Few studies report that invasive plant species may also be adapted to low nutrient availabilities (e.g. Musil 1993), may be shade-tolerant (e.g. Martin and Marks 2006), or may colonize harsher habitats (e.g. Burke 2003, Dietz 2005).

Woody invasive species in tropical forests A more sophisticated understanding of species invasiveness interacting with habitat invasibility may be particularly important in the case of woody invasive species in the tropics. New woody plant species are introduced to the tropics at very high rates (Richardson 1998, Richardson et al. 2004). While many of them are already highly invasive on oceanic islands (e.g. Meyer 2000, Haysom and Murphy 2003, Kueffer et al. 2004a, 2004b, Lugo 2004), mainland tropical forests have so far been relatively unaffected by plant invasions (Fine 2002). It is possible that the unique biogeographic situation of oceanic islands (small land masses with isolated floras and low species diversity) contributes to these differences (Whittaker 1998, Denslow 2003). Alternatively, there may be a lower degree of habitat disturbance and/or lower propagule pressure of invasive species in mainland tropical forests (cf. Kueffer 2006). In the latter case, a better understanding of successful woody plant invasions on oceanic islands is highly relevant, not only for the protection and restoration of island habitats, but also for the prevention of future invasions into mainland habitats.

The response of tropical invasive woody species to resource gradients In general, light is the most important resource limiting plant growth in a tropical forest (e.g. Whitmore 1998, Turner 2001). For many tree species regeneration is only possible in forest gaps, though small, shade tolerant seedlings may persist in the form of a “seedling bank”. It has been hypothesised that the resistance of mainland tropical forests to plant invasions can be explained by the low frequency of forest gaps (Fine 2002). In fact, tropical forests on the mainland and on oceanic islands seem to be more invaded both after anthropogenic (e.g. Kendle and Rose 200, Kueffer et al. 2004a, Baret

and Strasberg 2005, Totland et al. 2005) and natural (e.g. Leps et al. 2002, Bellingham et al. 2005) disturbances of the canopy. Many tropical invasive species are lightdemanding (e.g. McDonald et al. 1991, Meyer 1998, Lavergne et al. 1999, Weber 2003, Baret et al. 2004, Meyer and Lavergne 2004). Ecophysiological experiments in Hawaii 

General Introduction showed that many invasive plant species use light more efficiently than native species, particularly in high-light environments (Pattison et al. 1998, Baruch et al. 2000), while under low light native species may outcompete invasive species (Fleischmann 1999). Soils are generally very nutrient poor in tropical forests (Vitousek and Sanford 1986), especially on oceanic islands (Vitousek 2004). Soil infertility may therefore also be another important factor limiting invasions into tropical forests (cf. Kueffer 2006). For instance, fertilization increased abundance of alien species in tropical montane forests in Hawaii (Ostertag and Verville 2002). Tropical invasive plants may also be more efficient in sequestering the nutrient needed for growth. For example, Baruch & Goldstein (1999) demonstrated that invasive plants in Hawaii made more efficient use of increased nutrient resources and could achieve higher growth rates than the native Hawaiian flora. Vitousek and Walker (1989) found that the invasive species Psidium cattleianum showed stronger growth responses to elevated nutrient availability (shoot

height increment and dry mass accumulation) than did the native species Metrosideros polymorpha. Generally, it seems that fast-growing species make more efficient use of increased nutrient availability, whereas shade-tolerant species often do not respond (Burslem et al. 1995, Raaimakers and Lambers 1996, Turner 2001). Light and nutrient availability may also interact in their effects on plant growth in tropical forests. For example, Thompson et al. (1992) showed that with decreasing light, photosynthetic adjustments resulted in a lower light compensation point, and that these were accentuated by low nutrient supply. Grubb (1996) hypothesized that competition for nutrients is most important for plants in canopy gaps where they grow faster than in shade, and that there is commonly an interaction between nutrient supply and degree of shade in their influence on growth rate. Hence, interactions between nutrients and light may also influence invasion success. Duggin and Gentle (1998), for example, showed that the invasive shrub Lantana camara significantly increased growth under high light conditions, especially in combination with high nutrient levels. During dry spells, mortality of tropical forest seedlings is very high (Turner 1990), particularly in the understorey (Veenendaal et al. 1996, Poorter and Hayashida-Oliver 2000). The root systems of seedlings in gaps are usually more extensive and thus allow better access to water (Fisher et al. 1991), whereas in deep shade seedlings do not grow fast enough to produce a substantial root system (Coomes and Grubb 2000, Poorter and Hayashida-Oliver 2000). Seedlings adapted to both shade and drought often have long-lived, thick leaves, relatively inflexible low SLA, and low LAR that allow them to persist in deep shade (Coomes and Grubb 2000). Thus, water stress could be of importance even for plant invasions into tropical forests (cf. Condit 1998). However, 10

General Introduction there have been no studies comparing the growth of native and invasive tropical tree species under conditions of water stress. Overall, the knowledge from studies on tropical tree seedlings indicates that light, nutrients, water and their combinations may have an important influence on the regeneration of invasive woody plants in tropical forests. However, very few studies have investigated the responses of tropical invasive tree species to variation in resource availabilities, and even fewer studies have studied the combined effect of several resources in short supply, or explicitly compared the performance of a broad range of native and invasive species.

Study system and species Geography The Republic of Seychelles is formed by two types of oceanic islands, c. 70 coralline (or ‘outer’) and 40 granitic (or ‘inner’) islands (Fig. 1A). The granitic Seychelles (4°-5° S, 55°-56° E) cover a total land area of c. 235 km2 (Fig. 1B). In contrast to the comparatively young volcanic or coralline rocks of other oceanic islands, the granitic islands of the Seychelles are formed of ancient, metamorphic bedrock (several 100 Mio year old granite) (Stoddart 1984). They are very isolated from continents, lying nearly 1600 km from the eastern coast of Africa, 930 km from Madagascar and 1700 from India. A Western Indian Ocean

B Granitic Seychelles

C Mahe Island

India

Mare aux Cochons

INDIAN OCEAN Africa

EQUATOR

Praslin

Granitic Seychelles

La Digue

Sans Souci

Silhouette

Ma

dag

asc

ar

Outer Seychelles

Fig. 1

Mahe Mascarenes 10 km

Maps of the geography of the Seychelles islands. A. The territory of the republic of Seychelles in the Western Indian Ocean with the two principal island groups: the outer, coralline and the inner, granitic islands. B. The four main islands Mahé, Praslin, Silhouette and La Digue of the group with the granitic islands. C. The largest island Mahé with the forestry station in Sans Souci where the two common garden experiments were performed and with the field site “Mare aux Cochons” where the transplant and phytometer experiments were performed.

11

General Introduction Mahé is the largest granitic island, comprising an area of 154 km2 or two thirds of the total land area of the granitic islands, and rising to 914 m asl. at its highest point – Morne Seychellois (Fig. 1C). During glacial periods, the last ending some 10’000 years ago, most of the land between and around the islands (bank of Seychelles) was free of water so that the granitic Seychelles formed a single landmass with an area of several 10’000 km2 (Cazes-Duvat and Robert 2001).

Biogeography The granitic islands of the Seychelles are of Gondwanan origin, and remained part of����������������������������������������������������������������������������������� the Indian land mass until 65 million years ago, having separated with India from Madagascar some 20–30 million years earlier (Briggs 2003). They are unique among the world oceans’ islands in their combination of isolation and continental origin. Due to this continental origin, the present flora of the Seychelles evolved from an established flora, rather than migrating to the islands. And the long period of isolation – longer than that of any other oceanic island – has led to a very high level of endemism in the flora. This history explains for instance the presence of species such as amphibians or the large seeded palm Lodoicea maldivica (Coco-de-Mer) that are not adapted to longdistance dispersal over the ocean. As a result of the former connection to both AfricaMadagacar and Asia, the flora includes species of both African and Asian affinity (Stoddart 1984).

Climate The climate of the Seychelles is equatorial with an annual rainfall of 1600 mm (at sea level on flat islands) to c. 3500 mm (on top of the highest peak, Morne Seychellois). Humidity is uniformly high, and mean temperatures at sea level range from 24°C to 30°C. The prevailing winds bring the wet northwest monsoon from December to March and the drier southeast monsoon from May to October (Stoddart 1984, CazesDuvat and Robert 2001).

The main habitats of the granitic islands of the Seychelles The vegetation of Seychelles granitic islands may be classified into seven main habitat types (Vesey-Fitzgerald 1940, Procter 1984, Carlstroem 1996, Wise 1998): coastal vegetation, mangrove forest, lowland forest (below c. 200 m asl.), intermediate-altitude forest (c. 200 - 650 m asl.), mountain mist forest (c. 650 - 900 m asl.), and, along the whole altitudinal gradient, palm forest and inselberg (or glacis) vegetation (i.e. sparse shrub vegetation on exposed granite outcrops). All of these habitats, except inselberg vegetation and some small pockets of mountain mist forest, have been heavily altered 12

General Introduction through human activities and most parts of the island had been completely deforested by the beginning of the 20th century.

Study species On the granitic islands, Friedmann (1994) and Robertson (1989) recorded a total of 370 dicotyledonous and monocotyledonous woody plant species. Among these species 105 were native (of which 54 percent were endemic) and 265 introduced. Besides the shrub and tree species, this includes six palm and four pandans (Pandanus spp.) species that are characteristic components of many habitats. These were not considered in this study. Grasses and herbs are of low importance in the flora. All the native tree species used in this study are common in intermediate-altitude forests (Tab. 1) apart from the two rather rare endemic species Syzygium wrightii and Psychotria pervillei, that were included in the pot experiment with varying light and nutrient levels (chapter 1). The set of native species covers some of the most important plant families of the Seychelles flora, including the Rubiaceae that is the family with by far the highest number of endemic species. Important families in the Seychelles flora not included in this study are the Araliaceae, Dilleniaceae, Dipterocarpaceae, Moraceae, and Sapotaceae. Apart from Alstonia macrophylla, all of the seven invasive tree species studied were introduced to Seychelles at least six decades ago (Tab. 1). They were introduced for a variety of purposes such as reforestation (e.g. Alstonia macrophylla, Sandoricum koetjape, Tabebuia pallida), fruit trees (e.g. Psidium gujava, Psidium cattleianum, Syzygium jambos), or spice crops (e.g. Cinnamomum verum). The four invasive tree species A. macrophylla, C. verum, P. cattleianum and S. jambos are among the six woody plant species which are the most problematic invasive species on the granitic islands of the Seychelles (Kueffer and Vos 2004). T. pallida is also considered as a problematic invasive species, but chiefly in lowland sites. S. koetjape is naturalized and widely planted but only slowly spreading because its large seeds are not actively dispersed. And the alien fruit tree Psidium guajava is currently invasive only on some small offshore islands of typically a few 100 hectares (Hill 2002).

13

14 Flacourtiaceae Rubiaceae Erythroxylaceae Melastomataceae Rubiaceae Rubiaceae Myrtaceae

Bois Merle

Bois Dur

Café Marron Petite Feuille

Bois Calou

Café Marron Grande Feuille

Bois Coulevre

Bois de Pomme

Aphloia theiformis

Canthium bibracteatum

Erythroxylum sechellarum

Memecylon eleagni

Paragenipa wrightii

Psychotria pervillei

Syzygium wrightii

Native

Bigogniaceae

Calice du Pape

Meliaceae

Santol

Sandoricum koetjape

Tabebuia pallida

Myrtaceae

Guave

Psidium guajava

Myrtaceae

Myrtaceae

Guave de Chine

Psidium cattleianum

Jambrosa

Lauraceae

Cinnamon

Cinnamomum verum

Syzygium jambos

Apocynaceae

Family name

Bois Jaune

Creole name

Alstonia macrophylla

Invasive

Species

native (endemic) native (endemic) native (endemic) native (endemic) native (endemic)

native

native

alien

alien

alien

alien

alien

alien

alien

Status

19th century (?) early 20th century

tropical America IndoMalaysia IndoMalaysia tropical America 1911

1870 or before

1870 or before

1772

c. 1960

Year of introduction

Brazil

IndoMalaysia Sri Lanka, India

Native range

20

4

6

10

7

8

12

10

10

25

10

7

15

15

Maximal height (m)

G, I G, I

birds fruit bats

G, I

G, L, I

G, I

G, L, I, M

G, I, M

C, L, P

L, I, M

L, I

C, L

100-700

200-900

0-700

0-900

0-900

0-700

0-800

0-200

100-900

0- 400

0-200

100-900

0-900

all habitats L, I ,M, P

0-600

Altitudinal range (m asl.)

C, G, L

Invaded habitats

birds

birds

birds

birds

birds

birds

wind

fruit bats

humans

birds fruit bats birds fruit bats

birds

wind

Mode of dispersal

x

x

x

x

x x

x

x

x x

x

x

x

x

x

x

x

x

RW

x

x

x

x

x

x

RN

x

x

x

x

x

x

x

x

x

TP

Tab. 1 Characterization of the species used in the common garden and transplant experiments. Invaded habitats: coastal forests (C), glacis (inselbergs) (G), lowland forest (L), intermediate-altitude forest (I), mountain forest (M) and palm forest (P). And their use in the different experiments: common garden with light and nutrients (RN), light and water (RW) and the transplant experiment (TP). Nomenclature and maximal stem height was taken from Friedmann (1994) and altitudinal range adapted from Wise (1999).

General Introduction

General Introduction

Outline of the thesis This is the first extensive comparative study of growth responses of seedlings of native and invasive tropical tree species to varying light, nutrient and water conditions. The results of this study are complementary to investigations on the invasibility of forest habitats in the same study system (Kueffer 2006). The main objective was to investigate whether invasive trees share certain growth characteristics that differentiate them from native species. To investigate the functional traits of species, I performed two common garden pot experiments (chapter 1 and 2) and a field transplant experiment (chapter 3) in which seedlings of invasive and native species were grown under varying light and nutrient or water conditions. In chapter 1 – Influence of light and nutrient conditions on seedling growth of native and

invasive trees in the Seychelles – I compare the performance of seedlings of invasive and native species (e.g. relative growth rate of dry weight) under different light and nutrient conditions. The main objective of this study was to investigate whether species invasive in very nutrient-poor tropical forests show a similarly high ability to exploit increased levels of light and nutrients comparable to that described for invasive plant species from nutrient-rich habitats. In chapter 2 – Influence of drought and shade on seedling growth of native and invasive trees in the Seychelles – I compare the performance of seedlings of invasive and native species under different light and water conditions. The main aim of this study was to test whether native and invasive species from a tropical oceanic island with a very high variability of rainfall patterns at the scale of a few square kilometres would generally differ in their tolerance of dry spells. In chapter 3 – The role of forest gaps in woody plant invasions into secondary forests in the Seychelles – I compare the responses of seedlings of invasive and native species to gap conditions in an intermediate-altitude forest on Mahé, and I compare the results of this transplant experiment with those from a pot experiment where the same species set and similar light and nutrient conditions were used (see chapter 1). The aims of this study were (i) to investigate the role of forest gaps in plant invasions into secondary tropical forests dominated by the alien tree Cinnamomum verum, to (ii) evaluate the effects of varying environmental factors from the understorey to the gap centre on seedlings growth, and (iii) to test the transferability of results gained under controlled conditions in a pot experiment to near-to-natural conditions in the field.

15

General Introduction

Literature

Alpert, P., E. Bone, and C. Holzapfel. 2000. Invasiveness, invasibility and the role of environmental stress in the spread of non-native plants. Perspectives in Plant Ecology, Evolution and Systematics 3:52-66. Baker, H. G. 1974. The evolution of weeds. Annual Review of Ecology and Systematics 5:1-24. Baret, S., S. Maurice, T. Le Bourgeois, and D. Strasberg. 2004. Altitudinal variation in fertility and vegetative growth in the invasive plant Rubus alceifolius Poiret (Rosaceae), on Réunion island. Plant Ecology 172:265-273. Baret, S., and D. Strasberg. 2005. The effect of opening trails on exotic plant invasion in protected areas on La Réunion island (Mascarene Archipelago, Indian Ocean). Revue d’écologie - La terre et la vie 60:325-332. Baruch, Z., and G. Goldstein. 1999. Leaf construction cost, nutrient concentration, and net CO2 assimilation of native and invasive species in Hawaii. Oecologia 121:183-192. Baruch, Z., R. R. Pattison, and G. Goldstein. 2000. Responses to light and water availability of four invasive Melastomataceae in the Hawaiian islands. International Journal of Plant Sciences 161:107-118. Bellingham, P. J., E. V. J. Tanner, and J. R. Healey. 2005. Hurricane disturbance accelerate invasion by the alien tree Pittosporum undulatum in Jamaican montane rain forests. Journal of Vegetation Science 16:675-684. Briggs, J. C. 2003. The biogeographic and tectonic history of India. Journal of Biogeography 30:381-388. Brock, M. J., and C. Galen. 2005. Drought tolerance in the Alpine Dandelion, Taraxacum ceratophorum (Asteraceae), its exotic congener T. officinale, and interspecific hybrids under natural and experimental conditions. American Journal of Botany 92:1311-1321. Burke, A. 2003. Inselbergs in a changing world - global trends. Diversity and Distributions 9:375-383. Burslem, D. F. R. P., P. J. Grubb, and I. M. Turner. 1995. Responses to nutrient addition among shade-tolerant tree seedlings of lowland tropical forest in Singapore. Journal of Ecology 83:113-122. Callaway, R. M., and E. T. Aschehoug. 2000. Invasive plants versus their new and old neighbors: A mechanism for exotic invasion. Science 290:521-523. Carlstroem, A. 1996. Endemic and threatened plant species on the granitic Seychelles. Conservation & National Parks Section, Division of Environment, Ministry of Foreign Affairs, Planning and Environment, Mahé, Seychelles. Cazes-Duvat, V., and R. Robert. 2001. Atlas de l’environnement côtier des îles granitiques de l’archipel des Seychelles. Université de La Réunion & CIRAD-Emvt, St. Denis, La Réunion & Montpellier, France. Condit, R. 1998. Ecological implications of changes in drought patterns: Shifts in forest composition in Panama. Climatic Change 39:413-427. Coomes, D. A., and P. J. Grubb. 2000. Impacts of root competition in forests and woodlands: A theoretical framework and review of experiments. Ecological Monographs 70:171-207. Daehler, C. C. 2003. Performance comparisons of co-occurring native and alien invasive plants: Implications for conservation and restoration. Annual Review of Ecology and Systematics 34:183-211. Davis, M. A., J. P. Grime, and K. Thompson. 2000. Fluctuating resources in plant communities: A general theory of invasibility. Journal of Ecology 88:528-534.

16

General Introduction Denslow, J. S. 2003. Weeds in paradise: Thoughts on the invasibility of tropical islands. Annals of the Missouri Botanical Garden 90:119-127. Dietz, H. 2005. A mountain invasions special issue. Perspectives in Plant Ecology, Evolution and Systematics 7:135-136. Dietz, H., and P. J. Edwards. 2006. Recognition that causal processes change during plant invasion helps explain conflicts in evidence. Ecology 87:1359-1367. Duggin, J. A., and C. B. Gentle. 1998. Experimental evidence on the importance of disturbance intensity for invasion of Lantana camara L. in dry rainforest-open forest ecotones in northeastern NSW; Australia. Forest Ecology and Management 109:279-292. Dukes, J. S., and H. A. Mooney. 1999. Does global change increase the success of biological invaders? Trends in Ecology and Evolution 14:135-139. Dukes, J. S., and H. A. Mooney. 2004. Disruption of ecosystem processes in western North America by invasive species. Revista Chilena de Historia Natural 77:411-437. Falk-Petersen, J., T. Born, and O. T. Sandlund. 2006. On the numerous concepts in invasion biology. Biological Invasions 8:1409-1424. Fine, P. V. A. 2002. The invasibility of tropical forests by exotic plants. Journal of Tropical Ecology 18:687-705. Fisher, B. L., H. F. Howe, and S. J. Wright. 1991. Survival and growth of Virola surinamensis yearlings: Water augmentation in gap and understorey. Oecologia 86:292–297. Fleischmann, K. 1999. Relations between the invasive Cinnamomum verum and the endemic Phoenicophorium borsigianum on Mahé island, Seychelles. Applied Vegetation Science 2:3746. Friedmann, F. 1994. Flore des Seychelles. Orstom, Paris. Grime, J. P. 2001. Plant Strategies, Vegetation Processes, and Ecosystem Properties, 2nd edition. John Wiley & Sons, Chichester, New York, Toronto. Grotkopp, E., M. Rejmanek, and T. L. Rost. 2002. Toward a causal explanation of plant invasiveness: Seedling growth and life-history strategies of 29 Pine (Pinus) species. American Naturalist 159:396-419. Grubb, P. J. 1996. Rainforest dynamics: The need for new paradigms. Pages 215-233 in D. S. Edwards, W. E. Booth, and S. C. Choy, editors. Tropical Rainforest Research - Current Issues. Kluwer Academic Publishers, Dordrecht. Hamilton, M. A., B. G. Murray, M. W. Cadotte, G. C. Hose, A. C. Baker, C. J. Harris, and D. Licari. 2005. Life-history correlates of plant invasiveness at regional and continental scales. Ecology Letters 8:1066-1074. Haysom, K. A., and S. T. Murphy. 2003. A Global Review of the Status of Invasiveness of Forest Tree Species Outside their Natural Habitat. Forest Biosecurity and Protection Working Paper FBS/3E Forestry Department, FAO, Rome. Hill, M. J. 2002. Biodiversity surveys and conservation potential of Inner Seychelles islands. Atoll Research Bulletin:495. Huston, M. A. 2004. Management strategies for plant invasions: Manipulating productivity, disturbance, and competition. Diversity and Distributions 10:167-178. Keane, R. M., and M. J. Crawley. 2002. Exotic plant invasions and the enemy release hypothesis. Trends in Ecology and Evolution 17:164-170. Kendle, A. D., and J. E. Rose. 200. Invasive Plants on Land Recovering from Desertification on Saint Helena Island. Pages 311-318 in G. Brundu, J. Brock, I. Camarda, L. Child, and M. Wade, editors. Plant invasions: Species Ecology and Ecosystem Management. Backhuys Publishers, Leiden, The Netherlands.

17

General Introduction Kironomos, J. N. 2002. Feedback with soil biota contributes to plant rarity and invasiveness in communities. Nature 417:67-70. Kolar, C. S., and T. S. Lodge. 2001. Progress in invasion biology: Predicting invaders. Trends in Ecology and Evolution 16:199-204. Kueffer, C. 2006. Impacts of woody invasive species on tropical forests of the Seychelles. PhD Thesis, ETH Zurich, Zurich. Kueffer, C., P. J. Edwards, K. Fleischmann, E. Schumacher, and H. Dietz. 2003. Invasion of woody plants into the Seychelles tropical forests: Habitat invasibility and propagule pressure. Bulletin of the Geobotanical Institute ETH 69:65-75. Kueffer, C., and P. Vos. 2004. Case Studies on the Status of Invasive Woody Plant Species in the Western Indian Ocean: 5. Seychelles. Forest Health & Biosecurity Working Papers FBS/45E Forestry Department, Food and Agriculture Organization of the United Nations, Rome, Italy. Kueffer, C., P. Vos, C. Lavergne, and J. Mauremootoo. 2004a. Case Studies on the Status of Invasive Woody Plant Species in the Western Indian Ocean. 1. Synthesis. Forest Health and Biosecurity Working Papers FBS/4-1E Forestry Department, Food and Agriculture Organization of the United Nations, Rome, Italy. Kueffer, C., P. Vos, C. Lavergne, and J. Mauremootoo. 2004b. Woody invasive species in the Western Indian Ocean: A regional assessment. Forest Genetic Resources Bulletin (FAO) 31:2530. Lavergne, C., J.-C. Rameau, and J. Figier. 1999. The invasive woody weed Ligustrum robustrum subsp. walkeri threatens native forests on La Réunion. Biological Invasions 1:377-392. Lee, C. E. 2002. Evolutionary genetics of invasive species. Trends in Ecology & Evolution 17:386-391. Leps, J., V. Novovotny, L. Cizek, K. Molem, B. Isua, W. Boen, R. Kutil, J. Auga, M. Kasbal, M. Manumbor, and S. Hiuk. 2002. Successful invasion of the neotropical species Piper aduncum in rain forests in Papua New Guinea. Applied Vegetation Science 5:255-262. Levine, J. M., M. Vila, C. M. D’Antonio, J. S. Dukes, K. Grigulis, and S. Lavorel. 2003. Mechanisms underlying the impacts of exotic plant invasions. Proceedings of the Royal Society of London Series B-Biological Sciences 270:775-781. Lonsdale, W. M. 1999. Global patterns of plant invasions and the concept of invasibility. Ecology 80:1522-1536. Lugo, A. E. 2004. The outcome of alien tree invasions in Puerto Rico. Frontiers in Ecology and Environment 2:265-273. Mack, R. N., D. Simberloff, W. M. Lonsdale, H. C. Evans, M. Clout, and F. A. Bazzaz. 2000. Biotic invasions: Causes, epidemiology, global consequences and control. Ecological Applications 10:689-710. Martin, P. H., and P. L. Marks. 2006. Intact forests provide only weak resistance to a shadetolerant invasive Norway maple (Acer platanoides L.). Journal of Ecology 94:1070-1079. McDonald, I. A. W., C. Thébaud, W. A. Strahm, and D. Strasberg. 1991. Effects of alien plant invasions on native vegetation remnants on La Réunion (Mascarene Islands, Indian Ocean). Environmental Conservation 18:51-61. McKinney, M. L., and J. L. Lockwood. 1999. Biotic homogenization: A few winners replacing many losers in the next mass extinction. Trends in Ecology & Evolution 14:450-453. McNeely, J. A., H. A. Mooney, L. E. Neville, P. Schei, and J. K. Waage, editors. 2001. A Global Strategy on Invasive Alien Species. IUCN, Gland, Switzerland.

18

General Introduction Meyer, J.-Y. 2000. Preliminary Review of the Invasive Plants in the Pacific islands (SPREP Member Countries). Pages 85-114 in G. Sherley, editor. Invasive species in the Pacific: A Technical Review and Draft Regional Strategy. SPREP, Apia, Samoa. Meyer, J.-Y., and C. Lavergne. 2004. Beautés fatales: Acanthaceae species as invasive alien plants on tropical Indo-Pacific Islands. Diversity and Distributions 10:333-347. Meyer, J. Y. 1998. Observations on the reproductive biology of Miconia calvescens DC (Melastomataceae), an alien invasive tree on the island of Tahiti (South Pacific Ocean). Biotropica 30:609-624. Milchunas, D. G., and W. K. Lauenroth. 1995. Inertia in plant community structure: State changes after cessation of nutrient enrichment stress. Ecological Applications 5:452-458. Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well-being: Biodiversity Synthesis. World Resources Institute, Washington, DC. Mooney, H. A., R. N. Mack, J. A. McNeely, L. E. Neville, P. J. Schei, and J. K. Waage, editors. 2005. Invasive Alien Species. A New Synthesis. Island Press, Washington, London. Musil, C. F. 1993. Effect of invasive Australian acacias on the regeneration, growth and nutrient chemistry of South African lowland fynbos. Journal of Applied Ecology 30:361-372. Olden, J. D., et. al. 2004. Ecological and evolutionary consequences of biotic homogenization. Trends in Ecology & Evolution 19:18-24. Ostertag, R., and J. H. Verville. 2002. Fertilization with nitrogen and phosphorus increases abundance of non-native species in Hawaiian montane forests. Plant Ecology 162:77-90. Pattison, R. R., G. Goldstein, and A. Ares. 1998. Growth, biomass allocation and photosynthesis of invasive and native Hawaiian rainforest species. Oecologia 117:449-459. Pimentel, D., R. Zuniga, and D. Morrison. 2005. Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecological Economics 52:273-288. Poorter, L., and Y. Hayashida-Oliver. 2000. Effects of seasonal drought on gap and understorey seedlings in a Bolivian moist forest. Journal of Tropical Ecology 16:481-498. Pratt, R. B., and R. A. Black. 2006. Do invasive trees have a hydraulic advantage over native trees? Biological Invasions 8:1331-1341. Procter, J. 1984. Vegetation of the Granitic Islands of the Seychelles. Pages 193-208 in D. R. Stoddart, editor. Biogeography and Ecology of the Seychelles Islands. DR W. Junk Publishers, The Hague, Boston, Lancaster. Raaimakers, D., and H. Lambers. 1996. Response to phosphorus supply of tropical tree seedlings: A comparision between a pioneer species Tapirira obtusa and a climax species Lecythis corrugata. New Phytologist 132:97-102. Rejmanek, M. 1996. A theory of seed plants invasiveness: The first sketch. Biological Conservation 78:171-181. Rejmanek, M., D. M. Richardson, J. I. Higgins, M. J. Pitcairn, and E. Grotkopp. 2005. Ecology of Invasive Plants: State of the Art. in H. A. Mooney, R. N. Mack, J. A. McNeely, L. E. Neville, P. J. Schei, and J. K. Waage, editors. Invasive Alien Species. A New Synthesis. Island Press, Washington, London. Richardson, D. M. 1998. Forestry trees as invasive aliens. Conservation Biology 12:18-26. Richardson, D. M., P. Binggeli, and G. Schroth. 2004. Plant Invasions - Problems and Solutions in Agroforestry. Pages 371-396 in G. F. Schroth G., C.A. Harvey, C. Gascon, H. Vasconcelos and A.M. Izac, editor. Agroforestry and Biodiversity Conservation in Tropical Landscapes. Island Press, Washington.

19

General Introduction Richardson, D. M., and P. Pysek. 2006. Plant invasions: Merging the concepts of species invasiveness and community invasibility. Progress in Physical Geography 30:409-431. Richardson, D. M., P. Pysek, M. Rejmánek, M. G. Barbour, F. D. Panetta, and C. J. West. 2000. Naturalization and invasion of alien plants: Concepts and definitions. Diversity and Distributions 6:93-107. Robertson, S. A. 1989. Flowering Plants of Seychelles. Royal Botanic Gardens, Kew. Schumacher, E., H. Dietz, K. Fleischmann, C. Kueffer, and P. J. Edwards. 2003. Invasion of woody plants into the Seychelles tropical forests: Species traits in the establishment phase. Bulletin of the Geobotanical Institute ETH 69:77-86. Stoddart, D. R., editor. 1984. Biogeography and Ecology of the Seychelles Islands. DR W. Junk Publishers, The Hague, Boston, Lancaster. Stohlgren, T. J., Y. Otsuki, C. A. Villa, M. Lee, and J. Belnap. 2001. Patterns of plant invasions: A case example in native species hotspots and rare habitats. Biological Invasions 3:37-50. Stratton, L. C., and G. Goldstein. 2001. Carbon uptake, growth and resource-use efficiency in one invasive and six native Hawaiian dry forest tree species. Tree Physiology 21:1327-1334. Thompson, W. A., P. E. Kriedemann, and I. E. Craig. 1992. Photosynthetic response to light and nutrients in sun-tolerant and shade-tolerant rain forest trees. 2. Leaf gas exchange and component processes of photosynthesis. Australian Journal of Plant Physiology 19:19-42. Totland, O., P. Nyeko, A.-L. Bjerknes, S. J. Hegland, and A. Nielsen. 2005. Does forest gap size affect population size, plant size, reproductive success and pollinator visitation in Lantana camara, a tropical invasive shrub? Forest Ecology and Management 215:329-338. Turner, I. M. 1990. The seedling survivorship and growth of three Shorea species in a Malaysian tropical rain forest. Journal of Tropical Ecology 6:469-478. Turner, I. M. 2001. The Ecology of Trees in the Tropical Rain Forest. Cambridge University Press, Cambridge. Veenendaal, E. M., M. D. Swaine, V. K. Agyeman, D. Blay, I. K. Abebrese, and C. E. Mullins. 1996. Differences in plant and soil water relations in and around a forest gap in West Africa during the dry season may influence seedling establishment and survival. Journal of Ecology 84:83-90. Vesey-Fitzgerald, D. 1940. On the vegetation of the Seychelles. Journal of Ecology 28:465-483. Vitousek, P. M. 2004. Nutrient Cycling and Limitation. Hawai’i as a Model System. Princeton University Press, Princeton. Vitousek, P. M., H. A. Mooney, J. Lubchenco, and J. M. Melillo. 1997. Human domination of earth’s ecosystems. Science 277:494-499. Vitousek, P. M., and R. E. Sanford. 1986. Nutrient cycling in moist tropical forests. Annual Review of Ecology and Systematics 17:137-167. Vitousek, P. M., and L. R. Walker. 1989. Biological invasion by Myrica faya in Hawai’i: Plant demography, nitrogen fixation, ecosystem effects. Ecological Monographs 59:247-265. Vitousek, P. M., L. R. Walker, L. D. Whiteaker, D. Mueller-Dombois, and P. A. Matson. 1987. Biological Invasion by Myrica faya Alters Ecosystem Development in Hawaii. Science 238:802804. Weber, E. 2003. Invasive Plant Species of the World. A Reference Guide to Environmental Weeds. CABI Publishing, Oxon, UK & Cambridge, USA. Whitmore, T. C. 1998. An Introduction to Tropical Rain Forests, 2nd edition. Oxford University Press, Oxford, New York, Tokyo. Whittaker, R. J. 1998. Island Biogeography. Ecology, Evolution, and Conservation. Oxford University Press, Oxford.

20

General Introduction Williamson, M. 1996. Biological Invasions. Chapman & Hall, London, New York ,Tokyo. Wise, R. 1998. A Fragile Eden: Portraits of the Endemic Flowering Plants of the Granitic Seychelles. Princeton University Press.

21

Chapter 1 Influence of light and nutrient conditions on seedling growth of native and invasive trees in the Seychelles

23

Chapter 1

Abstract 1 An increasing number of studies suggest that plant invasions may also occur in stressed and relatively undisturbed habitats. It is, therefore, important to know whether and how traits of plant species invasive in such habitats differ from those of species invasive in disturbed and resource rich habitats. 2 We studied the growth performance of seedlings of native and invasive tree species from nutrient-poor secondary forests in the tropical Seychelles. We hypothesised that the relative performance of native and invasive species would change predictably along resource gradients, with native species performing relatively better at very low levels of resource availability and invasive species performing better at higher levels. 3 To test this hypothesis, we performed a common garden (pot) experiment using seedlings of six invasive and seven native tree species grown under three levels of light (65%, 11% and 3.5% of ambient light) and two levels of nutrients (low and high). 4 The seedlings of invasive species tended to have higher growth rates (RGR), higher specific leaf areas (SLA) and higher leaf nutrient contents than the seedlings of native species. They also exhibited greater plasticity in biomass and nutrient allocation (i.e. greater plasticity in SLA, LAR, RSR and leaf nutrient contents) in response to varying resource availability. However, differences in mean values between the groups of invasive and native species were generally small compared with variation within groups. 5

We conclude that successful invaders of closed-canopy secondary forests on nutrient-poor soils in the Seychelles appear to be either stress-tolerant, possessing growth traits similar to those of the native species, or fast-growing but with particular adaptations to nutrient-poor soils. In contrast, the more typical, fastgrowing alien species with no particular adaptations to nutrient-poor soils seem to be restricted to relative nutrient-rich sites in the lowlands.

Keywords Biomass allocation, common garden experiment, growth plasticity, invasiveness, tropical tree seedlings, light availability, nutrient limitation, Seychelles

24

Light & Nutrient

Introduction Although only a small fraction of introduced species become invasive (Williamson 1996), these few species may cause enormous ecological damage and economic costs (Millennium Ecosystem Assessment 2005, Pimentel et al. 2005). For this reason, much research has been devoted to understanding what makes some species invasive (invasiveness, Kolar and Lodge 2001, Grotkopp et al. 2002, Daehler 2003). One idea that continues to attract interest is that invasive species share a common suite of functional traits (Baker 1974). Traits such as high growth rate, high reproductive allocation, high plasticity and high potential for acclimation have been associated with invasive species (e.g. Rejmanek 1996, Williamson 1996). And indeed, one general pattern that has emerged is that invasive species tend to produce leaves with a higher specific leaf area (SLA) than native or non-invasive relatives (Baruch and Goldstein 1999, Daehler 2003, Richardson and Pysek 2006). And, the most important predictor of invasiveness in 29 Pinus species was relative growth rate (RGR) of seedlings, and the main trait responsible for differences in RGR between invasive and non-invasive pines was SLA (Grotkopp et al. 2002). Invasive species also tend to produce leaves with high nitrogen content (Dukes and Mooney 1999, Ehrenfeld 2003, Niinemets et al. 2003), and are capable of fast growth under conditions of high resource availability (for tropical woody invaders, e.g. Baruch et al. 2000). In addition, invaders often show higher phenotypic plasticity than native species (Daehler 2003, Richards et al. 2006, Richardson and Pysek 2006 and references therein). However, despite these trends, it is still not possible to generalise about what makes some species invasive (Kolar and Lodge 2001, Daehler 2003). One reason may be that the traits that make a species a successful invader vary according to both habitat conditions and the stage of the invasion (Alpert et al. 2000, Dietz and Edwards 2006). Daehler (2003) found that alien invaders were not always competitively superior to native species, but rather that the competitive hierarchy between native and alien species shifted according to both resource availability and disturbance regime. On the basis of such observations, Richards et al. (2006) proposed a scheme for classifying invaders according to their performance relative to co-occuring native species under stressed and favourable conditions. They called an invader that outperforms native species under stressed conditions a ‘jack-of-all-trades’, while one that succeeds only under favourable conditions is a ‘master-of-some’; and a species that succeeds under both kinds of conditions is a ‘jack-and-master’. The fact that this range of outcomes is possible makes it important that studies on invasiveness include a wide range of different habitat conditions. 25

Chapter 1 Because alien plants tend to be most abundant in disturbed, resource-rich habitats (compare Alpert et al. 2000, Davis et al. 2000), most research on invasive plants has been conducted in such sites or stages (primary invasion sensu Dietz and Edwards 2006). Thus, some of the common traits of invaders, such as high SLA or RGR, may simply be interpreted as typical features of species adapted to disturbed and resource-rich sites (in tropical trees see e.g. Veneklaas and Poorter 1998). However, an increasing number of studies show that invasions also occur in stressed and/or relatively undisturbed habitats (e.g. Stohlgren et al. 1999, Burke 2003, Dietz 2005, Martin and Marks 2006). Interesting examples of this phenomenon are tropical forests on oceanic islands, which are among the most heavily invaded of habitats (Denslow 2003), even though their soils are often very nutrient poor (Vitousek 2004, Kueffer 2006). Further, plant invasions on oceanic islands sometimes occur in closed vegetation (e.g. Huenneke and Vitousek 1990, Fleischmann 1997), while invasions in continental tropical forests have often been attributed to higher light levels resulting from disturbance (Fine 2002). The granitic island of Mahé (Republic of Seychelles) offers a wide diversity of terrestrial

habitats, all of which have to some extent been invaded by alien species. These habitats range from relatively nutrient-rich, open vegetation in the lowlands to very nutrientpoor, closed secondary forest in the uplands. However, despite the high environmental heterogeneity, the total number of woody species is rather small, with around 20 abundant native species and even fewer invasive species. As a consequence, many species occur across a broad ecological range, with the same set of invasive and native species growing under both resource-rich and resource-poor conditions. Based on the work of Fleischmann (1999), we hypothesised that the relative performance of native and invasive species would change predictably along resource gradients, with native species outperforming invasive species under conditions of low light and low nutrients but with invasive species being better able to exploit higher levels of these resources. To test this hypothesis, we grew seedlings of six invasive and seven native species under a range of light and nutrient conditions, and analysed the data to compare growth rates and patterns of biomass and nutrient allocation under contrasting resource conditions.

26

Light & Nutrient

Methods General study area The island of Mahé (4° S, 55° E, 154 km2, 0 - 900 m asl) is the largest island in the inner Seychelles group. It is built of granite 550 to 650 Mio years old that has never been covered by the ocean. The soils are typically ferrasols and have a pH of c. 4.5; and due the very long history of weathering (Braithwaite 1984), they are poor in most nutrients and especially phosphorus (cf. Kueffer 2006). Inland forests in Seychelles are characterized by a tropical climate with a mean annual rainfall of between 1600 and 3500 mm depending on altitude (Stoddart 1984). Although there is no pronounced seasonality in precipitation, the period from May to October is generally drier than the rest of the year. Monthly mean temperatures range from 26°C to 28°C (Meteo Seychelles). In most inland forests the invasive tree Cinnamomum verum dominates the canopy.

Species We selected six invasive and seven native tree species that are abundant in the secondary forests of the inner islands of the Seychelles and for which seeds were available at the time of the experiment (Tab. 1). The native species included two indigenous and five endemic species. The majority of the invasive species were introduced to the islands in the late 19th or early 20th century, but Cinnamomum verum and Syzygium jambos have been present for more than 200 years (Kueffer and Vos 2004). In choosing species for the study, we avoided including closely related (congeneric) species pairs within one group. All species have small to medium sized seeds 2 to 10 mm in diameter except for the invasive Sandoricum koetjape and Syzygium jambos, which have seeds of 15 to 20 mm diameter. Nomenclature follows Friedmann (1994).

27

Chapter 1 Table 1 Characterization of the species used in the common garden experiment. The three different experimental runs S1, S2 and S3 were started on 27.10.2002 (duration 201-226 days), 3.6.2003 (duration 180 days) and 6.12.2003 (duration 210 days), respectively. Nomenclature and maximal stem height was taken from Friedmann (1994). Species

Family

Maximal stem height (m)

Experimental run

Invasives Alstonia macrophylla

Apocynaceae

15

S2

Cinnamomum verum

Lauraceae

15

S2

Psidium cattleianum*

Myrtaceae

7

S1

Sandoricum koetjape

Meliaceae

25

S2

Syzygium jambos*

Myrtaceae

10

S2

Tabebuia pallida*

Bigogniaceae

10

S1

Aphloia theiformis+

Flacourtiaceae

12

S3

Canthium bibracteatum

Rubiaceae

8

S3

Erythroxylum sechellarum†

Erythroxylaceae

7

S1

Melastomataceae

10

S3

Natives

Memecylon eleagni†

Rubiaceae

6

S1

†

Rubiaceae

4

S3

Syzygium wrightii†

Myrtaceae

20

S1

Paragenipa wrightii† Psychotria pervillei

subsp. madascariensis var. seychellensis * species that are also invasive in many other tropical regions † species endemic to the Seychelles +

Common garden experiment A common garden experiment was performed to compare the growth of native and invasive juvenile trees under varying light and nutrient levels. The experiment was conducted on a flat and unshaded lawn on the eastern slope of Morne Seychellois at the Sans Souci forestry station (4°38’S and 55°27’E; 380 m asl., Fig. 1A). Except for Alstonia macrophylla, all seedlings were grown from seed. Between 5 and 15 trees per species were sampled in forest vegetation on Mahé, the ripe fruits being collected either directly from parent trees or from the ground (Syzygium jambos). Immediately after collection the mixed seed of each species was sown into trays. For Alstonia macrophylla it was not possible to collect seed, and seedlings were collected from different forest sites on Mahé. Care was taken to select seedlings similar in size to those grown from seed. 28

Light & Nutrient tents

fence

building

small wall

car pass

tree

N

B 30cm

shading tissue

HR

poles

Base planks

LR

IR

30cm

10cm

160cm

A

pots

30cm 11cm

LR

HR Side view

IR

HR

LR

IR

IR

LR

HR

40cm

HR

50cm

LR

Fig. 1

180cm

50cm

180cm

20cm

20cm

160cm

100cm

LR

100cm

IR

140cm

HR

Front view

50cm

160cm

180cm

50cm

IR

30cm

A Plan view of the arrangement of shading tents used in the common garden experiment at Sans Souci forestry station: HR (high light), IR (intermediate light), LR (low light). B Design of the shading tents and the arrangement of pots within them. See text for further information.

When the seedlings had developed their first true leaves (three to six months after sowing), they were transplanted into 1-litre pots filled with local forest soil (see below). The plants were allowed to adjust to the pot environment for two weeks before the experiments were started. Because the timing of emergence varied, both between invasive and native species (compare also Schmitt and Riviere 2002) and between small- and larger-seeded species, similar-aged individuals of different species differed considerably in their biomass. To control for any influence of size variation at the start of the experiment on the results, we analysed not only relative growth rates (i.e. logistic growth) but also linear growth (i.e. (biomassEnd - biomassStart)/time) and total biomass at the end of the experiment, and we found no qualitative differences in the results.

We chose three levels of irradiance to represent typical light conditions within the forests: 65% ambient light for gap conditions (high radiation, HR), 11% ambient light for disturbed understorey (intermediate radiation, IR), and 3.5% ambient light for closed native forest (low radiation, LR). These light conditions were achieved by constructing wooden frames with sloping roofs (height 1 to 1.4 m; area 1.6 m x 1.8 m; Fig. 1B) and covering them with green shading cloths with the appropriate transmittance (Agroflor, Austria). These ‘tents’, as we called them, were calibrated using a PARsensor to determine the light level inside as a percentage of that outside. To prevent 29

Chapter 1 the humidity and temperature within the tents rising above the ambient levels, there was no shading cloth around the bottom 50 cm of the frame. In practice, temperatures tended to be somewhat higher (36°C vs. 32°C) and relative humidities lower (60% vs. 75%) under HR compared to IR and LR (measurements made around noon on sunny days). Two nutrient levels were chosen that were intended to correspond to conditions of low (LN) and high nutrient (HN) availability in typical forest soils on Mahé. For the LN treatment we mixed organic forest topsoil with laterite soil (35% organic soil, 65% laterite soil, vol%), resulting in soil that was poorer in nutrients than most forest soils in the Seychelles (Kjedahl N 1.5 mg g-1; P 0.4 mg g-1). For the HN treatment the same mixture was used but 1 g of a slow release N-P-K-fertilizer (Osmocote 16:11:11, Osmocote, Scotland) was applied to each pot every two months. The tents were arranged in a block design with six replicates per light treatment (Fig. 1A). A split-plot design was used with light as the main-plot factor and nutrients as the split-plot factor. In each tent there were two plants per species, with one for each nutrient treatment. Aphloia theiformis was not included in the low-light treatment because of a

shortage of seedlings. The plants were redistributed monthly within the tents to avoid local position effects. Due to the changing availability of seeds and seedlings of the different species, the experiment was conducted in three series starting in October 2002, June 2003 and December 2003. Each series lasted for six to seven months (Tab. 1).

Data collection At the onset of each series, 4 to 6 randomly chosen seedlings of each species were harvested to determine the initial total dry weight. Thereafter, the following parameters were measured on all plants at one- to two-month intervals: stem height, number of leaves, leaf length and breadth, and stem diameter (using callipers). To estimate leaf area, linear regressions of leaf area against the product of leaf length and breadth were calculated for a sample of > 100 leaves per species. The sample leaves were placed beneath a glass plate and photographed with a digital camera (Nikon Coolpix 995, resolution at 2048 x 1536 pixels); the images were then used to determine leaf length and breadth using Adobe IllustratorTM 10, and area using Adobe PhotoshopTM 7.0 (cf. Dietz and Steinlein 1996). At the end of the experiment, plants were harvested and divided into leaves, stems plus

petioles, and roots. All material was then oven-dried at 80°C for 48 hours. Subsamples of leaf material were digested at 420°C with 98% H2SO4 and Merck Kjeltabs, and total 30

Light & Nutrient nitrogen and phosphorus concentrations of leaves were determined colorimetrically using a flow injection analyzer (FIA, TECATOR, Höganäs, Sweden). The raw data were used to calculate the following growth parameters: RGRDW Relative growth rate of total dry weight

ln (dry weight at end)-ln(dry weight at start) duration of experiment ln (leaf area at end)-ln(leaf area at start) duration of experiment

RGRLA

Relative growth rate of leaf area

SLA

Specific leaf area

leaf area dry leaf biomass

LAR

Leaf area ratio

leaf area dry plant biomass

RSR

Root:shoot ratio

dry root biomass dry shoot biomass



Stastistical analysis Due to the death of many seedlings in the low light / high nutrient combination, we performed two separate analyses of the data. In the first analysis, only the data for the low nutrient treatment were analysed. General linear models were used with light level, species status (native or invasive) and the corresponding interaction as fixed factors, and species identity (nested in species status) and shading tents (nested in light treatments) as random factors. To account for initial differences in size, leaf area of each plant at the start of the experiment was included as a covariable. The second analysis included plants growing at intermediate and high light levels with both fertilized and unfertilized pots. A similar statistical model was used, but with nutrient level and corresponding interactions as additional fixed factors. Plant growth (relative growth rate of total dry weight and leaf area), changes in plant morphology or allocation (SLA, LAR, RSR) and leaf nutrient contents (nitrogen N and phosphorus P and N:P ratio) were included as dependent variables. To remove heteroscedasticity, SLA, LAR and RSR were log-transformed, while N and P contents were square-root transformed prior to analysis. Syzygium wrightii was excluded from all analyses due to very high mortality in most treatments, and Aphloia theiformis was omitted from the first analysis because there was no low light treatment. All statistical analyses were performed with JMP V 6.0 (SAS Institute Inc., 2005). 31

Chapter 1

Results An average of 48% of seedlings subjected to both low light (LR) and high nutrients (HN) died. However, in the other treatments the mortality was much lower, ranging from 0 to 26% according to species and treatment, and with an overall mean value of 10%. There were no consistent differences in mortality between native and invasive species. For example, among the invasive species, all plants of Syzygium jambos died in the low light and high nutrient treatment, while all plants of Cinnamomum verum and Psidium cattleianum survived.

Responses of the species to variation in light and nutrient availability Relative growth rate The seedlings of both native and invasive species usually developed more biomass (relative growth rate of total dry weight, RGRDW) under higher light conditions (P < 0.001, Fig. 2A, Tab. 2), and the magnitude of the response was similar for both species groups (species status x light interaction, P = 0.95, Tab. 2). Mean RGRDW of native species was about 50% lower than that of invasive species under low light and 15% lower under intermediate and high light; however, neither of these differences was significant (P = 0.5) because the variation within both species groups was so high (Fig. 2A). Thus, within the group of invasive species, RGRDW ranged from 0.3 to 11.0 mg g-1 d-1 under low light, and from 6.2 to 20.6 mg g-1 d-1 under high light; the equivalent values for native species were 0.7 to 6.9 mg g-1 d-1 under low light, and 2.5 to 21.4 mg g-1 d-1 under high light. At intermediate (IR) and high light levels (HR) the seedlings of both native and invasive species had a higher RGRDW in the fertilized pots (HN) than in the unfertilized pots (LN) (P < 0.001, Fig. 2B, Tab. 3). This effect did not differ between light treatments (light x nutrient, P = 0.2), but invasive species responded more strongly to added nutrients than natives (species group x nutrient, P < 0.05, Tab. 3). Again, the mean growth rates of the invasive species were not consistently and significantly higher than those of the native species (P = 0.6) (Fig 2B, Tab. 3), partly because of the large variation within species groups. With the addition of nutrients, several species were able to almost double their growth rate; these included the native Memecylon eleagni and invasive Sandoricum koetjape under IR, and the native Paragenipa wrightii and invasive Syzygium jambos under HR. In contrast, nutrient addition reduced the RGRDW of some species; examples are the native Paragenipa wrightii and the invasive Tabebuia pallida under IR, and the native Psychotria pervillei under HR. 32

Light & Nutrient The relative growth rates of height (RGRH) and total leaf area (RGRLA) were both significantly correlated with RGRDW (r > 0.5, P < 0.01) and the data are therefore not shown.

Table 2 Results of ANOVA across all three light levels. Indicated are the F-ratios and significance levels (***: P < 0.001, **: P < 0.01: *: P < 0.05, significant ones in bold) of main effects and interactions for the following parameters: relative growth rate of dry weight (RGRDW), specific leaf area (SLA), leaf area ratio (LAR), root:shoot ratio (RSR), nitrogen to phosphorus ratio (N:P), N & P concentrations in leaves. RGRDW

SLA

LAR

RSR

N:P

N

P

0.6

2.9

3.4

0.1

0.4

4.7

5.7 *

116.5***

253.2 ***

101.9 ***

23.2 ***

27 ***

71.6 ***

12.7 ***

SxL

0.1

0.2

3.6 *

6.0 *

1.7

1.2

0.4

Initial leaf area

1.3

0.1

4.2 *

1

0.1

0.6

2.5

Species group† (S) Light (L)

†

native vs. invasive

Table 3 Results of ANOVA across two light and two nutrient levels. Indicated are the F-ratios and significance levels (***: P < 0.001, **: P < 0.01: *: P < 0.05, significant ones in bold) of main effects and interactions for the following parameters: RGRDW, SLA, LAR, RSR, N:P ratio, N & P concentrations in leaves. See table 2 for acronyms and further information. RGRDW

SLA

LAR

RSR

N:P

N

P

0.3

6.0 *

6.4 *

0.0

0.2

8.8 *

15.9 **

172.4 ***

397.4 ***

194.3 ***

64.0 ***

122.1 ***

8.6 ***

0.1

0.1

0.4

3.0

7.9 *

0.1

0.4

2.4

24.8 ***

3

0.3

8.2 *

16.4 **

61.6 ***

28.9 ***

SxN

5.3 *

0.3

3.4

3.5

0.1

8.3 *

16.4 **

LxN

1.6

1.1

4.6 *

3.9 *

0.7

13.3 ***

28.1 ***

Initial leaf area

1.4

5.0 *

7.1 **

1.2

0.5

0.1

3.0

Species group† (S) Light (L) SxL Nutrient (N)

†

native vs. invasive

33

A

20

RGRDW mg g-1 d-1

Chapter 1

15

B

Invasive Native

Intermediate radiation

High radiation

10

5

0

Low

Intermediate

High

High

Radiation

Fig. 2

Low

High

Low

Nutrient

RGRDW of invasive (black bars) and native (grey bars) species (mean ± SE). A Plants growing under three light levels with no fertilizer added. B Plants growing under two light levels and two nutrient levels. See text for further information.

Biomass allocation Specific leaf area (SLA) and leaf area ratio (LAR) were strongly affected by light conditions and also varied widely among species. For instance, the SLA of the invasive Alstonia macrophylla under low light was about three-times the value under high light (550 cm2 g-1 v. 170 cm2 g-1). In contrast, light conditions had a much smaller effect on the SLA of another invasive species Syzygium jambos, and values were always much lower than those of A. macrophylla (LR 120 cm2 g-1 , HR 95 cm2 g-1). Among the native species, the highest values of SLA were for Psychotria pervillei (c. 290 cm2 g-1 v. c. 120 cm2 g-1) and the lowest for Memecylon eleagni (c. 110 cm2 g-1 v. c. 75 cm2 g-1). In both species groups, mean SLA (Fig. 3) and LAR (data not shown) decreased strongly with increasing light availability (P < 0.001, Tab. 2), ���������������������������������� but there were ������������������������ no clear responses of SLA and LAR to nutrient addition (P > 0.1, Tab. 2). Mean values of SLA and LAR were about 50% higher in the invasive than in the native species, with both differences being significant (P < 0.05, across two light and nutrient levels; Tab. 3). In addition, the invasive species showed a relatively higher LAR than the native species under LR and IR than under HR (significant species status x light availability interaction P < 0.05; Tab. 2). The root:shoot ratios (RSR, Fig. 4) varied widely from 0.2 under IR-HN to 0.9 under

HR-LN. In all species, RSR was higher under HR than IR (P < 0.001), and was lower in fertilized pots (P = 0.02), particularly at HR (light x nutrient, P = 0.05). Overall, the RSR of invasive species responded much more strongly to nutrient and light addition than 34

Light & Nutrient did that of native species (interaction effects of species status with light or nutrient level, P = 0.01 v. P = 0.09, Tab. 2 and 3, Fig. 4).

400

Invasive Native

SLA cm2 g-1

300

200

100

0

Low

Intermediate

High

Radiation

Fig. 3

Specific leaf area (SLA) of invasive (black bars) and native (grey bars) species (mean ± SE). Plants growing under three light levels with no fertilizer added. See text for further information.

0.8

Low nutrient

High nutrient

0.7 0.6

RSR

0.5 0.4 0.3 0.2 0.1 0

Intermediate

High

Intermediate

High

Radiation

Fig. 4

Root:shoot ratio (RSR) of invasive (black bars) and native (grey bars) tree species (mean ± SE). Data for plants growing under two light and two nutrient levels. See text for further information.

35

Chapter 1 Leaf nutrient contents In the low nutrient treatments, leaf nitrogen (N) and phosphorus (P) concentrations tended to decrease with increasing light (P < 0.001, Tab. 2, Fig. 5), particularly between IR and HR. Among the invasive species, mean N concentrations ranged from 24 mg g-1 under LR to 13 mg g-1 under HR, while the corresponding range for P was from 1.2 mg g-1 to 1.0 mg g-1; under all light levels N and P concentrations were 20-50% higher than for the native species (P < 0.06, Tab. 2). The difference in N content between species groups was most pronounced under LR, resulting in a c. 30% higher N:P ratio for invasive (21) than for native species (16); in contrast, the N:P ratios did not differ under IR and HR (16 and 13.5, respectively). Again, differences between single species were very pronounced, especially in the LR treatment (invasive species: 17 - 42 mg g-1 (N), 0.8 - 1.8 mg g-1 (P); native species: 9 - 26 mg g-1 (N), 0.9 - 1.1 mg g-1 (P)).

A

30

Intermediate radiation

High radiation

B

2.0

Intermediate radiation

High radiation

1.8 1.6

20

P mg g-1 g-1

N mg g-1 g-1

25

15 10

1.2 1.0 0.8 0.6 0.4

5

0.2 0

0

High

Low

High

High

Low

Low

High

Low

Nutrient

Nutrient

Fig. 5

1.4

Leaf nutrient contents of invasive (black bars) and native (grey bars) tree species (mean ± SE). Data for plants growing under two light and two nutrient levels. See text for further information. A Nitrogen (N). B Phosphorus (P).

Adding nutrients led to significant increases in the concentrations of both N and P in the leaves (P < 0.001, Tab. 3, Fig. 5), and the N:P ratio in these plants was also 15-18% higher than in the no nutrient treatment (P = 0.002). There was also a significant interaction with status group, as invasive species tended to profit more from adding nutrients (species status x nutrients, P ≤ 0.02); however, this species group effect occurred almost exclusively under HR, leading to a significant three-way interaction (status x light x nutrients; P < 0.001). Under HR conditions, the invasive species increased their leaf N and P contents in response to nutrient addition by c. 75% and 50%, respectively, while the corresponding increases in the native species were much lower (c. 25% and 10%, respectively; Fig. 5). 36

Light & Nutrient

Discussion The responses of seedlings to the light and nutrient treatments suggest that juvenile tree growth in Seychelles forests is strongly limited by both light and nutrients. Increasing either of these resources along the resource availability gradient typical in the forests (i.e. the levels used in the low light and low nutrient versus high nutrient and high light treatments) led to a strongly increased growth. In addition, there were changes in biomass allocation favouring the plant organs that acquire whichever resource was more limiting; thus, with an increase in light there was an increase in RSR and a decrease in LAR, while the reverse trends were found when nutrients were added. Although it is not possible from this experiment to determine which nutrient (or nutrients) is limiting, the finding that both leaf N and P contents increased after fertilisation with an N-P-K fertilizer suggest that both these nutrients (and possibly also K, compare Kueffer 2006) limit growth. Further support for this suggestion is provided by the leaf N:P values, which are in the range that would be expected with co-limitation (i.e. c. 15 Koerselman and Meuleman 1996, Güsewell and Koerselman 2002). Compared with their native counterparts, seedlings of invasive species tended to have higher growth rates (RGR), higher specific leaf areas (SLA) and higher leaf nutrient contents. They also exhibited greater plasticity in biomass and nutrient allocation (i.e. greater plasticity in SLA, LAR, RSR, leaf nutrient contents) in response to resource availability; this was reflected in an increased differentiation between the status groups under high resource conditions (Fig. 2B, 4, 5). While these patterns fit with current generalizations about what makes some plant species invasive (Daehler 2003, Niinemets et al. 2003, Richards et al. 2006, Richardson and Pysek 2006), the differences in mean values between the two species groups were generally small compared with the variation within groups.

Higher variability of growth characteristics among invasive species There were interesting patterns of variation in plant traits (SLA, leaf nutrient contents, Fig. 6) and growth rates (RGR, data not shown) within both groups of species. Two invasive species, Alstonia macrophylla, Tabebuia pallida, produced leaves with notably high SLA across all light and nutrient treatments (Fig. 6), and their relative growth rates (RGR) under high light availability were 25-50% higher than those of most of the other species. In contrast, SLA and leaf nutrient contents of the other four invasive species were either similar to those of the native species under all treatments (Psidium cattleianum, Syzygium jambos), or under all but the high light / high nutrient treatment (Cinnamomum verum, Sandoricum koetjape) (Fig. 6). 37

Chapter 1 Nutrient High

Low

Intermediate

SLA cm2 g-1

400 300 200

Radiation

100 0

300

High

SLA cm2 g-1

400

200 100 0 0

5

10

15

20

25

30

35

40

0

N mg g-1

Fig. 6

5

10

15

20

25

30

35

40

N mg g-1

Nitrogen leaf content (N) versus specific leaf area (SLA) for plants growing under two light and two nutrient levels. Invasive species are indicated with filled symbols: triangles for the two fast-growing (A. macrophylla & T. pallida) and diamonds for the four stress-tolerant invaders. The native species are shown in open symbols: squares for the two indigenous species (A. theiformis & C. bibracteatum) and circles for the four endemic species.

These results, together with data on spatial distributions of these species in the field (Fleischmann 1997, Kueffer and Vos 2004), indicate that the species invading closed forest on nutrient-poor soils in the Seychelles have traits associated with stress tolerance, while the more typical ‘fast-growing’ invaders are restricted to more nutrient-rich, disturbed sites, especially in the lowlands (with the notable exception of A. macrophylla, see below). This conclusion is supported by the results of an additional experimental series that we performed with the alien species Psidium guajava using the same setup. P. guajava is invasive in many subtropical and tropical countries but has only become naturalized in the Seychelles in a few coastal habitats. The two congeneric species P. guajava and P. cattleianum differed strongly in their growth performance; in particular, P. guajava had a higher SLA (data not shown) and RGR than P. cattleianum under high light, while under low light P. guajava had the same RGRDW as P. cattleianum but a lower RGRLA (Fig. 7). Thus, P. guajava has traits characteristic of many woody invasive species in the tropics but is not invasive in Seychelles, while P. cattleianum has more stress tolerant characteristics and is invasive.

38

Light & Nutrient Among the native species, within-group variation was smaller, and even the two non-endemic species, Aphloia theiformis and Canthium bibracteatum, did not differ from endemic species in their leaf traits (Fig. 6). It is interesting, however, that these species - which typically occur in disturbed environments - had higher RGR’s than the other native species. It has been hypothesised that one of the reasons why oceanic islands are especially prone to invasion is that the native plants tend to exhibit relatively low environmental specialisation and a correspondingly low competitiveness (see Denslow 2003). Our study indicates that the invasive flora may be better able to exploit high resource conditions than the native flora, but more studies are needed in which the growth characteristics of a wider array of island plants are screened in order to test this hypothesis.

A

30

B

Psidium cattleianum Psidium guajava

0.02

RGRLA cm2 cm-2 d-1

RGRDW mg g-1 d-1

25

20

15

10

0.015

0.01

0.005 5

0

Low

Intermediate

0

High

Radiation

Fig. 7

Low

Intermediate

High

Radiation

Relative growth rates of dry weight (RGRDW, A) and of leaf area (RGRLA, B) of the invasive Psidium cattleianum and the non-invasive alien Psidium guajava under LR (low radiation), IR (intermediate radiation) and HR (high radiation).

Fast-growing invasive species The two fast-growing invasive species, A. macrophylla and T. pallida, showed ecological characteristics to be expected of a successful plant invader (e.g. Baker 1974): they grew particularly well under high resource availabilities, and in the field they are found mainly (A. macrophylla) or exclusively (T. pallida) in highly disturbed environments. In our study, these species grew so fast that the pot size may have restricted growth towards the end of the experiment, especially under the high resource treatments; however, this effect would not have altered the overall ranking of these species. Surprisingly, A. macrophylla also survived in the low light treatment and was the fastest 39

Chapter 1 growing species under intermediate and low light. One explanation for its relatively high shade tolerance is a high phenotypic plasticity: A. macrophylla adapted RSR and SLA by a factor of three from low to high resource conditions, and consequently had one of the lowest RSR (0.3) and the highest SLA (550 cm2 g-1) under low light. By allocating more of its biomass to leaves it could greatly increase its capacity to absorb light in the shade. Furthermore, the leaf N content of A. macrophylla under low light (data not shown) was about twice that of any other species, suggesting that it has a high ability to take up nutrients from infertile soils (compare Kueffer 2006). However, despite this evidence for high phenotypic plasticity, seedlings of A. macrophylla are not found in the least disturbed forests in Seychelles, in contrast to many endemic species and stress tolerant invasive species such as Psidium cattleianum.

Stress-tolerant invasive species The four stress tolerant alien species in our experiment included the three alien trees C. verum, P. cattleianum and S. jambos that invade relatively undisturbed mountain forests in the Seychelles. Indeed, two of them (P. cattleianum, S. jambos) can regenerate so vigorously in closed forest that they sometimes prevent regeneration of native species (Kueffer and Vos 2004, Kueffer 2006). These two species are also problematic invaders across the whole tropics (e.g. Weber 2003). In contrast, the fourth species, S. koetjape, is restricted to the understorey of plantation forests where it forms dense sapling layers; its restricted distribution can probably be explained by the fact that its large seeds are not readily dispersed. Although stress-tolerant invasive species tended to respond to changes in resource availability more plastically than native species, these traits alone may not be sufficient to predict their invasiveness. All of the successful invaders exhibit other features that enables them to dominate the understory: these include vegetative reproduction (P. cattleianum), large seeds that produce large seedlings (S. jambos, S. koetjape) and abundant production of bird-dispersed seeds (C. verum) (Kueffer, 2007; Kueffer, 2006). And on a landscape scale, the fact that these trees grow under a wide range of resource conditions may also contribute to their success by enhancing propagule pressure. Some of the species invasive in closed forest, such as C. verum and P. cattleianum, are also common on the many granitic rocky outcrops known as inselbergs. These plants produce large seed crops that may subsequently be dispersed to the surrounding forests by birds (compare Kueffer and Vos 2004, Kueffer 2006).

40

Light & Nutrient

A finer picture of plant invasiveness The successful plant invaders of closed-canopy secondary forests on nutrient-poor soils in the Seychelles (and probably also in many other tropical areas) appear to be of two types: stress-tolerant invaders (e.g. P. cattleianum, S. jambos) and fast-growing invaders with particular adaptations to nutrient-poor soils (e.g. A. macrophylla). In contrast, the more typical, fast-growing alien species are restricted to the relatively nutrient-rich coastal plateau (e.g. Psidium guajava, Tabebuia pallida). Such diverging specialisations among invaders could also explain why no clear trends have been found in the ecological traits of invasive species in other nutrient-poor areas (e.g. Bellingham et al. 2004). This conclusion has important implications for invasion biology in general. It becomes increasingly clear that invasive species are not characterised by a common set of traits (Daehler 2003). Rather than being ‘super-weeds’ that out-compete native species under all conditions, it seems that particular invasive species are most competitive under a certain set of habitat conditions; if this is the case, then different invasive species may exhibit more extreme features than native species at opposite ends of the ecological spectrum (Crawley et al. 1996, Richardson and Pysek 2006). Rather than reflecting general features of all invasive species, therefore, the traits that have been associated with successful invaders through broad comparisons of invasive and native floras could merely reflect the fact that invasions in disturbed habitats are more common (Maskell, 2006) and have received more attention. However, as in the Seychelles, invasions also happen in undisturbed and stressed ecosystems (e.g. Stohlgren et al. 1999, Dietz and Edwards 2006, Martin and Marks 2006 and references therein), where they often have a negative impact on biodiversity and ecosystem functioning. Understanding why these invasions occur, therefore, is not only a matter of scientific interest (Dietz and Edwards 2006) but of practical importance for management.

41

Chapter 1

Acknowledgements We thank the Seychelles Ministry of Environment and Natural Resources for their support with the conducting of the experiment, and particularly the staff of the Sans Souci forestry station for their assistance with the common garden experiment. At the Institute of Integrative Biology (ETH Zurich), Sabine Guesewell provided statistical advice, and Rose Trachsler and Marilyn Gaschen assisted with the chemical analyses of leaf and soil samples. Funding was provided by a research grant from the Swiss Federal Institute of Technology (ETH Zurich).

Literature

Alpert, P., E. Bone, and C. Holzapfel. 2000. Invasiveness, invasibility and the role of environmental stress in the spread of non-native plants. Perspectives in Plant Ecology, Evolution and Systematics 3:52-66. Baker, H. G. 1974. The evolution of weeds. Annual Review of Ecology and Systematics 5:1-24.

Baruch, Z., and G. Goldstein. 1999. Leaf construction cost, nutrient concentration, and net CO2 assimilation of native and invasive species in Hawaii. Oecologia 121:183-192.

Baruch, Z., R. R. Pattison, and G. Goldstein. 2000. Responses to light and water availability of four invasive Melastomataceae in the Hawaiian islands. International Journal of Plant Sciences 161:107-118. Bellingham, P. J., R. P. Duncan, W. G. Lee, and R. P. Buxton. 2004. Seedling growth rate and survival do not predict invasiveness in naturalized woody plants in New Zealand. Oikos 106:308-316. Burke, A. 2003. Inselbergs in a changing world - global trends. Diversity and Distributions 9:375-383.

Crawley, M. J., P. H. Harvey, and A. Purvis. 1996. Comparative ecology of the native and alien flora of the British Isles. Philosophical Transactions of the Royal Society London B 351. Daehler, C. C. 2003. Performance comparisons of co-occurring native and alien invasive plants: Implications for conservation and restoration. Annual Review of Ecology and Systematics 34:183-211. Davis, M. A., J. P. Grime, and K. Thompson. 2000. Fluctuating resources in plant communities: A general theory of invasibility. Journal of Ecology 88:528-534. Denslow, J. S. 2003. Weeds in paradise: Thoughts on the invasibility of tropical islands. Annals of the Missouri Botanical Garden 90:119-127.

Dietz, H. 2005. A mountain invasions special issue. Perspectives in Plant Ecology, Evolution and Systematics 7:135-136. Dietz, H., and P. J. Edwards. 2006. Recognition that causal processes change during plant invasion helps explain conflicts in evidence. Ecology 87:1359-1367.

42

Light & Nutrient Dietz, H., and T. Steinlein. 1996. Determination of plant species cover by means of image analysis. Journal of Vegetation Science 7:131-136. Dukes, J. S., and H. A. Mooney. 1999. Does global change increase the success of biological invaders? Trends in Ecology and Evolution 14:135-139.

Ehrenfeld, J. G. 2003. Effects of exotic plant invasions on soil nutrient cycling processes. Ecosystems 6:503-523. Fine, P. V. A. 2002. The invasibility of tropical forests by exotic plants. Journal of Tropical Ecology 18:687-705.

Fleischmann, K. 1997. Invasion of alien woody plants on the islands of Mahé and Silhouette, Seychelles. Journal of Vegetation Science 8:5-12. Fleischmann, K. 1999. Relations between the invasive Cinnamomum verum and the endemic Phoenicophorium borsigianum on Mahé island, Seychelles. Applied Vegetation Science 2:3746. Friedmann, F. 1994. Flore des Seychelles. Orstom, Paris.

Grotkopp, E., M. Rejmanek, and T. L. Rost. 2002. Toward a causal explanation of plant invasiveness: Seedling growth and life-history strategies of 29 Pine (Pinus) species. American Naturalist 159:396-419. Güsewell, S., and W. Koerselman. 2002. Variation in nitrogen and phosphorus concentrations of wetland plants. Perspectives in Plant Ecology, Evolution and Systematics 5:37-61.

Huenneke, L. F., and P. M. Vitousek. 1990. Seedling and clonal recruitment of the invasive tree Psidium cattleianum: Implications for management of native Hawaiian forests. Biological Conservation 53:199-211. Koerselman, W., and A. F. M. Meuleman. 1996. The vegetation N:P ratio: A new tool to detect the nature of nutrient limitation. Journal of Applied Ecology 33:1441-1450.

Kolar, C. S., and T. S. Lodge. 2001. Progress in invasion biology: Predicting invaders. Trends in Ecology and Evolution 16:199-204.

Kueffer, C., E. Schumacher, K. Fleischmann, P. J. Edwards, and H. Dietz. 2007. Strong belowground competition shapes tree regeneration in invasive Cinnamomum verum forests. Journal of Ecology 95:273–282. Kueffer, C. 2006. Impacts of woody invasive species on tropical forests of the Seychelles. PhD Thesis, ETH Zurich, Zurich.

Kueffer, C., and P. Vos. 2004. Case Studies on the Status of Invasive Woody Plant Species in the Western Indian Ocean: 5. Seychelles. Forest Health & Biosecurity Working Papers FBS/45E Forestry Department, Food and Agriculture Organization of the United Nations, Rome, Italy. Martin, P. H., and P. L. Marks. 2006. Intact forests provide only weak resistance to a shadetolerant invasive Norway maple (Acer platanoides L.). Journal of Ecology 94:1070-1079.

Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well-being: Biodiversity Synthesis. World Resources Institute, Washington, DC.

Niinemets, U., F. Valladares, and R. Ceulemans. 2003. Leaf-level phenotypic variability and plasticity of invasive Rhododendron ponticum and non-invasive Ilex aquifolium co- occurring at two contrasting European sites. Plant Cell and Environment 26:941-956.

Pimentel, D., R. Zuniga, and D. Morrison. 2005. Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecological Economics 52:273-288.

43

Chapter 1 Rejmanek, M. 1996. A theory of seed plants invasiveness: The first sketch. Biological Conservation 78:171-181. Richards, A. J., O. Bossdorf, N. Z. Muth, J. Gurevitch, and M. Pigliucci. 2006. Jack of all trades, master of some? On the role of phenotypic plasticity in plant invasions. Ecology Letters 9:981-993. Richardson, D. M., and P. Pysek. 2006. Plant invasions: Merging the concepts of species invasiveness and community invasibility. Progress in Physical Geography 30:409-431. Schmitt, L., and J. N. Riviere. 2002. Comparative life-history traits of two Syzygium species (Myrtaceae): One invasive alien in La Réunion, the other native. Acta Botanica Gallica 149:457-466. Stoddart, D. R., editor. 1984. Biogeography and Ecology of the Seychelles Islands. DR W. Junk Publishers, The Hague, Boston, Lancaster.

Stohlgren, T. J., D. Binkley, G. W. Chong, K. M. A, L. D. Schell, K. A. Bull, Y. Otskul, G. Newman, M. Bashkin, and Y. Son. 1999. Exotic plant species invade hot spots of native plant diversity. Ecological Monographs 69:25-46.

Veneklaas, E. J., and L. Poorter. 1998. Growth and Carbon Partitioning of Tropical Tree Seedlings in Contrasting Light Environments. in H. Lambers, H. Poorter, and M. M. I. Van Vuuren, editors. Inherent Variation in Plant Growth: Physiological Mechanisms and Ecological Consequences. Backhuys, Leiden. Vitousek, P. M. 2004. Nutrient Cycling and Limitation. Hawai’i as a Model System. Princeton University Press, Princeton. Weber, E. 2003. Invasive Plant Species of the World. A Reference Guide to Environmental Weeds. CABI Publishing, Oxon, UK & Cambridge, USA. Williamson, M. 1996. Biological invasions. Chapman & Hall, London, New York ,Tokyo.

44

Chapter 2 Influence of drought and shade on seedling growth of native and invasive trees in the Seychelles

45

Chapter 2

Abstract 1

Data on the role of water availability in reducing the invasibility of habitats to plants are scarce. However, such information may be important for predicting plant invasions, particularly in regions where water availability varies widely over short distances due to high variation in rainfall or topography.

2 We studied the growth of seedlings of native and invasive tree species from secondary tropical forests on Mahé (Seychelles). Our hypothesis was that the relative tolerance of native and invasive tree species to drought varies according to light conditions, with native species performing better at low water and low light availability and invasive species performing better at higher resource availabilities. 3 To test this hypothesis a common garden (pot) experiment was conducted comparing the growth characteristics of seedlings of five native and five invasive tree species under two levels of light (7% and 60% of ambient light) and two levels of water availability (ambient watered and drought). 4 Differences in the response of native and invasive species to varying light and water levels were mostly small and not statistically significant. In both groups of species, mean relative growth rates were reduced only slightly by intermittent drought that led to wilting of leaves. However, native species performed relatively better than invasive species in the treatment with both low light and low water availability. 5 Our results suggest that some fast-growing invasive species cope with moderate stress through high phenotypic plasticity in growth allocation patterns. As such plastic responses may not be possible under conditions of severe or multiple stresses, the invasive species may be disadvantaged in such situations relative to slower growing but more stress-tolerant native species. 6

We conclude that moderately water-stressed habitats may be more invasible than had been previously thought. However the influence of water availability on invasion is complex, partly because of the spatio-temporal variability in water availability and partly because of the wide variation among alien species in tolerance to drought.

Keywords Biomass allocation, common garden experiment, drought stress, invasiveness, light availability, oceanic islands, RGR, Seychelles, tropical tree seedlings

46

Drought Stress

Introduction Why some habitats are less invaded than others has long been a core question in invasion biology, and there are still no fully satisfying answers. Both biotic (Levine et al. 2004) and abiotic (Alpert et al. 2000, Davis et al. 2000) factors are usually assumed to explain differences in invasibility among habitats. While the exact role of biotic factors is still debated (Levine et al. 2004), much evidence has accumulated suggesting that many invasive species profit from increased levels of abiotic resources, particularly light and nutrients (e.g. Davis et al. 2000, Lake and Leishman 2004, Maskell et al. 2006, and chapter 1). For instance, nutrient enrichment strongly promotes plant invasions in low fertility bushland in south-eastern Australia (Lake and Leishman 2004), and increased light availability after anthropogenic disturbance seems to be the main factor enabling plant invasions into tropical forests (Fine 2002). Based on such empirical results, increased resource availability is now seen as a core factor favouring plant invasions (Alpert et al. 2000, Davis et al. 2000, Richardson and Pysek 2006). It has been proposed that pulses of resources exceeding the immediate needs of the established community increase the vulnerability of habitats to invasion (Davis et al. 2000), while very low levels of resources make habitats resistant to invasion (Alpert et al. 2000). Both Davis et al. (2000) and Alpert et al. (2000) have suggested that not only low levels of light and nutrients but also shortage of water are likely to reduce invasibility; conversely, if there is an increase in rainfall in dry habitats, perhaps as a result of climatic change, the risk of invasion may be enhanced (Dukes and Mooney 1999). However, despite its potential importance, there have been fewer studies on water as an ecological factor in plant invasions than there have been on light and nutrients. These few studies do indeed suggest that low water availability is associated with low vulnerability of a habitat to invasion (Alpert et al. 2000, Stohlgren et al. 2001 and references therein), and that water pulses can facilitate invasions (Milchunas and Lauenroth 1995). For example, in growth experiments comparing the performance of native and invasive species under varying environmental conditions, native species appeared to suffer less from water stress than invasive species (Daehler 2003). In tropical forests, patterns in the distribution of tree species often reflect a gradient in water availability (e.g. Bongers et al. 1999). The seedlings of some pioneer trees are very sensitive to drought, and even short dry spells can cause high mortality (Engelbrecht et al. 2006). However, seedlings growing in shady conditions are likely to be most vulnerable to drought, because of their low growth rates and thin leaves (e.g. Veenendaal et al. 1996, Turner 2001). Shifts in the frequency of dry periods due to 47

Chapter 2 climate change could have a considerable impact on the species composition of tropical forests (Condit 1998, Whitmore 1998), and climate change may possibly also reduce the resistance of tropical forests to plant invasions. It is therefore important to understand better how water availability influences plant invasions into tropical forests. The mountainous island of Mahé (Republic of Seychelles, Indian Ocean) offers a suitable system for studying the influence of water availability on plant invasions into tropical forests. This is because it offers a diversity of habitats - ranging from relatively dry, sun-exposed inselberg habitats (sparse shrub vegetation on granite outcrops) to montane mist forests, all of which have been invaded to some degree by alien plants. Moreover, despite its high environmental heterogeneity, the total number of woody species on the Seychelles is rather small, with around 20 abundant native species and even fewer invasive species. This situation makes it possible to compare the growth characteristics of the same invasive and native woody species along broad gradients of water and light availability.

The main objective of this study was to investigate whether native or invasive woody plant species are better able to withstand environmental stresses on the oceanic islands of the Seychelles. The flora of the Seychelles has experienced a long history of major climatic changes (Stoddart 1984, Briggs 2003), and it has been suggested that the native species are particularly well adapted to a broad range of conditions from dry to very humid (Stoddart 1984, Schumacher et al. 2003). We therefore expected native trees to be more drought tolerant than invasive species, especially in the shade. To test this hypothesis, we compared the growth of seedlings of five native and five invasive species in a common garden experiment with two light and two water conditions (watered and drought stressed). The data were analysed to investigate possible differences between the two groups of species in relative growth rates, patterns of biomass and nutrient allocation, and phenotypic plasticity.

48

Drought Stress

Methods General study area The study was carried out on the island of Mahé in the Republic of Seychelles (4° S, 55° E, total area 154 km2, 0-900 m asl). The inner islands of the Seychelles are built of granitic rock that is 550 to 650 Mio years old and has never been covered by the ocean, so that both rocks and soils have been continuously weathered for over 500 Mio years (Stoddart 1984). Thus, the vegetation on the inner islands of the Seychelles grows on soils (typically ferrasols with a pH of about 4.5) that are very poor in nutrients, particularly phosphorus, making them amongst the most infertile of all tropical forest soils (cf. Kueffer 2006). All vegetation in the Seychelles has been heavily affected by human activities. No

native forest remains in the lowlands, but in some areas a dense secondary forest has developed, much of it dominated by the invasive tree Cinnamomum verum. In contrast, the montane mist forests retain a higher proportion of native species, though even these are heavily invaded. The very nutrient poor, sun-exposed granitic rock outcrops (inselbergs) exhibits the lowest penetration by alien woody plant species (Fleischmann et al. 1996, Fleischmann 1997). The inland habitats have a humid tropical climate with a mean annual rainfall ranging from 1600 to 3500 mm depending on altitude (Stoddart 1984, Cazes-Duvat and Robert 2001). Although there is no pronounced seasonality in rainfall, the period between June and September is generally drier than that between November and February. At low to intermediate altitudes ( 100 leaves per species. The sample leaves were placed beneath a glass plate and photographed with a digital camera (Nikon Coolpix 995, resolution at 2048 x 1536 pixels); these images were then used to determine leaf length and breadth using Adobe IllustratorTM 10, and area using Adobe PhotoshopTM 7.0 (cf. Dietz and Steinlein 1996). At the end of the experiment, all plants were harvested, separated into leaves, stems plus petioles, and roots, and oven-dried at 80°C for 48 hours. Subsamples of leaf material were digested at 420°C with 98% H2SO4 and Merck Kjeltabs, and total nitrogen and phosphorus concentrations of leaves were determined colorimetrically using a flow injection analyzer (FIA, TECATOR, Höganäs, Sweden). The raw data were used to calculate the following growth parameters: RGRDW Relative growth rate of total dry weight

ln (dry weight at end)-ln(dry weight at start) duration of experiment ln (leaf area at end)-ln(leaf area at start) duration of experiment

RGRLA

Relative growth rate of leaf area

SLA

Specific leaf area

leaf area dry leaf biomass

LAR

Leaf area ratio

leaf area dry plant biomass

RSR

Root:shoot ratio

dry root biomass dry shoot biomass

Stastistical analysis We used general linear models with light level, water level, species status (native or invasive) and their interactions as fixed factors, and species identity (nested in species status) as a random factor. The leaf area of each plant at the start of the experiment

52

Drought Stress was included as a covariable to account for differences in initial plant size. Dependent variables were plant growth (relative growth rate of total dry weight and leaf area), allocation (SLA, LAR, RSR) and leaf nutrient contents (nitrogen N and phosphorus P and N:P ratio). To remove heteroscedasticity, SLA, LAR and RSR were log-transformed, and nitrogen and phosphorus leaf contents were square-root transformed. All statistical analyses were performed with JMP V 6.0 (SAS Institute Inc., 2005).

Results Seedling mortality during the course of the experiment was generally low, with an

average 8% of invasive plants and 12% of native plants dying, mainly in the treatment with low light and low water availability. The mortality of two invasive species, Alstonia macrophylla and Tabebuia pallida, was relatively high (≥ 20%) under low light, while most individuals of the other three species survived in all treatments. Among the native species, the relatively few deaths occurred mainly in the drought treatments (15-30% mortality). An exception was the endemic Memecylon eleagni, with between 15% and 70% of seedlings dying in the various treatments.

Responses of the species to variation in light and water availability Relative growth rate Variation in relative ��������������������������������������������� growth rate of total dry weight (RGRDW) ��������������������� among species within -1 -1 groups was large, ranging from -0.9 to c. 12 mg ���� g d in both species groups under low light, and from 9 to 24 ���� mg g-1 d-1 for invasive plants and 3 ���� mg g-1 d-1 to 20 ���� mg g-1 d-1 for native plants under high light. �������������������������������������������������� Both native and invasive species grew fastest (RGRDW and relative growth rate of total leaf area (RGRLA)) under high light (P < 0.001, Fig. 1, Tab. 2), but the strength of the response to light tended to be greater among the invasive species (species group x light, P ≤ 0.09). Compared to ambient water conditions (W), drought (D) reduced mean growth rates (both RGRDW and RGRLA) by 4% under low light and 28 % under high light (P ≤ 0.004, Fig. 1, Tab. 2 & 3). This relatively greater reduction in growth under high light was reflected in a significant interaction between light and water in the ANOVA (P < 0.01, Tab. 2). 53

Chapter 2

Fig. 1

Relative growth rate of total dry weight (RGRDW) of invasive (black bars) and native (grey bars) species (mean ± SE) under two light levels and two water treatments (drought stressed vs. watered). See text for further information. 20

Low radiation

High radiation a

b RGRDW mg g-1 d-1

15

abc de

cd

10

e

Invasive Native

de de

5

0

Drought stressed

Watered

Drought stressed

Watered

Table 2 Results of ANOVA across two levels of light and watering. Indicated are the F-ratios and significance levels (***: P < 0.001, **: P < 0.01: *: P < 0.05) of main effects and interactions for the following parameters: relative growth rates of total dry weight (RGRDW) and leaf area (RGRLA), root dry weight (RootDW) specific leaf area (SLA), leaf area ratio (LAR), root:shoot ratio (RSR), leaf nitrogen (N) and phosphorus (P) contents and N:P ratio in leaves. Significant values are given in bold. RGRDW

RGRLA

RootDW

SLA

LAR

RSR

N

P

N:P

1.8

3.4

17.2 **

2.2

1.2

3.1

2.9

2.4

0.6

216.3 ***

24.7 ***

143.0 ***

244.2 ***

110.9 ***

32.4 ***

18.9 ***

0.1

9.5 **

3.6

1.1

22.5 **

3.9

1.8

2.3

2.8

2.1

0.0

24.5 **

26.1 **

9.6 *

14.9 *

7.3 *

3.6

5.3

7.5 *

21.7 **

SxW

2.3

1.8

2.6

2.1

0.5

0.0

0.3

0.3

2.1

LxW

7.8 **

7.1 **

14.1 ***

0.0

0.5

0.1

2.1

5.7 *

14.4 ***

SxLxW

7.4 **

8.1 **

0.9

0.3

0.9

2.7

7.1 **

7.8 **

0.5

17.2 ***

21.8***

140.9 ***

1.1

5.2 *

3.2

1.5

0.3

0.1

Species group† (S) Light (L) SxL Water (W)

Initial leaf area †

native vs. invasive

54

Drought Stress The mean RGRDW of invasive species was 40 to 80% higher than that of the native species under all treatments except LR/D, but these differences between species groups were not significant (overall species group effect P = 0.2, Fig. 1, Tab. 2). However, while drought reduced the mean growth rate of invasive species by c. 20% irrespective of light availability, it only reduced the growth of native species under high light, and growth rates under low light were actually 40% higher. Thus, there was a significant three-way interaction between water, light and species group (P < 0.008, Fig. 1). Individual native species, however, varied widely in their response to drought stress at low light: three species (Paragenipa wrightii, Canthium bibracteatum and Aphloia theiformis) increased their RGRDW by 33 to 115%, whereas Erythroxylum sechellarum and Memecylon elagnii reduced their growth by c. 30%. The ���������������������������������������� relative growth rates of height (RGRH) and total leaf area (RGRLA) showed very similar patterns and were mostly highly correlated with RGRDW (r > 0.6, P < 0.01; data not shown). Biomass allocation Specific leaf area. Mean SLA ranged from 75 to 245 cm ��2 g-1 under HR, and from 80 to 550 �� cm2 g-1 under LR. Overall, ������������������������������������������������������������� means of SLA did not differ between the two species groups (P = 0.2)���������������������������������������������������������������������� , though mean values of SLA were about 60% higher in invasive than in native plants under HR, and 20 % higher under LR. Thus, invasive plants did exhibit significantly greater phenotypic plasticity in SLA in response to light (species group x light interaction, P = 0.09). While in both species groups SLA was lower under high light (P < 0.001, Tab. 2 & 3) the difference was 45% in invasive plants compared to 25% in native plants. SLA was also reduced by the drought treatment compared to the watered controls, being 10 % lower in invasive plants and 5 % lower in native plants (P ��� < 0.05)��������������������������������������������������������������������� ; once again the difference due to status group was not significant��. L�������������� eaf area ratio. LAR was positively correlated with SLA (r > 0.4, P < 0.001, Tab. 2 & 3) and showed similar trends in the various treatments. The results are therefore not presented in detail. Root dry weight and root:shoot ratio. Under LR, most seedlings showed only a small increase in root dry weight (Fig. 2). However, under HR the roots of invasive species grew more strongly, and by the end of the experiment they had a much higher root dry weight than native species. Thus, there was a significant interaction between species group and light (P = 0.002; Tab. 2, Fig. 2).

55

Chapter 2

Table 3 Means of response variables per species group (native vs. invasive). with standard errors in parentheses. See table 2 for acronyms. Low radiation Drought

Watered

Drought

Watered

RGRLA (cm2 cm-2 d-1)

invasive

0.006 (0.002)

0.008 (0.005)

0.009 (0.003)

0.012 (0.003)

native

0.005 (0.002)

0.003 (0.002)

0.003 (0.003)

0.008 (0.003)

SLA (cm2 g-1)

invasive

272 (63)

309 (80)

148 (19)

166 (26)

native

168 (25)

174 (28)

126 (16)

133 (15)

LAR (cm2 g-1)

invasive

96 (21)

117 (42)

53 (8)

61 (10)

native

73 (11)

78 (15)

49 (6)

56 (10)

invasive

0.52 (0.06)

0.46 (0.06)

0.65 (0.05)

0.61 (0.02)

native

0.40 (0.05)

0.39 (0.03)

0.49 (0.07)

0.45 (0.05)

invasive

15.5 (1.4)

14.4 (0.9)

15.8 (0.7)

11.2 (0.4)

native

13.2 (3.3)

13.3 (2.8)

14.1 (1.3)

9.9 (1.5)

RSR N:P

Fig. 2

High radiation

Root dry weight (RootDW) of invasive (black bars) and native (grey bars) species (mean ± SE) at the start and end of the experiment. 0.7

Start

End Low radiation

High radiation Invasive Native

0.6

Root dry weight g

0.5

0.4

0.3

0.2

0.1

0 Watered

Drought stressed

Watered

Drought stressed

Watered

The mean root:shoot ratio (RSR) of native species was lower than that of invasive species, but the difference was not significant (P = 0.1; Tab. 2). Both species groups showed similar trends in response to the treatments; the mean values for native species ranged from 0.4 in treatment LR/W to 0.5 in HR/D, and the corresponding means for the invasive plants were 0.45 and 0.65, respectively (Tab. 3). In both groups, RSR was 56

Drought Stress significantly higher at higher light availability (P < 0.001), but the effects of the water treatment were not significant (P = 0.1; Tab. 2). In both groups, variation in RSR among species was large (0.25 to 0.6 under LR, and 0.3 to 0.8 under HR, respectively). Leaf nutrient contents Leaf nitrogen (N) content ranged from 11.9 (natives under HR) to 19.3 mg g-1 (invasives under LR) and phosphorus (P) content ranged from 0.9 to 1.3 mg g-1. Mean N concentrations were lower under HR than LR (P < 0.001, Tab. 2, Fig 3A), but P concentrations were unaffected by light level (P = 0.8, Tab. 2, Fig. 3B). N contents were generally slightly lower and P contents slightly higher in watered plants (P ≤ 0.05), and there was a general trend for slightly higher nutrient contents in the invasive than in the native species (P ≤ 0.2). Differences in N concentrations among individual species were greater in treatments with low light and drought than in those with ambient conditions, and there was a three-way interaction of status x light x water (P < 0.01). For example, the N content of the invasive A. macrophylla in the LR/D treatment was almost ten times higher than that of the native P. wrightii. For P content, there was also a three-way interaction (status x light x water; P < 0.01), but the pattern was more complicated: in response to water addition, the invasive species had a higher leaf P content under both LR and HR, whereas leaves of the native species had a lower P content under LR but a higher P content under HR (Fig. 3B, Tab. 3). The mean N:P ratios in the leaves mostly varied between 13 and 15 and did not differ between species groups (P = 0.4). However, in the HR/W treatment, the N:P ratios were considerably lower in both species groups, being 10 for the natives and 11 for the invasives (light x water P < 0.001). Fig. 3

A

24

Leaf nutrient contents of invasive (black bars) and native (grey bars) species (mean ± SE) under two light levels and two water treatments (drought stressed vs. watered). See text for further information. A Nitrogen (N). B Phosphorus (P). Low radiation

High radiation

B

1.6

Low radiation

High radiation

1.4

20

1.2

P mg g-1

N mg g-1

16 12 8

1 0.8 0.6 0.4

4

0.2

0

0

Drought stressed

Watered

Drought stressed

Watered

Drought stressed

57

Watered

Drought stressed

Watered

Chapter 2

Discussion In general, species responded more strongly to the light treatments than to the water treatments. While there was considerable variation in growth responses to the light and water treatments in both groups of species, the differences in the mean performance of native and invasive species were mostly small. However, a closer analysis reveals important differences in the response of the two groups to specific treatment combinations.

Effects of water stress at varying light availability on growth performance of native and invasive juvenile trees in Seychelles As expected, both native and invasive tree seedlings showed lower relative growth rates (RGR) under drought conditions. And both groups also showed similar morphological responses to drought, with plants having thicker leaves and tending to produce lower leaf area (LAR) and higher root:shoot ratios (RSR) (compare ������������������������� e.g. Burslem et al. 1996, Baruch et al. 2000, Poorter and Hayashida-Oliver 2000)������������������� . However, most of these differences were rather small. Growth rates, for instance, were reduced by only c. 5-20% in response to the water stress treatment, while increasing light availability from levels typical for the understorey to light levels found under gap conditions led to a doubling of growth rates. This may seem surprising when we consider that the seedlings were exposed to periodic dry spells of a duration that induced leaf wilting, and indicates that the seedlings of most species were well able to recover from short episodes of drought.���������������������������������������������������������������� ��������������������������������������������������������������� In fact, the drought treatment in our experiment only produced stressed conditions intermittently, while the low light treatment represented a continuous resource reduction. Hence, in places where dry spells are mostly short, as in the Seychelles where periods > 10 days with no rain recur on average every second year ������������������������������������������������������������������������������������ (Meteo Seychelles), occasional shortages of water are likely to have less impact on tree regeneration than persistently shady conditions. We expected native tree species to be more tolerant of water stress because plant species on isolated islands cannot migrate in response to climate change; thus, the survivors must have persisted in situ through major climate fluctuations such as those that are known to have occurred in the Seychelles (Briggs 2003). Additionally, microclimatic variations on Mahé from below 1000 to above 3500 mm rainfall a year on a scale of a few square kilometres require a broad environmental tolerance from the widely distributed native species. In contrast, the invasive species included in this study and the invasive 58

Drought Stress woody flora of Seychelles in general consist of tropical species with a preference for moist to wet habitats (Kueffer and Vos 2004), suggesting a lower tolerance of drought conditions. Contrary to our hypothesis, the growth rates of invasive species were not more affected by water stress than those of native species. However, the data on both relative growth rate and leaf phosphorus contents reveal more complex interactions between species group and the supplies of water and light. Under low light, invasive species reduced RGR more than native species in response to the drought treatment. The higher vulnerability of invasive than native species to drought stress under low light availability might be related to differences between the two species groups in their strategies for coping with such stress. Invasive species may actually have a lower physiological ability to tolerate resource shortage, as predicted by our hypothesis, but they compensate for this to some degree through their higher morphological plasticity;

this allows them to increase allocation to the organs responsible for the uptake of the limiting resources, i.e. a high specific leaf area (SLA) under low light and a high RSR in drought conditions. As a result, for both native and invasive species the relative difference between plants grown under ample resources and those experiencing one resource in short supply were rather similar. However, when there are deficiencies of both light and water, morphological plasticity may not be a viable strategy because the optimal allocation patterns conflict with each other (compare e.g. Veenendaal et al. 1996, Niinemets and Valladares 2006 and references therein). If this interpretation is correct, it is easy to understand why the particularly fast growing and plastic invader Alstonia macrophylla was the species that suffered most from the combination of drought and low light. In contrast, among the native species, growth and leaf phosphorus content tended to be higher under this treatment than when just one stress was imposed, suggesting that the native species did not suffer from the same conflict. Other studies also indicate that phenotypic plasticity is one mechanism by which invasive species cope with reduced resource availabilities (compare Richards et al. 2006 and chapter 1). For instance, in dry subtropical forests of Hawaii the invasive tree Schinus terebinthifolius showed a higher plasticity than native species in many different traits in response to wet versus dry seasons (Stratton and Goldstein 2001), and in an alpine ecosystem an alien species of Taraxacum was less drought tolerant but more plastic in the water use efficiency than a native species (Brock and Galen 2005). In contrast, many native species in Seychelles possess thick, sclerophyllous leaves (Friedmann 1994), and their phenotypic plasticity in response to increased levels of light and nutrients is low (see chapter 1). In their combination of slow growth rates, 59

Chapter 2 even under optimal conditions, and high tolerance of resource deficiencies, these species exhibit the typical characteristics of the stress tolerance strategy (Grime 2001). However, the dichotomy between plastic and fast-growing species and stress-tolerant species may not be a general distinction between native and invasive species, but rather both types may occur within both species groups (compare chapter 1). In fact, the range of soil moisture contents at leaf wilting was similar in both species groups (see Tab. 1), and the species with the lowest wilting point (6% soil moisture) was the invasive Psidium cattleianum, which also proved to be very shade-tolerant (compare chapter 1). Furthermore, this species is very invasive in both shady, montane mist forests and on the very exposed, dry inselbergs (Fleischmann 1997, Kueffer and Vos 2004). It might be argued that the reason that invasive species were able to survive water stress was because they produced more roots than native species in the pot experiments

(see Fig. 2). However, in our experimental design we reduced soil moisture content separately for each species to a value where the particular species showed signs of stress. In this way, we excluded the possibility that plants avoided water shortage by producing a larger root mass. Overall, we conclude that high phenotypic plasticity allows invasive species to compete successfully with more stress tolerant native species in areas where resources are in moderately short supply; however, this strategy does not enable them to grow under conditions where several resources are strongly limiting.

The relevance of dry spells for regeneration dynamics of native versus invasive tree species in Seychelles Water shortage is unlikely to be an important factor limiting regeneration in the tropical forests of the Seychelles. Mortality rates of the juveniles in the pot experiments were mostly low even at maximum drought stress, when soil moisture contents were between 6 and 12%. In the sandy loam laterite soils used in the experiment (c. 70% sand, 15% silt, 15% clay, unpublished data), these water contents correspond to a water potentials of well below -1.5 MPa (derived from soil-moisture characteristic curves of laterite and organic soils collected in secondary forests in Seychelles, unpublished data), which is cited as the permanent wilting point of many plants (e.g. Lambers et al. 1998). Further, the average intervals between watering in the drought treatment - 10 days in HR and 15 days in LR - were also longer than typical dry spells on Mahé;

60

Drought Stress indeed, since records began over 30 years ago, there have only been 18 dry periods (i.e. < 0.1 mm per day) of ≥ 10 days, and 6 of ≥ 15 days (Meteo Seychelles; records are made at sea level at Mahé airport). However, in evaluating the relative importance of water as an ecological factor for tree regeneration in the Seychelles, two important factors must be considered. First, water availability varies widely according to substrate conditions and rainfall. On lower islands such as Praslin (the second largest island) where rainfall is lower and fires are more frequent, large areas of eroded laterite soils are now covered in sparse secondary scrub; this vegetation is more prone to drought than the intermediate and mountain forests on the more mountainous islands of Mahé and Silhouette (the third largest island). But even on these islands, trees regenerating on inselbergs grow in very shallow soils and are exposed to high levels of irradiance and sometimes also to salt spray. These conditions contrast strongly to those of the moist understorey of closed canopy forests. Within the mid-altitude secondary forests of the Seychelles, variation on a micro-scale (i.e. tenth of metres) from organic matter rich soils to sandy clay loam (c. 50% sand, 25% silt, 25% clay) and sandy soils (> 95% sand) is common (Kueffer 2006, and unpublished data), and significantly affects the water holding capacity of the soil. On a landscape scale, the spatial distribution of invasive species may be related to these patterns of varying water availability. In particular, the fact that the driest habitats in the Seychelles, the inselbergs, are also the least invaded (Fleischmann 1997), suggests that water stress may be a factor reducing plant invasions in some areas.

Second, the frequency and intensity of drought events are important. While juveniles of invasive tree species may suffer more than their native counterparts from moderate droughts if growing under low light availability, more extreme drought events that occur every c. 5-10 years (Meteo Seychelles, see above) could be more detrimental for slow growing native species (cf. chapter 1). The survival of seedlings during severe droughts depends mainly on the size of the root system (e.g. Veenendaal et al. 1996, Turner 2001), and the roots of invasive species tend to grow faster than those of native species. For instance, seedlings of the invasive C. verum and S. jambos produced root dry weights of c. 1-2 g within 6 months in forest gaps, while the root dry weight of the native species was on average only c. 0.25 g over the same period (see chapter 3). And because their root system develops more slowly, seedlings of native species could be vulnerable to drought over a longer period than those of invasive species (cf. Fig. 3).

61

Chapter 2

Conclusions Our results suggest that in a diverse tropical system like the islands of the Seychelles, the influence of water availability on the invasion potential of introduced trees is complex. In general, invasive species appear to be better able to cope with moderate stress than had been previously thought; and as a result, native species may only have a relative advantage over invasive species in conditions of severe resource shortages that cannot be compensated for by higher growth plasticity. However, more research is needed to establish whether this finding is generally true. Even in the Seychelles, high stress tolerance among young invasive trees seems to be a trait of particular species rather than a general tendency. Because of this, and because there may be strong spatiotemporal variation in the occurrence of severe dry periods, the influence of drought upon the regeneration success of invasive and native trees in tropical forest may be hard to predict.

Acknowledgements We thank the Seychelles Ministry of Environment and Natural Resources for their support with the conducting of the experiment, and particularly the staff of the Sans Souci forestry station for their assistance with the common garden experiment; and Sabine Güsewell for providing statistical advice. Funding was provided by a research grant from the Swiss Federal Institute of Technology (ETH Zurich).

62

Drought Stress

Literature

Alpert, P., E. Bone, and C. Holzapfel. 2000. Invasiveness, invasibility and the role of environmental stress in the spread of non-native plants. Perspectives in Plant Ecology, Evolution and Systematics 3:52-66. Baruch, Z., R. R. Pattison, and G. Goldstein. 2000. Responses to light and water availability of four invasive Melastomataceae in the Hawaiian islands. International Journal of Plant Sciences 161:107-118. Bongers, F., L. Poorter, R. S. A. R. Van Rompaey, and M. P. E. Parren. 1999. Distribution of twelve moist forest canopy tree species in Liberia and Côte d’Ivoire: Response curves to a climatic gradient. Journal of Vegetation Science 10:371-382. Briggs, J. C. 2003. The biogeographic and tectonic history of India. Journal of Biogeography 30:381-388. Brock, M. J., and C. Galen. 2005. Drought tolerance in the Alpine Dandelion, Taraxacum ceratophorum (Asteraceae), its exotic congener T. officinale, and interspecific hybrids under natural and experimental conditions. American Journal of Botany 92:1311-1321. Burslem, D. F. R. P., P. J. Grubb, and I. M. Turner. 1996. Responses to simulated drought and elevated nutrient supply among shade-tolerant tree seedlings of lowland tropical forest in Singapore. Biotropica 28:636-648. Cazes-Duvat, V., and R. Robert. 2001. Atlas de l’environnement côtier des îles granitiques de l’archipel des Seychelles. Université de La Réunion & CIRAD-Emvt, St. Denis, La Réunion & Montpellier, France. Condit, R. 1998. Ecological implications of changes in drought patterns: Shift in forest composition in Panama. Climatic Change 39:413-427. Daehler, C. C. 2003. Performance comparisons of co-occurring native and alien invasive plants: Implications for conservation and restoration. Annual Review of Ecology and Systematics 34:183-211. Davis, M. A., J. P. Grime, and K. Thompson. 2000. Fluctuating resources in plant communities: A general theory of invasibility. Journal of Ecology 88:528-534. Dietz, H., and T. Steinlein. 1996. Determination of plant species cover by means of image analysis. Journal of Vegetation Science 7:131-136. Dukes, J. S., and H. A. Mooney. 1999. Does global change increase the success of biological invaders? Trends in Ecology and Evolution 14:135-139. Engelbrecht, B. M., J. W. Dalling, R. H. Pearson, R. L. Wolf, D. A. Galvez, T. Koehler, M. T. Tyree, and T. A. Kursar. 2006. Short dry spells in the wet season increase mortality of tropical pioneer seedlings. Oecologia 148:258-269. Fine, P. V. A. 2002. The invasibility of tropical forests by exotic plants. Journal of Tropical Ecology 18:687-705. Fleischmann, K. 1997. Invasion of alien woody plants on the islands of Mahé and Silhouette, Seychelles. Journal of Vegetation Science 8:5-12. Fleischmann, K., S. Porembski, N. Biedinger, and W. Barthlott. 1996. Inselbergs in the sea: Vegetation of granite outcrops on the islands of Mahe, Praslin and Silhouette (Seychelles). Bulletin of the Geobotanical Institute ETH 62:61-74. Friedmann, F. 1994. Flore des Seychelles. Orstom, Paris. Grime, J. P. 2001. Plant Strategies, Vegetation Processes, and Ecosystem Properties, 2nd edition. John Wiley& Sons, Chichester, New York, Toronto.

63

Chapter 2 Kueffer, C. 2006. Impacts of woody invasive species on tropical forests of the Seychelles. PhD Thesis ETH Zurich, Zurich. Kueffer, C., and P. Vos. 2004. Case Studies on the Status of Invasive Woody Plant Species in the Western Indian Ocean: 5. Seychelles. Forest Health & Biosecurity Working Papers FBS/45E Forestry Department, Food and Agriculture Organization of the United Nations, Rome, Italy. Lake, J. C., and M. R. Leishman. 2004. Invasion success of exotic plants in natural ecosystems: The role of disturbance, plant attributes and freedom from herbivores. Biological Conservation 117:215-226. Lambers, H., F. S. Chapin III, and T. L. Pons. 1998. Plant Physiological Ecology. Springer, New York, Berlin, Tokyo. Levine, J. M., P. B. Adler, and S. G. Yelenik. 2004. A meta-analysis of biotic resistance to exotic plant invasions. Ecology Letters 7:975-989. Maskell, L. C., L. G. Firbank, K. Thompson, J. M. Bullock, and S. M. Smart. 2006. Interactions between non-native plant species and the floristic composition of common habitats. Journal of Ecology 94:1052-1060. Milchunas, D. G., and W. K. Lauenroth. 1995. Inertia in plant community structure: State changes after cessation of nutrient enrichment stress. Ecological Applications 5:452-458. Niinemets, U., and F. Valladares. 2006. Tolerance to shade, drought, and waterlogging of temperate northern hemisphere trees and shrubs. Ecological Monographs 76:521-547. Poorter, L., and Y. Hayashida-Oliver. 2000. Effects of seasonal drought on gap and understorey seedlings in a Bolivian moist forest. Journal of Tropical Ecology 16:481-498. Richards, A. J., O. Bossdorf, N. Z. Muth, J. Gurevitch, and M. Pigliucci. 2006. Jack of all trades, master of some? On the role of phenotypic plasticity in plant invasions. Ecology Letters 9:981-993. Richardson, D. M., and P. Pysek. 2006. Plant invasions: Merging the concepts of species invasiveness and community invasibility. Progress in Physical Geography 30:409-431. Schumacher, E., H. Dietz, K. Fleischmann, C. Kueffer, and P. J. Edwards. 2003. Invasion of woody plants into the Seychelles tropical forests: Species traits in the establishment phase. Bulletin of the Geobotanical Institute ETH 69:77-86. Stoddart, D. R., editor. 1984. Biogeography and Ecology of the Seychelles Islands. DR W. Junk Publishers, The Hague, Boston, Lancaster. Stohlgren, T. J., Y. Otsuki, C. A. Villa, M. Lee, and J. Belnap. 2001. Patterns of plant invasions: A case example in native species hotspots and rare habitats. Biological Invasions 3:37-50. Stratton, L. C., and G. Goldstein. 2001. Carbon uptake, growth and resource-use efficiency in one invasive and six native Hawaiian dry forest tree species. Tree Physiology 21:1327-1334. Turner, I. M. 2001. The Ecology of Trees in the Tropical Rain Forest. Cambridge University Press, Cambridge. Veenendaal, E. M., M. D. Swaine, V. K. Agyeman, D. Blay, I. K. Abebrese, and C. E. Mullins. 1996. Differences in plant and soil water relations in and around a forest gap in West Africa during the dry season may influence seedling establishment and survival. Journal of Ecology 84:83-90. Whitmore, T. C. 1998. Potential impact of climatic change on tropical rain forest seedlings and forest regeneration. Climatic Change 39:429-438.

64

Chapter 3 The role of forest gaps in woody plant invasions into secondary forests in the Seychelles

65

Chapter 3

Abstract

1 The aim of this study was to understand better the role played by gaps in enabling alien tree species to invade tropical secondary forest. The work was conducted on the island of Mahé in the Seychelles, in mid-altitude forests dominated by the invasive tree Cinnamomum verum. 2 We monitored the growth of seedlings of five invasive and four native species transplanted into either artificially created forest gaps of c. 15 m diameter or into the undisturbed understorey. Half of the seedlings were fertilized with additional nutrients. We also monitored the growth of self-sown seedlings of Cinnamomum verum along transects from the forest understorey to the interior of the gaps. 3 Both the native and invasive seedlings grew faster in the gaps than in the understorey, and they also profited more from added nutrients in the gaps. The mean growth responses did not differ significantly between the invasive and native species. The growth of plants in the gaps was similar to that of seedlings in the high light treatment in an earlier common garden experiment with potted plants. However, plants in the understorey plots grew much more slowly than those in the low light treatment of the pot experiment, and also responded less to nutrients. 4 The growth of Cinnamomum verum phytometer seedlings was 5 to 10 times faster in the gap centre than in the forest understorey. However, changes in growth along the transects were more gradual than might have been expected from the steep changes in light conditions at the edge of the gaps. This result was attributed to root competition in the gaps from established Cinnamomum verum trees. 5 Three important conclusions can be drawn from our study. First, it is probable that only a fraction of the invasive trees in Seychelles have the growth characteristics that allow them to take advantage of forest gaps. However, those species may grow considerably faster in gaps than most native species. Second, even fairly small disturbances to the canopy may facilitate the spread of invasive species. Light levels in the understorey of secondary Cinnamomum verum forests are relatively high (c. 10% of ambient), and our results show that seedlings respond strongly to even small increases of light at these light levels. Third, the effect of forest gap formation on belowground root competition is as important as the effect on light availability for seedling growth in Cinnamomum verum secondary forests. Keywords Cinnamomum verum, invasiveness, forest gap, light, nutrient, oceanic islands, RGR, root competition, Seychelles, transplant experiment, tropical forest management, tropical tree seedlings 66

Field Transplant Experiment

Introduction The spread of alien species is one of the main reasons for the loss of biodiversity (Millennium Ecosystem Assessment 2005), and the urgent need to prevent further invasions is now internationally recognized (cf. McNeely et al. 2001). One possible management strategy to achieve this is to identify potentially invasive species before they spread into natural areas, for example by preventing the import of species at national borders (Daehler et al. 2004). However, many invasive plant species are already widely established and abundant in areas of high conservation value (Stohlgren et al. 2001, Denslow 2003), and in these cases control measures need to be integrated into strategies for habitat management (e.g. Zavaleta et al. 2001, D`Antonio and Meyerson 2002). In order to define appropriate management strategies, it is important to understand what factors make a habitat vulnerable to invasion by alien plant species (i.e. habitat invasibility, Lonsdale 1999). In particular cases a variety of different factors may be important: abiotic factors such as resource availability (e.g. Davis et al. 2000), harsh environmental conditions (Alpert et al. 2000), disturbance (e.g. Elton 1958, Hobbs and Huenneke 1992, Huston 2004), and biotic factors such as species diversity (Levine et al. 2004). Recent syntheses emphasise the role of disturbances in releasing pulses of resources while reducing competition for these resources by the established plant community (Davis et al. 2000, Huston 2004, Richardson and Pysek 2006). An understanding of the role of disturbance is particularly important for the conservation of tropical forests. This is because the spread of alien species tends to be associated with both natural (e.g. Leps et al. 2002, Bellingham et al. 2005) and anthropogenic (e.g. Lavergne et al. 1999, Baret and Strasberg 2005, Totland et al. 2005) disturbance of the canopy. However, while many mainland tropical forests have proved highly resistant to plant invasions (Fine 2002), most forests on oceanic islands are heavily invaded (Denslow 2003). This difference may partly reflect the fact that deforestation on oceanic islands has been particularly severe, but other factors related to plant evolution on isolated islands may also be important.

Developing an appropriate management strategy for uninvaded and undisturbed forest is straightforward - prevent anthropogenic disturbance. However, the situation in disturbed, secondary forests on oceanic islands is more complex because the vegetation is often dominated by alien species, and only small patches of native species survive (e.g. Wiser et al. 2002, Kueffer and Vos 2004, Lugo 2004). In these circumstances, there is a balance to be struck between removing alien species in order to facilitate the 67

Chapter 3 regeneration of native species and avoiding disturbance of the forest canopy, which may enhance further plant invasions (compare Adler et al. 1998, Zavaleta et al. 2001, Hata et al. 2006, Kueffer et al. 2007). The mountainous island of Mahé (Republic of Seychelles, Indian Ocean) represents a suitable system for investigations into the role of forest gaps in the rehabilitation of secondary forests dominated by alien species. The canopy trees in the intermediatealtitude forests typically include 70-90% alien species, mainly Cinnamomum verum, but native plants survive in scattered patches and juveniles of these species can be found in the understorey (Fleischmann 1997, Kueffer et al. 2007). One strategy that has been proposed for restoring these secondary forests is to cut small gaps and replant them with native species (Kueffer 2003). The overall aim of this study was to understand how disturbance affects the balance of regeneration between native and invasive species in secondary forest in the Seychelles.

Based on the work of Fleischmann (1999), we hypothesised that the relative performance of native and invasive species would change predictably along resource gradients, with native species outperforming invasive species under conditions of low light and low nutrients but with invasive species being better able to exploit higher levels of these resources. In a previous common garden experiment, we tested this hypothesis by growing seedlings of several native and invasive species in pots under a range of light and nutrient conditions, and comparing growth rates and patterns of biomass and nutrient allocation. The results suggested that juvenile tree growth in Seychelles forests is strongly limited by both light and nutrients, and that increasing either of these resources to levels above those typical in the forest leads to strongly increased growth. Compared to native species, seedlings of invasive species tended to have higher growth rates (RGR), higher specific leaf areas (SLA) and higher leaf nutrient contents. They also exhibited greater plasticity in biomass and nutrient allocation (i.e. greater plasticity in SLA, LAR, RSR, leaf nutrient content) in response to resource availability; this was reflected in an increased differentiation between the status groups under high resource conditions. However, while these patterns fit with current generalizations about what makes some plant species invasive (Daehler 2003, Richardson and Pysek 2006), the differences in mean values between the two species groups were generally small compared with the variation within groups. The aim of the work described here was to find out whether seedlings growing under more natural conditions in forest gaps respond to varying light and nutrient conditions similarly to those in a pot experiment. Deviant patterns under field conditions would indicate that other factors also influence tree regeneration. We therefore monitored 68

Field Transplant Experiment the growth of tree seedlings transplanted into artificially created gaps and into the undisturbed understorey and subjected them to one of two nutrient conditions (ambient vs. fertilized). The main experiment used five invasive and four native species. In addition, we monitored the growth of self-sown seedlings of the invasive tree Cinnamomum verum along transects from the forest understorey to the interior of the gaps; these served as phytometers to evaluate how tree growth responded to variation in environmental conditions created by the gaps.

Methods Study area The study was carried out in a mid-altitude forest at Mare aux Cochons (MC) within the Morne Seychellois National Park on the island of Mahé (4° S, 55° E, 154 km2, Republic of Seychelles). The Mare aux Cochons is in a forested upland plateau (c. 450 m asl.) situated in the mountainous North-West of Mahé (Fig. 1), where altitude increases rapidly from sea level to the highest peak of the Seychelles, Morne Seychellois, 914 m asl.. The native vegetation at MC was almost completely destroyed at the beginning of the 20th century and Cinnamomum verum (cinnamon) was cropped there until the 1970s. In our study area, c. 85% of the canopy was formed by C. verum trees, and the canopy also included a few patches of the alien Syzygium jambos and the native Northea hornei. The forest had a very dense layer of tree seedlings (> 100 per m2), mainly of C. verum. The soils at MC are ferrasols that have developed on granitic bedrock. The pH is around 4.5, and nutrient availabilities are generally low. Plant growth appears to be limited mainly by the availability of phosphorus, and possibly by other nutrients including potassium (cf. Kueffer 2006). Mean leaf nutrient concentrations (and ranges) in seedlings of two native and two invasive species in the forest understorey at MC were 12.5 (10.9-15.2) mg g-1 N and 1.1 (0.9-1.2) mg g-1 P. During the study period, the annual rainfall was c. 3400 mm. Although there is no pronounced rainfall seasonality on Mahé, the period between June and September is generally drier than that between November and February. Monthly mean rainfall

69

Chapter 3 typically ranges from 80 to 150 mm during the ‘dry’ period to 300 to 450 mm in the ‘wet’ period at low to mid altitudes (< 600 m asl.) (Cazes-Duvat and Robert 2001). Over the past 30 years, c. 10 dry days per month (< 0.1 mm rainfall per day) were recorded during the wet period and 15 during the dry period (near the coast, Meteo Seychelles). At this mid-altitude site the mean annual air temperature 1 m above ground under closed canopy was 23°C (seasonal range of mean maximum daily temperature: 23-26 °C) and mean annual air humidity was 98% (daily range: 95-100%; data obtained by an Onset HOBO Pro RH/Temp sensor) Mare aux Cochons Area (Morne Seychellois National Park) Trail

U2

N

G2 plot

G1

Rive

U1

r

LN

subplots

G3 U3

U4

G4 HN

2m

Mahé Island Altitude (m)

N Victoria

900 600 400 200 0 mountain pass

Swamp

U5 1k

m

Mare aux Cochons G5

5 km

Fig. 1

100m

Locations of gap (G) and understorey (U) plots used in the transplant experiment. Each plot was subdivided into two subplots with a high nutrient (HN) and low nutrient (LN) treatment. The inlet map indicates the location of the study site at Mare aux Cochons on the island Mahé.

Species We selected four native and five invasive tree species that are abundant in the forests on Mahé and for which either seeds or seedlings were available at the time of the experiment (Tab. 1). The native species included two indigenous and two endemic species. The majority of the invasive species were introduced to the islands in the late 19th or early 20th century. Only Cinnamomum verum and Syzygium jambos were introduced 70

Field Transplant Experiment more than 200 years ago (Kueffer and Vos 2004). We avoided including closely related species pairs within groups. All species have small to medium sized seeds (2 - 10 mm in diameter) except S. jambos, which has seeds of 15 - 20 mm diameter. For the phytometer experiment we used self-sown C. verum seedlings in the experimental gaps and surrounding forest. Nomenclature follows Friedmann (1994). Table 1 Characterization of the tree species used in the transplant experiment. The experiment started between 02.03.2004 and 24.03.2004 (duration 305-324 days). The plants were either grown from seeds (S), or collected from seedling banks in the forest (F). Nomenclature and maximal stem height were taken from Friedmann (1994). Species

Acronyms

Family

Maximal stem Seedling height (m) origin

Invasives Alstonia macrophylla

Am

Apocynaceae

15

F

Cinnamomum verum

Cv

Lauraceae

15

S

Psidium cattleianum*

Pc

Myrtaceae

7

F

Syzygium jambos*

Sj

Myrtaceae

10

F

Tabebuia pallida*

Tp

Bigogniaceae

10

S

Aphloia theiformis+

At

Flacourtiaceae

12

F

Canthium bibracteatum

Cb

Rubiaceae

8

S

Erythroxylum sechellarum†

Es

Erythroxylaceae

7

F

Memecylon eleagni†

Me

Melastomataceae

10

S

Natives

subsp. madascariensis var. seychellensis * species that are also invasive in many other tropical regions † species endemic to the Seychelles +

Experiment 1: Transplant experiment Experimental setup Five clearings of approximately 15 by 15 m2 (area c. 225 m2) were made to achieve the kind of canopy opening that is common following moderate anthropogenic disturbance (e.g. due to control of invasive species or commercial exploitation of feral Cinnamomum verum). To create these clearings all adult trees (mainly C. verum) were cut with chainsaws and the fallen trees were then removed and piled up at the edge 71

Chapter 3 of the gap. Care was taken to keep disturbance to the forest floor to a minimum. In the vicinity of each gap, an understorey plot was chosen under undisturbed C. verum (Fig. 1). Four of the species in the experiment - C. verum and Tabebuia pallida (invasive) and Canthium bibracteatum and Memecylon eleagni (native) - were grown from seed collected in the forest. Ripe fruits were collected directly from 5 - 15 parent trees per species, and the mixed seed sown into seedling trays immediately after collection. When the seedlings had developed the first true leaves (three to six months after sowing) the plants were transplanted into 1-litre pots filled with forest soils. For the other species - the invasives Alstonia macrophylla, Psidium cattleianum and S. jambos, and the natives Aphloia theiformis and Erythroxylum sechellarum - no seed was available and we therefore collected in the field young plants similar in size as those grown from seed. All plants were allowed to adapt to the pot environment for two weeks before being transplanted into the forest at the onset of the experiment. A split-plot design was used, with light availability as the main-plot factor and nutrient treatment as the split-plot factor. Mean light levels were 10% (range 8.2 - 12.6%) of ambient light in the understorey plots and 63% (56 - 69 %) in the gap plots. The mean total N concentration in the soil at the plots (understorey and gap sites combined) was 1.4 mg g-1 (range: 0.1 - 2.8 mg g-1), while that of P was 0.34 mg g-1 (range: 0.14 - 0.58 mg g-1). Two subplots (2 x 2 m2) spaced 2 metres apart were placed in the centre of each gap and understorey plot. All herbs and seedlings were removed from the subplots, but the sparse leaf litter layer was left. Two seedlings of each species (i.e. 18 seedlings in all) were planted at randomly selected points on a regular grid with a distance of 35 cm between neighbouring seedlings. One of each pair of subplots was randomly assigned to the ambient nutrient (LN, no nutrients added) and the other to the high nutrient treatment (HN, addition of fertilizer). For the HN treatment, 1 g of slow release N-PK-fertilizer (Osmocote 16:11:11, Osmocote, Scotland) was added around each seedling every two months. The experiment was started in early March 2004 and ran until January 2005. During this period, seedlings were watered as necessary and any unwanted plants were removed. Data collection At the beginning of the experiment, six seedlings per species were selected at random from the surplus plants and harvested to determine the initial dry weight. Thereafter, the following measurements were made on the experimental plants at two-month 72

Field Transplant Experiment intervals: stem height, number of leaves and stem diameter. The lengths and breadths of all leaves were measured at the beginning and end of the experiment, and the data used to estimate total leaf area. To do this, linear regressions of leaf area against the product of leaf length and breadth were calculated for a sample of leaves (> 100 per species). These leaves were placed beneath a glass plate and photographed with a digital camera (Nikon Coolpix 995, resolution at 2048 x 1536 pixels); the images were then used to determine leaf length and breadth using Adobe IllustratorTM 10, and area using Adobe PhotoshopTM 7.0 (cf. Dietz and Steinlein 1996). At the end of the experiment, the plants were harvested and divided into leaves, stems plus petioles, and roots. All material was then oven-dried at 80°C for 48 hours. Subsamples of leaf material of two native (A. theiformis and C. bibracteatum) and two invasive (C. verum and P. cattleianum) species were digested at 420°C with 98% H2SO4 and Merck Kjeltabs, and total nitrogen and phosphorus concentrations of leaves were determined colorimetrically using a flow injection analyzer (FIA, TECATOR, Höganäs, Sweden). The raw data were used to calculate the following growth parameters: RGRDW Relative growth rate of total dry weight

ln (dry weight at end)-ln(dry weight at start) duration of experiment ln (leaf area at end)-ln(leaf area at start) duration of experiment

RGRLA

Relative growth rate of leaf area

SLA

Specific leaf area

leaf area dry leaf biomass

LAR

Leaf area ratio

leaf area dry plant biomass

RSR

Root:shoot ratio

dry root biomass dry shoot biomass

Experiment 2: Cinnamomum verum phytometer experiment Between July 2004 and February 2005, we monitored the growth of established Cinnamomum verum seedlings in three of the five gaps (Gaps 1, 2 and 3; Fig. 1). The seedlings studied lay along two perpendicular transects (orientation S-N and E-W) passing through the centre of each gap and extending 7 m into the forest on both sides; for the analysis the transects in each gap were treated as four sub-transects running

73

Chapter 3 from the gap into the forest. Established seedlings c. 20-30 cm tall were chosen at the following distances along each subtransect: -7, -5, 0, 2, 4 and 6 metres (where 0 represents the edge of the gap and negative values are in the forest; Fig. 2). At each sampling point three randomly chosen seedlings were marked. Additionally, at two spots > 30 m from any gap we monitored growth of 8 seedlings (control plants). Number of leaves, stem height and percentage leaf area damaged by herbivory were monitored every two months. Dry aboveground leaf and stem biomass and leaf area were determined at the end of the monitoring period. Growth parameters (RGR of aboveground biomass and SLA) were calculated as described for the transplant experiment above. Leaf number and stem height growth were calculated as the difference of the final minus the first measurement (i.e. assuming linear growth). Fig. 2

Lay-out of the transects and the positions of the sampling points used in the phytometer experiment with Cinnamomum verum seedlings. C. verum seedlings were also monitored at positions more than 30 m distance from a gap. N 3 seedlings of C. verum per sampling point C. verum seedlings in the understorey (control plants)

Forest

2m

5m

Forest

2m

2m

Gap

2m

5m

E

2m

W

m

30

Forest

Forest

S

74

Field Transplant Experiment

Stastistical analysis Experiment 1: Transplant experiment In each gap and understorey plot, the average of the two replicate plants was used for the analysis. We used general linear models with light level, nutrient level, species status (native vs. invasive) and their interactions as fixed factors, and species identity (nested in species status) as well as plot type (gap vs. understorey, nested in light treatments) as random factors. The leaf area of each plant at the start of the experiment was included as a covariable to account for differences in initial plant size. Dependent variables were relative growth rates of total dry weight and leaf area, and the allocation parameters SLA, LAR, RSR. To remove heteroscedasticity, SLA, LAR and RSR were log-transformed. Experiment 2: Phytometer experiment The analyses used the mean values of the three replicate C. verum seedlings at each sampling position. To avoid pseudo-replication we pooled the data for the four subtransects in each gap, resulting in one value per transect position. We used general linear models with plot (i.e. 3 replicate gaps) and distance from gap edge as fixed factors. Various models were made using total dry weight, number of leaves and stem height produced over the experimental period, and SLA as the dependent variables. Differences between positions were tested with a Tukey test. All statistical analyses were performed with JMP V 6.0 (SAS Institute Inc., 2005).

75

Chapter 3

Results Growth responses to varying light and nutrient availability: gap versus understory plots Mortality Overall seedling mortality during the experiment was 28% for the invasives species and 21% for the natives. More seedlings died in the understorey plots than in the gaps (29% and 21%, respectively), and in sub-plots with added nutrients than in the controls (30% and 20%, respectively). Two invasive species, Tabebuia pallida and Alstonia macrophylla, suffered the highest overall mortality (40% and 85%, respectively), with no plants of A. macrophylla surviving in the understorey plots. Among the native species, Aphloia theiformis and Memecylon eleagni showed relatively high mortalities, with ≥ 30% of plants dying in both the understorey and gap plots, while for Erythroxylum sechellarum plants 30 % died in the understorey plots but none of the plants in the gaps. Table 2 Results of ANOVA across two light (gap and understorey) and two nutrient levels (ambient and high nutrients). Indicated are the F-ratios and significance levels (***: P < 0.001, **: P < 0.01: *: P < 0.05, significant ones in bold) of main or interaction effects on relative growth rate of total dry weight (RGRDW) and leaf area (RGRLA), root dry weight (RootDW), specific leaf area (SLA), leaf area ratio (LAR) and root:shoot ratio (RSR). See text for further information. RGRDW

RGRLA

RootDW

SLA

LAR

RSR

0.1

0.3

2.6

0.6

0.2

0.0

55.5 ***

34.0 ***

21.2 ***

18.8 **

2.9

5.6 *

0.1

0.0

3.2

0.1

0.3

0.5

11.2 *

11.7 *

4.9

2.3

0.3

0.4

SxN

0.3

3.0

1.8

0.7

0.0

6.6

LxN

9.1 **

5.4 *

6.5 *

0.0

0.0

2.5

1.1

3.7

2.3

0.9

1.0

0.0

Species group† (S) Light (L) SxL Nutrient (N)

Initial leaf area †

native vs. invasive

Relative growth rate Both native and invasive species grew faster (relative growth rate of total dry weight, RGRDW) in the gap plots than in the understorey plots (P < 0.001, Fig. 3, Tab. 2), and plants in gaps also profited more from added nutrients (significant interaction between 76

Field Transplant Experiment light and nutrient, P = 0.003, Fig. 3, Tab. 2). The range of RGRDW values among species was greater in the gaps (2 - 16 mg g-1 d-1) than in the understorey (-0.5 - 5 mg g-1 d-1; Fig. 4A).

RGRDW mg g-1 d-1

A

20

Gap

Understorey

15

Invasive Native 10

5

0

High

Low

High

Low

Nutrient availability

B

2.5

Start

End

Gap

Understorey

RootDW g

2.0

1.5

1.0

0.5

0.0

High

Low

High

Low

Nutrient availability

Fig. 3

A Relative growth rate of total dry weight (RGRDW) and B root dry weight at the end (RootDW) of invasive (black bars) and native (grey bars) seedlings (mean ± SE). Data are shown for plants growing under two light levels (forest understorey and gap) and two levels of nutrients (low and fertilized). See text for further information.

77

Chapter 3 Species showed significantly different responses to light (species x light, P < 0.001) but not to nutrients (species x nutrient, P = 0.9). However, RGRDW did not differ significantly between the invasive and native species in any treatment (P ≥ 0.3), and the variation among species was independent of species status (species effect in separate ANOVA, P < 0.001). In the gaps, the invasives Alstonia macrophylla and Psidium cattleianum were the fastest growing species, whereas in the understorey Aphloia theiformis and Canthium bibracteatum (both natives) were among the three fastest growing species. There was a stronger tendency for the invasive species to respond positively to nutrient addition (Fig. 4A), but the results were not fully consistent. Thus, both the native Erythroxylum sechellarum and the invasive Tabebuia pallida showed negative growth rates under low nutrient availability and profited strongly from nutrient addition, whereas the invasive Syzygium jambos grew more slowly with added nutrients. The two species with the highest RGRDW under low nutrient availability - A. theiformis and C. bibracteatum (both natives) - also had a lower RGRDW in response to nutrient addition (Fig. 4A).

The data for relative growth rates of height (RGRH) and total leaf area (RGRLA) showed very similar patterns and were highly correlated with RGRDW (r > 0.6, P < 0.01; data not shown). Table 3 Mean values per species group (native vs. invasive) of different growth parameters. See table 2 for explanation of acronyms. High radiation (Gap)

Low radiation

High nutrient

Low nutrient

High nutrient

Low nutrient

invasive

8.7 (0.27)

6.9 (0.22)

0.9 (0.05)

0.7 (0.06)

native

7.6 (0.25)

6.2 (0.19)

1.4 (0.07)

2.0 (0.13)

invasive

7.6 x 10-3 (2.3 x 10-3)

5.1 x 10-3 (2.4 x 10-3)

0.6 x 10-3 (0. 7 x 10-3)

0.7 x 10-3 (0. 6 x 10-3)

native

6.6 x 10-3 (2.0 x 10-3)

5.4 x 10-3 (1.5 x 10-3)

1.3 x 10-3 (0.9 x 10-3)

1.6 x 10-3 (1.7 x 10-3)

invasive

1.75 (0.40)

0.81 (0.12)

0.19 (0.03)

0.18 (0.03)

(g)

native

0.49 (0.12)

0.25 (0.06)

0.05 (0.01)

0.05 (0.01)

SLA

invasive

132 (26)

130 (22)

202 (54)

206 (59)

(cm2 g-1)

native

108 (11)

110 (11)

179 (46)

184 (38)

LAR

invasive

53 (12)

55 (16)

60 (10)

66 (8)

native

47 (5)

51 (9)

74 (26)

77 (29)

invasive

0.49 (0.10)

0.53 (0.09)

0.66 (0.10)

0.60 (0.10)

native

0.49 (0.11)

0.54 (0.11)

0.66 (0.11)

0.61 (0.14)

RGRDW

(mg g-1 d-1) RGRLA

(cm2 cm-2 d-1) RootDW

(cm2 g-1) RSR

78

Field Transplant Experiment A

30

Understorey

Gap

Am Cv Pc Sj Tp

RGRDW mg g-1 d-1

25 20 15

At Cb Es Me

10 5 0 High

Low

High

Low

Nutrient availability

B

30

Low radiation

High radiation

RGRDW mg g-1 d-1

25 20 15 10 5 0 High

Low

High

Low

Nutrient availability

Fig. 4

RGRDW of the species grown in the transplant experiment (A) compared with that of the same species set in the common garden experiment (B, data from chapter 1). Filled symbols, invasives species; open symbols, native species. In the understorey plots all individuals of Alstonia macrophylla (Am) died. For acronyms of species names see table 1.

Biomass allocation The specific leaf area (SLA) of all species was significantly lower in gaps than in understorey plots (100 - 130 cm2 g-1 vs. 170 - 200 cm2 g-1; respectively; P = 0.003). Leaf area ratio (LAR) also tended to be lower in gaps than in the understorey (50 cm2 g-1 vs. 60 cm2 g-1; P = 0.1). However, nutrient addition had no effect on either SLA or LAR (P ≥ 0.2), and neither of these parameters varied significantly between the native and invasive species in any treatment (P ≥ 0.4, Tab. 3). 79

Chapter 3 Plants in the understorey plots developed a slightly higher mean root:shoot ratio (RSR) than those in gap plots (Tab. 3, P = 0.04), but this parameter was not affected by the nutrient treatment (P = 0.5). Total root dry weight (RootDW) of plants in the understory scarcely changed during the experiment (Fig. 3B), while in the gaps it increased by a factor of 5 in the low nutrient treatment and 10 in the high nutrient treatment (light effect P < 0.001; light x nutrient P = 0.01; Tab. 2, Fig. 3B). However, RSR did not differ significantly between native and invasive species (P = 0.9), though there was a nonsignificant trend for native species to respond more strongly to nutrients (P = 0.06).

Growth responses of Cinnamomum verum seedlings along understorey to gap gradients On the phytometer transects, light intensity at ground level in the gaps ranged from 40 to 55% of ambient (P < 0.001) (Fig. 5). At the gap edge, light levels decreased abruptly and soon levelled off in the interior of the forest, where they ranged from c. 9-14% of full daylight (Fig. 5). Growth of Cinnamomum verum seedlings was low in the understorey part of the transect but increased from the edge of the gap towards the centre (Fig. 5): final seedling biomass (P = 0.002), number of new leaves (P = 0.05) and absolute stem height growth (P < 0.001) all increased by a factor of 5 to 10 from the understorey to the gap centre. In contrast, SLA varied only slightly along the transects, from c. 130 cm2 g-1 in the understorey to c. 110 cm2 g-1 in the gap centre (P = 0.06). The changes in growth along the transects were more gradual than might have been expected from the changes in light conditions. Only differences in growth parameters between points in the forest interior (i.e. ≥ 5 m from the gap edge) and the gap centre (i.e. ≥ 4 m from the gap edge) were significant (Fig. 5).

80

Field Transplant Experiment

8

60

c

Aboveground biomass

c

% Ambient light

7

50

Aboveground biomass g

bc

6 b

40

5 4

30

3

20 a

2

a

a

10

1 a

a

a

ab

- 30m

- 7m

- 5m

edge

0

B

ab

b

+ 2m

+ 4m

60

c

c

Height

12

bc

bc

c

50

10 b

40

8

abc

30 6 ab

4

a

a a a a

a

20

a

a

a

2

a a

10

a

a

0

0 - 30m

- 7m

- 5m

edge

+ 2m

+ 4m

+ 6m

Gap

Understorey

Fig. 5

0

+ 6m

14 Number of leaves

Number of leaves Height cm

b

Ambient light transmitted to the ground %

A

Growth of naturally established Cinnamomum verum juveniles along an environmental gradient from the understorey to the centre of the gap plot. Shown are dry aboveground biomass at the final measurement (A) and increase of number of leaves and stem height during 6 months (B; mean ± SE). Responses in growth are shown in relation to light levels (% ambient light transmitted to the ground). Distance from the gap edge into the forest understorey are indicated by negative numbers and those into the gap interior by positive numbers. Different letters indicate significant differences (Tukey test).

81

Chapter 3

Discussion The gaps had a strong positive effect on the growth of seedlings, both in the transplant experiment and in the transect study with Cinnamomum verum seedlings. In addition, the seedlings in the gaps responded positively to nutrient addition. These are well known patterns of response in tropical forests (e.g. Coomes and Grubb 2000, Turner 2001 and references therein). The relative growth rates of transplanted and self-sown C. verum seedlings were very similar (data not shown), indicating that process of transplanting had little effect on the results. Contrary to the notion that gap formation strongly favours invasive species (e.g. Fine 2002, Daehler 2003, Totland et al. 2005), the difference in performance of native and invasive species in gaps was relatively small (Fig. 3, Fig. 4). In fact, three of the four species that profited most from the gaps (the invasive Alstonia macrophylla and the two native, non-endemic Aphloia theiformis and Canthium bibracteatum) are typically found

in high light environments in the Seychelles (Fleischmann 1997, Kueffer and Vos 2004), suggesting that the response to gap conditions is related more to the growth strategies of individual species rather than their status as natives or invasives. The other lightdemanding invasive species in our experiment, Tabebuia pallida, did not grow well, even under gap conditions. This species is restricted in the Seychelles to the coastal and lowland plateau, where it is highly invasive (Kueffer and Vos 2004), and it has been suggested that its range is restricted by the low nutrient conditions of the inland forests in Seychelles (Kueffer 2006). However, this suggestion is not confirmed by our experiment, since nutrient addition had only a minor positive effect on the growth of T. pallida. Further research is needed to understand the poor performance in upland forests of T. pallida and other alien species such as Psidium guajava that are only invasive in the lowlands (see chapter 1). Unexpectedly, the data show no general growth advantage of native over invasive species in the understorey plots. This may be because we did not include any very shade-tolerant native species in our study, these being either very rare or having very large seeds (e.g. Northea hornei, seed diameter: c. 6 cm). Furthermore, some of the most shade tolerant trees in the Seychelles are palms and pandans, which have a very different growth form from the tree species we used. The mean relative growth rate for all species varied rather little among the five gaps, with the highest value being 1.5-times greater than the lowest (P = 0.5); in contrast,

among the understorey plots the mean values varied by a factor of more than 30 (P = 0.002). This higher variation in the understorey could have been due to differences

82

Field Transplant Experiment in light availability or soil conditions or biotic factors (e.g. herbivory). The substrates of the sites varied from coarse sand to well-developed organic soils, while nutrient availability in the soil varied widely both among and within sites (Kjeldahl P: 0.10.6 mg g-1, total N: 0.1-2.8 mg g-1). However, since soil conditions also varied among the gap plots, the most likely reason for the higher variation among the understorey plots is varying light conditions. Separate linear regressions of RGR against light were calculated for each species using data from the five gap plots, and the results show no strong tendency for seedling growth to increase in the range of light conditions represented in the gaps (55% to 70% ambient light; r < 0.2, P > 0.5). However, a corresponding analysis for the understorey plots shows that most species increased their RGRDW by a factor of two or three along the light gradient from c. 8% to 12.5% (r > 0.6, P < 0.1). This result is consistent with other studies which show that the response of tropical tree seedlings to increased light is usually strongest below 10-20% of ambient and levels off quickly at higher light levels (Turner 2001). It indicates that even if no gaps are formed, varying light availability may have a considerable influence on relative growth rates of native and invasive tree seedlings. According to this analysis, some of the responses of individual species to light were unexpected. In the gaps, the growth rate of the invasive C. verum declined by almost 50% along the gradient from 55% to 70% of ambient light (r = 0.8, P = 0.04). And in the understorey plots the two endemic species Erythroxylum sechellarum and Memecylon eleagni responded to high light only when fertilized, contrary to our expectation that native species would be well adapted to low nutrient availabilities (see chapter 1).

What is the role of root competition by canopy trees in gap versus understorey seedling growth? To evaluate the interacting factors shaping juvenile tree growth under field conditions it is helpful to compare the results presented here with those obtained in a previous study using potted plants grown under comparable light and nutrient levels in a common garden (cf. chapter 1). Under high light, the growth rates of all species were slightly lower in the field (i.e. in gaps) than in the pots (Fig. 4), which could be due to differences in the substrate or to the presence of additional stressors in the field, such as occasional exposure to direct sunlight. However, the rank order of species RGR values was similar in both experiments. Also, the means for biomass allocation parameters (SLA, LAR and RSR) in the native and the invasive species differed by less than 15% in the two experiments.

83

Chapter 3 This indicates a good transferability of the results obtained in the pot experiment to near-to-natural field conditions.

A

0.8 0.7

B

Invasive Native

120

100

0.6 80

LAR cm2 g-1

RSR

0.5 0.4 0.3

60

40

0.2 20

0.1

0

0 Common garden

Fig. 6

Transplant

Common garden

Transplant

Biomass allocation patterns under low light of plants in the common garden experiment and in the transplant experiment (mean ± SE). Invasive species, black bars; native species, grey bars.

In contrast, relative growth rates in the forest understorey were much lower than in plants grown under low light in pots; and nutrient addition to the understorey had no clear effect on seedling growth, whereas in the pots most invasive species responded positively. These differences are probably due to strong root competition by adult trees of Cinnamomum verum (cf. Kueffer et al. 2007). Species that exhibit high plasticity in response to resource levels may suffer from conflicting allocation priorities where there are multiple stresses (cf. chapter 2); while high allocation below ground is the necessary response to strong root competition, high allocation above ground is needed under low light conditions (compare Cahill 1999). In support of this argument, biomass allocation to the roots of the seedlings was far higher in the understorey than in the pot-grown seedlings (Fig. 6A), particularly for the invasive species. And the invasive species also reduced biomass allocation to the leaves in the field (i.e. a lower leaf area ratio, Fig. 6B), which may indicate that belowground needs for carbon may have restricted the leaf production needed by these fast-growing species to survive under low light. Overall, the results of this study are consistent with the hypothesis that seedlings of invasive species suffer more from strong belowground competition than those of native species (Kueffer et al. 2007). And the potential allocation conflicts when both nutrients and 84

Field Transplant Experiment light are in short supply could explain why all understorey plants of fast-growing invasive species, A. macrophylla, died. The phytometer experiment also provides evidence that root competition is an important factor limiting seedling growth in secondary Cinnamomum verum forests. The changes in growth rate from the understorey to the gap centre were rather gradual, with the increase from the understorey to the gap edge being of similar magnitude to the increase from the edge of the gap to the centre. However, these changes in growth along the transect did not match closely the trend in light intensity, which increased abruptly at the edge of the gap; and the growth trends in the gaps were also not consistent with the known responses of seedlings to light intensities in the range from 40% to 55% (Fig. 3A). The reason for these dicrepancies is probably that root competition from C. verum trees also limits seedling growth in the gaps, with the importance of this factor declining with increasing distance from the gap edge. If root competition by C. verum is an important factor in gaps, then conditions for tree regeneration are likely to be very different in C. verum forests compared with vegetation dominated by native trees, where root competition is probably much less. In both types of forest, light conditions in a small gap (e.g. one formed by the death of a single tree) would be the same, but only in the C. verum forests would there also be strong suppression of seedling growth by root competition. The significance of gap size for seedling growth in C. verum forests may also vary over time. We used gaps of recently felled trees, and the duration of the experiment probably did not allow time for the roots of surrounding trees to spread into the gap centres. However, such release from root competition may be only transient except in much larger gaps.

The importance of forest gaps for managing woody plant invasions in Seychelles Two important conclusions for habitat management can be drawn from our study. First, it is probable that only a fraction of the invasive trees in Seychelles have the growth characteristics that allow them to take advantage of forest gaps of intermediate size (c. 15 metres in diameter). However, those species may grow considerably faster in gaps than most native species. Some of these invasive trees - notably Alstonia macrophylla and Falcataria moluccana - produce large quantities of small, wind-dispersed seeds that can colonize relatively distant gaps. Both these species are widespread in mid-altitude

85

Chapter 3 forests on Mahé, and within four months of the start of the experiment large numbers of their seedlings (>100 per m2) had become established in some gaps. Recently, rather new invasive species such as Clidemia hirta and Dillenia suffruticosa have been rapidly spreading under gap conditions (Kueffer and Vos 2004). Creation of gaps for the exploitation of Cinnamon bark or habitat rehabilitation purposes will therefore require major follow-up work, including regular weeding and replanting with native species. Second, even fairly small disturbances to the canopy may facilitate the spread of invasive species. Light levels in the understorey of secondary C. verum forest are relatively high (c. 10% of ambient light), and our results show that seedlings respond strongly to even small increases of light at these light levels. And because of lower root competition from canopy trees in native forests, the formation of small gaps may have a stronger positive effect on seedling growth, particularly of invasive species. It must, therefore, be a high conservation priority to avoid any canopy disturbances in relatively undisturbed native forests. In secondary Cinnamomum verum forest, a high priority may be to weed and rehabilitate the remaining small pockets of native vegetation.

Acknowledgements We thank the Seychelles Ministry of Environment and Natural Resources for their support with the conducting of the experiment, and particularly the staff of the Morne Seychellois National Park unit for permission for the creation of the forest gaps, the creation of the gaps, and assistance with data collection. Especially, we would like to acknowledge the assistance with the fieldwork by Unels Bristol and Terence Valentin, and the provision of statistical advice by Sabine Güsewell. Funding was provided by a research grant from the Swiss Federal Institute of Technology (ETH Zurich).

Literature

Adler, P. B., C. M. D’Antonio, and J. T. Tunison. 1998. Understorey succession following a dieback of Myrica faya in Hawai’i Volcanoes National Park. Pacific Science 52:69-78. Alpert, P., E. Bone, and C. Holzapfel. 2000. Invasiveness, invasibility and the role of environmental stress in the spread of non-native plants. Perspectives in Plant Ecology, Evolution and Systematics 3:52-66.

Baret, S., and D. Strasberg. 2005. The effect of opening trails on exotic plant invasion in protected areas on La Réunion Island (Mascarene Archipelago, Indian Ocean). Revue d’écologie - La Terre et la Vie 60:325-332.

86

Field Transplant Experiment Bellingham, P. J., E. V. J. Tanner, and J. R. Healey. 2005. Hurricane disturbance accelerate invasion by the alien tree Pittosporum undulatum in Jamaican montane rain forests. Journal of Vegetation Science 16:675-684. Cahill, J. F. 1999. Fertilization effects on interactions between above- and belowground competition in an old field. Ecology 80:466-480.

Cazes-Duvat, V., and R. Robert. 2001. Atlas de l’environnement côtier des îles granitiques de l›archipel des Seychelles. Université de La Réunion & CIRAD-Emvt, St. Denis, La Réunion & Montpellier, France.

Coomes, D. A., and P. J. Grubb. 2000. Impacts of root competition in forests and woodlands: A theoretical framework and review of experiments. Ecological Monographs 70:171-207. D’Antonio, C. M., and L. A. Meyerson. 2002. Exotic plant species as problems and solutions in ecological restoration: A synthesis. Restoration Ecology 10:703-713. Daehler, C. C. 2003. Performance comparisons of co-occurring native and alien invasive plants: Implications for conservation and restoration. Annual Review of Ecology and Systematics 34:183-211.

Daehler, C. C., J. S. Denslow, S. Ansari, and H.-C. Kuo. 2004. A risk-assessment system for screening out invasive pest plants from Hawaii and other Pacific Islands. Conservation Biology 18:360-368. Davis, M. A., J. P. Grime, and K. Thompson. 2000. Fluctuating resources in plant communities: A general theory of invasibility. Journal of Ecology 88:528-534. Denslow, J. S. 2003. Weeds in paradise: Thoughts on the invasibility of tropical islands. Annals of the Missouri Botanical Garden 90:119-127.

Dietz, H., and T. Steinlein. 1996. Determination of plant species cover by means of image analysis. Journal of Vegetation Science 7:131-136. Elton, C. S. 1958. The Ecology of Invasions by Animals and Plants. Methuen, London.

Fine, P. V. A. 2002. The invasibility of tropical forests by exotic plants. Journal of Tropical Ecology 18:687-705.

Fleischmann, K. 1997. Invasion of alien woody plants on the islands of Mahé and Silhouette, Seychelles. Journal of Vegetation Science 8:5-12.

Fleischmann, K. 1999. Relations between the invasive Cinnamomum verum and the endemic Phoenicophorium borsigianum on Mahé island, Seychelles. Applied Vegetation Science 2:3746. Friedmann, F. 1994. Flore des Seychelles. Orstom, Paris.

Hata, K., J. I. Suzuki, N. Kachi, and Y. Yamamura. 2006. A 19-year study of the dynamics of an invasive alien tree, Bischofia javanica, on a subtropical oceanic island. Pacific Science 60:455470. Hobbs, R. J., and L. F. Huenneke. 1992. Disturbance, diversity, and invasion: Implications for conservation. Conservation Biology 6:324-327.

Huston, M. A. 2004. Management strategies for plant invasions: Manipulating productivity, disturbance, and competition. Diversity and Distributions 10:167-178.

Kueffer, C. 2003. Habitat Restoration of Mid-altitude Secondary Cinnamon Forests in the Seychelles. Pages 147-155 in J. R. Mauremootoo, editor. Proceedings of the Regional Workshop on Invasive Alien Species and Terrestrial Ecosystem Rehabilitation in Western Indian Ocean Island States. Sharing Experience, Identifying Priorities and Defining Joint Action. Indian Ocean Commission, Quatre Bornes, Mauritius.

87

Chapter 3 Kueffer, C. 2006. Impacts of woody invasive species on tropical forests of the Seychelles. PhD Thesis, ETH Zurich, Zurich.

Kueffer, C., E. Schumacher, K. Fleischmann, P. J. Edwards, and H. Dietz. 2007. Strong belowground competition shapes tree regeneration in invasive Cinnamomum verum forests. Journal of Ecology 95:273–282.

Kueffer, C., and P. Vos. 2004. Case Studies on the Status of Invasive Woody Plant Species in the Western Indian Ocean: 5. Seychelles. Forest Health & Biosecurity Working Papers FBS/45E Forestry Department, Food and Agriculture Organization of the United Nations, Rome, Italy. Lavergne, C., J.-C. Rameau, and J. Figier. 1999. The invasive woody weed Ligustrum robustrum subsp. walkeri threatens native forests on La Réunion. Biological Invasions 1:377-392. Leps, J., V. Novovotny, L. Cizek, K. Molem, B. Isua, W. Boen, R. Kutil, J. Auga, M. Kasbal, M. Manumbor, and S. Hiuk. 2002. Successful invasion of the neotropical species Piper aduncum in rain forests in Papua New Guinea. Applied Vegetation Science 5:255-262.

Levine, J. M., P. B. Adler, and S. G. Yelenik. 2004. A meta-analysis of biotic resistance to exotic plant invasions. Ecology Letters 7:975-989.

Lonsdale, W. M. 1999. Global patterns of plant invasions and the concept of invasibility. Ecology 80:1522-1536.

Lugo, A. E. 2004. The outcome of alien tree invasions in Puerto Rico. Frontiers in Ecology and Environment 2:265-273. McNeely, J. A., H. A. Mooney, L. E. Neville, P. Schei, and J. K. Waage, editors. 2001. A Global Strategy on Invasive Alien Species. IUCN, Gland, Switzerland.

Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well-being: Biodiversity Synthesis. World Resources Institute, Washington, DC.

Richardson, D. M., and P. Pysek. 2006. Plant invasions: Merging the concepts of species invasiveness and community invasibility. Progress in Physical Geography 30:409-431.

Stohlgren, T. J., Y. Otsuki, C. A. Villa, M. Lee, and J. Belnap. 2001. Patterns of plant invasions: A case example in native species hotspots and rare habitats. Biological Invasions 3:37-50. Totland, O., P. Nyeko, A.-L. Bjerknes, S. J. Hegland, and A. Nielsen. 2005. Does forest gap size affect population size, plant size, reproductive success and pollinator visitation in Lantana camara, a tropical invasive shrub? Forest Ecology and Management 215:329-338. Turner, I. M. 2001. The Ecology of Trees in the Tropical Rain Forest. Cambridge University Press, Cambridge.

Wiser, S. K., D. R. Drake, L. E. Burrows, and W. R. Sykes. 2002. The potential for long-term persistence of forest fragments on Tongatapu, a large island in western Polynesia. Journal of Biogeography 29:767-787. Zavaleta, E., R. J. Hobbs, and H. A. Mooney. 2001. Viewing invasive species removal in a wholeecosystem context. Trends in Ecology & Evolution 16:454-459.

88

General Discussion A typology of invasive tree species in nutrient-poor oceanic island habitats The mechanisms behind the remarkable success of some alien plant species that spread rapidly into natural areas and compete efficiently with the native species are still only poorly understood. It is not likely that invasive plant species belong to a class of ‘superplants’ that can generally outperform native species. Why should such superior plant types not have evolved in the native flora? Rather, most invasive plant species probably exploit temporary windows of opportunity. For instance, through major disturbance, or environmental change in general, habitat conditions may change to a degree that the native species lose their advantage of being particularly well adapted to the local conditions (Byers 2002). For oceanic islands, it has often been argued that invasive species may profit from the presence of empty niches (e.g. Cronk and Fuller 1995, Whittaker 1998, Denslow 2003). This hypothesis seems to be supported in the cases of introduced predators (e.g. Fritts and Rodda 1998) and fixation of atmospheric nitrogen by invasive shrubs (Vitousek et al. 1987). This study indicates that the alien invasive tree flora in nutrient-poor tropical forests includes rather distinct growth strategies that defy straightforward classification of a ‘typical invader’ (see chapters 1 and 2). It may be best to characterize different types of invasive tree species that appear to exploit different windows of opportunities. If we do so, the invasive tree species of the Seychelles may be assigned to at least three different types: fast-growing invaders with high resource demands, fast-growing invaders with special adaptations to nutrient-poor soils, and stress-tolerant invaders (Tab. 1).

89

General Discussion Fast-growing invaders with high resource demands such as Tabebuia pallida or Psidium guajava invade relatively nutrient-rich and highly disturbed habitats in the lowlands but are not present in nutrient-poor and less disturbed upland forests. They represent the more classical type of an invasive plant and exploit the opportunity of human disturbance in relatively nutrient rich habitats through a highly plastic growth response to increased resource levels (see chapter 1). Lowland habitats in Seychelles are typically almost completely composed of such pantropical invasive plants. Fast-growing invaders with special adaptations to nutrient-poor soils such as Alstonia macrophylla and Falcataria moluccana show a similarly plastic response to increased resource levels (see chapters 1 and 3) and a distribution in the lowlands similar to that of the first type, but are also invasive in upland forests on nutrient-poor soil, probably because of special adaptations to nutrient-poor soils (e.g. N-fixation by F. moluccana), mycorrhiza or the development of a large root mat in the topsoil (cf. Kueffer 2006). Their success may be interpreted as the exploitation of an empty niche because these particular adaptations did not evolve in fast growing native plants on oceanic islands. Stress-tolerant invaders such as Cinnamomum verum, Psidium cattleianum and Syzygium jambos are particularly problematic in shaded habitats (e.g. the understorey of mountain mist forests) and those characterised by generally harsh environmental conditions (e.g. inselbergs). They show growth characteristics (e.g. low relative growth rates and low SLA) that are similar to many native species and exhibit a high stress tolerance (see chapters 1 and 2). Although this type of invader does not generally outperform native species (see chapters 1 to 3), its success is probably due to efficient production and dispersal of propagules, either by prolific production of frugivore-dispersed seeds (C. verum, P. cattleianum, S. jambos) or through vegetative reproduction (esp. P. cattleianum). Thanks to its high stress tolerance, this type of invader has the potential to maintain high propagule pressure at a landscape scale by establishing reproducing populations in very different types of habitat. In contrast, the population sizes of many native tree species have been dramatically reduced through anthropogenic disturbance. Correspondingly, the fruit crops of the native species are also smaller than those of the invasive species. Therefore native propagule pressure is very low. This window of opportunity of low native propagule pressure is best illustrated using C. verum and P. cattleianum as examples. Despite high seedling mortality in mid-altitude secondary C. verum forests, the species sustains dominance due to its extremely high seed rain (cf. Kueffer et al. 2007). In mountain mist forests, juvenile regeneration from seeds is generally very poor, but P. cattleianum manages to invade the understorey through the prolific production of vegetative suckers (cf. Kueffer 2006). 90

General Discussion Table 1 A proposed classification scheme of invasive woody plant species on oceanic islands, derived from the results of this study gained from successful plant invaders on the granitic oceanic islands of the Seychelles. Examples of invasive plant species abundant in Seychelles are given for each type of invader. See main text for further explanations.

Type

Characterization

‘Windows of opportunity’ exploited

Fast-growing invader with high resource demands

Not shade tolerant, very high growth rates under high light

Disturbance of lowland habitats; anthropogenic nutrient inputs

Lantana camara

Fast-growing invader with special adaptations to nutrient-poor soils

High plasticity in response to increased light

‘Add-on’ traits allow fast growth in low resource habitats (empty niche)

Alstonia macrophylla

Stress-tolerant invaders

Shade tolerant, but not very plastic response to increased light

Low native propagule pressure but high propagule pressure of the invasive species

Cinnamomum verum

High production of bird-dispersed seeds, or production of large seeds or vegetative growth

Examples

Psidium guajava Tabebuia pallida

Falcataria moluccana

Psidium cattleianum Syzygium jambos

Conclusions for management Weed risk assessment systems Knowledge of the traits of successful plant invaders has allowed to develop weed risk assessment systems for predicting new potential invasive species before they are actually introduced to a new geographical area (e.g. Pheloung et al. 1999, Wittenberg and Cock 2001, Daehler et al. 2004, Mooney et al. 2005). For Hawaii it has been stated that such a weed risk assessment system can correctly predict 95% of the current invasive flora (Daehler et al. 2004). However, the results of this study indicate that successful plant invaders in the Seychelles cover a very broad range of growth properties (see chapters 1 to 3). Apparently, many of the most problematic species are either very well adapted to conditions of low nutrient and/or low light. Especially adaptation to particular soil conditions is not 91

General Discussion well recognized in weed risk assessment systems (cf. Rejmanek et al. 2005). Species that are invasive in nutrient-poor or highly shaded tropical habitats should therefore be a particular focus of such weed risk assessment systems in the future. The results of this study from a tropical oceanic island may be an early warning for mainland tropical forest ecosystems that are also typically characterized by low light and low nutrient availability. In this sense my results agree with a recent review that discusses invasive species as an “emerging threat to tropical forests” (Laurance et al. 2006) and support one of the conclusions of the Millennium Ecosystem Assessment, namely that the impact of invasive species in tropical forests will “very rapidly increase” (Millennium Ecosystem Assessment 2005). For these reasons, it is particularly worrying that species with a high shade-tolerance or adaptations to nutrient-poor soils are currently introduced at a high rate to tropical areas for forestry or agroforestry purposes (Richardson et al. 2004, Laurance et al. 2006).

Habitat management and rehabilitation This study has shown that even closed-canopy forests are highly threatened by several invasive tree species that are currently present in the Seychelles (see chapters 1 and 3). Only the few completely undisturbed patches of native vegetation, e.g. mountain mist forests, may be relatively resistant to plant invasions (compare Fleischmann 1999), and these must therefore be protected from even minor disturbances. In habitat rehabilitation projects, it has to be recognized that only severe or multiple stress factors may prevent the spread of all invasive tree species present. In the short term, a habitat rehabilitation strategy should account for the fact that dominant established alien tree species such as Cinnamomum verum may themselves contribute to habitat resistance against further invasions, e.g. by intense root competition (compare Zavaleta et al. 2001, D’Antonio and Meyerson 2002, Kueffer 2003). However, in the end, the success of habitat rehabilitation will depend on managing propagule pressures at a site. The more that propagules of native species, and the less that propagules of problematic invaders, are dispersed to a site, the more successful habitat rehabilitation will be. Depending on abiotic habitat conditions, the composition of the canopy and propagule pressure, habitat management strategies have to be adapted. In undisturbed forest, the avoidance of any disturbance should receive the highest priority. In secondary forests dominated by alien species such as C. verum, a replacement of the alien vegetation with native trees will only be successful if the proportion of native species among the propagules can be increased, and low light and nutrient levels can be maintained along the whole habitat rehabilitation trajectory. 92

General Discussion

Literature Byers, J. E. 2002. Impact of non-indigenous species on natives enhanced by anthropogenic alteration of selection regimes. Oikos 97:449-458. Cronk, Q. C. B., and J. L. Fuller. 1995. Plant Invaders. Chapman & Hall, London, Glasgow, New York, Tokyo.

D’Antonio, C. M., and L. A. Meyerson. 2002. Exotic plant species as problems and solutions in ecological restoration: A synthesis. Restoration Ecology 10:703-713. Daehler, C. C., J. S. Denslow, S. Ansari, and H.-C. Kuo. 2004. A risk-assessment system for screening out invasive pest plants from Hawaii and other Pacific Islands. Conservation Biology 18:360-368. Denslow, J. S. 2003. Weeds in paradise: Thoughts on the invasibility of tropical islands. Annals of the Missouri Botanical Garden 90:119-127.

Fleischmann, K. 1999. Relations between the invasive Cinnamomum verum and the endemic Phoenicophorium borsigianum on Mahé island, Seychelles. Applied Vegetation Science 2:3746. Fritts, T. H., and G. H. Rodda. 1998. The role of introduced species in the degradation of island ecosystems: A case history of guam. Annual Review of Ecology and Systematics 29:113-140.

Kueffer, C. 2003. Habitat Restoration of Mid-altitude Secondary Cinnamon Forests in the Seychelles. Pages 147-155 in J. R. Mauremootoo, editor. Proceedings of the Regional Workshop on Invasive Alien Species and Terrestrial Ecosystem Rehabilitation in Western Indian Ocean Island States. Sharing Experience, Identifying Priorities and Defining Joint Action. Indian Ocean Commission, Quatre Bornes, Mauritius.

Kueffer, C. 2006. Impacts of woody invasive species on tropical forests of the Seychelles. PhD Thesis, ETH Zurich, Zurich.

Kueffer, C., E. Schumacher, K. Fleischmann, P. J. Edwards, and H. Dietz. 2007. Strong belowground competition shapes tree regeneration in invasive Cinnamomum verum forests. Journal of Ecology 95:273–282. Laurance, S. G., C. A. Peres, P. A. Jansen, and L. D’Croz. 2006. Emerging Threats to Tropical Forests: What We Know and What We Don’t Know. in S. G. Laurance, and C. A. Peres, editors. Emerging Threats to Tropical Forests. Chicago University Press, Chicago.

Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well-being: Biodiversity Synthesis. World Resources Institute, Washington, DC. Mooney, H. A., R. N. Mack, J. A. McNeely, L. E. Neville, P. J. Schei, and J. K. Waage, editors. 2005. Invasive Alien Species. A New Synthesis. Island Press, Washington, London.

Pheloung, P. C., P. A. Williams, and S. R. Halloy. 1999. A weed risk assessment model for ouse as a biosecurity tool evaluating plant introductions. Journal of Environmental Management 57:239-251. Rejmanek, M., D. M. Richardson, J. I. Higgins, M. J. Pitcairn, and E. Grotkopp. 2005. Ecology of Invasive Plants: State of the Art. in H. A. Mooney, R. N. Mack, J. A. McNeely, L. E. Neville, P. J. Schei, and J. K. Waage, editors. Invasive Alien Species. A New Synthesis. Island Press, Washington, London. Richardson, D. M., P. Binggeli, and G. Schroth. 2004. Plant Invasions - Problems and Solutions in Agroforestry. Pages 371-396 in G. F. Schroth G., C.A. Harvey, C. Gascon, H. Vasconcelos and A.M. Izac, editors. Agroforestry and Biodiversity Conservation in Tropical Landscapes. Island Press, Washington.

93

General Discussion Vitousek, P. M., L. R. Walker, L. D. Whiteaker, D. Mueller-Dombois, and P. A. Matson. 1987. Biological Invasion by Myrica faya Alters Ecosystem Development in Hawaii. Science 238:802804.

Whittaker, R. J. 1998. Island Biogeography. Ecology, Evolution, and Conservation. Oxford University Press, Oxford. Wittenberg, R., and M. J. W. Cock. 2001. Invasive Alien Species. How to Address One of the Greatest Threats to Biodiversity: A Toolkit of Best Prevention and Management Practices. CAB International, Wallingford, Oxon, UK. Zavaleta, E., R. J. Hobbs, and H. A. Mooney. 2001. Viewing invasive species removal in a wholeecosystem context. Trends in Ecology & Evolution 16:454-459.

94

Acknowledgements I would like to thank my supervisors at the Geobotanical Institute – Hansjörg Dietz, Peter Edwards and Karl Fleischmann for their overall support of my PhD project, and Luc Gigord for being the co-examiner of my thesis.

Many thanks to Peter for offering me the opportunity to do my PhD on the Seychelles islands, and for having given me the freedom to combine plant research and conservation during the time of my thesis. Hansjörg’s optimism and encouragement in many ways was the supporting leg of this study. Thank you very much for your assistance over long and short distances. Through Karl it all began. It was him who first started to do research projects in the Seychelles, and he encouraged me already in 1999 to do my master thesis in the Seychelles. Thank you for opening for me the door to a beautiful tropical island! Monika Tobler and Veronika Gmür did their master thesis on the effect of drought stress on native and invasive species in Seychelles, and thereby contributed much to chapter 2 of my thesis. Many thanks to both of you! Many thanks also to the staff and peer PhD students at the Geobotanical Institute for their help in many ways, especially Sabine for her constant and important statistical support, Rose & Marylin for the assistance with the chemical analyses, Hans-Heini & Karsten for their prompt IT support and René & Tino for their help in finding the right field equipment. The project would not have been possible without the constant and grand support of the Seychelles Ministry of Environment and Natural Resources, and particularly of the Forests and National Park section. I would like to thank Michel Vielle and Frauke Fleischer Dogley for having enabled the collaboration in the beginning, and Rolph Payet, Didier Dogley, and Joseph Francois for having supported it.

95

My experimental research at Sans Souci and in the Mare aux Cochons depended heavily on the staff of the Forests and National Parks, Conservation and Botanic Gardens sections in Victoria, Le Niol, Sans Souci, Grand Anse, Barbarons and on Praslin. Thanks a lot to Basil, Bodrigue, Breina, Cliff, Charles, Damien, Denis, Edwina, Elena, Emmanuel, Georgette, Hansel, Irena, Jacques, James, Lina, Marc, Norbert, Paul, Perley, Peter, Robert, Ronnie(s), Roy, Selby, Serge, Simon, Steven(s), Terence(s), Walter. Many thanks to the members of the plant conservation NGO Plant Conservation Action group (PCA): Charles, Denis, Didier, Frauke, James, Katy, Lindsay and Pat. Many thanks also to our friends in the Seychelles who did contribute to wonderful three years on the islands in many different ways such as car repairing, cooking great meals, counting Cinnamon seedlings, providing accommodation, giving medical advices, organising sport activities and much much more! Thank you Alex, Andy and family, Angelika, Ashley, Chris, Dave, Emilia, Georges, Hardy, Harry, the Seychelles Hash Runners, James, Joanna, Justin, Kenneth, Laura, Margaret, Marie-Therese, Marylene, Maureen, Michel, Mike, Peter, Pierre, Precy, Rachel, Robin, Terence, Unels, Van and all their kids! I also thank Terence Coopoosamy and the Seychelles Bureau of Standards for the support of the project, Peter Lalande and the National Archive for their source of local references and Robert Lajoie from Meteo Seychelles for providing meteorological data. The project was supported financially or in kind by an ETH Zurich research grant, Air Seychelles, the Stiftung Rübel, and the Seychelles Ministry of Environment and Natural Resources. During the whole time of my PhD I also very much depended on the help, (technical) assistance, moral support and delivery of goods (like chocolate, spare parts, “Tagesanzeiger”-magazines…) of friends and family. My thanks go to: Anna, Barbara, Christian, Christina, Claudia, Dave, Esther, Felix, Jean-Pierre, Judith, Jürg, Kuese, Georg v A, Georg K, Hanni, Hans-Heini, Helen, Holger, Isabella, Kaspar, Kowi, Lea, Lisa, Lilian, Luzia, Mario, Markus H, Markus S, Meltem, Moni, Monika W, Michel, Michi, Naomi, Priska, Renate, Rene, Roman, Saskia, Silke, Silvan, Simone, Dominik, Stefan B, Stefan W, Stefan Z, Tania, Thomas, Tomas, Ueli, Urs, Ursula and particularly my parents Peter & Rita Schumacher, and Ursula & Urs! Last but not least my sincere thanks to Christoph for his overall support and superb working together in this project and in nature conservation in the Indian Ocean region. 96

Curriculum Vitae Name:

Eva Schumacher

Date of Birth:

10th of October 1973

Nationality:

Citizen of Vilters-Wangs SG, Switzerland

Education 1994 – 2000

Study of Environmental Natural Sciences at ETH Zurich



Degree: Dipl. Umwelt-natw. ETH Thesis title: Vegetation Survey of Cousin and Cousine Island in the Seychelles

1994

Matura (high school graduation), Gymnasium Frauenfeld

Work experience 2006 –

Scientific assistant, Institute of Integrative Biology, ETH Zurich

Oct. – Nov. 2005

Scientific assistant, Swiss Federal Institute of Forest, Snow and Landscape Research WSL, Birmensdorf

2003 –

A founder and editor, Kapisen – Seychelles Plant Conservation Newsletter www.plantecology.ethz.ch/publications/books/kapisen

2002 – 2005

Consultant for invasive species management, Seychelles Ministry of Environment and Natural Resources

Jan. – March 2002 Pilot study for the research project Invasions of woody plants into Seychelles tropical forests, Geobotanical Institute, ETH Zurich 2000 – 2005

Scientific assistant, Geobotanical Institute, ETH Zurich

2000 – Jan. 2002

Scientific assistant, Foundation for Research on Information Technology in Society (IT IS), Zurich

June – Oct. 2000

Scientific assistant, Vogelwarte Sempach

97

Publications KUEFFER, C., E. Schumacher, K. FLEISCHMANN, P. J. EDWARDS & H. DIETZ. 2007. Strong below-ground competition shapes tree regeneration in invasive Cinnamomum verum forests. Journal of Ecology 95:273–282. Kueffer, C. & E. Schumacher. 2005. Consultancy on invasive species management in Seychelles - Final report. Seychelles Ministry of Environment and Natural Resources, Victoria, Seychelles. Kueffer, C., P. J. Edwards, K. Fleischmann, E. Schumacher & H. Dietz. 2003. Invasion of woody plants into the Seychelles tropical forests: Habitat invasibility and propagule pressure. Bulletin of the Geobotanical Institute ETH 69:65-75. Schumacher, E., H. Dietz, K. Fleischmann, C. Kueffer & P. J. Edwards. 2003. Invasion of woody plants into the Seychelles tropical forests: Species traits in the establishment phase. Bulletin of the Geobotanical Institute ETH 69:77-86.

Conference contributions Kueffer, C. & E. Schumacher 2005. Research collaborations for plant conservation: Concepts, tools and examples. The Global Partnership for Plant Conservation - Plants 2010 Conference. October 23-25, Dublin, Irland. (poster). Kueffer, C., G. Klingler, K. Zirfass, E. Schumacher, S. Güsewell & P. J. Edwards 2005. Woody invasive species increase nutrient cycling in the granitic Seychelles. International Botanical Congress (IBC). July 17-23, Vienna, Austria. (oral). Schumacher, E., P. J. Edwards, M. Tobler, C. Kueffer & H. Dietz 2005. Tree invasions into Seychelles tropical forests: ecophysiology of native vs. invasive species. International Botanical Congress (IBC). July 17-23, Vienna, Austria. (poster). Kueffer, C., P.J. Edwards, K. Fleischmann, E. Schumacher & H. Dietz. 2004. Habitat invasibility of tropical forests in the Seychelles: The influence of belowground competition. 3rd International Conference on Biological Invasions NEOBIOTA - From Ecology to Control. September 30 – October 1, Bern, Switzerland. (poster). Kueffer, C., P. J. Edwards, K. Fleischmann, E. Schumacher, & H. Dietz. 2003. Dynamics of native vs. invasive woody plants in Seychelles tropical forests: a comparative study of recruitment limitation, and sapling performance under varying environmental conditions. Biotic Interactions in the Tropics: A Special Symposium of the British Ecological Society and the Annual Meeting of the Association fo Tropical Biology and Conservation. July 7-10, Aberdeen, Scotland. (oral). Schumacher, E., H. Dietz, K. Fleischmann, C. Kueffer & P. J. Edwards. 2003. Woody invasions into the Seychelles tropical forests: species traits in the establishment phase. Biotic Interactions in the Tropics: A Special Symposium of the British Ecological Society and the Annual Meeting of the Association fo Tropical Biology and Conservation. July 7-10, Aberdeen, Scotland. (oral).

98